diff options
Diffstat (limited to 'kernel/trace')
129 files changed, 26147 insertions, 8153 deletions
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index d7043043f59c..bfa2ec46e075 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -31,6 +31,14 @@ config HAVE_FUNCTION_GRAPH_TRACER help See Documentation/trace/ftrace-design.rst +config HAVE_FUNCTION_GRAPH_FREGS + bool + +config HAVE_FTRACE_GRAPH_FUNC + bool + help + True if ftrace_graph_func() is defined. + config HAVE_DYNAMIC_FTRACE bool help @@ -42,6 +50,15 @@ config HAVE_DYNAMIC_FTRACE_WITH_REGS config HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS bool +config HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS + bool + +config HAVE_EXTRA_IPI_TRACEPOINTS + bool + help + For architectures that use ipi_raise, ipi_entry and ipi_exit + tracepoints. + config HAVE_DYNAMIC_FTRACE_WITH_ARGS bool help @@ -51,16 +68,23 @@ config HAVE_DYNAMIC_FTRACE_WITH_ARGS This allows for use of ftrace_regs_get_argument() and ftrace_regs_get_stack_pointer(). +config HAVE_FTRACE_REGS_HAVING_PT_REGS + bool + help + If this is set, ftrace_regs has pt_regs, thus it can convert to + pt_regs without allocating memory. + config HAVE_DYNAMIC_FTRACE_NO_PATCHABLE bool help If the architecture generates __patchable_function_entries sections but does not want them included in the ftrace locations. -config HAVE_FTRACE_MCOUNT_RECORD +config HAVE_DYNAMIC_FTRACE_WITH_JMP bool help - See Documentation/trace/ftrace-design.rst + If the architecture supports to replace the __fentry__ with a + "jmp" instruction. config HAVE_SYSCALL_TRACEPOINTS bool @@ -157,7 +181,7 @@ config TRACING select BINARY_PRINTF select EVENT_TRACING select TRACE_CLOCK - select TASKS_RCU if PREEMPTION + select NEED_TASKS_RCU config GENERIC_TRACER bool @@ -182,6 +206,19 @@ menuconfig FTRACE if FTRACE +config TRACEFS_AUTOMOUNT_DEPRECATED + bool "Automount tracefs on debugfs [DEPRECATED]" + depends on TRACING + default y + help + The tracing interface was moved from /sys/kernel/debug/tracing + to /sys/kernel/tracing in 2015, but the tracing file system + was still automounted in /sys/kernel/debug for backward + compatibility with tooling. + + The new interface has been around for more than 10 years and + the old debug mount will soon be removed. + config BOOTTIME_TRACING bool "Boot-time Tracing support" depends on TRACING @@ -198,7 +235,7 @@ config FUNCTION_TRACER select GENERIC_TRACER select CONTEXT_SWITCH_TRACER select GLOB - select TASKS_RCU if PREEMPTION + select NEED_TASKS_RCU select TASKS_RUDE_RCU help Enable the kernel to trace every kernel function. This is done @@ -224,8 +261,41 @@ config FUNCTION_GRAPH_TRACER the return value. This is done by setting the current return address on the current task structure into a stack of calls. +config FUNCTION_GRAPH_RETVAL + bool "Kernel Function Graph Return Value" + depends on HAVE_FUNCTION_GRAPH_FREGS + depends on FUNCTION_GRAPH_TRACER + default n + help + Support recording and printing the function return value when + using function graph tracer. It can be helpful to locate functions + that return errors. This feature is off by default, and you can + enable it via the trace option funcgraph-retval. + See Documentation/trace/ftrace.rst + +config FUNCTION_GRAPH_RETADDR + bool "Kernel Function Graph Return Address" + depends on FUNCTION_GRAPH_TRACER + default n + help + Support recording and printing the function return address when + using function graph tracer. It can be helpful to locate code line that + the function is called. This feature is off by default, and you can + enable it via the trace option funcgraph-retaddr. + +config FUNCTION_TRACE_ARGS + bool + depends on PROBE_EVENTS_BTF_ARGS + default y + help + If supported with function argument access API and BTF, then + the function tracer and function graph tracer will support printing + of function arguments. This feature is off by default, and can be + enabled via the trace option func-args (for the function tracer) and + funcgraph-args (for the function graph tracer) + config DYNAMIC_FTRACE - bool "enable/disable function tracing dynamically" + bool depends on FUNCTION_TRACER depends on HAVE_DYNAMIC_FTRACE default y @@ -239,7 +309,7 @@ config DYNAMIC_FTRACE enabled, and the functions not enabled will not affect performance of the system. - See the files in /sys/kernel/debug/tracing: + See the files in /sys/kernel/tracing: available_filter_functions set_ftrace_filter set_ftrace_notrace @@ -254,20 +324,43 @@ config DYNAMIC_FTRACE_WITH_REGS config DYNAMIC_FTRACE_WITH_DIRECT_CALLS def_bool y - depends on DYNAMIC_FTRACE_WITH_REGS + depends on DYNAMIC_FTRACE_WITH_REGS || DYNAMIC_FTRACE_WITH_ARGS depends on HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS +config DYNAMIC_FTRACE_WITH_CALL_OPS + def_bool y + depends on HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS + config DYNAMIC_FTRACE_WITH_ARGS def_bool y depends on DYNAMIC_FTRACE depends on HAVE_DYNAMIC_FTRACE_WITH_ARGS +config DYNAMIC_FTRACE_WITH_JMP + def_bool y + depends on DYNAMIC_FTRACE + depends on DYNAMIC_FTRACE_WITH_DIRECT_CALLS + depends on HAVE_DYNAMIC_FTRACE_WITH_JMP + +config FUNCTION_SELF_TRACING + bool "Function trace tracing code" + depends on FUNCTION_TRACER + help + Normally all the tracing code is set to notrace, where the function + tracer will ignore all the tracing functions. Sometimes it is useful + for debugging to trace some of the tracing infratructure itself. + Enable this to allow some of the tracing infrastructure to be traced + by the function tracer. Note, this will likely add noise to function + tracing if events and other tracing features are enabled along with + function tracing. + + If unsure, say N. + config FPROBE bool "Kernel Function Probe (fprobe)" - depends on FUNCTION_TRACER - depends on DYNAMIC_FTRACE_WITH_REGS - depends on HAVE_RETHOOK - select RETHOOK + depends on HAVE_FUNCTION_GRAPH_FREGS && HAVE_FTRACE_GRAPH_FUNC + depends on DYNAMIC_FTRACE_WITH_ARGS + select FUNCTION_GRAPH_TRACER default n help This option enables kernel function probe (fprobe) based on ftrace. @@ -299,7 +392,7 @@ config STACK_TRACER select KALLSYMS help This special tracer records the maximum stack footprint of the - kernel and displays it in /sys/kernel/debug/tracing/stack_trace. + kernel and displays it in /sys/kernel/tracing/stack_trace. This tracer works by hooking into every function call that the kernel executes, and keeping a maximum stack depth value and @@ -339,7 +432,7 @@ config IRQSOFF_TRACER disabled by default and can be runtime (re-)started via: - echo 0 > /sys/kernel/debug/tracing/tracing_max_latency + echo 0 > /sys/kernel/tracing/tracing_max_latency (Note that kernel size and overhead increase with this option enabled. This option and the preempt-off timing option can be @@ -363,7 +456,7 @@ config PREEMPT_TRACER disabled by default and can be runtime (re-)started via: - echo 0 > /sys/kernel/debug/tracing/tracing_max_latency + echo 0 > /sys/kernel/tracing/tracing_max_latency (Note that kernel size and overhead increase with this option enabled. This option and the irqs-off timing option can be @@ -508,6 +601,20 @@ config FTRACE_SYSCALLS help Basic tracer to catch the syscall entry and exit events. +config TRACE_SYSCALL_BUF_SIZE_DEFAULT + int "System call user read max size" + range 0 165 + default 63 + depends on FTRACE_SYSCALLS + help + Some system call trace events will record the data from a user + space address that one of the parameters point to. The amount of + data per event is limited. That limit is set by this config and + this config also affects how much user space data perf can read. + + For a tracing instance, this size may be changed by writing into + its syscall_user_buf_size file. + config TRACER_SNAPSHOT bool "Create a snapshot trace buffer" select TRACER_MAX_TRACE @@ -515,7 +622,7 @@ config TRACER_SNAPSHOT Allow tracing users to take snapshot of the current buffer using the ftrace interface, e.g.: - echo 1 > /sys/kernel/debug/tracing/snapshot + echo 1 > /sys/kernel/tracing/snapshot cat snapshot config TRACER_SNAPSHOT_PER_CPU_SWAP @@ -527,7 +634,7 @@ config TRACER_SNAPSHOT_PER_CPU_SWAP full swap (all buffers). If this is set, then the following is allowed: - echo 1 > /sys/kernel/debug/tracing/per_cpu/cpu2/snapshot + echo 1 > /sys/kernel/tracing/per_cpu/cpu2/snapshot After which, only the tracing buffer for CPU 2 was swapped with the main tracing buffer, and the other CPU buffers remain the same. @@ -574,7 +681,7 @@ config PROFILE_ANNOTATED_BRANCHES This tracer profiles all likely and unlikely macros in the kernel. It will display the results in: - /sys/kernel/debug/tracing/trace_stat/branch_annotated + /sys/kernel/tracing/trace_stat/branch_annotated Note: this will add a significant overhead; only turn this on if you need to profile the system's use of these macros. @@ -587,7 +694,7 @@ config PROFILE_ALL_BRANCHES taken in the kernel is recorded whether it hit or miss. The results will be displayed in: - /sys/kernel/debug/tracing/trace_stat/branch_all + /sys/kernel/tracing/trace_stat/branch_all This option also enables the likely/unlikely profiler. @@ -638,11 +745,37 @@ config BLK_DEV_IO_TRACE Tracing also is possible using the ftrace interface, e.g.: echo 1 > /sys/block/sda/sda1/trace/enable - echo blk > /sys/kernel/debug/tracing/current_tracer - cat /sys/kernel/debug/tracing/trace_pipe + echo blk > /sys/kernel/tracing/current_tracer + cat /sys/kernel/tracing/trace_pipe If unsure, say N. +config FPROBE_EVENTS + depends on FPROBE + depends on HAVE_REGS_AND_STACK_ACCESS_API + bool "Enable fprobe-based dynamic events" + select TRACING + select PROBE_EVENTS + select DYNAMIC_EVENTS + default y + help + This allows user to add tracing events on the function entry and + exit via ftrace interface. The syntax is same as the kprobe events + and the kprobe events on function entry and exit will be + transparently converted to this fprobe events. + +config PROBE_EVENTS_BTF_ARGS + depends on HAVE_FUNCTION_ARG_ACCESS_API + depends on FPROBE_EVENTS || KPROBE_EVENTS + depends on DEBUG_INFO_BTF && BPF_SYSCALL + bool "Support BTF function arguments for probe events" + default y + help + The user can specify the arguments of the probe event using the names + of the arguments of the probed function, when the probe location is a + kernel function entry or a tracepoint. + This is available only if BTF (BPF Type Format) support is enabled. + config KPROBE_EVENTS depends on KPROBES depends on HAVE_REGS_AND_STACK_ACCESS_API @@ -700,6 +833,20 @@ config UPROBE_EVENTS This option is required if you plan to use perf-probe subcommand of perf tools on user space applications. +config EPROBE_EVENTS + bool "Enable event-based dynamic events" + depends on TRACING + depends on HAVE_REGS_AND_STACK_ACCESS_API + select PROBE_EVENTS + select DYNAMIC_EVENTS + default y + help + Eprobes are dynamic events that can be placed on other existing + events. It can be used to limit what fields are recorded in + an event or even dereference a field of an event. It can + convert the type of an event field. For example, turn an + address into a string. + config BPF_EVENTS depends on BPF_SYSCALL depends on (KPROBE_EVENTS || UPROBE_EVENTS) && PERF_EVENTS @@ -724,27 +871,22 @@ config BPF_KPROBE_OVERRIDE Allows BPF to override the execution of a probed function and set a different return value. This is used for error injection. -config FTRACE_MCOUNT_RECORD - def_bool y - depends on DYNAMIC_FTRACE - depends on HAVE_FTRACE_MCOUNT_RECORD - config FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY bool - depends on FTRACE_MCOUNT_RECORD + depends on DYNAMIC_FTRACE config FTRACE_MCOUNT_USE_CC def_bool y depends on $(cc-option,-mrecord-mcount) depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY - depends on FTRACE_MCOUNT_RECORD + depends on DYNAMIC_FTRACE config FTRACE_MCOUNT_USE_OBJTOOL def_bool y depends on HAVE_OBJTOOL_MCOUNT depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY depends on !FTRACE_MCOUNT_USE_CC - depends on FTRACE_MCOUNT_RECORD + depends on DYNAMIC_FTRACE select OBJTOOL config FTRACE_MCOUNT_USE_RECORDMCOUNT @@ -752,7 +894,7 @@ config FTRACE_MCOUNT_USE_RECORDMCOUNT depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY depends on !FTRACE_MCOUNT_USE_CC depends on !FTRACE_MCOUNT_USE_OBJTOOL - depends on FTRACE_MCOUNT_RECORD + depends on DYNAMIC_FTRACE config TRACING_MAP bool @@ -785,15 +927,15 @@ config USER_EVENTS bool "User trace events" select TRACING select DYNAMIC_EVENTS - depends on BROKEN || COMPILE_TEST # API needs to be straighten out help User trace events are user-defined trace events that can be used like an existing kernel trace event. User trace events are generated by writing to a tracefs file. User processes can determine if their tracing events should be - generated by memory mapping a tracefs file and checking for - an associated byte being non-zero. + generated by registering a value and bit with the kernel + that reflects when it is enabled or not. + See Documentation/trace/user_events.rst. If in doubt, say N. config HIST_TRIGGERS @@ -917,7 +1059,7 @@ config FTRACE_RECORD_RECURSION config FTRACE_RECORD_RECURSION_SIZE int "Max number of recursed functions to record" - default 128 + default 128 depends on FTRACE_RECORD_RECURSION help This defines the limit of number of functions that can be @@ -926,6 +1068,19 @@ config FTRACE_RECORD_RECURSION_SIZE This file can be reset, but the limit can not change in size at runtime. +config FTRACE_VALIDATE_RCU_IS_WATCHING + bool "Validate RCU is on during ftrace execution" + depends on FUNCTION_TRACER + depends on ARCH_WANTS_NO_INSTR + help + All callbacks that attach to the function tracing have some sort of + protection against recursion. This option is only to verify that + ftrace (and other users of ftrace_test_recursion_trylock()) are not + called outside of RCU, as if they are, it can cause a race. But it + also has a noticeable overhead when enabled. + + If unsure, say N + config RING_BUFFER_RECORD_RECURSION bool "Record functions that recurse in the ring buffer" depends on FTRACE_RECORD_RECURSION @@ -1075,7 +1230,7 @@ config PREEMPTIRQ_DELAY_TEST config SYNTH_EVENT_GEN_TEST tristate "Test module for in-kernel synthetic event generation" - depends on SYNTH_EVENTS + depends on SYNTH_EVENTS && m help This option creates a test module to check the base functionality of in-kernel synthetic event definition and @@ -1088,7 +1243,7 @@ config SYNTH_EVENT_GEN_TEST config KPROBE_EVENT_GEN_TEST tristate "Test module for in-kernel kprobe event generation" - depends on KPROBE_EVENTS + depends on KPROBE_EVENTS && m help This option creates a test module to check the base functionality of in-kernel kprobe event definition. diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index c6651e16b557..fc5dcc888e13 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -16,6 +16,23 @@ obj-y += trace_selftest_dynamic.o endif endif +# Allow some files to be function traced +ifdef CONFIG_FUNCTION_SELF_TRACING +CFLAGS_trace_output.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_seq.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_stat.o = $(CC_FLAGS_FTRACE) +CFLAGS_tracing_map.o = $(CC_FLAGS_FTRACE) +CFLAGS_synth_event_gen_test.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_events.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_syscalls.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_events_filter.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_events_trigger.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_events_synth.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_events_hist.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_events_user.o = $(CC_FLAGS_FTRACE) +CFLAGS_trace_dynevent.o = $(CC_FLAGS_FTRACE) +endif + ifdef CONFIG_FTRACE_STARTUP_TEST CFLAGS_trace_kprobe_selftest.o = $(CC_FLAGS_FTRACE) obj-$(CONFIG_KPROBE_EVENTS) += trace_kprobe_selftest.o @@ -82,7 +99,7 @@ obj-$(CONFIG_EVENT_TRACING) += trace_event_perf.o endif obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_EVENT_TRACING) += trace_events_trigger.o -obj-$(CONFIG_PROBE_EVENTS) += trace_eprobe.o +obj-$(CONFIG_EPROBE_EVENTS) += trace_eprobe.o obj-$(CONFIG_TRACE_EVENT_INJECT) += trace_events_inject.o obj-$(CONFIG_SYNTH_EVENTS) += trace_events_synth.o obj-$(CONFIG_HIST_TRIGGERS) += trace_events_hist.o @@ -99,11 +116,13 @@ obj-$(CONFIG_KGDB_KDB) += trace_kdb.o endif obj-$(CONFIG_DYNAMIC_EVENTS) += trace_dynevent.o obj-$(CONFIG_PROBE_EVENTS) += trace_probe.o +obj-$(CONFIG_PROBE_EVENTS_BTF_ARGS) += trace_btf.o obj-$(CONFIG_UPROBE_EVENTS) += trace_uprobe.o obj-$(CONFIG_BOOTTIME_TRACING) += trace_boot.o obj-$(CONFIG_FTRACE_RECORD_RECURSION) += trace_recursion_record.o obj-$(CONFIG_FPROBE) += fprobe.o obj-$(CONFIG_RETHOOK) += rethook.o +obj-$(CONFIG_FPROBE_EVENTS) += trace_fprobe.o obj-$(CONFIG_TRACEPOINT_BENCHMARK) += trace_benchmark.o obj-$(CONFIG_RV) += rv/ diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 918a7d12df8f..d031c8d80be4 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -63,13 +63,116 @@ static int blk_probes_ref; static void blk_register_tracepoints(void); static void blk_unregister_tracepoints(void); +static void record_blktrace_event(struct blk_io_trace *t, pid_t pid, int cpu, + sector_t sector, int bytes, u64 what, + dev_t dev, int error, u64 cgid, + ssize_t cgid_len, void *pdu_data, int pdu_len) + +{ + /* + * These two are not needed in ftrace as they are in the + * generic trace_entry, filled by tracing_generic_entry_update, + * but for the trace_event->bin() synthesizer benefit we do it + * here too. + */ + t->cpu = cpu; + t->pid = pid; + + t->sector = sector; + t->bytes = bytes; + t->action = lower_32_bits(what); + t->device = dev; + t->error = error; + t->pdu_len = pdu_len + cgid_len; + + if (cgid_len) + memcpy((void *)t + sizeof(*t), &cgid, cgid_len); + if (pdu_len) + memcpy((void *)t + sizeof(*t) + cgid_len, pdu_data, pdu_len); +} + +static void record_blktrace_event2(struct blk_io_trace2 *t2, pid_t pid, int cpu, + sector_t sector, int bytes, u64 what, + dev_t dev, int error, u64 cgid, + ssize_t cgid_len, void *pdu_data, + int pdu_len) +{ + t2->pid = pid; + t2->cpu = cpu; + + t2->sector = sector; + t2->bytes = bytes; + t2->action = what; + t2->device = dev; + t2->error = error; + t2->pdu_len = pdu_len + cgid_len; + + if (cgid_len) + memcpy((void *)t2 + sizeof(*t2), &cgid, cgid_len); + if (pdu_len) + memcpy((void *)t2 + sizeof(*t2) + cgid_len, pdu_data, pdu_len); +} + +static void relay_blktrace_event1(struct blk_trace *bt, unsigned long sequence, + pid_t pid, int cpu, sector_t sector, int bytes, + u64 what, int error, u64 cgid, + ssize_t cgid_len, void *pdu_data, int pdu_len) +{ + struct blk_io_trace *t; + size_t trace_len = sizeof(*t) + pdu_len + cgid_len; + + t = relay_reserve(bt->rchan, trace_len); + if (!t) + return; + + t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; + t->sequence = sequence; + t->time = ktime_to_ns(ktime_get()); + + record_blktrace_event(t, pid, cpu, sector, bytes, what, bt->dev, error, + cgid, cgid_len, pdu_data, pdu_len); +} + +static void relay_blktrace_event2(struct blk_trace *bt, unsigned long sequence, + pid_t pid, int cpu, sector_t sector, + int bytes, u64 what, int error, u64 cgid, + ssize_t cgid_len, void *pdu_data, int pdu_len) +{ + struct blk_io_trace2 *t; + size_t trace_len = sizeof(struct blk_io_trace2) + pdu_len + cgid_len; + + t = relay_reserve(bt->rchan, trace_len); + if (!t) + return; + + t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE2_VERSION; + t->sequence = sequence; + t->time = ktime_to_ns(ktime_get()); + + record_blktrace_event2(t, pid, cpu, sector, bytes, what, bt->dev, error, + cgid, cgid_len, pdu_data, pdu_len); +} + +static void relay_blktrace_event(struct blk_trace *bt, unsigned long sequence, + pid_t pid, int cpu, sector_t sector, int bytes, + u64 what, int error, u64 cgid, + ssize_t cgid_len, void *pdu_data, int pdu_len) +{ + if (bt->version == 2) + return relay_blktrace_event2(bt, sequence, pid, cpu, sector, + bytes, what, error, cgid, cgid_len, + pdu_data, pdu_len); + return relay_blktrace_event1(bt, sequence, pid, cpu, sector, bytes, + what, error, cgid, cgid_len, pdu_data, + pdu_len); +} + /* * Send out a notify message. */ -static void trace_note(struct blk_trace *bt, pid_t pid, int action, +static void trace_note(struct blk_trace *bt, pid_t pid, u64 action, const void *data, size_t len, u64 cgid) { - struct blk_io_trace *t; struct ring_buffer_event *event = NULL; struct trace_buffer *buffer = NULL; unsigned int trace_ctx = 0; @@ -77,38 +180,30 @@ static void trace_note(struct blk_trace *bt, pid_t pid, int action, bool blk_tracer = blk_tracer_enabled; ssize_t cgid_len = cgid ? sizeof(cgid) : 0; + action = lower_32_bits(action | (cgid ? __BLK_TN_CGROUP : 0)); if (blk_tracer) { + struct blk_io_trace2 *t; + size_t trace_len = sizeof(*t) + cgid_len + len; + buffer = blk_tr->array_buffer.buffer; trace_ctx = tracing_gen_ctx_flags(0); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, - sizeof(*t) + len + cgid_len, - trace_ctx); + trace_len, trace_ctx); if (!event) return; t = ring_buffer_event_data(event); - goto record_it; + record_blktrace_event2(t, pid, cpu, 0, 0, + action, bt->dev, 0, cgid, cgid_len, + (void *)data, len); + trace_buffer_unlock_commit(blk_tr, buffer, event, trace_ctx); + return; } if (!bt->rchan) return; - t = relay_reserve(bt->rchan, sizeof(*t) + len + cgid_len); - if (t) { - t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; - t->time = ktime_to_ns(ktime_get()); -record_it: - t->device = bt->dev; - t->action = action | (cgid ? __BLK_TN_CGROUP : 0); - t->pid = pid; - t->cpu = cpu; - t->pdu_len = len + cgid_len; - if (cgid_len) - memcpy((void *)t + sizeof(*t), &cgid, cgid_len); - memcpy((void *) t + sizeof(*t) + cgid_len, data, len); - - if (blk_tracer) - trace_buffer_unlock_commit(blk_tr, buffer, event, trace_ctx); - } + relay_blktrace_event(bt, 0, pid, cpu, 0, 0, action, 0, cgid, + cgid_len, (void *)data, len); } /* @@ -182,7 +277,7 @@ void __blk_trace_note_message(struct blk_trace *bt, } EXPORT_SYMBOL_GPL(__blk_trace_note_message); -static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector, +static int act_log_check(struct blk_trace *bt, u64 what, sector_t sector, pid_t pid) { if (((bt->act_mask << BLK_TC_SHIFT) & what) == 0) @@ -213,13 +308,12 @@ static const u32 ddir_act[2] = { BLK_TC_ACT(BLK_TC_READ), * blk_io_trace structure and places it in a per-cpu subbuffer. */ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, - const blk_opf_t opf, u32 what, int error, + const blk_opf_t opf, u64 what, int error, int pdu_len, void *pdu_data, u64 cgid) { struct task_struct *tsk = current; struct ring_buffer_event *event = NULL; struct trace_buffer *buffer = NULL; - struct blk_io_trace *t; unsigned long flags = 0; unsigned long *sequence; unsigned int trace_ctx = 0; @@ -228,6 +322,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, bool blk_tracer = blk_tracer_enabled; ssize_t cgid_len = cgid ? sizeof(cgid) : 0; const enum req_op op = opf & REQ_OP_MASK; + size_t trace_len; if (unlikely(bt->trace_state != Blktrace_running && !blk_tracer)) return; @@ -238,10 +333,47 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, what |= MASK_TC_BIT(opf, META); what |= MASK_TC_BIT(opf, PREFLUSH); what |= MASK_TC_BIT(opf, FUA); - if (op == REQ_OP_DISCARD || op == REQ_OP_SECURE_ERASE) + + switch (op) { + case REQ_OP_DISCARD: + case REQ_OP_SECURE_ERASE: what |= BLK_TC_ACT(BLK_TC_DISCARD); - if (op == REQ_OP_FLUSH) + break; + case REQ_OP_FLUSH: what |= BLK_TC_ACT(BLK_TC_FLUSH); + break; + case REQ_OP_ZONE_APPEND: + what |= BLK_TC_ACT(BLK_TC_ZONE_APPEND); + break; + case REQ_OP_ZONE_RESET: + what |= BLK_TC_ACT(BLK_TC_ZONE_RESET); + break; + case REQ_OP_ZONE_RESET_ALL: + what |= BLK_TC_ACT(BLK_TC_ZONE_RESET_ALL); + break; + case REQ_OP_ZONE_FINISH: + what |= BLK_TC_ACT(BLK_TC_ZONE_FINISH); + break; + case REQ_OP_ZONE_OPEN: + what |= BLK_TC_ACT(BLK_TC_ZONE_OPEN); + break; + case REQ_OP_ZONE_CLOSE: + what |= BLK_TC_ACT(BLK_TC_ZONE_CLOSE); + break; + case REQ_OP_WRITE_ZEROES: + what |= BLK_TC_ACT(BLK_TC_WRITE_ZEROES); + break; + default: + break; + } + + /* Drop trace events for zone operations with blktrace v1 */ + if (bt->version == 1 && (what >> BLK_TC_SHIFT) > BLK_TC_END_V1) { + pr_debug_ratelimited("blktrace v1 cannot trace zone operation 0x%llx\n", + (unsigned long long)what); + return; + } + if (cgid) what |= __BLK_TA_CGROUP; @@ -255,13 +387,68 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, buffer = blk_tr->array_buffer.buffer; trace_ctx = tracing_gen_ctx_flags(0); + switch (bt->version) { + case 1: + trace_len = sizeof(struct blk_io_trace); + break; + case 2: + default: + /* + * ftrace always uses v2 (blk_io_trace2) format. + * + * For sysfs-enabled tracing path (enabled via + * /sys/block/DEV/trace/enable), blk_trace_setup_queue() + * never initializes bt->version, leaving it 0 from + * kzalloc(). We must handle version==0 safely here. + * + * Fall through to default to ensure we never hit the + * old bug where default set trace_len=0, causing + * buffer underflow and memory corruption. + * + * Always use v2 format for ftrace and normalize + * bt->version to 2 when uninitialized. + */ + trace_len = sizeof(struct blk_io_trace2); + if (bt->version == 0) + bt->version = 2; + break; + } + trace_len += pdu_len + cgid_len; event = trace_buffer_lock_reserve(buffer, TRACE_BLK, - sizeof(*t) + pdu_len + cgid_len, - trace_ctx); + trace_len, trace_ctx); if (!event) return; - t = ring_buffer_event_data(event); - goto record_it; + + switch (bt->version) { + case 1: + record_blktrace_event(ring_buffer_event_data(event), + pid, cpu, sector, bytes, + what, bt->dev, error, cgid, cgid_len, + pdu_data, pdu_len); + break; + case 2: + default: + /* + * Use v2 recording function (record_blktrace_event2) + * which writes blk_io_trace2 structure with correct + * field layout: + * - 32-bit pid at offset 28 + * - 64-bit action at offset 32 + * + * Fall through to default handles version==0 case + * (from sysfs path), ensuring we always use correct + * v2 recording function to match the v2 buffer + * allocated above. + */ + record_blktrace_event2(ring_buffer_event_data(event), + pid, cpu, sector, bytes, + what, bt->dev, error, cgid, cgid_len, + pdu_data, pdu_len); + break; + } + + trace_buffer_unlock_commit(blk_tr, buffer, event, trace_ctx); + return; } if (unlikely(tsk->btrace_seq != blktrace_seq)) @@ -273,41 +460,10 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, * from coming in and stepping on our toes. */ local_irq_save(flags); - t = relay_reserve(bt->rchan, sizeof(*t) + pdu_len + cgid_len); - if (t) { - sequence = per_cpu_ptr(bt->sequence, cpu); - - t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; - t->sequence = ++(*sequence); - t->time = ktime_to_ns(ktime_get()); -record_it: - /* - * These two are not needed in ftrace as they are in the - * generic trace_entry, filled by tracing_generic_entry_update, - * but for the trace_event->bin() synthesizer benefit we do it - * here too. - */ - t->cpu = cpu; - t->pid = pid; - - t->sector = sector; - t->bytes = bytes; - t->action = what; - t->device = bt->dev; - t->error = error; - t->pdu_len = pdu_len + cgid_len; - - if (cgid_len) - memcpy((void *)t + sizeof(*t), &cgid, cgid_len); - if (pdu_len) - memcpy((void *)t + sizeof(*t) + cgid_len, pdu_data, pdu_len); - - if (blk_tracer) { - trace_buffer_unlock_commit(blk_tr, buffer, event, trace_ctx); - return; - } - } - + sequence = per_cpu_ptr(bt->sequence, cpu); + (*sequence)++; + relay_blktrace_event(bt, *sequence, pid, cpu, sector, bytes, + what, error, cgid, cgid_len, pdu_data, pdu_len); local_irq_restore(flags); } @@ -320,8 +476,8 @@ static void blk_trace_free(struct request_queue *q, struct blk_trace *bt) * under 'q->debugfs_dir', thus lookup and remove them. */ if (!bt->dir) { - debugfs_remove(debugfs_lookup("dropped", q->debugfs_dir)); - debugfs_remove(debugfs_lookup("msg", q->debugfs_dir)); + debugfs_lookup_and_remove("dropped", q->debugfs_dir); + debugfs_lookup_and_remove("msg", q->debugfs_dir); } else { debugfs_remove(bt->dir); } @@ -415,9 +571,10 @@ static ssize_t blk_dropped_read(struct file *filp, char __user *buffer, size_t count, loff_t *ppos) { struct blk_trace *bt = filp->private_data; + size_t dropped = relay_stats(bt->rchan, RELAY_STATS_BUF_FULL); char buf[16]; - snprintf(buf, sizeof(buf), "%u\n", atomic_read(&bt->dropped)); + snprintf(buf, sizeof(buf), "%zu\n", dropped); return simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf)); } @@ -456,23 +613,6 @@ static const struct file_operations blk_msg_fops = { .llseek = noop_llseek, }; -/* - * Keep track of how many times we encountered a full subbuffer, to aid - * the user space app in telling how many lost events there were. - */ -static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf, - void *prev_subbuf, size_t prev_padding) -{ - struct blk_trace *bt; - - if (!relay_buf_full(buf)) - return 1; - - bt = buf->chan->private_data; - atomic_inc(&bt->dropped); - return 0; -} - static int blk_remove_buf_file_callback(struct dentry *dentry) { debugfs_remove(dentry); @@ -491,7 +631,6 @@ static struct dentry *blk_create_buf_file_callback(const char *filename, } static const struct rchan_callbacks blk_relay_callbacks = { - .subbuf_start = blk_subbuf_start_callback, .create_buf_file = blk_create_buf_file_callback, .remove_buf_file = blk_remove_buf_file_callback, }; @@ -511,9 +650,10 @@ static void blk_trace_setup_lba(struct blk_trace *bt, /* * Setup everything required to start tracing */ -static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, - struct block_device *bdev, - struct blk_user_trace_setup *buts) +static struct blk_trace *blk_trace_setup_prepare(struct request_queue *q, + char *name, dev_t dev, + u32 buf_size, u32 buf_nr, + struct block_device *bdev) { struct blk_trace *bt = NULL; struct dentry *dir = NULL; @@ -521,32 +661,19 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, lockdep_assert_held(&q->debugfs_mutex); - if (!buts->buf_size || !buts->buf_nr) - return -EINVAL; - - strncpy(buts->name, name, BLKTRACE_BDEV_SIZE); - buts->name[BLKTRACE_BDEV_SIZE - 1] = '\0'; - - /* - * some device names have larger paths - convert the slashes - * to underscores for this to work as expected - */ - strreplace(buts->name, '/', '_'); - /* * bdev can be NULL, as with scsi-generic, this is a helpful as * we can be. */ if (rcu_dereference_protected(q->blk_trace, lockdep_is_held(&q->debugfs_mutex))) { - pr_warn("Concurrent blktraces are not allowed on %s\n", - buts->name); - return -EBUSY; + pr_warn("Concurrent blktraces are not allowed on %s\n", name); + return ERR_PTR(-EBUSY); } bt = kzalloc(sizeof(*bt), GFP_KERNEL); if (!bt) - return -ENOMEM; + return ERR_PTR(-ENOMEM); ret = -ENOMEM; bt->sequence = alloc_percpu(unsigned long); @@ -566,7 +693,7 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, if (bdev && !bdev_is_partition(bdev)) dir = q->debugfs_dir; else - bt->dir = dir = debugfs_create_dir(buts->name, blk_debugfs_root); + bt->dir = dir = debugfs_create_dir(name, blk_debugfs_root); /* * As blktrace relies on debugfs for its interface the debugfs directory @@ -574,31 +701,52 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, * files or directories. */ if (IS_ERR_OR_NULL(dir)) { - pr_warn("debugfs_dir not present for %s so skipping\n", - buts->name); + pr_warn("debugfs_dir not present for %s so skipping\n", name); ret = -ENOENT; goto err; } bt->dev = dev; - atomic_set(&bt->dropped, 0); INIT_LIST_HEAD(&bt->running_list); ret = -EIO; debugfs_create_file("dropped", 0444, dir, bt, &blk_dropped_fops); debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops); - bt->rchan = relay_open("trace", dir, buts->buf_size, - buts->buf_nr, &blk_relay_callbacks, bt); + bt->rchan = relay_open("trace", dir, buf_size, buf_nr, + &blk_relay_callbacks, bt); if (!bt->rchan) goto err; + blk_trace_setup_lba(bt, bdev); + + return bt; + +err: + blk_trace_free(q, bt); + + return ERR_PTR(ret); +} + +static void blk_trace_setup_finalize(struct request_queue *q, + char *name, int version, + struct blk_trace *bt, + struct blk_user_trace_setup2 *buts) + +{ + strscpy_pad(buts->name, name, BLKTRACE_BDEV_SIZE2); + + /* + * some device names have larger paths - convert the slashes + * to underscores for this to work as expected + */ + strreplace(buts->name, '/', '_'); + + bt->version = version; bt->act_mask = buts->act_mask; if (!bt->act_mask) bt->act_mask = (u16) -1; - blk_trace_setup_lba(bt, bdev); - /* overwrite with user settings */ if (buts->start_lba) bt->start_lba = buts->start_lba; @@ -610,62 +758,100 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, rcu_assign_pointer(q->blk_trace, bt); get_probe_ref(); - - ret = 0; -err: - if (ret) - blk_trace_free(q, bt); - return ret; } -static int __blk_trace_setup(struct request_queue *q, char *name, dev_t dev, - struct block_device *bdev, char __user *arg) +int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, + char __user *arg) { + struct blk_user_trace_setup2 buts2; struct blk_user_trace_setup buts; + struct blk_trace *bt; int ret; ret = copy_from_user(&buts, arg, sizeof(buts)); if (ret) return -EFAULT; - ret = do_blk_trace_setup(q, name, dev, bdev, &buts); - if (ret) - return ret; + if (!buts.buf_size || !buts.buf_nr) + return -EINVAL; + + buts2 = (struct blk_user_trace_setup2) { + .act_mask = buts.act_mask, + .buf_size = buts.buf_size, + .buf_nr = buts.buf_nr, + .start_lba = buts.start_lba, + .end_lba = buts.end_lba, + .pid = buts.pid, + }; + + mutex_lock(&q->debugfs_mutex); + bt = blk_trace_setup_prepare(q, name, dev, buts.buf_size, buts.buf_nr, + bdev); + if (IS_ERR(bt)) { + mutex_unlock(&q->debugfs_mutex); + return PTR_ERR(bt); + } + blk_trace_setup_finalize(q, name, 1, bt, &buts2); + strcpy(buts.name, buts2.name); + mutex_unlock(&q->debugfs_mutex); if (copy_to_user(arg, &buts, sizeof(buts))) { - __blk_trace_remove(q); + blk_trace_remove(q); return -EFAULT; } return 0; } +EXPORT_SYMBOL_GPL(blk_trace_setup); -int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, - struct block_device *bdev, - char __user *arg) +static int blk_trace_setup2(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, char __user *arg) { - int ret; + struct blk_user_trace_setup2 buts2; + struct blk_trace *bt; + + if (copy_from_user(&buts2, arg, sizeof(buts2))) + return -EFAULT; + + if (!buts2.buf_size || !buts2.buf_nr) + return -EINVAL; + + if (buts2.flags != 0) + return -EINVAL; mutex_lock(&q->debugfs_mutex); - ret = __blk_trace_setup(q, name, dev, bdev, arg); + bt = blk_trace_setup_prepare(q, name, dev, buts2.buf_size, buts2.buf_nr, + bdev); + if (IS_ERR(bt)) { + mutex_unlock(&q->debugfs_mutex); + return PTR_ERR(bt); + } + blk_trace_setup_finalize(q, name, 2, bt, &buts2); mutex_unlock(&q->debugfs_mutex); - return ret; + if (copy_to_user(arg, &buts2, sizeof(buts2))) { + blk_trace_remove(q); + return -EFAULT; + } + return 0; } -EXPORT_SYMBOL_GPL(blk_trace_setup); #if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64) static int compat_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, struct block_device *bdev, char __user *arg) { - struct blk_user_trace_setup buts; + struct blk_user_trace_setup2 buts2; struct compat_blk_user_trace_setup cbuts; - int ret; + struct blk_trace *bt; if (copy_from_user(&cbuts, arg, sizeof(cbuts))) return -EFAULT; - buts = (struct blk_user_trace_setup) { + if (!cbuts.buf_size || !cbuts.buf_nr) + return -EINVAL; + + buts2 = (struct blk_user_trace_setup2) { .act_mask = cbuts.act_mask, .buf_size = cbuts.buf_size, .buf_nr = cbuts.buf_nr, @@ -674,12 +860,18 @@ static int compat_blk_trace_setup(struct request_queue *q, char *name, .pid = cbuts.pid, }; - ret = do_blk_trace_setup(q, name, dev, bdev, &buts); - if (ret) - return ret; + mutex_lock(&q->debugfs_mutex); + bt = blk_trace_setup_prepare(q, name, dev, buts2.buf_size, buts2.buf_nr, + bdev); + if (IS_ERR(bt)) { + mutex_unlock(&q->debugfs_mutex); + return PTR_ERR(bt); + } + blk_trace_setup_finalize(q, name, 1, bt, &buts2); + mutex_unlock(&q->debugfs_mutex); - if (copy_to_user(arg, &buts.name, ARRAY_SIZE(buts.name))) { - __blk_trace_remove(q); + if (copy_to_user(arg, &buts2.name, ARRAY_SIZE(buts2.name))) { + blk_trace_remove(q); return -EFAULT; } @@ -729,20 +921,18 @@ EXPORT_SYMBOL_GPL(blk_trace_startstop); **/ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) { - struct request_queue *q; + struct request_queue *q = bdev_get_queue(bdev); int ret, start = 0; char b[BDEVNAME_SIZE]; - q = bdev_get_queue(bdev); - if (!q) - return -ENXIO; - - mutex_lock(&q->debugfs_mutex); - switch (cmd) { + case BLKTRACESETUP2: + snprintf(b, sizeof(b), "%pg", bdev); + ret = blk_trace_setup2(q, b, bdev->bd_dev, bdev, arg); + break; case BLKTRACESETUP: snprintf(b, sizeof(b), "%pg", bdev); - ret = __blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); + ret = blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); break; #if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64) case BLKTRACESETUP32: @@ -754,17 +944,15 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) start = 1; fallthrough; case BLKTRACESTOP: - ret = __blk_trace_startstop(q, start); + ret = blk_trace_startstop(q, start); break; case BLKTRACETEARDOWN: - ret = __blk_trace_remove(q); + ret = blk_trace_remove(q); break; default: ret = -ENOTTY; break; } - - mutex_unlock(&q->debugfs_mutex); return ret; } @@ -829,7 +1017,7 @@ blk_trace_request_get_cgid(struct request *rq) * **/ static void blk_add_trace_rq(struct request *rq, blk_status_t error, - unsigned int nr_bytes, u32 what, u64 cgid) + unsigned int nr_bytes, u64 what, u64 cgid) { struct blk_trace *bt; @@ -881,6 +1069,22 @@ static void blk_add_trace_rq_complete(void *ignore, struct request *rq, blk_trace_request_get_cgid(rq)); } +static void blk_add_trace_zone_update_request(void *ignore, struct request *rq) +{ + struct blk_trace *bt; + + rcu_read_lock(); + bt = rcu_dereference(rq->q->blk_trace); + if (likely(!bt) || bt->version < 2) { + rcu_read_unlock(); + return; + } + rcu_read_unlock(); + + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_ZONE_APPEND, + blk_trace_request_get_cgid(rq)); +} + /** * blk_add_trace_bio - Add a trace for a bio oriented action * @q: queue the io is for @@ -893,7 +1097,7 @@ static void blk_add_trace_rq_complete(void *ignore, struct request *rq, * **/ static void blk_add_trace_bio(struct request_queue *q, struct bio *bio, - u32 what, int error) + u64 what, int error) { struct blk_trace *bt; @@ -910,11 +1114,6 @@ static void blk_add_trace_bio(struct request_queue *q, struct bio *bio, rcu_read_unlock(); } -static void blk_add_trace_bio_bounce(void *ignore, struct bio *bio) -{ - blk_add_trace_bio(bio->bi_bdev->bd_disk->queue, bio, BLK_TA_BOUNCE, 0); -} - static void blk_add_trace_bio_complete(void *ignore, struct request_queue *q, struct bio *bio) { @@ -964,7 +1163,7 @@ static void blk_add_trace_unplug(void *ignore, struct request_queue *q, bt = rcu_dereference(q->blk_trace); if (bt) { __be64 rpdu = cpu_to_be64(depth); - u32 what; + u64 what; if (explicit) what = BLK_TA_UNPLUG_IO; @@ -976,6 +1175,37 @@ static void blk_add_trace_unplug(void *ignore, struct request_queue *q, rcu_read_unlock(); } +static void blk_add_trace_zone_plug(void *ignore, struct request_queue *q, + unsigned int zno, sector_t sector, + unsigned int sectors) +{ + struct blk_trace *bt; + + rcu_read_lock(); + bt = rcu_dereference(q->blk_trace); + if (bt && bt->version >= 2) + __blk_add_trace(bt, sector, sectors << SECTOR_SHIFT, 0, + BLK_TA_ZONE_PLUG, 0, 0, NULL, 0); + rcu_read_unlock(); + + return; +} + +static void blk_add_trace_zone_unplug(void *ignore, struct request_queue *q, + unsigned int zno, sector_t sector, + unsigned int sectors) +{ + struct blk_trace *bt; + + rcu_read_lock(); + bt = rcu_dereference(q->blk_trace); + if (bt && bt->version >= 2) + __blk_add_trace(bt, sector, sectors << SECTOR_SHIFT, 0, + BLK_TA_ZONE_UNPLUG, 0, 0, NULL, 0); + rcu_read_unlock(); + return; +} + static void blk_add_trace_split(void *ignore, struct bio *bio, unsigned int pdu) { struct request_queue *q = bio->bi_bdev->bd_disk->queue; @@ -1106,8 +1336,6 @@ static void blk_register_tracepoints(void) WARN_ON(ret); ret = register_trace_block_rq_complete(blk_add_trace_rq_complete, NULL); WARN_ON(ret); - ret = register_trace_block_bio_bounce(blk_add_trace_bio_bounce, NULL); - WARN_ON(ret); ret = register_trace_block_bio_complete(blk_add_trace_bio_complete, NULL); WARN_ON(ret); ret = register_trace_block_bio_backmerge(blk_add_trace_bio_backmerge, NULL); @@ -1118,6 +1346,15 @@ static void blk_register_tracepoints(void) WARN_ON(ret); ret = register_trace_block_getrq(blk_add_trace_getrq, NULL); WARN_ON(ret); + ret = register_trace_blk_zone_append_update_request_bio( + blk_add_trace_zone_update_request, NULL); + WARN_ON(ret); + ret = register_trace_disk_zone_wplug_add_bio(blk_add_trace_zone_plug, + NULL); + WARN_ON(ret); + ret = register_trace_blk_zone_wplug_bio(blk_add_trace_zone_unplug, + NULL); + WARN_ON(ret); ret = register_trace_block_plug(blk_add_trace_plug, NULL); WARN_ON(ret); ret = register_trace_block_unplug(blk_add_trace_unplug, NULL); @@ -1137,12 +1374,15 @@ static void blk_unregister_tracepoints(void) unregister_trace_block_split(blk_add_trace_split, NULL); unregister_trace_block_unplug(blk_add_trace_unplug, NULL); unregister_trace_block_plug(blk_add_trace_plug, NULL); + unregister_trace_blk_zone_wplug_bio(blk_add_trace_zone_unplug, NULL); + unregister_trace_disk_zone_wplug_add_bio(blk_add_trace_zone_plug, NULL); + unregister_trace_blk_zone_append_update_request_bio( + blk_add_trace_zone_update_request, NULL); unregister_trace_block_getrq(blk_add_trace_getrq, NULL); unregister_trace_block_bio_queue(blk_add_trace_bio_queue, NULL); unregister_trace_block_bio_frontmerge(blk_add_trace_bio_frontmerge, NULL); unregister_trace_block_bio_backmerge(blk_add_trace_bio_backmerge, NULL); unregister_trace_block_bio_complete(blk_add_trace_bio_complete, NULL); - unregister_trace_block_bio_bounce(blk_add_trace_bio_bounce, NULL); unregister_trace_block_rq_complete(blk_add_trace_rq_complete, NULL); unregister_trace_block_rq_requeue(blk_add_trace_rq_requeue, NULL); unregister_trace_block_rq_merge(blk_add_trace_rq_merge, NULL); @@ -1156,7 +1396,7 @@ static void blk_unregister_tracepoints(void) * struct blk_io_tracer formatting routines */ -static void fill_rwbs(char *rwbs, const struct blk_io_trace *t) +static void fill_rwbs(char *rwbs, const struct blk_io_trace2 *t) { int i = 0; int tc = t->action >> BLK_TC_SHIFT; @@ -1171,7 +1411,10 @@ static void fill_rwbs(char *rwbs, const struct blk_io_trace *t) if (tc & BLK_TC_DISCARD) rwbs[i++] = 'D'; - else if (tc & BLK_TC_WRITE) + else if (tc & BLK_TC_WRITE_ZEROES) { + rwbs[i++] = 'W'; + rwbs[i++] = 'Z'; + } else if (tc & BLK_TC_WRITE) rwbs[i++] = 'W'; else if (t->bytes) rwbs[i++] = 'R'; @@ -1191,9 +1434,9 @@ out: } static inline -const struct blk_io_trace *te_blk_io_trace(const struct trace_entry *ent) +const struct blk_io_trace2 *te_blk_io_trace(const struct trace_entry *ent) { - return (const struct blk_io_trace *)ent; + return (const struct blk_io_trace2 *)ent; } static inline const void *pdu_start(const struct trace_entry *ent, bool has_cg) @@ -1252,7 +1495,7 @@ static void blk_log_action_classic(struct trace_iterator *iter, const char *act, unsigned long long ts = iter->ts; unsigned long nsec_rem = do_div(ts, NSEC_PER_SEC); unsigned secs = (unsigned long)ts; - const struct blk_io_trace *t = te_blk_io_trace(iter->ent); + const struct blk_io_trace2 *t = te_blk_io_trace(iter->ent); fill_rwbs(rwbs, t); @@ -1266,7 +1509,7 @@ static void blk_log_action(struct trace_iterator *iter, const char *act, bool has_cg) { char rwbs[RWBS_LEN]; - const struct blk_io_trace *t = te_blk_io_trace(iter->ent); + const struct blk_io_trace2 *t = te_blk_io_trace(iter->ent); fill_rwbs(rwbs, t); if (has_cg) { @@ -1479,7 +1722,6 @@ static const struct { [__BLK_TA_UNPLUG_TIMER] = {{ "UT", "unplug_timer" }, blk_log_unplug }, [__BLK_TA_INSERT] = {{ "I", "insert" }, blk_log_generic }, [__BLK_TA_SPLIT] = {{ "X", "split" }, blk_log_split }, - [__BLK_TA_BOUNCE] = {{ "B", "bounce" }, blk_log_generic }, [__BLK_TA_REMAP] = {{ "A", "remap" }, blk_log_remap }, }; @@ -1488,7 +1730,7 @@ static enum print_line_t print_one_line(struct trace_iterator *iter, { struct trace_array *tr = iter->tr; struct trace_seq *s = &iter->seq; - const struct blk_io_trace *t; + const struct blk_io_trace2 *t; u16 what; bool long_act; blk_log_action_t *log_action; @@ -1496,7 +1738,7 @@ static enum print_line_t print_one_line(struct trace_iterator *iter, t = te_blk_io_trace(iter->ent); what = (t->action & ((1 << BLK_TC_SHIFT) - 1)) & ~__BLK_TA_CGROUP; - long_act = !!(tr->trace_flags & TRACE_ITER_VERBOSE); + long_act = !!(tr->trace_flags & TRACE_ITER(VERBOSE)); log_action = classic ? &blk_log_action_classic : &blk_log_action; has_cg = t->action & __BLK_TA_CGROUP; @@ -1525,8 +1767,8 @@ static enum print_line_t blk_trace_event_print(struct trace_iterator *iter, static void blk_trace_synthesize_old_trace(struct trace_iterator *iter) { struct trace_seq *s = &iter->seq; - struct blk_io_trace *t = (struct blk_io_trace *)iter->ent; - const int offset = offsetof(struct blk_io_trace, sector); + struct blk_io_trace2 *t = (struct blk_io_trace2 *)iter->ent; + const int offset = offsetof(struct blk_io_trace2, sector); struct blk_io_trace old = { .magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION, .time = iter->ts, @@ -1561,9 +1803,9 @@ blk_tracer_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set) /* don't output context-info for blk_classic output */ if (bit == TRACE_BLK_OPT_CLASSIC) { if (set) - tr->trace_flags &= ~TRACE_ITER_CONTEXT_INFO; + tr->trace_flags &= ~TRACE_ITER(CONTEXT_INFO); else - tr->trace_flags |= TRACE_ITER_CONTEXT_INFO; + tr->trace_flags |= TRACE_ITER(CONTEXT_INFO); } return 0; } @@ -1603,6 +1845,10 @@ static int __init init_blk_tracer(void) return 1; } + BUILD_BUG_ON(__alignof__(struct blk_user_trace_setup2) % + __alignof__(long)); + BUILD_BUG_ON(__alignof__(struct blk_io_trace2) % __alignof__(long)); + return 0; } @@ -1711,6 +1957,7 @@ static const struct { { BLK_TC_DISCARD, "discard" }, { BLK_TC_DRV_DATA, "drv_data" }, { BLK_TC_FUA, "fua" }, + { BLK_TC_WRITE_ZEROES, "write-zeroes" }, }; static int blk_trace_str2mask(const char *str) @@ -1901,6 +2148,33 @@ void blk_fill_rwbs(char *rwbs, blk_opf_t opf) case REQ_OP_READ: rwbs[i++] = 'R'; break; + case REQ_OP_ZONE_APPEND: + rwbs[i++] = 'Z'; + rwbs[i++] = 'A'; + break; + case REQ_OP_ZONE_RESET: + case REQ_OP_ZONE_RESET_ALL: + rwbs[i++] = 'Z'; + rwbs[i++] = 'R'; + if ((opf & REQ_OP_MASK) == REQ_OP_ZONE_RESET_ALL) + rwbs[i++] = 'A'; + break; + case REQ_OP_ZONE_FINISH: + rwbs[i++] = 'Z'; + rwbs[i++] = 'F'; + break; + case REQ_OP_ZONE_OPEN: + rwbs[i++] = 'Z'; + rwbs[i++] = 'O'; + break; + case REQ_OP_ZONE_CLOSE: + rwbs[i++] = 'Z'; + rwbs[i++] = 'C'; + break; + case REQ_OP_WRITE_ZEROES: + rwbs[i++] = 'W'; + rwbs[i++] = 'Z'; + break; default: rwbs[i++] = 'N'; } @@ -1913,6 +2187,10 @@ void blk_fill_rwbs(char *rwbs, blk_opf_t opf) rwbs[i++] = 'S'; if (opf & REQ_META) rwbs[i++] = 'M'; + if (opf & REQ_ATOMIC) + rwbs[i++] = 'U'; + + WARN_ON_ONCE(i >= RWBS_LEN); rwbs[i] = '\0'; } diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index f47274de012b..d57727abaade 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -22,7 +22,7 @@ #include <linux/bsearch.h> #include <linux/sort.h> #include <linux/key.h> -#include <linux/verification.h> +#include <linux/namei.h> #include <net/bpf_sk_storage.h> @@ -40,6 +40,9 @@ #define bpf_event_rcu_dereference(p) \ rcu_dereference_protected(p, lockdep_is_held(&bpf_event_mutex)) +#define MAX_UPROBE_MULTI_CNT (1U << 20) +#define MAX_KPROBE_MULTI_CNT (1U << 20) + #ifdef CONFIG_MODULES struct bpf_trace_module { struct module *module; @@ -86,6 +89,9 @@ static int bpf_btf_printf_prepare(struct btf_ptr *ptr, u32 btf_ptr_size, static u64 bpf_kprobe_multi_cookie(struct bpf_run_ctx *ctx); static u64 bpf_kprobe_multi_entry_ip(struct bpf_run_ctx *ctx); +static u64 bpf_uprobe_multi_cookie(struct bpf_run_ctx *ctx); +static u64 bpf_uprobe_multi_entry_ip(struct bpf_run_ctx *ctx); + /** * trace_call_bpf - invoke BPF program * @call: tracepoint event @@ -113,6 +119,9 @@ unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx) * and don't send kprobe event into ring-buffer, * so return zero here */ + rcu_read_lock(); + bpf_prog_inc_misses_counters(rcu_dereference(call->prog_array)); + rcu_read_unlock(); ret = 0; goto out; } @@ -223,17 +232,6 @@ const struct bpf_func_proto bpf_probe_read_user_str_proto = { .arg3_type = ARG_ANYTHING, }; -static __always_inline int -bpf_probe_read_kernel_common(void *dst, u32 size, const void *unsafe_ptr) -{ - int ret; - - ret = copy_from_kernel_nofault(dst, unsafe_ptr, size); - if (unlikely(ret < 0)) - memset(dst, 0, size); - return ret; -} - BPF_CALL_3(bpf_probe_read_kernel, void *, dst, u32, size, const void *, unsafe_ptr) { @@ -358,19 +356,6 @@ static const struct bpf_func_proto bpf_probe_write_user_proto = { .arg3_type = ARG_CONST_SIZE, }; -static const struct bpf_func_proto *bpf_get_probe_write_proto(void) -{ - if (!capable(CAP_SYS_ADMIN)) - return NULL; - - pr_warn_ratelimited("%s[%d] is installing a program with bpf_probe_write_user helper that may corrupt user memory!", - current->comm, task_pid_nr(current)); - - return &bpf_probe_write_user_proto; -} - -static DEFINE_RAW_SPINLOCK(trace_printk_lock); - #define MAX_TRACE_PRINTK_VARARGS 3 #define BPF_TRACE_PRINTK_SIZE 1024 @@ -378,23 +363,22 @@ BPF_CALL_5(bpf_trace_printk, char *, fmt, u32, fmt_size, u64, arg1, u64, arg2, u64, arg3) { u64 args[MAX_TRACE_PRINTK_VARARGS] = { arg1, arg2, arg3 }; - u32 *bin_args; - static char buf[BPF_TRACE_PRINTK_SIZE]; - unsigned long flags; + struct bpf_bprintf_data data = { + .get_bin_args = true, + .get_buf = true, + }; int ret; - ret = bpf_bprintf_prepare(fmt, fmt_size, args, &bin_args, - MAX_TRACE_PRINTK_VARARGS); + ret = bpf_bprintf_prepare(fmt, fmt_size, args, + MAX_TRACE_PRINTK_VARARGS, &data); if (ret < 0) return ret; - raw_spin_lock_irqsave(&trace_printk_lock, flags); - ret = bstr_printf(buf, sizeof(buf), fmt, bin_args); + ret = bstr_printf(data.buf, MAX_BPRINTF_BUF, fmt, data.bin_args); - trace_bpf_trace_printk(buf); - raw_spin_unlock_irqrestore(&trace_printk_lock, flags); + trace_bpf_trace_printk(data.buf); - bpf_bprintf_cleanup(); + bpf_bprintf_cleanup(&data); return ret; } @@ -407,7 +391,7 @@ static const struct bpf_func_proto bpf_trace_printk_proto = { .arg2_type = ARG_CONST_SIZE, }; -static void __set_printk_clr_event(void) +static void __set_printk_clr_event(struct work_struct *work) { /* * This program might be calling bpf_trace_printk, @@ -420,37 +404,37 @@ static void __set_printk_clr_event(void) if (trace_set_clr_event("bpf_trace", "bpf_trace_printk", 1)) pr_warn_ratelimited("could not enable bpf_trace_printk events"); } +static DECLARE_WORK(set_printk_work, __set_printk_clr_event); const struct bpf_func_proto *bpf_get_trace_printk_proto(void) { - __set_printk_clr_event(); + schedule_work(&set_printk_work); return &bpf_trace_printk_proto; } -BPF_CALL_4(bpf_trace_vprintk, char *, fmt, u32, fmt_size, const void *, data, +BPF_CALL_4(bpf_trace_vprintk, char *, fmt, u32, fmt_size, const void *, args, u32, data_len) { - static char buf[BPF_TRACE_PRINTK_SIZE]; - unsigned long flags; + struct bpf_bprintf_data data = { + .get_bin_args = true, + .get_buf = true, + }; int ret, num_args; - u32 *bin_args; if (data_len & 7 || data_len > MAX_BPRINTF_VARARGS * 8 || - (data_len && !data)) + (data_len && !args)) return -EINVAL; num_args = data_len / 8; - ret = bpf_bprintf_prepare(fmt, fmt_size, data, &bin_args, num_args); + ret = bpf_bprintf_prepare(fmt, fmt_size, args, num_args, &data); if (ret < 0) return ret; - raw_spin_lock_irqsave(&trace_printk_lock, flags); - ret = bstr_printf(buf, sizeof(buf), fmt, bin_args); + ret = bstr_printf(data.buf, MAX_BPRINTF_BUF, fmt, data.bin_args); - trace_bpf_trace_printk(buf); - raw_spin_unlock_irqrestore(&trace_printk_lock, flags); + trace_bpf_trace_printk(data.buf); - bpf_bprintf_cleanup(); + bpf_bprintf_cleanup(&data); return ret; } @@ -467,28 +451,30 @@ static const struct bpf_func_proto bpf_trace_vprintk_proto = { const struct bpf_func_proto *bpf_get_trace_vprintk_proto(void) { - __set_printk_clr_event(); + schedule_work(&set_printk_work); return &bpf_trace_vprintk_proto; } BPF_CALL_5(bpf_seq_printf, struct seq_file *, m, char *, fmt, u32, fmt_size, - const void *, data, u32, data_len) + const void *, args, u32, data_len) { + struct bpf_bprintf_data data = { + .get_bin_args = true, + }; int err, num_args; - u32 *bin_args; if (data_len & 7 || data_len > MAX_BPRINTF_VARARGS * 8 || - (data_len && !data)) + (data_len && !args)) return -EINVAL; num_args = data_len / 8; - err = bpf_bprintf_prepare(fmt, fmt_size, data, &bin_args, num_args); + err = bpf_bprintf_prepare(fmt, fmt_size, args, num_args, &data); if (err < 0) return err; - seq_bprintf(m, fmt, bin_args); + seq_bprintf(m, fmt, data.bin_args); - bpf_bprintf_cleanup(); + bpf_bprintf_cleanup(&data); return seq_has_overflowed(m) ? -EOVERFLOW : 0; } @@ -585,7 +571,7 @@ BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) return value; } -static const struct bpf_func_proto bpf_perf_event_read_proto = { +const struct bpf_func_proto bpf_perf_event_read_proto = { .func = bpf_perf_event_read, .gpl_only = true, .ret_type = RET_INTEGER, @@ -620,9 +606,15 @@ static const struct bpf_func_proto bpf_perf_event_read_value_proto = { .arg4_type = ARG_CONST_SIZE, }; +const struct bpf_func_proto *bpf_get_perf_event_read_value_proto(void) +{ + return &bpf_perf_event_read_value_proto; +} + static __always_inline u64 __bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map, - u64 flags, struct perf_sample_data *sd) + u64 flags, struct perf_raw_record *raw, + struct perf_sample_data *sd) { struct bpf_array *array = container_of(map, struct bpf_array, map); unsigned int cpu = smp_processor_id(); @@ -647,6 +639,8 @@ __bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map, if (unlikely(event->oncpu != cpu)) return -EOPNOTSUPP; + perf_sample_save_raw_data(sd, event, raw); + return perf_event_output(event, sd, regs); } @@ -663,8 +657,7 @@ static DEFINE_PER_CPU(int, bpf_trace_nest_level); BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map, u64, flags, void *, data, u64, size) { - struct bpf_trace_sample_data *sds = this_cpu_ptr(&bpf_trace_sds); - int nest_level = this_cpu_inc_return(bpf_trace_nest_level); + struct bpf_trace_sample_data *sds; struct perf_raw_record raw = { .frag = { .size = size, @@ -672,7 +665,11 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map, }, }; struct perf_sample_data *sd; - int err; + int nest_level, err; + + preempt_disable(); + sds = this_cpu_ptr(&bpf_trace_sds); + nest_level = this_cpu_inc_return(bpf_trace_nest_level); if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(sds->sds))) { err = -EBUSY; @@ -687,13 +684,11 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map, } perf_sample_data_init(sd, 0, 0); - sd->raw = &raw; - sd->sample_flags |= PERF_SAMPLE_RAW; - - err = __bpf_perf_event_output(regs, map, flags, sd); + err = __bpf_perf_event_output(regs, map, flags, &raw, sd); out: this_cpu_dec(bpf_trace_nest_level); + preempt_enable(); return err; } @@ -718,7 +713,6 @@ static DEFINE_PER_CPU(struct bpf_trace_sample_data, bpf_misc_sds); u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { - int nest_level = this_cpu_inc_return(bpf_event_output_nest_level); struct perf_raw_frag frag = { .copy = ctx_copy, .size = ctx_size, @@ -735,8 +729,12 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, }; struct perf_sample_data *sd; struct pt_regs *regs; + int nest_level; u64 ret; + preempt_disable(); + nest_level = this_cpu_inc_return(bpf_event_output_nest_level); + if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(bpf_misc_sds.sds))) { ret = -EBUSY; goto out; @@ -746,12 +744,11 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, perf_fetch_caller_regs(regs); perf_sample_data_init(sd, 0, 0); - sd->raw = &raw; - sd->sample_flags |= PERF_SAMPLE_RAW; - ret = __bpf_perf_event_output(regs, map, flags, sd); + ret = __bpf_perf_event_output(regs, map, flags, &raw, sd); out: this_cpu_dec(bpf_event_output_nest_level); + preempt_enable(); return ret; } @@ -783,8 +780,7 @@ BPF_CALL_1(bpf_task_pt_regs, struct task_struct *, task) return (unsigned long) task_pt_regs(task); } -BTF_ID_LIST(bpf_task_pt_regs_ids) -BTF_ID(struct, pt_regs) +BTF_ID_LIST_SINGLE(bpf_task_pt_regs_ids, struct, pt_regs) const struct bpf_func_proto bpf_task_pt_regs_proto = { .func = bpf_task_pt_regs, @@ -795,34 +791,13 @@ const struct bpf_func_proto bpf_task_pt_regs_proto = { .ret_btf_id = &bpf_task_pt_regs_ids[0], }; -BPF_CALL_2(bpf_current_task_under_cgroup, struct bpf_map *, map, u32, idx) -{ - struct bpf_array *array = container_of(map, struct bpf_array, map); - struct cgroup *cgrp; - - if (unlikely(idx >= array->map.max_entries)) - return -E2BIG; - - cgrp = READ_ONCE(array->ptrs[idx]); - if (unlikely(!cgrp)) - return -EAGAIN; - - return task_under_cgroup_hierarchy(current, cgrp); -} - -static const struct bpf_func_proto bpf_current_task_under_cgroup_proto = { - .func = bpf_current_task_under_cgroup, - .gpl_only = false, - .ret_type = RET_INTEGER, - .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_ANYTHING, -}; - struct send_signal_irq_work { struct irq_work irq_work; struct task_struct *task; u32 sig; enum pid_type type; + bool has_siginfo; + struct kernel_siginfo info; }; static DEFINE_PER_CPU(struct send_signal_irq_work, send_signal_work); @@ -830,29 +805,49 @@ static DEFINE_PER_CPU(struct send_signal_irq_work, send_signal_work); static void do_bpf_send_signal(struct irq_work *entry) { struct send_signal_irq_work *work; + struct kernel_siginfo *siginfo; work = container_of(entry, struct send_signal_irq_work, irq_work); - group_send_sig_info(work->sig, SEND_SIG_PRIV, work->task, work->type); + siginfo = work->has_siginfo ? &work->info : SEND_SIG_PRIV; + + group_send_sig_info(work->sig, siginfo, work->task, work->type); + put_task_struct(work->task); } -static int bpf_send_signal_common(u32 sig, enum pid_type type) +static int bpf_send_signal_common(u32 sig, enum pid_type type, struct task_struct *task, u64 value) { struct send_signal_irq_work *work = NULL; + struct kernel_siginfo info; + struct kernel_siginfo *siginfo; + + if (!task) { + task = current; + siginfo = SEND_SIG_PRIV; + } else { + clear_siginfo(&info); + info.si_signo = sig; + info.si_errno = 0; + info.si_code = SI_KERNEL; + info.si_pid = 0; + info.si_uid = 0; + info.si_value.sival_ptr = (void *)(unsigned long)value; + siginfo = &info; + } /* Similar to bpf_probe_write_user, task needs to be * in a sound condition and kernel memory access be * permitted in order to send signal to the current * task. */ - if (unlikely(current->flags & (PF_KTHREAD | PF_EXITING))) + if (unlikely(task->flags & (PF_KTHREAD | PF_EXITING))) return -EPERM; if (unlikely(!nmi_uaccess_okay())) return -EPERM; /* Task should not be pid=1 to avoid kernel panic. */ - if (unlikely(is_global_init(current))) + if (unlikely(is_global_init(task))) return -EPERM; - if (irqs_disabled()) { + if (preempt_count() != 0 || irqs_disabled()) { /* Do an early check on signal validity. Otherwise, * the error is lost in deferred irq_work. */ @@ -867,22 +862,25 @@ static int bpf_send_signal_common(u32 sig, enum pid_type type) * to the irq_work. The current task may change when queued * irq works get executed. */ - work->task = current; + work->task = get_task_struct(task); + work->has_siginfo = siginfo == &info; + if (work->has_siginfo) + copy_siginfo(&work->info, &info); work->sig = sig; work->type = type; irq_work_queue(&work->irq_work); return 0; } - return group_send_sig_info(sig, SEND_SIG_PRIV, current, type); + return group_send_sig_info(sig, siginfo, task, type); } BPF_CALL_1(bpf_send_signal, u32, sig) { - return bpf_send_signal_common(sig, PIDTYPE_TGID); + return bpf_send_signal_common(sig, PIDTYPE_TGID, NULL, 0); } -static const struct bpf_func_proto bpf_send_signal_proto = { +const struct bpf_func_proto bpf_send_signal_proto = { .func = bpf_send_signal, .gpl_only = false, .ret_type = RET_INTEGER, @@ -891,25 +889,35 @@ static const struct bpf_func_proto bpf_send_signal_proto = { BPF_CALL_1(bpf_send_signal_thread, u32, sig) { - return bpf_send_signal_common(sig, PIDTYPE_PID); + return bpf_send_signal_common(sig, PIDTYPE_PID, NULL, 0); } -static const struct bpf_func_proto bpf_send_signal_thread_proto = { +const struct bpf_func_proto bpf_send_signal_thread_proto = { .func = bpf_send_signal_thread, .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_ANYTHING, }; -BPF_CALL_3(bpf_d_path, struct path *, path, char *, buf, u32, sz) +BPF_CALL_3(bpf_d_path, const struct path *, path, char *, buf, u32, sz) { + struct path copy; long len; char *p; if (!sz) return 0; - p = d_path(path, buf, sz); + /* + * The path pointer is verified as trusted and safe to use, + * but let's double check it's valid anyway to workaround + * potentially broken verifier. + */ + len = copy_from_kernel_nofault(©, path, sizeof(*path)); + if (len < 0) + return len; + + p = d_path(©, buf, sz); if (IS_ERR(p)) { len = PTR_ERR(p); } else { @@ -1034,25 +1042,27 @@ static const struct bpf_func_proto bpf_get_func_ip_proto_tracing = { .arg1_type = ARG_PTR_TO_CTX, }; -#ifdef CONFIG_X86_KERNEL_IBT -static unsigned long get_entry_ip(unsigned long fentry_ip) +static inline unsigned long get_entry_ip(unsigned long fentry_ip) { - u32 instr; - - /* Being extra safe in here in case entry ip is on the page-edge. */ - if (get_kernel_nofault(instr, (u32 *) fentry_ip - 1)) - return fentry_ip; - if (is_endbr(instr)) +#ifdef CONFIG_X86_KERNEL_IBT + if (is_endbr((void *)(fentry_ip - ENDBR_INSN_SIZE))) fentry_ip -= ENDBR_INSN_SIZE; +#endif return fentry_ip; } -#else -#define get_entry_ip(fentry_ip) fentry_ip -#endif BPF_CALL_1(bpf_get_func_ip_kprobe, struct pt_regs *, regs) { - struct kprobe *kp = kprobe_running(); + struct bpf_trace_run_ctx *run_ctx __maybe_unused; + struct kprobe *kp; + +#ifdef CONFIG_UPROBES + run_ctx = container_of(current->bpf_ctx, struct bpf_trace_run_ctx, run_ctx); + if (run_ctx->is_uprobe) + return ((struct uprobe_dispatch_data *)current->utask->vaddr)->bp_addr; +#endif + + kp = kprobe_running(); if (!kp || !(kp->flags & KPROBE_FLAG_ON_FUNC_ENTRY)) return 0; @@ -1091,6 +1101,30 @@ static const struct bpf_func_proto bpf_get_attach_cookie_proto_kmulti = { .arg1_type = ARG_PTR_TO_CTX, }; +BPF_CALL_1(bpf_get_func_ip_uprobe_multi, struct pt_regs *, regs) +{ + return bpf_uprobe_multi_entry_ip(current->bpf_ctx); +} + +static const struct bpf_func_proto bpf_get_func_ip_proto_uprobe_multi = { + .func = bpf_get_func_ip_uprobe_multi, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, +}; + +BPF_CALL_1(bpf_get_attach_cookie_uprobe_multi, struct pt_regs *, regs) +{ + return bpf_uprobe_multi_cookie(current->bpf_ctx); +} + +static const struct bpf_func_proto bpf_get_attach_cookie_proto_umulti = { + .func = bpf_get_attach_cookie_uprobe_multi, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, +}; + BPF_CALL_1(bpf_get_attach_cookie_trace, void *, ctx) { struct bpf_trace_run_ctx *run_ctx; @@ -1135,9 +1169,6 @@ static const struct bpf_func_proto bpf_get_attach_cookie_proto_tracing = { BPF_CALL_3(bpf_get_branch_snapshot, void *, buf, u32, size, u64, flags) { -#ifndef CONFIG_X86 - return -ENOENT; -#else static const u32 br_entry_size = sizeof(struct perf_branch_entry); u32 entry_cnt = size / br_entry_size; @@ -1150,10 +1181,9 @@ BPF_CALL_3(bpf_get_branch_snapshot, void *, buf, u32, size, u64, flags) return -ENOENT; return entry_cnt * br_entry_size; -#endif } -static const struct bpf_func_proto bpf_get_branch_snapshot_proto = { +const struct bpf_func_proto bpf_get_branch_snapshot_proto = { .func = bpf_get_branch_snapshot, .gpl_only = true, .ret_type = RET_INTEGER, @@ -1177,7 +1207,8 @@ static const struct bpf_func_proto bpf_get_func_arg_proto = { .ret_type = RET_INTEGER, .arg1_type = ARG_PTR_TO_CTX, .arg2_type = ARG_ANYTHING, - .arg3_type = ARG_PTR_TO_LONG, + .arg3_type = ARG_PTR_TO_FIXED_SIZE_MEM | MEM_UNINIT | MEM_WRITE | MEM_ALIGNED, + .arg3_size = sizeof(u64), }; BPF_CALL_2(get_func_ret, void *, ctx, u64 *, value) @@ -1193,7 +1224,8 @@ static const struct bpf_func_proto bpf_get_func_ret_proto = { .func = get_func_ret, .ret_type = RET_INTEGER, .arg1_type = ARG_PTR_TO_CTX, - .arg2_type = ARG_PTR_TO_LONG, + .arg2_type = ARG_PTR_TO_FIXED_SIZE_MEM | MEM_UNINIT | MEM_WRITE | MEM_ALIGNED, + .arg2_size = sizeof(u64), }; BPF_CALL_1(get_func_arg_cnt, void *, ctx) @@ -1208,318 +1240,64 @@ static const struct bpf_func_proto bpf_get_func_arg_cnt_proto = { .arg1_type = ARG_PTR_TO_CTX, }; -#ifdef CONFIG_KEYS -__diag_push(); -__diag_ignore_all("-Wmissing-prototypes", - "kfuncs which will be used in BPF programs"); - -/** - * bpf_lookup_user_key - lookup a key by its serial - * @serial: key handle serial number - * @flags: lookup-specific flags - * - * Search a key with a given *serial* and the provided *flags*. - * If found, increment the reference count of the key by one, and - * return it in the bpf_key structure. - * - * The bpf_key structure must be passed to bpf_key_put() when done - * with it, so that the key reference count is decremented and the - * bpf_key structure is freed. - * - * Permission checks are deferred to the time the key is used by - * one of the available key-specific kfuncs. - * - * Set *flags* with KEY_LOOKUP_CREATE, to attempt creating a requested - * special keyring (e.g. session keyring), if it doesn't yet exist. - * Set *flags* with KEY_LOOKUP_PARTIAL, to lookup a key without waiting - * for the key construction, and to retrieve uninstantiated keys (keys - * without data attached to them). - * - * Return: a bpf_key pointer with a valid key pointer if the key is found, a - * NULL pointer otherwise. - */ -struct bpf_key *bpf_lookup_user_key(u32 serial, u64 flags) +static const struct bpf_func_proto * +bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) { - key_ref_t key_ref; - struct bpf_key *bkey; - - if (flags & ~KEY_LOOKUP_ALL) - return NULL; - - /* - * Permission check is deferred until the key is used, as the - * intent of the caller is unknown here. - */ - key_ref = lookup_user_key(serial, flags, KEY_DEFER_PERM_CHECK); - if (IS_ERR(key_ref)) - return NULL; + const struct bpf_func_proto *func_proto; - bkey = kmalloc(sizeof(*bkey), GFP_KERNEL); - if (!bkey) { - key_put(key_ref_to_ptr(key_ref)); - return NULL; + switch (func_id) { + case BPF_FUNC_get_smp_processor_id: + return &bpf_get_smp_processor_id_proto; +#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE + case BPF_FUNC_probe_read: + return security_locked_down(LOCKDOWN_BPF_READ_KERNEL) < 0 ? + NULL : &bpf_probe_read_compat_proto; + case BPF_FUNC_probe_read_str: + return security_locked_down(LOCKDOWN_BPF_READ_KERNEL) < 0 ? + NULL : &bpf_probe_read_compat_str_proto; +#endif + case BPF_FUNC_get_func_ip: + return &bpf_get_func_ip_proto_tracing; + default: + break; } - bkey->key = key_ref_to_ptr(key_ref); - bkey->has_ref = true; - - return bkey; -} + func_proto = bpf_base_func_proto(func_id, prog); + if (func_proto) + return func_proto; -/** - * bpf_lookup_system_key - lookup a key by a system-defined ID - * @id: key ID - * - * Obtain a bpf_key structure with a key pointer set to the passed key ID. - * The key pointer is marked as invalid, to prevent bpf_key_put() from - * attempting to decrement the key reference count on that pointer. The key - * pointer set in such way is currently understood only by - * verify_pkcs7_signature(). - * - * Set *id* to one of the values defined in include/linux/verification.h: - * 0 for the primary keyring (immutable keyring of system keys); - * VERIFY_USE_SECONDARY_KEYRING for both the primary and secondary keyring - * (where keys can be added only if they are vouched for by existing keys - * in those keyrings); VERIFY_USE_PLATFORM_KEYRING for the platform - * keyring (primarily used by the integrity subsystem to verify a kexec'ed - * kerned image and, possibly, the initramfs signature). - * - * Return: a bpf_key pointer with an invalid key pointer set from the - * pre-determined ID on success, a NULL pointer otherwise - */ -struct bpf_key *bpf_lookup_system_key(u64 id) -{ - struct bpf_key *bkey; - - if (system_keyring_id_check(id) < 0) + if (!bpf_token_capable(prog->aux->token, CAP_SYS_ADMIN)) return NULL; - bkey = kmalloc(sizeof(*bkey), GFP_ATOMIC); - if (!bkey) + switch (func_id) { + case BPF_FUNC_probe_write_user: + return security_locked_down(LOCKDOWN_BPF_WRITE_USER) < 0 ? + NULL : &bpf_probe_write_user_proto; + default: return NULL; - - bkey->key = (struct key *)(unsigned long)id; - bkey->has_ref = false; - - return bkey; + } } -/** - * bpf_key_put - decrement key reference count if key is valid and free bpf_key - * @bkey: bpf_key structure - * - * Decrement the reference count of the key inside *bkey*, if the pointer - * is valid, and free *bkey*. - */ -void bpf_key_put(struct bpf_key *bkey) +static bool is_kprobe_multi(const struct bpf_prog *prog) { - if (bkey->has_ref) - key_put(bkey->key); - - kfree(bkey); + return prog->expected_attach_type == BPF_TRACE_KPROBE_MULTI || + prog->expected_attach_type == BPF_TRACE_KPROBE_SESSION; } -#ifdef CONFIG_SYSTEM_DATA_VERIFICATION -/** - * bpf_verify_pkcs7_signature - verify a PKCS#7 signature - * @data_ptr: data to verify - * @sig_ptr: signature of the data - * @trusted_keyring: keyring with keys trusted for signature verification - * - * Verify the PKCS#7 signature *sig_ptr* against the supplied *data_ptr* - * with keys in a keyring referenced by *trusted_keyring*. - * - * Return: 0 on success, a negative value on error. - */ -int bpf_verify_pkcs7_signature(struct bpf_dynptr_kern *data_ptr, - struct bpf_dynptr_kern *sig_ptr, - struct bpf_key *trusted_keyring) +static inline bool is_kprobe_session(const struct bpf_prog *prog) { - int ret; - - if (trusted_keyring->has_ref) { - /* - * Do the permission check deferred in bpf_lookup_user_key(). - * See bpf_lookup_user_key() for more details. - * - * A call to key_task_permission() here would be redundant, as - * it is already done by keyring_search() called by - * find_asymmetric_key(). - */ - ret = key_validate(trusted_keyring->key); - if (ret < 0) - return ret; - } - - return verify_pkcs7_signature(data_ptr->data, - bpf_dynptr_get_size(data_ptr), - sig_ptr->data, - bpf_dynptr_get_size(sig_ptr), - trusted_keyring->key, - VERIFYING_UNSPECIFIED_SIGNATURE, NULL, - NULL); + return prog->expected_attach_type == BPF_TRACE_KPROBE_SESSION; } -#endif /* CONFIG_SYSTEM_DATA_VERIFICATION */ - -__diag_pop(); -BTF_SET8_START(key_sig_kfunc_set) -BTF_ID_FLAGS(func, bpf_lookup_user_key, KF_ACQUIRE | KF_RET_NULL | KF_SLEEPABLE) -BTF_ID_FLAGS(func, bpf_lookup_system_key, KF_ACQUIRE | KF_RET_NULL) -BTF_ID_FLAGS(func, bpf_key_put, KF_RELEASE) -#ifdef CONFIG_SYSTEM_DATA_VERIFICATION -BTF_ID_FLAGS(func, bpf_verify_pkcs7_signature, KF_SLEEPABLE) -#endif -BTF_SET8_END(key_sig_kfunc_set) - -static const struct btf_kfunc_id_set bpf_key_sig_kfunc_set = { - .owner = THIS_MODULE, - .set = &key_sig_kfunc_set, -}; - -static int __init bpf_key_sig_kfuncs_init(void) +static inline bool is_uprobe_multi(const struct bpf_prog *prog) { - return register_btf_kfunc_id_set(BPF_PROG_TYPE_TRACING, - &bpf_key_sig_kfunc_set); + return prog->expected_attach_type == BPF_TRACE_UPROBE_MULTI || + prog->expected_attach_type == BPF_TRACE_UPROBE_SESSION; } -late_initcall(bpf_key_sig_kfuncs_init); -#endif /* CONFIG_KEYS */ - -static const struct bpf_func_proto * -bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) +static inline bool is_uprobe_session(const struct bpf_prog *prog) { - switch (func_id) { - case BPF_FUNC_map_lookup_elem: - return &bpf_map_lookup_elem_proto; - case BPF_FUNC_map_update_elem: - return &bpf_map_update_elem_proto; - case BPF_FUNC_map_delete_elem: - return &bpf_map_delete_elem_proto; - case BPF_FUNC_map_push_elem: - return &bpf_map_push_elem_proto; - case BPF_FUNC_map_pop_elem: - return &bpf_map_pop_elem_proto; - case BPF_FUNC_map_peek_elem: - return &bpf_map_peek_elem_proto; - case BPF_FUNC_map_lookup_percpu_elem: - return &bpf_map_lookup_percpu_elem_proto; - case BPF_FUNC_ktime_get_ns: - return &bpf_ktime_get_ns_proto; - case BPF_FUNC_ktime_get_boot_ns: - return &bpf_ktime_get_boot_ns_proto; - case BPF_FUNC_tail_call: - return &bpf_tail_call_proto; - case BPF_FUNC_get_current_pid_tgid: - return &bpf_get_current_pid_tgid_proto; - case BPF_FUNC_get_current_task: - return &bpf_get_current_task_proto; - case BPF_FUNC_get_current_task_btf: - return &bpf_get_current_task_btf_proto; - case BPF_FUNC_task_pt_regs: - return &bpf_task_pt_regs_proto; - case BPF_FUNC_get_current_uid_gid: - return &bpf_get_current_uid_gid_proto; - case BPF_FUNC_get_current_comm: - return &bpf_get_current_comm_proto; - case BPF_FUNC_trace_printk: - return bpf_get_trace_printk_proto(); - case BPF_FUNC_get_smp_processor_id: - return &bpf_get_smp_processor_id_proto; - case BPF_FUNC_get_numa_node_id: - return &bpf_get_numa_node_id_proto; - case BPF_FUNC_perf_event_read: - return &bpf_perf_event_read_proto; - case BPF_FUNC_current_task_under_cgroup: - return &bpf_current_task_under_cgroup_proto; - case BPF_FUNC_get_prandom_u32: - return &bpf_get_prandom_u32_proto; - case BPF_FUNC_probe_write_user: - return security_locked_down(LOCKDOWN_BPF_WRITE_USER) < 0 ? - NULL : bpf_get_probe_write_proto(); - case BPF_FUNC_probe_read_user: - return &bpf_probe_read_user_proto; - case BPF_FUNC_probe_read_kernel: - return security_locked_down(LOCKDOWN_BPF_READ_KERNEL) < 0 ? - NULL : &bpf_probe_read_kernel_proto; - case BPF_FUNC_probe_read_user_str: - return &bpf_probe_read_user_str_proto; - case BPF_FUNC_probe_read_kernel_str: - return security_locked_down(LOCKDOWN_BPF_READ_KERNEL) < 0 ? - NULL : &bpf_probe_read_kernel_str_proto; -#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE - case BPF_FUNC_probe_read: - return security_locked_down(LOCKDOWN_BPF_READ_KERNEL) < 0 ? - NULL : &bpf_probe_read_compat_proto; - case BPF_FUNC_probe_read_str: - return security_locked_down(LOCKDOWN_BPF_READ_KERNEL) < 0 ? - NULL : &bpf_probe_read_compat_str_proto; -#endif -#ifdef CONFIG_CGROUPS - case BPF_FUNC_get_current_cgroup_id: - return &bpf_get_current_cgroup_id_proto; - case BPF_FUNC_get_current_ancestor_cgroup_id: - return &bpf_get_current_ancestor_cgroup_id_proto; - case BPF_FUNC_cgrp_storage_get: - return &bpf_cgrp_storage_get_proto; - case BPF_FUNC_cgrp_storage_delete: - return &bpf_cgrp_storage_delete_proto; -#endif - case BPF_FUNC_send_signal: - return &bpf_send_signal_proto; - case BPF_FUNC_send_signal_thread: - return &bpf_send_signal_thread_proto; - case BPF_FUNC_perf_event_read_value: - return &bpf_perf_event_read_value_proto; - case BPF_FUNC_get_ns_current_pid_tgid: - return &bpf_get_ns_current_pid_tgid_proto; - case BPF_FUNC_ringbuf_output: - return &bpf_ringbuf_output_proto; - case BPF_FUNC_ringbuf_reserve: - return &bpf_ringbuf_reserve_proto; - case BPF_FUNC_ringbuf_submit: - return &bpf_ringbuf_submit_proto; - case BPF_FUNC_ringbuf_discard: - return &bpf_ringbuf_discard_proto; - case BPF_FUNC_ringbuf_query: - return &bpf_ringbuf_query_proto; - case BPF_FUNC_jiffies64: - return &bpf_jiffies64_proto; - case BPF_FUNC_get_task_stack: - return &bpf_get_task_stack_proto; - case BPF_FUNC_copy_from_user: - return &bpf_copy_from_user_proto; - case BPF_FUNC_copy_from_user_task: - return &bpf_copy_from_user_task_proto; - case BPF_FUNC_snprintf_btf: - return &bpf_snprintf_btf_proto; - case BPF_FUNC_per_cpu_ptr: - return &bpf_per_cpu_ptr_proto; - case BPF_FUNC_this_cpu_ptr: - return &bpf_this_cpu_ptr_proto; - case BPF_FUNC_task_storage_get: - if (bpf_prog_check_recur(prog)) - return &bpf_task_storage_get_recur_proto; - return &bpf_task_storage_get_proto; - case BPF_FUNC_task_storage_delete: - if (bpf_prog_check_recur(prog)) - return &bpf_task_storage_delete_recur_proto; - return &bpf_task_storage_delete_proto; - case BPF_FUNC_for_each_map_elem: - return &bpf_for_each_map_elem_proto; - case BPF_FUNC_snprintf: - return &bpf_snprintf_proto; - case BPF_FUNC_get_func_ip: - return &bpf_get_func_ip_proto_tracing; - case BPF_FUNC_get_branch_snapshot: - return &bpf_get_branch_snapshot_proto; - case BPF_FUNC_find_vma: - return &bpf_find_vma_proto; - case BPF_FUNC_trace_vprintk: - return bpf_get_trace_vprintk_proto(); - default: - return bpf_base_func_proto(func_id); - } + return prog->expected_attach_type == BPF_TRACE_UPROBE_SESSION; } static const struct bpf_func_proto * @@ -1531,19 +1309,23 @@ kprobe_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) case BPF_FUNC_get_stackid: return &bpf_get_stackid_proto; case BPF_FUNC_get_stack: - return &bpf_get_stack_proto; + return prog->sleepable ? &bpf_get_stack_sleepable_proto : &bpf_get_stack_proto; #ifdef CONFIG_BPF_KPROBE_OVERRIDE case BPF_FUNC_override_return: return &bpf_override_return_proto; #endif case BPF_FUNC_get_func_ip: - return prog->expected_attach_type == BPF_TRACE_KPROBE_MULTI ? - &bpf_get_func_ip_proto_kprobe_multi : - &bpf_get_func_ip_proto_kprobe; + if (is_kprobe_multi(prog)) + return &bpf_get_func_ip_proto_kprobe_multi; + if (is_uprobe_multi(prog)) + return &bpf_get_func_ip_proto_uprobe_multi; + return &bpf_get_func_ip_proto_kprobe; case BPF_FUNC_get_attach_cookie: - return prog->expected_attach_type == BPF_TRACE_KPROBE_MULTI ? - &bpf_get_attach_cookie_proto_kmulti : - &bpf_get_attach_cookie_proto_trace; + if (is_kprobe_multi(prog)) + return &bpf_get_attach_cookie_proto_kmulti; + if (is_uprobe_multi(prog)) + return &bpf_get_attach_cookie_proto_umulti; + return &bpf_get_attach_cookie_proto_trace; default: return bpf_tracing_func_proto(func_id, prog); } @@ -1556,8 +1338,6 @@ static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type { if (off < 0 || off >= sizeof(struct pt_regs)) return false; - if (type != BPF_READ) - return false; if (off % size != 0) return false; /* @@ -1567,6 +1347,9 @@ static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type if (off + size > sizeof(struct pt_regs)) return false; + if (type == BPF_WRITE) + prog->aux->kprobe_write_ctx = true; + return true; } @@ -1787,7 +1570,7 @@ static struct pt_regs *get_bpf_raw_tp_regs(void) struct bpf_raw_tp_regs *tp_regs = this_cpu_ptr(&bpf_raw_tp_regs); int nest_level = this_cpu_inc_return(bpf_raw_tp_nest_level); - if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(tp_regs->regs))) { + if (nest_level > ARRAY_SIZE(tp_regs->regs)) { this_cpu_dec(bpf_raw_tp_nest_level); return ERR_PTR(-EBUSY); } @@ -1893,6 +1676,8 @@ raw_tp_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) return &bpf_get_stackid_proto_raw_tp; case BPF_FUNC_get_stack: return &bpf_get_stack_proto_raw_tp; + case BPF_FUNC_get_attach_cookie: + return &bpf_get_attach_cookie_proto_tracing; default: return bpf_tracing_func_proto(func_id, prog); } @@ -1955,6 +1740,9 @@ tracing_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) case BPF_FUNC_get_func_arg_cnt: return bpf_prog_has_trampoline(prog) ? &bpf_get_func_arg_cnt_proto : NULL; case BPF_FUNC_get_attach_cookie: + if (prog->type == BPF_PROG_TYPE_TRACING && + prog->expected_attach_type == BPF_TRACE_RAW_TP) + return &bpf_get_attach_cookie_proto_tracing; return bpf_prog_has_trampoline(prog) ? &bpf_get_attach_cookie_proto_tracing : NULL; default: fn = raw_tp_prog_func_proto(func_id, prog); @@ -2163,6 +1951,7 @@ void perf_event_detach_bpf_prog(struct perf_event *event) { struct bpf_prog_array *old_array; struct bpf_prog_array *new_array; + struct bpf_prog *prog = NULL; int ret; mutex_lock(&bpf_event_mutex); @@ -2171,9 +1960,10 @@ void perf_event_detach_bpf_prog(struct perf_event *event) goto unlock; old_array = bpf_event_rcu_dereference(event->tp_event->prog_array); + if (!old_array) + goto put; + ret = bpf_prog_array_copy(old_array, event->prog, NULL, 0, &new_array); - if (ret == -ENOENT) - goto unlock; if (ret < 0) { bpf_prog_array_delete_safe(old_array, event->prog); } else { @@ -2181,11 +1971,23 @@ void perf_event_detach_bpf_prog(struct perf_event *event) bpf_prog_array_free_sleepable(old_array); } - bpf_prog_put(event->prog); +put: + prog = event->prog; event->prog = NULL; unlock: mutex_unlock(&bpf_event_mutex); + + if (prog) { + /* + * It could be that the bpf_prog is not sleepable (and will be freed + * via normal RCU), but is called from a point that supports sleepable + * programs and uses tasks-trace-RCU. + */ + synchronize_rcu_tasks_trace(); + + bpf_prog_put(prog); + } } int perf_event_query_prog_array(struct perf_event *event, void __user *info) @@ -2248,23 +2050,32 @@ void bpf_put_raw_tracepoint(struct bpf_raw_event_map *btp) { struct module *mod; - preempt_disable(); + guard(rcu)(); mod = __module_address((unsigned long)btp); module_put(mod); - preempt_enable(); } static __always_inline -void __bpf_trace_run(struct bpf_prog *prog, u64 *args) +void __bpf_trace_run(struct bpf_raw_tp_link *link, u64 *args) { + struct bpf_prog *prog = link->link.prog; + struct bpf_run_ctx *old_run_ctx; + struct bpf_trace_run_ctx run_ctx; + cant_sleep(); if (unlikely(this_cpu_inc_return(*(prog->active)) != 1)) { bpf_prog_inc_misses_counter(prog); goto out; } + + run_ctx.bpf_cookie = link->cookie; + old_run_ctx = bpf_set_run_ctx(&run_ctx.run_ctx); + rcu_read_lock(); (void) bpf_prog_run(prog, args); rcu_read_unlock(); + + bpf_reset_run_ctx(old_run_ctx); out: this_cpu_dec(*(prog->active)); } @@ -2293,12 +2104,12 @@ out: #define __SEQ_0_11 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 #define BPF_TRACE_DEFN_x(x) \ - void bpf_trace_run##x(struct bpf_prog *prog, \ + void bpf_trace_run##x(struct bpf_raw_tp_link *link, \ REPEAT(x, SARG, __DL_COM, __SEQ_0_11)) \ { \ u64 args[x]; \ REPEAT(x, COPY, __DL_SEM, __SEQ_0_11); \ - __bpf_trace_run(prog, args); \ + __bpf_trace_run(link, args); \ } \ EXPORT_SYMBOL_GPL(bpf_trace_run##x) BPF_TRACE_DEFN_x(1); @@ -2314,9 +2125,10 @@ BPF_TRACE_DEFN_x(10); BPF_TRACE_DEFN_x(11); BPF_TRACE_DEFN_x(12); -static int __bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_prog *prog) +int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_raw_tp_link *link) { struct tracepoint *tp = btp->tp; + struct bpf_prog *prog = link->link.prog; /* * check that program doesn't access arguments beyond what's @@ -2328,23 +2140,18 @@ static int __bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_prog * if (prog->aux->max_tp_access > btp->writable_size) return -EINVAL; - return tracepoint_probe_register_may_exist(tp, (void *)btp->bpf_func, - prog); -} - -int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_prog *prog) -{ - return __bpf_probe_register(btp, prog); + return tracepoint_probe_register_may_exist(tp, (void *)btp->bpf_func, link); } -int bpf_probe_unregister(struct bpf_raw_event_map *btp, struct bpf_prog *prog) +int bpf_probe_unregister(struct bpf_raw_event_map *btp, struct bpf_raw_tp_link *link) { - return tracepoint_probe_unregister(btp->tp, (void *)btp->bpf_func, prog); + return tracepoint_probe_unregister(btp->tp, (void *)btp->bpf_func, link); } int bpf_get_perf_event_info(const struct perf_event *event, u32 *prog_id, u32 *fd_type, const char **buf, - u64 *probe_offset, u64 *probe_addr) + u64 *probe_offset, u64 *probe_addr, + unsigned long *missed) { bool is_tracepoint, is_syscall_tp; struct bpf_prog *prog; @@ -2366,22 +2173,26 @@ int bpf_get_perf_event_info(const struct perf_event *event, u32 *prog_id, if (is_tracepoint || is_syscall_tp) { *buf = is_tracepoint ? event->tp_event->tp->name : event->tp_event->name; - *fd_type = BPF_FD_TYPE_TRACEPOINT; - *probe_offset = 0x0; - *probe_addr = 0x0; + /* We allow NULL pointer for tracepoint */ + if (fd_type) + *fd_type = BPF_FD_TYPE_TRACEPOINT; + if (probe_offset) + *probe_offset = 0x0; + if (probe_addr) + *probe_addr = 0x0; } else { /* kprobe/uprobe */ err = -EOPNOTSUPP; #ifdef CONFIG_KPROBE_EVENTS if (flags & TRACE_EVENT_FL_KPROBE) err = bpf_get_kprobe_info(event, fd_type, buf, - probe_offset, probe_addr, + probe_offset, probe_addr, missed, event->attr.type == PERF_TYPE_TRACEPOINT); #endif #ifdef CONFIG_UPROBE_EVENTS if (flags & TRACE_EVENT_FL_UPROBE) err = bpf_get_uprobe_info(event, fd_type, buf, - probe_offset, + probe_offset, probe_addr, event->attr.type == PERF_TYPE_TRACEPOINT); #endif } @@ -2457,6 +2268,12 @@ static int __init bpf_event_init(void) fs_initcall(bpf_event_init); #endif /* CONFIG_MODULES */ +struct bpf_session_run_ctx { + struct bpf_run_ctx run_ctx; + bool is_return; + void *data; +}; + #ifdef CONFIG_FPROBE struct bpf_kprobe_multi_link { struct bpf_link link; @@ -2469,7 +2286,7 @@ struct bpf_kprobe_multi_link { }; struct bpf_kprobe_multi_run_ctx { - struct bpf_run_ctx run_ctx; + struct bpf_session_run_ctx session_ctx; struct bpf_kprobe_multi_link *link; unsigned long entry_ip; }; @@ -2479,6 +2296,20 @@ struct user_syms { char *buf; }; +#ifndef CONFIG_HAVE_FTRACE_REGS_HAVING_PT_REGS +static DEFINE_PER_CPU(struct pt_regs, bpf_kprobe_multi_pt_regs); +#define bpf_kprobe_multi_pt_regs_ptr() this_cpu_ptr(&bpf_kprobe_multi_pt_regs) +#else +#define bpf_kprobe_multi_pt_regs_ptr() (NULL) +#endif + +static unsigned long ftrace_get_entry_ip(unsigned long fentry_ip) +{ + unsigned long ip = ftrace_get_symaddr(fentry_ip); + + return ip ? : fentry_ip; +} + static int copy_user_syms(struct user_syms *us, unsigned long __user *usyms, u32 cnt) { unsigned long __user usymbol; @@ -2555,9 +2386,87 @@ static void bpf_kprobe_multi_link_dealloc(struct bpf_link *link) kfree(kmulti_link); } +static int bpf_kprobe_multi_link_fill_link_info(const struct bpf_link *link, + struct bpf_link_info *info) +{ + u64 __user *ucookies = u64_to_user_ptr(info->kprobe_multi.cookies); + u64 __user *uaddrs = u64_to_user_ptr(info->kprobe_multi.addrs); + struct bpf_kprobe_multi_link *kmulti_link; + u32 ucount = info->kprobe_multi.count; + int err = 0, i; + + if (!uaddrs ^ !ucount) + return -EINVAL; + if (ucookies && !ucount) + return -EINVAL; + + kmulti_link = container_of(link, struct bpf_kprobe_multi_link, link); + info->kprobe_multi.count = kmulti_link->cnt; + info->kprobe_multi.flags = kmulti_link->link.flags; + info->kprobe_multi.missed = kmulti_link->fp.nmissed; + + if (!uaddrs) + return 0; + if (ucount < kmulti_link->cnt) + err = -ENOSPC; + else + ucount = kmulti_link->cnt; + + if (ucookies) { + if (kmulti_link->cookies) { + if (copy_to_user(ucookies, kmulti_link->cookies, ucount * sizeof(u64))) + return -EFAULT; + } else { + for (i = 0; i < ucount; i++) { + if (put_user(0, ucookies + i)) + return -EFAULT; + } + } + } + + if (kallsyms_show_value(current_cred())) { + if (copy_to_user(uaddrs, kmulti_link->addrs, ucount * sizeof(u64))) + return -EFAULT; + } else { + for (i = 0; i < ucount; i++) { + if (put_user(0, uaddrs + i)) + return -EFAULT; + } + } + return err; +} + +#ifdef CONFIG_PROC_FS +static void bpf_kprobe_multi_show_fdinfo(const struct bpf_link *link, + struct seq_file *seq) +{ + struct bpf_kprobe_multi_link *kmulti_link; + + kmulti_link = container_of(link, struct bpf_kprobe_multi_link, link); + + seq_printf(seq, + "kprobe_cnt:\t%u\n" + "missed:\t%lu\n", + kmulti_link->cnt, + kmulti_link->fp.nmissed); + + seq_printf(seq, "%s\t %s\n", "cookie", "func"); + for (int i = 0; i < kmulti_link->cnt; i++) { + seq_printf(seq, + "%llu\t %pS\n", + kmulti_link->cookies[i], + (void *)kmulti_link->addrs[i]); + } +} +#endif + static const struct bpf_link_ops bpf_kprobe_multi_link_lops = { .release = bpf_kprobe_multi_link_release, - .dealloc = bpf_kprobe_multi_link_dealloc, + .dealloc_deferred = bpf_kprobe_multi_link_dealloc, + .fill_link_info = bpf_kprobe_multi_link_fill_link_info, +#ifdef CONFIG_PROC_FS + .show_fdinfo = bpf_kprobe_multi_show_fdinfo, +#endif }; static void bpf_kprobe_multi_cookie_swap(void *a, void *b, int size, const void *priv) @@ -2597,7 +2506,8 @@ static u64 bpf_kprobe_multi_cookie(struct bpf_run_ctx *ctx) if (WARN_ON_ONCE(!ctx)) return 0; - run_ctx = container_of(current->bpf_ctx, struct bpf_kprobe_multi_run_ctx, run_ctx); + run_ctx = container_of(current->bpf_ctx, struct bpf_kprobe_multi_run_ctx, + session_ctx.run_ctx); link = run_ctx->link; if (!link->cookies) return 0; @@ -2614,47 +2524,77 @@ static u64 bpf_kprobe_multi_entry_ip(struct bpf_run_ctx *ctx) { struct bpf_kprobe_multi_run_ctx *run_ctx; - run_ctx = container_of(current->bpf_ctx, struct bpf_kprobe_multi_run_ctx, run_ctx); + run_ctx = container_of(current->bpf_ctx, struct bpf_kprobe_multi_run_ctx, + session_ctx.run_ctx); return run_ctx->entry_ip; } -static int +static __always_inline int kprobe_multi_link_prog_run(struct bpf_kprobe_multi_link *link, - unsigned long entry_ip, struct pt_regs *regs) + unsigned long entry_ip, struct ftrace_regs *fregs, + bool is_return, void *data) { struct bpf_kprobe_multi_run_ctx run_ctx = { + .session_ctx = { + .is_return = is_return, + .data = data, + }, .link = link, .entry_ip = entry_ip, }; struct bpf_run_ctx *old_run_ctx; + struct pt_regs *regs; int err; + /* + * graph tracer framework ensures we won't migrate, so there is no need + * to use migrate_disable for bpf_prog_run again. The check here just for + * __this_cpu_inc_return. + */ + cant_sleep(); + if (unlikely(__this_cpu_inc_return(bpf_prog_active) != 1)) { - err = 0; + bpf_prog_inc_misses_counter(link->link.prog); + err = 1; goto out; } - migrate_disable(); rcu_read_lock(); - old_run_ctx = bpf_set_run_ctx(&run_ctx.run_ctx); + regs = ftrace_partial_regs(fregs, bpf_kprobe_multi_pt_regs_ptr()); + old_run_ctx = bpf_set_run_ctx(&run_ctx.session_ctx.run_ctx); err = bpf_prog_run(link->link.prog, regs); bpf_reset_run_ctx(old_run_ctx); rcu_read_unlock(); - migrate_enable(); out: __this_cpu_dec(bpf_prog_active); return err; } -static void +static int kprobe_multi_link_handler(struct fprobe *fp, unsigned long fentry_ip, - struct pt_regs *regs) + unsigned long ret_ip, struct ftrace_regs *fregs, + void *data) { struct bpf_kprobe_multi_link *link; + int err; link = container_of(fp, struct bpf_kprobe_multi_link, fp); - kprobe_multi_link_prog_run(link, get_entry_ip(fentry_ip), regs); + err = kprobe_multi_link_prog_run(link, ftrace_get_entry_ip(fentry_ip), + fregs, false, data); + return is_kprobe_session(link->link.prog) ? err : 0; +} + +static void +kprobe_multi_link_exit_handler(struct fprobe *fp, unsigned long fentry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, + void *data) +{ + struct bpf_kprobe_multi_link *link; + + link = container_of(fp, struct bpf_kprobe_multi_link, fp); + kprobe_multi_link_prog_run(link, ftrace_get_entry_ip(fentry_ip), + fregs, true, data); } static int symbols_cmp_r(const void *a, const void *b, const void *priv) @@ -2687,69 +2627,91 @@ static void symbols_swap_r(void *a, void *b, int size, const void *priv) } } -struct module_addr_args { - unsigned long *addrs; - u32 addrs_cnt; +struct modules_array { struct module **mods; int mods_cnt; int mods_cap; }; -static int module_callback(void *data, const char *name, - struct module *mod, unsigned long addr) +static int add_module(struct modules_array *arr, struct module *mod) { - struct module_addr_args *args = data; struct module **mods; - /* We iterate all modules symbols and for each we: - * - search for it in provided addresses array - * - if found we check if we already have the module pointer stored - * (we iterate modules sequentially, so we can check just the last - * module pointer) - * - take module reference and store it - */ - if (!bsearch(&addr, args->addrs, args->addrs_cnt, sizeof(addr), - bpf_kprobe_multi_addrs_cmp)) - return 0; - - if (args->mods && args->mods[args->mods_cnt - 1] == mod) - return 0; - - if (args->mods_cnt == args->mods_cap) { - args->mods_cap = max(16, args->mods_cap * 3 / 2); - mods = krealloc_array(args->mods, args->mods_cap, sizeof(*mods), GFP_KERNEL); + if (arr->mods_cnt == arr->mods_cap) { + arr->mods_cap = max(16, arr->mods_cap * 3 / 2); + mods = krealloc_array(arr->mods, arr->mods_cap, sizeof(*mods), GFP_KERNEL); if (!mods) return -ENOMEM; - args->mods = mods; + arr->mods = mods; } - if (!try_module_get(mod)) - return -EINVAL; - - args->mods[args->mods_cnt] = mod; - args->mods_cnt++; + arr->mods[arr->mods_cnt] = mod; + arr->mods_cnt++; return 0; } +static bool has_module(struct modules_array *arr, struct module *mod) +{ + int i; + + for (i = arr->mods_cnt - 1; i >= 0; i--) { + if (arr->mods[i] == mod) + return true; + } + return false; +} + static int get_modules_for_addrs(struct module ***mods, unsigned long *addrs, u32 addrs_cnt) { - struct module_addr_args args = { - .addrs = addrs, - .addrs_cnt = addrs_cnt, - }; - int err; + struct modules_array arr = {}; + u32 i, err = 0; + + for (i = 0; i < addrs_cnt; i++) { + bool skip_add = false; + struct module *mod; + + scoped_guard(rcu) { + mod = __module_address(addrs[i]); + /* Either no module or it's already stored */ + if (!mod || has_module(&arr, mod)) { + skip_add = true; + break; /* scoped_guard */ + } + if (!try_module_get(mod)) + err = -EINVAL; + } + if (skip_add) + continue; + if (err) + break; + err = add_module(&arr, mod); + if (err) { + module_put(mod); + break; + } + } /* We return either err < 0 in case of error, ... */ - err = module_kallsyms_on_each_symbol(module_callback, &args); if (err) { - kprobe_multi_put_modules(args.mods, args.mods_cnt); - kfree(args.mods); + kprobe_multi_put_modules(arr.mods, arr.mods_cnt); + kfree(arr.mods); return err; } /* or number of modules found if everything is ok. */ - *mods = args.mods; - return args.mods_cnt; + *mods = arr.mods; + return arr.mods_cnt; +} + +static int addrs_check_error_injection_list(unsigned long *addrs, u32 cnt) +{ + u32 i; + + for (i = 0; i < cnt; i++) { + if (!within_error_injection_list(addrs[i])) + return -EINVAL; + } + return 0; } int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) @@ -2768,7 +2730,14 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr if (sizeof(u64) != sizeof(void *)) return -EOPNOTSUPP; - if (prog->expected_attach_type != BPF_TRACE_KPROBE_MULTI) + if (attr->link_create.flags) + return -EINVAL; + + if (!is_kprobe_multi(prog)) + return -EINVAL; + + /* Writing to context is not allowed for kprobes. */ + if (prog->aux->kprobe_write_ctx) return -EINVAL; flags = attr->link_create.kprobe_multi.flags; @@ -2783,6 +2752,8 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr cnt = attr->link_create.kprobe_multi.cnt; if (!cnt) return -EINVAL; + if (cnt > MAX_KPROBE_MULTI_CNT) + return -E2BIG; size = cnt * sizeof(*addrs); addrs = kvmalloc_array(cnt, sizeof(*addrs), GFP_KERNEL); @@ -2829,6 +2800,11 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr goto error; } + if (prog->kprobe_override && addrs_check_error_injection_list(addrs, cnt)) { + err = -EINVAL; + goto error; + } + link = kzalloc(sizeof(*link), GFP_KERNEL); if (!link) { err = -ENOMEM; @@ -2836,20 +2812,23 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr } bpf_link_init(&link->link, BPF_LINK_TYPE_KPROBE_MULTI, - &bpf_kprobe_multi_link_lops, prog); + &bpf_kprobe_multi_link_lops, prog, attr->link_create.attach_type); err = bpf_link_prime(&link->link, &link_primer); if (err) goto error; - if (flags & BPF_F_KPROBE_MULTI_RETURN) - link->fp.exit_handler = kprobe_multi_link_handler; - else + if (!(flags & BPF_F_KPROBE_MULTI_RETURN)) link->fp.entry_handler = kprobe_multi_link_handler; + if ((flags & BPF_F_KPROBE_MULTI_RETURN) || is_kprobe_session(prog)) + link->fp.exit_handler = kprobe_multi_link_exit_handler; + if (is_kprobe_session(prog)) + link->fp.entry_data_size = sizeof(u64); link->addrs = addrs; link->cookies = cookies; link->cnt = cnt; + link->link.flags = flags; if (cookies) { /* @@ -2862,13 +2841,6 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr bpf_kprobe_multi_cookie_cmp, bpf_kprobe_multi_cookie_swap, link); - } else { - /* - * We need to sort addrs array even if there are no cookies - * provided, to allow bsearch in get_modules_for_addrs. - */ - sort(addrs, cnt, sizeof(*addrs), - bpf_kprobe_multi_addrs_cmp, NULL); } err = get_modules_for_addrs(&link->mods, addrs, cnt); @@ -2907,3 +2879,697 @@ static u64 bpf_kprobe_multi_entry_ip(struct bpf_run_ctx *ctx) return 0; } #endif + +#ifdef CONFIG_UPROBES +struct bpf_uprobe_multi_link; + +struct bpf_uprobe { + struct bpf_uprobe_multi_link *link; + loff_t offset; + unsigned long ref_ctr_offset; + u64 cookie; + struct uprobe *uprobe; + struct uprobe_consumer consumer; + bool session; +}; + +struct bpf_uprobe_multi_link { + struct path path; + struct bpf_link link; + u32 cnt; + struct bpf_uprobe *uprobes; + struct task_struct *task; +}; + +struct bpf_uprobe_multi_run_ctx { + struct bpf_session_run_ctx session_ctx; + unsigned long entry_ip; + struct bpf_uprobe *uprobe; +}; + +static void bpf_uprobe_unregister(struct bpf_uprobe *uprobes, u32 cnt) +{ + u32 i; + + for (i = 0; i < cnt; i++) + uprobe_unregister_nosync(uprobes[i].uprobe, &uprobes[i].consumer); + + if (cnt) + uprobe_unregister_sync(); +} + +static void bpf_uprobe_multi_link_release(struct bpf_link *link) +{ + struct bpf_uprobe_multi_link *umulti_link; + + umulti_link = container_of(link, struct bpf_uprobe_multi_link, link); + bpf_uprobe_unregister(umulti_link->uprobes, umulti_link->cnt); + if (umulti_link->task) + put_task_struct(umulti_link->task); + path_put(&umulti_link->path); +} + +static void bpf_uprobe_multi_link_dealloc(struct bpf_link *link) +{ + struct bpf_uprobe_multi_link *umulti_link; + + umulti_link = container_of(link, struct bpf_uprobe_multi_link, link); + kvfree(umulti_link->uprobes); + kfree(umulti_link); +} + +static int bpf_uprobe_multi_link_fill_link_info(const struct bpf_link *link, + struct bpf_link_info *info) +{ + u64 __user *uref_ctr_offsets = u64_to_user_ptr(info->uprobe_multi.ref_ctr_offsets); + u64 __user *ucookies = u64_to_user_ptr(info->uprobe_multi.cookies); + u64 __user *uoffsets = u64_to_user_ptr(info->uprobe_multi.offsets); + u64 __user *upath = u64_to_user_ptr(info->uprobe_multi.path); + u32 upath_size = info->uprobe_multi.path_size; + struct bpf_uprobe_multi_link *umulti_link; + u32 ucount = info->uprobe_multi.count; + int err = 0, i; + char *p, *buf; + long left = 0; + + if (!upath ^ !upath_size) + return -EINVAL; + + if ((uoffsets || uref_ctr_offsets || ucookies) && !ucount) + return -EINVAL; + + umulti_link = container_of(link, struct bpf_uprobe_multi_link, link); + info->uprobe_multi.count = umulti_link->cnt; + info->uprobe_multi.flags = umulti_link->link.flags; + info->uprobe_multi.pid = umulti_link->task ? + task_pid_nr_ns(umulti_link->task, task_active_pid_ns(current)) : 0; + + upath_size = upath_size ? min_t(u32, upath_size, PATH_MAX) : PATH_MAX; + buf = kmalloc(upath_size, GFP_KERNEL); + if (!buf) + return -ENOMEM; + p = d_path(&umulti_link->path, buf, upath_size); + if (IS_ERR(p)) { + kfree(buf); + return PTR_ERR(p); + } + upath_size = buf + upath_size - p; + + if (upath) + left = copy_to_user(upath, p, upath_size); + kfree(buf); + if (left) + return -EFAULT; + info->uprobe_multi.path_size = upath_size; + + if (!uoffsets && !ucookies && !uref_ctr_offsets) + return 0; + + if (ucount < umulti_link->cnt) + err = -ENOSPC; + else + ucount = umulti_link->cnt; + + for (i = 0; i < ucount; i++) { + if (uoffsets && + put_user(umulti_link->uprobes[i].offset, uoffsets + i)) + return -EFAULT; + if (uref_ctr_offsets && + put_user(umulti_link->uprobes[i].ref_ctr_offset, uref_ctr_offsets + i)) + return -EFAULT; + if (ucookies && + put_user(umulti_link->uprobes[i].cookie, ucookies + i)) + return -EFAULT; + } + + return err; +} + +#ifdef CONFIG_PROC_FS +static void bpf_uprobe_multi_show_fdinfo(const struct bpf_link *link, + struct seq_file *seq) +{ + struct bpf_uprobe_multi_link *umulti_link; + char *p, *buf; + pid_t pid; + + umulti_link = container_of(link, struct bpf_uprobe_multi_link, link); + + buf = kmalloc(PATH_MAX, GFP_KERNEL); + if (!buf) + return; + + p = d_path(&umulti_link->path, buf, PATH_MAX); + if (IS_ERR(p)) { + kfree(buf); + return; + } + + pid = umulti_link->task ? + task_pid_nr_ns(umulti_link->task, task_active_pid_ns(current)) : 0; + seq_printf(seq, + "uprobe_cnt:\t%u\n" + "pid:\t%u\n" + "path:\t%s\n", + umulti_link->cnt, pid, p); + + seq_printf(seq, "%s\t %s\t %s\n", "cookie", "offset", "ref_ctr_offset"); + for (int i = 0; i < umulti_link->cnt; i++) { + seq_printf(seq, + "%llu\t %#llx\t %#lx\n", + umulti_link->uprobes[i].cookie, + umulti_link->uprobes[i].offset, + umulti_link->uprobes[i].ref_ctr_offset); + } + + kfree(buf); +} +#endif + +static const struct bpf_link_ops bpf_uprobe_multi_link_lops = { + .release = bpf_uprobe_multi_link_release, + .dealloc_deferred = bpf_uprobe_multi_link_dealloc, + .fill_link_info = bpf_uprobe_multi_link_fill_link_info, +#ifdef CONFIG_PROC_FS + .show_fdinfo = bpf_uprobe_multi_show_fdinfo, +#endif +}; + +static int uprobe_prog_run(struct bpf_uprobe *uprobe, + unsigned long entry_ip, + struct pt_regs *regs, + bool is_return, void *data) +{ + struct bpf_uprobe_multi_link *link = uprobe->link; + struct bpf_uprobe_multi_run_ctx run_ctx = { + .session_ctx = { + .is_return = is_return, + .data = data, + }, + .entry_ip = entry_ip, + .uprobe = uprobe, + }; + struct bpf_prog *prog = link->link.prog; + bool sleepable = prog->sleepable; + struct bpf_run_ctx *old_run_ctx; + int err; + + if (link->task && !same_thread_group(current, link->task)) + return 0; + + if (sleepable) + rcu_read_lock_trace(); + else + rcu_read_lock(); + + migrate_disable(); + + old_run_ctx = bpf_set_run_ctx(&run_ctx.session_ctx.run_ctx); + err = bpf_prog_run(link->link.prog, regs); + bpf_reset_run_ctx(old_run_ctx); + + migrate_enable(); + + if (sleepable) + rcu_read_unlock_trace(); + else + rcu_read_unlock(); + return err; +} + +static bool +uprobe_multi_link_filter(struct uprobe_consumer *con, struct mm_struct *mm) +{ + struct bpf_uprobe *uprobe; + + uprobe = container_of(con, struct bpf_uprobe, consumer); + return uprobe->link->task->mm == mm; +} + +static int +uprobe_multi_link_handler(struct uprobe_consumer *con, struct pt_regs *regs, + __u64 *data) +{ + struct bpf_uprobe *uprobe; + int ret; + + uprobe = container_of(con, struct bpf_uprobe, consumer); + ret = uprobe_prog_run(uprobe, instruction_pointer(regs), regs, false, data); + if (uprobe->session) + return ret ? UPROBE_HANDLER_IGNORE : 0; + return 0; +} + +static int +uprobe_multi_link_ret_handler(struct uprobe_consumer *con, unsigned long func, struct pt_regs *regs, + __u64 *data) +{ + struct bpf_uprobe *uprobe; + + uprobe = container_of(con, struct bpf_uprobe, consumer); + uprobe_prog_run(uprobe, func, regs, true, data); + return 0; +} + +static u64 bpf_uprobe_multi_entry_ip(struct bpf_run_ctx *ctx) +{ + struct bpf_uprobe_multi_run_ctx *run_ctx; + + run_ctx = container_of(current->bpf_ctx, struct bpf_uprobe_multi_run_ctx, + session_ctx.run_ctx); + return run_ctx->entry_ip; +} + +static u64 bpf_uprobe_multi_cookie(struct bpf_run_ctx *ctx) +{ + struct bpf_uprobe_multi_run_ctx *run_ctx; + + run_ctx = container_of(current->bpf_ctx, struct bpf_uprobe_multi_run_ctx, + session_ctx.run_ctx); + return run_ctx->uprobe->cookie; +} + +int bpf_uprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) +{ + struct bpf_uprobe_multi_link *link = NULL; + unsigned long __user *uref_ctr_offsets; + struct bpf_link_primer link_primer; + struct bpf_uprobe *uprobes = NULL; + struct task_struct *task = NULL; + unsigned long __user *uoffsets; + u64 __user *ucookies; + void __user *upath; + u32 flags, cnt, i; + struct path path; + char *name; + pid_t pid; + int err; + + /* no support for 32bit archs yet */ + if (sizeof(u64) != sizeof(void *)) + return -EOPNOTSUPP; + + if (attr->link_create.flags) + return -EINVAL; + + if (!is_uprobe_multi(prog)) + return -EINVAL; + + flags = attr->link_create.uprobe_multi.flags; + if (flags & ~BPF_F_UPROBE_MULTI_RETURN) + return -EINVAL; + + /* + * path, offsets and cnt are mandatory, + * ref_ctr_offsets and cookies are optional + */ + upath = u64_to_user_ptr(attr->link_create.uprobe_multi.path); + uoffsets = u64_to_user_ptr(attr->link_create.uprobe_multi.offsets); + cnt = attr->link_create.uprobe_multi.cnt; + pid = attr->link_create.uprobe_multi.pid; + + if (!upath || !uoffsets || !cnt || pid < 0) + return -EINVAL; + if (cnt > MAX_UPROBE_MULTI_CNT) + return -E2BIG; + + uref_ctr_offsets = u64_to_user_ptr(attr->link_create.uprobe_multi.ref_ctr_offsets); + ucookies = u64_to_user_ptr(attr->link_create.uprobe_multi.cookies); + + name = strndup_user(upath, PATH_MAX); + if (IS_ERR(name)) { + err = PTR_ERR(name); + return err; + } + + err = kern_path(name, LOOKUP_FOLLOW, &path); + kfree(name); + if (err) + return err; + + if (!d_is_reg(path.dentry)) { + err = -EBADF; + goto error_path_put; + } + + if (pid) { + rcu_read_lock(); + task = get_pid_task(find_vpid(pid), PIDTYPE_TGID); + rcu_read_unlock(); + if (!task) { + err = -ESRCH; + goto error_path_put; + } + } + + err = -ENOMEM; + + link = kzalloc(sizeof(*link), GFP_KERNEL); + uprobes = kvcalloc(cnt, sizeof(*uprobes), GFP_KERNEL); + + if (!uprobes || !link) + goto error_free; + + for (i = 0; i < cnt; i++) { + if (__get_user(uprobes[i].offset, uoffsets + i)) { + err = -EFAULT; + goto error_free; + } + if (uprobes[i].offset < 0) { + err = -EINVAL; + goto error_free; + } + if (uref_ctr_offsets && __get_user(uprobes[i].ref_ctr_offset, uref_ctr_offsets + i)) { + err = -EFAULT; + goto error_free; + } + if (ucookies && __get_user(uprobes[i].cookie, ucookies + i)) { + err = -EFAULT; + goto error_free; + } + + uprobes[i].link = link; + + if (!(flags & BPF_F_UPROBE_MULTI_RETURN)) + uprobes[i].consumer.handler = uprobe_multi_link_handler; + if (flags & BPF_F_UPROBE_MULTI_RETURN || is_uprobe_session(prog)) + uprobes[i].consumer.ret_handler = uprobe_multi_link_ret_handler; + if (is_uprobe_session(prog)) + uprobes[i].session = true; + if (pid) + uprobes[i].consumer.filter = uprobe_multi_link_filter; + } + + link->cnt = cnt; + link->uprobes = uprobes; + link->path = path; + link->task = task; + link->link.flags = flags; + + bpf_link_init(&link->link, BPF_LINK_TYPE_UPROBE_MULTI, + &bpf_uprobe_multi_link_lops, prog, attr->link_create.attach_type); + + for (i = 0; i < cnt; i++) { + uprobes[i].uprobe = uprobe_register(d_real_inode(link->path.dentry), + uprobes[i].offset, + uprobes[i].ref_ctr_offset, + &uprobes[i].consumer); + if (IS_ERR(uprobes[i].uprobe)) { + err = PTR_ERR(uprobes[i].uprobe); + link->cnt = i; + goto error_unregister; + } + } + + err = bpf_link_prime(&link->link, &link_primer); + if (err) + goto error_unregister; + + return bpf_link_settle(&link_primer); + +error_unregister: + bpf_uprobe_unregister(uprobes, link->cnt); + +error_free: + kvfree(uprobes); + kfree(link); + if (task) + put_task_struct(task); +error_path_put: + path_put(&path); + return err; +} +#else /* !CONFIG_UPROBES */ +int bpf_uprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) +{ + return -EOPNOTSUPP; +} +static u64 bpf_uprobe_multi_cookie(struct bpf_run_ctx *ctx) +{ + return 0; +} +static u64 bpf_uprobe_multi_entry_ip(struct bpf_run_ctx *ctx) +{ + return 0; +} +#endif /* CONFIG_UPROBES */ + +__bpf_kfunc_start_defs(); + +__bpf_kfunc bool bpf_session_is_return(void) +{ + struct bpf_session_run_ctx *session_ctx; + + session_ctx = container_of(current->bpf_ctx, struct bpf_session_run_ctx, run_ctx); + return session_ctx->is_return; +} + +__bpf_kfunc __u64 *bpf_session_cookie(void) +{ + struct bpf_session_run_ctx *session_ctx; + + session_ctx = container_of(current->bpf_ctx, struct bpf_session_run_ctx, run_ctx); + return session_ctx->data; +} + +__bpf_kfunc_end_defs(); + +BTF_KFUNCS_START(kprobe_multi_kfunc_set_ids) +BTF_ID_FLAGS(func, bpf_session_is_return) +BTF_ID_FLAGS(func, bpf_session_cookie) +BTF_KFUNCS_END(kprobe_multi_kfunc_set_ids) + +static int bpf_kprobe_multi_filter(const struct bpf_prog *prog, u32 kfunc_id) +{ + if (!btf_id_set8_contains(&kprobe_multi_kfunc_set_ids, kfunc_id)) + return 0; + + if (!is_kprobe_session(prog) && !is_uprobe_session(prog)) + return -EACCES; + + return 0; +} + +static const struct btf_kfunc_id_set bpf_kprobe_multi_kfunc_set = { + .owner = THIS_MODULE, + .set = &kprobe_multi_kfunc_set_ids, + .filter = bpf_kprobe_multi_filter, +}; + +static int __init bpf_kprobe_multi_kfuncs_init(void) +{ + return register_btf_kfunc_id_set(BPF_PROG_TYPE_KPROBE, &bpf_kprobe_multi_kfunc_set); +} + +late_initcall(bpf_kprobe_multi_kfuncs_init); + +typedef int (*copy_fn_t)(void *dst, const void *src, u32 size, struct task_struct *tsk); + +/* + * The __always_inline is to make sure the compiler doesn't + * generate indirect calls into callbacks, which is expensive, + * on some kernel configurations. This allows compiler to put + * direct calls into all the specific callback implementations + * (copy_user_data_sleepable, copy_user_data_nofault, and so on) + */ +static __always_inline int __bpf_dynptr_copy_str(struct bpf_dynptr *dptr, u64 doff, u64 size, + const void *unsafe_src, + copy_fn_t str_copy_fn, + struct task_struct *tsk) +{ + struct bpf_dynptr_kern *dst; + u64 chunk_sz, off; + void *dst_slice; + int cnt, err; + char buf[256]; + + dst_slice = bpf_dynptr_slice_rdwr(dptr, doff, NULL, size); + if (likely(dst_slice)) + return str_copy_fn(dst_slice, unsafe_src, size, tsk); + + dst = (struct bpf_dynptr_kern *)dptr; + if (bpf_dynptr_check_off_len(dst, doff, size)) + return -E2BIG; + + for (off = 0; off < size; off += chunk_sz - 1) { + chunk_sz = min_t(u64, sizeof(buf), size - off); + /* Expect str_copy_fn to return count of copied bytes, including + * zero terminator. Next iteration increment off by chunk_sz - 1 to + * overwrite NUL. + */ + cnt = str_copy_fn(buf, unsafe_src + off, chunk_sz, tsk); + if (cnt < 0) + return cnt; + err = __bpf_dynptr_write(dst, doff + off, buf, cnt, 0); + if (err) + return err; + if (cnt < chunk_sz || chunk_sz == 1) /* we are done */ + return off + cnt; + } + return off; +} + +static __always_inline int __bpf_dynptr_copy(const struct bpf_dynptr *dptr, u64 doff, + u64 size, const void *unsafe_src, + copy_fn_t copy_fn, struct task_struct *tsk) +{ + struct bpf_dynptr_kern *dst; + void *dst_slice; + char buf[256]; + u64 off, chunk_sz; + int err; + + dst_slice = bpf_dynptr_slice_rdwr(dptr, doff, NULL, size); + if (likely(dst_slice)) + return copy_fn(dst_slice, unsafe_src, size, tsk); + + dst = (struct bpf_dynptr_kern *)dptr; + if (bpf_dynptr_check_off_len(dst, doff, size)) + return -E2BIG; + + for (off = 0; off < size; off += chunk_sz) { + chunk_sz = min_t(u64, sizeof(buf), size - off); + err = copy_fn(buf, unsafe_src + off, chunk_sz, tsk); + if (err) + return err; + err = __bpf_dynptr_write(dst, doff + off, buf, chunk_sz, 0); + if (err) + return err; + } + return 0; +} + +static __always_inline int copy_user_data_nofault(void *dst, const void *unsafe_src, + u32 size, struct task_struct *tsk) +{ + return copy_from_user_nofault(dst, (const void __user *)unsafe_src, size); +} + +static __always_inline int copy_user_data_sleepable(void *dst, const void *unsafe_src, + u32 size, struct task_struct *tsk) +{ + int ret; + + if (!tsk) { /* Read from the current task */ + ret = copy_from_user(dst, (const void __user *)unsafe_src, size); + if (ret) + return -EFAULT; + return 0; + } + + ret = access_process_vm(tsk, (unsigned long)unsafe_src, dst, size, 0); + if (ret != size) + return -EFAULT; + return 0; +} + +static __always_inline int copy_kernel_data_nofault(void *dst, const void *unsafe_src, + u32 size, struct task_struct *tsk) +{ + return copy_from_kernel_nofault(dst, unsafe_src, size); +} + +static __always_inline int copy_user_str_nofault(void *dst, const void *unsafe_src, + u32 size, struct task_struct *tsk) +{ + return strncpy_from_user_nofault(dst, (const void __user *)unsafe_src, size); +} + +static __always_inline int copy_user_str_sleepable(void *dst, const void *unsafe_src, + u32 size, struct task_struct *tsk) +{ + int ret; + + if (unlikely(size == 0)) + return 0; + + if (tsk) { + ret = copy_remote_vm_str(tsk, (unsigned long)unsafe_src, dst, size, 0); + } else { + ret = strncpy_from_user(dst, (const void __user *)unsafe_src, size - 1); + /* strncpy_from_user does not guarantee NUL termination */ + if (ret >= 0) + ((char *)dst)[ret] = '\0'; + } + + if (ret < 0) + return ret; + return ret + 1; +} + +static __always_inline int copy_kernel_str_nofault(void *dst, const void *unsafe_src, + u32 size, struct task_struct *tsk) +{ + return strncpy_from_kernel_nofault(dst, unsafe_src, size); +} + +__bpf_kfunc_start_defs(); + +__bpf_kfunc int bpf_send_signal_task(struct task_struct *task, int sig, enum pid_type type, + u64 value) +{ + if (type != PIDTYPE_PID && type != PIDTYPE_TGID) + return -EINVAL; + + return bpf_send_signal_common(sig, type, task, value); +} + +__bpf_kfunc int bpf_probe_read_user_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void __user *unsafe_ptr__ign) +{ + return __bpf_dynptr_copy(dptr, off, size, (const void *)unsafe_ptr__ign, + copy_user_data_nofault, NULL); +} + +__bpf_kfunc int bpf_probe_read_kernel_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void *unsafe_ptr__ign) +{ + return __bpf_dynptr_copy(dptr, off, size, unsafe_ptr__ign, + copy_kernel_data_nofault, NULL); +} + +__bpf_kfunc int bpf_probe_read_user_str_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void __user *unsafe_ptr__ign) +{ + return __bpf_dynptr_copy_str(dptr, off, size, (const void *)unsafe_ptr__ign, + copy_user_str_nofault, NULL); +} + +__bpf_kfunc int bpf_probe_read_kernel_str_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void *unsafe_ptr__ign) +{ + return __bpf_dynptr_copy_str(dptr, off, size, unsafe_ptr__ign, + copy_kernel_str_nofault, NULL); +} + +__bpf_kfunc int bpf_copy_from_user_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void __user *unsafe_ptr__ign) +{ + return __bpf_dynptr_copy(dptr, off, size, (const void *)unsafe_ptr__ign, + copy_user_data_sleepable, NULL); +} + +__bpf_kfunc int bpf_copy_from_user_str_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void __user *unsafe_ptr__ign) +{ + return __bpf_dynptr_copy_str(dptr, off, size, (const void *)unsafe_ptr__ign, + copy_user_str_sleepable, NULL); +} + +__bpf_kfunc int bpf_copy_from_user_task_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void __user *unsafe_ptr__ign, + struct task_struct *tsk) +{ + return __bpf_dynptr_copy(dptr, off, size, (const void *)unsafe_ptr__ign, + copy_user_data_sleepable, tsk); +} + +__bpf_kfunc int bpf_copy_from_user_task_str_dynptr(struct bpf_dynptr *dptr, u64 off, + u64 size, const void __user *unsafe_ptr__ign, + struct task_struct *tsk) +{ + return __bpf_dynptr_copy_str(dptr, off, size, (const void *)unsafe_ptr__ign, + copy_user_str_sleepable, tsk); +} + +__bpf_kfunc_end_defs(); diff --git a/kernel/trace/bpf_trace.h b/kernel/trace/bpf_trace.h index 9acbc11ac7bb..c4075b56becc 100644 --- a/kernel/trace/bpf_trace.h +++ b/kernel/trace/bpf_trace.h @@ -19,7 +19,7 @@ TRACE_EVENT(bpf_trace_printk, ), TP_fast_assign( - __assign_str(bpf_string, bpf_string); + __assign_str(bpf_string); ), TP_printk("%s", __get_str(bpf_string)) diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 218cd95bf8e4..cc48d16be43e 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -7,28 +7,496 @@ * * Highly modified by Steven Rostedt (VMware). */ +#include <linux/bits.h> #include <linux/jump_label.h> #include <linux/suspend.h> #include <linux/ftrace.h> +#include <linux/static_call.h> #include <linux/slab.h> #include <trace/events/sched.h> #include "ftrace_internal.h" +#include "trace.h" -#ifdef CONFIG_DYNAMIC_FTRACE -#define ASSIGN_OPS_HASH(opsname, val) \ - .func_hash = val, \ - .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), -#else -#define ASSIGN_OPS_HASH(opsname, val) -#endif +/* + * FGRAPH_FRAME_SIZE: Size in bytes of the meta data on the shadow stack + * FGRAPH_FRAME_OFFSET: Size in long words of the meta data frame + */ +#define FGRAPH_FRAME_SIZE sizeof(struct ftrace_ret_stack) +#define FGRAPH_FRAME_OFFSET DIV_ROUND_UP(FGRAPH_FRAME_SIZE, sizeof(long)) + +/* + * On entry to a function (via function_graph_enter()), a new fgraph frame + * (ftrace_ret_stack) is pushed onto the stack as well as a word that + * holds a bitmask and a type (called "bitmap"). The bitmap is defined as: + * + * bits: 0 - 9 offset in words from the previous ftrace_ret_stack + * + * bits: 10 - 11 Type of storage + * 0 - reserved + * 1 - bitmap of fgraph_array index + * 2 - reserved data + * + * For type with "bitmap of fgraph_array index" (FGRAPH_TYPE_BITMAP): + * bits: 12 - 27 The bitmap of fgraph_ops fgraph_array index + * That is, it's a bitmask of 0-15 (16 bits) + * where if a corresponding ops in the fgraph_array[] + * expects a callback from the return of the function + * it's corresponding bit will be set. + * + * + * The top of the ret_stack (when not empty) will always have a reference + * word that points to the last fgraph frame that was saved. + * + * For reserved data: + * bits: 12 - 17 The size in words that is stored + * bits: 18 - 23 The index of fgraph_array, which shows who is stored + * + * That is, at the end of function_graph_enter, if the first and forth + * fgraph_ops on the fgraph_array[] (index 0 and 3) needs their retfunc called + * on the return of the function being traced, and the forth fgraph_ops + * stored two words of data, this is what will be on the task's shadow + * ret_stack: (the stack grows upward) + * + * ret_stack[SHADOW_STACK_OFFSET] + * | SHADOW_STACK_TASK_VARS(ret_stack)[15] | + * ... + * | SHADOW_STACK_TASK_VARS(ret_stack)[0] | + * ret_stack[SHADOW_STACK_MAX_OFFSET] + * ... + * | | <- task->curr_ret_stack + * +--------------------------------------------+ + * | (3 << 12) | (3 << 10) | FGRAPH_FRAME_OFFSET| + * | *or put another way* | + * | (3 << FGRAPH_DATA_INDEX_SHIFT)| \ | This is for fgraph_ops[3]. + * | ((2 - 1) << FGRAPH_DATA_SHIFT)| \ | The data size is 2 words. + * | (FGRAPH_TYPE_DATA << FGRAPH_TYPE_SHIFT)| \ | + * | (offset2:FGRAPH_FRAME_OFFSET+3) | <- the offset2 is from here + * +--------------------------------------------+ ( It is 4 words from the ret_stack) + * | STORED DATA WORD 2 | + * | STORED DATA WORD 1 | + * +--------------------------------------------+ + * | (9 << 12) | (1 << 10) | FGRAPH_FRAME_OFFSET| + * | *or put another way* | + * | (BIT(3)|BIT(0)) << FGRAPH_INDEX_SHIFT | \ | + * | FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT| \ | + * | (offset1:FGRAPH_FRAME_OFFSET) | <- the offset1 is from here + * +--------------------------------------------+ + * | struct ftrace_ret_stack | + * | (stores the saved ret pointer) | <- the offset points here + * +--------------------------------------------+ + * | (X) | (N) | ( N words away from + * | | previous ret_stack) + * ... + * ret_stack[0] + * + * If a backtrace is required, and the real return pointer needs to be + * fetched, then it looks at the task's curr_ret_stack offset, if it + * is greater than zero (reserved, or right before popped), it would mask + * the value by FGRAPH_FRAME_OFFSET_MASK to get the offset of the + * ftrace_ret_stack structure stored on the shadow stack. + */ + +/* + * The following is for the top word on the stack: + * + * FGRAPH_FRAME_OFFSET (0-9) holds the offset delta to the fgraph frame + * FGRAPH_TYPE (10-11) holds the type of word this is. + * (RESERVED or BITMAP) + */ +#define FGRAPH_FRAME_OFFSET_BITS 10 +#define FGRAPH_FRAME_OFFSET_MASK GENMASK(FGRAPH_FRAME_OFFSET_BITS - 1, 0) + +#define FGRAPH_TYPE_BITS 2 +#define FGRAPH_TYPE_MASK GENMASK(FGRAPH_TYPE_BITS - 1, 0) +#define FGRAPH_TYPE_SHIFT FGRAPH_FRAME_OFFSET_BITS + +enum { + FGRAPH_TYPE_RESERVED = 0, + FGRAPH_TYPE_BITMAP = 1, + FGRAPH_TYPE_DATA = 2, +}; + +/* + * For BITMAP type: + * FGRAPH_INDEX (12-27) bits holding the gops index wanting return callback called + */ +#define FGRAPH_INDEX_BITS 16 +#define FGRAPH_INDEX_MASK GENMASK(FGRAPH_INDEX_BITS - 1, 0) +#define FGRAPH_INDEX_SHIFT (FGRAPH_TYPE_SHIFT + FGRAPH_TYPE_BITS) + +/* + * For DATA type: + * FGRAPH_DATA (12-17) bits hold the size of data (in words) + * FGRAPH_INDEX (18-23) bits hold the index for which gops->idx the data is for + * + * Note: + * data_size == 0 means 1 word, and 31 (=2^5 - 1) means 32 words. + */ +#define FGRAPH_DATA_BITS 5 +#define FGRAPH_DATA_MASK GENMASK(FGRAPH_DATA_BITS - 1, 0) +#define FGRAPH_DATA_SHIFT (FGRAPH_TYPE_SHIFT + FGRAPH_TYPE_BITS) +#define FGRAPH_MAX_DATA_SIZE (sizeof(long) * (1 << FGRAPH_DATA_BITS)) + +#define FGRAPH_DATA_INDEX_BITS 4 +#define FGRAPH_DATA_INDEX_MASK GENMASK(FGRAPH_DATA_INDEX_BITS - 1, 0) +#define FGRAPH_DATA_INDEX_SHIFT (FGRAPH_DATA_SHIFT + FGRAPH_DATA_BITS) + +#define FGRAPH_MAX_INDEX \ + ((FGRAPH_INDEX_SIZE << FGRAPH_DATA_BITS) + FGRAPH_RET_INDEX) + +#define FGRAPH_ARRAY_SIZE FGRAPH_INDEX_BITS + +/* + * SHADOW_STACK_SIZE: The size in bytes of the entire shadow stack + * SHADOW_STACK_OFFSET: The size in long words of the shadow stack + * SHADOW_STACK_MAX_OFFSET: The max offset of the stack for a new frame to be added + */ +#define SHADOW_STACK_SIZE (4096) +#define SHADOW_STACK_OFFSET (SHADOW_STACK_SIZE / sizeof(long)) +/* Leave on a buffer at the end */ +#define SHADOW_STACK_MAX_OFFSET \ + (SHADOW_STACK_OFFSET - (FGRAPH_FRAME_OFFSET + 1 + FGRAPH_ARRAY_SIZE)) + +/* RET_STACK(): Return the frame from a given @offset from task @t */ +#define RET_STACK(t, offset) ((struct ftrace_ret_stack *)(&(t)->ret_stack[offset])) + +/* + * Each fgraph_ops has a reserved unsigned long at the end (top) of the + * ret_stack to store task specific state. + */ +#define SHADOW_STACK_TASK_VARS(ret_stack) \ + ((unsigned long *)(&(ret_stack)[SHADOW_STACK_OFFSET - FGRAPH_ARRAY_SIZE])) DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph); int ftrace_graph_active; -/* Both enabled by default (can be cleared by function_graph tracer flags */ -static bool fgraph_sleep_time = true; +static struct kmem_cache *fgraph_stack_cachep; + +static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE]; +static unsigned long fgraph_array_bitmask; + +/* LRU index table for fgraph_array */ +static int fgraph_lru_table[FGRAPH_ARRAY_SIZE]; +static int fgraph_lru_next; +static int fgraph_lru_last; + +/* Initialize fgraph_lru_table with unused index */ +static void fgraph_lru_init(void) +{ + int i; + + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) + fgraph_lru_table[i] = i; +} + +/* Release the used index to the LRU table */ +static int fgraph_lru_release_index(int idx) +{ + if (idx < 0 || idx >= FGRAPH_ARRAY_SIZE || + WARN_ON_ONCE(fgraph_lru_table[fgraph_lru_last] != -1)) + return -1; + + fgraph_lru_table[fgraph_lru_last] = idx; + fgraph_lru_last = (fgraph_lru_last + 1) % FGRAPH_ARRAY_SIZE; + + clear_bit(idx, &fgraph_array_bitmask); + return 0; +} + +/* Allocate a new index from LRU table */ +static int fgraph_lru_alloc_index(void) +{ + int idx = fgraph_lru_table[fgraph_lru_next]; + + /* No id is available */ + if (idx == -1) + return -1; + + fgraph_lru_table[fgraph_lru_next] = -1; + fgraph_lru_next = (fgraph_lru_next + 1) % FGRAPH_ARRAY_SIZE; + + set_bit(idx, &fgraph_array_bitmask); + return idx; +} + +/* Get the offset to the fgraph frame from a ret_stack value */ +static inline int __get_offset(unsigned long val) +{ + return val & FGRAPH_FRAME_OFFSET_MASK; +} + +/* Get the type of word from a ret_stack value */ +static inline int __get_type(unsigned long val) +{ + return (val >> FGRAPH_TYPE_SHIFT) & FGRAPH_TYPE_MASK; +} + +/* Get the data_index for a DATA type ret_stack word */ +static inline int __get_data_index(unsigned long val) +{ + return (val >> FGRAPH_DATA_INDEX_SHIFT) & FGRAPH_DATA_INDEX_MASK; +} + +/* Get the data_size for a DATA type ret_stack word */ +static inline int __get_data_size(unsigned long val) +{ + return ((val >> FGRAPH_DATA_SHIFT) & FGRAPH_DATA_MASK) + 1; +} + +/* Get the word from the ret_stack at @offset */ +static inline unsigned long get_fgraph_entry(struct task_struct *t, int offset) +{ + return t->ret_stack[offset]; +} + +/* Get the FRAME_OFFSET from the word from the @offset on ret_stack */ +static inline int get_frame_offset(struct task_struct *t, int offset) +{ + return __get_offset(t->ret_stack[offset]); +} + +/* For BITMAP type: get the bitmask from the @offset at ret_stack */ +static inline unsigned long +get_bitmap_bits(struct task_struct *t, int offset) +{ + return (t->ret_stack[offset] >> FGRAPH_INDEX_SHIFT) & FGRAPH_INDEX_MASK; +} + +/* Write the bitmap to the ret_stack at @offset (does index, offset and bitmask) */ +static inline void +set_bitmap(struct task_struct *t, int offset, unsigned long bitmap) +{ + t->ret_stack[offset] = (bitmap << FGRAPH_INDEX_SHIFT) | + (FGRAPH_TYPE_BITMAP << FGRAPH_TYPE_SHIFT) | FGRAPH_FRAME_OFFSET; +} + +/* For DATA type: get the data saved under the ret_stack word at @offset */ +static inline void *get_data_type_data(struct task_struct *t, int offset) +{ + unsigned long val = t->ret_stack[offset]; + + if (__get_type(val) != FGRAPH_TYPE_DATA) + return NULL; + offset -= __get_data_size(val); + return (void *)&t->ret_stack[offset]; +} + +/* Create the ret_stack word for a DATA type */ +static inline unsigned long make_data_type_val(int idx, int size, int offset) +{ + return (idx << FGRAPH_DATA_INDEX_SHIFT) | + ((size - 1) << FGRAPH_DATA_SHIFT) | + (FGRAPH_TYPE_DATA << FGRAPH_TYPE_SHIFT) | offset; +} + +/* ftrace_graph_entry set to this to tell some archs to run function graph */ +static int entry_run(struct ftrace_graph_ent *trace, struct fgraph_ops *ops, + struct ftrace_regs *fregs) +{ + return 0; +} + +/* ftrace_graph_return set to this to tell some archs to run function graph */ +static void return_run(struct ftrace_graph_ret *trace, struct fgraph_ops *ops, + struct ftrace_regs *fregs) +{ +} + +static void ret_stack_set_task_var(struct task_struct *t, int idx, long val) +{ + unsigned long *gvals = SHADOW_STACK_TASK_VARS(t->ret_stack); + + gvals[idx] = val; +} + +static unsigned long * +ret_stack_get_task_var(struct task_struct *t, int idx) +{ + unsigned long *gvals = SHADOW_STACK_TASK_VARS(t->ret_stack); + + return &gvals[idx]; +} + +static void ret_stack_init_task_vars(unsigned long *ret_stack) +{ + unsigned long *gvals = SHADOW_STACK_TASK_VARS(ret_stack); + + memset(gvals, 0, sizeof(*gvals) * FGRAPH_ARRAY_SIZE); +} + +/** + * fgraph_reserve_data - Reserve storage on the task's ret_stack + * @idx: The index of fgraph_array + * @size_bytes: The size in bytes to reserve + * + * Reserves space of up to FGRAPH_MAX_DATA_SIZE bytes on the + * task's ret_stack shadow stack, for a given fgraph_ops during + * the entryfunc() call. If entryfunc() returns zero, the storage + * is discarded. An entryfunc() can only call this once per iteration. + * The fgraph_ops retfunc() can retrieve this stored data with + * fgraph_retrieve_data(). + * + * Returns: On success, a pointer to the data on the stack. + * Otherwise, NULL if there's not enough space left on the + * ret_stack for the data, or if fgraph_reserve_data() was called + * more than once for a single entryfunc() call. + */ +void *fgraph_reserve_data(int idx, int size_bytes) +{ + unsigned long val; + void *data; + int curr_ret_stack = current->curr_ret_stack; + int data_size; + + if (size_bytes > FGRAPH_MAX_DATA_SIZE) + return NULL; + + /* Convert the data size to number of longs. */ + data_size = (size_bytes + sizeof(long) - 1) >> (sizeof(long) == 4 ? 2 : 3); + + val = get_fgraph_entry(current, curr_ret_stack - 1); + data = ¤t->ret_stack[curr_ret_stack]; + + curr_ret_stack += data_size + 1; + if (unlikely(curr_ret_stack >= SHADOW_STACK_MAX_OFFSET)) + return NULL; + + val = make_data_type_val(idx, data_size, __get_offset(val) + data_size + 1); + + /* Set the last word to be reserved */ + current->ret_stack[curr_ret_stack - 1] = val; + + /* Make sure interrupts see this */ + barrier(); + current->curr_ret_stack = curr_ret_stack; + /* Again sync with interrupts, and reset reserve */ + current->ret_stack[curr_ret_stack - 1] = val; + + return data; +} + +/** + * fgraph_retrieve_data - Retrieve stored data from fgraph_reserve_data() + * @idx: the index of fgraph_array (fgraph_ops::idx) + * @size_bytes: pointer to retrieved data size. + * + * This is to be called by a fgraph_ops retfunc(), to retrieve data that + * was stored by the fgraph_ops entryfunc() on the function entry. + * That is, this will retrieve the data that was reserved on the + * entry of the function that corresponds to the exit of the function + * that the fgraph_ops retfunc() is called on. + * + * Returns: The stored data from fgraph_reserve_data() called by the + * matching entryfunc() for the retfunc() this is called from. + * Or NULL if there was nothing stored. + */ +void *fgraph_retrieve_data(int idx, int *size_bytes) +{ + return fgraph_retrieve_parent_data(idx, size_bytes, 0); +} + +/** + * fgraph_get_task_var - retrieve a task specific state variable + * @gops: The ftrace_ops that owns the task specific variable + * + * Every registered fgraph_ops has a task state variable + * reserved on the task's ret_stack. This function returns the + * address to that variable. + * + * Returns the address to the fgraph_ops @gops tasks specific + * unsigned long variable. + */ +unsigned long *fgraph_get_task_var(struct fgraph_ops *gops) +{ + return ret_stack_get_task_var(current, gops->idx); +} + +/* + * @offset: The offset into @t->ret_stack to find the ret_stack entry + * @frame_offset: Where to place the offset into @t->ret_stack of that entry + * + * Returns a pointer to the previous ret_stack below @offset or NULL + * when it reaches the bottom of the stack. + * + * Calling this with: + * + * offset = task->curr_ret_stack; + * do { + * ret_stack = get_ret_stack(task, offset, &offset); + * } while (ret_stack); + * + * Will iterate through all the ret_stack entries from curr_ret_stack + * down to the first one. + */ +static inline struct ftrace_ret_stack * +get_ret_stack(struct task_struct *t, int offset, int *frame_offset) +{ + int offs; + + BUILD_BUG_ON(FGRAPH_FRAME_SIZE % sizeof(long)); + + if (unlikely(offset <= 0)) + return NULL; + + offs = get_frame_offset(t, --offset); + if (WARN_ON_ONCE(offs <= 0 || offs > offset)) + return NULL; + + offset -= offs; + + *frame_offset = offset; + return RET_STACK(t, offset); +} + +/** + * fgraph_retrieve_parent_data - get data from a parent function + * @idx: The index into the fgraph_array (fgraph_ops::idx) + * @size_bytes: A pointer to retrieved data size + * @depth: The depth to find the parent (0 is the current function) + * + * This is similar to fgraph_retrieve_data() but can be used to retrieve + * data from a parent caller function. + * + * Return: a pointer to the specified parent data or NULL if not found + */ +void *fgraph_retrieve_parent_data(int idx, int *size_bytes, int depth) +{ + struct ftrace_ret_stack *ret_stack = NULL; + int offset = current->curr_ret_stack; + unsigned long val; + + if (offset <= 0) + return NULL; + + for (;;) { + int next_offset; + + ret_stack = get_ret_stack(current, offset, &next_offset); + if (!ret_stack || --depth < 0) + break; + offset = next_offset; + } + + if (!ret_stack) + return NULL; + + offset--; + + val = get_fgraph_entry(current, offset); + while (__get_type(val) == FGRAPH_TYPE_DATA) { + if (__get_data_index(val) == idx) + goto found; + offset -= __get_data_size(val) + 1; + val = get_fgraph_entry(current, offset); + } + return NULL; +found: + if (size_bytes) + *size_bytes = __get_data_size(val) * sizeof(long); + return get_data_type_data(current, offset); +} #ifdef CONFIG_DYNAMIC_FTRACE /* @@ -50,6 +518,29 @@ int __weak ftrace_disable_ftrace_graph_caller(void) } #endif +int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + return 0; +} + +static void ftrace_graph_ret_stub(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ +} + +static struct fgraph_ops fgraph_stub = { + .entryfunc = ftrace_graph_entry_stub, + .retfunc = ftrace_graph_ret_stub, +}; + +static struct fgraph_ops *fgraph_direct_gops = &fgraph_stub; +DEFINE_STATIC_CALL(fgraph_func, ftrace_graph_entry_stub); +DEFINE_STATIC_CALL(fgraph_retfunc, ftrace_graph_ret_stub); +static DEFINE_STATIC_KEY_TRUE(fgraph_do_direct); + /** * ftrace_graph_stop - set to permanently disable function graph tracing * @@ -66,10 +557,12 @@ void ftrace_graph_stop(void) /* Add a function return address to the trace stack on thread info.*/ static int ftrace_push_return_trace(unsigned long ret, unsigned long func, - unsigned long frame_pointer, unsigned long *retp) + unsigned long frame_pointer, unsigned long *retp, + int fgraph_idx) { - unsigned long long calltime; - int index; + struct ftrace_ret_stack *ret_stack; + unsigned long val; + int offset; if (unlikely(ftrace_graph_is_dead())) return -EBUSY; @@ -77,32 +570,64 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, if (!current->ret_stack) return -EBUSY; + BUILD_BUG_ON(SHADOW_STACK_SIZE % sizeof(long)); + + /* Set val to "reserved" with the delta to the new fgraph frame */ + val = (FGRAPH_TYPE_RESERVED << FGRAPH_TYPE_SHIFT) | FGRAPH_FRAME_OFFSET; + /* * We must make sure the ret_stack is tested before we read * anything else. */ smp_rmb(); - /* The return trace stack is full */ - if (current->curr_ret_stack == FTRACE_RETFUNC_DEPTH - 1) { + /* + * Check if there's room on the shadow stack to fit a fraph frame + * and a bitmap word. + */ + if (current->curr_ret_stack + FGRAPH_FRAME_OFFSET + 1 >= SHADOW_STACK_MAX_OFFSET) { atomic_inc(¤t->trace_overrun); return -EBUSY; } - calltime = trace_clock_local(); + offset = READ_ONCE(current->curr_ret_stack); + ret_stack = RET_STACK(current, offset); + offset += FGRAPH_FRAME_OFFSET; - index = ++current->curr_ret_stack; + /* ret offset = FGRAPH_FRAME_OFFSET ; type = reserved */ + current->ret_stack[offset] = val; + ret_stack->ret = ret; + /* + * The unwinders expect curr_ret_stack to point to either zero + * or an offset where to find the next ret_stack. Even though the + * ret stack might be bogus, we want to write the ret and the + * offset to find the ret_stack before we increment the stack point. + * If an interrupt comes in now before we increment the curr_ret_stack + * it may blow away what we wrote. But that's fine, because the + * offset will still be correct (even though the 'ret' won't be). + * What we worry about is the offset being correct after we increment + * the curr_ret_stack and before we update that offset, as if an + * interrupt comes in and does an unwind stack dump, it will need + * at least a correct offset! + */ + barrier(); + WRITE_ONCE(current->curr_ret_stack, offset + 1); + /* + * This next barrier is to ensure that an interrupt coming in + * will not corrupt what we are about to write. + */ barrier(); - current->ret_stack[index].ret = ret; - current->ret_stack[index].func = func; - current->ret_stack[index].calltime = calltime; + + /* Still keep it reserved even if an interrupt came in */ + current->ret_stack[offset] = val; + + ret_stack->ret = ret; + ret_stack->func = func; #ifdef HAVE_FUNCTION_GRAPH_FP_TEST - current->ret_stack[index].fp = frame_pointer; + ret_stack->fp = frame_pointer; #endif -#ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR - current->ret_stack[index].retp = retp; -#endif - return 0; + ret_stack->retp = retp; + return offset; } /* @@ -119,55 +644,92 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, # define MCOUNT_INSN_SIZE 0 #endif -int function_graph_enter(unsigned long ret, unsigned long func, - unsigned long frame_pointer, unsigned long *retp) +/* If the caller does not use ftrace, call this function. */ +int function_graph_enter_regs(unsigned long ret, unsigned long func, + unsigned long frame_pointer, unsigned long *retp, + struct ftrace_regs *fregs) { struct ftrace_graph_ent trace; + unsigned long bitmap = 0; + int offset; + int bit; + int i; -#ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS - /* - * Skip graph tracing if the return location is served by direct trampoline, - * since call sequence and return addresses are unpredictable anyway. - * Ex: BPF trampoline may call original function and may skip frame - * depending on type of BPF programs attached. - */ - if (ftrace_direct_func_count && - ftrace_find_rec_direct(ret - MCOUNT_INSN_SIZE)) + bit = ftrace_test_recursion_trylock(func, ret); + if (bit < 0) return -EBUSY; -#endif + trace.func = func; trace.depth = ++current->curr_ret_depth; - if (ftrace_push_return_trace(ret, func, frame_pointer, retp)) + offset = ftrace_push_return_trace(ret, func, frame_pointer, retp, 0); + if (offset < 0) goto out; - /* Only trace if the calling function expects to */ - if (!ftrace_graph_entry(&trace)) +#ifdef CONFIG_HAVE_STATIC_CALL + if (static_branch_likely(&fgraph_do_direct)) { + int save_curr_ret_stack = current->curr_ret_stack; + + if (static_call(fgraph_func)(&trace, fgraph_direct_gops, fregs)) + bitmap |= BIT(fgraph_direct_gops->idx); + else + /* Clear out any saved storage */ + current->curr_ret_stack = save_curr_ret_stack; + } else +#endif + { + for_each_set_bit(i, &fgraph_array_bitmask, + sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { + struct fgraph_ops *gops = READ_ONCE(fgraph_array[i]); + int save_curr_ret_stack; + + if (gops == &fgraph_stub) + continue; + + save_curr_ret_stack = current->curr_ret_stack; + if (ftrace_ops_test(&gops->ops, func, NULL) && + gops->entryfunc(&trace, gops, fregs)) + bitmap |= BIT(i); + else + /* Clear out any saved storage */ + current->curr_ret_stack = save_curr_ret_stack; + } + } + + if (!bitmap) goto out_ret; + /* + * Since this function uses fgraph_idx = 0 as a tail-call checking + * flag, set that bit always. + */ + set_bitmap(current, offset, bitmap | BIT(0)); + ftrace_test_recursion_unlock(bit); return 0; out_ret: - current->curr_ret_stack--; + current->curr_ret_stack -= FGRAPH_FRAME_OFFSET + 1; out: current->curr_ret_depth--; + ftrace_test_recursion_unlock(bit); return -EBUSY; } /* Retrieve a function return address to the trace stack on thread info.*/ -static void +static struct ftrace_ret_stack * ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, - unsigned long frame_pointer) + unsigned long frame_pointer, int *offset) { - int index; + struct ftrace_ret_stack *ret_stack; - index = current->curr_ret_stack; + ret_stack = get_ret_stack(current, current->curr_ret_stack, offset); - if (unlikely(index < 0 || index >= FTRACE_RETFUNC_DEPTH)) { + if (unlikely(!ret_stack)) { ftrace_graph_stop(); - WARN_ON(1); + WARN(1, "Bad function graph ret_stack pointer: %d", + current->curr_ret_stack); /* Might as well panic, otherwise we have no where to go */ *ret = (unsigned long)panic; - return; + return NULL; } #ifdef HAVE_FUNCTION_GRAPH_FP_TEST @@ -185,30 +747,32 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, * Note, -mfentry does not use frame pointers, and this test * is not needed if CC_USING_FENTRY is set. */ - if (unlikely(current->ret_stack[index].fp != frame_pointer)) { + if (unlikely(ret_stack->fp != frame_pointer)) { ftrace_graph_stop(); WARN(1, "Bad frame pointer: expected %lx, received %lx\n" " from func %ps return to %lx\n", - current->ret_stack[index].fp, + ret_stack->fp, frame_pointer, - (void *)current->ret_stack[index].func, - current->ret_stack[index].ret); + (void *)ret_stack->func, + ret_stack->ret); *ret = (unsigned long)panic; - return; + return NULL; } #endif - *ret = current->ret_stack[index].ret; - trace->func = current->ret_stack[index].func; - trace->calltime = current->ret_stack[index].calltime; + *offset += FGRAPH_FRAME_OFFSET; + *ret = ret_stack->ret; + trace->func = ret_stack->func; trace->overrun = atomic_read(¤t->trace_overrun); - trace->depth = current->curr_ret_depth--; + trace->depth = current->curr_ret_depth; /* * We still want to trace interrupts coming in if * max_depth is set to 1. Make sure the decrement is * seen before ftrace_graph_return. */ barrier(); + + return ret_stack; } /* @@ -240,35 +804,95 @@ static struct notifier_block ftrace_suspend_notifier = { * Send the trace to the ring-buffer. * @return the original return address. */ -unsigned long ftrace_return_to_handler(unsigned long frame_pointer) +static inline unsigned long +__ftrace_return_to_handler(struct ftrace_regs *fregs, unsigned long frame_pointer) { + struct ftrace_ret_stack *ret_stack; struct ftrace_graph_ret trace; + unsigned long bitmap; unsigned long ret; + int offset; + int bit; + int i; + + ret_stack = ftrace_pop_return_trace(&trace, &ret, frame_pointer, &offset); + + if (unlikely(!ret_stack)) { + ftrace_graph_stop(); + WARN_ON(1); + /* Might as well panic. What else to do? */ + return (unsigned long)panic; + } + + if (fregs) + ftrace_regs_set_instruction_pointer(fregs, ret); - ftrace_pop_return_trace(&trace, &ret, frame_pointer); - trace.rettime = trace_clock_local(); - ftrace_graph_return(&trace); + bit = ftrace_test_recursion_trylock(trace.func, ret); + /* + * This can fail because ftrace_test_recursion_trylock() allows one nest + * call. If we are already in a nested call, then we don't probe this and + * just return the original return address. + */ + if (unlikely(bit < 0)) + goto out; + +#ifdef CONFIG_FUNCTION_GRAPH_RETVAL + trace.retval = ftrace_regs_get_return_value(fregs); +#endif + + bitmap = get_bitmap_bits(current, offset); + +#ifdef CONFIG_HAVE_STATIC_CALL + if (static_branch_likely(&fgraph_do_direct)) { + if (test_bit(fgraph_direct_gops->idx, &bitmap)) + static_call(fgraph_retfunc)(&trace, fgraph_direct_gops, fregs); + } else +#endif + { + for_each_set_bit(i, &bitmap, sizeof(bitmap) * BITS_PER_BYTE) { + struct fgraph_ops *gops = READ_ONCE(fgraph_array[i]); + + if (gops == &fgraph_stub) + continue; + + gops->retfunc(&trace, gops, fregs); + } + } + + ftrace_test_recursion_unlock(bit); +out: /* * The ftrace_graph_return() may still access the current * ret_stack structure, we need to make sure the update of * curr_ret_stack is after that. */ barrier(); - current->curr_ret_stack--; - - if (unlikely(!ret)) { - ftrace_graph_stop(); - WARN_ON(1); - /* Might as well panic. What else to do? */ - ret = (unsigned long)panic; - } + current->curr_ret_stack = offset - FGRAPH_FRAME_OFFSET; + current->curr_ret_depth--; return ret; } +/* + * After all architectures have selected HAVE_FUNCTION_GRAPH_FREGS, we can + * leave only ftrace_return_to_handler(fregs). + */ +#ifdef CONFIG_HAVE_FUNCTION_GRAPH_FREGS +unsigned long ftrace_return_to_handler(struct ftrace_regs *fregs) +{ + return __ftrace_return_to_handler(fregs, + ftrace_regs_get_frame_pointer(fregs)); +} +#else +unsigned long ftrace_return_to_handler(unsigned long frame_pointer) +{ + return __ftrace_return_to_handler(NULL, frame_pointer); +} +#endif + /** * ftrace_graph_get_ret_stack - return the entry of the shadow stack - * @task: The task to read the shadow stack from + * @task: The task to read the shadow stack from. * @idx: Index down the shadow stack * * Return the ret_struct on the shadow stack of the @task at the @@ -280,115 +904,147 @@ unsigned long ftrace_return_to_handler(unsigned long frame_pointer) struct ftrace_ret_stack * ftrace_graph_get_ret_stack(struct task_struct *task, int idx) { - idx = task->curr_ret_stack - idx; + struct ftrace_ret_stack *ret_stack = NULL; + int offset = task->curr_ret_stack; - if (idx >= 0 && idx <= task->curr_ret_stack) - return &task->ret_stack[idx]; + if (offset < 0) + return NULL; - return NULL; + do { + ret_stack = get_ret_stack(task, offset, &offset); + } while (ret_stack && --idx >= 0); + + return ret_stack; +} + +/** + * ftrace_graph_top_ret_addr - return the top return address in the shadow stack + * @task: The task to read the shadow stack from. + * + * Return the first return address on the shadow stack of the @task, which is + * not the fgraph's return_to_handler. + */ +unsigned long ftrace_graph_top_ret_addr(struct task_struct *task) +{ + unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler); + struct ftrace_ret_stack *ret_stack = NULL; + int offset = task->curr_ret_stack; + + if (offset < 0) + return 0; + + do { + ret_stack = get_ret_stack(task, offset, &offset); + } while (ret_stack && ret_stack->ret == return_handler); + + return ret_stack ? ret_stack->ret : 0; } /** - * ftrace_graph_ret_addr - convert a potentially modified stack return address - * to its original value + * ftrace_graph_ret_addr - return the original value of the return address + * @task: The task the unwinder is being executed on + * @idx: An initialized pointer to the next stack index to use + * @ret: The current return address (likely pointing to return_handler) + * @retp: The address on the stack of the current return location * * This function can be called by stack unwinding code to convert a found stack - * return address ('ret') to its original value, in case the function graph + * return address (@ret) to its original value, in case the function graph * tracer has modified it to be 'return_to_handler'. If the address hasn't - * been modified, the unchanged value of 'ret' is returned. + * been modified, the unchanged value of @ret is returned. * - * 'idx' is a state variable which should be initialized by the caller to zero - * before the first call. + * @idx holds the last index used to know where to start from. It should be + * initialized to zero for the first iteration as that will mean to start + * at the top of the shadow stack. If the location is found, this pointer + * will be assigned that location so that if called again, it will continue + * where it left off. * - * 'retp' is a pointer to the return address on the stack. It's ignored if - * the arch doesn't have HAVE_FUNCTION_GRAPH_RET_ADDR_PTR defined. + * @retp is a pointer to the return address on the stack. */ -#ifdef HAVE_FUNCTION_GRAPH_RET_ADDR_PTR unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, unsigned long ret, unsigned long *retp) { - int index = task->curr_ret_stack; + struct ftrace_ret_stack *ret_stack; + unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler); int i; - if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) + if (ret != return_handler) return ret; - if (index < 0) + if (!idx) return ret; - for (i = 0; i <= index; i++) - if (task->ret_stack[i].retp == retp) - return task->ret_stack[i].ret; + i = *idx ? : task->curr_ret_stack; + while (i > 0) { + ret_stack = get_ret_stack(task, i, &i); + if (!ret_stack) + break; + /* + * For the tail-call, there would be 2 or more ftrace_ret_stacks on + * the ret_stack, which records "return_to_handler" as the return + * address except for the last one. + * But on the real stack, there should be 1 entry because tail-call + * reuses the return address on the stack and jump to the next function. + * Thus we will continue to find real return address. + */ + if (ret_stack->retp == retp && + ret_stack->ret != return_handler) { + *idx = i; + return ret_stack->ret; + } + } return ret; } -#else /* !HAVE_FUNCTION_GRAPH_RET_ADDR_PTR */ -unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx, - unsigned long ret, unsigned long *retp) -{ - int task_idx; - - if (ret != (unsigned long)dereference_kernel_function_descriptor(return_to_handler)) - return ret; - - task_idx = task->curr_ret_stack; - - if (!task->ret_stack || task_idx < *idx) - return ret; - - task_idx -= *idx; - (*idx)++; - - return task->ret_stack[task_idx].ret; -} -#endif /* HAVE_FUNCTION_GRAPH_RET_ADDR_PTR */ static struct ftrace_ops graph_ops = { .func = ftrace_graph_func, - .flags = FTRACE_OPS_FL_INITIALIZED | - FTRACE_OPS_FL_PID | - FTRACE_OPS_GRAPH_STUB, + .flags = FTRACE_OPS_GRAPH_STUB, #ifdef FTRACE_GRAPH_TRAMP_ADDR .trampoline = FTRACE_GRAPH_TRAMP_ADDR, /* trampoline_size is only needed for dynamically allocated tramps */ #endif - ASSIGN_OPS_HASH(graph_ops, &global_ops.local_hash) }; -void ftrace_graph_sleep_time_control(bool enable) +void fgraph_init_ops(struct ftrace_ops *dst_ops, + struct ftrace_ops *src_ops) { - fgraph_sleep_time = enable; -} + dst_ops->flags = FTRACE_OPS_FL_PID | FTRACE_OPS_GRAPH_STUB; -int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) -{ - return 0; +#ifdef CONFIG_DYNAMIC_FTRACE + if (src_ops) { + dst_ops->func_hash = &src_ops->local_hash; + mutex_init(&dst_ops->local_hash.regex_lock); + INIT_LIST_HEAD(&dst_ops->subop_list); + dst_ops->flags |= FTRACE_OPS_FL_INITIALIZED; + dst_ops->private = src_ops->private; + } +#endif } /* * Simply points to ftrace_stub, but with the proper protocol. * Defined by the linker script in linux/vmlinux.lds.h */ -extern void ftrace_stub_graph(struct ftrace_graph_ret *); +void ftrace_stub_graph(struct ftrace_graph_ret *trace, struct fgraph_ops *gops, + struct ftrace_regs *fregs); /* The callbacks that hook a function */ trace_func_graph_ret_t ftrace_graph_return = ftrace_stub_graph; trace_func_graph_ent_t ftrace_graph_entry = ftrace_graph_entry_stub; -static trace_func_graph_ent_t __ftrace_graph_entry = ftrace_graph_entry_stub; /* Try to assign a return stack array on FTRACE_RETSTACK_ALLOC_SIZE tasks. */ -static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) +static int alloc_retstack_tasklist(unsigned long **ret_stack_list) { int i; int ret = 0; int start = 0, end = FTRACE_RETSTACK_ALLOC_SIZE; struct task_struct *g, *t; + if (WARN_ON_ONCE(!fgraph_stack_cachep)) + return -ENOMEM; + for (i = 0; i < FTRACE_RETSTACK_ALLOC_SIZE; i++) { - ret_stack_list[i] = - kmalloc_array(FTRACE_RETFUNC_DEPTH, - sizeof(struct ftrace_ret_stack), - GFP_KERNEL); + ret_stack_list[i] = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL); if (!ret_stack_list[i]) { start = 0; end = i; @@ -406,9 +1062,10 @@ static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) if (t->ret_stack == NULL) { atomic_set(&t->trace_overrun, 0); - t->curr_ret_stack = -1; + ret_stack_init_task_vars(ret_stack_list[start]); + t->curr_ret_stack = 0; t->curr_ret_depth = -1; - /* Make sure the tasks see the -1 first: */ + /* Make sure the tasks see the 0 first: */ smp_wmb(); t->ret_stack = ret_stack_list[start++]; } @@ -418,7 +1075,7 @@ unlock: rcu_read_unlock(); free: for (i = start; i < end; i++) - kfree(ret_stack_list[i]); + kmem_cache_free(fgraph_stack_cachep, ret_stack_list[i]); return ret; } @@ -429,13 +1086,12 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, unsigned int prev_state) { unsigned long long timestamp; - int index; /* * Does the user want to count the time a function was asleep. * If so, do not update the time stamps. */ - if (fgraph_sleep_time) + if (!fgraph_no_sleep_time) return; timestamp = trace_clock_local(); @@ -446,63 +1102,19 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt, if (!next->ftrace_timestamp) return; - /* - * Update all the counters in next to make up for the - * time next was sleeping. - */ - timestamp -= next->ftrace_timestamp; - - for (index = next->curr_ret_stack; index >= 0; index--) - next->ret_stack[index].calltime += timestamp; -} - -static int ftrace_graph_entry_test(struct ftrace_graph_ent *trace) -{ - if (!ftrace_ops_test(&global_ops, trace->func, NULL)) - return 0; - return __ftrace_graph_entry(trace); -} - -/* - * The function graph tracer should only trace the functions defined - * by set_ftrace_filter and set_ftrace_notrace. If another function - * tracer ops is registered, the graph tracer requires testing the - * function against the global ops, and not just trace any function - * that any ftrace_ops registered. - */ -void update_function_graph_func(void) -{ - struct ftrace_ops *op; - bool do_test = false; - - /* - * The graph and global ops share the same set of functions - * to test. If any other ops is on the list, then - * the graph tracing needs to test if its the function - * it should call. - */ - do_for_each_ftrace_op(op, ftrace_ops_list) { - if (op != &global_ops && op != &graph_ops && - op != &ftrace_list_end) { - do_test = true; - /* in double loop, break out with goto */ - goto out; - } - } while_for_each_ftrace_op(op); - out: - if (do_test) - ftrace_graph_entry = ftrace_graph_entry_test; - else - ftrace_graph_entry = __ftrace_graph_entry; + next->ftrace_sleeptime += timestamp - next->ftrace_timestamp; } -static DEFINE_PER_CPU(struct ftrace_ret_stack *, idle_ret_stack); +static DEFINE_PER_CPU(unsigned long *, idle_ret_stack); static void -graph_init_task(struct task_struct *t, struct ftrace_ret_stack *ret_stack) +graph_init_task(struct task_struct *t, unsigned long *ret_stack) { atomic_set(&t->trace_overrun, 0); + ret_stack_init_task_vars(ret_stack); t->ftrace_timestamp = 0; + t->curr_ret_stack = 0; + t->curr_ret_depth = -1; /* make curr_ret_stack visible before we add the ret_stack */ smp_wmb(); t->ret_stack = ret_stack; @@ -514,7 +1126,7 @@ graph_init_task(struct task_struct *t, struct ftrace_ret_stack *ret_stack) */ void ftrace_graph_init_idle_task(struct task_struct *t, int cpu) { - t->curr_ret_stack = -1; + t->curr_ret_stack = 0; t->curr_ret_depth = -1; /* * The idle task has no parent, it either has its own @@ -524,14 +1136,14 @@ void ftrace_graph_init_idle_task(struct task_struct *t, int cpu) WARN_ON(t->ret_stack != per_cpu(idle_ret_stack, cpu)); if (ftrace_graph_active) { - struct ftrace_ret_stack *ret_stack; + unsigned long *ret_stack; + + if (WARN_ON_ONCE(!fgraph_stack_cachep)) + return; ret_stack = per_cpu(idle_ret_stack, cpu); if (!ret_stack) { - ret_stack = - kmalloc_array(FTRACE_RETFUNC_DEPTH, - sizeof(struct ftrace_ret_stack), - GFP_KERNEL); + ret_stack = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL); if (!ret_stack) return; per_cpu(idle_ret_stack, cpu) = ret_stack; @@ -545,15 +1157,16 @@ void ftrace_graph_init_task(struct task_struct *t) { /* Make sure we do not use the parent ret_stack */ t->ret_stack = NULL; - t->curr_ret_stack = -1; + t->curr_ret_stack = 0; t->curr_ret_depth = -1; if (ftrace_graph_active) { - struct ftrace_ret_stack *ret_stack; + unsigned long *ret_stack; + + if (WARN_ON_ONCE(!fgraph_stack_cachep)) + return; - ret_stack = kmalloc_array(FTRACE_RETFUNC_DEPTH, - sizeof(struct ftrace_ret_stack), - GFP_KERNEL); + ret_stack = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL); if (!ret_stack) return; graph_init_task(t, ret_stack); @@ -562,24 +1175,67 @@ void ftrace_graph_init_task(struct task_struct *t) void ftrace_graph_exit_task(struct task_struct *t) { - struct ftrace_ret_stack *ret_stack = t->ret_stack; + unsigned long *ret_stack = t->ret_stack; t->ret_stack = NULL; /* NULL must become visible to IRQs before we free it: */ barrier(); - kfree(ret_stack); + if (ret_stack) { + if (WARN_ON_ONCE(!fgraph_stack_cachep)) + return; + kmem_cache_free(fgraph_stack_cachep, ret_stack); + } +} + +#ifdef CONFIG_DYNAMIC_FTRACE +static int fgraph_pid_func(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + struct trace_array *tr = gops->ops.private; + int pid; + + if (tr) { + pid = this_cpu_read(tr->array_buffer.data->ftrace_ignore_pid); + if (pid == FTRACE_PID_IGNORE) + return 0; + if (pid != FTRACE_PID_TRACE && + pid != current->pid) + return 0; + } + + return gops->saved_func(trace, gops, fregs); } +void fgraph_update_pid_func(void) +{ + struct fgraph_ops *gops; + struct ftrace_ops *op; + + if (!(graph_ops.flags & FTRACE_OPS_FL_INITIALIZED)) + return; + + list_for_each_entry(op, &graph_ops.subop_list, list) { + if (op->flags & FTRACE_OPS_FL_PID) { + gops = container_of(op, struct fgraph_ops, ops); + gops->entryfunc = ftrace_pids_enabled(op) ? + fgraph_pid_func : gops->saved_func; + if (ftrace_graph_active == 1) + static_call_update(fgraph_func, gops->entryfunc); + } + } +} +#endif + /* Allocate a return stack for each task */ static int start_graph_tracing(void) { - struct ftrace_ret_stack **ret_stack_list; + unsigned long **ret_stack_list; int ret, cpu; - ret_stack_list = kmalloc_array(FTRACE_RETSTACK_ALLOC_SIZE, - sizeof(struct ftrace_ret_stack *), - GFP_KERNEL); + ret_stack_list = kcalloc(FTRACE_RETSTACK_ALLOC_SIZE, + sizeof(*ret_stack_list), GFP_KERNEL); if (!ret_stack_list) return -ENOMEM; @@ -605,60 +1261,199 @@ static int start_graph_tracing(void) return ret; } +static void init_task_vars(int idx) +{ + struct task_struct *g, *t; + int cpu; + + for_each_online_cpu(cpu) { + if (idle_task(cpu)->ret_stack) + ret_stack_set_task_var(idle_task(cpu), idx, 0); + } + + read_lock(&tasklist_lock); + for_each_process_thread(g, t) { + if (t->ret_stack) + ret_stack_set_task_var(t, idx, 0); + } + read_unlock(&tasklist_lock); +} + +static void ftrace_graph_enable_direct(bool enable_branch, struct fgraph_ops *gops) +{ + trace_func_graph_ent_t func = NULL; + trace_func_graph_ret_t retfunc = NULL; + int i; + + if (gops) { + func = gops->entryfunc; + retfunc = gops->retfunc; + fgraph_direct_gops = gops; + } else { + for_each_set_bit(i, &fgraph_array_bitmask, + sizeof(fgraph_array_bitmask) * BITS_PER_BYTE) { + func = fgraph_array[i]->entryfunc; + retfunc = fgraph_array[i]->retfunc; + fgraph_direct_gops = fgraph_array[i]; + } + } + if (WARN_ON_ONCE(!func)) + return; + + static_call_update(fgraph_func, func); + static_call_update(fgraph_retfunc, retfunc); + if (enable_branch) + static_branch_disable(&fgraph_do_direct); +} + +static void ftrace_graph_disable_direct(bool disable_branch) +{ + if (disable_branch) + static_branch_disable(&fgraph_do_direct); + static_call_update(fgraph_func, ftrace_graph_entry_stub); + static_call_update(fgraph_retfunc, ftrace_graph_ret_stub); + fgraph_direct_gops = &fgraph_stub; +} + +/* The cpu_boot init_task->ret_stack will never be freed */ +static int fgraph_cpu_init(unsigned int cpu) +{ + if (!idle_task(cpu)->ret_stack) + ftrace_graph_init_idle_task(idle_task(cpu), cpu); + return 0; +} + int register_ftrace_graph(struct fgraph_ops *gops) { + static bool fgraph_initialized; + int command = 0; int ret = 0; + int i = -1; - mutex_lock(&ftrace_lock); + if (WARN_ONCE(gops->ops.flags & FTRACE_OPS_FL_GRAPH, + "function graph ops registered again")) + return -EBUSY; - /* we currently allow only one tracer registered at a time */ - if (ftrace_graph_active) { - ret = -EBUSY; - goto out; + guard(mutex)(&ftrace_lock); + + if (!fgraph_stack_cachep) { + fgraph_stack_cachep = kmem_cache_create("fgraph_stack", + SHADOW_STACK_SIZE, + SHADOW_STACK_SIZE, 0, NULL); + if (!fgraph_stack_cachep) + return -ENOMEM; + } + + if (!fgraph_initialized) { + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "fgraph:online", + fgraph_cpu_init, NULL); + if (ret < 0) { + pr_warn("fgraph: Error to init cpu hotplug support\n"); + return ret; + } + fgraph_initialized = true; + ret = 0; + } + + if (!fgraph_array[0]) { + /* The array must always have real data on it */ + for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) + fgraph_array[i] = &fgraph_stub; + fgraph_lru_init(); } - register_pm_notifier(&ftrace_suspend_notifier); + i = fgraph_lru_alloc_index(); + if (i < 0 || WARN_ON_ONCE(fgraph_array[i] != &fgraph_stub)) + return -ENOSPC; + gops->idx = i; ftrace_graph_active++; - ret = start_graph_tracing(); - if (ret) { - ftrace_graph_active--; - goto out; + + /* Always save the function, and reset at unregistering */ + gops->saved_func = gops->entryfunc; +#ifdef CONFIG_DYNAMIC_FTRACE + if (ftrace_pids_enabled(&gops->ops)) + gops->entryfunc = fgraph_pid_func; +#endif + + if (ftrace_graph_active == 2) + ftrace_graph_disable_direct(true); + + if (ftrace_graph_active == 1) { + ftrace_graph_enable_direct(false, gops); + register_pm_notifier(&ftrace_suspend_notifier); + ret = start_graph_tracing(); + if (ret) + goto error; + /* + * Some archs just test to see if these are not + * the default function + */ + ftrace_graph_return = return_run; + ftrace_graph_entry = entry_run; + command = FTRACE_START_FUNC_RET; + } else { + init_task_vars(gops->idx); } - ftrace_graph_return = gops->retfunc; + gops->ops.flags |= FTRACE_OPS_FL_GRAPH; - /* - * Update the indirect function to the entryfunc, and the - * function that gets called to the entry_test first. Then - * call the update fgraph entry function to determine if - * the entryfunc should be called directly or not. - */ - __ftrace_graph_entry = gops->entryfunc; - ftrace_graph_entry = ftrace_graph_entry_test; - update_function_graph_func(); + ret = ftrace_startup_subops(&graph_ops, &gops->ops, command); + if (!ret) + fgraph_array[i] = gops; - ret = ftrace_startup(&graph_ops, FTRACE_START_FUNC_RET); -out: - mutex_unlock(&ftrace_lock); +error: + if (ret) { + ftrace_graph_active--; + gops->saved_func = NULL; + fgraph_lru_release_index(i); + if (!ftrace_graph_active) + unregister_pm_notifier(&ftrace_suspend_notifier); + } return ret; } void unregister_ftrace_graph(struct fgraph_ops *gops) { - mutex_lock(&ftrace_lock); + int command = 0; + + if (WARN_ONCE(!(gops->ops.flags & FTRACE_OPS_FL_GRAPH), + "function graph ops unregistered without registering")) + return; + + guard(mutex)(&ftrace_lock); if (unlikely(!ftrace_graph_active)) goto out; + if (unlikely(gops->idx < 0 || gops->idx >= FGRAPH_ARRAY_SIZE || + fgraph_array[gops->idx] != gops)) + goto out; + + if (fgraph_lru_release_index(gops->idx) < 0) + goto out; + + fgraph_array[gops->idx] = &fgraph_stub; + ftrace_graph_active--; - ftrace_graph_return = ftrace_stub_graph; - ftrace_graph_entry = ftrace_graph_entry_stub; - __ftrace_graph_entry = ftrace_graph_entry_stub; - ftrace_shutdown(&graph_ops, FTRACE_STOP_FUNC_RET); - unregister_pm_notifier(&ftrace_suspend_notifier); - unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); + if (!ftrace_graph_active) + command = FTRACE_STOP_FUNC_RET; + + ftrace_shutdown_subops(&graph_ops, &gops->ops, command); + + if (ftrace_graph_active == 1) + ftrace_graph_enable_direct(true, NULL); + else if (!ftrace_graph_active) + ftrace_graph_disable_direct(false); + + if (!ftrace_graph_active) { + ftrace_graph_return = ftrace_stub_graph; + ftrace_graph_entry = ftrace_graph_entry_stub; + unregister_pm_notifier(&ftrace_suspend_notifier); + unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); + } + gops->saved_func = NULL; out: - mutex_unlock(&ftrace_lock); + gops->ops.flags &= ~FTRACE_OPS_FL_GRAPH; } diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c index e8143e368074..1188eefef07c 100644 --- a/kernel/trace/fprobe.c +++ b/kernel/trace/fprobe.c @@ -8,82 +8,632 @@ #include <linux/fprobe.h> #include <linux/kallsyms.h> #include <linux/kprobes.h> -#include <linux/rethook.h> +#include <linux/list.h> +#include <linux/mutex.h> +#include <linux/rhashtable.h> #include <linux/slab.h> #include <linux/sort.h> +#include <asm/fprobe.h> + #include "trace.h" -struct fprobe_rethook_node { - struct rethook_node node; - unsigned long entry_ip; +#define FPROBE_IP_HASH_BITS 8 +#define FPROBE_IP_TABLE_SIZE (1 << FPROBE_IP_HASH_BITS) + +#define FPROBE_HASH_BITS 6 +#define FPROBE_TABLE_SIZE (1 << FPROBE_HASH_BITS) + +#define SIZE_IN_LONG(x) ((x + sizeof(long) - 1) >> (sizeof(long) == 8 ? 3 : 2)) + +/* + * fprobe_table: hold 'fprobe_hlist::hlist' for checking the fprobe still + * exists. The key is the address of fprobe instance. + * fprobe_ip_table: hold 'fprobe_hlist::array[*]' for searching the fprobe + * instance related to the function address. The key is the ftrace IP + * address. + * + * When unregistering the fprobe, fprobe_hlist::fp and fprobe_hlist::array[*].fp + * are set NULL and delete those from both hash tables (by hlist_del_rcu). + * After an RCU grace period, the fprobe_hlist itself will be released. + * + * fprobe_table and fprobe_ip_table can be accessed from either + * - Normal hlist traversal and RCU add/del under 'fprobe_mutex' is held. + * - RCU hlist traversal under disabling preempt + */ +static struct hlist_head fprobe_table[FPROBE_TABLE_SIZE]; +static struct rhltable fprobe_ip_table; +static DEFINE_MUTEX(fprobe_mutex); +static struct fgraph_ops fprobe_graph_ops; + +static u32 fprobe_node_hashfn(const void *data, u32 len, u32 seed) +{ + return hash_ptr(*(unsigned long **)data, 32); +} + +static int fprobe_node_cmp(struct rhashtable_compare_arg *arg, + const void *ptr) +{ + unsigned long key = *(unsigned long *)arg->key; + const struct fprobe_hlist_node *n = ptr; + + return n->addr != key; +} + +static u32 fprobe_node_obj_hashfn(const void *data, u32 len, u32 seed) +{ + const struct fprobe_hlist_node *n = data; + + return hash_ptr((void *)n->addr, 32); +} + +static const struct rhashtable_params fprobe_rht_params = { + .head_offset = offsetof(struct fprobe_hlist_node, hlist), + .key_offset = offsetof(struct fprobe_hlist_node, addr), + .key_len = sizeof_field(struct fprobe_hlist_node, addr), + .hashfn = fprobe_node_hashfn, + .obj_hashfn = fprobe_node_obj_hashfn, + .obj_cmpfn = fprobe_node_cmp, + .automatic_shrinking = true, }; -static void fprobe_handler(unsigned long ip, unsigned long parent_ip, - struct ftrace_ops *ops, struct ftrace_regs *fregs) +/* Node insertion and deletion requires the fprobe_mutex */ +static int insert_fprobe_node(struct fprobe_hlist_node *node) +{ + lockdep_assert_held(&fprobe_mutex); + + return rhltable_insert(&fprobe_ip_table, &node->hlist, fprobe_rht_params); +} + +/* Return true if there are synonims */ +static bool delete_fprobe_node(struct fprobe_hlist_node *node) { - struct fprobe_rethook_node *fpr; - struct rethook_node *rh; + lockdep_assert_held(&fprobe_mutex); + bool ret; + + /* Avoid double deleting */ + if (READ_ONCE(node->fp) != NULL) { + WRITE_ONCE(node->fp, NULL); + rhltable_remove(&fprobe_ip_table, &node->hlist, + fprobe_rht_params); + } + + rcu_read_lock(); + ret = !!rhltable_lookup(&fprobe_ip_table, &node->addr, + fprobe_rht_params); + rcu_read_unlock(); + + return ret; +} + +/* Check existence of the fprobe */ +static bool is_fprobe_still_exist(struct fprobe *fp) +{ + struct hlist_head *head; + struct fprobe_hlist *fph; + + head = &fprobe_table[hash_ptr(fp, FPROBE_HASH_BITS)]; + hlist_for_each_entry_rcu(fph, head, hlist, + lockdep_is_held(&fprobe_mutex)) { + if (fph->fp == fp) + return true; + } + return false; +} +NOKPROBE_SYMBOL(is_fprobe_still_exist); + +static int add_fprobe_hash(struct fprobe *fp) +{ + struct fprobe_hlist *fph = fp->hlist_array; + struct hlist_head *head; + + lockdep_assert_held(&fprobe_mutex); + + if (WARN_ON_ONCE(!fph)) + return -EINVAL; + + if (is_fprobe_still_exist(fp)) + return -EEXIST; + + head = &fprobe_table[hash_ptr(fp, FPROBE_HASH_BITS)]; + hlist_add_head_rcu(&fp->hlist_array->hlist, head); + return 0; +} + +static int del_fprobe_hash(struct fprobe *fp) +{ + struct fprobe_hlist *fph = fp->hlist_array; + + lockdep_assert_held(&fprobe_mutex); + + if (WARN_ON_ONCE(!fph)) + return -EINVAL; + + if (!is_fprobe_still_exist(fp)) + return -ENOENT; + + fph->fp = NULL; + hlist_del_rcu(&fph->hlist); + return 0; +} + +#ifdef ARCH_DEFINE_ENCODE_FPROBE_HEADER + +/* The arch should encode fprobe_header info into one unsigned long */ +#define FPROBE_HEADER_SIZE_IN_LONG 1 + +static inline bool write_fprobe_header(unsigned long *stack, + struct fprobe *fp, unsigned int size_words) +{ + if (WARN_ON_ONCE(size_words > MAX_FPROBE_DATA_SIZE_WORD || + !arch_fprobe_header_encodable(fp))) + return false; + + *stack = arch_encode_fprobe_header(fp, size_words); + return true; +} + +static inline void read_fprobe_header(unsigned long *stack, + struct fprobe **fp, unsigned int *size_words) +{ + *fp = arch_decode_fprobe_header_fp(*stack); + *size_words = arch_decode_fprobe_header_size(*stack); +} + +#else + +/* Generic fprobe_header */ +struct __fprobe_header { struct fprobe *fp; - int bit; + unsigned long size_words; +} __packed; - fp = container_of(ops, struct fprobe, ops); - if (fprobe_disabled(fp)) - return; +#define FPROBE_HEADER_SIZE_IN_LONG SIZE_IN_LONG(sizeof(struct __fprobe_header)) - bit = ftrace_test_recursion_trylock(ip, parent_ip); - if (bit < 0) { +static inline bool write_fprobe_header(unsigned long *stack, + struct fprobe *fp, unsigned int size_words) +{ + struct __fprobe_header *fph = (struct __fprobe_header *)stack; + + if (WARN_ON_ONCE(size_words > MAX_FPROBE_DATA_SIZE_WORD)) + return false; + + fph->fp = fp; + fph->size_words = size_words; + return true; +} + +static inline void read_fprobe_header(unsigned long *stack, + struct fprobe **fp, unsigned int *size_words) +{ + struct __fprobe_header *fph = (struct __fprobe_header *)stack; + + *fp = fph->fp; + *size_words = fph->size_words; +} + +#endif + +/* + * fprobe shadow stack management: + * Since fprobe shares a single fgraph_ops, it needs to share the stack entry + * among the probes on the same function exit. Note that a new probe can be + * registered before a target function is returning, we can not use the hash + * table to find the corresponding probes. Thus the probe address is stored on + * the shadow stack with its entry data size. + * + */ +static inline int __fprobe_handler(unsigned long ip, unsigned long parent_ip, + struct fprobe *fp, struct ftrace_regs *fregs, + void *data) +{ + if (!fp->entry_handler) + return 0; + + return fp->entry_handler(fp, ip, parent_ip, fregs, data); +} + +static inline int __fprobe_kprobe_handler(unsigned long ip, unsigned long parent_ip, + struct fprobe *fp, struct ftrace_regs *fregs, + void *data) +{ + int ret; + /* + * This user handler is shared with other kprobes and is not expected to be + * called recursively. So if any other kprobe handler is running, this will + * exit as kprobe does. See the section 'Share the callbacks with kprobes' + * in Documentation/trace/fprobe.rst for more information. + */ + if (unlikely(kprobe_running())) { fp->nmissed++; + return 0; + } + + kprobe_busy_begin(); + ret = __fprobe_handler(ip, parent_ip, fp, fregs, data); + kprobe_busy_end(); + return ret; +} + +#if defined(CONFIG_DYNAMIC_FTRACE_WITH_ARGS) || defined(CONFIG_DYNAMIC_FTRACE_WITH_REGS) +/* ftrace_ops callback, this processes fprobes which have only entry_handler. */ +static void fprobe_ftrace_entry(unsigned long ip, unsigned long parent_ip, + struct ftrace_ops *ops, struct ftrace_regs *fregs) +{ + struct fprobe_hlist_node *node; + struct rhlist_head *head, *pos; + struct fprobe *fp; + int bit; + + bit = ftrace_test_recursion_trylock(ip, parent_ip); + if (bit < 0) return; + + /* + * ftrace_test_recursion_trylock() disables preemption, but + * rhltable_lookup() checks whether rcu_read_lcok is held. + * So we take rcu_read_lock() here. + */ + rcu_read_lock(); + head = rhltable_lookup(&fprobe_ip_table, &ip, fprobe_rht_params); + + rhl_for_each_entry_rcu(node, pos, head, hlist) { + if (node->addr != ip) + break; + fp = READ_ONCE(node->fp); + if (unlikely(!fp || fprobe_disabled(fp) || fp->exit_handler)) + continue; + + if (fprobe_shared_with_kprobes(fp)) + __fprobe_kprobe_handler(ip, parent_ip, fp, fregs, NULL); + else + __fprobe_handler(ip, parent_ip, fp, fregs, NULL); } + rcu_read_unlock(); + ftrace_test_recursion_unlock(bit); +} +NOKPROBE_SYMBOL(fprobe_ftrace_entry); + +static struct ftrace_ops fprobe_ftrace_ops = { + .func = fprobe_ftrace_entry, + .flags = FTRACE_OPS_FL_SAVE_ARGS, +}; +static int fprobe_ftrace_active; + +static int fprobe_ftrace_add_ips(unsigned long *addrs, int num) +{ + int ret; + + lockdep_assert_held(&fprobe_mutex); - if (fp->entry_handler) - fp->entry_handler(fp, ip, ftrace_get_regs(fregs)); + ret = ftrace_set_filter_ips(&fprobe_ftrace_ops, addrs, num, 0, 0); + if (ret) + return ret; - if (fp->exit_handler) { - rh = rethook_try_get(fp->rethook); - if (!rh) { - fp->nmissed++; - goto out; + if (!fprobe_ftrace_active) { + ret = register_ftrace_function(&fprobe_ftrace_ops); + if (ret) { + ftrace_free_filter(&fprobe_ftrace_ops); + return ret; } - fpr = container_of(rh, struct fprobe_rethook_node, node); - fpr->entry_ip = ip; - rethook_hook(rh, ftrace_get_regs(fregs), true); } + fprobe_ftrace_active++; + return 0; +} -out: - ftrace_test_recursion_unlock(bit); +static void fprobe_ftrace_remove_ips(unsigned long *addrs, int num) +{ + lockdep_assert_held(&fprobe_mutex); + + fprobe_ftrace_active--; + if (!fprobe_ftrace_active) + unregister_ftrace_function(&fprobe_ftrace_ops); + if (num) + ftrace_set_filter_ips(&fprobe_ftrace_ops, addrs, num, 1, 0); } -NOKPROBE_SYMBOL(fprobe_handler); -static void fprobe_kprobe_handler(unsigned long ip, unsigned long parent_ip, - struct ftrace_ops *ops, struct ftrace_regs *fregs) +static bool fprobe_is_ftrace(struct fprobe *fp) { - struct fprobe *fp = container_of(ops, struct fprobe, ops); + return !fp->exit_handler; +} - if (unlikely(kprobe_running())) { - fp->nmissed++; +#ifdef CONFIG_MODULES +static void fprobe_set_ips(unsigned long *ips, unsigned int cnt, int remove, + int reset) +{ + ftrace_set_filter_ips(&fprobe_graph_ops.ops, ips, cnt, remove, reset); + ftrace_set_filter_ips(&fprobe_ftrace_ops, ips, cnt, remove, reset); +} +#endif +#else +static int fprobe_ftrace_add_ips(unsigned long *addrs, int num) +{ + return -ENOENT; +} + +static void fprobe_ftrace_remove_ips(unsigned long *addrs, int num) +{ +} + +static bool fprobe_is_ftrace(struct fprobe *fp) +{ + return false; +} + +#ifdef CONFIG_MODULES +static void fprobe_set_ips(unsigned long *ips, unsigned int cnt, int remove, + int reset) +{ + ftrace_set_filter_ips(&fprobe_graph_ops.ops, ips, cnt, remove, reset); +} +#endif +#endif /* !CONFIG_DYNAMIC_FTRACE_WITH_ARGS && !CONFIG_DYNAMIC_FTRACE_WITH_REGS */ + +/* fgraph_ops callback, this processes fprobes which have exit_handler. */ +static int fprobe_fgraph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + unsigned long *fgraph_data = NULL; + unsigned long func = trace->func; + struct fprobe_hlist_node *node; + struct rhlist_head *head, *pos; + unsigned long ret_ip; + int reserved_words; + struct fprobe *fp; + int used, ret; + + if (WARN_ON_ONCE(!fregs)) + return 0; + + guard(rcu)(); + head = rhltable_lookup(&fprobe_ip_table, &func, fprobe_rht_params); + reserved_words = 0; + rhl_for_each_entry_rcu(node, pos, head, hlist) { + if (node->addr != func) + continue; + fp = READ_ONCE(node->fp); + if (!fp || !fp->exit_handler) + continue; + /* + * Since fprobe can be enabled until the next loop, we ignore the + * fprobe's disabled flag in this loop. + */ + reserved_words += + FPROBE_HEADER_SIZE_IN_LONG + SIZE_IN_LONG(fp->entry_data_size); + } + if (reserved_words) { + fgraph_data = fgraph_reserve_data(gops->idx, reserved_words * sizeof(long)); + if (unlikely(!fgraph_data)) { + rhl_for_each_entry_rcu(node, pos, head, hlist) { + if (node->addr != func) + continue; + fp = READ_ONCE(node->fp); + if (fp && !fprobe_disabled(fp) && !fprobe_is_ftrace(fp)) + fp->nmissed++; + } + return 0; + } + } + + /* + * TODO: recursion detection has been done in the fgraph. Thus we need + * to add a callback to increment missed counter. + */ + ret_ip = ftrace_regs_get_return_address(fregs); + used = 0; + rhl_for_each_entry_rcu(node, pos, head, hlist) { + int data_size; + void *data; + + if (node->addr != func) + continue; + fp = READ_ONCE(node->fp); + if (unlikely(!fp || fprobe_disabled(fp) || fprobe_is_ftrace(fp))) + continue; + + data_size = fp->entry_data_size; + if (data_size && fp->exit_handler) + data = fgraph_data + used + FPROBE_HEADER_SIZE_IN_LONG; + else + data = NULL; + + if (fprobe_shared_with_kprobes(fp)) + ret = __fprobe_kprobe_handler(func, ret_ip, fp, fregs, data); + else + ret = __fprobe_handler(func, ret_ip, fp, fregs, data); + + /* If entry_handler returns !0, nmissed is not counted but skips exit_handler. */ + if (!ret && fp->exit_handler) { + int size_words = SIZE_IN_LONG(data_size); + + if (write_fprobe_header(&fgraph_data[used], fp, size_words)) + used += FPROBE_HEADER_SIZE_IN_LONG + size_words; + } + } + if (used < reserved_words) + memset(fgraph_data + used, 0, reserved_words - used); + + /* If any exit_handler is set, data must be used. */ + return used != 0; +} +NOKPROBE_SYMBOL(fprobe_fgraph_entry); + +static void fprobe_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + unsigned long *fgraph_data = NULL; + unsigned long ret_ip; + struct fprobe *fp; + int size, curr; + int size_words; + + fgraph_data = (unsigned long *)fgraph_retrieve_data(gops->idx, &size); + if (WARN_ON_ONCE(!fgraph_data)) return; + size_words = SIZE_IN_LONG(size); + ret_ip = ftrace_regs_get_instruction_pointer(fregs); + + preempt_disable_notrace(); + + curr = 0; + while (size_words > curr) { + read_fprobe_header(&fgraph_data[curr], &fp, &size); + if (!fp) + break; + curr += FPROBE_HEADER_SIZE_IN_LONG; + if (is_fprobe_still_exist(fp) && !fprobe_disabled(fp)) { + if (WARN_ON_ONCE(curr + size > size_words)) + break; + fp->exit_handler(fp, trace->func, ret_ip, fregs, + size ? fgraph_data + curr : NULL); + } + curr += size; } - kprobe_busy_begin(); - fprobe_handler(ip, parent_ip, ops, fregs); - kprobe_busy_end(); + preempt_enable_notrace(); +} +NOKPROBE_SYMBOL(fprobe_return); + +static struct fgraph_ops fprobe_graph_ops = { + .entryfunc = fprobe_fgraph_entry, + .retfunc = fprobe_return, +}; +static int fprobe_graph_active; + +/* Add @addrs to the ftrace filter and register fgraph if needed. */ +static int fprobe_graph_add_ips(unsigned long *addrs, int num) +{ + int ret; + + lockdep_assert_held(&fprobe_mutex); + + ret = ftrace_set_filter_ips(&fprobe_graph_ops.ops, addrs, num, 0, 0); + if (ret) + return ret; + + if (!fprobe_graph_active) { + ret = register_ftrace_graph(&fprobe_graph_ops); + if (WARN_ON_ONCE(ret)) { + ftrace_free_filter(&fprobe_graph_ops.ops); + return ret; + } + } + fprobe_graph_active++; + return 0; } -static void fprobe_exit_handler(struct rethook_node *rh, void *data, - struct pt_regs *regs) +/* Remove @addrs from the ftrace filter and unregister fgraph if possible. */ +static void fprobe_graph_remove_ips(unsigned long *addrs, int num) { - struct fprobe *fp = (struct fprobe *)data; - struct fprobe_rethook_node *fpr; + lockdep_assert_held(&fprobe_mutex); - if (!fp || fprobe_disabled(fp)) + fprobe_graph_active--; + /* Q: should we unregister it ? */ + if (!fprobe_graph_active) + unregister_ftrace_graph(&fprobe_graph_ops); + + if (num) + ftrace_set_filter_ips(&fprobe_graph_ops.ops, addrs, num, 1, 0); +} + +#ifdef CONFIG_MODULES + +#define FPROBE_IPS_BATCH_INIT 8 +/* instruction pointer address list */ +struct fprobe_addr_list { + int index; + int size; + unsigned long *addrs; +}; + +static int fprobe_addr_list_add(struct fprobe_addr_list *alist, unsigned long addr) +{ + unsigned long *addrs; + + /* Previously we failed to expand the list. */ + if (alist->index == alist->size) + return -ENOSPC; + + alist->addrs[alist->index++] = addr; + if (alist->index < alist->size) + return 0; + + /* Expand the address list */ + addrs = kcalloc(alist->size * 2, sizeof(*addrs), GFP_KERNEL); + if (!addrs) + return -ENOMEM; + + memcpy(addrs, alist->addrs, alist->size * sizeof(*addrs)); + alist->size *= 2; + kfree(alist->addrs); + alist->addrs = addrs; + + return 0; +} + +static void fprobe_remove_node_in_module(struct module *mod, struct fprobe_hlist_node *node, + struct fprobe_addr_list *alist) +{ + if (!within_module(node->addr, mod)) + return; + if (delete_fprobe_node(node)) return; + /* + * If failed to update alist, just continue to update hlist. + * Therefore, at list user handler will not hit anymore. + */ + fprobe_addr_list_add(alist, node->addr); +} + +/* Handle module unloading to manage fprobe_ip_table. */ +static int fprobe_module_callback(struct notifier_block *nb, + unsigned long val, void *data) +{ + struct fprobe_addr_list alist = {.size = FPROBE_IPS_BATCH_INIT}; + struct fprobe_hlist_node *node; + struct rhashtable_iter iter; + struct module *mod = data; + + if (val != MODULE_STATE_GOING) + return NOTIFY_DONE; + + alist.addrs = kcalloc(alist.size, sizeof(*alist.addrs), GFP_KERNEL); + /* If failed to alloc memory, we can not remove ips from hash. */ + if (!alist.addrs) + return NOTIFY_DONE; + + mutex_lock(&fprobe_mutex); + rhltable_walk_enter(&fprobe_ip_table, &iter); + do { + rhashtable_walk_start(&iter); + + while ((node = rhashtable_walk_next(&iter)) && !IS_ERR(node)) + fprobe_remove_node_in_module(mod, node, &alist); + + rhashtable_walk_stop(&iter); + } while (node == ERR_PTR(-EAGAIN)); + rhashtable_walk_exit(&iter); + + if (alist.index > 0) + fprobe_set_ips(alist.addrs, alist.index, 1, 0); + mutex_unlock(&fprobe_mutex); - fpr = container_of(rh, struct fprobe_rethook_node, node); + kfree(alist.addrs); - fp->exit_handler(fp, fpr->entry_ip, regs); + return NOTIFY_DONE; } -NOKPROBE_SYMBOL(fprobe_exit_handler); + +static struct notifier_block fprobe_module_nb = { + .notifier_call = fprobe_module_callback, + .priority = 0, +}; + +static int __init init_fprobe_module(void) +{ + return register_module_notifier(&fprobe_module_nb); +} +early_initcall(init_fprobe_module); +#endif static int symbols_cmp(const void *a, const void *b) { @@ -113,58 +663,118 @@ static unsigned long *get_ftrace_locations(const char **syms, int num) return ERR_PTR(-ENOENT); } -static void fprobe_init(struct fprobe *fp) +struct filter_match_data { + const char *filter; + const char *notfilter; + size_t index; + size_t size; + unsigned long *addrs; + struct module **mods; +}; + +static int filter_match_callback(void *data, const char *name, unsigned long addr) { - fp->nmissed = 0; - if (fprobe_shared_with_kprobes(fp)) - fp->ops.func = fprobe_kprobe_handler; - else - fp->ops.func = fprobe_handler; - fp->ops.flags |= FTRACE_OPS_FL_SAVE_REGS; + struct filter_match_data *match = data; + + if (!glob_match(match->filter, name) || + (match->notfilter && glob_match(match->notfilter, name))) + return 0; + + if (!ftrace_location(addr)) + return 0; + + if (match->addrs) { + struct module *mod = __module_text_address(addr); + + if (mod && !try_module_get(mod)) + return 0; + + match->mods[match->index] = mod; + match->addrs[match->index] = addr; + } + match->index++; + return match->index == match->size; } -static int fprobe_init_rethook(struct fprobe *fp, int num) +/* + * Make IP list from the filter/no-filter glob patterns. + * Return the number of matched symbols, or errno. + * If @addrs == NULL, this just counts the number of matched symbols. If @addrs + * is passed with an array, we need to pass the an @mods array of the same size + * to increment the module refcount for each symbol. + * This means we also need to call `module_put` for each element of @mods after + * using the @addrs. + */ +static int get_ips_from_filter(const char *filter, const char *notfilter, + unsigned long *addrs, struct module **mods, + size_t size) { - int i, size; + struct filter_match_data match = { .filter = filter, .notfilter = notfilter, + .index = 0, .size = size, .addrs = addrs, .mods = mods}; + int ret; - if (num < 0) + if (addrs && !mods) return -EINVAL; - if (!fp->exit_handler) { - fp->rethook = NULL; - return 0; + ret = kallsyms_on_each_symbol(filter_match_callback, &match); + if (ret < 0) + return ret; + if (IS_ENABLED(CONFIG_MODULES)) { + ret = module_kallsyms_on_each_symbol(NULL, filter_match_callback, &match); + if (ret < 0) + return ret; } - /* Initialize rethook if needed */ - size = num * num_possible_cpus() * 2; - if (size < 0) + return match.index ?: -ENOENT; +} + +static void fprobe_fail_cleanup(struct fprobe *fp) +{ + kfree(fp->hlist_array); + fp->hlist_array = NULL; +} + +/* Initialize the fprobe data structure. */ +static int fprobe_init(struct fprobe *fp, unsigned long *addrs, int num) +{ + struct fprobe_hlist *hlist_array; + unsigned long addr; + int size, i; + + if (!fp || !addrs || num <= 0) + return -EINVAL; + + size = ALIGN(fp->entry_data_size, sizeof(long)); + if (size > MAX_FPROBE_DATA_SIZE) return -E2BIG; + fp->entry_data_size = size; - fp->rethook = rethook_alloc((void *)fp, fprobe_exit_handler); - if (!fp->rethook) + hlist_array = kzalloc(struct_size(hlist_array, array, num), GFP_KERNEL); + if (!hlist_array) return -ENOMEM; - for (i = 0; i < size; i++) { - struct fprobe_rethook_node *node; - - node = kzalloc(sizeof(*node), GFP_KERNEL); - if (!node) { - rethook_free(fp->rethook); - fp->rethook = NULL; - return -ENOMEM; + + fp->nmissed = 0; + + hlist_array->size = num; + fp->hlist_array = hlist_array; + hlist_array->fp = fp; + for (i = 0; i < num; i++) { + hlist_array->array[i].fp = fp; + addr = ftrace_location(addrs[i]); + if (!addr) { + fprobe_fail_cleanup(fp); + return -ENOENT; } - rethook_add_node(fp->rethook, &node->node); + hlist_array->array[i].addr = addr; } return 0; } -static void fprobe_fail_cleanup(struct fprobe *fp) +#define FPROBE_IPS_MAX INT_MAX + +int fprobe_count_ips_from_filter(const char *filter, const char *notfilter) { - if (fp->rethook) { - /* Don't need to cleanup rethook->handler because this is not used. */ - rethook_free(fp->rethook); - fp->rethook = NULL; - } - ftrace_free_filter(&fp->ops); + return get_ips_from_filter(filter, notfilter, NULL, NULL, FPROBE_IPS_MAX); } /** @@ -180,46 +790,35 @@ static void fprobe_fail_cleanup(struct fprobe *fp) */ int register_fprobe(struct fprobe *fp, const char *filter, const char *notfilter) { - struct ftrace_hash *hash; - unsigned char *str; - int ret, len; + unsigned long *addrs __free(kfree) = NULL; + struct module **mods __free(kfree) = NULL; + int ret, num; if (!fp || !filter) return -EINVAL; - fprobe_init(fp); + num = get_ips_from_filter(filter, notfilter, NULL, NULL, FPROBE_IPS_MAX); + if (num < 0) + return num; - len = strlen(filter); - str = kstrdup(filter, GFP_KERNEL); - ret = ftrace_set_filter(&fp->ops, str, len, 0); - kfree(str); - if (ret) - return ret; + addrs = kcalloc(num, sizeof(*addrs), GFP_KERNEL); + if (!addrs) + return -ENOMEM; - if (notfilter) { - len = strlen(notfilter); - str = kstrdup(notfilter, GFP_KERNEL); - ret = ftrace_set_notrace(&fp->ops, str, len, 0); - kfree(str); - if (ret) - goto out; - } + mods = kcalloc(num, sizeof(*mods), GFP_KERNEL); + if (!mods) + return -ENOMEM; - /* TODO: - * correctly calculate the total number of filtered symbols - * from both filter and notfilter. - */ - hash = rcu_access_pointer(fp->ops.local_hash.filter_hash); - if (WARN_ON_ONCE(!hash)) - goto out; + ret = get_ips_from_filter(filter, notfilter, addrs, mods, num); + if (ret < 0) + return ret; - ret = fprobe_init_rethook(fp, (int)hash->count); - if (!ret) - ret = register_ftrace_function(&fp->ops); + ret = register_fprobe_ips(fp, addrs, ret); -out: - if (ret) - fprobe_fail_cleanup(fp); + for (int i = 0; i < num; i++) { + if (mods[i]) + module_put(mods[i]); + } return ret; } EXPORT_SYMBOL_GPL(register_fprobe); @@ -227,7 +826,7 @@ EXPORT_SYMBOL_GPL(register_fprobe); /** * register_fprobe_ips() - Register fprobe to ftrace by address. * @fp: A fprobe data structure to be registered. - * @addrs: An array of target ftrace location addresses. + * @addrs: An array of target function address. * @num: The number of entries of @addrs. * * Register @fp to ftrace for enabling the probe on the address given by @addrs. @@ -239,23 +838,39 @@ EXPORT_SYMBOL_GPL(register_fprobe); */ int register_fprobe_ips(struct fprobe *fp, unsigned long *addrs, int num) { - int ret; - - if (!fp || !addrs || num <= 0) - return -EINVAL; - - fprobe_init(fp); + struct fprobe_hlist *hlist_array; + int ret, i; - ret = ftrace_set_filter_ips(&fp->ops, addrs, num, 0, 0); + ret = fprobe_init(fp, addrs, num); if (ret) return ret; - ret = fprobe_init_rethook(fp, num); - if (!ret) - ret = register_ftrace_function(&fp->ops); + mutex_lock(&fprobe_mutex); + + hlist_array = fp->hlist_array; + if (fprobe_is_ftrace(fp)) + ret = fprobe_ftrace_add_ips(addrs, num); + else + ret = fprobe_graph_add_ips(addrs, num); + + if (!ret) { + add_fprobe_hash(fp); + for (i = 0; i < hlist_array->size; i++) { + ret = insert_fprobe_node(&hlist_array->array[i]); + if (ret) + break; + } + /* fallback on insert error */ + if (ret) { + for (i--; i >= 0; i--) + delete_fprobe_node(&hlist_array->array[i]); + } + } + mutex_unlock(&fprobe_mutex); if (ret) fprobe_fail_cleanup(fp); + return ret; } EXPORT_SYMBOL_GPL(register_fprobe_ips); @@ -291,8 +906,15 @@ int register_fprobe_syms(struct fprobe *fp, const char **syms, int num) } EXPORT_SYMBOL_GPL(register_fprobe_syms); +bool fprobe_is_registered(struct fprobe *fp) +{ + if (!fp || !fp->hlist_array) + return false; + return true; +} + /** - * unregister_fprobe() - Unregister fprobe from ftrace + * unregister_fprobe() - Unregister fprobe. * @fp: A fprobe data structure to be unregistered. * * Unregister fprobe (and remove ftrace hooks from the function entries). @@ -301,27 +923,50 @@ EXPORT_SYMBOL_GPL(register_fprobe_syms); */ int unregister_fprobe(struct fprobe *fp) { - int ret; + struct fprobe_hlist *hlist_array; + unsigned long *addrs = NULL; + int ret = 0, i, count; - if (!fp || (fp->ops.saved_func != fprobe_handler && - fp->ops.saved_func != fprobe_kprobe_handler)) - return -EINVAL; + mutex_lock(&fprobe_mutex); + if (!fp || !is_fprobe_still_exist(fp)) { + ret = -EINVAL; + goto out; + } - /* - * rethook_free() starts disabling the rethook, but the rethook handlers - * may be running on other processors at this point. To make sure that all - * current running handlers are finished, call unregister_ftrace_function() - * after this. - */ - if (fp->rethook) - rethook_free(fp->rethook); + hlist_array = fp->hlist_array; + addrs = kcalloc(hlist_array->size, sizeof(unsigned long), GFP_KERNEL); + if (!addrs) { + ret = -ENOMEM; /* TODO: Fallback to one-by-one loop */ + goto out; + } - ret = unregister_ftrace_function(&fp->ops); - if (ret < 0) - return ret; + /* Remove non-synonim ips from table and hash */ + count = 0; + for (i = 0; i < hlist_array->size; i++) { + if (!delete_fprobe_node(&hlist_array->array[i])) + addrs[count++] = hlist_array->array[i].addr; + } + del_fprobe_hash(fp); - ftrace_free_filter(&fp->ops); + if (fprobe_is_ftrace(fp)) + fprobe_ftrace_remove_ips(addrs, count); + else + fprobe_graph_remove_ips(addrs, count); + + kfree_rcu(hlist_array, rcu); + fp->hlist_array = NULL; +out: + mutex_unlock(&fprobe_mutex); + + kfree(addrs); return ret; } EXPORT_SYMBOL_GPL(unregister_fprobe); + +static int __init fprobe_initcall(void) +{ + rhltable_init(&fprobe_ip_table, &fprobe_rht_params); + return 0; +} +core_initcall(fprobe_initcall); diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 750aa3f08b25..3ec2033c0774 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -45,6 +45,10 @@ #include "trace_output.h" #include "trace_stat.h" +/* Flags that do not get reset */ +#define FTRACE_NOCLEAR_FLAGS (FTRACE_FL_DISABLED | FTRACE_FL_TOUCHED | \ + FTRACE_FL_MODIFIED) + #define FTRACE_INVALID_FUNCTION "__ftrace_invalid_address__" #define FTRACE_WARN_ON(cond) \ @@ -70,7 +74,8 @@ #ifdef CONFIG_DYNAMIC_FTRACE #define INIT_OPS_HASH(opsname) \ .func_hash = &opsname.local_hash, \ - .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), + .local_hash.regex_lock = __MUTEX_INITIALIZER(opsname.local_hash.regex_lock), \ + .subop_list = LIST_HEAD_INIT(opsname.subop_list), #else #define INIT_OPS_HASH(opsname) #endif @@ -95,7 +100,7 @@ struct ftrace_ops *function_trace_op __read_mostly = &ftrace_list_end; /* What to set function_trace_op to */ static struct ftrace_ops *set_function_trace_op; -static bool ftrace_pids_enabled(struct ftrace_ops *ops) +bool ftrace_pids_enabled(struct ftrace_ops *ops) { struct trace_array *tr; @@ -117,7 +122,7 @@ static int ftrace_disabled __read_mostly; DEFINE_MUTEX(ftrace_lock); -struct ftrace_ops __rcu *ftrace_ops_list __read_mostly = &ftrace_list_end; +struct ftrace_ops __rcu *ftrace_ops_list __read_mostly = (struct ftrace_ops __rcu *)&ftrace_list_end; ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; struct ftrace_ops global_ops; @@ -125,17 +130,46 @@ struct ftrace_ops global_ops; void ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs); +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_CALL_OPS +/* + * Stub used to invoke the list ops without requiring a separate trampoline. + */ +const struct ftrace_ops ftrace_list_ops = { + .func = ftrace_ops_list_func, + .flags = FTRACE_OPS_FL_STUB, +}; + +static void ftrace_ops_nop_func(unsigned long ip, unsigned long parent_ip, + struct ftrace_ops *op, + struct ftrace_regs *fregs) +{ + /* do nothing */ +} + +/* + * Stub used when a call site is disabled. May be called transiently by threads + * which have made it into ftrace_caller but haven't yet recovered the ops at + * the point the call site is disabled. + */ +const struct ftrace_ops ftrace_nop_ops = { + .func = ftrace_ops_nop_func, + .flags = FTRACE_OPS_FL_STUB, +}; +#endif + static inline void ftrace_ops_init(struct ftrace_ops *ops) { #ifdef CONFIG_DYNAMIC_FTRACE if (!(ops->flags & FTRACE_OPS_FL_INITIALIZED)) { mutex_init(&ops->local_hash.regex_lock); + INIT_LIST_HEAD(&ops->subop_list); ops->func_hash = &ops->local_hash; ops->flags |= FTRACE_OPS_FL_INITIALIZED; } #endif } +/* Call this function for when a callback filters on set_ftrace_pid */ static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs) { @@ -154,7 +188,7 @@ static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip, op->saved_func(ip, parent_ip, op, fregs); } -static void ftrace_sync_ipi(void *data) +void ftrace_sync_ipi(void *data) { /* Probably not needed, but do it anyway */ smp_rmb(); @@ -204,8 +238,6 @@ static void update_ftrace_function(void) func = ftrace_ops_list_func; } - update_function_graph_func(); - /* If there's no change, then do nothing more here */ if (ftrace_trace_function == func) return; @@ -279,7 +311,7 @@ static int remove_ftrace_ops(struct ftrace_ops __rcu **list, lockdep_is_held(&ftrace_lock)) == ops && rcu_dereference_protected(ops->next, lockdep_is_held(&ftrace_lock)) == &ftrace_list_end) { - *list = &ftrace_list_end; + rcu_assign_pointer(*list, &ftrace_list_end); return 0; } @@ -375,6 +407,8 @@ static void ftrace_update_pid_func(void) } } while_for_each_ftrace_op(op); + fgraph_update_pid_func(); + update_ftrace_function(); } @@ -500,51 +534,69 @@ static int function_stat_headers(struct seq_file *m) static int function_stat_show(struct seq_file *m, void *v) { + struct trace_array *tr = trace_get_global_array(); struct ftrace_profile *rec = v; + const char *refsymbol = NULL; char str[KSYM_SYMBOL_LEN]; - int ret = 0; #ifdef CONFIG_FUNCTION_GRAPH_TRACER static struct trace_seq s; unsigned long long avg; unsigned long long stddev; + unsigned long long stddev_denom; #endif - mutex_lock(&ftrace_profile_lock); + guard(mutex)(&ftrace_profile_lock); /* we raced with function_profile_reset() */ - if (unlikely(rec->counter == 0)) { - ret = -EBUSY; - goto out; - } + if (unlikely(rec->counter == 0)) + return -EBUSY; #ifdef CONFIG_FUNCTION_GRAPH_TRACER avg = div64_ul(rec->time, rec->counter); if (tracing_thresh && (avg < tracing_thresh)) - goto out; + return 0; #endif - kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); + if (tr->trace_flags & TRACE_ITER(PROF_TEXT_OFFSET)) { + unsigned long offset; + + if (core_kernel_text(rec->ip)) { + refsymbol = "_text"; + offset = rec->ip - (unsigned long)_text; + } else { + struct module *mod; + + guard(rcu)(); + mod = __module_text_address(rec->ip); + if (mod) { + refsymbol = mod->name; + /* Calculate offset from module's text entry address. */ + offset = rec->ip - (unsigned long)mod->mem[MOD_TEXT].base; + } + } + if (refsymbol) + snprintf(str, sizeof(str), " %s+%#lx", refsymbol, offset); + } + if (!refsymbol) + kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); + seq_printf(m, " %-30.30s %10lu", str, rec->counter); #ifdef CONFIG_FUNCTION_GRAPH_TRACER seq_puts(m, " "); - /* Sample standard deviation (s^2) */ - if (rec->counter <= 1) - stddev = 0; - else { - /* - * Apply Welford's method: - * s^2 = 1 / (n * (n-1)) * (n * \Sum (x_i)^2 - (\Sum x_i)^2) - */ + /* + * Variance formula: + * s^2 = 1 / (n * (n-1)) * (n * \Sum (x_i)^2 - (\Sum x_i)^2) + * Maybe Welford's method is better here? + * Divide only by 1000 for ns^2 -> us^2 conversion. + * trace_print_graph_duration will divide by 1000 again. + */ + stddev = 0; + stddev_denom = rec->counter * (rec->counter - 1) * 1000; + if (stddev_denom) { stddev = rec->counter * rec->time_squared - rec->time * rec->time; - - /* - * Divide only 1000 for ns^2 -> us^2 conversion. - * trace_print_graph_duration will divide 1000 again. - */ - stddev = div64_ul(stddev, - rec->counter * (rec->counter - 1) * 1000); + stddev = div64_ul(stddev, stddev_denom); } trace_seq_init(&s); @@ -556,10 +608,8 @@ static int function_stat_show(struct seq_file *m, void *v) trace_print_seq(m, &s); #endif seq_putc(m, '\n'); -out: - mutex_unlock(&ftrace_profile_lock); - return ret; + return 0; } static void ftrace_profile_reset(struct ftrace_profile_stat *stat) @@ -755,27 +805,24 @@ function_profile_call(unsigned long ip, unsigned long parent_ip, { struct ftrace_profile_stat *stat; struct ftrace_profile *rec; - unsigned long flags; if (!ftrace_profile_enabled) return; - local_irq_save(flags); + guard(preempt_notrace)(); stat = this_cpu_ptr(&ftrace_profile_stats); if (!stat->hash || !ftrace_profile_enabled) - goto out; + return; rec = ftrace_find_profiled_func(stat, ip); if (!rec) { rec = ftrace_profile_alloc(stat, ip); if (!rec) - goto out; + return; } rec->counter++; - out: - local_irq_restore(flags); } #ifdef CONFIG_FUNCTION_GRAPH_TRACER @@ -786,9 +833,17 @@ void ftrace_graph_graph_time_control(bool enable) fgraph_graph_time = enable; } -static int profile_graph_entry(struct ftrace_graph_ent *trace) +struct profile_fgraph_data { + unsigned long long calltime; + unsigned long long subtime; + unsigned long long sleeptime; +}; + +static int profile_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { - struct ftrace_ret_stack *ret_stack; + struct profile_fgraph_data *profile_data; function_profile_call(trace->func, 0, NULL, NULL); @@ -796,42 +851,59 @@ static int profile_graph_entry(struct ftrace_graph_ent *trace) if (!current->ret_stack) return 0; - ret_stack = ftrace_graph_get_ret_stack(current, 0); - if (ret_stack) - ret_stack->subtime = 0; + profile_data = fgraph_reserve_data(gops->idx, sizeof(*profile_data)); + if (!profile_data) + return 0; + + profile_data->subtime = 0; + profile_data->sleeptime = current->ftrace_sleeptime; + profile_data->calltime = trace_clock_local(); return 1; } -static void profile_graph_return(struct ftrace_graph_ret *trace) +bool fprofile_no_sleep_time; + +static void profile_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { - struct ftrace_ret_stack *ret_stack; + struct profile_fgraph_data *profile_data; struct ftrace_profile_stat *stat; unsigned long long calltime; + unsigned long long rettime = trace_clock_local(); struct ftrace_profile *rec; - unsigned long flags; + int size; + + guard(preempt_notrace)(); - local_irq_save(flags); stat = this_cpu_ptr(&ftrace_profile_stats); if (!stat->hash || !ftrace_profile_enabled) - goto out; + return; + + profile_data = fgraph_retrieve_data(gops->idx, &size); /* If the calltime was zero'd ignore it */ - if (!trace->calltime) - goto out; + if (!profile_data || !profile_data->calltime) + return; - calltime = trace->rettime - trace->calltime; + calltime = rettime - profile_data->calltime; + + if (fprofile_no_sleep_time) { + if (current->ftrace_sleeptime) + calltime -= current->ftrace_sleeptime - profile_data->sleeptime; + } if (!fgraph_graph_time) { + struct profile_fgraph_data *parent_data; /* Append this call time to the parent time to subtract */ - ret_stack = ftrace_graph_get_ret_stack(current, 1); - if (ret_stack) - ret_stack->subtime += calltime; + parent_data = fgraph_retrieve_parent_data(gops->idx, &size, 1); + if (parent_data) + parent_data->subtime += calltime; - ret_stack = ftrace_graph_get_ret_stack(current, 0); - if (ret_stack && ret_stack->subtime < calltime) - calltime -= ret_stack->subtime; + if (profile_data->subtime && profile_data->subtime < calltime) + calltime -= profile_data->subtime; else calltime = 0; } @@ -841,9 +913,6 @@ static void profile_graph_return(struct ftrace_graph_ret *trace) rec->time += calltime; rec->time_squared += calltime * calltime; } - - out: - local_irq_restore(flags); } static struct fgraph_ops fprofiler_ops = { @@ -853,6 +922,7 @@ static struct fgraph_ops fprofiler_ops = { static int register_ftrace_profiler(void) { + ftrace_ops_set_global_filter(&fprofiler_ops.ops); return register_ftrace_graph(&fprofiler_ops); } @@ -863,12 +933,11 @@ static void unregister_ftrace_profiler(void) #else static struct ftrace_ops ftrace_profile_ops __read_mostly = { .func = function_profile_call, - .flags = FTRACE_OPS_FL_INITIALIZED, - INIT_OPS_HASH(ftrace_profile_ops) }; static int register_ftrace_profiler(void) { + ftrace_ops_set_global_filter(&ftrace_profile_ops); return register_ftrace_function(&ftrace_profile_ops); } @@ -891,20 +960,16 @@ ftrace_profile_write(struct file *filp, const char __user *ubuf, val = !!val; - mutex_lock(&ftrace_profile_lock); + guard(mutex)(&ftrace_profile_lock); if (ftrace_profile_enabled ^ val) { if (val) { ret = ftrace_profile_init(); - if (ret < 0) { - cnt = ret; - goto out; - } + if (ret < 0) + return ret; ret = register_ftrace_profiler(); - if (ret < 0) { - cnt = ret; - goto out; - } + if (ret < 0) + return ret; ftrace_profile_enabled = 1; } else { ftrace_profile_enabled = 0; @@ -915,8 +980,6 @@ ftrace_profile_write(struct file *filp, const char __user *ubuf, unregister_ftrace_profiler(); } } - out: - mutex_unlock(&ftrace_profile_lock); *ppos += cnt; @@ -1005,10 +1068,6 @@ static struct ftrace_ops *removed_ops; */ static bool update_all_ops; -#ifndef CONFIG_FTRACE_MCOUNT_RECORD -# error Dynamic ftrace depends on MCOUNT_RECORD -#endif - struct ftrace_func_probe { struct ftrace_probe_ops *probe_ops; struct ftrace_ops ops; @@ -1129,7 +1188,7 @@ __ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) * Search a given @hash to see if a given instruction pointer (@ip) * exists in it. * - * Returns the entry that holds the @ip if found. NULL otherwise. + * Returns: the entry that holds the @ip if found. NULL otherwise. */ struct ftrace_func_entry * ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) @@ -1152,18 +1211,19 @@ static void __add_hash_entry(struct ftrace_hash *hash, hash->count++; } -static int add_hash_entry(struct ftrace_hash *hash, unsigned long ip) +static struct ftrace_func_entry * +add_hash_entry(struct ftrace_hash *hash, unsigned long ip) { struct ftrace_func_entry *entry; entry = kmalloc(sizeof(*entry), GFP_KERNEL); if (!entry) - return -ENOMEM; + return NULL; entry->ip = ip; __add_hash_entry(hash, entry); - return 0; + return entry; } static void @@ -1250,13 +1310,17 @@ static void free_ftrace_hash_rcu(struct ftrace_hash *hash) /** * ftrace_free_filter - remove all filters for an ftrace_ops - * @ops - the ops to remove the filters from + * @ops: the ops to remove the filters from */ void ftrace_free_filter(struct ftrace_ops *ops) { ftrace_ops_init(ops); + if (WARN_ON(ops->flags & FTRACE_OPS_FL_ENABLED)) + return; free_ftrace_hash(ops->func_hash->filter_hash); free_ftrace_hash(ops->func_hash->notrace_hash); + ops->func_hash->filter_hash = EMPTY_HASH; + ops->func_hash->notrace_hash = EMPTY_HASH; } EXPORT_SYMBOL_GPL(ftrace_free_filter); @@ -1282,7 +1346,7 @@ static struct ftrace_hash *alloc_ftrace_hash(int size_bits) return hash; } - +/* Used to save filters on functions for modules not loaded yet */ static int ftrace_add_mod(struct trace_array *tr, const char *func, const char *module, int enable) @@ -1318,7 +1382,6 @@ alloc_and_copy_ftrace_hash(int size_bits, struct ftrace_hash *hash) struct ftrace_func_entry *entry; struct ftrace_hash *new_hash; int size; - int ret; int i; new_hash = alloc_ftrace_hash(size_bits); @@ -1335,8 +1398,7 @@ alloc_and_copy_ftrace_hash(int size_bits, struct ftrace_hash *hash) size = 1 << hash->size_bits; for (i = 0; i < size; i++) { hlist_for_each_entry(entry, &hash->buckets[i], hlist) { - ret = add_hash_entry(new_hash, entry->ip); - if (ret < 0) + if (add_hash_entry(new_hash, entry->ip) == NULL) goto free_hash; } } @@ -1350,15 +1412,17 @@ alloc_and_copy_ftrace_hash(int size_bits, struct ftrace_hash *hash) return NULL; } -static void -ftrace_hash_rec_disable_modify(struct ftrace_ops *ops, int filter_hash); -static void -ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, int filter_hash); +static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops); +static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops); static int ftrace_hash_ipmodify_update(struct ftrace_ops *ops, struct ftrace_hash *new_hash); -static struct ftrace_hash *dup_hash(struct ftrace_hash *src, int size) +/* + * Allocate a new hash and remove entries from @src and move them to the new hash. + * On success, the @src hash will be empty and should be freed. + */ +static struct ftrace_hash *__move_hash(struct ftrace_hash *src, int size) { struct ftrace_func_entry *entry; struct ftrace_hash *new_hash; @@ -1394,6 +1458,7 @@ static struct ftrace_hash *dup_hash(struct ftrace_hash *src, int size) return new_hash; } +/* Move the @src entries to a newly allocated hash */ static struct ftrace_hash * __ftrace_hash_move(struct ftrace_hash *src) { @@ -1405,9 +1470,29 @@ __ftrace_hash_move(struct ftrace_hash *src) if (ftrace_hash_empty(src)) return EMPTY_HASH; - return dup_hash(src, size); + return __move_hash(src, size); } +/** + * ftrace_hash_move - move a new hash to a filter and do updates + * @ops: The ops with the hash that @dst points to + * @enable: True if for the filter hash, false for the notrace hash + * @dst: Points to the @ops hash that should be updated + * @src: The hash to update @dst with + * + * This is called when an ftrace_ops hash is being updated and the + * the kernel needs to reflect this. Note, this only updates the kernel + * function callbacks if the @ops is enabled (not to be confused with + * @enable above). If the @ops is enabled, its hash determines what + * callbacks get called. This function gets called when the @ops hash + * is updated and it requires new callbacks. + * + * On success the elements of @src is moved to @dst, and @dst is updated + * properly, as well as the functions determined by the @ops hashes + * are now calling the @ops callback function. + * + * Regardless of return type, @src should be freed with free_ftrace_hash(). + */ static int ftrace_hash_move(struct ftrace_ops *ops, int enable, struct ftrace_hash **dst, struct ftrace_hash *src) @@ -1437,11 +1522,11 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, * Remove the current set, update the hash and add * them back. */ - ftrace_hash_rec_disable_modify(ops, enable); + ftrace_hash_rec_disable_modify(ops); rcu_assign_pointer(*dst, new_hash); - ftrace_hash_rec_enable_modify(ops, enable); + ftrace_hash_rec_enable_modify(ops); return 0; } @@ -1537,7 +1622,8 @@ static struct dyn_ftrace *lookup_rec(unsigned long start, unsigned long end) key.flags = end; /* overload flags, as it is unsigned long */ for (pg = ftrace_pages_start; pg; pg = pg->next) { - if (end < pg->records[0].ip || + if (pg->index == 0 || + end < pg->records[0].ip || start >= (pg->records[pg->index - 1].ip + MCOUNT_INSN_SIZE)) continue; rec = bsearch(&key, pg->records, pg->index, @@ -1556,7 +1642,7 @@ static struct dyn_ftrace *lookup_rec(unsigned long start, unsigned long end) * @end: end of range to search (inclusive). @end points to the last byte * to check. * - * Returns rec->ip if the related ftrace location is a least partly within + * Returns: rec->ip if the related ftrace location is a least partly within * the given address range. That is, the first address of the instruction * that is either a NOP or call to the function tracer. It checks the ftrace * internal tables to determine if the address belongs or not. @@ -1564,43 +1650,42 @@ static struct dyn_ftrace *lookup_rec(unsigned long start, unsigned long end) unsigned long ftrace_location_range(unsigned long start, unsigned long end) { struct dyn_ftrace *rec; + unsigned long ip = 0; + rcu_read_lock(); rec = lookup_rec(start, end); if (rec) - return rec->ip; + ip = rec->ip; + rcu_read_unlock(); - return 0; + return ip; } /** * ftrace_location - return the ftrace location * @ip: the instruction pointer to check * - * If @ip matches the ftrace location, return @ip. - * If @ip matches sym+0, return sym's ftrace location. - * Otherwise, return 0. + * Returns: + * * If @ip matches the ftrace location, return @ip. + * * If @ip matches sym+0, return sym's ftrace location. + * * Otherwise, return 0. */ unsigned long ftrace_location(unsigned long ip) { - struct dyn_ftrace *rec; + unsigned long loc; unsigned long offset; unsigned long size; - rec = lookup_rec(ip, ip); - if (!rec) { + loc = ftrace_location_range(ip, ip); + if (!loc) { if (!kallsyms_lookup_size_offset(ip, &size, &offset)) - goto out; + return 0; /* map sym+0 to __fentry__ */ if (!offset) - rec = lookup_rec(ip, ip + size - 1); + loc = ftrace_location_range(ip, ip + size - 1); } - - if (rec) - return rec->ip; - -out: - return 0; + return loc; } /** @@ -1608,7 +1693,7 @@ out: * @start: start of range to search * @end: end of range to search (inclusive). @end points to the last byte to check. * - * Returns 1 if @start and @end contains a ftrace location. + * Returns: 1 if @start and @end contains a ftrace location. * That is, the instruction that is either a NOP or call to * the function tracer. It checks the ftrace internal tables to * determine if the address belongs or not. @@ -1662,12 +1747,21 @@ static bool skip_record(struct dyn_ftrace *rec) !(rec->flags & FTRACE_FL_ENABLED); } +/* + * This is the main engine to the ftrace updates to the dyn_ftrace records. + * + * It will iterate through all the available ftrace functions + * (the ones that ftrace can have callbacks to) and set the flags + * in the associated dyn_ftrace records. + * + * @inc: If true, the functions associated to @ops are added to + * the dyn_ftrace records, otherwise they are removed. + */ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, - int filter_hash, bool inc) { struct ftrace_hash *hash; - struct ftrace_hash *other_hash; + struct ftrace_hash *notrace_hash; struct ftrace_page *pg; struct dyn_ftrace *rec; bool update = false; @@ -1679,35 +1773,16 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, return false; /* - * In the filter_hash case: * If the count is zero, we update all records. * Otherwise we just update the items in the hash. - * - * In the notrace_hash case: - * We enable the update in the hash. - * As disabling notrace means enabling the tracing, - * and enabling notrace means disabling, the inc variable - * gets inversed. */ - if (filter_hash) { - hash = ops->func_hash->filter_hash; - other_hash = ops->func_hash->notrace_hash; - if (ftrace_hash_empty(hash)) - all = true; - } else { - inc = !inc; - hash = ops->func_hash->notrace_hash; - other_hash = ops->func_hash->filter_hash; - /* - * If the notrace hash has no items, - * then there's nothing to do. - */ - if (ftrace_hash_empty(hash)) - return false; - } + hash = ops->func_hash->filter_hash; + notrace_hash = ops->func_hash->notrace_hash; + if (ftrace_hash_empty(hash)) + all = true; do_for_each_ftrace_rec(pg, rec) { - int in_other_hash = 0; + int in_notrace_hash = 0; int in_hash = 0; int match = 0; @@ -1719,26 +1794,17 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, * Only the filter_hash affects all records. * Update if the record is not in the notrace hash. */ - if (!other_hash || !ftrace_lookup_ip(other_hash, rec->ip)) + if (!notrace_hash || !ftrace_lookup_ip(notrace_hash, rec->ip)) match = 1; } else { in_hash = !!ftrace_lookup_ip(hash, rec->ip); - in_other_hash = !!ftrace_lookup_ip(other_hash, rec->ip); + in_notrace_hash = !!ftrace_lookup_ip(notrace_hash, rec->ip); /* - * If filter_hash is set, we want to match all functions - * that are in the hash but not in the other hash. - * - * If filter_hash is not set, then we are decrementing. - * That means we match anything that is in the hash - * and also in the other_hash. That is, we need to turn - * off functions in the other hash because they are disabled - * by this hash. + * We want to match all functions that are in the hash but + * not in the other hash. */ - if (filter_hash && in_hash && !in_other_hash) - match = 1; - else if (!filter_hash && in_hash && - (in_other_hash || ftrace_hash_empty(other_hash))) + if (in_hash && !in_notrace_hash) match = 1; } if (!match) @@ -1819,6 +1885,18 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, * if rec count is zero. */ } + + /* + * If the rec has a single associated ops, and ops->func can be + * called directly, allow the call site to call via the ops. + */ + if (IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_CALL_OPS) && + ftrace_rec_count(rec) == 1 && + ftrace_ops_get_func(ops) == ops->func) + rec->flags |= FTRACE_FL_CALL_OPS; + else + rec->flags &= ~FTRACE_FL_CALL_OPS; + count++; /* Must match FTRACE_UPDATE_CALLS in ftrace_modify_all_code() */ @@ -1832,24 +1910,48 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, return update; } -static bool ftrace_hash_rec_disable(struct ftrace_ops *ops, - int filter_hash) +/* + * This is called when an ops is removed from tracing. It will decrement + * the counters of the dyn_ftrace records for all the functions that + * the @ops attached to. + */ +static bool ftrace_hash_rec_disable(struct ftrace_ops *ops) { - return __ftrace_hash_rec_update(ops, filter_hash, 0); + return __ftrace_hash_rec_update(ops, false); } -static bool ftrace_hash_rec_enable(struct ftrace_ops *ops, - int filter_hash) +/* + * This is called when an ops is added to tracing. It will increment + * the counters of the dyn_ftrace records for all the functions that + * the @ops attached to. + */ +static bool ftrace_hash_rec_enable(struct ftrace_ops *ops) { - return __ftrace_hash_rec_update(ops, filter_hash, 1); + return __ftrace_hash_rec_update(ops, true); } -static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, - int filter_hash, int inc) +/* + * This function will update what functions @ops traces when its filter + * changes. + * + * The @inc states if the @ops callbacks are going to be added or removed. + * When one of the @ops hashes are updated to a "new_hash" the dyn_ftrace + * records are update via: + * + * ftrace_hash_rec_disable_modify(ops); + * ops->hash = new_hash + * ftrace_hash_rec_enable_modify(ops); + * + * Where the @ops is removed from all the records it is tracing using + * its old hash. The @ops hash is updated to the new hash, and then + * the @ops is added back to the records so that it is tracing all + * the new functions. + */ +static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, bool inc) { struct ftrace_ops *op; - __ftrace_hash_rec_update(ops, filter_hash, inc); + __ftrace_hash_rec_update(ops, inc); if (ops->func_hash != &global_ops.local_hash) return; @@ -1863,20 +1965,18 @@ static void ftrace_hash_rec_update_modify(struct ftrace_ops *ops, if (op == ops) continue; if (op->func_hash == &global_ops.local_hash) - __ftrace_hash_rec_update(op, filter_hash, inc); + __ftrace_hash_rec_update(op, inc); } while_for_each_ftrace_op(op); } -static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops, - int filter_hash) +static void ftrace_hash_rec_disable_modify(struct ftrace_ops *ops) { - ftrace_hash_rec_update_modify(ops, filter_hash, 0); + ftrace_hash_rec_update_modify(ops, false); } -static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, - int filter_hash) +static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops) { - ftrace_hash_rec_update_modify(ops, filter_hash, 1); + ftrace_hash_rec_update_modify(ops, true); } /* @@ -1897,7 +1997,8 @@ static void ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, */ static int __ftrace_hash_update_ipmodify(struct ftrace_ops *ops, struct ftrace_hash *old_hash, - struct ftrace_hash *new_hash) + struct ftrace_hash *new_hash, + bool update_target) { struct ftrace_page *pg; struct dyn_ftrace *rec, *end = NULL; @@ -1932,10 +2033,13 @@ static int __ftrace_hash_update_ipmodify(struct ftrace_ops *ops, if (rec->flags & FTRACE_FL_DISABLED) continue; - /* We need to update only differences of filter_hash */ + /* + * Unless we are updating the target of a direct function, + * we only need to update differences of filter_hash + */ in_old = !!ftrace_lookup_ip(old_hash, rec->ip); in_new = !!ftrace_lookup_ip(new_hash, rec->ip); - if (in_old == in_new) + if (!update_target && (in_old == in_new)) continue; if (in_new) { @@ -1946,7 +2050,16 @@ static int __ftrace_hash_update_ipmodify(struct ftrace_ops *ops, if (is_ipmodify) goto rollback; - FTRACE_WARN_ON(rec->flags & FTRACE_FL_DIRECT); + /* + * If this is called by __modify_ftrace_direct() + * then it is only changing where the direct + * pointer is jumping to, and the record already + * points to a direct trampoline. If it isn't, + * then it is a bug to update ipmodify on a direct + * caller. + */ + FTRACE_WARN_ON(!update_target && + (rec->flags & FTRACE_FL_DIRECT)); /* * Another ops with IPMODIFY is already @@ -1979,7 +2092,7 @@ rollback: continue; if (rec == end) - goto err_out; + return -EBUSY; in_old = !!ftrace_lookup_ip(old_hash, rec->ip); in_new = !!ftrace_lookup_ip(new_hash, rec->ip); @@ -1992,7 +2105,6 @@ rollback: rec->flags |= FTRACE_FL_IPMODIFY; } while_for_each_ftrace_rec(); -err_out: return -EBUSY; } @@ -2003,7 +2115,7 @@ static int ftrace_hash_ipmodify_enable(struct ftrace_ops *ops) if (ftrace_hash_empty(hash)) hash = NULL; - return __ftrace_hash_update_ipmodify(ops, EMPTY_HASH, hash); + return __ftrace_hash_update_ipmodify(ops, EMPTY_HASH, hash, false); } /* Disabling always succeeds */ @@ -2014,7 +2126,7 @@ static void ftrace_hash_ipmodify_disable(struct ftrace_ops *ops) if (ftrace_hash_empty(hash)) hash = NULL; - __ftrace_hash_update_ipmodify(ops, hash, EMPTY_HASH); + __ftrace_hash_update_ipmodify(ops, hash, EMPTY_HASH, false); } static int ftrace_hash_ipmodify_update(struct ftrace_ops *ops, @@ -2028,7 +2140,7 @@ static int ftrace_hash_ipmodify_update(struct ftrace_ops *ops, if (ftrace_hash_empty(new_hash)) new_hash = NULL; - return __ftrace_hash_update_ipmodify(ops, old_hash, new_hash); + return __ftrace_hash_update_ipmodify(ops, old_hash, new_hash, false); } static void print_ip_ins(const char *fmt, const unsigned char *p) @@ -2113,8 +2225,9 @@ void ftrace_bug(int failed, struct dyn_ftrace *rec) struct ftrace_ops *ops = NULL; pr_info("ftrace record flags: %lx\n", rec->flags); - pr_cont(" (%ld)%s", ftrace_rec_count(rec), - rec->flags & FTRACE_FL_REGS ? " R" : " "); + pr_cont(" (%ld)%s%s", ftrace_rec_count(rec), + rec->flags & FTRACE_FL_REGS ? " R" : " ", + rec->flags & FTRACE_FL_CALL_OPS ? " O" : " "); if (rec->flags & FTRACE_FL_TRAMP_EN) { ops = ftrace_find_tramp_ops_any(rec); if (ops) { @@ -2182,6 +2295,7 @@ static int ftrace_check_record(struct dyn_ftrace *rec, bool enable, bool update) * want the direct enabled (it will be done via the * direct helper). But if DIRECT_EN is set, and * the count is not one, we need to clear it. + * */ if (ftrace_rec_count(rec) == 1) { if (!(rec->flags & FTRACE_FL_DIRECT) != @@ -2190,6 +2304,19 @@ static int ftrace_check_record(struct dyn_ftrace *rec, bool enable, bool update) } else if (rec->flags & FTRACE_FL_DIRECT_EN) { flag |= FTRACE_FL_DIRECT; } + + /* + * Ops calls are special, as count matters. + * As with direct calls, they must only be enabled when count + * is one, otherwise they'll be handled via the list ops. + */ + if (ftrace_rec_count(rec) == 1) { + if (!(rec->flags & FTRACE_FL_CALL_OPS) != + !(rec->flags & FTRACE_FL_CALL_OPS_EN)) + flag |= FTRACE_FL_CALL_OPS; + } else if (rec->flags & FTRACE_FL_CALL_OPS_EN) { + flag |= FTRACE_FL_CALL_OPS; + } } /* If the state of this record hasn't changed, then do nothing */ @@ -2201,7 +2328,7 @@ static int ftrace_check_record(struct dyn_ftrace *rec, bool enable, bool update) flag ^= rec->flags & FTRACE_FL_ENABLED; if (update) { - rec->flags |= FTRACE_FL_ENABLED; + rec->flags |= FTRACE_FL_ENABLED | FTRACE_FL_TOUCHED; if (flag & FTRACE_FL_REGS) { if (rec->flags & FTRACE_FL_REGS) rec->flags |= FTRACE_FL_REGS_EN; @@ -2215,6 +2342,10 @@ static int ftrace_check_record(struct dyn_ftrace *rec, bool enable, bool update) rec->flags &= ~FTRACE_FL_TRAMP_EN; } + /* Keep track of anything that modifies the function */ + if (rec->flags & (FTRACE_FL_DIRECT | FTRACE_FL_IPMODIFY)) + rec->flags |= FTRACE_FL_MODIFIED; + if (flag & FTRACE_FL_DIRECT) { /* * If there's only one user (direct_ops helper) @@ -2234,6 +2365,21 @@ static int ftrace_check_record(struct dyn_ftrace *rec, bool enable, bool update) rec->flags &= ~FTRACE_FL_DIRECT_EN; } } + + if (flag & FTRACE_FL_CALL_OPS) { + if (ftrace_rec_count(rec) == 1) { + if (rec->flags & FTRACE_FL_CALL_OPS) + rec->flags |= FTRACE_FL_CALL_OPS_EN; + else + rec->flags &= ~FTRACE_FL_CALL_OPS_EN; + } else { + /* + * Can only call directly if there's + * only one set of associated ops. + */ + rec->flags &= ~FTRACE_FL_CALL_OPS_EN; + } + } } /* @@ -2256,14 +2402,15 @@ static int ftrace_check_record(struct dyn_ftrace *rec, bool enable, bool update) if (update) { /* If there's no more users, clear all flags */ if (!ftrace_rec_count(rec)) - rec->flags &= FTRACE_FL_DISABLED; + rec->flags &= FTRACE_NOCLEAR_FLAGS; else /* * Just disable the record, but keep the ops TRAMP * and REGS states. The _EN flags must be disabled though. */ rec->flags &= ~(FTRACE_FL_ENABLED | FTRACE_FL_TRAMP_EN | - FTRACE_FL_REGS_EN | FTRACE_FL_DIRECT_EN); + FTRACE_FL_REGS_EN | FTRACE_FL_DIRECT_EN | + FTRACE_FL_CALL_OPS_EN); } ftrace_bug_type = FTRACE_BUG_NOP; @@ -2436,11 +2583,29 @@ ftrace_find_tramp_ops_new(struct dyn_ftrace *rec) return NULL; } +struct ftrace_ops * +ftrace_find_unique_ops(struct dyn_ftrace *rec) +{ + struct ftrace_ops *op, *found = NULL; + unsigned long ip = rec->ip; + + do_for_each_ftrace_op(op, ftrace_ops_list) { + + if (hash_contains_ip(ip, op->func_hash)) { + if (found) + return NULL; + found = op; + } + + } while_for_each_ftrace_op(op); + + return found; +} + #ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS /* Protected by rcu_tasks for reading, and direct_mutex for writing */ -static struct ftrace_hash *direct_functions = EMPTY_HASH; +static struct ftrace_hash __rcu *direct_functions = EMPTY_HASH; static DEFINE_MUTEX(direct_mutex); -int ftrace_direct_func_count; /* * Search the direct_functions hash to see if the given instruction pointer @@ -2457,64 +2622,16 @@ unsigned long ftrace_find_rec_direct(unsigned long ip) return entry->direct; } -static struct ftrace_func_entry* -ftrace_add_rec_direct(unsigned long ip, unsigned long addr, - struct ftrace_hash **free_hash) -{ - struct ftrace_func_entry *entry; - - if (ftrace_hash_empty(direct_functions) || - direct_functions->count > 2 * (1 << direct_functions->size_bits)) { - struct ftrace_hash *new_hash; - int size = ftrace_hash_empty(direct_functions) ? 0 : - direct_functions->count + 1; - - if (size < 32) - size = 32; - - new_hash = dup_hash(direct_functions, size); - if (!new_hash) - return NULL; - - *free_hash = direct_functions; - direct_functions = new_hash; - } - - entry = kmalloc(sizeof(*entry), GFP_KERNEL); - if (!entry) - return NULL; - - entry->ip = ip; - entry->direct = addr; - __add_hash_entry(direct_functions, entry); - return entry; -} - static void call_direct_funcs(unsigned long ip, unsigned long pip, struct ftrace_ops *ops, struct ftrace_regs *fregs) { - unsigned long addr; + unsigned long addr = READ_ONCE(ops->direct_call); - addr = ftrace_find_rec_direct(ip); if (!addr) return; arch_ftrace_set_direct_caller(fregs, addr); } - -struct ftrace_ops direct_ops = { - .func = call_direct_funcs, - .flags = FTRACE_OPS_FL_DIRECT | FTRACE_OPS_FL_SAVE_REGS - | FTRACE_OPS_FL_PERMANENT, - /* - * By declaring the main trampoline as this trampoline - * it will never have one allocated for it. Allocated - * trampolines should not call direct functions. - * The direct_ops should only be called by the builtin - * ftrace_regs_caller trampoline. - */ - .trampoline = FTRACE_REGS_ADDR, -}; #endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */ /** @@ -2525,7 +2642,7 @@ struct ftrace_ops direct_ops = { * wants to convert to a callback that saves all regs. If FTRACE_FL_REGS * is not set, then it wants to convert to the normal callback. * - * Returns the address of the trampoline to set to + * Returns: the address of the trampoline to set to */ unsigned long ftrace_get_addr_new(struct dyn_ftrace *rec) { @@ -2566,7 +2683,7 @@ unsigned long ftrace_get_addr_new(struct dyn_ftrace *rec) * a function that saves all the regs. Basically the '_EN' version * represents the current state of the function. * - * Returns the address of the trampoline that is currently being called + * Returns: the address of the trampoline that is currently being called */ unsigned long ftrace_get_addr_curr(struct dyn_ftrace *rec) { @@ -2670,7 +2787,7 @@ struct ftrace_rec_iter { /** * ftrace_rec_iter_start - start up iterating over traced functions * - * Returns an iterator handle that is used to iterate over all + * Returns: an iterator handle that is used to iterate over all * the records that represent address locations where functions * are traced. * @@ -2702,7 +2819,7 @@ struct ftrace_rec_iter *ftrace_rec_iter_start(void) * ftrace_rec_iter_next - get the next record to process. * @iter: The handle to the iterator. * - * Returns the next iterator after the given iterator @iter. + * Returns: the next iterator after the given iterator @iter. */ struct ftrace_rec_iter *ftrace_rec_iter_next(struct ftrace_rec_iter *iter) { @@ -2727,7 +2844,7 @@ struct ftrace_rec_iter *ftrace_rec_iter_next(struct ftrace_rec_iter *iter) * ftrace_rec_iter_record - get the record at the iterator location * @iter: The current iterator location * - * Returns the record that the current @iter is at. + * Returns: the record that the current @iter is at. */ struct dyn_ftrace *ftrace_rec_iter_record(struct ftrace_rec_iter *iter) { @@ -2994,7 +3111,7 @@ int ftrace_startup(struct ftrace_ops *ops, int command) return ret; } - if (ftrace_hash_rec_enable(ops, 1)) + if (ftrace_hash_rec_enable(ops)) command |= FTRACE_UPDATE_CALLS; ftrace_startup_enable(command); @@ -3036,7 +3153,7 @@ int ftrace_shutdown(struct ftrace_ops *ops, int command) /* Disabling ipmodify never fails */ ftrace_hash_ipmodify_disable(ops); - if (ftrace_hash_rec_disable(ops, 1)) + if (ftrace_hash_rec_disable(ops)) command |= FTRACE_UPDATE_CALLS; ops->flags &= ~FTRACE_OPS_FL_ENABLED; @@ -3072,7 +3189,7 @@ int ftrace_shutdown(struct ftrace_ops *ops, int command) struct dyn_ftrace *rec; do_for_each_ftrace_rec(pg, rec) { - if (FTRACE_WARN_ON_ONCE(rec->flags & ~FTRACE_FL_DISABLED)) + if (FTRACE_WARN_ON_ONCE(rec->flags & ~FTRACE_NOCLEAR_FLAGS)) pr_warn(" %pS flags:%lx\n", (void *)rec->ip, rec->flags); } while_for_each_ftrace_rec(); @@ -3107,8 +3224,7 @@ out: * synchronize_rcu_tasks() will wait for those tasks to * execute and either schedule voluntarily or enter user space. */ - if (IS_ENABLED(CONFIG_PREEMPTION)) - synchronize_rcu_tasks(); + synchronize_rcu_tasks(); ftrace_trampoline_free(ops); } @@ -3116,7 +3232,535 @@ out: return 0; } -static u64 ftrace_update_time; +/* Simply make a copy of @src and return it */ +static struct ftrace_hash *copy_hash(struct ftrace_hash *src) +{ + if (ftrace_hash_empty(src)) + return EMPTY_HASH; + + return alloc_and_copy_ftrace_hash(src->size_bits, src); +} + +/* + * Append @new_hash entries to @hash: + * + * If @hash is the EMPTY_HASH then it traces all functions and nothing + * needs to be done. + * + * If @new_hash is the EMPTY_HASH, then make *hash the EMPTY_HASH so + * that it traces everything. + * + * Otherwise, go through all of @new_hash and add anything that @hash + * doesn't already have, to @hash. + * + * The filter_hash updates uses just the append_hash() function + * and the notrace_hash does not. + */ +static int append_hash(struct ftrace_hash **hash, struct ftrace_hash *new_hash, + int size_bits) +{ + struct ftrace_func_entry *entry; + int size; + int i; + + if (*hash) { + /* An empty hash does everything */ + if (ftrace_hash_empty(*hash)) + return 0; + } else { + *hash = alloc_ftrace_hash(size_bits); + if (!*hash) + return -ENOMEM; + } + + /* If new_hash has everything make hash have everything */ + if (ftrace_hash_empty(new_hash)) { + free_ftrace_hash(*hash); + *hash = EMPTY_HASH; + return 0; + } + + size = 1 << new_hash->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &new_hash->buckets[i], hlist) { + /* Only add if not already in hash */ + if (!__ftrace_lookup_ip(*hash, entry->ip) && + add_hash_entry(*hash, entry->ip) == NULL) + return -ENOMEM; + } + } + return 0; +} + +/* + * Remove functions from @hash that are in @notrace_hash + */ +static void remove_hash(struct ftrace_hash *hash, struct ftrace_hash *notrace_hash) +{ + struct ftrace_func_entry *entry; + struct hlist_node *tmp; + int size; + int i; + + /* If the notrace hash is empty, there's nothing to do */ + if (ftrace_hash_empty(notrace_hash)) + return; + + size = 1 << hash->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry_safe(entry, tmp, &hash->buckets[i], hlist) { + if (!__ftrace_lookup_ip(notrace_hash, entry->ip)) + continue; + remove_hash_entry(hash, entry); + kfree(entry); + } + } +} + +/* + * Add to @hash only those that are in both @new_hash1 and @new_hash2 + * + * The notrace_hash updates uses just the intersect_hash() function + * and the filter_hash does not. + */ +static int intersect_hash(struct ftrace_hash **hash, struct ftrace_hash *new_hash1, + struct ftrace_hash *new_hash2) +{ + struct ftrace_func_entry *entry; + int size; + int i; + + /* + * If new_hash1 or new_hash2 is the EMPTY_HASH then make the hash + * empty as well as empty for notrace means none are notraced. + */ + if (ftrace_hash_empty(new_hash1) || ftrace_hash_empty(new_hash2)) { + free_ftrace_hash(*hash); + *hash = EMPTY_HASH; + return 0; + } + + size = 1 << new_hash1->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &new_hash1->buckets[i], hlist) { + /* Only add if in both @new_hash1 and @new_hash2 */ + if (__ftrace_lookup_ip(new_hash2, entry->ip) && + add_hash_entry(*hash, entry->ip) == NULL) + return -ENOMEM; + } + } + /* If nothing intersects, make it the empty set */ + if (ftrace_hash_empty(*hash)) { + free_ftrace_hash(*hash); + *hash = EMPTY_HASH; + } + return 0; +} + +static bool ops_equal(struct ftrace_hash *A, struct ftrace_hash *B) +{ + struct ftrace_func_entry *entry; + int size; + int i; + + if (ftrace_hash_empty(A)) + return ftrace_hash_empty(B); + + if (ftrace_hash_empty(B)) + return ftrace_hash_empty(A); + + if (A->count != B->count) + return false; + + size = 1 << A->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &A->buckets[i], hlist) { + if (!__ftrace_lookup_ip(B, entry->ip)) + return false; + } + } + + return true; +} + +static void ftrace_ops_update_code(struct ftrace_ops *ops, + struct ftrace_ops_hash *old_hash); + +static int __ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, + struct ftrace_hash **orig_hash, + struct ftrace_hash *hash, + int enable) +{ + struct ftrace_ops_hash old_hash_ops; + struct ftrace_hash *old_hash; + int ret; + + old_hash = *orig_hash; + old_hash_ops.filter_hash = ops->func_hash->filter_hash; + old_hash_ops.notrace_hash = ops->func_hash->notrace_hash; + ret = ftrace_hash_move(ops, enable, orig_hash, hash); + if (!ret) { + ftrace_ops_update_code(ops, &old_hash_ops); + free_ftrace_hash_rcu(old_hash); + } + return ret; +} + +static int ftrace_update_ops(struct ftrace_ops *ops, struct ftrace_hash *filter_hash, + struct ftrace_hash *notrace_hash) +{ + int ret; + + if (!ops_equal(filter_hash, ops->func_hash->filter_hash)) { + ret = __ftrace_hash_move_and_update_ops(ops, &ops->func_hash->filter_hash, + filter_hash, 1); + if (ret < 0) + return ret; + } + + if (!ops_equal(notrace_hash, ops->func_hash->notrace_hash)) { + ret = __ftrace_hash_move_and_update_ops(ops, &ops->func_hash->notrace_hash, + notrace_hash, 0); + if (ret < 0) + return ret; + } + + return 0; +} + +static int add_first_hash(struct ftrace_hash **filter_hash, struct ftrace_hash **notrace_hash, + struct ftrace_ops_hash *func_hash) +{ + /* If the filter hash is not empty, simply remove the nohash from it */ + if (!ftrace_hash_empty(func_hash->filter_hash)) { + *filter_hash = copy_hash(func_hash->filter_hash); + if (!*filter_hash) + return -ENOMEM; + remove_hash(*filter_hash, func_hash->notrace_hash); + *notrace_hash = EMPTY_HASH; + + } else { + *notrace_hash = copy_hash(func_hash->notrace_hash); + if (!*notrace_hash) + return -ENOMEM; + *filter_hash = EMPTY_HASH; + } + return 0; +} + +static int add_next_hash(struct ftrace_hash **filter_hash, struct ftrace_hash **notrace_hash, + struct ftrace_ops_hash *ops_hash, struct ftrace_ops_hash *subops_hash) +{ + int size_bits; + int ret; + + /* If the subops trace all functions so must the main ops */ + if (ftrace_hash_empty(ops_hash->filter_hash) || + ftrace_hash_empty(subops_hash->filter_hash)) { + *filter_hash = EMPTY_HASH; + } else { + /* + * The main ops filter hash is not empty, so its + * notrace_hash had better be, as the notrace hash + * is only used for empty main filter hashes. + */ + WARN_ON_ONCE(!ftrace_hash_empty(ops_hash->notrace_hash)); + + size_bits = max(ops_hash->filter_hash->size_bits, + subops_hash->filter_hash->size_bits); + + /* Copy the subops hash */ + *filter_hash = alloc_and_copy_ftrace_hash(size_bits, subops_hash->filter_hash); + if (!*filter_hash) + return -ENOMEM; + /* Remove any notrace functions from the copy */ + remove_hash(*filter_hash, subops_hash->notrace_hash); + + ret = append_hash(filter_hash, ops_hash->filter_hash, + size_bits); + if (ret < 0) { + free_ftrace_hash(*filter_hash); + *filter_hash = EMPTY_HASH; + return ret; + } + } + + /* + * Only process notrace hashes if the main filter hash is empty + * (tracing all functions), otherwise the filter hash will just + * remove the notrace hash functions, and the notrace hash is + * not needed. + */ + if (ftrace_hash_empty(*filter_hash)) { + /* + * Intersect the notrace functions. That is, if two + * subops are not tracing a set of functions, the + * main ops will only not trace the functions that are + * in both subops, but has to trace the functions that + * are only notrace in one of the subops, for the other + * subops to be able to trace them. + */ + size_bits = max(ops_hash->notrace_hash->size_bits, + subops_hash->notrace_hash->size_bits); + *notrace_hash = alloc_ftrace_hash(size_bits); + if (!*notrace_hash) + return -ENOMEM; + + ret = intersect_hash(notrace_hash, ops_hash->notrace_hash, + subops_hash->notrace_hash); + if (ret < 0) { + free_ftrace_hash(*notrace_hash); + *notrace_hash = EMPTY_HASH; + return ret; + } + } + return 0; +} + +/** + * ftrace_startup_subops - enable tracing for subops of an ops + * @ops: Manager ops (used to pick all the functions of its subops) + * @subops: A new ops to add to @ops + * @command: Extra commands to use to enable tracing + * + * The @ops is a manager @ops that has the filter that includes all the functions + * that its list of subops are tracing. Adding a new @subops will add the + * functions of @subops to @ops. + */ +int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + struct ftrace_hash *filter_hash = EMPTY_HASH; + struct ftrace_hash *notrace_hash = EMPTY_HASH; + struct ftrace_hash *save_filter_hash; + struct ftrace_hash *save_notrace_hash; + int ret; + + if (unlikely(ftrace_disabled)) + return -ENODEV; + + ftrace_ops_init(ops); + ftrace_ops_init(subops); + + if (WARN_ON_ONCE(subops->flags & FTRACE_OPS_FL_ENABLED)) + return -EBUSY; + + /* Make everything canonical (Just in case!) */ + if (!ops->func_hash->filter_hash) + ops->func_hash->filter_hash = EMPTY_HASH; + if (!ops->func_hash->notrace_hash) + ops->func_hash->notrace_hash = EMPTY_HASH; + if (!subops->func_hash->filter_hash) + subops->func_hash->filter_hash = EMPTY_HASH; + if (!subops->func_hash->notrace_hash) + subops->func_hash->notrace_hash = EMPTY_HASH; + + /* For the first subops to ops just enable it normally */ + if (list_empty(&ops->subop_list)) { + + /* The ops was empty, should have empty hashes */ + WARN_ON_ONCE(!ftrace_hash_empty(ops->func_hash->filter_hash)); + WARN_ON_ONCE(!ftrace_hash_empty(ops->func_hash->notrace_hash)); + + ret = add_first_hash(&filter_hash, ¬race_hash, subops->func_hash); + if (ret < 0) + return ret; + + save_filter_hash = ops->func_hash->filter_hash; + save_notrace_hash = ops->func_hash->notrace_hash; + + ops->func_hash->filter_hash = filter_hash; + ops->func_hash->notrace_hash = notrace_hash; + list_add(&subops->list, &ops->subop_list); + ret = ftrace_startup(ops, command); + if (ret < 0) { + list_del(&subops->list); + ops->func_hash->filter_hash = save_filter_hash; + ops->func_hash->notrace_hash = save_notrace_hash; + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + } else { + free_ftrace_hash(save_filter_hash); + free_ftrace_hash(save_notrace_hash); + subops->flags |= FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP; + subops->managed = ops; + } + return ret; + } + + /* + * Here there's already something attached. Here are the rules: + * If the new subops and main ops filter hashes are not empty: + * o Make a copy of the subops filter hash + * o Remove all functions in the nohash from it. + * o Add in the main hash filter functions + * o Remove any of these functions from the main notrace hash + */ + + ret = add_next_hash(&filter_hash, ¬race_hash, ops->func_hash, subops->func_hash); + if (ret < 0) + return ret; + + list_add(&subops->list, &ops->subop_list); + + ret = ftrace_update_ops(ops, filter_hash, notrace_hash); + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + if (ret < 0) { + list_del(&subops->list); + } else { + subops->flags |= FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP; + subops->managed = ops; + } + return ret; +} + +static int rebuild_hashes(struct ftrace_hash **filter_hash, struct ftrace_hash **notrace_hash, + struct ftrace_ops *ops) +{ + struct ftrace_ops_hash temp_hash; + struct ftrace_ops *subops; + bool first = true; + int ret; + + temp_hash.filter_hash = EMPTY_HASH; + temp_hash.notrace_hash = EMPTY_HASH; + + list_for_each_entry(subops, &ops->subop_list, list) { + *filter_hash = EMPTY_HASH; + *notrace_hash = EMPTY_HASH; + + if (first) { + ret = add_first_hash(filter_hash, notrace_hash, subops->func_hash); + if (ret < 0) + return ret; + first = false; + } else { + ret = add_next_hash(filter_hash, notrace_hash, + &temp_hash, subops->func_hash); + if (ret < 0) { + free_ftrace_hash(temp_hash.filter_hash); + free_ftrace_hash(temp_hash.notrace_hash); + return ret; + } + } + + free_ftrace_hash(temp_hash.filter_hash); + free_ftrace_hash(temp_hash.notrace_hash); + + temp_hash.filter_hash = *filter_hash; + temp_hash.notrace_hash = *notrace_hash; + } + return 0; +} + +/** + * ftrace_shutdown_subops - Remove a subops from a manager ops + * @ops: A manager ops to remove @subops from + * @subops: The subops to remove from @ops + * @command: Any extra command flags to add to modifying the text + * + * Removes the functions being traced by the @subops from @ops. Note, it + * will not affect functions that are being traced by other subops that + * still exist in @ops. + * + * If the last subops is removed from @ops, then @ops is shutdown normally. + */ +int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + struct ftrace_hash *filter_hash = EMPTY_HASH; + struct ftrace_hash *notrace_hash = EMPTY_HASH; + int ret; + + if (unlikely(ftrace_disabled)) + return -ENODEV; + + if (WARN_ON_ONCE(!(subops->flags & FTRACE_OPS_FL_ENABLED))) + return -EINVAL; + + list_del(&subops->list); + + if (list_empty(&ops->subop_list)) { + /* Last one, just disable the current ops */ + + ret = ftrace_shutdown(ops, command); + if (ret < 0) { + list_add(&subops->list, &ops->subop_list); + return ret; + } + + subops->flags &= ~FTRACE_OPS_FL_ENABLED; + + free_ftrace_hash(ops->func_hash->filter_hash); + free_ftrace_hash(ops->func_hash->notrace_hash); + ops->func_hash->filter_hash = EMPTY_HASH; + ops->func_hash->notrace_hash = EMPTY_HASH; + subops->flags &= ~(FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP); + subops->managed = NULL; + + return 0; + } + + /* Rebuild the hashes without subops */ + ret = rebuild_hashes(&filter_hash, ¬race_hash, ops); + if (ret < 0) + return ret; + + ret = ftrace_update_ops(ops, filter_hash, notrace_hash); + if (ret < 0) { + list_add(&subops->list, &ops->subop_list); + } else { + subops->flags &= ~(FTRACE_OPS_FL_ENABLED | FTRACE_OPS_FL_SUBOP); + subops->managed = NULL; + } + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + return ret; +} + +static int ftrace_hash_move_and_update_subops(struct ftrace_ops *subops, + struct ftrace_hash **orig_subhash, + struct ftrace_hash *hash) +{ + struct ftrace_ops *ops = subops->managed; + struct ftrace_hash *notrace_hash; + struct ftrace_hash *filter_hash; + struct ftrace_hash *save_hash; + struct ftrace_hash *new_hash; + int ret; + + /* Manager ops can not be subops (yet) */ + if (WARN_ON_ONCE(!ops || ops->flags & FTRACE_OPS_FL_SUBOP)) + return -EINVAL; + + /* Move the new hash over to the subops hash */ + save_hash = *orig_subhash; + *orig_subhash = __ftrace_hash_move(hash); + if (!*orig_subhash) { + *orig_subhash = save_hash; + return -ENOMEM; + } + + ret = rebuild_hashes(&filter_hash, ¬race_hash, ops); + if (!ret) { + ret = ftrace_update_ops(ops, filter_hash, notrace_hash); + free_ftrace_hash(filter_hash); + free_ftrace_hash(notrace_hash); + } + + if (ret) { + /* Put back the original hash */ + new_hash = *orig_subhash; + *orig_subhash = save_hash; + free_ftrace_hash_rcu(new_hash); + } else { + free_ftrace_hash_rcu(save_hash); + } + return ret; +} + + +u64 ftrace_update_time; +u64 ftrace_total_mod_time; unsigned long ftrace_update_tot_cnt; unsigned long ftrace_number_of_pages; unsigned long ftrace_number_of_groups; @@ -3136,7 +3780,7 @@ static int ftrace_update_code(struct module *mod, struct ftrace_page *new_pgs) bool init_nop = ftrace_need_init_nop(); struct ftrace_page *pg; struct dyn_ftrace *p; - u64 start, stop; + u64 start, stop, update_time; unsigned long update_cnt = 0; unsigned long rec_flags = 0; int i; @@ -3180,7 +3824,11 @@ static int ftrace_update_code(struct module *mod, struct ftrace_page *new_pgs) } stop = ftrace_now(raw_smp_processor_id()); - ftrace_update_time = stop - start; + update_time = stop - start; + if (mod) + ftrace_total_mod_time += update_time; + else + ftrace_update_time = update_time; ftrace_update_tot_cnt += update_cnt; return 0; @@ -3222,6 +3870,22 @@ static int ftrace_allocate_records(struct ftrace_page *pg, int count) return cnt; } +static void ftrace_free_pages(struct ftrace_page *pages) +{ + struct ftrace_page *pg = pages; + + while (pg) { + if (pg->records) { + free_pages((unsigned long)pg->records, pg->order); + ftrace_number_of_pages -= 1 << pg->order; + } + pages = pg->next; + kfree(pg); + pg = pages; + ftrace_number_of_groups--; + } +} + static struct ftrace_page * ftrace_allocate_pages(unsigned long num_to_init) { @@ -3260,17 +3924,7 @@ ftrace_allocate_pages(unsigned long num_to_init) return start_pg; free_pages: - pg = start_pg; - while (pg) { - if (pg->records) { - free_pages((unsigned long)pg->records, pg->order); - ftrace_number_of_pages -= 1 << pg->order; - } - start_pg = pg->next; - kfree(pg); - pg = start_pg; - ftrace_number_of_groups--; - } + ftrace_free_pages(start_pg); pr_info("ftrace: FAILED to allocate memory for functions\n"); return NULL; } @@ -3523,7 +4177,10 @@ t_func_next(struct seq_file *m, loff_t *pos) !ftrace_lookup_ip(iter->hash, rec->ip)) || ((iter->flags & FTRACE_ITER_ENABLED) && - !(rec->flags & FTRACE_FL_ENABLED))) { + !(rec->flags & FTRACE_FL_ENABLED)) || + + ((iter->flags & FTRACE_ITER_TOUCHED) && + !(rec->flags & FTRACE_FL_TOUCHED))) { rec = NULL; goto retry; @@ -3751,6 +4408,42 @@ static inline int print_rec(struct seq_file *m, unsigned long ip) } #endif +static void print_subops(struct seq_file *m, struct ftrace_ops *ops, struct dyn_ftrace *rec) +{ + struct ftrace_ops *subops; + bool first = true; + + list_for_each_entry(subops, &ops->subop_list, list) { + if (!((subops->flags & FTRACE_OPS_FL_ENABLED) && + hash_contains_ip(rec->ip, subops->func_hash))) + continue; + if (first) { + seq_printf(m, "\tsubops:"); + first = false; + } +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + if (subops->flags & FTRACE_OPS_FL_GRAPH) { + struct fgraph_ops *gops; + + gops = container_of(subops, struct fgraph_ops, ops); + seq_printf(m, " {ent:%pS ret:%pS}", + (void *)gops->entryfunc, + (void *)gops->retfunc); + continue; + } +#endif + if (subops->trampoline) { + seq_printf(m, " {%pS (%pS)}", + (void *)subops->trampoline, + (void *)subops->func); + add_trampoline_func(m, subops, rec); + } else { + seq_printf(m, " {%pS}", + (void *)subops->func); + } + } +} + static int t_show(struct seq_file *m, void *v) { struct ftrace_iterator *iter = m->private; @@ -3775,6 +4468,9 @@ static int t_show(struct seq_file *m, void *v) if (!rec) return 0; + if (iter->flags & FTRACE_ITER_ADDRS) + seq_printf(m, "%lx ", rec->ip); + if (print_rec(m, rec->ip)) { /* This should only happen when a rec is disabled */ WARN_ON_ONCE(!(rec->flags & FTRACE_FL_DISABLED)); @@ -3782,14 +4478,16 @@ static int t_show(struct seq_file *m, void *v) return 0; } - if (iter->flags & FTRACE_ITER_ENABLED) { + if (iter->flags & (FTRACE_ITER_ENABLED | FTRACE_ITER_TOUCHED)) { struct ftrace_ops *ops; - seq_printf(m, " (%ld)%s%s%s", + seq_printf(m, " (%ld)%s%s%s%s%s", ftrace_rec_count(rec), rec->flags & FTRACE_FL_REGS ? " R" : " ", rec->flags & FTRACE_FL_IPMODIFY ? " I" : " ", - rec->flags & FTRACE_FL_DIRECT ? " D" : " "); + rec->flags & FTRACE_FL_DIRECT ? " D" : " ", + rec->flags & FTRACE_FL_CALL_OPS ? " O" : " ", + rec->flags & FTRACE_FL_MODIFIED ? " M " : " "); if (rec->flags & FTRACE_FL_TRAMP_EN) { ops = ftrace_find_tramp_ops_any(rec); if (ops) { @@ -3798,6 +4496,7 @@ static int t_show(struct seq_file *m, void *v) (void *)ops->trampoline, (void *)ops->func); add_trampoline_func(m, ops, rec); + print_subops(m, ops, rec); ops = ftrace_find_tramp_ops_next(rec, ops); } while (ops); } else @@ -3805,6 +4504,16 @@ static int t_show(struct seq_file *m, void *v) } else { add_trampoline_func(m, NULL, rec); } + if (rec->flags & FTRACE_FL_CALL_OPS_EN) { + ops = ftrace_find_unique_ops(rec); + if (ops) { + seq_printf(m, "\tops: %pS (%pS)", + ops, ops->func); + print_subops(m, ops, rec); + } else { + seq_puts(m, "\tops: ERROR!"); + } + } if (rec->flags & FTRACE_FL_DIRECT) { unsigned long direct; @@ -3874,6 +4583,55 @@ ftrace_enabled_open(struct inode *inode, struct file *file) return 0; } +static int +ftrace_touched_open(struct inode *inode, struct file *file) +{ + struct ftrace_iterator *iter; + + /* + * This shows us what functions have ever been enabled + * (traced, direct, patched, etc). Not sure if we want lockdown + * to hide such critical information for an admin. + * Although, perhaps it can show information we don't + * want people to see, but if something had traced + * something, we probably want to know about it. + */ + + iter = __seq_open_private(file, &show_ftrace_seq_ops, sizeof(*iter)); + if (!iter) + return -ENOMEM; + + iter->pg = ftrace_pages_start; + iter->flags = FTRACE_ITER_TOUCHED; + iter->ops = &global_ops; + + return 0; +} + +static int +ftrace_avail_addrs_open(struct inode *inode, struct file *file) +{ + struct ftrace_iterator *iter; + int ret; + + ret = security_locked_down(LOCKDOWN_TRACEFS); + if (ret) + return ret; + + if (unlikely(ftrace_disabled)) + return -ENODEV; + + iter = __seq_open_private(file, &show_ftrace_seq_ops, sizeof(*iter)); + if (!iter) + return -ENOMEM; + + iter->pg = ftrace_pages_start; + iter->flags = FTRACE_ITER_ADDRS; + iter->ops = &global_ops; + + return 0; +} + /** * ftrace_regex_open - initialize function tracer filter files * @ops: The ftrace_ops that hold the hash filters @@ -3889,6 +4647,8 @@ ftrace_enabled_open(struct inode *inode, struct file *file) * ftrace_notrace_write() if @flag has FTRACE_ITER_NOTRACE set. * tracing_lseek() should be used as the lseek routine, and * release must call ftrace_regex_release(). + * + * Returns: 0 on success or a negative errno value on failure */ int ftrace_regex_open(struct ftrace_ops *ops, int flag, @@ -3940,13 +4700,17 @@ ftrace_regex_open(struct ftrace_ops *ops, int flag, } else { iter->hash = alloc_and_copy_ftrace_hash(size_bits, hash); } + } else { + if (hash) + iter->hash = alloc_and_copy_ftrace_hash(hash->size_bits, hash); + else + iter->hash = EMPTY_HASH; + } - if (!iter->hash) { - trace_parser_put(&iter->parser); - goto out_unlock; - } - } else - iter->hash = hash; + if (!iter->hash) { + trace_parser_put(&iter->parser); + goto out_unlock; + } ret = 0; @@ -4068,8 +4832,8 @@ enter_record(struct ftrace_hash *hash, struct dyn_ftrace *rec, int clear_filter) /* Do nothing if it exists */ if (entry) return 0; - - ret = add_hash_entry(hash, rec->ip); + if (add_hash_entry(hash, rec->ip) == NULL) + ret = -ENOMEM; } return ret; } @@ -4078,12 +4842,12 @@ static int add_rec_by_index(struct ftrace_hash *hash, struct ftrace_glob *func_g, int clear_filter) { - long index = simple_strtoul(func_g->search, NULL, 0); + long index; struct ftrace_page *pg; struct dyn_ftrace *rec; /* The index starts at 1 */ - if (--index < 0) + if (kstrtoul(func_g->search, 0, &index) || --index < 0) return 0; do_for_each_ftrace_rec(pg, rec) { @@ -4185,15 +4949,13 @@ match_records(struct ftrace_hash *hash, char *func, int len, char *mod) mod_g.len = strlen(mod_g.search); } - mutex_lock(&ftrace_lock); + guard(mutex)(&ftrace_lock); if (unlikely(ftrace_disabled)) - goto out_unlock; + return 0; - if (func_g.type == MATCH_INDEX) { - found = add_rec_by_index(hash, &func_g, clear_filter); - goto out_unlock; - } + if (func_g.type == MATCH_INDEX) + return add_rec_by_index(hash, &func_g, clear_filter); do_for_each_ftrace_rec(pg, rec) { @@ -4202,16 +4964,12 @@ match_records(struct ftrace_hash *hash, char *func, int len, char *mod) if (ftrace_match_record(rec, &func_g, mod_match, exclude_mod)) { ret = enter_record(hash, rec, clear_filter); - if (ret < 0) { - found = ret; - goto out_unlock; - } + if (ret < 0) + return ret; found = 1; } cond_resched(); } while_for_each_ftrace_rec(); - out_unlock: - mutex_unlock(&ftrace_lock); return found; } @@ -4258,36 +5016,33 @@ static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, struct ftrace_hash *hash, int enable) { - struct ftrace_ops_hash old_hash_ops; - struct ftrace_hash *old_hash; - int ret; - - old_hash = *orig_hash; - old_hash_ops.filter_hash = ops->func_hash->filter_hash; - old_hash_ops.notrace_hash = ops->func_hash->notrace_hash; - ret = ftrace_hash_move(ops, enable, orig_hash, hash); - if (!ret) { - ftrace_ops_update_code(ops, &old_hash_ops); - free_ftrace_hash_rcu(old_hash); - } - return ret; -} + if (ops->flags & FTRACE_OPS_FL_SUBOP) + return ftrace_hash_move_and_update_subops(ops, orig_hash, hash); -static bool module_exists(const char *module) -{ - /* All modules have the symbol __this_module */ - static const char this_mod[] = "__this_module"; - char modname[MAX_PARAM_PREFIX_LEN + sizeof(this_mod) + 2]; - unsigned long val; - int n; + /* + * If this ops is not enabled, it could be sharing its filters + * with a subop. If that's the case, update the subop instead of + * this ops. Shared filters are only allowed to have one ops set + * at a time, and if we update the ops that is not enabled, + * it will not affect subops that share it. + */ + if (!(ops->flags & FTRACE_OPS_FL_ENABLED)) { + struct ftrace_ops *op; - n = snprintf(modname, sizeof(modname), "%s:%s", module, this_mod); + /* Check if any other manager subops maps to this hash */ + do_for_each_ftrace_op(op, ftrace_ops_list) { + struct ftrace_ops *subops; - if (n > sizeof(modname) - 1) - return false; + list_for_each_entry(subops, &op->subop_list, list) { + if ((subops->flags & FTRACE_OPS_FL_ENABLED) && + subops->func_hash == ops->func_hash) { + return ftrace_hash_move_and_update_subops(subops, orig_hash, hash); + } + } + } while_for_each_ftrace_op(op); + } - val = module_kallsyms_lookup_name(modname); - return val != 0; + return __ftrace_hash_move_and_update_ops(ops, orig_hash, hash, enable); } static int cache_mod(struct trace_array *tr, @@ -4295,14 +5050,14 @@ static int cache_mod(struct trace_array *tr, { struct ftrace_mod_load *ftrace_mod, *n; struct list_head *head = enable ? &tr->mod_trace : &tr->mod_notrace; - int ret; - mutex_lock(&ftrace_lock); + guard(mutex)(&ftrace_lock); /* We do not cache inverse filters */ if (func[0] == '!') { + int ret = -EINVAL; + func++; - ret = -EINVAL; /* Look to remove this hash */ list_for_each_entry_safe(ftrace_mod, n, head, list) { @@ -4318,26 +5073,17 @@ static int cache_mod(struct trace_array *tr, continue; } } - goto out; + return ret; } - ret = -EINVAL; /* We only care about modules that have not been loaded yet */ if (module_exists(module)) - goto out; + return -EINVAL; /* Save this string off, and execute it when the module is loaded */ - ret = ftrace_add_mod(tr, func, module, enable); - out: - mutex_unlock(&ftrace_lock); - - return ret; + return ftrace_add_mod(tr, func, module, enable); } -static int -ftrace_set_regex(struct ftrace_ops *ops, unsigned char *buf, int len, - int reset, int enable); - #ifdef CONFIG_MODULES static void process_mod_list(struct list_head *head, struct ftrace_ops *ops, char *mod, bool enable) @@ -4441,6 +5187,9 @@ ftrace_mod_callback(struct trace_array *tr, struct ftrace_hash *hash, char *func; int ret; + if (!tr) + return -ENODEV; + /* match_records() modifies func, and we need the original */ func = kstrdup(func_orig, GFP_KERNEL); if (!func) @@ -4498,14 +5247,18 @@ struct ftrace_func_map { void *data; }; +/* + * Note, ftrace_func_mapper is freed by free_ftrace_hash(&mapper->hash). + * The hash field must be the first field. + */ struct ftrace_func_mapper { - struct ftrace_hash hash; + struct ftrace_hash hash; /* Must be first! */ }; /** * allocate_ftrace_func_mapper - allocate a new ftrace_func_mapper * - * Returns a ftrace_func_mapper descriptor that can be used to map ips to data. + * Returns: a ftrace_func_mapper descriptor that can be used to map ips to data. */ struct ftrace_func_mapper *allocate_ftrace_func_mapper(void) { @@ -4525,7 +5278,7 @@ struct ftrace_func_mapper *allocate_ftrace_func_mapper(void) * @mapper: The mapper that has the ip maps * @ip: the instruction pointer to find the data for * - * Returns the data mapped to @ip if found otherwise NULL. The return + * Returns: the data mapped to @ip if found otherwise NULL. The return * is actually the address of the mapper data pointer. The address is * returned for use cases where the data is no bigger than a long, and * the user can use the data pointer as its data instead of having to @@ -4551,7 +5304,7 @@ void **ftrace_func_mapper_find_ip(struct ftrace_func_mapper *mapper, * @ip: The instruction pointer address to map @data to * @data: The data to map to @ip * - * Returns 0 on success otherwise an error. + * Returns: 0 on success otherwise an error. */ int ftrace_func_mapper_add_ip(struct ftrace_func_mapper *mapper, unsigned long ip, void *data) @@ -4580,7 +5333,7 @@ int ftrace_func_mapper_add_ip(struct ftrace_func_mapper *mapper, * @mapper: The mapper that has the ip maps * @ip: The instruction pointer address to remove the data from * - * Returns the data if it is found, otherwise NULL. + * Returns: the data if it is found, otherwise NULL. * Note, if the data pointer is used as the data itself, (see * ftrace_func_mapper_find_ip(), then the return value may be meaningless, * if the data pointer was set to zero. @@ -4634,6 +5387,7 @@ void free_ftrace_func_mapper(struct ftrace_func_mapper *mapper, } } } + /* This also frees the mapper itself */ free_ftrace_hash(&mapper->hash); } @@ -4641,7 +5395,7 @@ static void release_probe(struct ftrace_func_probe *probe) { struct ftrace_probe_ops *probe_ops; - mutex_lock(&ftrace_lock); + guard(mutex)(&ftrace_lock); WARN_ON(probe->ref <= 0); @@ -4659,7 +5413,6 @@ static void release_probe(struct ftrace_func_probe *probe) list_del(&probe->list); kfree(probe); } - mutex_unlock(&ftrace_lock); } static void acquire_probe_locked(struct ftrace_func_probe *probe) @@ -4965,20 +5718,15 @@ static DEFINE_MUTEX(ftrace_cmd_mutex); __init int register_ftrace_command(struct ftrace_func_command *cmd) { struct ftrace_func_command *p; - int ret = 0; - mutex_lock(&ftrace_cmd_mutex); + guard(mutex)(&ftrace_cmd_mutex); list_for_each_entry(p, &ftrace_commands, list) { - if (strcmp(cmd->name, p->name) == 0) { - ret = -EBUSY; - goto out_unlock; - } + if (strcmp(cmd->name, p->name) == 0) + return -EBUSY; } list_add(&cmd->list, &ftrace_commands); - out_unlock: - mutex_unlock(&ftrace_cmd_mutex); - return ret; + return 0; } /* @@ -4988,20 +5736,17 @@ __init int register_ftrace_command(struct ftrace_func_command *cmd) __init int unregister_ftrace_command(struct ftrace_func_command *cmd) { struct ftrace_func_command *p, *n; - int ret = -ENODEV; - mutex_lock(&ftrace_cmd_mutex); + guard(mutex)(&ftrace_cmd_mutex); + list_for_each_entry_safe(p, n, &ftrace_commands, list) { if (strcmp(cmd->name, p->name) == 0) { - ret = 0; list_del_init(&p->list); - goto out_unlock; + return 0; } } - out_unlock: - mutex_unlock(&ftrace_cmd_mutex); - return ret; + return -ENODEV; } static int ftrace_process_regex(struct ftrace_iterator *iter, @@ -5011,7 +5756,7 @@ static int ftrace_process_regex(struct ftrace_iterator *iter, struct trace_array *tr = iter->ops->private; char *func, *command, *next = buff; struct ftrace_func_command *p; - int ret = -EINVAL; + int ret; func = strsep(&next, ":"); @@ -5028,17 +5773,14 @@ static int ftrace_process_regex(struct ftrace_iterator *iter, command = strsep(&next, ":"); - mutex_lock(&ftrace_cmd_mutex); + guard(mutex)(&ftrace_cmd_mutex); + list_for_each_entry(p, &ftrace_commands, list) { - if (strcmp(p->name, command) == 0) { - ret = p->func(tr, hash, func, command, next, enable); - goto out_unlock; - } + if (strcmp(p->name, command) == 0) + return p->func(tr, hash, func, command, next, enable); } - out_unlock: - mutex_unlock(&ftrace_cmd_mutex); - return ret; + return -EINVAL; } static ssize_t @@ -5072,12 +5814,10 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, parser->idx, enable); trace_parser_clear(parser); if (ret < 0) - goto out; + return ret; } - ret = read; - out: - return ret; + return read; } ssize_t @@ -5109,9 +5849,13 @@ __ftrace_match_addr(struct ftrace_hash *hash, unsigned long ip, int remove) return -ENOENT; free_hash_entry(hash, entry); return 0; + } else if (__ftrace_lookup_ip(hash, ip) != NULL) { + /* Already exists */ + return 0; } - return add_hash_entry(hash, ip); + entry = add_hash_entry(hash, ip); + return entry ? 0 : -ENOMEM; } static int @@ -5137,7 +5881,7 @@ ftrace_match_addr(struct ftrace_hash *hash, unsigned long *ips, static int ftrace_set_hash(struct ftrace_ops *ops, unsigned char *buf, int len, unsigned long *ips, unsigned int cnt, - int remove, int reset, int enable) + int remove, int reset, int enable, char *mod) { struct ftrace_hash **orig_hash; struct ftrace_hash *hash; @@ -5163,7 +5907,15 @@ ftrace_set_hash(struct ftrace_ops *ops, unsigned char *buf, int len, goto out_regex_unlock; } - if (buf && !ftrace_match_records(hash, buf, len)) { + if (buf && !match_records(hash, buf, len, mod)) { + /* If this was for a module and nothing was enabled, flag it */ + if (mod) + (*orig_hash)->flags |= FTRACE_HASH_FL_MOD; + + /* + * Even if it is a mod, return error to let caller know + * nothing was added + */ ret = -EINVAL; goto out_regex_unlock; } @@ -5188,401 +5940,24 @@ static int ftrace_set_addr(struct ftrace_ops *ops, unsigned long *ips, unsigned int cnt, int remove, int reset, int enable) { - return ftrace_set_hash(ops, NULL, 0, ips, cnt, remove, reset, enable); + return ftrace_set_hash(ops, NULL, 0, ips, cnt, remove, reset, enable, NULL); } #ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS -struct ftrace_direct_func { - struct list_head next; - unsigned long addr; - int count; -}; - -static LIST_HEAD(ftrace_direct_funcs); - -/** - * ftrace_find_direct_func - test an address if it is a registered direct caller - * @addr: The address of a registered direct caller - * - * This searches to see if a ftrace direct caller has been registered - * at a specific address, and if so, it returns a descriptor for it. - * - * This can be used by architecture code to see if an address is - * a direct caller (trampoline) attached to a fentry/mcount location. - * This is useful for the function_graph tracer, as it may need to - * do adjustments if it traced a location that also has a direct - * trampoline attached to it. - */ -struct ftrace_direct_func *ftrace_find_direct_func(unsigned long addr) -{ - struct ftrace_direct_func *entry; - bool found = false; - - /* May be called by fgraph trampoline (protected by rcu tasks) */ - list_for_each_entry_rcu(entry, &ftrace_direct_funcs, next) { - if (entry->addr == addr) { - found = true; - break; - } - } - if (found) - return entry; - - return NULL; -} - -static struct ftrace_direct_func *ftrace_alloc_direct_func(unsigned long addr) -{ - struct ftrace_direct_func *direct; - - direct = kmalloc(sizeof(*direct), GFP_KERNEL); - if (!direct) - return NULL; - direct->addr = addr; - direct->count = 0; - list_add_rcu(&direct->next, &ftrace_direct_funcs); - ftrace_direct_func_count++; - return direct; -} - static int register_ftrace_function_nolock(struct ftrace_ops *ops); -/** - * register_ftrace_direct - Call a custom trampoline directly - * @ip: The address of the nop at the beginning of a function - * @addr: The address of the trampoline to call at @ip - * - * This is used to connect a direct call from the nop location (@ip) - * at the start of ftrace traced functions. The location that it calls - * (@addr) must be able to handle a direct call, and save the parameters - * of the function being traced, and restore them (or inject new ones - * if needed), before returning. - * - * Returns: - * 0 on success - * -EBUSY - Another direct function is already attached (there can be only one) - * -ENODEV - @ip does not point to a ftrace nop location (or not supported) - * -ENOMEM - There was an allocation failure. - */ -int register_ftrace_direct(unsigned long ip, unsigned long addr) -{ - struct ftrace_direct_func *direct; - struct ftrace_func_entry *entry; - struct ftrace_hash *free_hash = NULL; - struct dyn_ftrace *rec; - int ret = -ENODEV; - - mutex_lock(&direct_mutex); - - ip = ftrace_location(ip); - if (!ip) - goto out_unlock; - - /* See if there's a direct function at @ip already */ - ret = -EBUSY; - if (ftrace_find_rec_direct(ip)) - goto out_unlock; - - ret = -ENODEV; - rec = lookup_rec(ip, ip); - if (!rec) - goto out_unlock; - - /* - * Check if the rec says it has a direct call but we didn't - * find one earlier? - */ - if (WARN_ON(rec->flags & FTRACE_FL_DIRECT)) - goto out_unlock; - - /* Make sure the ip points to the exact record */ - if (ip != rec->ip) { - ip = rec->ip; - /* Need to check this ip for a direct. */ - if (ftrace_find_rec_direct(ip)) - goto out_unlock; - } - - ret = -ENOMEM; - direct = ftrace_find_direct_func(addr); - if (!direct) { - direct = ftrace_alloc_direct_func(addr); - if (!direct) - goto out_unlock; - } - - entry = ftrace_add_rec_direct(ip, addr, &free_hash); - if (!entry) - goto out_unlock; - - ret = ftrace_set_filter_ip(&direct_ops, ip, 0, 0); - - if (!ret && !(direct_ops.flags & FTRACE_OPS_FL_ENABLED)) { - ret = register_ftrace_function_nolock(&direct_ops); - if (ret) - ftrace_set_filter_ip(&direct_ops, ip, 1, 0); - } - - if (ret) { - remove_hash_entry(direct_functions, entry); - kfree(entry); - if (!direct->count) { - list_del_rcu(&direct->next); - synchronize_rcu_tasks(); - kfree(direct); - if (free_hash) - free_ftrace_hash(free_hash); - free_hash = NULL; - ftrace_direct_func_count--; - } - } else { - direct->count++; - } - out_unlock: - mutex_unlock(&direct_mutex); - - if (free_hash) { - synchronize_rcu_tasks(); - free_ftrace_hash(free_hash); - } - - return ret; -} -EXPORT_SYMBOL_GPL(register_ftrace_direct); - -static struct ftrace_func_entry *find_direct_entry(unsigned long *ip, - struct dyn_ftrace **recp) -{ - struct ftrace_func_entry *entry; - struct dyn_ftrace *rec; - - rec = lookup_rec(*ip, *ip); - if (!rec) - return NULL; - - entry = __ftrace_lookup_ip(direct_functions, rec->ip); - if (!entry) { - WARN_ON(rec->flags & FTRACE_FL_DIRECT); - return NULL; - } - - WARN_ON(!(rec->flags & FTRACE_FL_DIRECT)); - - /* Passed in ip just needs to be on the call site */ - *ip = rec->ip; - - if (recp) - *recp = rec; - - return entry; -} - -int unregister_ftrace_direct(unsigned long ip, unsigned long addr) -{ - struct ftrace_direct_func *direct; - struct ftrace_func_entry *entry; - struct ftrace_hash *hash; - int ret = -ENODEV; - - mutex_lock(&direct_mutex); - - ip = ftrace_location(ip); - if (!ip) - goto out_unlock; - - entry = find_direct_entry(&ip, NULL); - if (!entry) - goto out_unlock; - - hash = direct_ops.func_hash->filter_hash; - if (hash->count == 1) - unregister_ftrace_function(&direct_ops); - - ret = ftrace_set_filter_ip(&direct_ops, ip, 1, 0); - - WARN_ON(ret); - - remove_hash_entry(direct_functions, entry); - - direct = ftrace_find_direct_func(addr); - if (!WARN_ON(!direct)) { - /* This is the good path (see the ! before WARN) */ - direct->count--; - WARN_ON(direct->count < 0); - if (!direct->count) { - list_del_rcu(&direct->next); - synchronize_rcu_tasks(); - kfree(direct); - kfree(entry); - ftrace_direct_func_count--; - } - } - out_unlock: - mutex_unlock(&direct_mutex); - - return ret; -} -EXPORT_SYMBOL_GPL(unregister_ftrace_direct); - -static struct ftrace_ops stub_ops = { - .func = ftrace_stub, -}; - -/** - * ftrace_modify_direct_caller - modify ftrace nop directly - * @entry: The ftrace hash entry of the direct helper for @rec - * @rec: The record representing the function site to patch - * @old_addr: The location that the site at @rec->ip currently calls - * @new_addr: The location that the site at @rec->ip should call - * - * An architecture may overwrite this function to optimize the - * changing of the direct callback on an ftrace nop location. - * This is called with the ftrace_lock mutex held, and no other - * ftrace callbacks are on the associated record (@rec). Thus, - * it is safe to modify the ftrace record, where it should be - * currently calling @old_addr directly, to call @new_addr. - * - * This is called with direct_mutex locked. - * - * Safety checks should be made to make sure that the code at - * @rec->ip is currently calling @old_addr. And this must - * also update entry->direct to @new_addr. - */ -int __weak ftrace_modify_direct_caller(struct ftrace_func_entry *entry, - struct dyn_ftrace *rec, - unsigned long old_addr, - unsigned long new_addr) -{ - unsigned long ip = rec->ip; - int ret; - - lockdep_assert_held(&direct_mutex); - - /* - * The ftrace_lock was used to determine if the record - * had more than one registered user to it. If it did, - * we needed to prevent that from changing to do the quick - * switch. But if it did not (only a direct caller was attached) - * then this function is called. But this function can deal - * with attached callers to the rec that we care about, and - * since this function uses standard ftrace calls that take - * the ftrace_lock mutex, we need to release it. - */ - mutex_unlock(&ftrace_lock); - - /* - * By setting a stub function at the same address, we force - * the code to call the iterator and the direct_ops helper. - * This means that @ip does not call the direct call, and - * we can simply modify it. - */ - ret = ftrace_set_filter_ip(&stub_ops, ip, 0, 0); - if (ret) - goto out_lock; - - ret = register_ftrace_function_nolock(&stub_ops); - if (ret) { - ftrace_set_filter_ip(&stub_ops, ip, 1, 0); - goto out_lock; - } - - entry->direct = new_addr; - - /* - * By removing the stub, we put back the direct call, calling - * the @new_addr. - */ - unregister_ftrace_function(&stub_ops); - ftrace_set_filter_ip(&stub_ops, ip, 1, 0); - - out_lock: - mutex_lock(&ftrace_lock); - - return ret; -} - -/** - * modify_ftrace_direct - Modify an existing direct call to call something else - * @ip: The instruction pointer to modify - * @old_addr: The address that the current @ip calls directly - * @new_addr: The address that the @ip should call - * - * This modifies a ftrace direct caller at an instruction pointer without - * having to disable it first. The direct call will switch over to the - * @new_addr without missing anything. - * - * Returns: zero on success. Non zero on error, which includes: - * -ENODEV : the @ip given has no direct caller attached - * -EINVAL : the @old_addr does not match the current direct caller +/* + * If there are multiple ftrace_ops, use SAVE_REGS by default, so that direct + * call will be jumped from ftrace_regs_caller. Only if the architecture does + * not support ftrace_regs_caller but direct_call, use SAVE_ARGS so that it + * jumps from ftrace_caller for multiple ftrace_ops. */ -int modify_ftrace_direct(unsigned long ip, - unsigned long old_addr, unsigned long new_addr) -{ - struct ftrace_direct_func *direct, *new_direct = NULL; - struct ftrace_func_entry *entry; - struct dyn_ftrace *rec; - int ret = -ENODEV; - - mutex_lock(&direct_mutex); - - mutex_lock(&ftrace_lock); - - ip = ftrace_location(ip); - if (!ip) - goto out_unlock; - - entry = find_direct_entry(&ip, &rec); - if (!entry) - goto out_unlock; - - ret = -EINVAL; - if (entry->direct != old_addr) - goto out_unlock; - - direct = ftrace_find_direct_func(old_addr); - if (WARN_ON(!direct)) - goto out_unlock; - if (direct->count > 1) { - ret = -ENOMEM; - new_direct = ftrace_alloc_direct_func(new_addr); - if (!new_direct) - goto out_unlock; - direct->count--; - new_direct->count++; - } else { - direct->addr = new_addr; - } - - /* - * If there's no other ftrace callback on the rec->ip location, - * then it can be changed directly by the architecture. - * If there is another caller, then we just need to change the - * direct caller helper to point to @new_addr. - */ - if (ftrace_rec_count(rec) == 1) { - ret = ftrace_modify_direct_caller(entry, rec, old_addr, new_addr); - } else { - entry->direct = new_addr; - ret = 0; - } - - if (unlikely(ret && new_direct)) { - direct->count++; - list_del_rcu(&new_direct->next); - synchronize_rcu_tasks(); - kfree(new_direct); - ftrace_direct_func_count--; - } - - out_unlock: - mutex_unlock(&ftrace_lock); - mutex_unlock(&direct_mutex); - return ret; -} -EXPORT_SYMBOL_GPL(modify_ftrace_direct); - +#ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_REGS +#define MULTI_FLAGS (FTRACE_OPS_FL_DIRECT | FTRACE_OPS_FL_SAVE_ARGS) +#else #define MULTI_FLAGS (FTRACE_OPS_FL_DIRECT | FTRACE_OPS_FL_SAVE_REGS) +#endif static int check_direct_multi(struct ftrace_ops *ops) { @@ -5602,7 +5977,8 @@ static void remove_direct_functions_hash(struct ftrace_hash *hash, unsigned long for (i = 0; i < size; i++) { hlist_for_each_entry(entry, &hash->buckets[i], hlist) { del = __ftrace_lookup_ip(direct_functions, entry->ip); - if (del && del->direct == addr) { + if (del && ftrace_jmp_get(del->direct) == + ftrace_jmp_get(addr)) { remove_hash_entry(direct_functions, del); kfree(del); } @@ -5610,8 +5986,26 @@ static void remove_direct_functions_hash(struct ftrace_hash *hash, unsigned long } } +static void register_ftrace_direct_cb(struct rcu_head *rhp) +{ + struct ftrace_hash *fhp = container_of(rhp, struct ftrace_hash, rcu); + + free_ftrace_hash(fhp); +} + +static void reset_direct(struct ftrace_ops *ops, unsigned long addr) +{ + struct ftrace_hash *hash = ops->func_hash->filter_hash; + + remove_direct_functions_hash(hash, addr); + + /* cleanup for possible another register call */ + ops->func = NULL; + ops->trampoline = 0; +} + /** - * register_ftrace_direct_multi - Call a custom trampoline directly + * register_ftrace_direct - Call a custom trampoline directly * for multiple functions registered in @ops * @ops: The address of the struct ftrace_ops object * @addr: The address of the trampoline to call at @ops functions @@ -5632,9 +6026,9 @@ static void remove_direct_functions_hash(struct ftrace_hash *hash, unsigned long * -ENODEV - @ip does not point to a ftrace nop location (or not supported) * -ENOMEM - There was an allocation failure. */ -int register_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) +int register_ftrace_direct(struct ftrace_ops *ops, unsigned long addr) { - struct ftrace_hash *hash, *free_hash = NULL; + struct ftrace_hash *hash, *new_hash = NULL, *free_hash = NULL; struct ftrace_func_entry *entry, *new; int err = -EBUSY, size, i; @@ -5649,8 +6043,15 @@ int register_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) if (ftrace_hash_empty(hash)) return -EINVAL; + /* This is a "raw" address, and this should never happen. */ + if (WARN_ON_ONCE(ftrace_is_jmp(addr))) + return -EINVAL; + mutex_lock(&direct_mutex); + if (ops->flags & FTRACE_OPS_FL_JMP) + addr = ftrace_jmp_set(addr); + /* Make sure requested entries are not already registered.. */ size = 1 << hash->size_bits; for (i = 0; i < size; i++) { @@ -5660,42 +6061,73 @@ int register_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) } } - /* ... and insert them to direct_functions hash. */ err = -ENOMEM; + + /* Make a copy hash to place the new and the old entries in */ + size = hash->count + direct_functions->count; + size = fls(size); + if (size > FTRACE_HASH_MAX_BITS) + size = FTRACE_HASH_MAX_BITS; + new_hash = alloc_ftrace_hash(size); + if (!new_hash) + goto out_unlock; + + /* Now copy over the existing direct entries */ + size = 1 << direct_functions->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &direct_functions->buckets[i], hlist) { + new = add_hash_entry(new_hash, entry->ip); + if (!new) + goto out_unlock; + new->direct = entry->direct; + } + } + + /* ... and add the new entries */ + size = 1 << hash->size_bits; for (i = 0; i < size; i++) { hlist_for_each_entry(entry, &hash->buckets[i], hlist) { - new = ftrace_add_rec_direct(entry->ip, addr, &free_hash); + new = add_hash_entry(new_hash, entry->ip); if (!new) - goto out_remove; + goto out_unlock; + /* Update both the copy and the hash entry */ + new->direct = addr; entry->direct = addr; } } + free_hash = direct_functions; + rcu_assign_pointer(direct_functions, new_hash); + new_hash = NULL; + ops->func = call_direct_funcs; - ops->flags = MULTI_FLAGS; + ops->flags |= MULTI_FLAGS; ops->trampoline = FTRACE_REGS_ADDR; + ops->direct_call = addr; err = register_ftrace_function_nolock(ops); - - out_remove: if (err) - remove_direct_functions_hash(hash, addr); + reset_direct(ops, addr); out_unlock: mutex_unlock(&direct_mutex); - if (free_hash) { - synchronize_rcu_tasks(); - free_ftrace_hash(free_hash); - } + if (free_hash && free_hash != EMPTY_HASH) + call_rcu_tasks(&free_hash->rcu, register_ftrace_direct_cb); + + if (new_hash) + free_ftrace_hash(new_hash); + return err; } -EXPORT_SYMBOL_GPL(register_ftrace_direct_multi); +EXPORT_SYMBOL_GPL(register_ftrace_direct); /** - * unregister_ftrace_direct_multi - Remove calls to custom trampoline - * previously registered by register_ftrace_direct_multi for @ops object. + * unregister_ftrace_direct - Remove calls to custom trampoline + * previously registered by register_ftrace_direct for @ops object. * @ops: The address of the struct ftrace_ops object + * @addr: The address of the direct function that is called by the @ops functions + * @free_filters: Set to true to remove all filters for the ftrace_ops, false otherwise * * This is used to remove a direct calls to @addr from the nop locations * of the functions registered in @ops (with by ftrace_set_filter_ip @@ -5705,9 +6137,9 @@ EXPORT_SYMBOL_GPL(register_ftrace_direct_multi); * 0 on success * -EINVAL - The @ops object was not properly registered. */ -int unregister_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) +int unregister_ftrace_direct(struct ftrace_ops *ops, unsigned long addr, + bool free_filters) { - struct ftrace_hash *hash = ops->func_hash->filter_hash; int err; if (check_direct_multi(ops)) @@ -5717,20 +6149,19 @@ int unregister_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) mutex_lock(&direct_mutex); err = unregister_ftrace_function(ops); - remove_direct_functions_hash(hash, addr); + reset_direct(ops, addr); mutex_unlock(&direct_mutex); - /* cleanup for possible another register call */ - ops->func = NULL; - ops->trampoline = 0; + if (free_filters) + ftrace_free_filter(ops); return err; } -EXPORT_SYMBOL_GPL(unregister_ftrace_direct_multi); +EXPORT_SYMBOL_GPL(unregister_ftrace_direct); static int -__modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) +__modify_ftrace_direct(struct ftrace_ops *ops, unsigned long addr) { - struct ftrace_hash *hash; + struct ftrace_hash *hash = ops->func_hash->filter_hash; struct ftrace_func_entry *entry, *iter; static struct ftrace_ops tmp_ops = { .func = ftrace_stub, @@ -5741,21 +6172,37 @@ __modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) lockdep_assert_held_once(&direct_mutex); + /* This is a "raw" address, and this should never happen. */ + if (WARN_ON_ONCE(ftrace_is_jmp(addr))) + return -EINVAL; + + if (ops->flags & FTRACE_OPS_FL_JMP) + addr = ftrace_jmp_set(addr); + /* Enable the tmp_ops to have the same functions as the direct ops */ ftrace_ops_init(&tmp_ops); tmp_ops.func_hash = ops->func_hash; + tmp_ops.direct_call = addr; err = register_ftrace_function_nolock(&tmp_ops); if (err) return err; /* + * Call __ftrace_hash_update_ipmodify() here, so that we can call + * ops->ops_func for the ops. This is needed because the above + * register_ftrace_function_nolock() worked on tmp_ops. + */ + err = __ftrace_hash_update_ipmodify(ops, hash, hash, true); + if (err) + goto out; + + /* * Now the ftrace_ops_list_func() is called to do the direct callers. * We can safely change the direct functions attached to each entry. */ mutex_lock(&ftrace_lock); - hash = ops->func_hash->filter_hash; size = 1 << hash->size_bits; for (i = 0; i < size; i++) { hlist_for_each_entry(iter, &hash->buckets[i], hlist) { @@ -5765,9 +6212,12 @@ __modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) entry->direct = addr; } } + /* Prevent store tearing if a trampoline concurrently accesses the value */ + WRITE_ONCE(ops->direct_call, addr); mutex_unlock(&ftrace_lock); +out: /* Removing the tmp_ops will add the updated direct callers to the functions */ unregister_ftrace_function(&tmp_ops); @@ -5775,7 +6225,7 @@ __modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) } /** - * modify_ftrace_direct_multi_nolock - Modify an existing direct 'multi' call + * modify_ftrace_direct_nolock - Modify an existing direct 'multi' call * to call something else * @ops: The address of the struct ftrace_ops object * @addr: The address of the new trampoline to call at @ops functions @@ -5792,19 +6242,19 @@ __modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) * Returns: zero on success. Non zero on error, which includes: * -EINVAL - The @ops object was not properly registered. */ -int modify_ftrace_direct_multi_nolock(struct ftrace_ops *ops, unsigned long addr) +int modify_ftrace_direct_nolock(struct ftrace_ops *ops, unsigned long addr) { if (check_direct_multi(ops)) return -EINVAL; if (!(ops->flags & FTRACE_OPS_FL_ENABLED)) return -EINVAL; - return __modify_ftrace_direct_multi(ops, addr); + return __modify_ftrace_direct(ops, addr); } -EXPORT_SYMBOL_GPL(modify_ftrace_direct_multi_nolock); +EXPORT_SYMBOL_GPL(modify_ftrace_direct_nolock); /** - * modify_ftrace_direct_multi - Modify an existing direct 'multi' call + * modify_ftrace_direct - Modify an existing direct 'multi' call * to call something else * @ops: The address of the struct ftrace_ops object * @addr: The address of the new trampoline to call at @ops functions @@ -5818,7 +6268,7 @@ EXPORT_SYMBOL_GPL(modify_ftrace_direct_multi_nolock); * Returns: zero on success. Non zero on error, which includes: * -EINVAL - The @ops object was not properly registered. */ -int modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) +int modify_ftrace_direct(struct ftrace_ops *ops, unsigned long addr) { int err; @@ -5828,19 +6278,19 @@ int modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr) return -EINVAL; mutex_lock(&direct_mutex); - err = __modify_ftrace_direct_multi(ops, addr); + err = __modify_ftrace_direct(ops, addr); mutex_unlock(&direct_mutex); return err; } -EXPORT_SYMBOL_GPL(modify_ftrace_direct_multi); +EXPORT_SYMBOL_GPL(modify_ftrace_direct); #endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */ /** * ftrace_set_filter_ip - set a function to filter on in ftrace by address - * @ops - the ops to set the filter with - * @ip - the address to add to or remove from the filter. - * @remove - non zero to remove the ip from the filter - * @reset - non zero to reset all filters before applying this filter. + * @ops: the ops to set the filter with + * @ip: the address to add to or remove from the filter. + * @remove: non zero to remove the ip from the filter + * @reset: non zero to reset all filters before applying this filter. * * Filters denote which functions should be enabled when tracing is enabled * If @ip is NULL, it fails to update filter. @@ -5859,11 +6309,11 @@ EXPORT_SYMBOL_GPL(ftrace_set_filter_ip); /** * ftrace_set_filter_ips - set functions to filter on in ftrace by addresses - * @ops - the ops to set the filter with - * @ips - the array of addresses to add to or remove from the filter. - * @cnt - the number of addresses in @ips - * @remove - non zero to remove ips from the filter - * @reset - non zero to reset all filters before applying this filter. + * @ops: the ops to set the filter with + * @ips: the array of addresses to add to or remove from the filter. + * @cnt: the number of addresses in @ips + * @remove: non zero to remove ips from the filter + * @reset: non zero to reset all filters before applying this filter. * * Filters denote which functions should be enabled when tracing is enabled * If @ips array or any ip specified within is NULL , it fails to update filter. @@ -5882,7 +6332,7 @@ EXPORT_SYMBOL_GPL(ftrace_set_filter_ips); /** * ftrace_ops_set_global_filter - setup ops to use global filters - * @ops - the ops which will use the global filters + * @ops: the ops which will use the global filters * * ftrace users who need global function trace filtering should call this. * It can set the global filter only if ops were not initialized before. @@ -5901,15 +6351,46 @@ static int ftrace_set_regex(struct ftrace_ops *ops, unsigned char *buf, int len, int reset, int enable) { - return ftrace_set_hash(ops, buf, len, NULL, 0, 0, reset, enable); + char *mod = NULL, *func, *command, *next = buf; + char *tmp __free(kfree) = NULL; + struct trace_array *tr = ops->private; + int ret; + + func = strsep(&next, ":"); + + /* This can also handle :mod: parsing */ + if (next) { + if (!tr) + return -EINVAL; + + command = strsep(&next, ":"); + if (strcmp(command, "mod") != 0) + return -EINVAL; + + mod = next; + len = command - func; + /* Save the original func as ftrace_set_hash() can modify it */ + tmp = kstrdup(func, GFP_KERNEL); + } + + ret = ftrace_set_hash(ops, func, len, NULL, 0, 0, reset, enable, mod); + + if (tr && mod && ret < 0) { + /* Did tmp fail to allocate? */ + if (!tmp) + return -ENOMEM; + ret = cache_mod(tr, tmp, mod, enable); + } + + return ret; } /** * ftrace_set_filter - set a function to filter on in ftrace - * @ops - the ops to set the filter with - * @buf - the string that holds the function filter text. - * @len - the length of the string. - * @reset - non zero to reset all filters before applying this filter. + * @ops: the ops to set the filter with + * @buf: the string that holds the function filter text. + * @len: the length of the string. + * @reset: non-zero to reset all filters before applying this filter. * * Filters denote which functions should be enabled when tracing is enabled. * If @buf is NULL and reset is set, all functions will be enabled for tracing. @@ -5928,10 +6409,10 @@ EXPORT_SYMBOL_GPL(ftrace_set_filter); /** * ftrace_set_notrace - set a function to not trace in ftrace - * @ops - the ops to set the notrace filter with - * @buf - the string that holds the function notrace text. - * @len - the length of the string. - * @reset - non zero to reset all filters before applying this filter. + * @ops: the ops to set the notrace filter with + * @buf: the string that holds the function notrace text. + * @len: the length of the string. + * @reset: non-zero to reset all filters before applying this filter. * * Notrace Filters denote which functions should not be enabled when tracing * is enabled. If @buf is NULL and reset is set, all functions will be enabled @@ -5950,9 +6431,9 @@ int ftrace_set_notrace(struct ftrace_ops *ops, unsigned char *buf, EXPORT_SYMBOL_GPL(ftrace_set_notrace); /** * ftrace_set_global_filter - set a function to filter on with global tracers - * @buf - the string that holds the function filter text. - * @len - the length of the string. - * @reset - non zero to reset all filters before applying this filter. + * @buf: the string that holds the function filter text. + * @len: the length of the string. + * @reset: non-zero to reset all filters before applying this filter. * * Filters denote which functions should be enabled when tracing is enabled. * If @buf is NULL and reset is set, all functions will be enabled for tracing. @@ -5965,9 +6446,9 @@ EXPORT_SYMBOL_GPL(ftrace_set_global_filter); /** * ftrace_set_global_notrace - set a function to not trace with global tracers - * @buf - the string that holds the function notrace text. - * @len - the length of the string. - * @reset - non zero to reset all filters before applying this filter. + * @buf: the string that holds the function notrace text. + * @len: the length of the string. + * @reset: non-zero to reset all filters before applying this filter. * * Notrace Filters denote which functions should not be enabled when tracing * is enabled. If @buf is NULL and reset is set, all functions will be enabled @@ -5992,7 +6473,7 @@ bool ftrace_filter_param __initdata; static int __init set_ftrace_notrace(char *str) { ftrace_filter_param = true; - strlcpy(ftrace_notrace_buf, str, FTRACE_FILTER_SIZE); + strscpy(ftrace_notrace_buf, str, FTRACE_FILTER_SIZE); return 1; } __setup("ftrace_notrace=", set_ftrace_notrace); @@ -6000,7 +6481,7 @@ __setup("ftrace_notrace=", set_ftrace_notrace); static int __init set_ftrace_filter(char *str) { ftrace_filter_param = true; - strlcpy(ftrace_filter_buf, str, FTRACE_FILTER_SIZE); + strscpy(ftrace_filter_buf, str, FTRACE_FILTER_SIZE); return 1; } __setup("ftrace_filter=", set_ftrace_filter); @@ -6012,23 +6493,22 @@ static int ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer); static int __init set_graph_function(char *str) { - strlcpy(ftrace_graph_buf, str, FTRACE_FILTER_SIZE); + strscpy(ftrace_graph_buf, str, FTRACE_FILTER_SIZE); return 1; } __setup("ftrace_graph_filter=", set_graph_function); static int __init set_graph_notrace_function(char *str) { - strlcpy(ftrace_graph_notrace_buf, str, FTRACE_FILTER_SIZE); + strscpy(ftrace_graph_notrace_buf, str, FTRACE_FILTER_SIZE); return 1; } __setup("ftrace_graph_notrace=", set_graph_notrace_function); static int __init set_graph_max_depth_function(char *str) { - if (!str) + if (!str || kstrtouint(str, 0, &fgraph_max_depth)) return 0; - fgraph_max_depth = simple_strtoul(str, NULL, 0); return 1; } __setup("ftrace_graph_max_depth=", set_graph_max_depth_function); @@ -6066,6 +6546,14 @@ ftrace_set_early_filter(struct ftrace_ops *ops, char *buf, int enable) ftrace_ops_init(ops); + /* The trace_array is needed for caching module function filters */ + if (!ops->private) { + struct trace_array *tr = trace_get_global_array(); + + ops->private = tr; + ftrace_init_trace_array(tr); + } + while (buf) { func = strsep(&buf, ","); ftrace_set_regex(ops, func, strlen(func), 0, enable); @@ -6130,9 +6618,6 @@ int ftrace_regex_release(struct inode *inode, struct file *file) ftrace_hash_move_and_update_ops(iter->ops, orig_hash, iter->hash, filter_hash); mutex_unlock(&ftrace_lock); - } else { - /* For read only, the hash is the ops hash */ - iter->hash = NULL; } mutex_unlock(&iter->ops->func_hash->regex_lock); @@ -6158,6 +6643,20 @@ static const struct file_operations ftrace_enabled_fops = { .release = seq_release_private, }; +static const struct file_operations ftrace_touched_fops = { + .open = ftrace_touched_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +static const struct file_operations ftrace_avail_addrs_fops = { + .open = ftrace_avail_addrs_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + static const struct file_operations ftrace_filter_fops = { .open = ftrace_filter_open, .read = seq_read, @@ -6491,12 +6990,10 @@ ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer) func_g.len = strlen(func_g.search); - mutex_lock(&ftrace_lock); + guard(mutex)(&ftrace_lock); - if (unlikely(ftrace_disabled)) { - mutex_unlock(&ftrace_lock); + if (unlikely(ftrace_disabled)) return -ENODEV; - } do_for_each_ftrace_rec(pg, rec) { @@ -6511,8 +7008,8 @@ ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer) if (entry) continue; - if (add_hash_entry(hash, rec->ip) < 0) - goto out; + if (add_hash_entry(hash, rec->ip) == NULL) + return 0; } else { if (entry) { free_hash_entry(hash, entry); @@ -6520,14 +7017,10 @@ ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer) } } } + cond_resched(); } while_for_each_ftrace_rec(); -out: - mutex_unlock(&ftrace_lock); - - if (fail) - return -EINVAL; - return 0; + return fail ? -EINVAL : 0; } static ssize_t @@ -6619,9 +7112,15 @@ static __init int ftrace_init_dyn_tracefs(struct dentry *d_tracer) trace_create_file("available_filter_functions", TRACE_MODE_READ, d_tracer, NULL, &ftrace_avail_fops); + trace_create_file("available_filter_functions_addrs", TRACE_MODE_READ, + d_tracer, NULL, &ftrace_avail_addrs_fops); + trace_create_file("enabled_functions", TRACE_MODE_READ, d_tracer, NULL, &ftrace_enabled_fops); + trace_create_file("touched_functions", TRACE_MODE_READ, + d_tracer, NULL, &ftrace_touched_fops); + ftrace_create_filter_files(&global_ops, d_tracer); #ifdef CONFIG_FUNCTION_GRAPH_TRACER @@ -6673,13 +7172,16 @@ static int ftrace_process_locs(struct module *mod, unsigned long *start, unsigned long *end) { + struct ftrace_page *pg_unuse = NULL; struct ftrace_page *start_pg; struct ftrace_page *pg; struct dyn_ftrace *rec; + unsigned long skipped = 0; unsigned long count; unsigned long *p; unsigned long addr; unsigned long flags = 0; /* Shut up gcc */ + unsigned long pages; int ret = -ENOMEM; count = end - start; @@ -6687,6 +7189,8 @@ static int ftrace_process_locs(struct module *mod, if (!count) return 0; + pages = DIV_ROUND_UP(count, ENTRIES_PER_PAGE); + /* * Sorting mcount in vmlinux at build time depend on * CONFIG_BUILDTIME_MCOUNT_SORT, while mcount loc in @@ -6731,15 +7235,32 @@ static int ftrace_process_locs(struct module *mod, pg = start_pg; while (p < end) { unsigned long end_offset; - addr = ftrace_call_adjust(*p++); + + addr = *p++; + /* * Some architecture linkers will pad between * the different mcount_loc sections of different * object files to satisfy alignments. * Skip any NULL pointers. */ - if (!addr) + if (!addr) { + skipped++; + continue; + } + + /* + * If this is core kernel, make sure the address is in core + * or inittext, as weak functions get zeroed and KASLR can + * move them to something other than zero. It just will not + * move it to an area where kernel text is. + */ + if (!mod && !(is_kernel_text(addr) || is_kernel_inittext(addr))) { + skipped++; continue; + } + + addr = ftrace_call_adjust(addr); end_offset = (pg->index+1) * sizeof(pg->records[0]); if (end_offset > PAGE_SIZE << pg->order) { @@ -6753,8 +7274,10 @@ static int ftrace_process_locs(struct module *mod, rec->ip = addr; } - /* We should have used all pages */ - WARN_ON(pg->next); + if (pg->next) { + pg_unuse = pg->next; + pg->next = NULL; + } /* Assign the last page to ftrace_pages */ ftrace_pages = pg; @@ -6776,6 +7299,43 @@ static int ftrace_process_locs(struct module *mod, out: mutex_unlock(&ftrace_lock); + /* We should have used all pages unless we skipped some */ + if (pg_unuse) { + unsigned long pg_remaining, remaining = 0; + unsigned long skip; + + /* Count the number of entries unused and compare it to skipped. */ + pg_remaining = (ENTRIES_PER_PAGE << pg->order) - pg->index; + + if (!WARN(skipped < pg_remaining, "Extra allocated pages for ftrace")) { + + skip = skipped - pg_remaining; + + for (pg = pg_unuse; pg; pg = pg->next) + remaining += 1 << pg->order; + + pages -= remaining; + + skip = DIV_ROUND_UP(skip, ENTRIES_PER_PAGE); + + /* + * Check to see if the number of pages remaining would + * just fit the number of entries skipped. + */ + WARN(skip != remaining, "Extra allocated pages for ftrace: %lu with %lu skipped", + remaining, skipped); + } + /* Need to synchronize with ftrace_location_range() */ + synchronize_rcu(); + ftrace_free_pages(pg_unuse); + } + + if (!mod) { + count -= skipped; + pr_info("ftrace: allocating %ld entries in %ld pages\n", + count, pages); + } + return ret; } @@ -6808,8 +7368,8 @@ static int ftrace_get_trampoline_kallsym(unsigned int symnum, continue; *value = op->trampoline; *type = 't'; - strlcpy(name, FTRACE_TRAMPOLINE_SYM, KSYM_NAME_LEN); - strlcpy(module_name, FTRACE_TRAMPOLINE_MOD, MODULE_NAME_LEN); + strscpy(name, FTRACE_TRAMPOLINE_SYM, KSYM_NAME_LEN); + strscpy(module_name, FTRACE_TRAMPOLINE_MOD, MODULE_NAME_LEN); *exported = 0; return 0; } @@ -6946,9 +7506,10 @@ void ftrace_release_mod(struct module *mod) mutex_lock(&ftrace_lock); - if (ftrace_disabled) - goto out_unlock; - + /* + * To avoid the UAF problem after the module is unloaded, the + * 'mod_map' resource needs to be released unconditionally. + */ list_for_each_entry_safe(mod_map, n, &ftrace_mod_maps, list) { if (mod_map->mod == mod) { list_del_rcu(&mod_map->list); @@ -6957,6 +7518,9 @@ void ftrace_release_mod(struct module *mod) } } + if (ftrace_disabled) + goto out_unlock; + /* * Each module has its own ftrace_pages, remove * them from the list. @@ -6964,8 +7528,7 @@ void ftrace_release_mod(struct module *mod) last_pg = &ftrace_pages_start; for (pg = ftrace_pages_start; pg; pg = *last_pg) { rec = &pg->records[0]; - if (within_module_core(rec->ip, mod) || - within_module_init(rec->ip, mod)) { + if (within_module(rec->ip, mod)) { /* * As core pages are first, the first * page should never be a module page. @@ -6988,6 +7551,9 @@ void ftrace_release_mod(struct module *mod) out_unlock: mutex_unlock(&ftrace_lock); + /* Need to synchronize with ftrace_location_range() */ + if (tmp_page) + synchronize_rcu(); for (pg = tmp_page; pg; pg = tmp_page) { /* Needs to be called outside of ftrace_lock */ @@ -7037,10 +7603,11 @@ void ftrace_module_enable(struct module *mod) * not part of this module, then skip this pg, * which the "break" will do. */ - if (!within_module_core(rec->ip, mod) && - !within_module_init(rec->ip, mod)) + if (!within_module(rec->ip, mod)) break; + cond_resched(); + /* Weak functions should still be ignored */ if (!test_for_valid_rec(rec)) { /* Clear all other flags. Should not be enabled anyway */ @@ -7134,6 +7701,9 @@ allocate_ftrace_mod_map(struct module *mod, { struct ftrace_mod_map *mod_map; + if (ftrace_disabled) + return NULL; + mod_map = kmalloc(sizeof(*mod_map), GFP_KERNEL); if (!mod_map) return NULL; @@ -7150,7 +7720,7 @@ allocate_ftrace_mod_map(struct module *mod, return mod_map; } -static const char * +static int ftrace_func_address_lookup(struct ftrace_mod_map *mod_map, unsigned long addr, unsigned long *size, unsigned long *off, char *sym) @@ -7171,21 +7741,18 @@ ftrace_func_address_lookup(struct ftrace_mod_map *mod_map, *size = found_func->size; if (off) *off = addr - found_func->ip; - if (sym) - strlcpy(sym, found_func->name, KSYM_NAME_LEN); - - return found_func->name; + return strscpy(sym, found_func->name, KSYM_NAME_LEN); } - return NULL; + return 0; } -const char * +int ftrace_mod_address_lookup(unsigned long addr, unsigned long *size, unsigned long *off, char **modname, char *sym) { struct ftrace_mod_map *mod_map; - const char *ret = NULL; + int ret = 0; /* mod_map is freed via call_rcu() */ preempt_disable(); @@ -7226,8 +7793,8 @@ int ftrace_mod_get_kallsym(unsigned int symnum, unsigned long *value, *value = mod_func->ip; *type = 'T'; - strlcpy(name, mod_func->name, KSYM_NAME_LEN); - strlcpy(module_name, mod_map->mod->name, MODULE_NAME_LEN); + strscpy(name, mod_func->name, KSYM_NAME_LEN); + strscpy(module_name, mod_map->mod->name, MODULE_NAME_LEN); *exported = 1; preempt_enable(); return 0; @@ -7322,14 +7889,13 @@ void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) unsigned long start = (unsigned long)(start_ptr); unsigned long end = (unsigned long)(end_ptr); struct ftrace_page **last_pg = &ftrace_pages_start; + struct ftrace_page *tmp_page = NULL; struct ftrace_page *pg; struct dyn_ftrace *rec; struct dyn_ftrace key; struct ftrace_mod_map *mod_map = NULL; struct ftrace_init_func *func, *func_next; - struct list_head clear_hash; - - INIT_LIST_HEAD(&clear_hash); + LIST_HEAD(clear_hash); key.ip = start; key.flags = end; /* overload flags, as it is unsigned long */ @@ -7365,12 +7931,8 @@ void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) ftrace_update_tot_cnt--; if (!pg->index) { *last_pg = pg->next; - if (pg->records) { - free_pages((unsigned long)pg->records, pg->order); - ftrace_number_of_pages -= 1 << pg->order; - } - ftrace_number_of_groups--; - kfree(pg); + pg->next = tmp_page; + tmp_page = pg; pg = container_of(last_pg, struct ftrace_page, next); if (!(*last_pg)) ftrace_pages = pg; @@ -7387,6 +7949,11 @@ void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) clear_func_from_hashes(func); kfree(func); } + /* Need to synchronize with ftrace_location_range() */ + if (tmp_page) { + synchronize_rcu(); + ftrace_free_pages(tmp_page); + } } void __init ftrace_free_init_mem(void) @@ -7423,9 +7990,6 @@ void __init ftrace_init(void) goto failed; } - pr_info("ftrace: allocating %ld entries in %ld pages\n", - count, DIV_ROUND_UP(count, ENTRIES_PER_PAGE)); - ret = ftrace_process_locs(NULL, __start_mcount_loc, __stop_mcount_loc); @@ -7475,9 +8039,14 @@ static void ftrace_update_trampoline(struct ftrace_ops *ops) void ftrace_init_trace_array(struct trace_array *tr) { + if (tr->flags & TRACE_ARRAY_FL_MOD_INIT) + return; + INIT_LIST_HEAD(&tr->func_probes); INIT_LIST_HEAD(&tr->mod_trace); INIT_LIST_HEAD(&tr->mod_notrace); + + tr->flags |= TRACE_ARRAY_FL_MOD_INIT; } #else @@ -7506,8 +8075,10 @@ static void ftrace_update_trampoline(struct ftrace_ops *ops) __init void ftrace_init_global_array_ops(struct trace_array *tr) { tr->ops = &global_ops; - tr->ops->private = tr; + if (!global_ops.private) + global_ops.private = tr; ftrace_init_trace_array(tr); + init_array_fgraph_ops(tr, tr->ops); } void ftrace_init_array_ops(struct trace_array *tr, ftrace_func_t func) @@ -7588,6 +8159,7 @@ out: void arch_ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs) { + kmsan_unpoison_memory(fregs, ftrace_regs_size()); __ftrace_ops_list_func(ip, parent_ip, NULL, fregs); } #else @@ -7628,7 +8200,7 @@ NOKPROBE_SYMBOL(ftrace_ops_assist_func); * have its own recursion protection, then it should call the * ftrace_ops_assist_func() instead. * - * Returns the function that the trampoline should call for @ops. + * Returns: the function that the trampoline should call for @ops. */ ftrace_func_t ftrace_ops_get_func(struct ftrace_ops *ops) { @@ -7946,7 +8518,7 @@ pid_write(struct file *filp, const char __user *ubuf, if (!cnt) return 0; - mutex_lock(&ftrace_lock); + guard(mutex)(&ftrace_lock); switch (type) { case TRACE_PIDS: @@ -7962,14 +8534,13 @@ pid_write(struct file *filp, const char __user *ubuf, lockdep_is_held(&ftrace_lock)); break; default: - ret = -EINVAL; WARN_ON_ONCE(1); - goto out; + return -EINVAL; } ret = trace_pid_write(filtered_pids, &pid_list, ubuf, cnt); if (ret < 0) - goto out; + return ret; switch (type) { case TRACE_PIDS: @@ -7998,11 +8569,8 @@ pid_write(struct file *filp, const char __user *ubuf, ftrace_update_pid_func(); ftrace_startup_all(0); - out: - mutex_unlock(&ftrace_lock); - if (ret > 0) - *ppos += ret; + *ppos += ret; return ret; } @@ -8077,12 +8645,13 @@ void ftrace_kill(void) ftrace_disabled = 1; ftrace_enabled = 0; ftrace_trace_function = ftrace_stub; + kprobe_ftrace_kill(); } /** * ftrace_is_dead - Test if ftrace is dead or not. * - * Returns 1 if ftrace is "dead", zero otherwise. + * Returns: 1 if ftrace is "dead", zero otherwise. */ int ftrace_is_dead(void) { @@ -8292,8 +8861,7 @@ struct kallsyms_data { * and returns 1 in case we resolved all the requested symbols, * 0 otherwise. */ -static int kallsyms_callback(void *data, const char *name, - struct module *mod, unsigned long addr) +static int kallsyms_callback(void *data, const char *name, unsigned long addr) { struct kallsyms_data *args = data; const char **sym; @@ -8328,8 +8896,7 @@ static int kallsyms_callback(void *data, const char *name, * @addrs array, which needs to be big enough to store at least @cnt * addresses. * - * This function returns 0 if all provided symbols are found, - * -ESRCH otherwise. + * Returns: 0 if all provided symbols are found, -ESRCH otherwise. */ int ftrace_lookup_symbols(const char **sorted_syms, size_t cnt, unsigned long *addrs) { @@ -8345,7 +8912,7 @@ int ftrace_lookup_symbols(const char **sorted_syms, size_t cnt, unsigned long *a found_all = kallsyms_on_each_symbol(kallsyms_callback, &args); if (found_all) return 0; - found_all = module_kallsyms_on_each_symbol(kallsyms_callback, &args); + found_all = module_kallsyms_on_each_symbol(NULL, kallsyms_callback, &args); return found_all ? 0 : -ESRCH; } @@ -8403,20 +8970,20 @@ static bool is_permanent_ops_registered(void) } static int -ftrace_enable_sysctl(struct ctl_table *table, int write, +ftrace_enable_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { - int ret = -ENODEV; + int ret; - mutex_lock(&ftrace_lock); + guard(mutex)(&ftrace_lock); if (unlikely(ftrace_disabled)) - goto out; + return -ENODEV; ret = proc_dointvec(table, write, buffer, lenp, ppos); if (ret || !write || (last_ftrace_enabled == !!ftrace_enabled)) - goto out; + return ret; if (ftrace_enabled) { @@ -8430,8 +8997,7 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, } else { if (is_permanent_ops_registered()) { ftrace_enabled = true; - ret = -EBUSY; - goto out; + return -EBUSY; } /* stopping ftrace calls (just send to ftrace_stub) */ @@ -8441,12 +9007,10 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, } last_ftrace_enabled = !!ftrace_enabled; - out: - mutex_unlock(&ftrace_lock); - return ret; + return 0; } -static struct ctl_table ftrace_sysctls[] = { +static const struct ctl_table ftrace_sysctls[] = { { .procname = "ftrace_enabled", .data = &ftrace_enabled, @@ -8454,7 +9018,6 @@ static struct ctl_table ftrace_sysctls[] = { .mode = 0644, .proc_handler = ftrace_enable_sysctl, }, - {} }; static int __init ftrace_sysctl_init(void) diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index 382775edf690..3235470e61b3 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -2,6 +2,9 @@ #ifndef _LINUX_KERNEL_FTRACE_INTERNAL_H #define _LINUX_KERNEL_FTRACE_INTERNAL_H +int __register_ftrace_function(struct ftrace_ops *ops); +int __unregister_ftrace_function(struct ftrace_ops *ops); + #ifdef CONFIG_FUNCTION_TRACER extern struct mutex ftrace_lock; @@ -12,11 +15,11 @@ extern struct ftrace_ops global_ops; int ftrace_startup(struct ftrace_ops *ops, int command); int ftrace_shutdown(struct ftrace_ops *ops, int command); int ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs); +int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); +int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command); #else /* !CONFIG_DYNAMIC_FTRACE */ -int __register_ftrace_function(struct ftrace_ops *ops); -int __unregister_ftrace_function(struct ftrace_ops *ops); /* Keep as macros so we do not need to define the commands */ # define ftrace_startup(ops, command) \ ({ \ @@ -37,14 +40,26 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) { return 1; } +static inline int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + return -EINVAL; +} +static inline int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command) +{ + return -EINVAL; +} #endif /* CONFIG_DYNAMIC_FTRACE */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER extern int ftrace_graph_active; -void update_function_graph_func(void); +# ifdef CONFIG_DYNAMIC_FTRACE +extern void fgraph_update_pid_func(void); +# else +static inline void fgraph_update_pid_func(void) {} +# endif #else /* !CONFIG_FUNCTION_GRAPH_TRACER */ # define ftrace_graph_active 0 -static inline void update_function_graph_func(void) { } +static inline void fgraph_update_pid_func(void) {} #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ #else /* !CONFIG_FUNCTION_TRACER */ diff --git a/kernel/trace/kprobe_event_gen_test.c b/kernel/trace/kprobe_event_gen_test.c index c736487fc0e4..5a4b722b5045 100644 --- a/kernel/trace/kprobe_event_gen_test.c +++ b/kernel/trace/kprobe_event_gen_test.c @@ -21,7 +21,7 @@ * Then: * * # insmod kernel/trace/kprobe_event_gen_test.ko - * # cat /sys/kernel/debug/tracing/trace + * # cat /sys/kernel/tracing/trace * * You should see many instances of the "gen_kprobe_test" and * "gen_kretprobe_test" events in the trace buffer. @@ -146,7 +146,7 @@ static int __init test_gen_kprobe_cmd(void) if (trace_event_file_is_valid(gen_kprobe_test)) gen_kprobe_test = NULL; /* We got an error after creating the event, delete it */ - ret = kprobe_event_delete("gen_kprobe_test"); + kprobe_event_delete("gen_kprobe_test"); goto out; } @@ -211,7 +211,7 @@ static int __init test_gen_kretprobe_cmd(void) if (trace_event_file_is_valid(gen_kretprobe_test)) gen_kretprobe_test = NULL; /* We got an error after creating the event, delete it */ - ret = kprobe_event_delete("gen_kretprobe_test"); + kprobe_event_delete("gen_kretprobe_test"); goto out; } diff --git a/kernel/trace/pid_list.c b/kernel/trace/pid_list.c index 95106d02b32d..dbee72d69d0a 100644 --- a/kernel/trace/pid_list.c +++ b/kernel/trace/pid_list.c @@ -3,6 +3,7 @@ * Copyright (C) 2021 VMware Inc, Steven Rostedt <rostedt@goodmis.org> */ #include <linux/spinlock.h> +#include <linux/seqlock.h> #include <linux/irq_work.h> #include <linux/slab.h> #include "trace.h" @@ -81,13 +82,9 @@ static inline bool upper_empty(union upper_chunk *chunk) { /* * If chunk->data has no lower chunks, it will be the same - * as a zeroed bitmask. Use find_first_bit() to test it - * and if it doesn't find any bits set, then the array - * is empty. + * as a zeroed bitmask. */ - int bit = find_first_bit((unsigned long *)chunk->data, - sizeof(chunk->data) * 8); - return bit >= sizeof(chunk->data) * 8; + return bitmap_empty((unsigned long *)chunk->data, BITS_PER_TYPE(chunk->data)); } static inline int pid_split(unsigned int pid, unsigned int *upper1, @@ -130,7 +127,7 @@ bool trace_pid_list_is_set(struct trace_pid_list *pid_list, unsigned int pid) { union upper_chunk *upper_chunk; union lower_chunk *lower_chunk; - unsigned long flags; + unsigned int seq; unsigned int upper1; unsigned int upper2; unsigned int lower; @@ -142,14 +139,16 @@ bool trace_pid_list_is_set(struct trace_pid_list *pid_list, unsigned int pid) if (pid_split(pid, &upper1, &upper2, &lower) < 0) return false; - raw_spin_lock_irqsave(&pid_list->lock, flags); - upper_chunk = pid_list->upper[upper1]; - if (upper_chunk) { - lower_chunk = upper_chunk->data[upper2]; - if (lower_chunk) - ret = test_bit(lower, lower_chunk->data); - } - raw_spin_unlock_irqrestore(&pid_list->lock, flags); + do { + seq = read_seqcount_begin(&pid_list->seqcount); + ret = false; + upper_chunk = pid_list->upper[upper1]; + if (upper_chunk) { + lower_chunk = upper_chunk->data[upper2]; + if (lower_chunk) + ret = test_bit(lower, lower_chunk->data); + } + } while (read_seqcount_retry(&pid_list->seqcount, seq)); return ret; } @@ -182,6 +181,7 @@ int trace_pid_list_set(struct trace_pid_list *pid_list, unsigned int pid) return -EINVAL; raw_spin_lock_irqsave(&pid_list->lock, flags); + write_seqcount_begin(&pid_list->seqcount); upper_chunk = pid_list->upper[upper1]; if (!upper_chunk) { upper_chunk = get_upper_chunk(pid_list); @@ -203,6 +203,7 @@ int trace_pid_list_set(struct trace_pid_list *pid_list, unsigned int pid) set_bit(lower, lower_chunk->data); ret = 0; out: + write_seqcount_end(&pid_list->seqcount); raw_spin_unlock_irqrestore(&pid_list->lock, flags); return ret; } @@ -234,6 +235,7 @@ int trace_pid_list_clear(struct trace_pid_list *pid_list, unsigned int pid) return -EINVAL; raw_spin_lock_irqsave(&pid_list->lock, flags); + write_seqcount_begin(&pid_list->seqcount); upper_chunk = pid_list->upper[upper1]; if (!upper_chunk) goto out; @@ -254,6 +256,7 @@ int trace_pid_list_clear(struct trace_pid_list *pid_list, unsigned int pid) } } out: + write_seqcount_end(&pid_list->seqcount); raw_spin_unlock_irqrestore(&pid_list->lock, flags); return 0; } @@ -344,8 +347,10 @@ static void pid_list_refill_irq(struct irq_work *iwork) again: raw_spin_lock(&pid_list->lock); + write_seqcount_begin(&pid_list->seqcount); upper_count = CHUNK_ALLOC - pid_list->free_upper_chunks; lower_count = CHUNK_ALLOC - pid_list->free_lower_chunks; + write_seqcount_end(&pid_list->seqcount); raw_spin_unlock(&pid_list->lock); if (upper_count <= 0 && lower_count <= 0) @@ -354,7 +359,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) while (upper_count-- > 0) { union upper_chunk *chunk; - chunk = kzalloc(sizeof(*chunk), GFP_KERNEL); + chunk = kzalloc(sizeof(*chunk), GFP_NOWAIT); if (!chunk) break; *upper_next = chunk; @@ -365,7 +370,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) while (lower_count-- > 0) { union lower_chunk *chunk; - chunk = kzalloc(sizeof(*chunk), GFP_KERNEL); + chunk = kzalloc(sizeof(*chunk), GFP_NOWAIT); if (!chunk) break; *lower_next = chunk; @@ -374,6 +379,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) } raw_spin_lock(&pid_list->lock); + write_seqcount_begin(&pid_list->seqcount); if (upper) { *upper_next = pid_list->upper_list; pid_list->upper_list = upper; @@ -384,6 +390,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) pid_list->lower_list = lower; pid_list->free_lower_chunks += lcnt; } + write_seqcount_end(&pid_list->seqcount); raw_spin_unlock(&pid_list->lock); /* @@ -414,7 +421,7 @@ struct trace_pid_list *trace_pid_list_alloc(void) int i; /* According to linux/thread.h, pids can be no bigger that 30 bits */ - WARN_ON_ONCE(pid_max > (1 << 30)); + WARN_ON_ONCE(init_pid_ns.pid_max > (1 << 30)); pid_list = kzalloc(sizeof(*pid_list), GFP_KERNEL); if (!pid_list) @@ -423,6 +430,7 @@ struct trace_pid_list *trace_pid_list_alloc(void) init_irq_work(&pid_list->refill_irqwork, pid_list_refill_irq); raw_spin_lock_init(&pid_list->lock); + seqcount_raw_spinlock_init(&pid_list->seqcount, &pid_list->lock); for (i = 0; i < CHUNK_ALLOC; i++) { union upper_chunk *chunk; @@ -451,6 +459,7 @@ struct trace_pid_list *trace_pid_list_alloc(void) /** * trace_pid_list_free - Frees an allocated pid_list. + * @pid_list: The pid list to free. * * Frees the memory for a pid_list that was allocated. */ diff --git a/kernel/trace/pid_list.h b/kernel/trace/pid_list.h index 62e73f1ac85f..0b45fb0eadb9 100644 --- a/kernel/trace/pid_list.h +++ b/kernel/trace/pid_list.h @@ -76,6 +76,7 @@ union upper_chunk { }; struct trace_pid_list { + seqcount_raw_spinlock_t seqcount; raw_spinlock_t lock; struct irq_work refill_irqwork; union upper_chunk *upper[UPPER1_SIZE]; // 1 or 2K in size diff --git a/kernel/trace/power-traces.c b/kernel/trace/power-traces.c index 21bb161c2316..f2fe33573e54 100644 --- a/kernel/trace/power-traces.c +++ b/kernel/trace/power-traces.c @@ -17,5 +17,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(suspend_resume); EXPORT_TRACEPOINT_SYMBOL_GPL(cpu_idle); EXPORT_TRACEPOINT_SYMBOL_GPL(cpu_frequency); -EXPORT_TRACEPOINT_SYMBOL_GPL(powernv_throttle); diff --git a/kernel/trace/preemptirq_delay_test.c b/kernel/trace/preemptirq_delay_test.c index 8c4ffd076162..acb0c971a408 100644 --- a/kernel/trace/preemptirq_delay_test.c +++ b/kernel/trace/preemptirq_delay_test.c @@ -34,8 +34,6 @@ MODULE_PARM_DESC(cpu_affinity, "Cpu num test is running on"); static struct completion done; -#define MIN(x, y) ((x) < (y) ? (x) : (y)) - static void busy_wait(ulong time) { u64 start, end; @@ -119,12 +117,15 @@ static int preemptirq_delay_run(void *data) { int i; int s = MIN(burst_size, NR_TEST_FUNCS); - struct cpumask cpu_mask; + cpumask_var_t cpu_mask; + + if (!alloc_cpumask_var(&cpu_mask, GFP_KERNEL)) + return -ENOMEM; if (cpu_affinity > -1) { - cpumask_clear(&cpu_mask); - cpumask_set_cpu(cpu_affinity, &cpu_mask); - if (set_cpus_allowed_ptr(current, &cpu_mask)) + cpumask_clear(cpu_mask); + cpumask_set_cpu(cpu_affinity, cpu_mask); + if (set_cpus_allowed_ptr(current, cpu_mask)) pr_err("cpu_affinity:%d, failed\n", cpu_affinity); } @@ -141,6 +142,8 @@ static int preemptirq_delay_run(void *data) __set_current_state(TASK_RUNNING); + free_cpumask_var(cpu_mask); + return 0; } @@ -215,4 +218,5 @@ static void __exit preemptirq_delay_exit(void) module_init(preemptirq_delay_init) module_exit(preemptirq_delay_exit) +MODULE_DESCRIPTION("Preempt / IRQ disable delay thread to test latency tracers"); MODULE_LICENSE("GPL v2"); diff --git a/kernel/trace/rethook.c b/kernel/trace/rethook.c index 32c3dfdb4d6a..30d224946881 100644 --- a/kernel/trace/rethook.c +++ b/kernel/trace/rethook.c @@ -8,7 +8,6 @@ #include <linux/preempt.h> #include <linux/rethook.h> #include <linux/slab.h> -#include <linux/sort.h> /* Return hook list (shadow stack by list) */ @@ -36,21 +35,20 @@ void rethook_flush_task(struct task_struct *tk) static void rethook_free_rcu(struct rcu_head *head) { struct rethook *rh = container_of(head, struct rethook, rcu); - struct rethook_node *rhn; - struct freelist_node *node; - int count = 1; - - node = rh->pool.head; - while (node) { - rhn = container_of(node, struct rethook_node, freelist); - node = node->next; - kfree(rhn); - count++; - } + objpool_fini(&rh->pool); +} - /* The rh->ref is the number of pooled node + 1 */ - if (refcount_sub_and_test(count, &rh->ref)) - kfree(rh); +/** + * rethook_stop() - Stop using a rethook. + * @rh: the struct rethook to stop. + * + * Stop using a rethook to prepare for freeing it. If you want to wait for + * all running rethook handler before calling rethook_free(), you need to + * call this first and wait RCU, and call rethook_free(). + */ +void rethook_stop(struct rethook *rh) +{ + rcu_assign_pointer(rh->handler, NULL); } /** @@ -65,59 +63,73 @@ static void rethook_free_rcu(struct rcu_head *head) */ void rethook_free(struct rethook *rh) { - WRITE_ONCE(rh->handler, NULL); + rethook_stop(rh); call_rcu(&rh->rcu, rethook_free_rcu); } +static int rethook_init_node(void *nod, void *context) +{ + struct rethook_node *node = nod; + + node->rethook = context; + return 0; +} + +static int rethook_fini_pool(struct objpool_head *head, void *context) +{ + kfree(context); + return 0; +} + +static inline rethook_handler_t rethook_get_handler(struct rethook *rh) +{ + return (rethook_handler_t)rcu_dereference_check(rh->handler, + rcu_read_lock_any_held()); +} + /** * rethook_alloc() - Allocate struct rethook. * @data: a data to pass the @handler when hooking the return. - * @handler: the return hook callback function. + * @handler: the return hook callback function, must NOT be NULL + * @size: node size: rethook node and additional data + * @num: number of rethook nodes to be preallocated * * Allocate and initialize a new rethook with @data and @handler. - * Return NULL if memory allocation fails or @handler is NULL. + * Return pointer of new rethook, or error codes for failures. + * * Note that @handler == NULL means this rethook is going to be freed. */ -struct rethook *rethook_alloc(void *data, rethook_handler_t handler) +struct rethook *rethook_alloc(void *data, rethook_handler_t handler, + int size, int num) { - struct rethook *rh = kzalloc(sizeof(struct rethook), GFP_KERNEL); + struct rethook *rh; - if (!rh || !handler) { - kfree(rh); - return NULL; - } + if (!handler || num <= 0 || size < sizeof(struct rethook_node)) + return ERR_PTR(-EINVAL); + + rh = kzalloc(sizeof(struct rethook), GFP_KERNEL); + if (!rh) + return ERR_PTR(-ENOMEM); rh->data = data; - rh->handler = handler; - rh->pool.head = NULL; - refcount_set(&rh->ref, 1); + rcu_assign_pointer(rh->handler, handler); + /* initialize the objpool for rethook nodes */ + if (objpool_init(&rh->pool, num, size, GFP_KERNEL, rh, + rethook_init_node, rethook_fini_pool)) { + kfree(rh); + return ERR_PTR(-ENOMEM); + } return rh; } -/** - * rethook_add_node() - Add a new node to the rethook. - * @rh: the struct rethook. - * @node: the struct rethook_node to be added. - * - * Add @node to @rh. User must allocate @node (as a part of user's - * data structure.) The @node fields are initialized in this function. - */ -void rethook_add_node(struct rethook *rh, struct rethook_node *node) -{ - node->rethook = rh; - freelist_add(&node->freelist, &rh->pool); - refcount_inc(&rh->ref); -} - static void free_rethook_node_rcu(struct rcu_head *head) { struct rethook_node *node = container_of(head, struct rethook_node, rcu); + struct rethook *rh = node->rethook; - if (refcount_dec_and_test(&node->rethook->ref)) - kfree(node->rethook); - kfree(node); + objpool_drop(node, &rh->pool); } /** @@ -129,10 +141,11 @@ static void free_rethook_node_rcu(struct rcu_head *head) */ void rethook_recycle(struct rethook_node *node) { - lockdep_assert_preemption_disabled(); + rethook_handler_t handler; - if (likely(READ_ONCE(node->rethook->handler))) - freelist_add(&node->freelist, &node->rethook->pool); + handler = rethook_get_handler(node->rethook); + if (likely(handler)) + objpool_push(node, &node->rethook->pool); else call_rcu(&node->rcu, free_rethook_node_rcu); } @@ -147,15 +160,13 @@ NOKPROBE_SYMBOL(rethook_recycle); */ struct rethook_node *rethook_try_get(struct rethook *rh) { - rethook_handler_t handler = READ_ONCE(rh->handler); - struct freelist_node *fn; - - lockdep_assert_preemption_disabled(); + rethook_handler_t handler = rethook_get_handler(rh); /* Check whether @rh is going to be freed. */ if (unlikely(!handler)) return NULL; +#if defined(CONFIG_FTRACE_VALIDATE_RCU_IS_WATCHING) || defined(CONFIG_KPROBE_EVENTS_ON_NOTRACE) /* * This expects the caller will set up a rethook on a function entry. * When the function returns, the rethook will eventually be reclaimed @@ -164,12 +175,9 @@ struct rethook_node *rethook_try_get(struct rethook *rh) */ if (unlikely(!rcu_is_watching())) return NULL; +#endif - fn = freelist_try_get(&rh->pool); - if (!fn) - return NULL; - - return container_of(fn, struct rethook_node, freelist); + return (struct rethook_node *)objpool_pop(&rh->pool); } NOKPROBE_SYMBOL(rethook_try_get); @@ -242,7 +250,7 @@ unsigned long rethook_find_ret_addr(struct task_struct *tsk, unsigned long frame if (WARN_ON_ONCE(!cur)) return 0; - if (WARN_ON_ONCE(tsk != current && task_is_running(tsk))) + if (tsk != current && task_is_running(tsk)) return 0; do { @@ -288,7 +296,7 @@ unsigned long rethook_trampoline_handler(struct pt_regs *regs, * These loops must be protected from rethook_free_rcu() because those * are accessing 'rhn->rethook'. */ - preempt_disable(); + preempt_disable_notrace(); /* * Run the handler on the shadow stack. Do not unlink the list here because @@ -299,9 +307,10 @@ unsigned long rethook_trampoline_handler(struct pt_regs *regs, rhn = container_of(first, struct rethook_node, llist); if (WARN_ON_ONCE(rhn->frame != frame)) break; - handler = READ_ONCE(rhn->rethook->handler); + handler = rethook_get_handler(rhn->rethook); if (handler) - handler(rhn, rhn->rethook->data, regs); + handler(rhn, rhn->rethook->data, + correct_ret_addr, regs); if (first == node) break; @@ -321,7 +330,7 @@ unsigned long rethook_trampoline_handler(struct pt_regs *regs, first = first->next; rethook_recycle(rhn); } - preempt_enable(); + preempt_enable_notrace(); return correct_ret_addr; } diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index c366a0a9ddba..41c9f5d079be 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -9,6 +9,7 @@ #include <linux/ring_buffer.h> #include <linux/trace_clock.h> #include <linux/sched/clock.h> +#include <linux/cacheflush.h> #include <linux/trace_seq.h> #include <linux/spinlock.h> #include <linux/irq_work.h> @@ -26,8 +27,13 @@ #include <linux/list.h> #include <linux/cpu.h> #include <linux/oom.h> +#include <linux/mm.h> +#include <asm/local64.h> #include <asm/local.h> +#include <asm/setup.h> + +#include "trace.h" /* * The "absolute" timestamp in the buffer is only 59 bits. @@ -39,6 +45,24 @@ static void update_pages_handler(struct work_struct *work); +#define RING_BUFFER_META_MAGIC 0xBADFEED + +struct ring_buffer_meta { + int magic; + int struct_sizes; + unsigned long total_size; + unsigned long buffers_offset; +}; + +struct ring_buffer_cpu_meta { + unsigned long first_buffer; + unsigned long head_buffer; + unsigned long commit_buffer; + __u32 subbuf_size; + __u32 nr_subbufs; + int buffers[]; +}; + /* * The ring buffer header is special. We must manually up keep it. */ @@ -163,7 +187,7 @@ enum { #define extended_time(event) \ (event->type_len >= RINGBUF_TYPE_TIME_EXTEND) -static inline int rb_null_event(struct ring_buffer_event *event) +static inline bool rb_null_event(struct ring_buffer_event *event) { return event->type_len == RINGBUF_TYPE_PADDING && !event->time_delta; } @@ -311,12 +335,19 @@ static u64 rb_event_time_stamp(struct ring_buffer_event *event) /* Missed count stored at end */ #define RB_MISSED_STORED (1 << 30) +#define RB_MISSED_MASK (3 << 30) + struct buffer_data_page { u64 time_stamp; /* page time stamp */ local_t commit; /* write committed index */ unsigned char data[] RB_ALIGN_DATA; /* data of buffer page */ }; +struct buffer_data_read_page { + unsigned order; /* order of the page */ + struct buffer_data_page *data; /* actual data, stored in this page */ +}; + /* * Note, the buffer_page list must be first. The buffer pages * are allocated in cache lines, which means that each buffer @@ -331,6 +362,9 @@ struct buffer_page { unsigned read; /* index for next read */ local_t entries; /* entries on this page */ unsigned long real_end; /* real end of data */ + unsigned order; /* order of the page */ + u32 id:30; /* ID for external mapping */ + u32 range:1; /* Mapped via a range */ struct buffer_data_page *page; /* Actual data page */ }; @@ -354,66 +388,67 @@ static void rb_init_page(struct buffer_data_page *bpage) local_set(&bpage->commit, 0); } -/* - * Also stolen from mm/slob.c. Thanks to Mathieu Desnoyers for pointing - * this issue out. - */ +static __always_inline unsigned int rb_page_commit(struct buffer_page *bpage) +{ + return local_read(&bpage->page->commit); +} + static void free_buffer_page(struct buffer_page *bpage) { - free_page((unsigned long)bpage->page); + /* Range pages are not to be freed */ + if (!bpage->range) + free_pages((unsigned long)bpage->page, bpage->order); kfree(bpage); } /* - * We need to fit the time_stamp delta into 27 bits. + * For best performance, allocate cpu buffer data cache line sized + * and per CPU. */ -static inline int test_time_stamp(u64 delta) -{ - if (delta & TS_DELTA_TEST) - return 1; - return 0; -} +#define alloc_cpu_buffer(cpu) (struct ring_buffer_per_cpu *) \ + kzalloc_node(ALIGN(sizeof(struct ring_buffer_per_cpu), \ + cache_line_size()), GFP_KERNEL, cpu_to_node(cpu)); -#define BUF_PAGE_SIZE (PAGE_SIZE - BUF_PAGE_HDR_SIZE) +#define alloc_cpu_page(cpu) (struct buffer_page *) \ + kzalloc_node(ALIGN(sizeof(struct buffer_page), \ + cache_line_size()), GFP_KERNEL, cpu_to_node(cpu)); -/* Max payload is BUF_PAGE_SIZE - header (8bytes) */ -#define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2)) - -int ring_buffer_print_page_header(struct trace_seq *s) +static struct buffer_data_page *alloc_cpu_data(int cpu, int order) { - struct buffer_data_page field; + struct buffer_data_page *dpage; + struct page *page; + gfp_t mflags; - trace_seq_printf(s, "\tfield: u64 timestamp;\t" - "offset:0;\tsize:%u;\tsigned:%u;\n", - (unsigned int)sizeof(field.time_stamp), - (unsigned int)is_signed_type(u64)); + /* + * __GFP_RETRY_MAYFAIL flag makes sure that the allocation fails + * gracefully without invoking oom-killer and the system is not + * destabilized. + */ + mflags = GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_COMP | __GFP_ZERO; - trace_seq_printf(s, "\tfield: local_t commit;\t" - "offset:%u;\tsize:%u;\tsigned:%u;\n", - (unsigned int)offsetof(typeof(field), commit), - (unsigned int)sizeof(field.commit), - (unsigned int)is_signed_type(long)); + page = alloc_pages_node(cpu_to_node(cpu), mflags, order); + if (!page) + return NULL; - trace_seq_printf(s, "\tfield: int overwrite;\t" - "offset:%u;\tsize:%u;\tsigned:%u;\n", - (unsigned int)offsetof(typeof(field), commit), - 1, - (unsigned int)is_signed_type(long)); + dpage = page_address(page); + rb_init_page(dpage); - trace_seq_printf(s, "\tfield: char data;\t" - "offset:%u;\tsize:%u;\tsigned:%u;\n", - (unsigned int)offsetof(typeof(field), data), - (unsigned int)BUF_PAGE_SIZE, - (unsigned int)is_signed_type(char)); + return dpage; +} - return !trace_seq_has_overflowed(s); +/* + * We need to fit the time_stamp delta into 27 bits. + */ +static inline bool test_time_stamp(u64 delta) +{ + return !!(delta & TS_DELTA_TEST); } struct rb_irq_work { struct irq_work work; wait_queue_head_t waiters; wait_queue_head_t full_waiters; - long wait_index; + atomic_t seq; bool waiters_pending; bool full_waiters_pending; bool wakeup_full; @@ -464,27 +499,9 @@ enum { RB_CTX_MAX }; -#if BITS_PER_LONG == 32 -#define RB_TIME_32 -#endif - -/* To test on 64 bit machines */ -//#define RB_TIME_32 - -#ifdef RB_TIME_32 - -struct rb_time_struct { - local_t cnt; - local_t top; - local_t bottom; - local_t msb; -}; -#else -#include <asm/local64.h> struct rb_time_struct { local64_t time; }; -#endif typedef struct rb_time_struct rb_time_t; #define MAX_NEST 5 @@ -504,6 +521,8 @@ struct ring_buffer_per_cpu { unsigned long nr_pages; unsigned int current_context; struct list_head *pages; + /* pages generation counter, incremented when the list changes */ + unsigned long cnt; struct buffer_page *head_page; /* read from head */ struct buffer_page *tail_page; /* write to tail */ struct buffer_page *commit_page; /* committed pages */ @@ -529,6 +548,16 @@ struct ring_buffer_per_cpu { rb_time_t before_stamp; u64 event_stamp[MAX_NEST]; u64 read_stamp; + /* pages removed since last reset */ + unsigned long pages_removed; + + unsigned int mapped; + unsigned int user_mapped; /* user space mapping */ + struct mutex mapping_lock; + unsigned long *subbuf_ids; /* ID to subbuf VA */ + struct trace_buffer_meta *meta_page; + struct ring_buffer_cpu_meta *ring_meta; + /* ring buffer pages to update, > 0 to add, < 0 to remove */ long nr_pages_to_update; struct list_head new_pages; /* new pages to add */ @@ -542,6 +571,7 @@ struct trace_buffer { unsigned flags; int cpus; atomic_t record_disabled; + atomic_t resizing; cpumask_var_t cpumask; struct lock_class_key *reader_lock_key; @@ -555,6 +585,15 @@ struct trace_buffer { struct rb_irq_work irq_work; bool time_stamp_abs; + + unsigned long range_addr_start; + unsigned long range_addr_end; + + struct ring_buffer_meta *meta; + + unsigned int subbuf_size; + unsigned int subbuf_order; + unsigned int max_data_size; }; struct ring_buffer_iter { @@ -564,202 +603,53 @@ struct ring_buffer_iter { struct buffer_page *head_page; struct buffer_page *cache_reader_page; unsigned long cache_read; + unsigned long cache_pages_removed; u64 read_stamp; u64 page_stamp; struct ring_buffer_event *event; + size_t event_size; int missed_events; }; -#ifdef RB_TIME_32 - -/* - * On 32 bit machines, local64_t is very expensive. As the ring - * buffer doesn't need all the features of a true 64 bit atomic, - * on 32 bit, it uses these functions (64 still uses local64_t). - * - * For the ring buffer, 64 bit required operations for the time is - * the following: - * - * - Reads may fail if it interrupted a modification of the time stamp. - * It will succeed if it did not interrupt another write even if - * the read itself is interrupted by a write. - * It returns whether it was successful or not. - * - * - Writes always succeed and will overwrite other writes and writes - * that were done by events interrupting the current write. - * - * - A write followed by a read of the same time stamp will always succeed, - * but may not contain the same value. - * - * - A cmpxchg will fail if it interrupted another write or cmpxchg. - * Other than that, it acts like a normal cmpxchg. - * - * The 60 bit time stamp is broken up by 30 bits in a top and bottom half - * (bottom being the least significant 30 bits of the 60 bit time stamp). - * - * The two most significant bits of each half holds a 2 bit counter (0-3). - * Each update will increment this counter by one. - * When reading the top and bottom, if the two counter bits match then the - * top and bottom together make a valid 60 bit number. - */ -#define RB_TIME_SHIFT 30 -#define RB_TIME_VAL_MASK ((1 << RB_TIME_SHIFT) - 1) -#define RB_TIME_MSB_SHIFT 60 - -static inline int rb_time_cnt(unsigned long val) -{ - return (val >> RB_TIME_SHIFT) & 3; -} - -static inline u64 rb_time_val(unsigned long top, unsigned long bottom) -{ - u64 val; - - val = top & RB_TIME_VAL_MASK; - val <<= RB_TIME_SHIFT; - val |= bottom & RB_TIME_VAL_MASK; - - return val; -} - -static inline bool __rb_time_read(rb_time_t *t, u64 *ret, unsigned long *cnt) +int ring_buffer_print_page_header(struct trace_buffer *buffer, struct trace_seq *s) { - unsigned long top, bottom, msb; - unsigned long c; - - /* - * If the read is interrupted by a write, then the cnt will - * be different. Loop until both top and bottom have been read - * without interruption. - */ - do { - c = local_read(&t->cnt); - top = local_read(&t->top); - bottom = local_read(&t->bottom); - msb = local_read(&t->msb); - } while (c != local_read(&t->cnt)); - - *cnt = rb_time_cnt(top); - - /* If top and bottom counts don't match, this interrupted a write */ - if (*cnt != rb_time_cnt(bottom)) - return false; - - /* The shift to msb will lose its cnt bits */ - *ret = rb_time_val(top, bottom) | ((u64)msb << RB_TIME_MSB_SHIFT); - return true; -} - -static bool rb_time_read(rb_time_t *t, u64 *ret) -{ - unsigned long cnt; - - return __rb_time_read(t, ret, &cnt); -} - -static inline unsigned long rb_time_val_cnt(unsigned long val, unsigned long cnt) -{ - return (val & RB_TIME_VAL_MASK) | ((cnt & 3) << RB_TIME_SHIFT); -} - -static inline void rb_time_split(u64 val, unsigned long *top, unsigned long *bottom, - unsigned long *msb) -{ - *top = (unsigned long)((val >> RB_TIME_SHIFT) & RB_TIME_VAL_MASK); - *bottom = (unsigned long)(val & RB_TIME_VAL_MASK); - *msb = (unsigned long)(val >> RB_TIME_MSB_SHIFT); -} - -static inline void rb_time_val_set(local_t *t, unsigned long val, unsigned long cnt) -{ - val = rb_time_val_cnt(val, cnt); - local_set(t, val); -} - -static void rb_time_set(rb_time_t *t, u64 val) -{ - unsigned long cnt, top, bottom, msb; - - rb_time_split(val, &top, &bottom, &msb); - - /* Writes always succeed with a valid number even if it gets interrupted. */ - do { - cnt = local_inc_return(&t->cnt); - rb_time_val_set(&t->top, top, cnt); - rb_time_val_set(&t->bottom, bottom, cnt); - rb_time_val_set(&t->msb, val >> RB_TIME_MSB_SHIFT, cnt); - } while (cnt != local_read(&t->cnt)); -} - -static inline bool -rb_time_read_cmpxchg(local_t *l, unsigned long expect, unsigned long set) -{ - unsigned long ret; - - ret = local_cmpxchg(l, expect, set); - return ret == expect; -} - -static int rb_time_cmpxchg(rb_time_t *t, u64 expect, u64 set) -{ - unsigned long cnt, top, bottom, msb; - unsigned long cnt2, top2, bottom2, msb2; - u64 val; - - /* The cmpxchg always fails if it interrupted an update */ - if (!__rb_time_read(t, &val, &cnt2)) - return false; - - if (val != expect) - return false; + struct buffer_data_page field; - cnt = local_read(&t->cnt); - if ((cnt & 3) != cnt2) - return false; + trace_seq_printf(s, "\tfield: u64 timestamp;\t" + "offset:0;\tsize:%u;\tsigned:%u;\n", + (unsigned int)sizeof(field.time_stamp), + (unsigned int)is_signed_type(u64)); - cnt2 = cnt + 1; + trace_seq_printf(s, "\tfield: local_t commit;\t" + "offset:%u;\tsize:%u;\tsigned:%u;\n", + (unsigned int)offsetof(typeof(field), commit), + (unsigned int)sizeof(field.commit), + (unsigned int)is_signed_type(long)); - rb_time_split(val, &top, &bottom, &msb); - top = rb_time_val_cnt(top, cnt); - bottom = rb_time_val_cnt(bottom, cnt); + trace_seq_printf(s, "\tfield: int overwrite;\t" + "offset:%u;\tsize:%u;\tsigned:%u;\n", + (unsigned int)offsetof(typeof(field), commit), + 1, + (unsigned int)is_signed_type(long)); - rb_time_split(set, &top2, &bottom2, &msb2); - top2 = rb_time_val_cnt(top2, cnt2); - bottom2 = rb_time_val_cnt(bottom2, cnt2); + trace_seq_printf(s, "\tfield: char data;\t" + "offset:%u;\tsize:%u;\tsigned:%u;\n", + (unsigned int)offsetof(typeof(field), data), + (unsigned int)buffer->subbuf_size, + (unsigned int)is_signed_type(char)); - if (!rb_time_read_cmpxchg(&t->cnt, cnt, cnt2)) - return false; - if (!rb_time_read_cmpxchg(&t->msb, msb, msb2)) - return false; - if (!rb_time_read_cmpxchg(&t->top, top, top2)) - return false; - if (!rb_time_read_cmpxchg(&t->bottom, bottom, bottom2)) - return false; - return true; + return !trace_seq_has_overflowed(s); } -#else /* 64 bits */ - -/* local64_t always succeeds */ - -static inline bool rb_time_read(rb_time_t *t, u64 *ret) +static inline void rb_time_read(rb_time_t *t, u64 *ret) { *ret = local64_read(&t->time); - return true; } static void rb_time_set(rb_time_t *t, u64 val) { local64_set(&t->time, val); } -static bool rb_time_cmpxchg(rb_time_t *t, u64 expect, u64 set) -{ - u64 val; - val = local64_cmpxchg(&t->time, expect, set); - return val == expect; -} -#endif - /* * Enable this to make sure that the event passed to * ring_buffer_event_time_stamp() is not committed and also @@ -865,27 +755,12 @@ u64 ring_buffer_event_time_stamp(struct trace_buffer *buffer, WARN_ONCE(1, "nest (%d) greater than max", nest); fail: - /* Can only fail on 32 bit */ - if (!rb_time_read(&cpu_buffer->write_stamp, &ts)) - /* Screw it, just read the current time */ - ts = rb_time_stamp(cpu_buffer->buffer); + rb_time_read(&cpu_buffer->write_stamp, &ts); return ts; } /** - * ring_buffer_nr_pages - get the number of buffer pages in the ring buffer - * @buffer: The ring_buffer to get the number of pages from - * @cpu: The cpu of the ring_buffer to get the number of pages from - * - * Returns the number of pages used by a per_cpu buffer of the ring buffer. - */ -size_t ring_buffer_nr_pages(struct trace_buffer *buffer, int cpu) -{ - return buffer->buffers[cpu]->nr_pages; -} - -/** * ring_buffer_nr_dirty_pages - get the number of used pages in the ring buffer * @buffer: The ring_buffer to get the number of pages from * @cpu: The cpu of the ring_buffer to get the number of pages from @@ -926,9 +801,14 @@ static __always_inline bool full_hit(struct trace_buffer *buffer, int cpu, int f if (!nr_pages || !full) return true; - dirty = ring_buffer_nr_dirty_pages(buffer, cpu); + /* + * Add one as dirty will never equal nr_pages, as the sub-buffer + * that the writer is on is not counted as dirty. + * This is needed if "buffer_percent" is set to 100. + */ + dirty = ring_buffer_nr_dirty_pages(buffer, cpu) + 1; - return (dirty * 100) > (full * nr_pages); + return (dirty * 100) >= (full * nr_pages); } /* @@ -941,10 +821,24 @@ static void rb_wake_up_waiters(struct irq_work *work) { struct rb_irq_work *rbwork = container_of(work, struct rb_irq_work, work); + /* For waiters waiting for the first wake up */ + (void)atomic_fetch_inc_release(&rbwork->seq); + wake_up_all(&rbwork->waiters); if (rbwork->full_waiters_pending || rbwork->wakeup_full) { + /* Only cpu_buffer sets the above flags */ + struct ring_buffer_per_cpu *cpu_buffer = + container_of(rbwork, struct ring_buffer_per_cpu, irq_work); + + /* Called from interrupt context */ + raw_spin_lock(&cpu_buffer->reader_lock); rbwork->wakeup_full = false; rbwork->full_waiters_pending = false; + + /* Waking up all waiters, they will reset the shortest full */ + cpu_buffer->shortest_full = 0; + raw_spin_unlock(&cpu_buffer->reader_lock); + wake_up_all(&rbwork->full_waiters); } } @@ -952,6 +846,7 @@ static void rb_wake_up_waiters(struct irq_work *work) /** * ring_buffer_wake_waiters - wake up any waiters on this ring buffer * @buffer: The ring buffer to wake waiters on + * @cpu: The CPU buffer to wake waiters on * * In the case of a file that represents a ring buffer is closing, * it is prudent to wake up any waiters that are on this. @@ -984,11 +879,94 @@ void ring_buffer_wake_waiters(struct trace_buffer *buffer, int cpu) rbwork = &cpu_buffer->irq_work; } - rbwork->wait_index++; - /* make sure the waiters see the new index */ - smp_wmb(); + /* This can be called in any context */ + irq_work_queue(&rbwork->work); +} + +static bool rb_watermark_hit(struct trace_buffer *buffer, int cpu, int full) +{ + struct ring_buffer_per_cpu *cpu_buffer; + bool ret = false; + + /* Reads of all CPUs always waits for any data */ + if (cpu == RING_BUFFER_ALL_CPUS) + return !ring_buffer_empty(buffer); - rb_wake_up_waiters(&rbwork->work); + cpu_buffer = buffer->buffers[cpu]; + + if (!ring_buffer_empty_cpu(buffer, cpu)) { + unsigned long flags; + bool pagebusy; + + if (!full) + return true; + + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + pagebusy = cpu_buffer->reader_page == cpu_buffer->commit_page; + ret = !pagebusy && full_hit(buffer, cpu, full); + + if (!ret && (!cpu_buffer->shortest_full || + cpu_buffer->shortest_full > full)) { + cpu_buffer->shortest_full = full; + } + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + } + return ret; +} + +static inline bool +rb_wait_cond(struct rb_irq_work *rbwork, struct trace_buffer *buffer, + int cpu, int full, ring_buffer_cond_fn cond, void *data) +{ + if (rb_watermark_hit(buffer, cpu, full)) + return true; + + if (cond(data)) + return true; + + /* + * The events can happen in critical sections where + * checking a work queue can cause deadlocks. + * After adding a task to the queue, this flag is set + * only to notify events to try to wake up the queue + * using irq_work. + * + * We don't clear it even if the buffer is no longer + * empty. The flag only causes the next event to run + * irq_work to do the work queue wake up. The worse + * that can happen if we race with !trace_empty() is that + * an event will cause an irq_work to try to wake up + * an empty queue. + * + * There's no reason to protect this flag either, as + * the work queue and irq_work logic will do the necessary + * synchronization for the wake ups. The only thing + * that is necessary is that the wake up happens after + * a task has been queued. It's OK for spurious wake ups. + */ + if (full) + rbwork->full_waiters_pending = true; + else + rbwork->waiters_pending = true; + + return false; +} + +struct rb_wait_data { + struct rb_irq_work *irq_work; + int seq; +}; + +/* + * The default wait condition for ring_buffer_wait() is to just to exit the + * wait loop the first time it is woken up. + */ +static bool rb_wait_once(void *data) +{ + struct rb_wait_data *rdata = data; + struct rb_irq_work *rbwork = rdata->irq_work; + + return atomic_read_acquire(&rbwork->seq) != rdata->seq; } /** @@ -996,17 +974,20 @@ void ring_buffer_wake_waiters(struct trace_buffer *buffer, int cpu) * @buffer: buffer to wait on * @cpu: the cpu buffer to wait on * @full: wait until the percentage of pages are available, if @cpu != RING_BUFFER_ALL_CPUS + * @cond: condition function to break out of wait (NULL to run once) + * @data: the data to pass to @cond. * * If @cpu == RING_BUFFER_ALL_CPUS then the task will wake up as soon * as data is added to any of the @buffer's cpu buffers. Otherwise * it will wait for data to be added to a specific cpu buffer. */ -int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full) +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full, + ring_buffer_cond_fn cond, void *data) { struct ring_buffer_per_cpu *cpu_buffer; - DEFINE_WAIT(wait); - struct rb_irq_work *work; - long wait_index; + struct wait_queue_head *waitq; + struct rb_irq_work *rbwork; + struct rb_wait_data rdata; int ret = 0; /* @@ -1015,90 +996,31 @@ int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full) * caller on the appropriate wait queue. */ if (cpu == RING_BUFFER_ALL_CPUS) { - work = &buffer->irq_work; + rbwork = &buffer->irq_work; /* Full only makes sense on per cpu reads */ full = 0; } else { if (!cpumask_test_cpu(cpu, buffer->cpumask)) return -ENODEV; cpu_buffer = buffer->buffers[cpu]; - work = &cpu_buffer->irq_work; - } - - wait_index = READ_ONCE(work->wait_index); - - while (true) { - if (full) - prepare_to_wait(&work->full_waiters, &wait, TASK_INTERRUPTIBLE); - else - prepare_to_wait(&work->waiters, &wait, TASK_INTERRUPTIBLE); - - /* - * The events can happen in critical sections where - * checking a work queue can cause deadlocks. - * After adding a task to the queue, this flag is set - * only to notify events to try to wake up the queue - * using irq_work. - * - * We don't clear it even if the buffer is no longer - * empty. The flag only causes the next event to run - * irq_work to do the work queue wake up. The worse - * that can happen if we race with !trace_empty() is that - * an event will cause an irq_work to try to wake up - * an empty queue. - * - * There's no reason to protect this flag either, as - * the work queue and irq_work logic will do the necessary - * synchronization for the wake ups. The only thing - * that is necessary is that the wake up happens after - * a task has been queued. It's OK for spurious wake ups. - */ - if (full) - work->full_waiters_pending = true; - else - work->waiters_pending = true; - - if (signal_pending(current)) { - ret = -EINTR; - break; - } - - if (cpu == RING_BUFFER_ALL_CPUS && !ring_buffer_empty(buffer)) - break; - - if (cpu != RING_BUFFER_ALL_CPUS && - !ring_buffer_empty_cpu(buffer, cpu)) { - unsigned long flags; - bool pagebusy; - bool done; - - if (!full) - break; - - raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); - pagebusy = cpu_buffer->reader_page == cpu_buffer->commit_page; - done = !pagebusy && full_hit(buffer, cpu, full); - - if (!cpu_buffer->shortest_full || - cpu_buffer->shortest_full > full) - cpu_buffer->shortest_full = full; - raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); - if (done) - break; - } - - schedule(); - - /* Make sure to see the new wait index */ - smp_rmb(); - if (wait_index != work->wait_index) - break; + rbwork = &cpu_buffer->irq_work; } if (full) - finish_wait(&work->full_waiters, &wait); + waitq = &rbwork->full_waiters; else - finish_wait(&work->waiters, &wait); + waitq = &rbwork->waiters; + + /* Set up to exit loop as soon as it is woken */ + if (!cond) { + cond = rb_wait_once; + rdata.irq_work = rbwork; + rdata.seq = atomic_read_acquire(&rbwork->seq); + data = &rdata; + } + + ret = wait_event_interruptible((*waitq), + rb_wait_cond(rbwork, buffer, cpu, full, cond, data)); return ret; } @@ -1122,27 +1044,44 @@ __poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu, struct file *filp, poll_table *poll_table, int full) { struct ring_buffer_per_cpu *cpu_buffer; - struct rb_irq_work *work; + struct rb_irq_work *rbwork; if (cpu == RING_BUFFER_ALL_CPUS) { - work = &buffer->irq_work; + rbwork = &buffer->irq_work; full = 0; } else { if (!cpumask_test_cpu(cpu, buffer->cpumask)) - return -EINVAL; + return EPOLLERR; cpu_buffer = buffer->buffers[cpu]; - work = &cpu_buffer->irq_work; + rbwork = &cpu_buffer->irq_work; } if (full) { - poll_wait(filp, &work->full_waiters, poll_table); - work->full_waiters_pending = true; - } else { - poll_wait(filp, &work->waiters, poll_table); - work->waiters_pending = true; + poll_wait(filp, &rbwork->full_waiters, poll_table); + + if (rb_watermark_hit(buffer, cpu, full)) + return EPOLLIN | EPOLLRDNORM; + /* + * Only allow full_waiters_pending update to be seen after + * the shortest_full is set (in rb_watermark_hit). If the + * writer sees the full_waiters_pending flag set, it will + * compare the amount in the ring buffer to shortest_full. + * If the amount in the ring buffer is greater than the + * shortest_full percent, it will call the irq_work handler + * to wake up this list. The irq_handler will reset shortest_full + * back to zero. That's done under the reader_lock, but + * the below smp_mb() makes sure that the update to + * full_waiters_pending doesn't leak up into the above. + */ + smp_mb(); + rbwork->full_waiters_pending = true; + return 0; } + poll_wait(filp, &rbwork->waiters, poll_table); + rbwork->waiters_pending = true; + /* * There's a tight race between setting the waiters_pending and * checking if the ring buffer is empty. Once the waiters_pending bit @@ -1158,9 +1097,6 @@ __poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu, */ smp_mb(); - if (full) - return full_hit(buffer, cpu, full) ? EPOLLIN | EPOLLRDNORM : 0; - if ((cpu == RING_BUFFER_ALL_CPUS && !ring_buffer_empty(buffer)) || (cpu != RING_BUFFER_ALL_CPUS && !ring_buffer_empty_cpu(buffer, cpu))) return EPOLLIN | EPOLLRDNORM; @@ -1191,7 +1127,7 @@ static inline u64 rb_time_stamp(struct trace_buffer *buffer) u64 ts; /* Skip retpolines :-( */ - if (IS_ENABLED(CONFIG_RETPOLINE) && likely(buffer->clock == trace_clock_local)) + if (IS_ENABLED(CONFIG_MITIGATION_RETPOLINE) && likely(buffer->clock == trace_clock_local)) ts = trace_clock_local(); else ts = buffer->clock(); @@ -1371,6 +1307,11 @@ static void rb_head_page_activate(struct ring_buffer_per_cpu *cpu_buffer) * Set the previous list pointer to have the HEAD flag. */ rb_set_list_to_head(head->list.prev); + + if (cpu_buffer->ring_meta) { + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + meta->head_buffer = (unsigned long)head->page; + } } static void rb_list_head_clear(struct list_head *list) @@ -1452,6 +1393,13 @@ static inline void rb_inc_page(struct buffer_page **bpage) *bpage = list_entry(p, struct buffer_page, list); } +static inline void rb_dec_page(struct buffer_page **bpage) +{ + struct list_head *p = rb_list_head((*bpage)->list.prev); + + *bpage = list_entry(p, struct buffer_page, list); +} + static struct buffer_page * rb_set_head_page(struct ring_buffer_per_cpu *cpu_buffer) { @@ -1490,19 +1438,16 @@ rb_set_head_page(struct ring_buffer_per_cpu *cpu_buffer) return NULL; } -static int rb_head_page_replace(struct buffer_page *old, +static bool rb_head_page_replace(struct buffer_page *old, struct buffer_page *new) { unsigned long *ptr = (unsigned long *)&old->list.prev->next; unsigned long val; - unsigned long ret; val = *ptr & ~RB_FLAG_MASK; val |= RB_PAGE_HEAD; - ret = cmpxchg(ptr, val, (unsigned long)&new->list); - - return ret == val; + return try_cmpxchg(ptr, &val, (unsigned long)&new->list); } /* @@ -1527,7 +1472,6 @@ static void rb_tail_page_update(struct ring_buffer_per_cpu *cpu_buffer, old_write = local_add_return(RB_WRITE_INTCNT, &next_page->write); old_entries = local_add_return(RB_WRITE_INTCNT, &next_page->entries); - local_inc(&cpu_buffer->pages_touched); /* * Just make sure we have seen our old_write and synchronize * with any interrupts that come in. @@ -1564,33 +1508,32 @@ static void rb_tail_page_update(struct ring_buffer_per_cpu *cpu_buffer, */ local_set(&next_page->page->commit, 0); - /* Again, either we update tail_page or an interrupt does */ - (void)cmpxchg(&cpu_buffer->tail_page, tail_page, next_page); + /* Either we update tail_page or an interrupt does */ + if (try_cmpxchg(&cpu_buffer->tail_page, &tail_page, next_page)) + local_inc(&cpu_buffer->pages_touched); } } -static int rb_check_bpage(struct ring_buffer_per_cpu *cpu_buffer, +static void rb_check_bpage(struct ring_buffer_per_cpu *cpu_buffer, struct buffer_page *bpage) { unsigned long val = (unsigned long)bpage; - if (RB_WARN_ON(cpu_buffer, val & RB_FLAG_MASK)) - return 1; - - return 0; + RB_WARN_ON(cpu_buffer, val & RB_FLAG_MASK); } -/** - * rb_check_list - make sure a pointer to a list has the last bits zero - */ -static int rb_check_list(struct ring_buffer_per_cpu *cpu_buffer, - struct list_head *list) +static bool rb_check_links(struct ring_buffer_per_cpu *cpu_buffer, + struct list_head *list) { - if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev) != list->prev)) - return 1; - if (RB_WARN_ON(cpu_buffer, rb_list_head(list->next) != list->next)) - return 1; - return 0; + if (RB_WARN_ON(cpu_buffer, + rb_list_head(rb_list_head(list->next)->prev) != list)) + return false; + + if (RB_WARN_ON(cpu_buffer, + rb_list_head(rb_list_head(list->prev)->next) != list)) + return false; + + return true; } /** @@ -1600,47 +1543,702 @@ static int rb_check_list(struct ring_buffer_per_cpu *cpu_buffer, * As a safety measure we check to make sure the data pages have not * been corrupted. */ -static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) +static void rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) { - struct list_head *head = cpu_buffer->pages; - struct buffer_page *bpage, *tmp; + struct list_head *head, *tmp; + unsigned long buffer_cnt; + unsigned long flags; + int nr_loops = 0; - /* Reset the head page if it exists */ - if (cpu_buffer->head_page) - rb_set_head_page(cpu_buffer); + /* + * Walk the linked list underpinning the ring buffer and validate all + * its next and prev links. + * + * The check acquires the reader_lock to avoid concurrent processing + * with code that could be modifying the list. However, the lock cannot + * be held for the entire duration of the walk, as this would make the + * time when interrupts are disabled non-deterministic, dependent on the + * ring buffer size. Therefore, the code releases and re-acquires the + * lock after checking each page. The ring_buffer_per_cpu.cnt variable + * is then used to detect if the list was modified while the lock was + * not held, in which case the check needs to be restarted. + * + * The code attempts to perform the check at most three times before + * giving up. This is acceptable because this is only a self-validation + * to detect problems early on. In practice, the list modification + * operations are fairly spaced, and so this check typically succeeds at + * most on the second try. + */ +again: + if (++nr_loops > 3) + return; - rb_head_page_deactivate(cpu_buffer); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + head = rb_list_head(cpu_buffer->pages); + if (!rb_check_links(cpu_buffer, head)) + goto out_locked; + buffer_cnt = cpu_buffer->cnt; + tmp = head; + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); - if (RB_WARN_ON(cpu_buffer, head->next->prev != head)) - return -1; - if (RB_WARN_ON(cpu_buffer, head->prev->next != head)) - return -1; + while (true) { + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); - if (rb_check_list(cpu_buffer, head)) - return -1; + if (buffer_cnt != cpu_buffer->cnt) { + /* The list was updated, try again. */ + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + goto again; + } - list_for_each_entry_safe(bpage, tmp, head, list) { - if (RB_WARN_ON(cpu_buffer, - bpage->list.next->prev != &bpage->list)) - return -1; - if (RB_WARN_ON(cpu_buffer, - bpage->list.prev->next != &bpage->list)) - return -1; - if (rb_check_list(cpu_buffer, &bpage->list)) + tmp = rb_list_head(tmp->next); + if (tmp == head) + /* The iteration circled back, all is done. */ + goto out_locked; + + if (!rb_check_links(cpu_buffer, tmp)) + goto out_locked; + + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + } + +out_locked: + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); +} + +/* + * Take an address, add the meta data size as well as the array of + * array subbuffer indexes, then align it to a subbuffer size. + * + * This is used to help find the next per cpu subbuffer within a mapped range. + */ +static unsigned long +rb_range_align_subbuf(unsigned long addr, int subbuf_size, int nr_subbufs) +{ + addr += sizeof(struct ring_buffer_cpu_meta) + + sizeof(int) * nr_subbufs; + return ALIGN(addr, subbuf_size); +} + +/* + * Return the ring_buffer_meta for a given @cpu. + */ +static void *rb_range_meta(struct trace_buffer *buffer, int nr_pages, int cpu) +{ + int subbuf_size = buffer->subbuf_size + BUF_PAGE_HDR_SIZE; + struct ring_buffer_cpu_meta *meta; + struct ring_buffer_meta *bmeta; + unsigned long ptr; + int nr_subbufs; + + bmeta = buffer->meta; + if (!bmeta) + return NULL; + + ptr = (unsigned long)bmeta + bmeta->buffers_offset; + meta = (struct ring_buffer_cpu_meta *)ptr; + + /* When nr_pages passed in is zero, the first meta has already been initialized */ + if (!nr_pages) { + nr_subbufs = meta->nr_subbufs; + } else { + /* Include the reader page */ + nr_subbufs = nr_pages + 1; + } + + /* + * The first chunk may not be subbuffer aligned, where as + * the rest of the chunks are. + */ + if (cpu) { + ptr = rb_range_align_subbuf(ptr, subbuf_size, nr_subbufs); + ptr += subbuf_size * nr_subbufs; + + /* We can use multiplication to find chunks greater than 1 */ + if (cpu > 1) { + unsigned long size; + unsigned long p; + + /* Save the beginning of this CPU chunk */ + p = ptr; + ptr = rb_range_align_subbuf(ptr, subbuf_size, nr_subbufs); + ptr += subbuf_size * nr_subbufs; + + /* Now all chunks after this are the same size */ + size = ptr - p; + ptr += size * (cpu - 2); + } + } + return (void *)ptr; +} + +/* Return the start of subbufs given the meta pointer */ +static void *rb_subbufs_from_meta(struct ring_buffer_cpu_meta *meta) +{ + int subbuf_size = meta->subbuf_size; + unsigned long ptr; + + ptr = (unsigned long)meta; + ptr = rb_range_align_subbuf(ptr, subbuf_size, meta->nr_subbufs); + + return (void *)ptr; +} + +/* + * Return a specific sub-buffer for a given @cpu defined by @idx. + */ +static void *rb_range_buffer(struct ring_buffer_per_cpu *cpu_buffer, int idx) +{ + struct ring_buffer_cpu_meta *meta; + unsigned long ptr; + int subbuf_size; + + meta = rb_range_meta(cpu_buffer->buffer, 0, cpu_buffer->cpu); + if (!meta) + return NULL; + + if (WARN_ON_ONCE(idx >= meta->nr_subbufs)) + return NULL; + + subbuf_size = meta->subbuf_size; + + /* Map this buffer to the order that's in meta->buffers[] */ + idx = meta->buffers[idx]; + + ptr = (unsigned long)rb_subbufs_from_meta(meta); + + ptr += subbuf_size * idx; + if (ptr + subbuf_size > cpu_buffer->buffer->range_addr_end) + return NULL; + + return (void *)ptr; +} + +/* + * See if the existing memory contains a valid meta section. + * if so, use that, otherwise initialize it. + */ +static bool rb_meta_init(struct trace_buffer *buffer, int scratch_size) +{ + unsigned long ptr = buffer->range_addr_start; + struct ring_buffer_meta *bmeta; + unsigned long total_size; + int struct_sizes; + + bmeta = (struct ring_buffer_meta *)ptr; + buffer->meta = bmeta; + + total_size = buffer->range_addr_end - buffer->range_addr_start; + + struct_sizes = sizeof(struct ring_buffer_cpu_meta); + struct_sizes |= sizeof(*bmeta) << 16; + + /* The first buffer will start word size after the meta page */ + ptr += sizeof(*bmeta); + ptr = ALIGN(ptr, sizeof(long)); + ptr += scratch_size; + + if (bmeta->magic != RING_BUFFER_META_MAGIC) { + pr_info("Ring buffer boot meta mismatch of magic\n"); + goto init; + } + + if (bmeta->struct_sizes != struct_sizes) { + pr_info("Ring buffer boot meta mismatch of struct size\n"); + goto init; + } + + if (bmeta->total_size != total_size) { + pr_info("Ring buffer boot meta mismatch of total size\n"); + goto init; + } + + if (bmeta->buffers_offset > bmeta->total_size) { + pr_info("Ring buffer boot meta mismatch of offset outside of total size\n"); + goto init; + } + + if (bmeta->buffers_offset != (void *)ptr - (void *)bmeta) { + pr_info("Ring buffer boot meta mismatch of first buffer offset\n"); + goto init; + } + + return true; + + init: + bmeta->magic = RING_BUFFER_META_MAGIC; + bmeta->struct_sizes = struct_sizes; + bmeta->total_size = total_size; + bmeta->buffers_offset = (void *)ptr - (void *)bmeta; + + /* Zero out the scratch pad */ + memset((void *)bmeta + sizeof(*bmeta), 0, bmeta->buffers_offset - sizeof(*bmeta)); + + return false; +} + +/* + * See if the existing memory contains valid ring buffer data. + * As the previous kernel must be the same as this kernel, all + * the calculations (size of buffers and number of buffers) + * must be the same. + */ +static bool rb_cpu_meta_valid(struct ring_buffer_cpu_meta *meta, int cpu, + struct trace_buffer *buffer, int nr_pages, + unsigned long *subbuf_mask) +{ + int subbuf_size = PAGE_SIZE; + struct buffer_data_page *subbuf; + unsigned long buffers_start; + unsigned long buffers_end; + int i; + + if (!subbuf_mask) + return false; + + buffers_start = meta->first_buffer; + buffers_end = meta->first_buffer + (subbuf_size * meta->nr_subbufs); + + /* Is the head and commit buffers within the range of buffers? */ + if (meta->head_buffer < buffers_start || + meta->head_buffer >= buffers_end) { + pr_info("Ring buffer boot meta [%d] head buffer out of range\n", cpu); + return false; + } + + if (meta->commit_buffer < buffers_start || + meta->commit_buffer >= buffers_end) { + pr_info("Ring buffer boot meta [%d] commit buffer out of range\n", cpu); + return false; + } + + subbuf = rb_subbufs_from_meta(meta); + + bitmap_clear(subbuf_mask, 0, meta->nr_subbufs); + + /* Is the meta buffers and the subbufs themselves have correct data? */ + for (i = 0; i < meta->nr_subbufs; i++) { + if (meta->buffers[i] < 0 || + meta->buffers[i] >= meta->nr_subbufs) { + pr_info("Ring buffer boot meta [%d] array out of range\n", cpu); + return false; + } + + if ((unsigned)local_read(&subbuf->commit) > subbuf_size) { + pr_info("Ring buffer boot meta [%d] buffer invalid commit\n", cpu); + return false; + } + + if (test_bit(meta->buffers[i], subbuf_mask)) { + pr_info("Ring buffer boot meta [%d] array has duplicates\n", cpu); + return false; + } + + set_bit(meta->buffers[i], subbuf_mask); + subbuf = (void *)subbuf + subbuf_size; + } + + return true; +} + +static int rb_meta_subbuf_idx(struct ring_buffer_cpu_meta *meta, void *subbuf); + +static int rb_read_data_buffer(struct buffer_data_page *dpage, int tail, int cpu, + unsigned long long *timestamp, u64 *delta_ptr) +{ + struct ring_buffer_event *event; + u64 ts, delta; + int events = 0; + int e; + + *delta_ptr = 0; + *timestamp = 0; + + ts = dpage->time_stamp; + + for (e = 0; e < tail; e += rb_event_length(event)) { + + event = (struct ring_buffer_event *)(dpage->data + e); + + switch (event->type_len) { + + case RINGBUF_TYPE_TIME_EXTEND: + delta = rb_event_time_stamp(event); + ts += delta; + break; + + case RINGBUF_TYPE_TIME_STAMP: + delta = rb_event_time_stamp(event); + delta = rb_fix_abs_ts(delta, ts); + if (delta < ts) { + *delta_ptr = delta; + *timestamp = ts; + return -1; + } + ts = delta; + break; + + case RINGBUF_TYPE_PADDING: + if (event->time_delta == 1) + break; + fallthrough; + case RINGBUF_TYPE_DATA: + events++; + ts += event->time_delta; + break; + + default: return -1; + } } + *timestamp = ts; + return events; +} - rb_head_page_activate(cpu_buffer); +static int rb_validate_buffer(struct buffer_data_page *dpage, int cpu) +{ + unsigned long long ts; + u64 delta; + int tail; + + tail = local_read(&dpage->commit); + return rb_read_data_buffer(dpage, tail, cpu, &ts, &delta); +} + +/* If the meta data has been validated, now validate the events */ +static void rb_meta_validate_events(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + struct buffer_page *head_page, *orig_head; + unsigned long entry_bytes = 0; + unsigned long entries = 0; + int ret; + u64 ts; + int i; + + if (!meta || !meta->head_buffer) + return; + + /* Do the reader page first */ + ret = rb_validate_buffer(cpu_buffer->reader_page->page, cpu_buffer->cpu); + if (ret < 0) { + pr_info("Ring buffer reader page is invalid\n"); + goto invalid; + } + entries += ret; + entry_bytes += local_read(&cpu_buffer->reader_page->page->commit); + local_set(&cpu_buffer->reader_page->entries, ret); + + orig_head = head_page = cpu_buffer->head_page; + ts = head_page->page->time_stamp; + + /* + * Try to rewind the head so that we can read the pages which already + * read in the previous boot. + */ + if (head_page == cpu_buffer->tail_page) + goto skip_rewind; + + rb_dec_page(&head_page); + for (i = 0; i < meta->nr_subbufs + 1; i++, rb_dec_page(&head_page)) { + + /* Rewind until tail (writer) page. */ + if (head_page == cpu_buffer->tail_page) + break; + + /* Ensure the page has older data than head. */ + if (ts < head_page->page->time_stamp) + break; + + ts = head_page->page->time_stamp; + /* Ensure the page has correct timestamp and some data. */ + if (!ts || rb_page_commit(head_page) == 0) + break; + + /* Stop rewind if the page is invalid. */ + ret = rb_validate_buffer(head_page->page, cpu_buffer->cpu); + if (ret < 0) + break; + + /* Recover the number of entries and update stats. */ + local_set(&head_page->entries, ret); + if (ret) + local_inc(&cpu_buffer->pages_touched); + entries += ret; + entry_bytes += rb_page_commit(head_page); + } + if (i) + pr_info("Ring buffer [%d] rewound %d pages\n", cpu_buffer->cpu, i); + + /* The last rewound page must be skipped. */ + if (head_page != orig_head) + rb_inc_page(&head_page); + + /* + * If the ring buffer was rewound, then inject the reader page + * into the location just before the original head page. + */ + if (head_page != orig_head) { + struct buffer_page *bpage = orig_head; + + rb_dec_page(&bpage); + /* + * Insert the reader_page before the original head page. + * Since the list encode RB_PAGE flags, general list + * operations should be avoided. + */ + cpu_buffer->reader_page->list.next = &orig_head->list; + cpu_buffer->reader_page->list.prev = orig_head->list.prev; + orig_head->list.prev = &cpu_buffer->reader_page->list; + bpage->list.next = &cpu_buffer->reader_page->list; + + /* Make the head_page the reader page */ + cpu_buffer->reader_page = head_page; + bpage = head_page; + rb_inc_page(&head_page); + head_page->list.prev = bpage->list.prev; + rb_dec_page(&bpage); + bpage->list.next = &head_page->list; + rb_set_list_to_head(&bpage->list); + cpu_buffer->pages = &head_page->list; + + cpu_buffer->head_page = head_page; + meta->head_buffer = (unsigned long)head_page->page; + + /* Reset all the indexes */ + bpage = cpu_buffer->reader_page; + meta->buffers[0] = rb_meta_subbuf_idx(meta, bpage->page); + bpage->id = 0; + + for (i = 1, bpage = head_page; i < meta->nr_subbufs; + i++, rb_inc_page(&bpage)) { + meta->buffers[i] = rb_meta_subbuf_idx(meta, bpage->page); + bpage->id = i; + } + + /* We'll restart verifying from orig_head */ + head_page = orig_head; + } + + skip_rewind: + /* If the commit_buffer is the reader page, update the commit page */ + if (meta->commit_buffer == (unsigned long)cpu_buffer->reader_page->page) { + cpu_buffer->commit_page = cpu_buffer->reader_page; + /* Nothing more to do, the only page is the reader page */ + goto done; + } + + /* Iterate until finding the commit page */ + for (i = 0; i < meta->nr_subbufs + 1; i++, rb_inc_page(&head_page)) { + + /* Reader page has already been done */ + if (head_page == cpu_buffer->reader_page) + continue; + + ret = rb_validate_buffer(head_page->page, cpu_buffer->cpu); + if (ret < 0) { + pr_info("Ring buffer meta [%d] invalid buffer page\n", + cpu_buffer->cpu); + goto invalid; + } + + /* If the buffer has content, update pages_touched */ + if (ret) + local_inc(&cpu_buffer->pages_touched); + + entries += ret; + entry_bytes += local_read(&head_page->page->commit); + local_set(&cpu_buffer->head_page->entries, ret); + + if (head_page == cpu_buffer->commit_page) + break; + } + + if (head_page != cpu_buffer->commit_page) { + pr_info("Ring buffer meta [%d] commit page not found\n", + cpu_buffer->cpu); + goto invalid; + } + done: + local_set(&cpu_buffer->entries, entries); + local_set(&cpu_buffer->entries_bytes, entry_bytes); + + pr_info("Ring buffer meta [%d] is from previous boot!\n", cpu_buffer->cpu); + return; + + invalid: + /* The content of the buffers are invalid, reset the meta data */ + meta->head_buffer = 0; + meta->commit_buffer = 0; + + /* Reset the reader page */ + local_set(&cpu_buffer->reader_page->entries, 0); + local_set(&cpu_buffer->reader_page->page->commit, 0); + + /* Reset all the subbuffers */ + for (i = 0; i < meta->nr_subbufs - 1; i++, rb_inc_page(&head_page)) { + local_set(&head_page->entries, 0); + local_set(&head_page->page->commit, 0); + } +} + +static void rb_range_meta_init(struct trace_buffer *buffer, int nr_pages, int scratch_size) +{ + struct ring_buffer_cpu_meta *meta; + unsigned long *subbuf_mask; + unsigned long delta; + void *subbuf; + bool valid = false; + int cpu; + int i; + + /* Create a mask to test the subbuf array */ + subbuf_mask = bitmap_alloc(nr_pages + 1, GFP_KERNEL); + /* If subbuf_mask fails to allocate, then rb_meta_valid() will return false */ + + if (rb_meta_init(buffer, scratch_size)) + valid = true; + + for (cpu = 0; cpu < nr_cpu_ids; cpu++) { + void *next_meta; + + meta = rb_range_meta(buffer, nr_pages, cpu); + + if (valid && rb_cpu_meta_valid(meta, cpu, buffer, nr_pages, subbuf_mask)) { + /* Make the mappings match the current address */ + subbuf = rb_subbufs_from_meta(meta); + delta = (unsigned long)subbuf - meta->first_buffer; + meta->first_buffer += delta; + meta->head_buffer += delta; + meta->commit_buffer += delta; + continue; + } + + if (cpu < nr_cpu_ids - 1) + next_meta = rb_range_meta(buffer, nr_pages, cpu + 1); + else + next_meta = (void *)buffer->range_addr_end; + + memset(meta, 0, next_meta - (void *)meta); + + meta->nr_subbufs = nr_pages + 1; + meta->subbuf_size = PAGE_SIZE; + + subbuf = rb_subbufs_from_meta(meta); + + meta->first_buffer = (unsigned long)subbuf; + + /* + * The buffers[] array holds the order of the sub-buffers + * that are after the meta data. The sub-buffers may + * be swapped out when read and inserted into a different + * location of the ring buffer. Although their addresses + * remain the same, the buffers[] array contains the + * index into the sub-buffers holding their actual order. + */ + for (i = 0; i < meta->nr_subbufs; i++) { + meta->buffers[i] = i; + rb_init_page(subbuf); + subbuf += meta->subbuf_size; + } + } + bitmap_free(subbuf_mask); +} + +static void *rbm_start(struct seq_file *m, loff_t *pos) +{ + struct ring_buffer_per_cpu *cpu_buffer = m->private; + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + unsigned long val; + + if (!meta) + return NULL; + + if (*pos > meta->nr_subbufs) + return NULL; + + val = *pos; + val++; + + return (void *)val; +} + +static void *rbm_next(struct seq_file *m, void *v, loff_t *pos) +{ + (*pos)++; + + return rbm_start(m, pos); +} + +static int rbm_show(struct seq_file *m, void *v) +{ + struct ring_buffer_per_cpu *cpu_buffer = m->private; + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + unsigned long val = (unsigned long)v; + + if (val == 1) { + seq_printf(m, "head_buffer: %d\n", + rb_meta_subbuf_idx(meta, (void *)meta->head_buffer)); + seq_printf(m, "commit_buffer: %d\n", + rb_meta_subbuf_idx(meta, (void *)meta->commit_buffer)); + seq_printf(m, "subbuf_size: %d\n", meta->subbuf_size); + seq_printf(m, "nr_subbufs: %d\n", meta->nr_subbufs); + return 0; + } + + val -= 2; + seq_printf(m, "buffer[%ld]: %d\n", val, meta->buffers[val]); return 0; } +static void rbm_stop(struct seq_file *m, void *p) +{ +} + +static const struct seq_operations rb_meta_seq_ops = { + .start = rbm_start, + .next = rbm_next, + .show = rbm_show, + .stop = rbm_stop, +}; + +int ring_buffer_meta_seq_init(struct file *file, struct trace_buffer *buffer, int cpu) +{ + struct seq_file *m; + int ret; + + ret = seq_open(file, &rb_meta_seq_ops); + if (ret) + return ret; + + m = file->private_data; + m->private = buffer->buffers[cpu]; + + return 0; +} + +/* Map the buffer_pages to the previous head and commit pages */ +static void rb_meta_buffer_update(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *bpage) +{ + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + + if (meta->head_buffer == (unsigned long)bpage->page) + cpu_buffer->head_page = bpage; + + if (meta->commit_buffer == (unsigned long)bpage->page) { + cpu_buffer->commit_page = bpage; + cpu_buffer->tail_page = bpage; + } +} + static int __rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, long nr_pages, struct list_head *pages) { + struct trace_buffer *buffer = cpu_buffer->buffer; + struct ring_buffer_cpu_meta *meta = NULL; struct buffer_page *bpage, *tmp; bool user_thread = current->mm != NULL; - gfp_t mflags; long i; /* @@ -1655,13 +2253,6 @@ static int __rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, return -ENOMEM; /* - * __GFP_RETRY_MAYFAIL flag makes sure that the allocation fails - * gracefully without invoking oom-killer and the system is not - * destabilized. - */ - mflags = GFP_KERNEL | __GFP_RETRY_MAYFAIL; - - /* * If a user thread allocates too much, and si_mem_available() * reports there's enough memory, even though there is not. * Make sure the OOM killer kills this thread. This can happen @@ -1672,23 +2263,41 @@ static int __rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, */ if (user_thread) set_current_oom_origin(); + + if (buffer->range_addr_start) + meta = rb_range_meta(buffer, nr_pages, cpu_buffer->cpu); + for (i = 0; i < nr_pages; i++) { - struct page *page; - bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), - mflags, cpu_to_node(cpu_buffer->cpu)); + bpage = alloc_cpu_page(cpu_buffer->cpu); if (!bpage) goto free_pages; rb_check_bpage(cpu_buffer, bpage); - list_add(&bpage->list, pages); - - page = alloc_pages_node(cpu_to_node(cpu_buffer->cpu), mflags, 0); - if (!page) - goto free_pages; - bpage->page = page_address(page); - rb_init_page(bpage->page); + /* + * Append the pages as for mapped buffers we want to keep + * the order + */ + list_add_tail(&bpage->list, pages); + + if (meta) { + /* A range was given. Use that for the buffer page */ + bpage->page = rb_range_buffer(cpu_buffer, i + 1); + if (!bpage->page) + goto free_pages; + /* If this is valid from a previous boot */ + if (meta->head_buffer) + rb_meta_buffer_update(cpu_buffer, bpage); + bpage->range = 1; + bpage->id = i + 1; + } else { + int order = cpu_buffer->buffer->subbuf_order; + bpage->page = alloc_cpu_data(cpu_buffer->cpu, order); + if (!bpage->page) + goto free_pages; + } + bpage->order = cpu_buffer->buffer->subbuf_order; if (user_thread && fatal_signal_pending(current)) goto free_pages; @@ -1737,13 +2346,12 @@ static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, static struct ring_buffer_per_cpu * rb_allocate_cpu_buffer(struct trace_buffer *buffer, long nr_pages, int cpu) { - struct ring_buffer_per_cpu *cpu_buffer; + struct ring_buffer_per_cpu *cpu_buffer __free(kfree) = + alloc_cpu_buffer(cpu); + struct ring_buffer_cpu_meta *meta; struct buffer_page *bpage; - struct page *page; int ret; - cpu_buffer = kzalloc_node(ALIGN(sizeof(*cpu_buffer), cache_line_size()), - GFP_KERNEL, cpu_to_node(cpu)); if (!cpu_buffer) return NULL; @@ -1757,20 +2365,35 @@ rb_allocate_cpu_buffer(struct trace_buffer *buffer, long nr_pages, int cpu) init_irq_work(&cpu_buffer->irq_work.work, rb_wake_up_waiters); init_waitqueue_head(&cpu_buffer->irq_work.waiters); init_waitqueue_head(&cpu_buffer->irq_work.full_waiters); + mutex_init(&cpu_buffer->mapping_lock); - bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), - GFP_KERNEL, cpu_to_node(cpu)); + bpage = alloc_cpu_page(cpu); if (!bpage) - goto fail_free_buffer; + return NULL; rb_check_bpage(cpu_buffer, bpage); cpu_buffer->reader_page = bpage; - page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL, 0); - if (!page) - goto fail_free_reader; - bpage->page = page_address(page); - rb_init_page(bpage->page); + + if (buffer->range_addr_start) { + /* + * Range mapped buffers have the same restrictions as memory + * mapped ones do. + */ + cpu_buffer->mapped = 1; + cpu_buffer->ring_meta = rb_range_meta(buffer, nr_pages, cpu); + bpage->page = rb_range_buffer(cpu_buffer, 0); + if (!bpage->page) + goto fail_free_reader; + if (cpu_buffer->ring_meta->head_buffer) + rb_meta_buffer_update(cpu_buffer, bpage); + bpage->range = 1; + } else { + int order = cpu_buffer->buffer->subbuf_order; + bpage->page = alloc_cpu_data(cpu, order); + if (!bpage->page) + goto fail_free_reader; + } INIT_LIST_HEAD(&cpu_buffer->reader_page->list); INIT_LIST_HEAD(&cpu_buffer->new_pages); @@ -1779,19 +2402,41 @@ rb_allocate_cpu_buffer(struct trace_buffer *buffer, long nr_pages, int cpu) if (ret < 0) goto fail_free_reader; - cpu_buffer->head_page - = list_entry(cpu_buffer->pages, struct buffer_page, list); - cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page; + rb_meta_validate_events(cpu_buffer); + + /* If the boot meta was valid then this has already been updated */ + meta = cpu_buffer->ring_meta; + if (!meta || !meta->head_buffer || + !cpu_buffer->head_page || !cpu_buffer->commit_page || !cpu_buffer->tail_page) { + if (meta && meta->head_buffer && + (cpu_buffer->head_page || cpu_buffer->commit_page || cpu_buffer->tail_page)) { + pr_warn("Ring buffer meta buffers not all mapped\n"); + if (!cpu_buffer->head_page) + pr_warn(" Missing head_page\n"); + if (!cpu_buffer->commit_page) + pr_warn(" Missing commit_page\n"); + if (!cpu_buffer->tail_page) + pr_warn(" Missing tail_page\n"); + } - rb_head_page_activate(cpu_buffer); + cpu_buffer->head_page + = list_entry(cpu_buffer->pages, struct buffer_page, list); + cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page; - return cpu_buffer; + rb_head_page_activate(cpu_buffer); + + if (cpu_buffer->ring_meta) + meta->commit_buffer = meta->head_buffer; + } else { + /* The valid meta buffer still needs to activate the head page */ + rb_head_page_activate(cpu_buffer); + } + + return_ptr(cpu_buffer); fail_free_reader: free_buffer_page(cpu_buffer->reader_page); - fail_free_buffer: - kfree(cpu_buffer); return NULL; } @@ -1800,6 +2445,8 @@ static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer) struct list_head *head = cpu_buffer->pages; struct buffer_page *bpage, *tmp; + irq_work_sync(&cpu_buffer->irq_work.work); + free_buffer_page(cpu_buffer->reader_page); if (head) { @@ -1813,25 +2460,20 @@ static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer) free_buffer_page(bpage); } + free_page((unsigned long)cpu_buffer->free_page); + kfree(cpu_buffer); } -/** - * __ring_buffer_alloc - allocate a new ring_buffer - * @size: the size in bytes per cpu that is needed. - * @flags: attributes to set for the ring buffer. - * @key: ring buffer reader_lock_key. - * - * Currently the only flag that is available is the RB_FL_OVERWRITE - * flag. This flag means that the buffer will overwrite old data - * when the buffer wraps. If this flag is not set, the buffer will - * drop data when the tail hits the head. - */ -struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, - struct lock_class_key *key) +static struct trace_buffer *alloc_buffer(unsigned long size, unsigned flags, + int order, unsigned long start, + unsigned long end, + unsigned long scratch_size, + struct lock_class_key *key) { - struct trace_buffer *buffer; + struct trace_buffer *buffer __free(kfree) = NULL; long nr_pages; + int subbuf_size; int bsize; int cpu; int ret; @@ -1843,9 +2485,15 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, return NULL; if (!zalloc_cpumask_var(&buffer->cpumask, GFP_KERNEL)) - goto fail_free_buffer; + return NULL; + + buffer->subbuf_order = order; + subbuf_size = (PAGE_SIZE << order); + buffer->subbuf_size = subbuf_size - BUF_PAGE_HDR_SIZE; + + /* Max payload is buffer page size - header (8bytes) */ + buffer->max_data_size = buffer->subbuf_size - (sizeof(u32) * 2); - nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE); buffer->flags = flags; buffer->clock = trace_clock_local; buffer->reader_lock_key = key; @@ -1853,10 +2501,6 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, init_irq_work(&buffer->irq_work.work, rb_wake_up_waiters); init_waitqueue_head(&buffer->irq_work.waiters); - /* need at least two pages */ - if (nr_pages < 2) - nr_pages = 2; - buffer->cpus = nr_cpu_ids; bsize = sizeof(void *) * nr_cpu_ids; @@ -1865,6 +2509,69 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, if (!buffer->buffers) goto fail_free_cpumask; + /* If start/end are specified, then that overrides size */ + if (start && end) { + unsigned long buffers_start; + unsigned long ptr; + int n; + + /* Make sure that start is word aligned */ + start = ALIGN(start, sizeof(long)); + + /* scratch_size needs to be aligned too */ + scratch_size = ALIGN(scratch_size, sizeof(long)); + + /* Subtract the buffer meta data and word aligned */ + buffers_start = start + sizeof(struct ring_buffer_cpu_meta); + buffers_start = ALIGN(buffers_start, sizeof(long)); + buffers_start += scratch_size; + + /* Calculate the size for the per CPU data */ + size = end - buffers_start; + size = size / nr_cpu_ids; + + /* + * The number of sub-buffers (nr_pages) is determined by the + * total size allocated minus the meta data size. + * Then that is divided by the number of per CPU buffers + * needed, plus account for the integer array index that + * will be appended to the meta data. + */ + nr_pages = (size - sizeof(struct ring_buffer_cpu_meta)) / + (subbuf_size + sizeof(int)); + /* Need at least two pages plus the reader page */ + if (nr_pages < 3) + goto fail_free_buffers; + + again: + /* Make sure that the size fits aligned */ + for (n = 0, ptr = buffers_start; n < nr_cpu_ids; n++) { + ptr += sizeof(struct ring_buffer_cpu_meta) + + sizeof(int) * nr_pages; + ptr = ALIGN(ptr, subbuf_size); + ptr += subbuf_size * nr_pages; + } + if (ptr > end) { + if (nr_pages <= 3) + goto fail_free_buffers; + nr_pages--; + goto again; + } + + /* nr_pages should not count the reader page */ + nr_pages--; + buffer->range_addr_start = start; + buffer->range_addr_end = end; + + rb_range_meta_init(buffer, nr_pages, scratch_size); + } else { + + /* need at least two pages */ + nr_pages = DIV_ROUND_UP(size, buffer->subbuf_size); + if (nr_pages < 2) + nr_pages = 2; + } + cpu = raw_smp_processor_id(); cpumask_set_cpu(cpu, buffer->cpumask); buffer->buffers[cpu] = rb_allocate_cpu_buffer(buffer, nr_pages, cpu); @@ -1877,7 +2584,7 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, mutex_init(&buffer->mutex); - return buffer; + return_ptr(buffer); fail_free_buffers: for_each_buffer_cpu(buffer, cpu) { @@ -1889,13 +2596,73 @@ struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, fail_free_cpumask: free_cpumask_var(buffer->cpumask); - fail_free_buffer: - kfree(buffer); return NULL; } + +/** + * __ring_buffer_alloc - allocate a new ring_buffer + * @size: the size in bytes per cpu that is needed. + * @flags: attributes to set for the ring buffer. + * @key: ring buffer reader_lock_key. + * + * Currently the only flag that is available is the RB_FL_OVERWRITE + * flag. This flag means that the buffer will overwrite old data + * when the buffer wraps. If this flag is not set, the buffer will + * drop data when the tail hits the head. + */ +struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, + struct lock_class_key *key) +{ + /* Default buffer page size - one system page */ + return alloc_buffer(size, flags, 0, 0, 0, 0, key); + +} EXPORT_SYMBOL_GPL(__ring_buffer_alloc); /** + * __ring_buffer_alloc_range - allocate a new ring_buffer from existing memory + * @size: the size in bytes per cpu that is needed. + * @flags: attributes to set for the ring buffer. + * @order: sub-buffer order + * @start: start of allocated range + * @range_size: size of allocated range + * @scratch_size: size of scratch area (for preallocated memory buffers) + * @key: ring buffer reader_lock_key. + * + * Currently the only flag that is available is the RB_FL_OVERWRITE + * flag. This flag means that the buffer will overwrite old data + * when the buffer wraps. If this flag is not set, the buffer will + * drop data when the tail hits the head. + */ +struct trace_buffer *__ring_buffer_alloc_range(unsigned long size, unsigned flags, + int order, unsigned long start, + unsigned long range_size, + unsigned long scratch_size, + struct lock_class_key *key) +{ + return alloc_buffer(size, flags, order, start, start + range_size, + scratch_size, key); +} + +void *ring_buffer_meta_scratch(struct trace_buffer *buffer, unsigned int *size) +{ + struct ring_buffer_meta *meta; + void *ptr; + + if (!buffer || !buffer->meta) + return NULL; + + meta = buffer->meta; + + ptr = (void *)ALIGN((unsigned long)meta + sizeof(*meta), sizeof(long)); + + if (size) + *size = (void *)meta + meta->buffers_offset - ptr; + + return ptr; +} + +/** * ring_buffer_free - free a ring buffer. * @buffer: the buffer to free. */ @@ -1906,6 +2673,8 @@ ring_buffer_free(struct trace_buffer *buffer) cpuhp_state_remove_instance(CPUHP_TRACE_RB_PREPARE, &buffer->node); + irq_work_sync(&buffer->irq_work.work); + for_each_buffer_cpu(buffer, cpu) rb_free_cpu_buffer(buffer->buffers[cpu]); @@ -1932,8 +2701,6 @@ bool ring_buffer_time_stamp_abs(struct trace_buffer *buffer) return buffer->time_stamp_abs; } -static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer); - static inline unsigned long rb_page_entries(struct buffer_page *bpage) { return local_read(&bpage->entries) & RB_WRITE_MASK; @@ -1944,7 +2711,7 @@ static inline unsigned long rb_page_write(struct buffer_page *bpage) return local_read(&bpage->write) & RB_WRITE_MASK; } -static int +static bool rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) { struct list_head *tail_page, *to_remove, *next_page; @@ -1985,6 +2752,8 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) to_remove = rb_list_head(to_remove)->next; head_bit |= (unsigned long)to_remove & RB_PAGE_HEAD; } + /* Read iterators need to reset themselves when some pages removed */ + cpu_buffer->pages_removed += nr_removed; next_page = rb_list_head(to_remove)->next; @@ -2000,18 +2769,13 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) /* make sure pages points to a valid page in the ring buffer */ cpu_buffer->pages = next_page; + cpu_buffer->cnt++; /* update head page */ if (head_bit) cpu_buffer->head_page = list_entry(next_page, struct buffer_page, list); - /* - * change read pointer to make sure any read iterators reset - * themselves - */ - cpu_buffer->read = 0; - /* pages are removed, resume tracing and then free the pages */ atomic_dec(&cpu_buffer->record_disabled); raw_spin_unlock_irq(&cpu_buffer->reader_lock); @@ -2039,7 +2803,7 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) * Increment overrun to account for the lost events. */ local_add(page_entries, &cpu_buffer->overrun); - local_sub(BUF_PAGE_SIZE, &cpu_buffer->entries_bytes); + local_sub(rb_page_commit(to_remove_page), &cpu_buffer->entries_bytes); local_inc(&cpu_buffer->pages_lost); } @@ -2057,12 +2821,13 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) return nr_removed == 0; } -static int +static bool rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer) { struct list_head *pages = &cpu_buffer->new_pages; - int retries, success; unsigned long flags; + bool success; + int retries; /* Can be called at early boot up, where interrupts must not been enabled */ raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); @@ -2081,15 +2846,16 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer) * spinning. */ retries = 10; - success = 0; + success = false; while (retries--) { - struct list_head *head_page, *prev_page, *r; + struct list_head *head_page, *prev_page; struct list_head *last_page, *first_page; struct list_head *head_page_with_bit; + struct buffer_page *hpage = rb_set_head_page(cpu_buffer); - head_page = &rb_set_head_page(cpu_buffer)->list; - if (!head_page) + if (!hpage) break; + head_page = &hpage->list; prev_page = head_page->prev; first_page = pages->next; @@ -2101,16 +2867,17 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer) last_page->next = head_page_with_bit; first_page->prev = prev_page; - r = cmpxchg(&prev_page->next, head_page_with_bit, first_page); - - if (r == head_page_with_bit) { + /* caution: head_page_with_bit gets updated on cmpxchg failure */ + if (try_cmpxchg(&prev_page->next, + &head_page_with_bit, first_page)) { /* * yay, we replaced the page pointer to our new list, * now, we just have to update to head page's prev * pointer to point to end of list */ head_page->prev = last_page; - success = 1; + cpu_buffer->cnt++; + success = true; break; } } @@ -2138,7 +2905,7 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer) static void rb_update_pages(struct ring_buffer_per_cpu *cpu_buffer) { - int success; + bool success; if (cpu_buffer->nr_pages_to_update > 0) success = rb_insert_pages(cpu_buffer); @@ -2164,7 +2931,7 @@ static void update_pages_handler(struct work_struct *work) * @size: the new size. * @cpu_id: the cpu buffer to resize * - * Minimum size is 2 * BUF_PAGE_SIZE. + * Minimum size is 2 * buffer->subbuf_size. * * Returns 0 on success and < 0 on failure. */ @@ -2186,15 +2953,21 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, !cpumask_test_cpu(cpu_id, buffer->cpumask)) return 0; - nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE); + nr_pages = DIV_ROUND_UP(size, buffer->subbuf_size); /* we need a minimum of two pages */ if (nr_pages < 2) nr_pages = 2; + /* + * Keep CPUs from coming online while resizing to synchronize + * with new per CPU buffers being created. + */ + guard(cpus_read_lock)(); + /* prevent another thread from changing buffer sizes */ mutex_lock(&buffer->mutex); - + atomic_inc(&buffer->resizing); if (cpu_id == RING_BUFFER_ALL_CPUS) { /* @@ -2232,9 +3005,10 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, err = -ENOMEM; goto out_err; } + + cond_resched(); } - cpus_read_lock(); /* * Fire off all the required work handlers * We can't schedule on offline CPUs, but it's not necessary @@ -2274,7 +3048,6 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, cpu_buffer->nr_pages_to_update = 0; } - cpus_read_unlock(); } else { cpu_buffer = buffer->buffers[cpu_id]; @@ -2302,8 +3075,6 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, goto out_err; } - cpus_read_lock(); - /* Can't run something on an offline CPU. */ if (!cpu_online(cpu_id)) rb_update_pages(cpu_buffer); @@ -2322,7 +3093,6 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, } cpu_buffer->nr_pages_to_update = 0; - cpus_read_unlock(); } out: @@ -2349,6 +3119,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, atomic_dec(&buffer->record_disabled); } + atomic_dec(&buffer->resizing); mutex_unlock(&buffer->mutex); return 0; @@ -2369,6 +3140,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, } } out_err_unlock: + atomic_dec(&buffer->resizing); mutex_unlock(&buffer->mutex); return err; } @@ -2397,11 +3169,6 @@ rb_reader_event(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->reader_page->read); } -static __always_inline unsigned rb_page_commit(struct buffer_page *bpage) -{ - return local_read(&bpage->page->commit); -} - static struct ring_buffer_event * rb_iter_head_event(struct ring_buffer_iter *iter) { @@ -2420,6 +3187,11 @@ rb_iter_head_event(struct ring_buffer_iter *iter) */ commit = rb_page_commit(iter_head_page); smp_rmb(); + + /* An event needs to be at least 8 bytes in size */ + if (iter->head > commit - 8) + goto reset; + event = __rb_page_index(iter_head_page, iter->head); length = rb_event_length(event); @@ -2429,7 +3201,7 @@ rb_iter_head_event(struct ring_buffer_iter *iter) */ barrier(); - if ((iter->head + length) > commit || length > BUF_MAX_DATA_SIZE) + if ((iter->head + length) > commit || length > iter->event_size) /* Writer corrupted the read? */ goto reset; @@ -2459,7 +3231,7 @@ rb_iter_head_event(struct ring_buffer_iter *iter) /* Size is determined by what has been committed */ static __always_inline unsigned rb_page_size(struct buffer_page *bpage) { - return rb_page_commit(bpage); + return rb_page_commit(bpage) & ~RB_MISSED_MASK; } static __always_inline unsigned @@ -2469,11 +3241,13 @@ rb_commit_index(struct ring_buffer_per_cpu *cpu_buffer) } static __always_inline unsigned -rb_event_index(struct ring_buffer_event *event) +rb_event_index(struct ring_buffer_per_cpu *cpu_buffer, struct ring_buffer_event *event) { unsigned long addr = (unsigned long)event; - return (addr & ~PAGE_MASK) - BUF_PAGE_HDR_SIZE; + addr &= (PAGE_SIZE << cpu_buffer->buffer->subbuf_order) - 1; + + return addr - BUF_PAGE_HDR_SIZE; } static void rb_inc_iter(struct ring_buffer_iter *iter) @@ -2496,6 +3270,52 @@ static void rb_inc_iter(struct ring_buffer_iter *iter) iter->next_event = 0; } +/* Return the index into the sub-buffers for a given sub-buffer */ +static int rb_meta_subbuf_idx(struct ring_buffer_cpu_meta *meta, void *subbuf) +{ + void *subbuf_array; + + subbuf_array = (void *)meta + sizeof(int) * meta->nr_subbufs; + subbuf_array = (void *)ALIGN((unsigned long)subbuf_array, meta->subbuf_size); + return (subbuf - subbuf_array) / meta->subbuf_size; +} + +static void rb_update_meta_head(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *next_page) +{ + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + unsigned long old_head = (unsigned long)next_page->page; + unsigned long new_head; + + rb_inc_page(&next_page); + new_head = (unsigned long)next_page->page; + + /* + * Only move it forward once, if something else came in and + * moved it forward, then we don't want to touch it. + */ + (void)cmpxchg(&meta->head_buffer, old_head, new_head); +} + +static void rb_update_meta_reader(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *reader) +{ + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + void *old_reader = cpu_buffer->reader_page->page; + void *new_reader = reader->page; + int id; + + id = reader->id; + cpu_buffer->reader_page->id = id; + reader->id = 0; + + meta->buffers[0] = rb_meta_subbuf_idx(meta, new_reader); + meta->buffers[id] = rb_meta_subbuf_idx(meta, old_reader); + + /* The head pointer is the one after the reader */ + rb_update_meta_head(cpu_buffer, reader); +} + /* * rb_handle_head_page - writer hit the head page * @@ -2542,9 +3362,11 @@ rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer, * the counters. */ local_add(entries, &cpu_buffer->overrun); - local_sub(BUF_PAGE_SIZE, &cpu_buffer->entries_bytes); + local_sub(rb_page_commit(next_page), &cpu_buffer->entries_bytes); local_inc(&cpu_buffer->pages_lost); + if (cpu_buffer->ring_meta) + rb_update_meta_head(cpu_buffer, next_page); /* * The entries will be zeroed out when we move the * tail page. @@ -2662,6 +3484,7 @@ static inline void rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, unsigned long tail, struct rb_event_info *info) { + unsigned long bsize = READ_ONCE(cpu_buffer->buffer->subbuf_size); struct buffer_page *tail_page = info->tail_page; struct ring_buffer_event *event; unsigned long length = info->length; @@ -2670,13 +3493,13 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, * Only the event that crossed the page boundary * must fill the old tail_page with padding. */ - if (tail >= BUF_PAGE_SIZE) { + if (tail >= bsize) { /* * If the page was filled, then we still need * to update the real_end. Reset it to zero * and the reader will ignore it. */ - if (tail == BUF_PAGE_SIZE) + if (tail == bsize) tail_page->real_end = 0; local_sub(length, &tail_page->write); @@ -2685,9 +3508,6 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, event = __rb_page_index(tail_page, tail); - /* account for padding bytes */ - local_add(BUF_PAGE_SIZE - tail, &cpu_buffer->entries_bytes); - /* * Save the original length to the meta data. * This will be used by the reader to add lost event @@ -2701,12 +3521,13 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, * write counter enough to allow another writer to slip * in on this page. * We put in a discarded commit instead, to make sure - * that this space is not used again. + * that this space is not used again, and this space will + * not be accounted into 'entries_bytes'. * * If we are less than the minimum size, we don't need to * worry about it. */ - if (tail > (BUF_PAGE_SIZE - RB_EVNT_MIN_SIZE)) { + if (tail > (bsize - RB_EVNT_MIN_SIZE)) { /* No room for any events */ /* Mark the rest of the page with padding */ @@ -2721,16 +3542,19 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, } /* Put in a discarded event */ - event->array[0] = (BUF_PAGE_SIZE - tail) - RB_EVNT_HDR_SIZE; + event->array[0] = (bsize - tail) - RB_EVNT_HDR_SIZE; event->type_len = RINGBUF_TYPE_PADDING; /* time delta must be non zero */ event->time_delta = 1; + /* account for padding bytes */ + local_add(bsize - tail, &cpu_buffer->entries_bytes); + /* Make sure the padding is visible before the tail_page->write update */ smp_wmb(); /* Set write to end of buffer */ - length = (tail + length) - BUF_PAGE_SIZE; + length = (tail + length) - bsize; local_sub(length, &tail_page->write); } @@ -2844,7 +3668,8 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, /* Slow path */ static struct ring_buffer_event * -rb_add_time_stamp(struct ring_buffer_event *event, u64 delta, bool abs) +rb_add_time_stamp(struct ring_buffer_per_cpu *cpu_buffer, + struct ring_buffer_event *event, u64 delta, bool abs) { if (abs) event->type_len = RINGBUF_TYPE_TIME_STAMP; @@ -2852,7 +3677,7 @@ rb_add_time_stamp(struct ring_buffer_event *event, u64 delta, bool abs) event->type_len = RINGBUF_TYPE_TIME_EXTEND; /* Not the first event on the page, or not delta? */ - if (abs || rb_event_index(event)) { + if (abs || rb_event_index(cpu_buffer, event)) { event->time_delta = delta & TS_MASK; event->array[0] = delta >> TS_SHIFT; } else { @@ -2882,11 +3707,11 @@ rb_check_timestamp(struct ring_buffer_per_cpu *cpu_buffer, (unsigned long long)info->ts, (unsigned long long)info->before, (unsigned long long)info->after, - (unsigned long long)(rb_time_read(&cpu_buffer->write_stamp, &write_stamp) ? write_stamp : 0), + (unsigned long long)({rb_time_read(&cpu_buffer->write_stamp, &write_stamp); write_stamp;}), sched_clock_stable() ? "" : "If you just came from a suspend/resume,\n" "please switch to the trace global clock:\n" - " echo global > /sys/kernel/debug/tracing/trace_clock\n" + " echo global > /sys/kernel/tracing/trace_clock\n" "or add trace_clock=global to the kernel command line\n"); } @@ -2926,7 +3751,7 @@ static void rb_add_timestamp(struct ring_buffer_per_cpu *cpu_buffer, if (!abs) info->delta = 0; } - *event = rb_add_time_stamp(*event, info->delta, abs); + *event = rb_add_time_stamp(cpu_buffer, *event, info->delta, abs); *length -= RB_LEN_TIME_EXTEND; *delta = 0; } @@ -3002,78 +3827,49 @@ static unsigned rb_calculate_event_length(unsigned length) return length; } -static u64 rb_time_delta(struct ring_buffer_event *event) -{ - switch (event->type_len) { - case RINGBUF_TYPE_PADDING: - return 0; - - case RINGBUF_TYPE_TIME_EXTEND: - return rb_event_time_stamp(event); - - case RINGBUF_TYPE_TIME_STAMP: - return 0; - - case RINGBUF_TYPE_DATA: - return event->time_delta; - default: - return 0; - } -} - -static inline int +static inline bool rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, struct ring_buffer_event *event) { unsigned long new_index, old_index; struct buffer_page *bpage; - unsigned long index; unsigned long addr; - u64 write_stamp; - u64 delta; - new_index = rb_event_index(event); + new_index = rb_event_index(cpu_buffer, event); old_index = new_index + rb_event_ts_length(event); addr = (unsigned long)event; - addr &= PAGE_MASK; + addr &= ~((PAGE_SIZE << cpu_buffer->buffer->subbuf_order) - 1); bpage = READ_ONCE(cpu_buffer->tail_page); - delta = rb_time_delta(event); - - if (!rb_time_read(&cpu_buffer->write_stamp, &write_stamp)) - return 0; - - /* Make sure the write stamp is read before testing the location */ - barrier(); - + /* + * Make sure the tail_page is still the same and + * the next write location is the end of this event + */ if (bpage->page == (void *)addr && rb_page_write(bpage) == old_index) { unsigned long write_mask = local_read(&bpage->write) & ~RB_WRITE_MASK; unsigned long event_length = rb_event_length(event); - /* Something came in, can't discard */ - if (!rb_time_cmpxchg(&cpu_buffer->write_stamp, - write_stamp, write_stamp - delta)) - return 0; - /* - * It's possible that the event time delta is zero - * (has the same time stamp as the previous event) - * in which case write_stamp and before_stamp could - * be the same. In such a case, force before_stamp - * to be different than write_stamp. It doesn't - * matter what it is, as long as its different. + * For the before_stamp to be different than the write_stamp + * to make sure that the next event adds an absolute + * value and does not rely on the saved write stamp, which + * is now going to be bogus. + * + * By setting the before_stamp to zero, the next event + * is not going to use the write_stamp and will instead + * create an absolute timestamp. This means there's no + * reason to update the wirte_stamp! */ - if (!delta) - rb_time_set(&cpu_buffer->before_stamp, 0); + rb_time_set(&cpu_buffer->before_stamp, 0); /* * If an event were to come in now, it would see that the * write_stamp and the before_stamp are different, and assume * that this event just added itself before updating * the write stamp. The interrupting event will fix the - * write stamp for us, and use the before stamp as its delta. + * write stamp for us, and use an absolute timestamp. */ /* @@ -3084,16 +3880,17 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, */ old_index += write_mask; new_index += write_mask; - index = local_cmpxchg(&bpage->write, old_index, new_index); - if (index == old_index) { + + /* caution: old_index gets updated on cmpxchg failure */ + if (local_try_cmpxchg(&bpage->write, &old_index, new_index)) { /* update counters */ local_sub(event_length, &cpu_buffer->entries_bytes); - return 1; + return true; } } /* could not discard */ - return 0; + return false; } static void rb_start_commit(struct ring_buffer_per_cpu *cpu_buffer) @@ -3124,15 +3921,25 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) if (RB_WARN_ON(cpu_buffer, rb_is_reader_page(cpu_buffer->tail_page))) return; + /* + * No need for a memory barrier here, as the update + * of the tail_page did it for this page. + */ local_set(&cpu_buffer->commit_page->page->commit, rb_page_write(cpu_buffer->commit_page)); rb_inc_page(&cpu_buffer->commit_page); + if (cpu_buffer->ring_meta) { + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + meta->commit_buffer = (unsigned long)cpu_buffer->commit_page->page; + } /* add barrier to keep gcc from optimizing too much */ barrier(); } while (rb_commit_index(cpu_buffer) != rb_page_write(cpu_buffer->commit_page)) { + /* Make sure the readers see the content of what is committed. */ + smp_wmb(); local_set(&cpu_buffer->commit_page->page->commit, rb_page_write(cpu_buffer->commit_page)); RB_WARN_ON(cpu_buffer, @@ -3308,7 +4115,7 @@ rb_wakeups(struct trace_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer) * Note: The TRANSITION bit only handles a single transition between context. */ -static __always_inline int +static __always_inline bool trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer) { unsigned int val = cpu_buffer->current_context; @@ -3325,14 +4132,14 @@ trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer) bit = RB_CTX_TRANSITION; if (val & (1 << (bit + cpu_buffer->nest))) { do_ring_buffer_record_recursion(); - return 1; + return true; } } val |= (1 << (bit + cpu_buffer->nest)); cpu_buffer->current_context = val; - return 0; + return false; } static __always_inline void @@ -3394,7 +4201,6 @@ void ring_buffer_nest_end(struct trace_buffer *buffer) /** * ring_buffer_unlock_commit - commit a reserved * @buffer: The buffer to commit to - * @event: The event pointer to commit. * * This commits the data to the ring buffer, and releases any locks held. * @@ -3423,6 +4229,76 @@ EXPORT_SYMBOL_GPL(ring_buffer_unlock_commit); #define CHECK_FULL_PAGE 1L #ifdef CONFIG_RING_BUFFER_VALIDATE_TIME_DELTAS + +static const char *show_irq_str(int bits) +{ + static const char * type[] = { + ".", // 0 + "s", // 1 + "h", // 2 + "Hs", // 3 + "n", // 4 + "Ns", // 5 + "Nh", // 6 + "NHs", // 7 + }; + + return type[bits]; +} + +/* Assume this is a trace event */ +static const char *show_flags(struct ring_buffer_event *event) +{ + struct trace_entry *entry; + int bits = 0; + + if (rb_event_data_length(event) - RB_EVNT_HDR_SIZE < sizeof(*entry)) + return "X"; + + entry = ring_buffer_event_data(event); + + if (entry->flags & TRACE_FLAG_SOFTIRQ) + bits |= 1; + + if (entry->flags & TRACE_FLAG_HARDIRQ) + bits |= 2; + + if (entry->flags & TRACE_FLAG_NMI) + bits |= 4; + + return show_irq_str(bits); +} + +static const char *show_irq(struct ring_buffer_event *event) +{ + struct trace_entry *entry; + + if (rb_event_data_length(event) - RB_EVNT_HDR_SIZE < sizeof(*entry)) + return ""; + + entry = ring_buffer_event_data(event); + if (entry->flags & TRACE_FLAG_IRQS_OFF) + return "d"; + return ""; +} + +static const char *show_interrupt_level(void) +{ + unsigned long pc = preempt_count(); + unsigned char level = 0; + + if (pc & SOFTIRQ_OFFSET) + level |= 1; + + if (pc & HARDIRQ_MASK) + level |= 2; + + if (pc & NMI_MASK) + level |= 4; + + return show_irq_str(level); +} + static void dump_buffer_page(struct buffer_data_page *bpage, struct rb_event_info *info, unsigned long tail) @@ -3443,34 +4319,57 @@ static void dump_buffer_page(struct buffer_data_page *bpage, case RINGBUF_TYPE_TIME_EXTEND: delta = rb_event_time_stamp(event); ts += delta; - pr_warn(" [%lld] delta:%lld TIME EXTEND\n", ts, delta); + pr_warn(" 0x%x: [%lld] delta:%lld TIME EXTEND\n", + e, ts, delta); break; case RINGBUF_TYPE_TIME_STAMP: delta = rb_event_time_stamp(event); ts = rb_fix_abs_ts(delta, ts); - pr_warn(" [%lld] absolute:%lld TIME STAMP\n", ts, delta); + pr_warn(" 0x%x: [%lld] absolute:%lld TIME STAMP\n", + e, ts, delta); break; case RINGBUF_TYPE_PADDING: ts += event->time_delta; - pr_warn(" [%lld] delta:%d PADDING\n", ts, event->time_delta); + pr_warn(" 0x%x: [%lld] delta:%d PADDING\n", + e, ts, event->time_delta); break; case RINGBUF_TYPE_DATA: ts += event->time_delta; - pr_warn(" [%lld] delta:%d\n", ts, event->time_delta); + pr_warn(" 0x%x: [%lld] delta:%d %s%s\n", + e, ts, event->time_delta, + show_flags(event), show_irq(event)); break; default: break; } } + pr_warn("expected end:0x%lx last event actually ended at:0x%x\n", tail, e); } static DEFINE_PER_CPU(atomic_t, checking); static atomic_t ts_dump; +#define buffer_warn_return(fmt, ...) \ + do { \ + /* If another report is happening, ignore this one */ \ + if (atomic_inc_return(&ts_dump) != 1) { \ + atomic_dec(&ts_dump); \ + goto out; \ + } \ + atomic_inc(&cpu_buffer->record_disabled); \ + pr_warn(fmt, ##__VA_ARGS__); \ + dump_buffer_page(bpage, info, tail); \ + atomic_dec(&ts_dump); \ + /* There's some cases in boot up that this can happen */ \ + if (WARN_ON_ONCE(system_state != SYSTEM_BOOTING)) \ + /* Do not re-enable checking */ \ + return; \ + } while (0) + /* * Check if the current event time stamp matches the deltas on * the buffer page. @@ -3479,11 +4378,10 @@ static void check_buffer(struct ring_buffer_per_cpu *cpu_buffer, struct rb_event_info *info, unsigned long tail) { - struct ring_buffer_event *event; struct buffer_data_page *bpage; u64 ts, delta; bool full = false; - int e; + int ret; bpage = info->tail_page->page; @@ -3504,60 +4402,26 @@ static void check_buffer(struct ring_buffer_per_cpu *cpu_buffer, return; /* - * If this interrupted another event, + * If this interrupted another event, */ if (atomic_inc_return(this_cpu_ptr(&checking)) != 1) goto out; - ts = bpage->time_stamp; - - for (e = 0; e < tail; e += rb_event_length(event)) { - - event = (struct ring_buffer_event *)(bpage->data + e); - - switch (event->type_len) { - - case RINGBUF_TYPE_TIME_EXTEND: - delta = rb_event_time_stamp(event); - ts += delta; - break; - - case RINGBUF_TYPE_TIME_STAMP: - delta = rb_event_time_stamp(event); - ts = rb_fix_abs_ts(delta, ts); - break; - - case RINGBUF_TYPE_PADDING: - if (event->time_delta == 1) - break; - fallthrough; - case RINGBUF_TYPE_DATA: - ts += event->time_delta; - break; - - default: - RB_WARN_ON(cpu_buffer, 1); + ret = rb_read_data_buffer(bpage, tail, cpu_buffer->cpu, &ts, &delta); + if (ret < 0) { + if (delta < ts) { + buffer_warn_return("[CPU: %d]ABSOLUTE TIME WENT BACKWARDS: last ts: %lld absolute ts: %lld\n", + cpu_buffer->cpu, ts, delta); + goto out; } } if ((full && ts > info->ts) || (!full && ts + info->delta != info->ts)) { - /* If another report is happening, ignore this one */ - if (atomic_inc_return(&ts_dump) != 1) { - atomic_dec(&ts_dump); - goto out; - } - atomic_inc(&cpu_buffer->record_disabled); - /* There's some cases in boot up that this can happen */ - WARN_ON_ONCE(system_state != SYSTEM_BOOTING); - pr_warn("[CPU: %d]TIME DOES NOT MATCH expected:%lld actual:%lld delta:%lld before:%lld after:%lld%s\n", - cpu_buffer->cpu, - ts + info->delta, info->ts, info->delta, - info->before, info->after, - full ? " (full)" : ""); - dump_buffer_page(bpage, info, tail); - atomic_dec(&ts_dump); - /* Do not re-enable checking */ - return; + buffer_warn_return("[CPU: %d]TIME DOES NOT MATCH expected:%lld actual:%lld delta:%lld before:%lld after:%lld%s context:%s\n", + cpu_buffer->cpu, + ts + info->delta, info->ts, info->delta, + info->before, info->after, + full ? " (full)" : "", show_interrupt_level()); } out: atomic_dec(this_cpu_ptr(&checking)); @@ -3577,16 +4441,14 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, struct ring_buffer_event *event; struct buffer_page *tail_page; unsigned long tail, write, w; - bool a_ok; - bool b_ok; /* Don't let the compiler play games with cpu_buffer->tail_page */ tail_page = info->tail_page = READ_ONCE(cpu_buffer->tail_page); /*A*/ w = local_read(&tail_page->write) & RB_WRITE_MASK; barrier(); - b_ok = rb_time_read(&cpu_buffer->before_stamp, &info->before); - a_ok = rb_time_read(&cpu_buffer->write_stamp, &info->after); + rb_time_read(&cpu_buffer->before_stamp, &info->before); + rb_time_read(&cpu_buffer->write_stamp, &info->after); barrier(); info->ts = rb_time_stamp(cpu_buffer->buffer); @@ -3598,7 +4460,10 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, * absolute timestamp. * Don't bother if this is the start of a new page (w == 0). */ - if (unlikely(!a_ok || !b_ok || (info->before != info->after && w))) { + if (!w) { + /* Use the sub-buffer timestamp */ + info->delta = 0; + } else if (unlikely(info->before != info->after)) { info->add_timestamp |= RB_ADD_STAMP_FORCE | RB_ADD_STAMP_EXTEND; info->length += RB_LEN_TIME_EXTEND; } else { @@ -3620,27 +4485,20 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, tail = write - info->length; /* See if we shot pass the end of this buffer page */ - if (unlikely(write > BUF_PAGE_SIZE)) { - /* before and after may now different, fix it up*/ - b_ok = rb_time_read(&cpu_buffer->before_stamp, &info->before); - a_ok = rb_time_read(&cpu_buffer->write_stamp, &info->after); - if (a_ok && b_ok && info->before != info->after) - (void)rb_time_cmpxchg(&cpu_buffer->before_stamp, - info->before, info->after); - if (a_ok && b_ok) - check_buffer(cpu_buffer, info, CHECK_FULL_PAGE); + if (unlikely(write > cpu_buffer->buffer->subbuf_size)) { + check_buffer(cpu_buffer, info, CHECK_FULL_PAGE); return rb_move_tail(cpu_buffer, tail, info); } if (likely(tail == w)) { - u64 save_before; - bool s_ok; - /* Nothing interrupted us between A and C */ /*D*/ rb_time_set(&cpu_buffer->write_stamp, info->ts); - barrier(); - /*E*/ s_ok = rb_time_read(&cpu_buffer->before_stamp, &save_before); - RB_WARN_ON(cpu_buffer, !s_ok); + /* + * If something came in between C and D, the write stamp + * may now not be in sync. But that's fine as the before_stamp + * will be different and then next event will just be forced + * to use an absolute timestamp. + */ if (likely(!(info->add_timestamp & (RB_ADD_STAMP_FORCE | RB_ADD_STAMP_ABSOLUTE)))) /* This did not interrupt any time update */ @@ -3648,41 +4506,37 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, else /* Just use full timestamp for interrupting event */ info->delta = info->ts; - barrier(); check_buffer(cpu_buffer, info, tail); - if (unlikely(info->ts != save_before)) { - /* SLOW PATH - Interrupted between C and E */ - - a_ok = rb_time_read(&cpu_buffer->write_stamp, &info->after); - RB_WARN_ON(cpu_buffer, !a_ok); - - /* Write stamp must only go forward */ - if (save_before > info->after) { - /* - * We do not care about the result, only that - * it gets updated atomically. - */ - (void)rb_time_cmpxchg(&cpu_buffer->write_stamp, - info->after, save_before); - } - } } else { u64 ts; /* SLOW PATH - Interrupted between A and C */ - a_ok = rb_time_read(&cpu_buffer->write_stamp, &info->after); - /* Was interrupted before here, write_stamp must be valid */ - RB_WARN_ON(cpu_buffer, !a_ok); + + /* Save the old before_stamp */ + rb_time_read(&cpu_buffer->before_stamp, &info->before); + + /* + * Read a new timestamp and update the before_stamp to make + * the next event after this one force using an absolute + * timestamp. This is in case an interrupt were to come in + * between E and F. + */ ts = rb_time_stamp(cpu_buffer->buffer); + rb_time_set(&cpu_buffer->before_stamp, ts); + + barrier(); + /*E*/ rb_time_read(&cpu_buffer->write_stamp, &info->after); barrier(); - /*E*/ if (write == (local_read(&tail_page->write) & RB_WRITE_MASK) && - info->after < ts && - rb_time_cmpxchg(&cpu_buffer->write_stamp, - info->after, ts)) { - /* Nothing came after this event between C and E */ + /*F*/ if (write == (local_read(&tail_page->write) & RB_WRITE_MASK) && + info->after == info->before && info->after < ts) { + /* + * Nothing came after this event between C and F, it is + * safe to use info->after for the delta as it + * matched info->before and is still valid. + */ info->delta = ts - info->after; } else { /* - * Interrupted between C and E: + * Interrupted between C and F: * Lost the previous events time stamp. Just set the * delta to zero, and this will be the same time as * the event this event interrupted. And the events that @@ -3733,6 +4587,17 @@ rb_reserve_next_event(struct trace_buffer *buffer, int nr_loops = 0; int add_ts_default; + /* + * ring buffer does cmpxchg as well as atomic64 operations + * (which some archs use locking for atomic64), make sure this + * is safe in NMI context + */ + if ((!IS_ENABLED(CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG) || + IS_ENABLED(CONFIG_GENERIC_ATOMIC64)) && + (unlikely(in_nmi()))) { + return NULL; + } + rb_start_commit(cpu_buffer); /* The commit page can not change after this */ @@ -3756,6 +4621,8 @@ rb_reserve_next_event(struct trace_buffer *buffer, if (ring_buffer_time_stamp_abs(cpu_buffer->buffer)) { add_ts_default = RB_ADD_STAMP_ABSOLUTE; info.length += RB_LEN_TIME_EXTEND; + if (info.length > cpu_buffer->buffer->max_data_size) + goto out_fail; } else { add_ts_default = RB_ADD_STAMP_NONE; } @@ -3829,7 +4696,7 @@ ring_buffer_lock_reserve(struct trace_buffer *buffer, unsigned long length) if (unlikely(atomic_read(&cpu_buffer->record_disabled))) goto out; - if (unlikely(length > BUF_MAX_DATA_SIZE)) + if (unlikely(length > buffer->max_data_size)) goto out; if (unlikely(trace_recursive_lock(cpu_buffer))) @@ -3863,7 +4730,7 @@ rb_decrement_entry(struct ring_buffer_per_cpu *cpu_buffer, struct buffer_page *bpage = cpu_buffer->commit_page; struct buffer_page *start; - addr &= PAGE_MASK; + addr &= ~((PAGE_SIZE << cpu_buffer->buffer->subbuf_order) - 1); /* Do the likely case first */ if (likely(bpage->page == (void *)addr)) { @@ -3928,10 +4795,7 @@ void ring_buffer_discard_commit(struct trace_buffer *buffer, RB_WARN_ON(buffer, !local_read(&cpu_buffer->committing)); rb_decrement_entry(cpu_buffer, event); - if (rb_try_to_discard(cpu_buffer, event)) - goto out; - - out: + rb_try_to_discard(cpu_buffer, event); rb_end_commit(cpu_buffer); trace_recursive_unlock(cpu_buffer); @@ -3964,26 +4828,26 @@ int ring_buffer_write(struct trace_buffer *buffer, int ret = -EBUSY; int cpu; - preempt_disable_notrace(); + guard(preempt_notrace)(); if (atomic_read(&buffer->record_disabled)) - goto out; + return -EBUSY; cpu = raw_smp_processor_id(); if (!cpumask_test_cpu(cpu, buffer->cpumask)) - goto out; + return -EBUSY; cpu_buffer = buffer->buffers[cpu]; if (atomic_read(&cpu_buffer->record_disabled)) - goto out; + return -EBUSY; - if (length > BUF_MAX_DATA_SIZE) - goto out; + if (length > buffer->max_data_size) + return -EBUSY; if (unlikely(trace_recursive_lock(cpu_buffer))) - goto out; + return -EBUSY; event = rb_reserve_next_event(buffer, cpu_buffer, length); if (!event) @@ -4001,48 +4865,26 @@ int ring_buffer_write(struct trace_buffer *buffer, out_unlock: trace_recursive_unlock(cpu_buffer); - - out: - preempt_enable_notrace(); - return ret; } EXPORT_SYMBOL_GPL(ring_buffer_write); -static bool rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer) +/* + * The total entries in the ring buffer is the running counter + * of entries entered into the ring buffer, minus the sum of + * the entries read from the ring buffer and the number of + * entries that were overwritten. + */ +static inline unsigned long +rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer) { - struct buffer_page *reader = cpu_buffer->reader_page; - struct buffer_page *head = rb_set_head_page(cpu_buffer); - struct buffer_page *commit = cpu_buffer->commit_page; - - /* In case of error, head will be NULL */ - if (unlikely(!head)) - return true; - - /* Reader should exhaust content in reader page */ - if (reader->read != rb_page_commit(reader)) - return false; - - /* - * If writers are committing on the reader page, knowing all - * committed content has been read, the ring buffer is empty. - */ - if (commit == reader) - return true; - - /* - * If writers are committing on a page other than reader page - * and head page, there should always be content to read. - */ - if (commit != head) - return false; + return local_read(&cpu_buffer->entries) - + (local_read(&cpu_buffer->overrun) + cpu_buffer->read); +} - /* - * Writers are committing on the head page, we just need - * to care about there're committed data, and the reader will - * swap reader page with head page when it is to read data. - */ - return rb_page_commit(commit) == 0; +static bool rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer) +{ + return !rb_num_of_entries(cpu_buffer); } /** @@ -4089,10 +4931,10 @@ void ring_buffer_record_off(struct trace_buffer *buffer) unsigned int rd; unsigned int new_rd; + rd = atomic_read(&buffer->record_disabled); do { - rd = atomic_read(&buffer->record_disabled); new_rd = rd | RB_BUFFER_OFF; - } while (atomic_cmpxchg(&buffer->record_disabled, rd, new_rd) != rd); + } while (!atomic_try_cmpxchg(&buffer->record_disabled, &rd, new_rd)); } EXPORT_SYMBOL_GPL(ring_buffer_record_off); @@ -4112,10 +4954,10 @@ void ring_buffer_record_on(struct trace_buffer *buffer) unsigned int rd; unsigned int new_rd; + rd = atomic_read(&buffer->record_disabled); do { - rd = atomic_read(&buffer->record_disabled); new_rd = rd & ~RB_BUFFER_OFF; - } while (atomic_cmpxchg(&buffer->record_disabled, rd, new_rd) != rd); + } while (!atomic_try_cmpxchg(&buffer->record_disabled, &rd, new_rd)); } EXPORT_SYMBOL_GPL(ring_buffer_record_on); @@ -4147,6 +4989,24 @@ bool ring_buffer_record_is_set_on(struct trace_buffer *buffer) } /** + * ring_buffer_record_is_on_cpu - return true if the ring buffer can write + * @buffer: The ring buffer to see if write is enabled + * @cpu: The CPU to test if the ring buffer can write too + * + * Returns true if the ring buffer is in a state that it accepts writes + * for a particular CPU. + */ +bool ring_buffer_record_is_on_cpu(struct trace_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + + cpu_buffer = buffer->buffers[cpu]; + + return ring_buffer_record_is_set_on(buffer) && + !atomic_read(&cpu_buffer->record_disabled); +} + +/** * ring_buffer_record_disable_cpu - stop all writes into the cpu_buffer * @buffer: The ring buffer to stop writes to. * @cpu: The CPU buffer to stop @@ -4188,19 +5048,6 @@ void ring_buffer_record_enable_cpu(struct trace_buffer *buffer, int cpu) } EXPORT_SYMBOL_GPL(ring_buffer_record_enable_cpu); -/* - * The total entries in the ring buffer is the running counter - * of entries entered into the ring buffer, minus the sum of - * the entries read from the ring buffer and the number of - * entries that were overwritten. - */ -static inline unsigned long -rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer) -{ - return local_read(&cpu_buffer->entries) - - (local_read(&cpu_buffer->overrun) + cpu_buffer->read); -} - /** * ring_buffer_oldest_event_ts - get the oldest event timestamp from the buffer * @buffer: The ring buffer @@ -4235,7 +5082,7 @@ u64 ring_buffer_oldest_event_ts(struct trace_buffer *buffer, int cpu) EXPORT_SYMBOL_GPL(ring_buffer_oldest_event_ts); /** - * ring_buffer_bytes_cpu - get the number of bytes consumed in a cpu buffer + * ring_buffer_bytes_cpu - get the number of bytes unconsumed in a cpu buffer * @buffer: The ring buffer * @cpu: The per CPU buffer to read from. */ @@ -4413,6 +5260,7 @@ static void rb_iter_reset(struct ring_buffer_iter *iter) iter->cache_reader_page = iter->head_page; iter->cache_read = cpu_buffer->read; + iter->cache_pages_removed = cpu_buffer->pages_removed; if (iter->head) { iter->read_stamp = cpu_buffer->read_stamp; @@ -4464,7 +5312,7 @@ int ring_buffer_iter_empty(struct ring_buffer_iter *iter) cpu_buffer = iter->cpu_buffer; reader = cpu_buffer->reader_page; head_page = cpu_buffer->head_page; - commit_page = cpu_buffer->commit_page; + commit_page = READ_ONCE(cpu_buffer->commit_page); commit_ts = commit_page->page->time_stamp; /* @@ -4490,7 +5338,7 @@ int ring_buffer_iter_empty(struct ring_buffer_iter *iter) return ((iter->head_page == commit_page && iter->head >= commit) || (iter->head_page == reader && commit_page == head_page && head_page->read == commit && - iter->head == rb_page_commit(cpu_buffer->reader_page))); + iter->head == rb_page_size(cpu_buffer->reader_page))); } EXPORT_SYMBOL_GPL(ring_buffer_iter_empty); @@ -4522,7 +5370,6 @@ rb_update_read_stamp(struct ring_buffer_per_cpu *cpu_buffer, default: RB_WARN_ON(cpu_buffer, 1); } - return; } static void @@ -4553,17 +5400,17 @@ rb_update_iter_read_stamp(struct ring_buffer_iter *iter, default: RB_WARN_ON(iter->cpu_buffer, 1); } - return; } static struct buffer_page * rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) { struct buffer_page *reader = NULL; + unsigned long bsize = READ_ONCE(cpu_buffer->buffer->subbuf_size); unsigned long overwrite; unsigned long flags; int nr_loops = 0; - int ret; + bool ret; local_irq_save(flags); arch_spin_lock(&cpu_buffer->lock); @@ -4605,7 +5452,6 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) */ local_set(&cpu_buffer->reader_page->write, 0); local_set(&cpu_buffer->reader_page->entries, 0); - local_set(&cpu_buffer->reader_page->page->commit, 0); cpu_buffer->reader_page->real_end = 0; spin: @@ -4648,7 +5494,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) * moving it. The page before the header page has the * flag bit '1' set if it is pointing to the page we want. * but if the writer is in the process of moving it - * than it will be '2' or already moved '0'. + * then it will be '2' or already moved '0'. */ ret = rb_head_page_replace(reader, cpu_buffer->reader_page); @@ -4659,6 +5505,9 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) if (!ret) goto spin; + if (cpu_buffer->ring_meta) + rb_update_meta_reader(cpu_buffer, reader); + /* * Yay! We succeeded in replacing the page. * @@ -4667,6 +5516,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) rb_list_head(reader->list.next)->prev = &cpu_buffer->reader_page->list; rb_inc_page(&cpu_buffer->head_page); + cpu_buffer->cnt++; local_inc(&cpu_buffer->pages_read); /* Finally update the reader page to the new head */ @@ -4695,7 +5545,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) #define USECS_WAIT 1000000 for (nr_loops = 0; nr_loops < USECS_WAIT; nr_loops++) { /* If the write is past the end of page, a writer is still updating it */ - if (likely(!reader || rb_page_write(reader) <= BUF_PAGE_SIZE)) + if (likely(!reader || rb_page_write(reader) <= bsize)) break; udelay(1); @@ -4710,7 +5560,12 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) /* * Make sure we see any padding after the write update - * (see rb_reset_tail()) + * (see rb_reset_tail()). + * + * In addition, a writer may be writing on the reader page + * if the page has not been fully filled, so the read barrier + * is also needed to make sure we see the content of what is + * committed by the writer (see rb_set_commit_to_write()). */ smp_rmb(); @@ -4739,6 +5594,7 @@ static void rb_advance_reader(struct ring_buffer_per_cpu *cpu_buffer) length = rb_event_length(event); cpu_buffer->reader_page->read += length; + cpu_buffer->read_bytes += length; } static void rb_advance_iter(struct ring_buffer_iter *iter) @@ -4864,12 +5720,13 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) buffer = cpu_buffer->buffer; /* - * Check if someone performed a consuming read to - * the buffer. A consuming read invalidates the iterator - * and we need to reset the iterator in this case. + * Check if someone performed a consuming read to the buffer + * or removed some pages from the buffer. In these cases, + * iterator was invalidated and we need to reset it. */ if (unlikely(iter->cache_read != cpu_buffer->read || - iter->cache_reader_page != cpu_buffer->reader_page)) + iter->cache_reader_page != cpu_buffer->reader_page || + iter->cache_pages_removed != cpu_buffer->pages_removed)) rb_iter_reset(iter); again: @@ -4968,7 +5825,6 @@ rb_reader_unlock(struct ring_buffer_per_cpu *cpu_buffer, bool locked) { if (likely(locked)) raw_spin_unlock(&cpu_buffer->reader_lock); - return; } /** @@ -5099,28 +5955,20 @@ ring_buffer_consume(struct trace_buffer *buffer, int cpu, u64 *ts, EXPORT_SYMBOL_GPL(ring_buffer_consume); /** - * ring_buffer_read_prepare - Prepare for a non consuming read of the buffer + * ring_buffer_read_start - start a non consuming read of the buffer * @buffer: The ring buffer to read from * @cpu: The cpu buffer to iterate over * @flags: gfp flags to use for memory allocation * - * This performs the initial preparations necessary to iterate - * through the buffer. Memory is allocated, buffer recording - * is disabled, and the iterator pointer is returned to the caller. + * This creates an iterator to allow non-consuming iteration through + * the buffer. If the buffer is disabled for writing, it will produce + * the same information each time, but if the buffer is still writing + * then the first hit of a write will cause the iteration to stop. * - * Disabling buffer recording prevents the reading from being - * corrupted. This is not a consuming read, so a producer is not - * expected. - * - * After a sequence of ring_buffer_read_prepare calls, the user is - * expected to make at least one call to ring_buffer_read_prepare_sync. - * Afterwards, ring_buffer_read_start is invoked to get things going - * for real. - * - * This overall must be paired with ring_buffer_read_finish. + * Must be paired with ring_buffer_read_finish. */ struct ring_buffer_iter * -ring_buffer_read_prepare(struct trace_buffer *buffer, int cpu, gfp_t flags) +ring_buffer_read_start(struct trace_buffer *buffer, int cpu, gfp_t flags) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_iter *iter; @@ -5132,7 +5980,9 @@ ring_buffer_read_prepare(struct trace_buffer *buffer, int cpu, gfp_t flags) if (!iter) return NULL; - iter->event = kmalloc(BUF_MAX_DATA_SIZE, flags); + /* Holds the entire event: data and meta data */ + iter->event_size = buffer->subbuf_size; + iter->event = kmalloc(iter->event_size, flags); if (!iter->event) { kfree(iter); return NULL; @@ -5144,51 +5994,12 @@ ring_buffer_read_prepare(struct trace_buffer *buffer, int cpu, gfp_t flags) atomic_inc(&cpu_buffer->resize_disabled); - return iter; -} -EXPORT_SYMBOL_GPL(ring_buffer_read_prepare); - -/** - * ring_buffer_read_prepare_sync - Synchronize a set of prepare calls - * - * All previously invoked ring_buffer_read_prepare calls to prepare - * iterators will be synchronized. Afterwards, read_buffer_read_start - * calls on those iterators are allowed. - */ -void -ring_buffer_read_prepare_sync(void) -{ - synchronize_rcu(); -} -EXPORT_SYMBOL_GPL(ring_buffer_read_prepare_sync); - -/** - * ring_buffer_read_start - start a non consuming read of the buffer - * @iter: The iterator returned by ring_buffer_read_prepare - * - * This finalizes the startup of an iteration through the buffer. - * The iterator comes from a call to ring_buffer_read_prepare and - * an intervening ring_buffer_read_prepare_sync must have been - * performed. - * - * Must be paired with ring_buffer_read_finish. - */ -void -ring_buffer_read_start(struct ring_buffer_iter *iter) -{ - struct ring_buffer_per_cpu *cpu_buffer; - unsigned long flags; - - if (!iter) - return; - - cpu_buffer = iter->cpu_buffer; - - raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + guard(raw_spinlock_irqsave)(&cpu_buffer->reader_lock); arch_spin_lock(&cpu_buffer->lock); rb_iter_reset(iter); arch_spin_unlock(&cpu_buffer->lock); - raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + return iter; } EXPORT_SYMBOL_GPL(ring_buffer_read_start); @@ -5196,24 +6007,15 @@ EXPORT_SYMBOL_GPL(ring_buffer_read_start); * ring_buffer_read_finish - finish reading the iterator of the buffer * @iter: The iterator retrieved by ring_buffer_start * - * This re-enables the recording to the buffer, and frees the - * iterator. + * This re-enables resizing of the buffer, and frees the iterator. */ void ring_buffer_read_finish(struct ring_buffer_iter *iter) { struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; - unsigned long flags; - /* - * Ring buffer is disabled from recording, here's a good place - * to check the integrity of the ring buffer. - * Must prevent readers from trying to read, as the check - * clears the HEAD page and readers require it. - */ - raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + /* Use this opportunity to check the integrity of the ring buffer. */ rb_check_pages(cpu_buffer); - raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); atomic_dec(&cpu_buffer->resize_disabled); kfree(iter->event); @@ -5248,41 +6050,110 @@ EXPORT_SYMBOL_GPL(ring_buffer_iter_advance); */ unsigned long ring_buffer_size(struct trace_buffer *buffer, int cpu) { - /* - * Earlier, this method returned - * BUF_PAGE_SIZE * buffer->nr_pages - * Since the nr_pages field is now removed, we have converted this to - * return the per cpu buffer value. - */ if (!cpumask_test_cpu(cpu, buffer->cpumask)) return 0; - return BUF_PAGE_SIZE * buffer->buffers[cpu]->nr_pages; + return buffer->subbuf_size * buffer->buffers[cpu]->nr_pages; } EXPORT_SYMBOL_GPL(ring_buffer_size); +/** + * ring_buffer_max_event_size - return the max data size of an event + * @buffer: The ring buffer. + * + * Returns the maximum size an event can be. + */ +unsigned long ring_buffer_max_event_size(struct trace_buffer *buffer) +{ + /* If abs timestamp is requested, events have a timestamp too */ + if (ring_buffer_time_stamp_abs(buffer)) + return buffer->max_data_size - RB_LEN_TIME_EXTEND; + return buffer->max_data_size; +} +EXPORT_SYMBOL_GPL(ring_buffer_max_event_size); + +static void rb_clear_buffer_page(struct buffer_page *page) +{ + local_set(&page->write, 0); + local_set(&page->entries, 0); + rb_init_page(page->page); + page->read = 0; +} + +/* + * When the buffer is memory mapped to user space, each sub buffer + * has a unique id that is used by the meta data to tell the user + * where the current reader page is. + * + * For a normal allocated ring buffer, the id is saved in the buffer page + * id field, and updated via this function. + * + * But for a fixed memory mapped buffer, the id is already assigned for + * fixed memory ordering in the memory layout and can not be used. Instead + * the index of where the page lies in the memory layout is used. + * + * For the normal pages, set the buffer page id with the passed in @id + * value and return that. + * + * For fixed memory mapped pages, get the page index in the memory layout + * and return that as the id. + */ +static int rb_page_id(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *bpage, int id) +{ + /* + * For boot buffers, the id is the index, + * otherwise, set the buffer page with this id + */ + if (cpu_buffer->ring_meta) + id = rb_meta_subbuf_idx(cpu_buffer->ring_meta, bpage->page); + else + bpage->id = id; + + return id; +} + +static void rb_update_meta_page(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct trace_buffer_meta *meta = cpu_buffer->meta_page; + + if (!meta) + return; + + meta->reader.read = cpu_buffer->reader_page->read; + meta->reader.id = rb_page_id(cpu_buffer, cpu_buffer->reader_page, + cpu_buffer->reader_page->id); + + meta->reader.lost_events = cpu_buffer->lost_events; + + meta->entries = local_read(&cpu_buffer->entries); + meta->overrun = local_read(&cpu_buffer->overrun); + meta->read = cpu_buffer->read; + + /* Some archs do not have data cache coherency between kernel and user-space */ + flush_kernel_vmap_range(cpu_buffer->meta_page, PAGE_SIZE); +} + static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) { + struct buffer_page *page; + rb_head_page_deactivate(cpu_buffer); cpu_buffer->head_page = list_entry(cpu_buffer->pages, struct buffer_page, list); - local_set(&cpu_buffer->head_page->write, 0); - local_set(&cpu_buffer->head_page->entries, 0); - local_set(&cpu_buffer->head_page->page->commit, 0); - - cpu_buffer->head_page->read = 0; + rb_clear_buffer_page(cpu_buffer->head_page); + list_for_each_entry(page, cpu_buffer->pages, list) { + rb_clear_buffer_page(page); + } cpu_buffer->tail_page = cpu_buffer->head_page; cpu_buffer->commit_page = cpu_buffer->head_page; INIT_LIST_HEAD(&cpu_buffer->reader_page->list); INIT_LIST_HEAD(&cpu_buffer->new_pages); - local_set(&cpu_buffer->reader_page->write, 0); - local_set(&cpu_buffer->reader_page->entries, 0); - local_set(&cpu_buffer->reader_page->page->commit, 0); - cpu_buffer->reader_page->read = 0; + rb_clear_buffer_page(cpu_buffer->reader_page); local_set(&cpu_buffer->entries_bytes, 0); local_set(&cpu_buffer->overrun, 0); @@ -5308,26 +6179,30 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->last_overrun = 0; rb_head_page_activate(cpu_buffer); + cpu_buffer->pages_removed = 0; + + if (cpu_buffer->mapped) { + rb_update_meta_page(cpu_buffer); + if (cpu_buffer->ring_meta) { + struct ring_buffer_cpu_meta *meta = cpu_buffer->ring_meta; + meta->commit_buffer = meta->head_buffer; + } + } } /* Must have disabled the cpu buffer then done a synchronize_rcu */ static void reset_disabled_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer) { - unsigned long flags; - - raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + guard(raw_spinlock_irqsave)(&cpu_buffer->reader_lock); if (RB_WARN_ON(cpu_buffer, local_read(&cpu_buffer->committing))) - goto out; + return; arch_spin_lock(&cpu_buffer->lock); rb_reset_cpu(cpu_buffer); arch_spin_unlock(&cpu_buffer->lock); - - out: - raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); } /** @@ -5360,10 +6235,12 @@ void ring_buffer_reset_cpu(struct trace_buffer *buffer, int cpu) } EXPORT_SYMBOL_GPL(ring_buffer_reset_cpu); +/* Flag to ensure proper resetting of atomic variables */ +#define RESET_BIT (1 << 30) + /** * ring_buffer_reset_online_cpus - reset a ring buffer per CPU buffer * @buffer: The ring buffer to reset a per cpu buffer of - * @cpu: The CPU buffer to be reset */ void ring_buffer_reset_online_cpus(struct trace_buffer *buffer) { @@ -5376,20 +6253,27 @@ void ring_buffer_reset_online_cpus(struct trace_buffer *buffer) for_each_online_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; - atomic_inc(&cpu_buffer->resize_disabled); + atomic_add(RESET_BIT, &cpu_buffer->resize_disabled); atomic_inc(&cpu_buffer->record_disabled); } /* Make sure all commits have finished */ synchronize_rcu(); - for_each_online_buffer_cpu(buffer, cpu) { + for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; + /* + * If a CPU came online during the synchronize_rcu(), then + * ignore it. + */ + if (!(atomic_read(&cpu_buffer->resize_disabled) & RESET_BIT)) + continue; + reset_disabled_cpu_buffer(cpu_buffer); atomic_dec(&cpu_buffer->record_disabled); - atomic_dec(&cpu_buffer->resize_disabled); + atomic_sub(RESET_BIT, &cpu_buffer->resize_disabled); } mutex_unlock(&buffer->mutex); @@ -5439,8 +6323,8 @@ bool ring_buffer_empty(struct trace_buffer *buffer) struct ring_buffer_per_cpu *cpu_buffer; unsigned long flags; bool dolock; + bool ret; int cpu; - int ret; /* yes this is racy, but if you don't like the race, lock the buffer */ for_each_buffer_cpu(buffer, cpu) { @@ -5469,7 +6353,7 @@ bool ring_buffer_empty_cpu(struct trace_buffer *buffer, int cpu) struct ring_buffer_per_cpu *cpu_buffer; unsigned long flags; bool dolock; - int ret; + bool ret; if (!cpumask_test_cpu(cpu, buffer->cpumask)) return true; @@ -5506,28 +6390,33 @@ int ring_buffer_swap_cpu(struct trace_buffer *buffer_a, if (!cpumask_test_cpu(cpu, buffer_a->cpumask) || !cpumask_test_cpu(cpu, buffer_b->cpumask)) - goto out; + return -EINVAL; cpu_buffer_a = buffer_a->buffers[cpu]; cpu_buffer_b = buffer_b->buffers[cpu]; + /* It's up to the callers to not try to swap mapped buffers */ + if (WARN_ON_ONCE(cpu_buffer_a->mapped || cpu_buffer_b->mapped)) + return -EBUSY; + /* At least make sure the two buffers are somewhat the same */ if (cpu_buffer_a->nr_pages != cpu_buffer_b->nr_pages) - goto out; + return -EINVAL; - ret = -EAGAIN; + if (buffer_a->subbuf_order != buffer_b->subbuf_order) + return -EINVAL; if (atomic_read(&buffer_a->record_disabled)) - goto out; + return -EAGAIN; if (atomic_read(&buffer_b->record_disabled)) - goto out; + return -EAGAIN; if (atomic_read(&cpu_buffer_a->record_disabled)) - goto out; + return -EAGAIN; if (atomic_read(&cpu_buffer_b->record_disabled)) - goto out; + return -EAGAIN; /* * We can't do a synchronize_rcu here because this @@ -5544,6 +6433,15 @@ int ring_buffer_swap_cpu(struct trace_buffer *buffer_a, if (local_read(&cpu_buffer_b->committing)) goto out_dec; + /* + * When resize is in progress, we cannot swap it because + * it will mess the state of the cpu buffer. + */ + if (atomic_read(&buffer_a->resizing)) + goto out_dec; + if (atomic_read(&buffer_b->resizing)) + goto out_dec; + buffer_a->buffers[cpu] = cpu_buffer_b; buffer_b->buffers[cpu] = cpu_buffer_a; @@ -5555,7 +6453,6 @@ int ring_buffer_swap_cpu(struct trace_buffer *buffer_a, out_dec: atomic_dec(&cpu_buffer_a->record_disabled); atomic_dec(&cpu_buffer_b->record_disabled); -out: return ret; } EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); @@ -5577,40 +6474,42 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); * Returns: * The page allocated, or ERR_PTR */ -void *ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu) +struct buffer_data_read_page * +ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; - struct buffer_data_page *bpage = NULL; + struct buffer_data_read_page *bpage = NULL; unsigned long flags; - struct page *page; if (!cpumask_test_cpu(cpu, buffer->cpumask)) return ERR_PTR(-ENODEV); + bpage = kzalloc(sizeof(*bpage), GFP_KERNEL); + if (!bpage) + return ERR_PTR(-ENOMEM); + + bpage->order = buffer->subbuf_order; cpu_buffer = buffer->buffers[cpu]; local_irq_save(flags); arch_spin_lock(&cpu_buffer->lock); if (cpu_buffer->free_page) { - bpage = cpu_buffer->free_page; + bpage->data = cpu_buffer->free_page; cpu_buffer->free_page = NULL; } arch_spin_unlock(&cpu_buffer->lock); local_irq_restore(flags); - if (bpage) - goto out; - - page = alloc_pages_node(cpu_to_node(cpu), - GFP_KERNEL | __GFP_NORETRY, 0); - if (!page) - return ERR_PTR(-ENOMEM); - - bpage = page_address(page); - - out: - rb_init_page(bpage); + if (bpage->data) { + rb_init_page(bpage->data); + } else { + bpage->data = alloc_cpu_data(cpu, cpu_buffer->buffer->subbuf_order); + if (!bpage->data) { + kfree(bpage); + return ERR_PTR(-ENOMEM); + } + } return bpage; } @@ -5620,19 +6519,29 @@ EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page); * ring_buffer_free_read_page - free an allocated read page * @buffer: the buffer the page was allocate for * @cpu: the cpu buffer the page came from - * @data: the page to free + * @data_page: the page to free * * Free a page allocated from ring_buffer_alloc_read_page. */ -void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data) +void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, + struct buffer_data_read_page *data_page) { - struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; - struct buffer_data_page *bpage = data; + struct ring_buffer_per_cpu *cpu_buffer; + struct buffer_data_page *bpage = data_page->data; struct page *page = virt_to_page(bpage); unsigned long flags; - /* If the page is still in use someplace else, we can't reuse it */ - if (page_ref_count(page) > 1) + if (!buffer || !buffer->buffers || !buffer->buffers[cpu]) + return; + + cpu_buffer = buffer->buffers[cpu]; + + /* + * If the page is still in use someplace else, or order of the page + * is different from the subbuffer order of the buffer - + * we can't reuse it + */ + if (page_ref_count(page) > 1 || data_page->order != buffer->subbuf_order) goto out; local_irq_save(flags); @@ -5647,7 +6556,8 @@ void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data local_irq_restore(flags); out: - free_page((unsigned long)bpage); + free_pages((unsigned long)bpage, data_page->order); + kfree(data_page); } EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); @@ -5668,9 +6578,10 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); * rpage = ring_buffer_alloc_read_page(buffer, cpu); * if (IS_ERR(rpage)) * return PTR_ERR(rpage); - * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); + * ret = ring_buffer_read_page(buffer, rpage, len, cpu, 0); * if (ret >= 0) - * process_page(rpage, ret); + * process_page(ring_buffer_read_page_data(rpage), ret); + * ring_buffer_free_read_page(buffer, cpu, rpage); * * When @full is set, the function will not return true unless * the writer is off the reader page. @@ -5685,48 +6596,50 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); * <0 if no data has been transferred. */ int ring_buffer_read_page(struct trace_buffer *buffer, - void **data_page, size_t len, int cpu, int full) + struct buffer_data_read_page *data_page, + size_t len, int cpu, int full) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; struct buffer_data_page *bpage; struct buffer_page *reader; unsigned long missed_events; - unsigned long flags; unsigned int commit; unsigned int read; u64 save_timestamp; - int ret = -1; if (!cpumask_test_cpu(cpu, buffer->cpumask)) - goto out; + return -1; /* * If len is not big enough to hold the page header, then * we can not copy anything. */ if (len <= BUF_PAGE_HDR_SIZE) - goto out; + return -1; len -= BUF_PAGE_HDR_SIZE; - if (!data_page) - goto out; + if (!data_page || !data_page->data) + return -1; + + if (data_page->order != buffer->subbuf_order) + return -1; - bpage = *data_page; + bpage = data_page->data; if (!bpage) - goto out; + return -1; - raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + guard(raw_spinlock_irqsave)(&cpu_buffer->reader_lock); reader = rb_get_reader_page(cpu_buffer); if (!reader) - goto out_unlock; + return -1; event = rb_reader_event(cpu_buffer); read = reader->read; - commit = rb_page_commit(reader); + commit = rb_page_size(reader); /* Check if any events were dropped */ missed_events = cpu_buffer->lost_events; @@ -5739,7 +6652,8 @@ int ring_buffer_read_page(struct trace_buffer *buffer, * Otherwise, we can simply swap the page with the one passed in. */ if (read || (len < (commit - read)) || - cpu_buffer->reader_page == cpu_buffer->commit_page) { + cpu_buffer->reader_page == cpu_buffer->commit_page || + cpu_buffer->mapped) { struct buffer_data_page *rpage = cpu_buffer->reader_page->page; unsigned int rpos = read; unsigned int pos = 0; @@ -5754,7 +6668,7 @@ int ring_buffer_read_page(struct trace_buffer *buffer, if (full && (!read || (len < (commit - read)) || cpu_buffer->reader_page == cpu_buffer->commit_page)) - goto out_unlock; + return -1; if (len > (commit - read)) len = (commit - read); @@ -5763,7 +6677,7 @@ int ring_buffer_read_page(struct trace_buffer *buffer, size = rb_event_ts_length(event); if (len < size) - goto out_unlock; + return -1; /* save the current timestamp, since the user will need it */ save_timestamp = cpu_buffer->read_stamp; @@ -5802,16 +6716,16 @@ int ring_buffer_read_page(struct trace_buffer *buffer, } else { /* update the entry counter */ cpu_buffer->read += rb_page_entries(reader); - cpu_buffer->read_bytes += BUF_PAGE_SIZE; + cpu_buffer->read_bytes += rb_page_size(reader); /* swap the pages */ rb_init_page(bpage); bpage = reader->page; - reader->page = *data_page; + reader->page = data_page->data; local_set(&reader->write, 0); local_set(&reader->entries, 0); reader->read = 0; - *data_page = bpage; + data_page->data = bpage; /* * Use the real_end for the data size, @@ -5821,7 +6735,6 @@ int ring_buffer_read_page(struct trace_buffer *buffer, if (reader->real_end) local_set(&bpage->commit, reader->real_end); } - ret = read; cpu_buffer->lost_events = 0; @@ -5833,7 +6746,7 @@ int ring_buffer_read_page(struct trace_buffer *buffer, /* If there is room at the end of the page to save the * missed events, then record it there. */ - if (BUF_PAGE_SIZE - commit >= sizeof(missed_events)) { + if (buffer->subbuf_size - commit >= sizeof(missed_events)) { memcpy(&bpage->data[commit], &missed_events, sizeof(missed_events)); local_add(RB_MISSED_STORED, &bpage->commit); @@ -5845,16 +6758,669 @@ int ring_buffer_read_page(struct trace_buffer *buffer, /* * This page may be off to user land. Zero it out here. */ - if (commit < BUF_PAGE_SIZE) - memset(&bpage->data[commit], 0, BUF_PAGE_SIZE - commit); + if (commit < buffer->subbuf_size) + memset(&bpage->data[commit], 0, buffer->subbuf_size - commit); + + return read; +} +EXPORT_SYMBOL_GPL(ring_buffer_read_page); + +/** + * ring_buffer_read_page_data - get pointer to the data in the page. + * @page: the page to get the data from + * + * Returns pointer to the actual data in this page. + */ +void *ring_buffer_read_page_data(struct buffer_data_read_page *page) +{ + return page->data; +} +EXPORT_SYMBOL_GPL(ring_buffer_read_page_data); + +/** + * ring_buffer_subbuf_size_get - get size of the sub buffer. + * @buffer: the buffer to get the sub buffer size from + * + * Returns size of the sub buffer, in bytes. + */ +int ring_buffer_subbuf_size_get(struct trace_buffer *buffer) +{ + return buffer->subbuf_size + BUF_PAGE_HDR_SIZE; +} +EXPORT_SYMBOL_GPL(ring_buffer_subbuf_size_get); + +/** + * ring_buffer_subbuf_order_get - get order of system sub pages in one buffer page. + * @buffer: The ring_buffer to get the system sub page order from + * + * By default, one ring buffer sub page equals to one system page. This parameter + * is configurable, per ring buffer. The size of the ring buffer sub page can be + * extended, but must be an order of system page size. + * + * Returns the order of buffer sub page size, in system pages: + * 0 means the sub buffer size is 1 system page and so forth. + * In case of an error < 0 is returned. + */ +int ring_buffer_subbuf_order_get(struct trace_buffer *buffer) +{ + if (!buffer) + return -EINVAL; + + return buffer->subbuf_order; +} +EXPORT_SYMBOL_GPL(ring_buffer_subbuf_order_get); + +/** + * ring_buffer_subbuf_order_set - set the size of ring buffer sub page. + * @buffer: The ring_buffer to set the new page size. + * @order: Order of the system pages in one sub buffer page + * + * By default, one ring buffer pages equals to one system page. This API can be + * used to set new size of the ring buffer page. The size must be order of + * system page size, that's why the input parameter @order is the order of + * system pages that are allocated for one ring buffer page: + * 0 - 1 system page + * 1 - 2 system pages + * 3 - 4 system pages + * ... + * + * Returns 0 on success or < 0 in case of an error. + */ +int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order) +{ + struct ring_buffer_per_cpu *cpu_buffer; + struct buffer_page *bpage, *tmp; + int old_order, old_size; + int nr_pages; + int psize; + int err; + int cpu; + + if (!buffer || order < 0) + return -EINVAL; + + if (buffer->subbuf_order == order) + return 0; + + psize = (1 << order) * PAGE_SIZE; + if (psize <= BUF_PAGE_HDR_SIZE) + return -EINVAL; + + /* Size of a subbuf cannot be greater than the write counter */ + if (psize > RB_WRITE_MASK + 1) + return -EINVAL; + + old_order = buffer->subbuf_order; + old_size = buffer->subbuf_size; + + /* prevent another thread from changing buffer sizes */ + guard(mutex)(&buffer->mutex); + atomic_inc(&buffer->record_disabled); + + /* Make sure all commits have finished */ + synchronize_rcu(); + + buffer->subbuf_order = order; + buffer->subbuf_size = psize - BUF_PAGE_HDR_SIZE; + + /* Make sure all new buffers are allocated, before deleting the old ones */ + for_each_buffer_cpu(buffer, cpu) { + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + continue; + + cpu_buffer = buffer->buffers[cpu]; + + if (cpu_buffer->mapped) { + err = -EBUSY; + goto error; + } + + /* Update the number of pages to match the new size */ + nr_pages = old_size * buffer->buffers[cpu]->nr_pages; + nr_pages = DIV_ROUND_UP(nr_pages, buffer->subbuf_size); + + /* we need a minimum of two pages */ + if (nr_pages < 2) + nr_pages = 2; + + cpu_buffer->nr_pages_to_update = nr_pages; + + /* Include the reader page */ + nr_pages++; + + /* Allocate the new size buffer */ + INIT_LIST_HEAD(&cpu_buffer->new_pages); + if (__rb_allocate_pages(cpu_buffer, nr_pages, + &cpu_buffer->new_pages)) { + /* not enough memory for new pages */ + err = -ENOMEM; + goto error; + } + } + + for_each_buffer_cpu(buffer, cpu) { + struct buffer_data_page *old_free_data_page; + struct list_head old_pages; + unsigned long flags; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + continue; + + cpu_buffer = buffer->buffers[cpu]; + + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + + /* Clear the head bit to make the link list normal to read */ + rb_head_page_deactivate(cpu_buffer); + + /* + * Collect buffers from the cpu_buffer pages list and the + * reader_page on old_pages, so they can be freed later when not + * under a spinlock. The pages list is a linked list with no + * head, adding old_pages turns it into a regular list with + * old_pages being the head. + */ + list_add(&old_pages, cpu_buffer->pages); + list_add(&cpu_buffer->reader_page->list, &old_pages); + + /* One page was allocated for the reader page */ + cpu_buffer->reader_page = list_entry(cpu_buffer->new_pages.next, + struct buffer_page, list); + list_del_init(&cpu_buffer->reader_page->list); + + /* Install the new pages, remove the head from the list */ + cpu_buffer->pages = cpu_buffer->new_pages.next; + list_del_init(&cpu_buffer->new_pages); + cpu_buffer->cnt++; + + cpu_buffer->head_page + = list_entry(cpu_buffer->pages, struct buffer_page, list); + cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page; + + cpu_buffer->nr_pages = cpu_buffer->nr_pages_to_update; + cpu_buffer->nr_pages_to_update = 0; + + old_free_data_page = cpu_buffer->free_page; + cpu_buffer->free_page = NULL; + + rb_head_page_activate(cpu_buffer); + + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + /* Free old sub buffers */ + list_for_each_entry_safe(bpage, tmp, &old_pages, list) { + list_del_init(&bpage->list); + free_buffer_page(bpage); + } + free_pages((unsigned long)old_free_data_page, old_order); + + rb_check_pages(cpu_buffer); + } + + atomic_dec(&buffer->record_disabled); + + return 0; + +error: + buffer->subbuf_order = old_order; + buffer->subbuf_size = old_size; + + atomic_dec(&buffer->record_disabled); + + for_each_buffer_cpu(buffer, cpu) { + cpu_buffer = buffer->buffers[cpu]; + + if (!cpu_buffer->nr_pages_to_update) + continue; + + list_for_each_entry_safe(bpage, tmp, &cpu_buffer->new_pages, list) { + list_del_init(&bpage->list); + free_buffer_page(bpage); + } + } + + return err; +} +EXPORT_SYMBOL_GPL(ring_buffer_subbuf_order_set); + +static int rb_alloc_meta_page(struct ring_buffer_per_cpu *cpu_buffer) +{ + struct page *page; + + if (cpu_buffer->meta_page) + return 0; + + page = alloc_page(GFP_USER | __GFP_ZERO); + if (!page) + return -ENOMEM; + + cpu_buffer->meta_page = page_to_virt(page); + + return 0; +} + +static void rb_free_meta_page(struct ring_buffer_per_cpu *cpu_buffer) +{ + unsigned long addr = (unsigned long)cpu_buffer->meta_page; + + free_page(addr); + cpu_buffer->meta_page = NULL; +} + +static void rb_setup_ids_meta_page(struct ring_buffer_per_cpu *cpu_buffer, + unsigned long *subbuf_ids) +{ + struct trace_buffer_meta *meta = cpu_buffer->meta_page; + unsigned int nr_subbufs = cpu_buffer->nr_pages + 1; + struct buffer_page *first_subbuf, *subbuf; + int cnt = 0; + int id = 0; + + id = rb_page_id(cpu_buffer, cpu_buffer->reader_page, id); + subbuf_ids[id++] = (unsigned long)cpu_buffer->reader_page->page; + cnt++; + + first_subbuf = subbuf = rb_set_head_page(cpu_buffer); + do { + id = rb_page_id(cpu_buffer, subbuf, id); + + if (WARN_ON(id >= nr_subbufs)) + break; + + subbuf_ids[id] = (unsigned long)subbuf->page; + + rb_inc_page(&subbuf); + id++; + cnt++; + } while (subbuf != first_subbuf); + + WARN_ON(cnt != nr_subbufs); + + /* install subbuf ID to kern VA translation */ + cpu_buffer->subbuf_ids = subbuf_ids; + + meta->meta_struct_len = sizeof(*meta); + meta->nr_subbufs = nr_subbufs; + meta->subbuf_size = cpu_buffer->buffer->subbuf_size + BUF_PAGE_HDR_SIZE; + meta->meta_page_size = meta->subbuf_size; + + rb_update_meta_page(cpu_buffer); +} + +static struct ring_buffer_per_cpu * +rb_get_mapped_buffer(struct trace_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return ERR_PTR(-EINVAL); + + cpu_buffer = buffer->buffers[cpu]; + + mutex_lock(&cpu_buffer->mapping_lock); + + if (!cpu_buffer->user_mapped) { + mutex_unlock(&cpu_buffer->mapping_lock); + return ERR_PTR(-ENODEV); + } + + return cpu_buffer; +} + +static void rb_put_mapped_buffer(struct ring_buffer_per_cpu *cpu_buffer) +{ + mutex_unlock(&cpu_buffer->mapping_lock); +} + +/* + * Fast-path for rb_buffer_(un)map(). Called whenever the meta-page doesn't need + * to be set-up or torn-down. + */ +static int __rb_inc_dec_mapped(struct ring_buffer_per_cpu *cpu_buffer, + bool inc) +{ + unsigned long flags; + + lockdep_assert_held(&cpu_buffer->mapping_lock); + + /* mapped is always greater or equal to user_mapped */ + if (WARN_ON(cpu_buffer->mapped < cpu_buffer->user_mapped)) + return -EINVAL; + + if (inc && cpu_buffer->mapped == UINT_MAX) + return -EBUSY; + + if (WARN_ON(!inc && cpu_buffer->user_mapped == 0)) + return -EINVAL; + + mutex_lock(&cpu_buffer->buffer->mutex); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + + if (inc) { + cpu_buffer->user_mapped++; + cpu_buffer->mapped++; + } else { + cpu_buffer->user_mapped--; + cpu_buffer->mapped--; + } - out_unlock: raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + mutex_unlock(&cpu_buffer->buffer->mutex); - out: - return ret; + return 0; +} + +/* + * +--------------+ pgoff == 0 + * | meta page | + * +--------------+ pgoff == 1 + * | subbuffer 0 | + * | | + * +--------------+ pgoff == (1 + (1 << subbuf_order)) + * | subbuffer 1 | + * | | + * ... + */ +#ifdef CONFIG_MMU +static int __rb_map_vma(struct ring_buffer_per_cpu *cpu_buffer, + struct vm_area_struct *vma) +{ + unsigned long nr_subbufs, nr_pages, nr_vma_pages, pgoff = vma->vm_pgoff; + unsigned int subbuf_pages, subbuf_order; + struct page **pages __free(kfree) = NULL; + int p = 0, s = 0; + int err; + + /* Refuse MP_PRIVATE or writable mappings */ + if (vma->vm_flags & VM_WRITE || vma->vm_flags & VM_EXEC || + !(vma->vm_flags & VM_MAYSHARE)) + return -EPERM; + + subbuf_order = cpu_buffer->buffer->subbuf_order; + subbuf_pages = 1 << subbuf_order; + + if (subbuf_order && pgoff % subbuf_pages) + return -EINVAL; + + /* + * Make sure the mapping cannot become writable later. Also tell the VM + * to not touch these pages (VM_DONTCOPY | VM_DONTEXPAND). + */ + vm_flags_mod(vma, VM_DONTCOPY | VM_DONTEXPAND | VM_DONTDUMP, + VM_MAYWRITE); + + lockdep_assert_held(&cpu_buffer->mapping_lock); + + nr_subbufs = cpu_buffer->nr_pages + 1; /* + reader-subbuf */ + nr_pages = ((nr_subbufs + 1) << subbuf_order); /* + meta-page */ + if (nr_pages <= pgoff) + return -EINVAL; + + nr_pages -= pgoff; + + nr_vma_pages = vma_pages(vma); + if (!nr_vma_pages || nr_vma_pages > nr_pages) + return -EINVAL; + + nr_pages = nr_vma_pages; + + pages = kcalloc(nr_pages, sizeof(*pages), GFP_KERNEL); + if (!pages) + return -ENOMEM; + + if (!pgoff) { + unsigned long meta_page_padding; + + pages[p++] = virt_to_page(cpu_buffer->meta_page); + + /* + * Pad with the zero-page to align the meta-page with the + * sub-buffers. + */ + meta_page_padding = subbuf_pages - 1; + while (meta_page_padding-- && p < nr_pages) { + unsigned long __maybe_unused zero_addr = + vma->vm_start + (PAGE_SIZE * p); + + pages[p++] = ZERO_PAGE(zero_addr); + } + } else { + /* Skip the meta-page */ + pgoff -= subbuf_pages; + + s += pgoff / subbuf_pages; + } + + while (p < nr_pages) { + struct page *page; + int off = 0; + + if (WARN_ON_ONCE(s >= nr_subbufs)) + return -EINVAL; + + page = virt_to_page((void *)cpu_buffer->subbuf_ids[s]); + + for (; off < (1 << (subbuf_order)); off++, page++) { + if (p >= nr_pages) + break; + + pages[p++] = page; + } + s++; + } + + err = vm_insert_pages(vma, vma->vm_start, pages, &nr_pages); + + return err; +} +#else +static int __rb_map_vma(struct ring_buffer_per_cpu *cpu_buffer, + struct vm_area_struct *vma) +{ + return -EOPNOTSUPP; +} +#endif + +int ring_buffer_map(struct trace_buffer *buffer, int cpu, + struct vm_area_struct *vma) +{ + struct ring_buffer_per_cpu *cpu_buffer; + unsigned long flags, *subbuf_ids; + int err; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return -EINVAL; + + cpu_buffer = buffer->buffers[cpu]; + + guard(mutex)(&cpu_buffer->mapping_lock); + + if (cpu_buffer->user_mapped) { + err = __rb_map_vma(cpu_buffer, vma); + if (!err) + err = __rb_inc_dec_mapped(cpu_buffer, true); + return err; + } + + /* prevent another thread from changing buffer/sub-buffer sizes */ + guard(mutex)(&buffer->mutex); + + err = rb_alloc_meta_page(cpu_buffer); + if (err) + return err; + + /* subbuf_ids include the reader while nr_pages does not */ + subbuf_ids = kcalloc(cpu_buffer->nr_pages + 1, sizeof(*subbuf_ids), GFP_KERNEL); + if (!subbuf_ids) { + rb_free_meta_page(cpu_buffer); + return -ENOMEM; + } + + atomic_inc(&cpu_buffer->resize_disabled); + + /* + * Lock all readers to block any subbuf swap until the subbuf IDs are + * assigned. + */ + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + rb_setup_ids_meta_page(cpu_buffer, subbuf_ids); + + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + err = __rb_map_vma(cpu_buffer, vma); + if (!err) { + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + /* This is the first time it is mapped by user */ + cpu_buffer->mapped++; + cpu_buffer->user_mapped = 1; + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + } else { + kfree(cpu_buffer->subbuf_ids); + cpu_buffer->subbuf_ids = NULL; + rb_free_meta_page(cpu_buffer); + atomic_dec(&cpu_buffer->resize_disabled); + } + + return err; +} + +int ring_buffer_unmap(struct trace_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + unsigned long flags; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return -EINVAL; + + cpu_buffer = buffer->buffers[cpu]; + + guard(mutex)(&cpu_buffer->mapping_lock); + + if (!cpu_buffer->user_mapped) { + return -ENODEV; + } else if (cpu_buffer->user_mapped > 1) { + __rb_inc_dec_mapped(cpu_buffer, false); + return 0; + } + + guard(mutex)(&buffer->mutex); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + + /* This is the last user space mapping */ + if (!WARN_ON_ONCE(cpu_buffer->mapped < cpu_buffer->user_mapped)) + cpu_buffer->mapped--; + cpu_buffer->user_mapped = 0; + + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + kfree(cpu_buffer->subbuf_ids); + cpu_buffer->subbuf_ids = NULL; + rb_free_meta_page(cpu_buffer); + atomic_dec(&cpu_buffer->resize_disabled); + + return 0; +} + +int ring_buffer_map_get_reader(struct trace_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + struct buffer_page *reader; + unsigned long missed_events; + unsigned long reader_size; + unsigned long flags; + + cpu_buffer = rb_get_mapped_buffer(buffer, cpu); + if (IS_ERR(cpu_buffer)) + return (int)PTR_ERR(cpu_buffer); + + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + +consume: + if (rb_per_cpu_empty(cpu_buffer)) + goto out; + + reader_size = rb_page_size(cpu_buffer->reader_page); + + /* + * There are data to be read on the current reader page, we can + * return to the caller. But before that, we assume the latter will read + * everything. Let's update the kernel reader accordingly. + */ + if (cpu_buffer->reader_page->read < reader_size) { + while (cpu_buffer->reader_page->read < reader_size) + rb_advance_reader(cpu_buffer); + goto out; + } + + /* Did the reader catch up with the writer? */ + if (cpu_buffer->reader_page == cpu_buffer->commit_page) + goto out; + + reader = rb_get_reader_page(cpu_buffer); + if (WARN_ON(!reader)) + goto out; + + /* Check if any events were dropped */ + missed_events = cpu_buffer->lost_events; + + if (missed_events) { + if (cpu_buffer->reader_page != cpu_buffer->commit_page) { + struct buffer_data_page *bpage = reader->page; + unsigned int commit; + /* + * Use the real_end for the data size, + * This gives us a chance to store the lost events + * on the page. + */ + if (reader->real_end) + local_set(&bpage->commit, reader->real_end); + /* + * If there is room at the end of the page to save the + * missed events, then record it there. + */ + commit = rb_page_size(reader); + if (buffer->subbuf_size - commit >= sizeof(missed_events)) { + memcpy(&bpage->data[commit], &missed_events, + sizeof(missed_events)); + local_add(RB_MISSED_STORED, &bpage->commit); + } + local_add(RB_MISSED_EVENTS, &bpage->commit); + } else if (!WARN_ONCE(cpu_buffer->reader_page == cpu_buffer->tail_page, + "Reader on commit with %ld missed events", + missed_events)) { + /* + * There shouldn't be any missed events if the tail_page + * is on the reader page. But if the tail page is not on the + * reader page and the commit_page is, that would mean that + * there's a commit_overrun (an interrupt preempted an + * addition of an event and then filled the buffer + * with new events). In this case it's not an + * error, but it should still be reported. + * + * TODO: Add missed events to the page for user space to know. + */ + pr_info("Ring buffer [%d] commit overrun lost %ld events at timestamp:%lld\n", + cpu, missed_events, cpu_buffer->reader_page->page->time_stamp); + } + } + + cpu_buffer->lost_events = 0; + + goto consume; + +out: + /* Some archs do not have data cache coherency between kernel and user-space */ + flush_kernel_vmap_range(cpu_buffer->reader_page->page, + buffer->subbuf_size + BUF_PAGE_HDR_SIZE); + + rb_update_meta_page(cpu_buffer); + + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + rb_put_mapped_buffer(cpu_buffer); + + return 0; } -EXPORT_SYMBOL_GPL(ring_buffer_read_page); /* * We only allocate new buffers, never free them if the CPU goes down. @@ -5979,9 +7545,9 @@ static __init int rb_write_something(struct rb_test_data *data, bool nested) /* Ignore dropped events before test starts. */ if (started) { if (nested) - data->bytes_dropped += len; - else data->bytes_dropped_nested += len; + else + data->bytes_dropped += len; } return len; } @@ -6103,7 +7669,7 @@ static __init int test_ringbuffer(void) /* * Show buffer is enabled before setting rb_test_started. * Yes there's a small race window where events could be - * dropped and the thread wont catch it. But when a ring + * dropped and the thread won't catch it. But when a ring * buffer gets enabled, there will always be some kind of * delay before other CPUs see it. Thus, we don't care about * those dropped events. We care about events dropped after @@ -6113,7 +7679,7 @@ static __init int test_ringbuffer(void) rb_test_started = true; set_current_state(TASK_INTERRUPTIBLE); - /* Just run for 10 seconds */; + /* Just run for 10 seconds */ schedule_timeout(10 * HZ); kthread_stop(rb_hammer); diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c index aef34673d79d..593e3b59e42e 100644 --- a/kernel/trace/ring_buffer_benchmark.c +++ b/kernel/trace/ring_buffer_benchmark.c @@ -104,10 +104,11 @@ static enum event_status read_event(int cpu) static enum event_status read_page(int cpu) { + struct buffer_data_read_page *bpage; struct ring_buffer_event *event; struct rb_page *rpage; unsigned long commit; - void *bpage; + int page_size; int *entry; int ret; int inc; @@ -117,14 +118,15 @@ static enum event_status read_page(int cpu) if (IS_ERR(bpage)) return EVENT_DROPPED; - ret = ring_buffer_read_page(buffer, &bpage, PAGE_SIZE, cpu, 1); + page_size = ring_buffer_subbuf_size_get(buffer); + ret = ring_buffer_read_page(buffer, bpage, page_size, cpu, 1); if (ret >= 0) { - rpage = bpage; + rpage = ring_buffer_read_page_data(bpage); /* The commit may have missed event flags set, clear them */ commit = local_read(&rpage->commit) & 0xfffff; for (i = 0; i < commit && !test_error ; i += inc) { - if (i >= (PAGE_SIZE - offsetof(struct rb_page, data))) { + if (i >= (page_size - offsetof(struct rb_page, data))) { TEST_ERROR(); break; } @@ -305,14 +307,14 @@ static void ring_buffer_producer(void) if (!disable_reader) { if (consumer_fifo) trace_printk("Running Consumer at SCHED_FIFO %s\n", - consumer_fifo == 1 ? "low" : "high"); + str_low_high(consumer_fifo == 1)); else trace_printk("Running Consumer at nice: %d\n", consumer_nice); } if (producer_fifo) trace_printk("Running Producer at SCHED_FIFO %s\n", - producer_fifo == 1 ? "low" : "high"); + str_low_high(producer_fifo == 1)); else trace_printk("Running Producer at nice: %d\n", producer_nice); @@ -431,7 +433,7 @@ static int __init ring_buffer_benchmark_init(void) { int ret; - /* make a one meg buffer in overwite mode */ + /* make a one meg buffer in overwrite mode */ buffer = ring_buffer_alloc(1000000, RB_FL_OVERWRITE); if (!buffer) return -ENOMEM; diff --git a/kernel/trace/rv/Kconfig b/kernel/trace/rv/Kconfig index 831779607e84..5b4be87ba59d 100644 --- a/kernel/trace/rv/Kconfig +++ b/kernel/trace/rv/Kconfig @@ -1,19 +1,31 @@ # SPDX-License-Identifier: GPL-2.0-only # -config DA_MON_EVENTS +config RV_MON_EVENTS + bool + +config RV_MON_MAINTENANCE_EVENTS bool config DA_MON_EVENTS_IMPLICIT - select DA_MON_EVENTS + select RV_MON_EVENTS + select RV_MON_MAINTENANCE_EVENTS bool config DA_MON_EVENTS_ID - select DA_MON_EVENTS + select RV_MON_EVENTS + select RV_MON_MAINTENANCE_EVENTS + bool + +config LTL_MON_EVENTS_ID + select RV_MON_EVENTS + bool + +config RV_LTL_MONITOR bool menuconfig RV bool "Runtime Verification" - depends on TRACING + select TRACING help Enable the kernel runtime verification infrastructure. RV is a lightweight (yet rigorous) method that complements classical @@ -25,30 +37,35 @@ menuconfig RV For further information, see: Documentation/trace/rv/runtime-verification.rst -config RV_MON_WIP +config RV_PER_TASK_MONITORS + int "Maximum number of per-task monitor" depends on RV - depends on PREEMPT_TRACER - select DA_MON_EVENTS_IMPLICIT - bool "wip monitor" + range 1 8 + default 2 help - Enable wip (wakeup in preemptive) sample monitor that illustrates - the usage of per-cpu monitors, and one limitation of the - preempt_disable/enable events. + This option configures the maximum number of per-task RV monitors that can run + simultaneously. - For further information, see: - Documentation/trace/rv/monitor_wip.rst +source "kernel/trace/rv/monitors/wip/Kconfig" +source "kernel/trace/rv/monitors/wwnr/Kconfig" -config RV_MON_WWNR - depends on RV - select DA_MON_EVENTS_ID - bool "wwnr monitor" - help - Enable wwnr (wakeup while not running) sample monitor, this is a - sample monitor that illustrates the usage of per-task monitor. - The model is borken on purpose: it serves to test reactors. +source "kernel/trace/rv/monitors/sched/Kconfig" +source "kernel/trace/rv/monitors/sco/Kconfig" +source "kernel/trace/rv/monitors/snroc/Kconfig" +source "kernel/trace/rv/monitors/scpd/Kconfig" +source "kernel/trace/rv/monitors/snep/Kconfig" +source "kernel/trace/rv/monitors/sts/Kconfig" +source "kernel/trace/rv/monitors/nrp/Kconfig" +source "kernel/trace/rv/monitors/sssw/Kconfig" +source "kernel/trace/rv/monitors/opid/Kconfig" +# Add new sched monitors here - For further information, see: - Documentation/trace/rv/monitor_wwnr.rst +source "kernel/trace/rv/monitors/rtapp/Kconfig" +source "kernel/trace/rv/monitors/pagefault/Kconfig" +source "kernel/trace/rv/monitors/sleep/Kconfig" +# Add new rtapp monitors here + +# Add new monitors here config RV_REACTORS bool "Runtime verification reactors" diff --git a/kernel/trace/rv/Makefile b/kernel/trace/rv/Makefile index 963d14875b45..750e4ad6fa0f 100644 --- a/kernel/trace/rv/Makefile +++ b/kernel/trace/rv/Makefile @@ -1,8 +1,23 @@ # SPDX-License-Identifier: GPL-2.0 +ccflags-y += -I $(src) # needed for trace events + obj-$(CONFIG_RV) += rv.o obj-$(CONFIG_RV_MON_WIP) += monitors/wip/wip.o obj-$(CONFIG_RV_MON_WWNR) += monitors/wwnr/wwnr.o +obj-$(CONFIG_RV_MON_SCHED) += monitors/sched/sched.o +obj-$(CONFIG_RV_MON_SCO) += monitors/sco/sco.o +obj-$(CONFIG_RV_MON_SNROC) += monitors/snroc/snroc.o +obj-$(CONFIG_RV_MON_SCPD) += monitors/scpd/scpd.o +obj-$(CONFIG_RV_MON_SNEP) += monitors/snep/snep.o +obj-$(CONFIG_RV_MON_RTAPP) += monitors/rtapp/rtapp.o +obj-$(CONFIG_RV_MON_PAGEFAULT) += monitors/pagefault/pagefault.o +obj-$(CONFIG_RV_MON_SLEEP) += monitors/sleep/sleep.o +obj-$(CONFIG_RV_MON_STS) += monitors/sts/sts.o +obj-$(CONFIG_RV_MON_NRP) += monitors/nrp/nrp.o +obj-$(CONFIG_RV_MON_SSSW) += monitors/sssw/sssw.o +obj-$(CONFIG_RV_MON_OPID) += monitors/opid/opid.o +# Add new monitors here obj-$(CONFIG_RV_REACTORS) += rv_reactors.o obj-$(CONFIG_RV_REACT_PRINTK) += reactor_printk.o obj-$(CONFIG_RV_REACT_PANIC) += reactor_panic.o diff --git a/kernel/trace/rv/monitors/nrp/Kconfig b/kernel/trace/rv/monitors/nrp/Kconfig new file mode 100644 index 000000000000..f5ec08f65535 --- /dev/null +++ b/kernel/trace/rv/monitors/nrp/Kconfig @@ -0,0 +1,16 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_NRP + depends on RV + depends on RV_MON_SCHED + default y if !ARM64 + select DA_MON_EVENTS_ID + bool "nrp monitor" + help + Monitor to ensure preemption requires need resched. + This monitor is part of the sched monitors collection. + + This monitor is unstable on arm64, say N unless you are testing it. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/nrp/nrp.c b/kernel/trace/rv/monitors/nrp/nrp.c new file mode 100644 index 000000000000..5a83b7171432 --- /dev/null +++ b/kernel/trace/rv/monitors/nrp/nrp.c @@ -0,0 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "nrp" + +#include <trace/events/irq.h> +#include <trace/events/sched.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "nrp.h" + +static struct rv_monitor rv_nrp; +DECLARE_DA_MON_PER_TASK(nrp, unsigned char); + +#ifdef CONFIG_X86_LOCAL_APIC +#include <asm/trace/irq_vectors.h> + +static void handle_vector_irq_entry(void *data, int vector) +{ + da_handle_event_nrp(current, irq_entry_nrp); +} + +static void attach_vector_irq(void) +{ + rv_attach_trace_probe("nrp", local_timer_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_IRQ_WORK)) + rv_attach_trace_probe("nrp", irq_work_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_SMP)) { + rv_attach_trace_probe("nrp", reschedule_entry, handle_vector_irq_entry); + rv_attach_trace_probe("nrp", call_function_entry, handle_vector_irq_entry); + rv_attach_trace_probe("nrp", call_function_single_entry, handle_vector_irq_entry); + } +} + +static void detach_vector_irq(void) +{ + rv_detach_trace_probe("nrp", local_timer_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_IRQ_WORK)) + rv_detach_trace_probe("nrp", irq_work_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_SMP)) { + rv_detach_trace_probe("nrp", reschedule_entry, handle_vector_irq_entry); + rv_detach_trace_probe("nrp", call_function_entry, handle_vector_irq_entry); + rv_detach_trace_probe("nrp", call_function_single_entry, handle_vector_irq_entry); + } +} + +#else +/* We assume irq_entry tracepoints are sufficient on other architectures */ +static void attach_vector_irq(void) { } +static void detach_vector_irq(void) { } +#endif + +static void handle_irq_entry(void *data, int irq, struct irqaction *action) +{ + da_handle_event_nrp(current, irq_entry_nrp); +} + +static void handle_sched_need_resched(void *data, struct task_struct *tsk, + int cpu, int tif) +{ + /* + * Although need_resched leads to both the rescheduling and preempt_irq + * states, it is safer to start the monitor always in preempt_irq, + * which may not mirror the system state but makes the monitor simpler, + */ + if (tif == TIF_NEED_RESCHED) + da_handle_start_event_nrp(tsk, sched_need_resched_nrp); +} + +static void handle_schedule_entry(void *data, bool preempt) +{ + if (preempt) + da_handle_event_nrp(current, schedule_entry_preempt_nrp); + else + da_handle_event_nrp(current, schedule_entry_nrp); +} + +static int enable_nrp(void) +{ + int retval; + + retval = da_monitor_init_nrp(); + if (retval) + return retval; + + rv_attach_trace_probe("nrp", irq_handler_entry, handle_irq_entry); + rv_attach_trace_probe("nrp", sched_set_need_resched_tp, handle_sched_need_resched); + rv_attach_trace_probe("nrp", sched_entry_tp, handle_schedule_entry); + attach_vector_irq(); + + return 0; +} + +static void disable_nrp(void) +{ + rv_nrp.enabled = 0; + + rv_detach_trace_probe("nrp", irq_handler_entry, handle_irq_entry); + rv_detach_trace_probe("nrp", sched_set_need_resched_tp, handle_sched_need_resched); + rv_detach_trace_probe("nrp", sched_entry_tp, handle_schedule_entry); + detach_vector_irq(); + + da_monitor_destroy_nrp(); +} + +static struct rv_monitor rv_nrp = { + .name = "nrp", + .description = "need resched preempts.", + .enable = enable_nrp, + .disable = disable_nrp, + .reset = da_monitor_reset_all_nrp, + .enabled = 0, +}; + +static int __init register_nrp(void) +{ + return rv_register_monitor(&rv_nrp, &rv_sched); +} + +static void __exit unregister_nrp(void) +{ + rv_unregister_monitor(&rv_nrp); +} + +module_init(register_nrp); +module_exit(unregister_nrp); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("nrp: need resched preempts."); diff --git a/kernel/trace/rv/monitors/nrp/nrp.h b/kernel/trace/rv/monitors/nrp/nrp.h new file mode 100644 index 000000000000..c9f12207cbf6 --- /dev/null +++ b/kernel/trace/rv/monitors/nrp/nrp.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of nrp automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_nrp { + preempt_irq_nrp = 0, + any_thread_running_nrp, + nested_preempt_nrp, + rescheduling_nrp, + state_max_nrp +}; + +#define INVALID_STATE state_max_nrp + +enum events_nrp { + irq_entry_nrp = 0, + sched_need_resched_nrp, + schedule_entry_nrp, + schedule_entry_preempt_nrp, + event_max_nrp +}; + +struct automaton_nrp { + char *state_names[state_max_nrp]; + char *event_names[event_max_nrp]; + unsigned char function[state_max_nrp][event_max_nrp]; + unsigned char initial_state; + bool final_states[state_max_nrp]; +}; + +static const struct automaton_nrp automaton_nrp = { + .state_names = { + "preempt_irq", + "any_thread_running", + "nested_preempt", + "rescheduling" + }, + .event_names = { + "irq_entry", + "sched_need_resched", + "schedule_entry", + "schedule_entry_preempt" + }, + .function = { + { + preempt_irq_nrp, + preempt_irq_nrp, + nested_preempt_nrp, + nested_preempt_nrp + }, + { + any_thread_running_nrp, + rescheduling_nrp, + any_thread_running_nrp, + INVALID_STATE + }, + { + nested_preempt_nrp, + preempt_irq_nrp, + any_thread_running_nrp, + any_thread_running_nrp + }, + { + preempt_irq_nrp, + rescheduling_nrp, + any_thread_running_nrp, + any_thread_running_nrp + }, + }, + .initial_state = preempt_irq_nrp, + .final_states = { 0, 1, 0, 0 }, +}; diff --git a/kernel/trace/rv/monitors/nrp/nrp_trace.h b/kernel/trace/rv/monitors/nrp/nrp_trace.h new file mode 100644 index 000000000000..2e13497de3b6 --- /dev/null +++ b/kernel/trace/rv/monitors/nrp/nrp_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_NRP +DEFINE_EVENT(event_da_monitor_id, event_nrp, + TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state), + TP_ARGS(id, state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor_id, error_nrp, + TP_PROTO(int id, char *state, char *event), + TP_ARGS(id, state, event)); +#endif /* CONFIG_RV_MON_NRP */ diff --git a/kernel/trace/rv/monitors/opid/Kconfig b/kernel/trace/rv/monitors/opid/Kconfig new file mode 100644 index 000000000000..561d32da572b --- /dev/null +++ b/kernel/trace/rv/monitors/opid/Kconfig @@ -0,0 +1,19 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_OPID + depends on RV + depends on TRACE_IRQFLAGS + depends on TRACE_PREEMPT_TOGGLE + depends on RV_MON_SCHED + default y if PREEMPT_RT + select DA_MON_EVENTS_IMPLICIT + bool "opid monitor" + help + Monitor to ensure operations like wakeup and need resched occur with + interrupts and preemption disabled or during IRQs, where preemption + may not be disabled explicitly. + + This monitor is unstable on !PREEMPT_RT, say N unless you are testing it. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/opid/opid.c b/kernel/trace/rv/monitors/opid/opid.c new file mode 100644 index 000000000000..50d64e7fb8c4 --- /dev/null +++ b/kernel/trace/rv/monitors/opid/opid.c @@ -0,0 +1,168 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "opid" + +#include <trace/events/sched.h> +#include <trace/events/irq.h> +#include <trace/events/preemptirq.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "opid.h" + +static struct rv_monitor rv_opid; +DECLARE_DA_MON_PER_CPU(opid, unsigned char); + +#ifdef CONFIG_X86_LOCAL_APIC +#include <asm/trace/irq_vectors.h> + +static void handle_vector_irq_entry(void *data, int vector) +{ + da_handle_event_opid(irq_entry_opid); +} + +static void attach_vector_irq(void) +{ + rv_attach_trace_probe("opid", local_timer_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_IRQ_WORK)) + rv_attach_trace_probe("opid", irq_work_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_SMP)) { + rv_attach_trace_probe("opid", reschedule_entry, handle_vector_irq_entry); + rv_attach_trace_probe("opid", call_function_entry, handle_vector_irq_entry); + rv_attach_trace_probe("opid", call_function_single_entry, handle_vector_irq_entry); + } +} + +static void detach_vector_irq(void) +{ + rv_detach_trace_probe("opid", local_timer_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_IRQ_WORK)) + rv_detach_trace_probe("opid", irq_work_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_SMP)) { + rv_detach_trace_probe("opid", reschedule_entry, handle_vector_irq_entry); + rv_detach_trace_probe("opid", call_function_entry, handle_vector_irq_entry); + rv_detach_trace_probe("opid", call_function_single_entry, handle_vector_irq_entry); + } +} + +#else +/* We assume irq_entry tracepoints are sufficient on other architectures */ +static void attach_vector_irq(void) { } +static void detach_vector_irq(void) { } +#endif + +static void handle_irq_disable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_opid(irq_disable_opid); +} + +static void handle_irq_enable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_opid(irq_enable_opid); +} + +static void handle_irq_entry(void *data, int irq, struct irqaction *action) +{ + da_handle_event_opid(irq_entry_opid); +} + +static void handle_preempt_disable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_opid(preempt_disable_opid); +} + +static void handle_preempt_enable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_opid(preempt_enable_opid); +} + +static void handle_sched_need_resched(void *data, struct task_struct *tsk, int cpu, int tif) +{ + /* The monitor's intitial state is not in_irq */ + if (this_cpu_read(hardirq_context)) + da_handle_event_opid(sched_need_resched_opid); + else + da_handle_start_event_opid(sched_need_resched_opid); +} + +static void handle_sched_waking(void *data, struct task_struct *p) +{ + /* The monitor's intitial state is not in_irq */ + if (this_cpu_read(hardirq_context)) + da_handle_event_opid(sched_waking_opid); + else + da_handle_start_event_opid(sched_waking_opid); +} + +static int enable_opid(void) +{ + int retval; + + retval = da_monitor_init_opid(); + if (retval) + return retval; + + rv_attach_trace_probe("opid", irq_disable, handle_irq_disable); + rv_attach_trace_probe("opid", irq_enable, handle_irq_enable); + rv_attach_trace_probe("opid", irq_handler_entry, handle_irq_entry); + rv_attach_trace_probe("opid", preempt_disable, handle_preempt_disable); + rv_attach_trace_probe("opid", preempt_enable, handle_preempt_enable); + rv_attach_trace_probe("opid", sched_set_need_resched_tp, handle_sched_need_resched); + rv_attach_trace_probe("opid", sched_waking, handle_sched_waking); + attach_vector_irq(); + + return 0; +} + +static void disable_opid(void) +{ + rv_opid.enabled = 0; + + rv_detach_trace_probe("opid", irq_disable, handle_irq_disable); + rv_detach_trace_probe("opid", irq_enable, handle_irq_enable); + rv_detach_trace_probe("opid", irq_handler_entry, handle_irq_entry); + rv_detach_trace_probe("opid", preempt_disable, handle_preempt_disable); + rv_detach_trace_probe("opid", preempt_enable, handle_preempt_enable); + rv_detach_trace_probe("opid", sched_set_need_resched_tp, handle_sched_need_resched); + rv_detach_trace_probe("opid", sched_waking, handle_sched_waking); + detach_vector_irq(); + + da_monitor_destroy_opid(); +} + +/* + * This is the monitor register section. + */ +static struct rv_monitor rv_opid = { + .name = "opid", + .description = "operations with preemption and irq disabled.", + .enable = enable_opid, + .disable = disable_opid, + .reset = da_monitor_reset_all_opid, + .enabled = 0, +}; + +static int __init register_opid(void) +{ + return rv_register_monitor(&rv_opid, &rv_sched); +} + +static void __exit unregister_opid(void) +{ + rv_unregister_monitor(&rv_opid); +} + +module_init(register_opid); +module_exit(unregister_opid); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("opid: operations with preemption and irq disabled."); diff --git a/kernel/trace/rv/monitors/opid/opid.h b/kernel/trace/rv/monitors/opid/opid.h new file mode 100644 index 000000000000..b4b8c2ff7f64 --- /dev/null +++ b/kernel/trace/rv/monitors/opid/opid.h @@ -0,0 +1,104 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of opid automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_opid { + disabled_opid = 0, + enabled_opid, + in_irq_opid, + irq_disabled_opid, + preempt_disabled_opid, + state_max_opid +}; + +#define INVALID_STATE state_max_opid + +enum events_opid { + irq_disable_opid = 0, + irq_enable_opid, + irq_entry_opid, + preempt_disable_opid, + preempt_enable_opid, + sched_need_resched_opid, + sched_waking_opid, + event_max_opid +}; + +struct automaton_opid { + char *state_names[state_max_opid]; + char *event_names[event_max_opid]; + unsigned char function[state_max_opid][event_max_opid]; + unsigned char initial_state; + bool final_states[state_max_opid]; +}; + +static const struct automaton_opid automaton_opid = { + .state_names = { + "disabled", + "enabled", + "in_irq", + "irq_disabled", + "preempt_disabled" + }, + .event_names = { + "irq_disable", + "irq_enable", + "irq_entry", + "preempt_disable", + "preempt_enable", + "sched_need_resched", + "sched_waking" + }, + .function = { + { + INVALID_STATE, + preempt_disabled_opid, + disabled_opid, + INVALID_STATE, + irq_disabled_opid, + disabled_opid, + disabled_opid + }, + { + irq_disabled_opid, + INVALID_STATE, + INVALID_STATE, + preempt_disabled_opid, + enabled_opid, + INVALID_STATE, + INVALID_STATE + }, + { + INVALID_STATE, + enabled_opid, + in_irq_opid, + INVALID_STATE, + INVALID_STATE, + in_irq_opid, + in_irq_opid + }, + { + INVALID_STATE, + enabled_opid, + in_irq_opid, + disabled_opid, + INVALID_STATE, + irq_disabled_opid, + INVALID_STATE + }, + { + disabled_opid, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + enabled_opid, + INVALID_STATE, + INVALID_STATE + }, + }, + .initial_state = disabled_opid, + .final_states = { 0, 1, 0, 0, 0 }, +}; diff --git a/kernel/trace/rv/monitors/opid/opid_trace.h b/kernel/trace/rv/monitors/opid/opid_trace.h new file mode 100644 index 000000000000..3df6ff955c30 --- /dev/null +++ b/kernel/trace/rv/monitors/opid/opid_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_OPID +DEFINE_EVENT(event_da_monitor, event_opid, + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + TP_ARGS(state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor, error_opid, + TP_PROTO(char *state, char *event), + TP_ARGS(state, event)); +#endif /* CONFIG_RV_MON_OPID */ diff --git a/kernel/trace/rv/monitors/pagefault/Kconfig b/kernel/trace/rv/monitors/pagefault/Kconfig new file mode 100644 index 000000000000..0e013f00c33b --- /dev/null +++ b/kernel/trace/rv/monitors/pagefault/Kconfig @@ -0,0 +1,21 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_PAGEFAULT + depends on RV + select RV_LTL_MONITOR + depends on RV_MON_RTAPP + depends on X86 || RISCV + depends on MMU + default y + select LTL_MON_EVENTS_ID + bool "pagefault monitor" + help + Monitor that real-time tasks do not raise page faults, causing + undesirable latency. + + If you are developing a real-time system and not entirely sure whether + the applications are designed correctly for real-time, you want to say + Y here. + + This monitor does not affect execution speed while it is not running, + therefore it is safe to enable this in production kernel. diff --git a/kernel/trace/rv/monitors/pagefault/pagefault.c b/kernel/trace/rv/monitors/pagefault/pagefault.c new file mode 100644 index 000000000000..9fe6123b2200 --- /dev/null +++ b/kernel/trace/rv/monitors/pagefault/pagefault.c @@ -0,0 +1,88 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/rv.h> +#include <linux/sched/deadline.h> +#include <linux/sched/rt.h> +#include <linux/tracepoint.h> +#include <rv/instrumentation.h> + +#define MODULE_NAME "pagefault" + +#include <rv_trace.h> +#include <trace/events/exceptions.h> +#include <monitors/rtapp/rtapp.h> + +#include "pagefault.h" +#include <rv/ltl_monitor.h> + +static void ltl_atoms_fetch(struct task_struct *task, struct ltl_monitor *mon) +{ + /* + * This includes "actual" real-time tasks and also PI-boosted + * tasks. A task being PI-boosted means it is blocking an "actual" + * real-task, therefore it should also obey the monitor's rule, + * otherwise the "actual" real-task may be delayed. + */ + ltl_atom_set(mon, LTL_RT, rt_or_dl_task(task)); +} + +static void ltl_atoms_init(struct task_struct *task, struct ltl_monitor *mon, bool task_creation) +{ + if (task_creation) + ltl_atom_set(mon, LTL_PAGEFAULT, false); +} + +static void handle_page_fault(void *data, unsigned long address, struct pt_regs *regs, + unsigned long error_code) +{ + ltl_atom_pulse(current, LTL_PAGEFAULT, true); +} + +static int enable_pagefault(void) +{ + int retval; + + retval = ltl_monitor_init(); + if (retval) + return retval; + + rv_attach_trace_probe("rtapp_pagefault", page_fault_kernel, handle_page_fault); + rv_attach_trace_probe("rtapp_pagefault", page_fault_user, handle_page_fault); + + return 0; +} + +static void disable_pagefault(void) +{ + rv_detach_trace_probe("rtapp_pagefault", page_fault_kernel, handle_page_fault); + rv_detach_trace_probe("rtapp_pagefault", page_fault_user, handle_page_fault); + + ltl_monitor_destroy(); +} + +static struct rv_monitor rv_pagefault = { + .name = "pagefault", + .description = "Monitor that RT tasks do not raise page faults", + .enable = enable_pagefault, + .disable = disable_pagefault, +}; + +static int __init register_pagefault(void) +{ + return rv_register_monitor(&rv_pagefault, &rv_rtapp); +} + +static void __exit unregister_pagefault(void) +{ + rv_unregister_monitor(&rv_pagefault); +} + +module_init(register_pagefault); +module_exit(unregister_pagefault); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nam Cao <namcao@linutronix.de>"); +MODULE_DESCRIPTION("pagefault: Monitor that RT tasks do not raise page faults"); diff --git a/kernel/trace/rv/monitors/pagefault/pagefault.h b/kernel/trace/rv/monitors/pagefault/pagefault.h new file mode 100644 index 000000000000..c580ec194009 --- /dev/null +++ b/kernel/trace/rv/monitors/pagefault/pagefault.h @@ -0,0 +1,64 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * C implementation of Buchi automaton, automatically generated by + * tools/verification/rvgen from the linear temporal logic specification. + * For further information, see kernel documentation: + * Documentation/trace/rv/linear_temporal_logic.rst + */ + +#include <linux/rv.h> + +#define MONITOR_NAME pagefault + +enum ltl_atom { + LTL_PAGEFAULT, + LTL_RT, + LTL_NUM_ATOM +}; +static_assert(LTL_NUM_ATOM <= RV_MAX_LTL_ATOM); + +static const char *ltl_atom_str(enum ltl_atom atom) +{ + static const char *const names[] = { + "pa", + "rt", + }; + + return names[atom]; +} + +enum ltl_buchi_state { + S0, + RV_NUM_BA_STATES +}; +static_assert(RV_NUM_BA_STATES <= RV_MAX_BA_STATES); + +static void ltl_start(struct task_struct *task, struct ltl_monitor *mon) +{ + bool pagefault = test_bit(LTL_PAGEFAULT, mon->atoms); + bool val3 = !pagefault; + bool rt = test_bit(LTL_RT, mon->atoms); + bool val1 = !rt; + bool val4 = val1 || val3; + + if (val4) + __set_bit(S0, mon->states); +} + +static void +ltl_possible_next_states(struct ltl_monitor *mon, unsigned int state, unsigned long *next) +{ + bool pagefault = test_bit(LTL_PAGEFAULT, mon->atoms); + bool val3 = !pagefault; + bool rt = test_bit(LTL_RT, mon->atoms); + bool val1 = !rt; + bool val4 = val1 || val3; + + switch (state) { + case S0: + if (val4) + __set_bit(S0, next); + break; + } +} diff --git a/kernel/trace/rv/monitors/pagefault/pagefault_trace.h b/kernel/trace/rv/monitors/pagefault/pagefault_trace.h new file mode 100644 index 000000000000..fe1f82597b1a --- /dev/null +++ b/kernel/trace/rv/monitors/pagefault/pagefault_trace.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_PAGEFAULT +DEFINE_EVENT(event_ltl_monitor_id, event_pagefault, + TP_PROTO(struct task_struct *task, char *states, char *atoms, char *next), + TP_ARGS(task, states, atoms, next)); +DEFINE_EVENT(error_ltl_monitor_id, error_pagefault, + TP_PROTO(struct task_struct *task), + TP_ARGS(task)); +#endif /* CONFIG_RV_MON_PAGEFAULT */ diff --git a/kernel/trace/rv/monitors/rtapp/Kconfig b/kernel/trace/rv/monitors/rtapp/Kconfig new file mode 100644 index 000000000000..1ce9370a9ba8 --- /dev/null +++ b/kernel/trace/rv/monitors/rtapp/Kconfig @@ -0,0 +1,11 @@ +config RV_MON_RTAPP + depends on RV + depends on RV_PER_TASK_MONITORS >= 2 + bool "rtapp monitor" + help + Collection of monitors to check for common problems with real-time + application that may cause unexpected latency. + + If you are developing a real-time system and not entirely sure whether + the applications are designed correctly for real-time, you want to say + Y here. diff --git a/kernel/trace/rv/monitors/rtapp/rtapp.c b/kernel/trace/rv/monitors/rtapp/rtapp.c new file mode 100644 index 000000000000..fd75fc927d65 --- /dev/null +++ b/kernel/trace/rv/monitors/rtapp/rtapp.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> + +#define MODULE_NAME "rtapp" + +#include "rtapp.h" + +struct rv_monitor rv_rtapp; + +struct rv_monitor rv_rtapp = { + .name = "rtapp", + .description = "Collection of monitors for detecting problems with real-time applications", +}; + +static int __init register_rtapp(void) +{ + return rv_register_monitor(&rv_rtapp, NULL); +} + +static void __exit unregister_rtapp(void) +{ + rv_unregister_monitor(&rv_rtapp); +} + +module_init(register_rtapp); +module_exit(unregister_rtapp); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nam Cao <namcao@linutronix.de>"); +MODULE_DESCRIPTION("Collection of monitors for detecting problems with real-time applications"); diff --git a/kernel/trace/rv/monitors/rtapp/rtapp.h b/kernel/trace/rv/monitors/rtapp/rtapp.h new file mode 100644 index 000000000000..4c200d67c7f6 --- /dev/null +++ b/kernel/trace/rv/monitors/rtapp/rtapp.h @@ -0,0 +1,3 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +extern struct rv_monitor rv_rtapp; diff --git a/kernel/trace/rv/monitors/sched/Kconfig b/kernel/trace/rv/monitors/sched/Kconfig new file mode 100644 index 000000000000..aa16456da864 --- /dev/null +++ b/kernel/trace/rv/monitors/sched/Kconfig @@ -0,0 +1,12 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SCHED + depends on RV + depends on RV_PER_TASK_MONITORS >= 3 + bool "sched monitor" + help + Collection of monitors to check the scheduler behaves according to specifications. + Enable this to enable all scheduler specification supported by the current kernel. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/sched/sched.c b/kernel/trace/rv/monitors/sched/sched.c new file mode 100644 index 000000000000..d04db4b543f9 --- /dev/null +++ b/kernel/trace/rv/monitors/sched/sched.c @@ -0,0 +1,37 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> + +#define MODULE_NAME "sched" + +#include "sched.h" + +struct rv_monitor rv_sched; + +struct rv_monitor rv_sched = { + .name = "sched", + .description = "container for several scheduler monitor specifications.", + .enable = NULL, + .disable = NULL, + .reset = NULL, + .enabled = 0, +}; + +static int __init register_sched(void) +{ + return rv_register_monitor(&rv_sched, NULL); +} + +static void __exit unregister_sched(void) +{ + rv_unregister_monitor(&rv_sched); +} + +module_init(register_sched); +module_exit(unregister_sched); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("sched: container for several scheduler monitor specifications."); diff --git a/kernel/trace/rv/monitors/sched/sched.h b/kernel/trace/rv/monitors/sched/sched.h new file mode 100644 index 000000000000..ba148dd8d48b --- /dev/null +++ b/kernel/trace/rv/monitors/sched/sched.h @@ -0,0 +1,3 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +extern struct rv_monitor rv_sched; diff --git a/kernel/trace/rv/monitors/sco/Kconfig b/kernel/trace/rv/monitors/sco/Kconfig new file mode 100644 index 000000000000..097c96cccdd7 --- /dev/null +++ b/kernel/trace/rv/monitors/sco/Kconfig @@ -0,0 +1,14 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SCO + depends on RV + depends on RV_MON_SCHED + default y + select DA_MON_EVENTS_IMPLICIT + bool "sco monitor" + help + Monitor to ensure sched_set_state happens only in thread context. + This monitor is part of the sched monitors collection. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/sco/sco.c b/kernel/trace/rv/monitors/sco/sco.c new file mode 100644 index 000000000000..04c36405e2e3 --- /dev/null +++ b/kernel/trace/rv/monitors/sco/sco.c @@ -0,0 +1,87 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "sco" + +#include <trace/events/sched.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "sco.h" + +static struct rv_monitor rv_sco; +DECLARE_DA_MON_PER_CPU(sco, unsigned char); + +static void handle_sched_set_state(void *data, struct task_struct *tsk, int state) +{ + da_handle_start_event_sco(sched_set_state_sco); +} + +static void handle_schedule_entry(void *data, bool preempt) +{ + da_handle_event_sco(schedule_entry_sco); +} + +static void handle_schedule_exit(void *data, bool is_switch) +{ + da_handle_start_event_sco(schedule_exit_sco); +} + +static int enable_sco(void) +{ + int retval; + + retval = da_monitor_init_sco(); + if (retval) + return retval; + + rv_attach_trace_probe("sco", sched_set_state_tp, handle_sched_set_state); + rv_attach_trace_probe("sco", sched_entry_tp, handle_schedule_entry); + rv_attach_trace_probe("sco", sched_exit_tp, handle_schedule_exit); + + return 0; +} + +static void disable_sco(void) +{ + rv_sco.enabled = 0; + + rv_detach_trace_probe("sco", sched_set_state_tp, handle_sched_set_state); + rv_detach_trace_probe("sco", sched_entry_tp, handle_schedule_entry); + rv_detach_trace_probe("sco", sched_exit_tp, handle_schedule_exit); + + da_monitor_destroy_sco(); +} + +static struct rv_monitor rv_sco = { + .name = "sco", + .description = "scheduling context operations.", + .enable = enable_sco, + .disable = disable_sco, + .reset = da_monitor_reset_all_sco, + .enabled = 0, +}; + +static int __init register_sco(void) +{ + return rv_register_monitor(&rv_sco, &rv_sched); +} + +static void __exit unregister_sco(void) +{ + rv_unregister_monitor(&rv_sco); +} + +module_init(register_sco); +module_exit(unregister_sco); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("sco: scheduling context operations."); diff --git a/kernel/trace/rv/monitors/sco/sco.h b/kernel/trace/rv/monitors/sco/sco.h new file mode 100644 index 000000000000..7a4c1f2d5ca1 --- /dev/null +++ b/kernel/trace/rv/monitors/sco/sco.h @@ -0,0 +1,47 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of sco automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_sco { + thread_context_sco = 0, + scheduling_context_sco, + state_max_sco +}; + +#define INVALID_STATE state_max_sco + +enum events_sco { + sched_set_state_sco = 0, + schedule_entry_sco, + schedule_exit_sco, + event_max_sco +}; + +struct automaton_sco { + char *state_names[state_max_sco]; + char *event_names[event_max_sco]; + unsigned char function[state_max_sco][event_max_sco]; + unsigned char initial_state; + bool final_states[state_max_sco]; +}; + +static const struct automaton_sco automaton_sco = { + .state_names = { + "thread_context", + "scheduling_context" + }, + .event_names = { + "sched_set_state", + "schedule_entry", + "schedule_exit" + }, + .function = { + { thread_context_sco, scheduling_context_sco, INVALID_STATE }, + { INVALID_STATE, INVALID_STATE, thread_context_sco }, + }, + .initial_state = thread_context_sco, + .final_states = { 1, 0 }, +}; diff --git a/kernel/trace/rv/monitors/sco/sco_trace.h b/kernel/trace/rv/monitors/sco/sco_trace.h new file mode 100644 index 000000000000..b711cd9024ec --- /dev/null +++ b/kernel/trace/rv/monitors/sco/sco_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_SCO +DEFINE_EVENT(event_da_monitor, event_sco, + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + TP_ARGS(state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor, error_sco, + TP_PROTO(char *state, char *event), + TP_ARGS(state, event)); +#endif /* CONFIG_RV_MON_SCO */ diff --git a/kernel/trace/rv/monitors/scpd/Kconfig b/kernel/trace/rv/monitors/scpd/Kconfig new file mode 100644 index 000000000000..682d0416188b --- /dev/null +++ b/kernel/trace/rv/monitors/scpd/Kconfig @@ -0,0 +1,15 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SCPD + depends on RV + depends on TRACE_PREEMPT_TOGGLE + depends on RV_MON_SCHED + default y + select DA_MON_EVENTS_IMPLICIT + bool "scpd monitor" + help + Monitor to ensure schedule is called with preemption disabled. + This monitor is part of the sched monitors collection. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/scpd/scpd.c b/kernel/trace/rv/monitors/scpd/scpd.c new file mode 100644 index 000000000000..1e351ba52fee --- /dev/null +++ b/kernel/trace/rv/monitors/scpd/scpd.c @@ -0,0 +1,95 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "scpd" + +#include <trace/events/sched.h> +#include <trace/events/preemptirq.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "scpd.h" + +static struct rv_monitor rv_scpd; +DECLARE_DA_MON_PER_CPU(scpd, unsigned char); + +static void handle_preempt_disable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_scpd(preempt_disable_scpd); +} + +static void handle_preempt_enable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_start_event_scpd(preempt_enable_scpd); +} + +static void handle_schedule_entry(void *data, bool preempt) +{ + da_handle_event_scpd(schedule_entry_scpd); +} + +static void handle_schedule_exit(void *data, bool is_switch) +{ + da_handle_event_scpd(schedule_exit_scpd); +} + +static int enable_scpd(void) +{ + int retval; + + retval = da_monitor_init_scpd(); + if (retval) + return retval; + + rv_attach_trace_probe("scpd", preempt_disable, handle_preempt_disable); + rv_attach_trace_probe("scpd", preempt_enable, handle_preempt_enable); + rv_attach_trace_probe("scpd", sched_entry_tp, handle_schedule_entry); + rv_attach_trace_probe("scpd", sched_exit_tp, handle_schedule_exit); + + return 0; +} + +static void disable_scpd(void) +{ + rv_scpd.enabled = 0; + + rv_detach_trace_probe("scpd", preempt_disable, handle_preempt_disable); + rv_detach_trace_probe("scpd", preempt_enable, handle_preempt_enable); + rv_detach_trace_probe("scpd", sched_entry_tp, handle_schedule_entry); + rv_detach_trace_probe("scpd", sched_exit_tp, handle_schedule_exit); + + da_monitor_destroy_scpd(); +} + +static struct rv_monitor rv_scpd = { + .name = "scpd", + .description = "schedule called with preemption disabled.", + .enable = enable_scpd, + .disable = disable_scpd, + .reset = da_monitor_reset_all_scpd, + .enabled = 0, +}; + +static int __init register_scpd(void) +{ + return rv_register_monitor(&rv_scpd, &rv_sched); +} + +static void __exit unregister_scpd(void) +{ + rv_unregister_monitor(&rv_scpd); +} + +module_init(register_scpd); +module_exit(unregister_scpd); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("scpd: schedule called with preemption disabled."); diff --git a/kernel/trace/rv/monitors/scpd/scpd.h b/kernel/trace/rv/monitors/scpd/scpd.h new file mode 100644 index 000000000000..295f735a5811 --- /dev/null +++ b/kernel/trace/rv/monitors/scpd/scpd.h @@ -0,0 +1,49 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of scpd automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_scpd { + cant_sched_scpd = 0, + can_sched_scpd, + state_max_scpd +}; + +#define INVALID_STATE state_max_scpd + +enum events_scpd { + preempt_disable_scpd = 0, + preempt_enable_scpd, + schedule_entry_scpd, + schedule_exit_scpd, + event_max_scpd +}; + +struct automaton_scpd { + char *state_names[state_max_scpd]; + char *event_names[event_max_scpd]; + unsigned char function[state_max_scpd][event_max_scpd]; + unsigned char initial_state; + bool final_states[state_max_scpd]; +}; + +static const struct automaton_scpd automaton_scpd = { + .state_names = { + "cant_sched", + "can_sched" + }, + .event_names = { + "preempt_disable", + "preempt_enable", + "schedule_entry", + "schedule_exit" + }, + .function = { + { can_sched_scpd, INVALID_STATE, INVALID_STATE, INVALID_STATE }, + { INVALID_STATE, cant_sched_scpd, can_sched_scpd, can_sched_scpd }, + }, + .initial_state = cant_sched_scpd, + .final_states = { 1, 0 }, +}; diff --git a/kernel/trace/rv/monitors/scpd/scpd_trace.h b/kernel/trace/rv/monitors/scpd/scpd_trace.h new file mode 100644 index 000000000000..6b0f4aa4732e --- /dev/null +++ b/kernel/trace/rv/monitors/scpd/scpd_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_SCPD +DEFINE_EVENT(event_da_monitor, event_scpd, + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + TP_ARGS(state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor, error_scpd, + TP_PROTO(char *state, char *event), + TP_ARGS(state, event)); +#endif /* CONFIG_RV_MON_SCPD */ diff --git a/kernel/trace/rv/monitors/sleep/Kconfig b/kernel/trace/rv/monitors/sleep/Kconfig new file mode 100644 index 000000000000..6b7a122e7b47 --- /dev/null +++ b/kernel/trace/rv/monitors/sleep/Kconfig @@ -0,0 +1,22 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SLEEP + depends on RV + select RV_LTL_MONITOR + depends on HAVE_SYSCALL_TRACEPOINTS + depends on RV_MON_RTAPP + select TRACE_IRQFLAGS + default y + select LTL_MON_EVENTS_ID + bool "sleep monitor" + help + Monitor that real-time tasks do not sleep in a manner that may + cause undesirable latency. + + If you are developing a real-time system and not entirely sure whether + the applications are designed correctly for real-time, you want to say + Y here. + + Enabling this monitor may have performance impact (due to select + TRACE_IRQFLAGS). Therefore, you probably should say N for + production kernel. diff --git a/kernel/trace/rv/monitors/sleep/sleep.c b/kernel/trace/rv/monitors/sleep/sleep.c new file mode 100644 index 000000000000..c1347da69e9d --- /dev/null +++ b/kernel/trace/rv/monitors/sleep/sleep.c @@ -0,0 +1,241 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/init.h> +#include <linux/irqflags.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/rv.h> +#include <linux/sched/deadline.h> +#include <linux/sched/rt.h> +#include <rv/instrumentation.h> + +#define MODULE_NAME "sleep" + +#include <trace/events/syscalls.h> +#include <trace/events/sched.h> +#include <trace/events/lock.h> +#include <uapi/linux/futex.h> +#include <rv_trace.h> +#include <monitors/rtapp/rtapp.h> + +#include "sleep.h" +#include <rv/ltl_monitor.h> + +static void ltl_atoms_fetch(struct task_struct *task, struct ltl_monitor *mon) +{ + /* + * This includes "actual" real-time tasks and also PI-boosted + * tasks. A task being PI-boosted means it is blocking an "actual" + * real-task, therefore it should also obey the monitor's rule, + * otherwise the "actual" real-task may be delayed. + */ + ltl_atom_set(mon, LTL_RT, rt_or_dl_task(task)); +} + +static void ltl_atoms_init(struct task_struct *task, struct ltl_monitor *mon, bool task_creation) +{ + ltl_atom_set(mon, LTL_SLEEP, false); + ltl_atom_set(mon, LTL_WAKE, false); + ltl_atom_set(mon, LTL_ABORT_SLEEP, false); + ltl_atom_set(mon, LTL_WOKEN_BY_HARDIRQ, false); + ltl_atom_set(mon, LTL_WOKEN_BY_NMI, false); + ltl_atom_set(mon, LTL_WOKEN_BY_EQUAL_OR_HIGHER_PRIO, false); + + if (task_creation) { + ltl_atom_set(mon, LTL_KTHREAD_SHOULD_STOP, false); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_MONOTONIC, false); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_TAI, false); + ltl_atom_set(mon, LTL_NANOSLEEP_TIMER_ABSTIME, false); + ltl_atom_set(mon, LTL_CLOCK_NANOSLEEP, false); + ltl_atom_set(mon, LTL_FUTEX_WAIT, false); + ltl_atom_set(mon, LTL_FUTEX_LOCK_PI, false); + ltl_atom_set(mon, LTL_BLOCK_ON_RT_MUTEX, false); + } + + if (task->flags & PF_KTHREAD) { + ltl_atom_set(mon, LTL_KERNEL_THREAD, true); + + /* kernel tasks do not do syscall */ + ltl_atom_set(mon, LTL_FUTEX_WAIT, false); + ltl_atom_set(mon, LTL_FUTEX_LOCK_PI, false); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_MONOTONIC, false); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_TAI, false); + ltl_atom_set(mon, LTL_NANOSLEEP_TIMER_ABSTIME, false); + ltl_atom_set(mon, LTL_CLOCK_NANOSLEEP, false); + + if (strstarts(task->comm, "migration/")) + ltl_atom_set(mon, LTL_TASK_IS_MIGRATION, true); + else + ltl_atom_set(mon, LTL_TASK_IS_MIGRATION, false); + + if (strstarts(task->comm, "rcu")) + ltl_atom_set(mon, LTL_TASK_IS_RCU, true); + else + ltl_atom_set(mon, LTL_TASK_IS_RCU, false); + } else { + ltl_atom_set(mon, LTL_KTHREAD_SHOULD_STOP, false); + ltl_atom_set(mon, LTL_KERNEL_THREAD, false); + ltl_atom_set(mon, LTL_TASK_IS_RCU, false); + ltl_atom_set(mon, LTL_TASK_IS_MIGRATION, false); + } + +} + +static void handle_sched_set_state(void *data, struct task_struct *task, int state) +{ + if (state & TASK_INTERRUPTIBLE) + ltl_atom_pulse(task, LTL_SLEEP, true); + else if (state == TASK_RUNNING) + ltl_atom_pulse(task, LTL_ABORT_SLEEP, true); +} + +static void handle_sched_wakeup(void *data, struct task_struct *task) +{ + ltl_atom_pulse(task, LTL_WAKE, true); +} + +static void handle_sched_waking(void *data, struct task_struct *task) +{ + if (this_cpu_read(hardirq_context)) { + ltl_atom_pulse(task, LTL_WOKEN_BY_HARDIRQ, true); + } else if (in_task()) { + if (current->prio <= task->prio) + ltl_atom_pulse(task, LTL_WOKEN_BY_EQUAL_OR_HIGHER_PRIO, true); + } else if (in_nmi()) { + ltl_atom_pulse(task, LTL_WOKEN_BY_NMI, true); + } +} + +static void handle_contention_begin(void *data, void *lock, unsigned int flags) +{ + if (flags & LCB_F_RT) + ltl_atom_update(current, LTL_BLOCK_ON_RT_MUTEX, true); +} + +static void handle_contention_end(void *data, void *lock, int ret) +{ + ltl_atom_update(current, LTL_BLOCK_ON_RT_MUTEX, false); +} + +static void handle_sys_enter(void *data, struct pt_regs *regs, long id) +{ + struct ltl_monitor *mon; + unsigned long args[6]; + int op, cmd; + + mon = ltl_get_monitor(current); + + switch (id) { +#ifdef __NR_clock_nanosleep + case __NR_clock_nanosleep: +#endif +#ifdef __NR_clock_nanosleep_time64 + case __NR_clock_nanosleep_time64: +#endif + syscall_get_arguments(current, regs, args); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_MONOTONIC, args[0] == CLOCK_MONOTONIC); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_TAI, args[0] == CLOCK_TAI); + ltl_atom_set(mon, LTL_NANOSLEEP_TIMER_ABSTIME, args[1] == TIMER_ABSTIME); + ltl_atom_update(current, LTL_CLOCK_NANOSLEEP, true); + break; + +#ifdef __NR_futex + case __NR_futex: +#endif +#ifdef __NR_futex_time64 + case __NR_futex_time64: +#endif + syscall_get_arguments(current, regs, args); + op = args[1]; + cmd = op & FUTEX_CMD_MASK; + + switch (cmd) { + case FUTEX_LOCK_PI: + case FUTEX_LOCK_PI2: + ltl_atom_update(current, LTL_FUTEX_LOCK_PI, true); + break; + case FUTEX_WAIT: + case FUTEX_WAIT_BITSET: + case FUTEX_WAIT_REQUEUE_PI: + ltl_atom_update(current, LTL_FUTEX_WAIT, true); + break; + } + break; + } +} + +static void handle_sys_exit(void *data, struct pt_regs *regs, long ret) +{ + struct ltl_monitor *mon = ltl_get_monitor(current); + + ltl_atom_set(mon, LTL_FUTEX_LOCK_PI, false); + ltl_atom_set(mon, LTL_FUTEX_WAIT, false); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_MONOTONIC, false); + ltl_atom_set(mon, LTL_NANOSLEEP_CLOCK_TAI, false); + ltl_atom_set(mon, LTL_NANOSLEEP_TIMER_ABSTIME, false); + ltl_atom_update(current, LTL_CLOCK_NANOSLEEP, false); +} + +static void handle_kthread_stop(void *data, struct task_struct *task) +{ + /* FIXME: this could race with other tracepoint handlers */ + ltl_atom_update(task, LTL_KTHREAD_SHOULD_STOP, true); +} + +static int enable_sleep(void) +{ + int retval; + + retval = ltl_monitor_init(); + if (retval) + return retval; + + rv_attach_trace_probe("rtapp_sleep", sched_waking, handle_sched_waking); + rv_attach_trace_probe("rtapp_sleep", sched_wakeup, handle_sched_wakeup); + rv_attach_trace_probe("rtapp_sleep", sched_set_state_tp, handle_sched_set_state); + rv_attach_trace_probe("rtapp_sleep", contention_begin, handle_contention_begin); + rv_attach_trace_probe("rtapp_sleep", contention_end, handle_contention_end); + rv_attach_trace_probe("rtapp_sleep", sched_kthread_stop, handle_kthread_stop); + rv_attach_trace_probe("rtapp_sleep", sys_enter, handle_sys_enter); + rv_attach_trace_probe("rtapp_sleep", sys_exit, handle_sys_exit); + return 0; +} + +static void disable_sleep(void) +{ + rv_detach_trace_probe("rtapp_sleep", sched_waking, handle_sched_waking); + rv_detach_trace_probe("rtapp_sleep", sched_wakeup, handle_sched_wakeup); + rv_detach_trace_probe("rtapp_sleep", sched_set_state_tp, handle_sched_set_state); + rv_detach_trace_probe("rtapp_sleep", contention_begin, handle_contention_begin); + rv_detach_trace_probe("rtapp_sleep", contention_end, handle_contention_end); + rv_detach_trace_probe("rtapp_sleep", sched_kthread_stop, handle_kthread_stop); + rv_detach_trace_probe("rtapp_sleep", sys_enter, handle_sys_enter); + rv_detach_trace_probe("rtapp_sleep", sys_exit, handle_sys_exit); + + ltl_monitor_destroy(); +} + +static struct rv_monitor rv_sleep = { + .name = "sleep", + .description = "Monitor that RT tasks do not undesirably sleep", + .enable = enable_sleep, + .disable = disable_sleep, +}; + +static int __init register_sleep(void) +{ + return rv_register_monitor(&rv_sleep, &rv_rtapp); +} + +static void __exit unregister_sleep(void) +{ + rv_unregister_monitor(&rv_sleep); +} + +module_init(register_sleep); +module_exit(unregister_sleep); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nam Cao <namcao@linutronix.de>"); +MODULE_DESCRIPTION("sleep: Monitor that RT tasks do not undesirably sleep"); diff --git a/kernel/trace/rv/monitors/sleep/sleep.h b/kernel/trace/rv/monitors/sleep/sleep.h new file mode 100644 index 000000000000..2ab46fd218d2 --- /dev/null +++ b/kernel/trace/rv/monitors/sleep/sleep.h @@ -0,0 +1,257 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * C implementation of Buchi automaton, automatically generated by + * tools/verification/rvgen from the linear temporal logic specification. + * For further information, see kernel documentation: + * Documentation/trace/rv/linear_temporal_logic.rst + */ + +#include <linux/rv.h> + +#define MONITOR_NAME sleep + +enum ltl_atom { + LTL_ABORT_SLEEP, + LTL_BLOCK_ON_RT_MUTEX, + LTL_CLOCK_NANOSLEEP, + LTL_FUTEX_LOCK_PI, + LTL_FUTEX_WAIT, + LTL_KERNEL_THREAD, + LTL_KTHREAD_SHOULD_STOP, + LTL_NANOSLEEP_CLOCK_MONOTONIC, + LTL_NANOSLEEP_CLOCK_TAI, + LTL_NANOSLEEP_TIMER_ABSTIME, + LTL_RT, + LTL_SLEEP, + LTL_TASK_IS_MIGRATION, + LTL_TASK_IS_RCU, + LTL_WAKE, + LTL_WOKEN_BY_EQUAL_OR_HIGHER_PRIO, + LTL_WOKEN_BY_HARDIRQ, + LTL_WOKEN_BY_NMI, + LTL_NUM_ATOM +}; +static_assert(LTL_NUM_ATOM <= RV_MAX_LTL_ATOM); + +static const char *ltl_atom_str(enum ltl_atom atom) +{ + static const char *const names[] = { + "ab_sl", + "bl_on_rt_mu", + "cl_na", + "fu_lo_pi", + "fu_wa", + "ker_th", + "kth_sh_st", + "na_cl_mo", + "na_cl_ta", + "na_ti_ab", + "rt", + "sl", + "ta_mi", + "ta_rc", + "wak", + "wo_eq_hi_pr", + "wo_ha", + "wo_nm", + }; + + return names[atom]; +} + +enum ltl_buchi_state { + S0, + S1, + S2, + S3, + S4, + S5, + S6, + S7, + RV_NUM_BA_STATES +}; +static_assert(RV_NUM_BA_STATES <= RV_MAX_BA_STATES); + +static void ltl_start(struct task_struct *task, struct ltl_monitor *mon) +{ + bool task_is_migration = test_bit(LTL_TASK_IS_MIGRATION, mon->atoms); + bool task_is_rcu = test_bit(LTL_TASK_IS_RCU, mon->atoms); + bool val40 = task_is_rcu || task_is_migration; + bool futex_lock_pi = test_bit(LTL_FUTEX_LOCK_PI, mon->atoms); + bool val41 = futex_lock_pi || val40; + bool block_on_rt_mutex = test_bit(LTL_BLOCK_ON_RT_MUTEX, mon->atoms); + bool val5 = block_on_rt_mutex || val41; + bool kthread_should_stop = test_bit(LTL_KTHREAD_SHOULD_STOP, mon->atoms); + bool abort_sleep = test_bit(LTL_ABORT_SLEEP, mon->atoms); + bool val32 = abort_sleep || kthread_should_stop; + bool woken_by_nmi = test_bit(LTL_WOKEN_BY_NMI, mon->atoms); + bool val33 = woken_by_nmi || val32; + bool woken_by_hardirq = test_bit(LTL_WOKEN_BY_HARDIRQ, mon->atoms); + bool val34 = woken_by_hardirq || val33; + bool woken_by_equal_or_higher_prio = test_bit(LTL_WOKEN_BY_EQUAL_OR_HIGHER_PRIO, + mon->atoms); + bool val14 = woken_by_equal_or_higher_prio || val34; + bool wake = test_bit(LTL_WAKE, mon->atoms); + bool val13 = !wake; + bool kernel_thread = test_bit(LTL_KERNEL_THREAD, mon->atoms); + bool nanosleep_clock_tai = test_bit(LTL_NANOSLEEP_CLOCK_TAI, mon->atoms); + bool nanosleep_clock_monotonic = test_bit(LTL_NANOSLEEP_CLOCK_MONOTONIC, mon->atoms); + bool val24 = nanosleep_clock_monotonic || nanosleep_clock_tai; + bool nanosleep_timer_abstime = test_bit(LTL_NANOSLEEP_TIMER_ABSTIME, mon->atoms); + bool val25 = nanosleep_timer_abstime && val24; + bool clock_nanosleep = test_bit(LTL_CLOCK_NANOSLEEP, mon->atoms); + bool val18 = clock_nanosleep && val25; + bool futex_wait = test_bit(LTL_FUTEX_WAIT, mon->atoms); + bool val9 = futex_wait || val18; + bool val11 = val9 || kernel_thread; + bool sleep = test_bit(LTL_SLEEP, mon->atoms); + bool val2 = !sleep; + bool rt = test_bit(LTL_RT, mon->atoms); + bool val1 = !rt; + bool val3 = val1 || val2; + + if (val3) + __set_bit(S0, mon->states); + if (val11 && val13) + __set_bit(S1, mon->states); + if (val11 && val14) + __set_bit(S4, mon->states); + if (val5) + __set_bit(S5, mon->states); +} + +static void +ltl_possible_next_states(struct ltl_monitor *mon, unsigned int state, unsigned long *next) +{ + bool task_is_migration = test_bit(LTL_TASK_IS_MIGRATION, mon->atoms); + bool task_is_rcu = test_bit(LTL_TASK_IS_RCU, mon->atoms); + bool val40 = task_is_rcu || task_is_migration; + bool futex_lock_pi = test_bit(LTL_FUTEX_LOCK_PI, mon->atoms); + bool val41 = futex_lock_pi || val40; + bool block_on_rt_mutex = test_bit(LTL_BLOCK_ON_RT_MUTEX, mon->atoms); + bool val5 = block_on_rt_mutex || val41; + bool kthread_should_stop = test_bit(LTL_KTHREAD_SHOULD_STOP, mon->atoms); + bool abort_sleep = test_bit(LTL_ABORT_SLEEP, mon->atoms); + bool val32 = abort_sleep || kthread_should_stop; + bool woken_by_nmi = test_bit(LTL_WOKEN_BY_NMI, mon->atoms); + bool val33 = woken_by_nmi || val32; + bool woken_by_hardirq = test_bit(LTL_WOKEN_BY_HARDIRQ, mon->atoms); + bool val34 = woken_by_hardirq || val33; + bool woken_by_equal_or_higher_prio = test_bit(LTL_WOKEN_BY_EQUAL_OR_HIGHER_PRIO, + mon->atoms); + bool val14 = woken_by_equal_or_higher_prio || val34; + bool wake = test_bit(LTL_WAKE, mon->atoms); + bool val13 = !wake; + bool kernel_thread = test_bit(LTL_KERNEL_THREAD, mon->atoms); + bool nanosleep_clock_tai = test_bit(LTL_NANOSLEEP_CLOCK_TAI, mon->atoms); + bool nanosleep_clock_monotonic = test_bit(LTL_NANOSLEEP_CLOCK_MONOTONIC, mon->atoms); + bool val24 = nanosleep_clock_monotonic || nanosleep_clock_tai; + bool nanosleep_timer_abstime = test_bit(LTL_NANOSLEEP_TIMER_ABSTIME, mon->atoms); + bool val25 = nanosleep_timer_abstime && val24; + bool clock_nanosleep = test_bit(LTL_CLOCK_NANOSLEEP, mon->atoms); + bool val18 = clock_nanosleep && val25; + bool futex_wait = test_bit(LTL_FUTEX_WAIT, mon->atoms); + bool val9 = futex_wait || val18; + bool val11 = val9 || kernel_thread; + bool sleep = test_bit(LTL_SLEEP, mon->atoms); + bool val2 = !sleep; + bool rt = test_bit(LTL_RT, mon->atoms); + bool val1 = !rt; + bool val3 = val1 || val2; + + switch (state) { + case S0: + if (val3) + __set_bit(S0, next); + if (val11 && val13) + __set_bit(S1, next); + if (val11 && val14) + __set_bit(S4, next); + if (val5) + __set_bit(S5, next); + break; + case S1: + if (val11 && val13) + __set_bit(S1, next); + if (val13 && val3) + __set_bit(S2, next); + if (val14 && val3) + __set_bit(S3, next); + if (val11 && val14) + __set_bit(S4, next); + if (val13 && val5) + __set_bit(S6, next); + if (val14 && val5) + __set_bit(S7, next); + break; + case S2: + if (val11 && val13) + __set_bit(S1, next); + if (val13 && val3) + __set_bit(S2, next); + if (val14 && val3) + __set_bit(S3, next); + if (val11 && val14) + __set_bit(S4, next); + if (val13 && val5) + __set_bit(S6, next); + if (val14 && val5) + __set_bit(S7, next); + break; + case S3: + if (val3) + __set_bit(S0, next); + if (val11 && val13) + __set_bit(S1, next); + if (val11 && val14) + __set_bit(S4, next); + if (val5) + __set_bit(S5, next); + break; + case S4: + if (val3) + __set_bit(S0, next); + if (val11 && val13) + __set_bit(S1, next); + if (val11 && val14) + __set_bit(S4, next); + if (val5) + __set_bit(S5, next); + break; + case S5: + if (val3) + __set_bit(S0, next); + if (val11 && val13) + __set_bit(S1, next); + if (val11 && val14) + __set_bit(S4, next); + if (val5) + __set_bit(S5, next); + break; + case S6: + if (val11 && val13) + __set_bit(S1, next); + if (val13 && val3) + __set_bit(S2, next); + if (val14 && val3) + __set_bit(S3, next); + if (val11 && val14) + __set_bit(S4, next); + if (val13 && val5) + __set_bit(S6, next); + if (val14 && val5) + __set_bit(S7, next); + break; + case S7: + if (val3) + __set_bit(S0, next); + if (val11 && val13) + __set_bit(S1, next); + if (val11 && val14) + __set_bit(S4, next); + if (val5) + __set_bit(S5, next); + break; + } +} diff --git a/kernel/trace/rv/monitors/sleep/sleep_trace.h b/kernel/trace/rv/monitors/sleep/sleep_trace.h new file mode 100644 index 000000000000..22eaf31da987 --- /dev/null +++ b/kernel/trace/rv/monitors/sleep/sleep_trace.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_SLEEP +DEFINE_EVENT(event_ltl_monitor_id, event_sleep, + TP_PROTO(struct task_struct *task, char *states, char *atoms, char *next), + TP_ARGS(task, states, atoms, next)); +DEFINE_EVENT(error_ltl_monitor_id, error_sleep, + TP_PROTO(struct task_struct *task), + TP_ARGS(task)); +#endif /* CONFIG_RV_MON_SLEEP */ diff --git a/kernel/trace/rv/monitors/snep/Kconfig b/kernel/trace/rv/monitors/snep/Kconfig new file mode 100644 index 000000000000..7dd54f434ff7 --- /dev/null +++ b/kernel/trace/rv/monitors/snep/Kconfig @@ -0,0 +1,15 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SNEP + depends on RV + depends on TRACE_PREEMPT_TOGGLE + depends on RV_MON_SCHED + default y + select DA_MON_EVENTS_IMPLICIT + bool "snep monitor" + help + Monitor to ensure schedule does not enable preempt. + This monitor is part of the sched monitors collection. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/snep/snep.c b/kernel/trace/rv/monitors/snep/snep.c new file mode 100644 index 000000000000..558950f524a5 --- /dev/null +++ b/kernel/trace/rv/monitors/snep/snep.c @@ -0,0 +1,95 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "snep" + +#include <trace/events/sched.h> +#include <trace/events/preemptirq.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "snep.h" + +static struct rv_monitor rv_snep; +DECLARE_DA_MON_PER_CPU(snep, unsigned char); + +static void handle_preempt_disable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_start_event_snep(preempt_disable_snep); +} + +static void handle_preempt_enable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_start_event_snep(preempt_enable_snep); +} + +static void handle_schedule_entry(void *data, bool preempt) +{ + da_handle_event_snep(schedule_entry_snep); +} + +static void handle_schedule_exit(void *data, bool is_switch) +{ + da_handle_start_event_snep(schedule_exit_snep); +} + +static int enable_snep(void) +{ + int retval; + + retval = da_monitor_init_snep(); + if (retval) + return retval; + + rv_attach_trace_probe("snep", preempt_disable, handle_preempt_disable); + rv_attach_trace_probe("snep", preempt_enable, handle_preempt_enable); + rv_attach_trace_probe("snep", sched_entry_tp, handle_schedule_entry); + rv_attach_trace_probe("snep", sched_exit_tp, handle_schedule_exit); + + return 0; +} + +static void disable_snep(void) +{ + rv_snep.enabled = 0; + + rv_detach_trace_probe("snep", preempt_disable, handle_preempt_disable); + rv_detach_trace_probe("snep", preempt_enable, handle_preempt_enable); + rv_detach_trace_probe("snep", sched_entry_tp, handle_schedule_entry); + rv_detach_trace_probe("snep", sched_exit_tp, handle_schedule_exit); + + da_monitor_destroy_snep(); +} + +static struct rv_monitor rv_snep = { + .name = "snep", + .description = "schedule does not enable preempt.", + .enable = enable_snep, + .disable = disable_snep, + .reset = da_monitor_reset_all_snep, + .enabled = 0, +}; + +static int __init register_snep(void) +{ + return rv_register_monitor(&rv_snep, &rv_sched); +} + +static void __exit unregister_snep(void) +{ + rv_unregister_monitor(&rv_snep); +} + +module_init(register_snep); +module_exit(unregister_snep); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("snep: schedule does not enable preempt."); diff --git a/kernel/trace/rv/monitors/snep/snep.h b/kernel/trace/rv/monitors/snep/snep.h new file mode 100644 index 000000000000..4cd9abb77b7b --- /dev/null +++ b/kernel/trace/rv/monitors/snep/snep.h @@ -0,0 +1,59 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of snep automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_snep { + non_scheduling_context_snep = 0, + scheduling_contex_snep, + state_max_snep +}; + +#define INVALID_STATE state_max_snep + +enum events_snep { + preempt_disable_snep = 0, + preempt_enable_snep, + schedule_entry_snep, + schedule_exit_snep, + event_max_snep +}; + +struct automaton_snep { + char *state_names[state_max_snep]; + char *event_names[event_max_snep]; + unsigned char function[state_max_snep][event_max_snep]; + unsigned char initial_state; + bool final_states[state_max_snep]; +}; + +static const struct automaton_snep automaton_snep = { + .state_names = { + "non_scheduling_context", + "scheduling_contex" + }, + .event_names = { + "preempt_disable", + "preempt_enable", + "schedule_entry", + "schedule_exit" + }, + .function = { + { + non_scheduling_context_snep, + non_scheduling_context_snep, + scheduling_contex_snep, + INVALID_STATE + }, + { + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + non_scheduling_context_snep + }, + }, + .initial_state = non_scheduling_context_snep, + .final_states = { 1, 0 }, +}; diff --git a/kernel/trace/rv/monitors/snep/snep_trace.h b/kernel/trace/rv/monitors/snep/snep_trace.h new file mode 100644 index 000000000000..01aad49a949a --- /dev/null +++ b/kernel/trace/rv/monitors/snep/snep_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_SNEP +DEFINE_EVENT(event_da_monitor, event_snep, + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + TP_ARGS(state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor, error_snep, + TP_PROTO(char *state, char *event), + TP_ARGS(state, event)); +#endif /* CONFIG_RV_MON_SNEP */ diff --git a/kernel/trace/rv/monitors/snroc/Kconfig b/kernel/trace/rv/monitors/snroc/Kconfig new file mode 100644 index 000000000000..6e4365a2fea3 --- /dev/null +++ b/kernel/trace/rv/monitors/snroc/Kconfig @@ -0,0 +1,14 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SNROC + depends on RV + depends on RV_MON_SCHED + default y + select DA_MON_EVENTS_ID + bool "snroc monitor" + help + Monitor to ensure sched_set_state happens only in the respective task's context. + This monitor is part of the sched monitors collection. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/snroc/snroc.c b/kernel/trace/rv/monitors/snroc/snroc.c new file mode 100644 index 000000000000..540e686e699f --- /dev/null +++ b/kernel/trace/rv/monitors/snroc/snroc.c @@ -0,0 +1,84 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "snroc" + +#include <trace/events/sched.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "snroc.h" + +static struct rv_monitor rv_snroc; +DECLARE_DA_MON_PER_TASK(snroc, unsigned char); + +static void handle_sched_set_state(void *data, struct task_struct *tsk, int state) +{ + da_handle_event_snroc(tsk, sched_set_state_snroc); +} + +static void handle_sched_switch(void *data, bool preempt, + struct task_struct *prev, + struct task_struct *next, + unsigned int prev_state) +{ + da_handle_start_event_snroc(prev, sched_switch_out_snroc); + da_handle_event_snroc(next, sched_switch_in_snroc); +} + +static int enable_snroc(void) +{ + int retval; + + retval = da_monitor_init_snroc(); + if (retval) + return retval; + + rv_attach_trace_probe("snroc", sched_set_state_tp, handle_sched_set_state); + rv_attach_trace_probe("snroc", sched_switch, handle_sched_switch); + + return 0; +} + +static void disable_snroc(void) +{ + rv_snroc.enabled = 0; + + rv_detach_trace_probe("snroc", sched_set_state_tp, handle_sched_set_state); + rv_detach_trace_probe("snroc", sched_switch, handle_sched_switch); + + da_monitor_destroy_snroc(); +} + +static struct rv_monitor rv_snroc = { + .name = "snroc", + .description = "set non runnable on its own context.", + .enable = enable_snroc, + .disable = disable_snroc, + .reset = da_monitor_reset_all_snroc, + .enabled = 0, +}; + +static int __init register_snroc(void) +{ + return rv_register_monitor(&rv_snroc, &rv_sched); +} + +static void __exit unregister_snroc(void) +{ + rv_unregister_monitor(&rv_snroc); +} + +module_init(register_snroc); +module_exit(unregister_snroc); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("snroc: set non runnable on its own context."); diff --git a/kernel/trace/rv/monitors/snroc/snroc.h b/kernel/trace/rv/monitors/snroc/snroc.h new file mode 100644 index 000000000000..c3650a2b1b10 --- /dev/null +++ b/kernel/trace/rv/monitors/snroc/snroc.h @@ -0,0 +1,47 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of snroc automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_snroc { + other_context_snroc = 0, + own_context_snroc, + state_max_snroc +}; + +#define INVALID_STATE state_max_snroc + +enum events_snroc { + sched_set_state_snroc = 0, + sched_switch_in_snroc, + sched_switch_out_snroc, + event_max_snroc +}; + +struct automaton_snroc { + char *state_names[state_max_snroc]; + char *event_names[event_max_snroc]; + unsigned char function[state_max_snroc][event_max_snroc]; + unsigned char initial_state; + bool final_states[state_max_snroc]; +}; + +static const struct automaton_snroc automaton_snroc = { + .state_names = { + "other_context", + "own_context" + }, + .event_names = { + "sched_set_state", + "sched_switch_in", + "sched_switch_out" + }, + .function = { + { INVALID_STATE, own_context_snroc, INVALID_STATE }, + { own_context_snroc, INVALID_STATE, other_context_snroc }, + }, + .initial_state = other_context_snroc, + .final_states = { 1, 0 }, +}; diff --git a/kernel/trace/rv/monitors/snroc/snroc_trace.h b/kernel/trace/rv/monitors/snroc/snroc_trace.h new file mode 100644 index 000000000000..50114cef5122 --- /dev/null +++ b/kernel/trace/rv/monitors/snroc/snroc_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_SNROC +DEFINE_EVENT(event_da_monitor_id, event_snroc, + TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state), + TP_ARGS(id, state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor_id, error_snroc, + TP_PROTO(int id, char *state, char *event), + TP_ARGS(id, state, event)); +#endif /* CONFIG_RV_MON_SNROC */ diff --git a/kernel/trace/rv/monitors/sssw/Kconfig b/kernel/trace/rv/monitors/sssw/Kconfig new file mode 100644 index 000000000000..23b7eeb38bbf --- /dev/null +++ b/kernel/trace/rv/monitors/sssw/Kconfig @@ -0,0 +1,15 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_SSSW + depends on RV + depends on RV_MON_SCHED + default y + select DA_MON_EVENTS_ID + bool "sssw monitor" + help + Monitor to ensure sched_set_state to sleepable leads to sleeping and + sleeping tasks require wakeup. + This monitor is part of the sched monitors collection. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/sssw/sssw.c b/kernel/trace/rv/monitors/sssw/sssw.c new file mode 100644 index 000000000000..84b8d890d9d4 --- /dev/null +++ b/kernel/trace/rv/monitors/sssw/sssw.c @@ -0,0 +1,116 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "sssw" + +#include <trace/events/sched.h> +#include <trace/events/signal.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "sssw.h" + +static struct rv_monitor rv_sssw; +DECLARE_DA_MON_PER_TASK(sssw, unsigned char); + +static void handle_sched_set_state(void *data, struct task_struct *tsk, int state) +{ + if (state == TASK_RUNNING) + da_handle_start_event_sssw(tsk, sched_set_state_runnable_sssw); + else + da_handle_event_sssw(tsk, sched_set_state_sleepable_sssw); +} + +static void handle_sched_switch(void *data, bool preempt, + struct task_struct *prev, + struct task_struct *next, + unsigned int prev_state) +{ + if (preempt) + da_handle_event_sssw(prev, sched_switch_preempt_sssw); + else if (prev_state == TASK_RUNNING) + da_handle_event_sssw(prev, sched_switch_yield_sssw); + else if (prev_state == TASK_RTLOCK_WAIT) + /* special case of sleeping task with racy conditions */ + da_handle_event_sssw(prev, sched_switch_blocking_sssw); + else + da_handle_event_sssw(prev, sched_switch_suspend_sssw); + da_handle_event_sssw(next, sched_switch_in_sssw); +} + +static void handle_sched_wakeup(void *data, struct task_struct *p) +{ + /* + * Wakeup can also lead to signal_wakeup although the system is + * actually runnable. The monitor can safely start with this event. + */ + da_handle_start_event_sssw(p, sched_wakeup_sssw); +} + +static void handle_signal_deliver(void *data, int sig, + struct kernel_siginfo *info, + struct k_sigaction *ka) +{ + da_handle_event_sssw(current, signal_deliver_sssw); +} + +static int enable_sssw(void) +{ + int retval; + + retval = da_monitor_init_sssw(); + if (retval) + return retval; + + rv_attach_trace_probe("sssw", sched_set_state_tp, handle_sched_set_state); + rv_attach_trace_probe("sssw", sched_switch, handle_sched_switch); + rv_attach_trace_probe("sssw", sched_wakeup, handle_sched_wakeup); + rv_attach_trace_probe("sssw", signal_deliver, handle_signal_deliver); + + return 0; +} + +static void disable_sssw(void) +{ + rv_sssw.enabled = 0; + + rv_detach_trace_probe("sssw", sched_set_state_tp, handle_sched_set_state); + rv_detach_trace_probe("sssw", sched_switch, handle_sched_switch); + rv_detach_trace_probe("sssw", sched_wakeup, handle_sched_wakeup); + rv_detach_trace_probe("sssw", signal_deliver, handle_signal_deliver); + + da_monitor_destroy_sssw(); +} + +static struct rv_monitor rv_sssw = { + .name = "sssw", + .description = "set state sleep and wakeup.", + .enable = enable_sssw, + .disable = disable_sssw, + .reset = da_monitor_reset_all_sssw, + .enabled = 0, +}; + +static int __init register_sssw(void) +{ + return rv_register_monitor(&rv_sssw, &rv_sched); +} + +static void __exit unregister_sssw(void) +{ + rv_unregister_monitor(&rv_sssw); +} + +module_init(register_sssw); +module_exit(unregister_sssw); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("sssw: set state sleep and wakeup."); diff --git a/kernel/trace/rv/monitors/sssw/sssw.h b/kernel/trace/rv/monitors/sssw/sssw.h new file mode 100644 index 000000000000..243d54050c94 --- /dev/null +++ b/kernel/trace/rv/monitors/sssw/sssw.h @@ -0,0 +1,105 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of sssw automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_sssw { + runnable_sssw = 0, + signal_wakeup_sssw, + sleepable_sssw, + sleeping_sssw, + state_max_sssw +}; + +#define INVALID_STATE state_max_sssw + +enum events_sssw { + sched_set_state_runnable_sssw = 0, + sched_set_state_sleepable_sssw, + sched_switch_blocking_sssw, + sched_switch_in_sssw, + sched_switch_preempt_sssw, + sched_switch_suspend_sssw, + sched_switch_yield_sssw, + sched_wakeup_sssw, + signal_deliver_sssw, + event_max_sssw +}; + +struct automaton_sssw { + char *state_names[state_max_sssw]; + char *event_names[event_max_sssw]; + unsigned char function[state_max_sssw][event_max_sssw]; + unsigned char initial_state; + bool final_states[state_max_sssw]; +}; + +static const struct automaton_sssw automaton_sssw = { + .state_names = { + "runnable", + "signal_wakeup", + "sleepable", + "sleeping" + }, + .event_names = { + "sched_set_state_runnable", + "sched_set_state_sleepable", + "sched_switch_blocking", + "sched_switch_in", + "sched_switch_preempt", + "sched_switch_suspend", + "sched_switch_yield", + "sched_wakeup", + "signal_deliver" + }, + .function = { + { + runnable_sssw, + sleepable_sssw, + sleeping_sssw, + runnable_sssw, + runnable_sssw, + INVALID_STATE, + runnable_sssw, + runnable_sssw, + runnable_sssw + }, + { + INVALID_STATE, + sleepable_sssw, + INVALID_STATE, + signal_wakeup_sssw, + signal_wakeup_sssw, + INVALID_STATE, + signal_wakeup_sssw, + signal_wakeup_sssw, + runnable_sssw + }, + { + runnable_sssw, + sleepable_sssw, + sleeping_sssw, + sleepable_sssw, + sleepable_sssw, + sleeping_sssw, + signal_wakeup_sssw, + runnable_sssw, + sleepable_sssw + }, + { + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + runnable_sssw, + INVALID_STATE + }, + }, + .initial_state = runnable_sssw, + .final_states = { 1, 0, 0, 0 }, +}; diff --git a/kernel/trace/rv/monitors/sssw/sssw_trace.h b/kernel/trace/rv/monitors/sssw/sssw_trace.h new file mode 100644 index 000000000000..6c03cfc6960b --- /dev/null +++ b/kernel/trace/rv/monitors/sssw/sssw_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_SSSW +DEFINE_EVENT(event_da_monitor_id, event_sssw, + TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state), + TP_ARGS(id, state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor_id, error_sssw, + TP_PROTO(int id, char *state, char *event), + TP_ARGS(id, state, event)); +#endif /* CONFIG_RV_MON_SSSW */ diff --git a/kernel/trace/rv/monitors/sts/Kconfig b/kernel/trace/rv/monitors/sts/Kconfig new file mode 100644 index 000000000000..7d1ff0f6fc91 --- /dev/null +++ b/kernel/trace/rv/monitors/sts/Kconfig @@ -0,0 +1,19 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_STS + depends on RV + depends on TRACE_IRQFLAGS + depends on RV_MON_SCHED + default y + select DA_MON_EVENTS_IMPLICIT + bool "sts monitor" + help + Monitor to ensure relationships between scheduler and task switches + * the scheduler is called and returns with interrupts disabled + * each call to the scheduler has up to one switch + * switches only happen inside the scheduler + * each call to the scheduler disables interrupts to switch + This monitor is part of the sched monitors collection. + + For further information, see: + Documentation/trace/rv/monitor_sched.rst diff --git a/kernel/trace/rv/monitors/sts/sts.c b/kernel/trace/rv/monitors/sts/sts.c new file mode 100644 index 000000000000..c4a9cd67c1d2 --- /dev/null +++ b/kernel/trace/rv/monitors/sts/sts.c @@ -0,0 +1,156 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ftrace.h> +#include <linux/tracepoint.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/rv.h> +#include <rv/instrumentation.h> +#include <rv/da_monitor.h> + +#define MODULE_NAME "sts" + +#include <trace/events/sched.h> +#include <trace/events/irq.h> +#include <trace/events/preemptirq.h> +#include <rv_trace.h> +#include <monitors/sched/sched.h> + +#include "sts.h" + +static struct rv_monitor rv_sts; +DECLARE_DA_MON_PER_CPU(sts, unsigned char); + +#ifdef CONFIG_X86_LOCAL_APIC +#include <asm/trace/irq_vectors.h> + +static void handle_vector_irq_entry(void *data, int vector) +{ + da_handle_event_sts(irq_entry_sts); +} + +static void attach_vector_irq(void) +{ + rv_attach_trace_probe("sts", local_timer_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_IRQ_WORK)) + rv_attach_trace_probe("sts", irq_work_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_SMP)) { + rv_attach_trace_probe("sts", reschedule_entry, handle_vector_irq_entry); + rv_attach_trace_probe("sts", call_function_entry, handle_vector_irq_entry); + rv_attach_trace_probe("sts", call_function_single_entry, handle_vector_irq_entry); + } +} + +static void detach_vector_irq(void) +{ + rv_detach_trace_probe("sts", local_timer_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_IRQ_WORK)) + rv_detach_trace_probe("sts", irq_work_entry, handle_vector_irq_entry); + if (IS_ENABLED(CONFIG_SMP)) { + rv_detach_trace_probe("sts", reschedule_entry, handle_vector_irq_entry); + rv_detach_trace_probe("sts", call_function_entry, handle_vector_irq_entry); + rv_detach_trace_probe("sts", call_function_single_entry, handle_vector_irq_entry); + } +} + +#else +/* We assume irq_entry tracepoints are sufficient on other architectures */ +static void attach_vector_irq(void) { } +static void detach_vector_irq(void) { } +#endif + +static void handle_irq_disable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_sts(irq_disable_sts); +} + +static void handle_irq_enable(void *data, unsigned long ip, unsigned long parent_ip) +{ + da_handle_event_sts(irq_enable_sts); +} + +static void handle_irq_entry(void *data, int irq, struct irqaction *action) +{ + da_handle_event_sts(irq_entry_sts); +} + +static void handle_sched_switch(void *data, bool preempt, + struct task_struct *prev, + struct task_struct *next, + unsigned int prev_state) +{ + da_handle_event_sts(sched_switch_sts); +} + +static void handle_schedule_entry(void *data, bool preempt) +{ + da_handle_event_sts(schedule_entry_sts); +} + +static void handle_schedule_exit(void *data, bool is_switch) +{ + da_handle_start_event_sts(schedule_exit_sts); +} + +static int enable_sts(void) +{ + int retval; + + retval = da_monitor_init_sts(); + if (retval) + return retval; + + rv_attach_trace_probe("sts", irq_disable, handle_irq_disable); + rv_attach_trace_probe("sts", irq_enable, handle_irq_enable); + rv_attach_trace_probe("sts", irq_handler_entry, handle_irq_entry); + rv_attach_trace_probe("sts", sched_switch, handle_sched_switch); + rv_attach_trace_probe("sts", sched_entry_tp, handle_schedule_entry); + rv_attach_trace_probe("sts", sched_exit_tp, handle_schedule_exit); + attach_vector_irq(); + + return 0; +} + +static void disable_sts(void) +{ + rv_sts.enabled = 0; + + rv_detach_trace_probe("sts", irq_disable, handle_irq_disable); + rv_detach_trace_probe("sts", irq_enable, handle_irq_enable); + rv_detach_trace_probe("sts", irq_handler_entry, handle_irq_entry); + rv_detach_trace_probe("sts", sched_switch, handle_sched_switch); + rv_detach_trace_probe("sts", sched_entry_tp, handle_schedule_entry); + rv_detach_trace_probe("sts", sched_exit_tp, handle_schedule_exit); + detach_vector_irq(); + + da_monitor_destroy_sts(); +} + +/* + * This is the monitor register section. + */ +static struct rv_monitor rv_sts = { + .name = "sts", + .description = "schedule implies task switch.", + .enable = enable_sts, + .disable = disable_sts, + .reset = da_monitor_reset_all_sts, + .enabled = 0, +}; + +static int __init register_sts(void) +{ + return rv_register_monitor(&rv_sts, &rv_sched); +} + +static void __exit unregister_sts(void) +{ + rv_unregister_monitor(&rv_sts); +} + +module_init(register_sts); +module_exit(unregister_sts); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gabriele Monaco <gmonaco@redhat.com>"); +MODULE_DESCRIPTION("sts: schedule implies task switch."); diff --git a/kernel/trace/rv/monitors/sts/sts.h b/kernel/trace/rv/monitors/sts/sts.h new file mode 100644 index 000000000000..3368b6599a00 --- /dev/null +++ b/kernel/trace/rv/monitors/sts/sts.h @@ -0,0 +1,117 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Automatically generated C representation of sts automaton + * For further information about this format, see kernel documentation: + * Documentation/trace/rv/deterministic_automata.rst + */ + +enum states_sts { + can_sched_sts = 0, + cant_sched_sts, + disable_to_switch_sts, + enable_to_exit_sts, + in_irq_sts, + scheduling_sts, + switching_sts, + state_max_sts +}; + +#define INVALID_STATE state_max_sts + +enum events_sts { + irq_disable_sts = 0, + irq_enable_sts, + irq_entry_sts, + sched_switch_sts, + schedule_entry_sts, + schedule_exit_sts, + event_max_sts +}; + +struct automaton_sts { + char *state_names[state_max_sts]; + char *event_names[event_max_sts]; + unsigned char function[state_max_sts][event_max_sts]; + unsigned char initial_state; + bool final_states[state_max_sts]; +}; + +static const struct automaton_sts automaton_sts = { + .state_names = { + "can_sched", + "cant_sched", + "disable_to_switch", + "enable_to_exit", + "in_irq", + "scheduling", + "switching" + }, + .event_names = { + "irq_disable", + "irq_enable", + "irq_entry", + "sched_switch", + "schedule_entry", + "schedule_exit" + }, + .function = { + { + cant_sched_sts, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + scheduling_sts, + INVALID_STATE + }, + { + INVALID_STATE, + can_sched_sts, + cant_sched_sts, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE + }, + { + INVALID_STATE, + enable_to_exit_sts, + in_irq_sts, + switching_sts, + INVALID_STATE, + INVALID_STATE + }, + { + enable_to_exit_sts, + enable_to_exit_sts, + enable_to_exit_sts, + INVALID_STATE, + INVALID_STATE, + can_sched_sts + }, + { + INVALID_STATE, + scheduling_sts, + in_irq_sts, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE + }, + { + disable_to_switch_sts, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE + }, + { + INVALID_STATE, + enable_to_exit_sts, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE, + INVALID_STATE + }, + }, + .initial_state = can_sched_sts, + .final_states = { 1, 0, 0, 0, 0, 0, 0 }, +}; diff --git a/kernel/trace/rv/monitors/sts/sts_trace.h b/kernel/trace/rv/monitors/sts/sts_trace.h new file mode 100644 index 000000000000..d78beb58d5b3 --- /dev/null +++ b/kernel/trace/rv/monitors/sts/sts_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_STS +DEFINE_EVENT(event_da_monitor, event_sts, + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + TP_ARGS(state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor, error_sts, + TP_PROTO(char *state, char *event), + TP_ARGS(state, event)); +#endif /* CONFIG_RV_MON_STS */ diff --git a/kernel/trace/rv/monitors/wip/Kconfig b/kernel/trace/rv/monitors/wip/Kconfig new file mode 100644 index 000000000000..87a26195792b --- /dev/null +++ b/kernel/trace/rv/monitors/wip/Kconfig @@ -0,0 +1,14 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_WIP + depends on RV + depends on TRACE_PREEMPT_TOGGLE + select DA_MON_EVENTS_IMPLICIT + bool "wip monitor" + help + Enable wip (wakeup in preemptive) sample monitor that illustrates + the usage of per-cpu monitors, and one limitation of the + preempt_disable/enable events. + + For further information, see: + Documentation/trace/rv/monitor_wip.rst diff --git a/kernel/trace/rv/monitors/wip/wip.c b/kernel/trace/rv/monitors/wip/wip.c index b2b49a27e886..4b4e99615a11 100644 --- a/kernel/trace/rv/monitors/wip/wip.c +++ b/kernel/trace/rv/monitors/wip/wip.c @@ -10,7 +10,7 @@ #define MODULE_NAME "wip" -#include <trace/events/rv.h> +#include <rv_trace.h> #include <trace/events/sched.h> #include <trace/events/preemptirq.h> @@ -71,8 +71,7 @@ static struct rv_monitor rv_wip = { static int __init register_wip(void) { - rv_register_monitor(&rv_wip); - return 0; + return rv_register_monitor(&rv_wip, NULL); } static void __exit unregister_wip(void) diff --git a/kernel/trace/rv/monitors/wip/wip.h b/kernel/trace/rv/monitors/wip/wip.h index 2e373f2c65ed..c7193748bf36 100644 --- a/kernel/trace/rv/monitors/wip/wip.h +++ b/kernel/trace/rv/monitors/wip/wip.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Automatically generated C representation of wip automaton * For further information about this format, see kernel documentation: diff --git a/kernel/trace/rv/monitors/wip/wip_trace.h b/kernel/trace/rv/monitors/wip/wip_trace.h new file mode 100644 index 000000000000..aa2162f47a4c --- /dev/null +++ b/kernel/trace/rv/monitors/wip/wip_trace.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_WIP +DEFINE_EVENT(event_da_monitor, event_wip, + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + TP_ARGS(state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor, error_wip, + TP_PROTO(char *state, char *event), + TP_ARGS(state, event)); +#endif /* CONFIG_RV_MON_WIP */ diff --git a/kernel/trace/rv/monitors/wwnr/Kconfig b/kernel/trace/rv/monitors/wwnr/Kconfig new file mode 100644 index 000000000000..d3bfc20037db --- /dev/null +++ b/kernel/trace/rv/monitors/wwnr/Kconfig @@ -0,0 +1,13 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +config RV_MON_WWNR + depends on RV + select DA_MON_EVENTS_ID + bool "wwnr monitor" + help + Enable wwnr (wakeup while not running) sample monitor, this is a + sample monitor that illustrates the usage of per-task monitor. + The model is borken on purpose: it serves to test reactors. + + For further information, see: + Documentation/trace/rv/monitor_wwnr.rst diff --git a/kernel/trace/rv/monitors/wwnr/wwnr.c b/kernel/trace/rv/monitors/wwnr/wwnr.c index 0e43dd2db685..4145bea2729e 100644 --- a/kernel/trace/rv/monitors/wwnr/wwnr.c +++ b/kernel/trace/rv/monitors/wwnr/wwnr.c @@ -10,7 +10,7 @@ #define MODULE_NAME "wwnr" -#include <trace/events/rv.h> +#include <rv_trace.h> #include <trace/events/sched.h> #include "wwnr.h" @@ -70,8 +70,7 @@ static struct rv_monitor rv_wwnr = { static int __init register_wwnr(void) { - rv_register_monitor(&rv_wwnr); - return 0; + return rv_register_monitor(&rv_wwnr, NULL); } static void __exit unregister_wwnr(void) diff --git a/kernel/trace/rv/monitors/wwnr/wwnr.h b/kernel/trace/rv/monitors/wwnr/wwnr.h index d0d9c4b8121b..0a59d23edf61 100644 --- a/kernel/trace/rv/monitors/wwnr/wwnr.h +++ b/kernel/trace/rv/monitors/wwnr/wwnr.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Automatically generated C representation of wwnr automaton * For further information about this format, see kernel documentation: diff --git a/kernel/trace/rv/monitors/wwnr/wwnr_trace.h b/kernel/trace/rv/monitors/wwnr/wwnr_trace.h new file mode 100644 index 000000000000..fc97ec7476ad --- /dev/null +++ b/kernel/trace/rv/monitors/wwnr/wwnr_trace.h @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +/* + * Snippet to be included in rv_trace.h + */ + +#ifdef CONFIG_RV_MON_WWNR +/* id is the pid of the task */ +DEFINE_EVENT(event_da_monitor_id, event_wwnr, + TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state), + TP_ARGS(id, state, event, next_state, final_state)); + +DEFINE_EVENT(error_da_monitor_id, error_wwnr, + TP_PROTO(int id, char *state, char *event), + TP_ARGS(id, state, event)); +#endif /* CONFIG_RV_MON_WWNR */ diff --git a/kernel/trace/rv/reactor_panic.c b/kernel/trace/rv/reactor_panic.c index d65f6c25a87c..76537b8a4343 100644 --- a/kernel/trace/rv/reactor_panic.c +++ b/kernel/trace/rv/reactor_panic.c @@ -13,9 +13,9 @@ #include <linux/init.h> #include <linux/rv.h> -static void rv_panic_reaction(char *msg) +__printf(1, 0) static void rv_panic_reaction(const char *msg, va_list args) { - panic(msg); + vpanic(msg, args); } static struct rv_reactor rv_panic = { @@ -38,6 +38,5 @@ static void __exit unregister_react_panic(void) module_init(register_react_panic); module_exit(unregister_react_panic); -MODULE_LICENSE("GPL"); MODULE_AUTHOR("Daniel Bristot de Oliveira"); MODULE_DESCRIPTION("panic rv reactor: panic if an exception is found."); diff --git a/kernel/trace/rv/reactor_printk.c b/kernel/trace/rv/reactor_printk.c index 4b6b7106a477..48c934e315b3 100644 --- a/kernel/trace/rv/reactor_printk.c +++ b/kernel/trace/rv/reactor_printk.c @@ -12,9 +12,9 @@ #include <linux/init.h> #include <linux/rv.h> -static void rv_printk_reaction(char *msg) +__printf(1, 0) static void rv_printk_reaction(const char *msg, va_list args) { - printk_deferred(msg); + vprintk_deferred(msg, args); } static struct rv_reactor rv_printk = { @@ -37,6 +37,5 @@ static void __exit unregister_react_printk(void) module_init(register_react_printk); module_exit(unregister_react_printk); -MODULE_LICENSE("GPL"); MODULE_AUTHOR("Daniel Bristot de Oliveira"); MODULE_DESCRIPTION("printk rv reactor: printk if an exception is hit."); diff --git a/kernel/trace/rv/rv.c b/kernel/trace/rv/rv.c index 7e9061828c24..ee4e68102f17 100644 --- a/kernel/trace/rv/rv.c +++ b/kernel/trace/rv/rv.c @@ -41,7 +41,7 @@ * per-task monitor, and so on), and the helper functions that glue the * monitor to the system via trace. Generally, a monitor includes some form * of trace output as a reaction for event parsing and exceptions, - * as depicted bellow: + * as depicted below: * * Linux +----- RV Monitor ----------------------------------+ Formal * Realm | | Realm @@ -143,9 +143,9 @@ #include <linux/init.h> #include <linux/slab.h> -#ifdef CONFIG_DA_MON_EVENTS +#ifdef CONFIG_RV_MON_EVENTS #define CREATE_TRACE_POINTS -#include <trace/events/rv.h> +#include <rv_trace.h> #endif #include "rv.h" @@ -162,10 +162,10 @@ struct dentry *get_monitors_root(void) /* * Interface for the monitor register. */ -static LIST_HEAD(rv_monitors_list); +LIST_HEAD(rv_monitors_list); static int task_monitor_count; -static bool task_monitor_slots[RV_PER_TASK_MONITORS]; +static bool task_monitor_slots[CONFIG_RV_PER_TASK_MONITORS]; int rv_get_task_monitor_slot(void) { @@ -173,12 +173,12 @@ int rv_get_task_monitor_slot(void) lockdep_assert_held(&rv_interface_lock); - if (task_monitor_count == RV_PER_TASK_MONITORS) + if (task_monitor_count == CONFIG_RV_PER_TASK_MONITORS) return -EBUSY; task_monitor_count++; - for (i = 0; i < RV_PER_TASK_MONITORS; i++) { + for (i = 0; i < CONFIG_RV_PER_TASK_MONITORS; i++) { if (task_monitor_slots[i] == false) { task_monitor_slots[i] = true; return i; @@ -194,7 +194,7 @@ void rv_put_task_monitor_slot(int slot) { lockdep_assert_held(&rv_interface_lock); - if (slot < 0 || slot >= RV_PER_TASK_MONITORS) { + if (slot < 0 || slot >= CONFIG_RV_PER_TASK_MONITORS) { WARN_ONCE(1, "RV releasing an invalid slot!: %d\n", slot); return; } @@ -207,15 +207,44 @@ void rv_put_task_monitor_slot(int slot) } /* + * Monitors with a parent are nested, + * Monitors without a parent could be standalone or containers. + */ +bool rv_is_nested_monitor(struct rv_monitor *mon) +{ + return mon->parent != NULL; +} + +/* + * We set our list to have nested monitors listed after their parent + * if a monitor has a child element its a container. + * Containers can be also identified based on their function pointers: + * as they are not real monitors they do not need function definitions + * for enable()/disable(). Use this condition to find empty containers. + * Keep both conditions in case we have some non-compliant containers. + */ +bool rv_is_container_monitor(struct rv_monitor *mon) +{ + struct rv_monitor *next; + + if (list_is_last(&mon->list, &rv_monitors_list)) + return false; + + next = list_next_entry(mon, list); + + return next->parent == mon || !mon->enable; +} + +/* * This section collects the monitor/ files and folders. */ static ssize_t monitor_enable_read_data(struct file *filp, char __user *user_buf, size_t count, loff_t *ppos) { - struct rv_monitor_def *mdef = filp->private_data; + struct rv_monitor *mon = filp->private_data; const char *buff; - buff = mdef->monitor->enabled ? "1\n" : "0\n"; + buff = mon->enabled ? "1\n" : "0\n"; return simple_read_from_buffer(user_buf, count, ppos, buff, strlen(buff)+1); } @@ -223,13 +252,14 @@ static ssize_t monitor_enable_read_data(struct file *filp, char __user *user_buf /* * __rv_disable_monitor - disabled an enabled monitor */ -static int __rv_disable_monitor(struct rv_monitor_def *mdef, bool sync) +static int __rv_disable_monitor(struct rv_monitor *mon, bool sync) { lockdep_assert_held(&rv_interface_lock); - if (mdef->monitor->enabled) { - mdef->monitor->enabled = 0; - mdef->monitor->disable(); + if (mon->enabled) { + mon->enabled = 0; + if (mon->disable) + mon->disable(); /* * Wait for the execution of all events to finish. @@ -243,35 +273,90 @@ static int __rv_disable_monitor(struct rv_monitor_def *mdef, bool sync) return 0; } +static void rv_disable_single(struct rv_monitor *mon) +{ + __rv_disable_monitor(mon, true); +} + +static int rv_enable_single(struct rv_monitor *mon) +{ + int retval; + + lockdep_assert_held(&rv_interface_lock); + + if (mon->enabled) + return 0; + + retval = mon->enable(); + + if (!retval) + mon->enabled = 1; + + return retval; +} + +static void rv_disable_container(struct rv_monitor *mon) +{ + struct rv_monitor *p = mon; + int enabled = 0; + + list_for_each_entry_continue(p, &rv_monitors_list, list) { + if (p->parent != mon) + break; + enabled += __rv_disable_monitor(p, false); + } + if (enabled) + tracepoint_synchronize_unregister(); + mon->enabled = 0; +} + +static int rv_enable_container(struct rv_monitor *mon) +{ + struct rv_monitor *p = mon; + int retval = 0; + + list_for_each_entry_continue(p, &rv_monitors_list, list) { + if (retval || p->parent != mon) + break; + retval = rv_enable_single(p); + } + if (retval) + rv_disable_container(mon); + else + mon->enabled = 1; + return retval; +} + /** * rv_disable_monitor - disable a given runtime monitor + * @mon: Pointer to the monitor definition structure. * * Returns 0 on success. */ -int rv_disable_monitor(struct rv_monitor_def *mdef) +int rv_disable_monitor(struct rv_monitor *mon) { - __rv_disable_monitor(mdef, true); + if (rv_is_container_monitor(mon)) + rv_disable_container(mon); + else + rv_disable_single(mon); + return 0; } /** * rv_enable_monitor - enable a given runtime monitor + * @mon: Pointer to the monitor definition structure. * * Returns 0 on success, error otherwise. */ -int rv_enable_monitor(struct rv_monitor_def *mdef) +int rv_enable_monitor(struct rv_monitor *mon) { int retval; - lockdep_assert_held(&rv_interface_lock); - - if (mdef->monitor->enabled) - return 0; - - retval = mdef->monitor->enable(); - - if (!retval) - mdef->monitor->enabled = 1; + if (rv_is_container_monitor(mon)) + retval = rv_enable_container(mon); + else + retval = rv_enable_single(mon); return retval; } @@ -282,7 +367,7 @@ int rv_enable_monitor(struct rv_monitor_def *mdef) static ssize_t monitor_enable_write_data(struct file *filp, const char __user *user_buf, size_t count, loff_t *ppos) { - struct rv_monitor_def *mdef = filp->private_data; + struct rv_monitor *mon = filp->private_data; int retval; bool val; @@ -290,23 +375,18 @@ static ssize_t monitor_enable_write_data(struct file *filp, const char __user *u if (retval) return retval; - retval = count; - - mutex_lock(&rv_interface_lock); + guard(mutex)(&rv_interface_lock); if (val) - retval = rv_enable_monitor(mdef); + retval = rv_enable_monitor(mon); else - retval = rv_disable_monitor(mdef); - - mutex_unlock(&rv_interface_lock); + retval = rv_disable_monitor(mon); return retval ? : count; } static const struct file_operations interface_enable_fops = { .open = simple_open, - .llseek = no_llseek, .write = monitor_enable_write_data, .read = monitor_enable_read_data, }; @@ -317,19 +397,18 @@ static const struct file_operations interface_enable_fops = { static ssize_t monitor_desc_read_data(struct file *filp, char __user *user_buf, size_t count, loff_t *ppos) { - struct rv_monitor_def *mdef = filp->private_data; + struct rv_monitor *mon = filp->private_data; char buff[256]; memset(buff, 0, sizeof(buff)); - snprintf(buff, sizeof(buff), "%s\n", mdef->monitor->description); + snprintf(buff, sizeof(buff), "%s\n", mon->description); return simple_read_from_buffer(user_buf, count, ppos, buff, strlen(buff) + 1); } static const struct file_operations interface_desc_fops = { .open = simple_open, - .llseek = no_llseek, .read = monitor_desc_read_data, }; @@ -338,38 +417,30 @@ static const struct file_operations interface_desc_fops = { * the monitor dir, where the specific options of the monitor * are exposed. */ -static int create_monitor_dir(struct rv_monitor_def *mdef) +static int create_monitor_dir(struct rv_monitor *mon, struct rv_monitor *parent) { - struct dentry *root = get_monitors_root(); - const char *name = mdef->monitor->name; + struct dentry *root = parent ? parent->root_d : get_monitors_root(); + struct dentry *dir __free(rv_remove) = rv_create_dir(mon->name, root); struct dentry *tmp; int retval; - mdef->root_d = rv_create_dir(name, root); - if (!mdef->root_d) + if (!dir) return -ENOMEM; - tmp = rv_create_file("enable", RV_MODE_WRITE, mdef->root_d, mdef, &interface_enable_fops); - if (!tmp) { - retval = -ENOMEM; - goto out_remove_root; - } + tmp = rv_create_file("enable", RV_MODE_WRITE, dir, mon, &interface_enable_fops); + if (!tmp) + return -ENOMEM; - tmp = rv_create_file("desc", RV_MODE_READ, mdef->root_d, mdef, &interface_desc_fops); - if (!tmp) { - retval = -ENOMEM; - goto out_remove_root; - } + tmp = rv_create_file("desc", RV_MODE_READ, dir, mon, &interface_desc_fops); + if (!tmp) + return -ENOMEM; - retval = reactor_populate_monitor(mdef); + retval = reactor_populate_monitor(mon, dir); if (retval) - goto out_remove_root; + return retval; + mon->root_d = no_free_ptr(dir); return 0; - -out_remove_root: - rv_remove(mdef->root_d); - return retval; } /* @@ -377,9 +448,12 @@ out_remove_root: */ static int monitors_show(struct seq_file *m, void *p) { - struct rv_monitor_def *mon_def = p; + struct rv_monitor *mon = container_of(p, struct rv_monitor, list); - seq_printf(m, "%s\n", mon_def->monitor->name); + if (mon->parent) + seq_printf(m, "%s:%s\n", mon->parent->name, mon->name); + else + seq_printf(m, "%s\n", mon->name); return 0; } @@ -411,13 +485,13 @@ static void *available_monitors_next(struct seq_file *m, void *p, loff_t *pos) */ static void *enabled_monitors_next(struct seq_file *m, void *p, loff_t *pos) { - struct rv_monitor_def *m_def = p; + struct rv_monitor *mon = container_of(p, struct rv_monitor, list); (*pos)++; - list_for_each_entry_continue(m_def, &rv_monitors_list, list) { - if (m_def->monitor->enabled) - return m_def; + list_for_each_entry_continue(mon, &rv_monitors_list, list) { + if (mon->enabled) + return &mon->list; } return NULL; @@ -425,7 +499,7 @@ static void *enabled_monitors_next(struct seq_file *m, void *p, loff_t *pos) static void *enabled_monitors_start(struct seq_file *m, loff_t *pos) { - struct rv_monitor_def *m_def; + struct list_head *head; loff_t l; mutex_lock(&rv_interface_lock); @@ -433,15 +507,15 @@ static void *enabled_monitors_start(struct seq_file *m, loff_t *pos) if (list_empty(&rv_monitors_list)) return NULL; - m_def = list_entry(&rv_monitors_list, struct rv_monitor_def, list); + head = &rv_monitors_list; for (l = 0; l <= *pos; ) { - m_def = enabled_monitors_next(m, m_def, &l); - if (!m_def) + head = enabled_monitors_next(m, head, &l); + if (!head) break; } - return m_def; + return head; } /* @@ -481,13 +555,13 @@ static const struct file_operations available_monitors_ops = { */ static void disable_all_monitors(void) { - struct rv_monitor_def *mdef; + struct rv_monitor *mon; int enabled = 0; - mutex_lock(&rv_interface_lock); + guard(mutex)(&rv_interface_lock); - list_for_each_entry(mdef, &rv_monitors_list, list) - enabled += __rv_disable_monitor(mdef, false); + list_for_each_entry(mon, &rv_monitors_list, list) + enabled += __rv_disable_monitor(mon, false); if (enabled) { /* @@ -497,8 +571,6 @@ static void disable_all_monitors(void) */ tracepoint_synchronize_unregister(); } - - mutex_unlock(&rv_interface_lock); } static int enabled_monitors_open(struct inode *inode, struct file *file) @@ -513,10 +585,10 @@ static ssize_t enabled_monitors_write(struct file *filp, const char __user *user size_t count, loff_t *ppos) { char buff[MAX_RV_MONITOR_NAME_SIZE + 2]; - struct rv_monitor_def *mdef; + struct rv_monitor *mon; int retval = -EINVAL; bool enable = true; - char *ptr; + char *ptr, *tmp; int len; if (count < 1 || count > MAX_RV_MONITOR_NAME_SIZE + 1) @@ -539,29 +611,32 @@ static ssize_t enabled_monitors_write(struct file *filp, const char __user *user if (!len) return count; - mutex_lock(&rv_interface_lock); + guard(mutex)(&rv_interface_lock); retval = -EINVAL; - list_for_each_entry(mdef, &rv_monitors_list, list) { - if (strcmp(ptr, mdef->monitor->name) != 0) + /* we support 1 nesting level, trim the parent */ + tmp = strstr(ptr, ":"); + if (tmp) + ptr = tmp+1; + + list_for_each_entry(mon, &rv_monitors_list, list) { + if (strcmp(ptr, mon->name) != 0) continue; /* * Monitor found! */ if (enable) - retval = rv_enable_monitor(mdef); + retval = rv_enable_monitor(mon); else - retval = rv_disable_monitor(mdef); + retval = rv_disable_monitor(mon); - if (!retval) - retval = count; - - break; + if (retval) + return retval; + return count; } - mutex_unlock(&rv_interface_lock); return retval; } @@ -585,8 +660,6 @@ static bool __read_mostly monitoring_on; */ bool rv_monitoring_on(void) { - /* Ensures that concurrent monitors read consistent monitoring_on */ - smp_rmb(); return READ_ONCE(monitoring_on); } @@ -606,25 +679,21 @@ static ssize_t monitoring_on_read_data(struct file *filp, char __user *user_buf, static void turn_monitoring_off(void) { WRITE_ONCE(monitoring_on, false); - /* Ensures that concurrent monitors read consistent monitoring_on */ - smp_wmb(); } static void reset_all_monitors(void) { - struct rv_monitor_def *mdef; + struct rv_monitor *mon; - list_for_each_entry(mdef, &rv_monitors_list, list) { - if (mdef->monitor->enabled) - mdef->monitor->reset(); + list_for_each_entry(mon, &rv_monitors_list, list) { + if (mon->enabled && mon->reset) + mon->reset(); } } static void turn_monitoring_on(void) { WRITE_ONCE(monitoring_on, true); - /* Ensures that concurrent monitors read consistent monitoring_on */ - smp_wmb(); } static void turn_monitoring_on_with_reset(void) @@ -654,7 +723,7 @@ static ssize_t monitoring_on_write_data(struct file *filp, const char __user *us if (retval) return retval; - mutex_lock(&rv_interface_lock); + guard(mutex)(&rv_interface_lock); if (val) turn_monitoring_on_with_reset(); @@ -667,70 +736,66 @@ static ssize_t monitoring_on_write_data(struct file *filp, const char __user *us */ tracepoint_synchronize_unregister(); - mutex_unlock(&rv_interface_lock); - return count; } static const struct file_operations monitoring_on_fops = { .open = simple_open, - .llseek = no_llseek, .write = monitoring_on_write_data, .read = monitoring_on_read_data, }; -static void destroy_monitor_dir(struct rv_monitor_def *mdef) +static void destroy_monitor_dir(struct rv_monitor *mon) { - reactor_cleanup_monitor(mdef); - rv_remove(mdef->root_d); + rv_remove(mon->root_d); } /** * rv_register_monitor - register a rv monitor. * @monitor: The rv_monitor to be registered. + * @parent: The parent of the monitor to be registered, NULL if not nested. * * Returns 0 if successful, error otherwise. */ -int rv_register_monitor(struct rv_monitor *monitor) +int rv_register_monitor(struct rv_monitor *monitor, struct rv_monitor *parent) { - struct rv_monitor_def *r; + struct rv_monitor *r; int retval = 0; if (strlen(monitor->name) >= MAX_RV_MONITOR_NAME_SIZE) { pr_info("Monitor %s has a name longer than %d\n", monitor->name, MAX_RV_MONITOR_NAME_SIZE); - return -1; + return -EINVAL; } - mutex_lock(&rv_interface_lock); + guard(mutex)(&rv_interface_lock); list_for_each_entry(r, &rv_monitors_list, list) { - if (strcmp(monitor->name, r->monitor->name) == 0) { + if (strcmp(monitor->name, r->name) == 0) { pr_info("Monitor %s is already registered\n", monitor->name); - retval = -1; - goto out_unlock; + return -EEXIST; } } - r = kzalloc(sizeof(struct rv_monitor_def), GFP_KERNEL); - if (!r) { - retval = -ENOMEM; - goto out_unlock; + if (parent && rv_is_nested_monitor(parent)) { + pr_info("Parent monitor %s is already nested, cannot nest further\n", + parent->name); + return -EINVAL; } - r->monitor = monitor; + monitor->parent = parent; - retval = create_monitor_dir(r); - if (retval) { - kfree(r); - goto out_unlock; - } + retval = create_monitor_dir(monitor, parent); + if (retval) + return retval; - list_add_tail(&r->list, &rv_monitors_list); + /* keep children close to the parent for easier visualisation */ + if (parent) + list_add(&monitor->list, &parent->list); + else + list_add_tail(&monitor->list, &rv_monitors_list); -out_unlock: - mutex_unlock(&rv_interface_lock); - return retval; + return 0; } /** @@ -741,19 +806,12 @@ out_unlock: */ int rv_unregister_monitor(struct rv_monitor *monitor) { - struct rv_monitor_def *ptr, *next; + guard(mutex)(&rv_interface_lock); - mutex_lock(&rv_interface_lock); - - list_for_each_entry_safe(ptr, next, &rv_monitors_list, list) { - if (strcmp(monitor->name, ptr->monitor->name) == 0) { - rv_disable_monitor(ptr); - list_del(&ptr->list); - destroy_monitor_dir(ptr); - } - } + rv_disable_monitor(monitor); + list_del(&monitor->list); + destroy_monitor_dir(monitor); - mutex_unlock(&rv_interface_lock); return 0; } @@ -761,39 +819,36 @@ int __init rv_init_interface(void) { struct dentry *tmp; int retval; + struct dentry *root_dir __free(rv_remove) = rv_create_dir("rv", NULL); - rv_root.root_dir = rv_create_dir("rv", NULL); - if (!rv_root.root_dir) - goto out_err; + if (!root_dir) + return 1; - rv_root.monitors_dir = rv_create_dir("monitors", rv_root.root_dir); + rv_root.monitors_dir = rv_create_dir("monitors", root_dir); if (!rv_root.monitors_dir) - goto out_err; + return 1; - tmp = rv_create_file("available_monitors", RV_MODE_READ, rv_root.root_dir, NULL, + tmp = rv_create_file("available_monitors", RV_MODE_READ, root_dir, NULL, &available_monitors_ops); if (!tmp) - goto out_err; + return 1; - tmp = rv_create_file("enabled_monitors", RV_MODE_WRITE, rv_root.root_dir, NULL, + tmp = rv_create_file("enabled_monitors", RV_MODE_WRITE, root_dir, NULL, &enabled_monitors_ops); if (!tmp) - goto out_err; + return 1; - tmp = rv_create_file("monitoring_on", RV_MODE_WRITE, rv_root.root_dir, NULL, + tmp = rv_create_file("monitoring_on", RV_MODE_WRITE, root_dir, NULL, &monitoring_on_fops); if (!tmp) - goto out_err; - retval = init_rv_reactors(rv_root.root_dir); + return 1; + retval = init_rv_reactors(root_dir); if (retval) - goto out_err; + return 1; turn_monitoring_on(); - return 0; + rv_root.root_dir = no_free_ptr(root_dir); -out_err: - rv_remove(rv_root.root_dir); - printk(KERN_ERR "RV: Error while creating the RV interface\n"); - return 1; + return 0; } diff --git a/kernel/trace/rv/rv.h b/kernel/trace/rv/rv.h index db6cb0913dbd..2c0f51ff9d5c 100644 --- a/kernel/trace/rv/rv.h +++ b/kernel/trace/rv/rv.h @@ -17,50 +17,29 @@ struct rv_interface { #define rv_create_file tracefs_create_file #define rv_remove tracefs_remove +DEFINE_FREE(rv_remove, struct dentry *, if (_T) rv_remove(_T)); + #define MAX_RV_MONITOR_NAME_SIZE 32 #define MAX_RV_REACTOR_NAME_SIZE 32 extern struct mutex rv_interface_lock; - -#ifdef CONFIG_RV_REACTORS -struct rv_reactor_def { - struct list_head list; - struct rv_reactor *reactor; - /* protected by the monitor interface lock */ - int counter; -}; -#endif - -struct rv_monitor_def { - struct list_head list; - struct rv_monitor *monitor; - struct dentry *root_d; -#ifdef CONFIG_RV_REACTORS - struct rv_reactor_def *rdef; - bool reacting; -#endif - bool task_monitor; -}; +extern struct list_head rv_monitors_list; struct dentry *get_monitors_root(void); -int rv_disable_monitor(struct rv_monitor_def *mdef); -int rv_enable_monitor(struct rv_monitor_def *mdef); +int rv_disable_monitor(struct rv_monitor *mon); +int rv_enable_monitor(struct rv_monitor *mon); +bool rv_is_container_monitor(struct rv_monitor *mon); +bool rv_is_nested_monitor(struct rv_monitor *mon); #ifdef CONFIG_RV_REACTORS -int reactor_populate_monitor(struct rv_monitor_def *mdef); -void reactor_cleanup_monitor(struct rv_monitor_def *mdef); +int reactor_populate_monitor(struct rv_monitor *mon, struct dentry *root); int init_rv_reactors(struct dentry *root_dir); #else -static inline int reactor_populate_monitor(struct rv_monitor_def *mdef) +static inline int reactor_populate_monitor(struct rv_monitor *mon, struct dentry *root) { return 0; } -static inline void reactor_cleanup_monitor(struct rv_monitor_def *mdef) -{ - return; -} - static inline int init_rv_reactors(struct dentry *root_dir) { return 0; diff --git a/kernel/trace/rv/rv_reactors.c b/kernel/trace/rv/rv_reactors.c index 6aae106695b6..460af07f7aba 100644 --- a/kernel/trace/rv/rv_reactors.c +++ b/kernel/trace/rv/rv_reactors.c @@ -61,6 +61,7 @@ * printk */ +#include <linux/lockdep.h> #include <linux/slab.h> #include "rv.h" @@ -70,12 +71,12 @@ */ static LIST_HEAD(rv_reactors_list); -static struct rv_reactor_def *get_reactor_rdef_by_name(char *name) +static struct rv_reactor *get_reactor_rdef_by_name(char *name) { - struct rv_reactor_def *r; + struct rv_reactor *r; list_for_each_entry(r, &rv_reactors_list, list) { - if (strcmp(name, r->reactor->name) == 0) + if (strcmp(name, r->name) == 0) return r; } return NULL; @@ -86,9 +87,9 @@ static struct rv_reactor_def *get_reactor_rdef_by_name(char *name) */ static int reactors_show(struct seq_file *m, void *p) { - struct rv_reactor_def *rea_def = p; + struct rv_reactor *reactor = container_of(p, struct rv_reactor, list); - seq_printf(m, "%s\n", rea_def->reactor->name); + seq_printf(m, "%s\n", reactor->name); return 0; } @@ -138,13 +139,13 @@ static const struct file_operations available_reactors_ops = { */ static int monitor_reactor_show(struct seq_file *m, void *p) { - struct rv_monitor_def *mdef = m->private; - struct rv_reactor_def *rdef = p; + struct rv_monitor *mon = m->private; + struct rv_reactor *reactor = container_of(p, struct rv_reactor, list); - if (mdef->rdef == rdef) - seq_printf(m, "[%s]\n", rdef->reactor->name); + if (mon->reactor == reactor) + seq_printf(m, "[%s]\n", reactor->name); else - seq_printf(m, "%s\n", rdef->reactor->name); + seq_printf(m, "%s\n", reactor->name); return 0; } @@ -158,29 +159,45 @@ static const struct seq_operations monitor_reactors_seq_ops = { .show = monitor_reactor_show }; -static void monitor_swap_reactors(struct rv_monitor_def *mdef, struct rv_reactor_def *rdef, - bool reacting) +static void monitor_swap_reactors_single(struct rv_monitor *mon, + struct rv_reactor *reactor, + bool nested) { bool monitor_enabled; /* nothing to do */ - if (mdef->rdef == rdef) + if (mon->reactor == reactor) return; - monitor_enabled = mdef->monitor->enabled; + monitor_enabled = mon->enabled; if (monitor_enabled) - rv_disable_monitor(mdef); + rv_disable_monitor(mon); - /* swap reactor's usage */ - mdef->rdef->counter--; - rdef->counter++; + mon->reactor = reactor; + mon->react = reactor->react; - mdef->rdef = rdef; - mdef->reacting = reacting; - mdef->monitor->react = rdef->reactor->react; + /* enable only once if iterating through a container */ + if (monitor_enabled && !nested) + rv_enable_monitor(mon); +} - if (monitor_enabled) - rv_enable_monitor(mdef); +static void monitor_swap_reactors(struct rv_monitor *mon, struct rv_reactor *reactor) +{ + struct rv_monitor *p = mon; + + if (rv_is_container_monitor(mon)) + list_for_each_entry_continue(p, &rv_monitors_list, list) { + if (p->parent != mon) + break; + monitor_swap_reactors_single(p, reactor, true); + } + /* + * This call enables and disables the monitor if they were active. + * In case of a container, we already disabled all and will enable all. + * All nested monitors are enabled also if they were off, we may refine + * this logic in the future. + */ + monitor_swap_reactors_single(mon, reactor, false); } static ssize_t @@ -188,11 +205,10 @@ monitor_reactors_write(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) { char buff[MAX_RV_REACTOR_NAME_SIZE + 2]; - struct rv_monitor_def *mdef; - struct rv_reactor_def *rdef; + struct rv_monitor *mon; + struct rv_reactor *reactor; struct seq_file *seq_f; int retval = -EINVAL; - bool enable; char *ptr; int len; @@ -215,30 +231,20 @@ monitor_reactors_write(struct file *file, const char __user *user_buf, * See monitor_reactors_open() */ seq_f = file->private_data; - mdef = seq_f->private; - - mutex_lock(&rv_interface_lock); + mon = seq_f->private; - retval = -EINVAL; + guard(mutex)(&rv_interface_lock); - list_for_each_entry(rdef, &rv_reactors_list, list) { - if (strcmp(ptr, rdef->reactor->name) != 0) + list_for_each_entry(reactor, &rv_reactors_list, list) { + if (strcmp(ptr, reactor->name) != 0) continue; - if (rdef == get_reactor_rdef_by_name("nop")) - enable = false; - else - enable = true; - - monitor_swap_reactors(mdef, rdef, enable); + monitor_swap_reactors(mon, reactor); - retval = count; - break; + return count; } - mutex_unlock(&rv_interface_lock); - - return retval; + return -EINVAL; } /* @@ -246,7 +252,7 @@ monitor_reactors_write(struct file *file, const char __user *user_buf, */ static int monitor_reactors_open(struct inode *inode, struct file *file) { - struct rv_monitor_def *mdef = inode->i_private; + struct rv_monitor *mon = inode->i_private; struct seq_file *seq_f; int ret; @@ -262,7 +268,7 @@ static int monitor_reactors_open(struct inode *inode, struct file *file) /* * Copy the create file "private" data to the seq_file private data. */ - seq_f->private = mdef; + seq_f->private = mon; return 0; }; @@ -277,23 +283,16 @@ static const struct file_operations monitor_reactors_ops = { static int __rv_register_reactor(struct rv_reactor *reactor) { - struct rv_reactor_def *r; + struct rv_reactor *r; list_for_each_entry(r, &rv_reactors_list, list) { - if (strcmp(reactor->name, r->reactor->name) == 0) { + if (strcmp(reactor->name, r->name) == 0) { pr_info("Reactor %s is already registered\n", reactor->name); return -EINVAL; } } - r = kzalloc(sizeof(struct rv_reactor_def), GFP_KERNEL); - if (!r) - return -ENOMEM; - - r->reactor = reactor; - r->counter = 0; - - list_add_tail(&r->list, &rv_reactors_list); + list_add_tail(&reactor->list, &rv_reactors_list); return 0; } @@ -306,18 +305,14 @@ static int __rv_register_reactor(struct rv_reactor *reactor) */ int rv_register_reactor(struct rv_reactor *reactor) { - int retval = 0; - if (strlen(reactor->name) >= MAX_RV_REACTOR_NAME_SIZE) { pr_info("Reactor %s has a name longer than %d\n", reactor->name, MAX_RV_MONITOR_NAME_SIZE); return -EINVAL; } - mutex_lock(&rv_interface_lock); - retval = __rv_register_reactor(reactor); - mutex_unlock(&rv_interface_lock); - return retval; + guard(mutex)(&rv_interface_lock); + return __rv_register_reactor(reactor); } /** @@ -328,30 +323,9 @@ int rv_register_reactor(struct rv_reactor *reactor) */ int rv_unregister_reactor(struct rv_reactor *reactor) { - struct rv_reactor_def *ptr, *next; - int ret = 0; - - mutex_lock(&rv_interface_lock); - - list_for_each_entry_safe(ptr, next, &rv_reactors_list, list) { - if (strcmp(reactor->name, ptr->reactor->name) == 0) { - - if (!ptr->counter) { - list_del(&ptr->list); - } else { - printk(KERN_WARNING - "rv: the rv_reactor %s is in use by %d monitor(s)\n", - ptr->reactor->name, ptr->counter); - printk(KERN_WARNING "rv: the rv_reactor %s cannot be removed\n", - ptr->reactor->name); - ret = -EBUSY; - break; - } - } - } - - mutex_unlock(&rv_interface_lock); - return ret; + guard(mutex)(&rv_interface_lock); + list_del(&reactor->list); + return 0; } /* @@ -364,7 +338,7 @@ static bool __read_mostly reacting_on; * * Returns 1 if on, 0 otherwise. */ -bool rv_reacting_on(void) +static bool rv_reacting_on(void) { /* Ensures that concurrent monitors read consistent reacting_on */ smp_rmb(); @@ -406,7 +380,7 @@ static ssize_t reacting_on_write_data(struct file *filp, const char __user *user if (retval) return retval; - mutex_lock(&rv_interface_lock); + guard(mutex)(&rv_interface_lock); if (val) turn_reacting_on(); @@ -419,57 +393,42 @@ static ssize_t reacting_on_write_data(struct file *filp, const char __user *user */ tracepoint_synchronize_unregister(); - mutex_unlock(&rv_interface_lock); - return count; } static const struct file_operations reacting_on_fops = { .open = simple_open, - .llseek = no_llseek, .write = reacting_on_write_data, .read = reacting_on_read_data, }; /** * reactor_populate_monitor - creates per monitor reactors file - * @mdef: monitor's definition. + * @mon: The monitor. + * @root: The directory of the monitor. * * Returns 0 if successful, error otherwise. */ -int reactor_populate_monitor(struct rv_monitor_def *mdef) +int reactor_populate_monitor(struct rv_monitor *mon, struct dentry *root) { struct dentry *tmp; - tmp = rv_create_file("reactors", RV_MODE_WRITE, mdef->root_d, mdef, &monitor_reactors_ops); + tmp = rv_create_file("reactors", RV_MODE_WRITE, root, mon, &monitor_reactors_ops); if (!tmp) return -ENOMEM; /* * Configure as the rv_nop reactor. */ - mdef->rdef = get_reactor_rdef_by_name("nop"); - mdef->rdef->counter++; - mdef->reacting = false; + mon->reactor = get_reactor_rdef_by_name("nop"); return 0; } -/** - * reactor_cleanup_monitor - cleanup a monitor reference - * @mdef: monitor's definition. - */ -void reactor_cleanup_monitor(struct rv_monitor_def *mdef) -{ - lockdep_assert_held(&rv_interface_lock); - mdef->rdef->counter--; - WARN_ON_ONCE(mdef->rdef->counter < 0); -} - /* * Nop reactor register */ -static void rv_nop_reaction(char *msg) +__printf(1, 0) static void rv_nop_reaction(const char *msg, va_list args) { } @@ -481,30 +440,42 @@ static struct rv_reactor rv_nop = { int init_rv_reactors(struct dentry *root_dir) { - struct dentry *available, *reacting; int retval; - available = rv_create_file("available_reactors", RV_MODE_READ, root_dir, NULL, - &available_reactors_ops); - if (!available) - goto out_err; + struct dentry *available __free(rv_remove) = + rv_create_file("available_reactors", RV_MODE_READ, root_dir, + NULL, &available_reactors_ops); + + struct dentry *reacting __free(rv_remove) = + rv_create_file("reacting_on", RV_MODE_WRITE, root_dir, NULL, &reacting_on_fops); - reacting = rv_create_file("reacting_on", RV_MODE_WRITE, root_dir, NULL, &reacting_on_fops); - if (!reacting) - goto rm_available; + if (!reacting || !available) + return -ENOMEM; retval = __rv_register_reactor(&rv_nop); if (retval) - goto rm_reacting; + return retval; turn_reacting_on(); + retain_and_null_ptr(available); + retain_and_null_ptr(reacting); return 0; +} + +void rv_react(struct rv_monitor *monitor, const char *msg, ...) +{ + static DEFINE_WAIT_OVERRIDE_MAP(rv_react_map, LD_WAIT_FREE); + va_list args; + + if (!rv_reacting_on() || !monitor->react) + return; + + va_start(args, msg); + + lock_map_acquire_try(&rv_react_map); + monitor->react(msg, args); + lock_map_release(&rv_react_map); -rm_reacting: - rv_remove(reacting); -rm_available: - rv_remove(available); -out_err: - return -ENOMEM; + va_end(args); } diff --git a/kernel/trace/rv/rv_trace.h b/kernel/trace/rv/rv_trace.h new file mode 100644 index 000000000000..4a6faddac614 --- /dev/null +++ b/kernel/trace/rv/rv_trace.h @@ -0,0 +1,212 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rv + +#if !defined(_TRACE_RV_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_RV_H + +#include <linux/rv.h> +#include <linux/tracepoint.h> + +#ifdef CONFIG_DA_MON_EVENTS_IMPLICIT +DECLARE_EVENT_CLASS(event_da_monitor, + + TP_PROTO(char *state, char *event, char *next_state, bool final_state), + + TP_ARGS(state, event, next_state, final_state), + + TP_STRUCT__entry( + __string( state, state ) + __string( event, event ) + __string( next_state, next_state ) + __field( bool, final_state ) + ), + + TP_fast_assign( + __assign_str(state); + __assign_str(event); + __assign_str(next_state); + __entry->final_state = final_state; + ), + + TP_printk("%s x %s -> %s%s", + __get_str(state), + __get_str(event), + __get_str(next_state), + __entry->final_state ? " (final)" : "") +); + +DECLARE_EVENT_CLASS(error_da_monitor, + + TP_PROTO(char *state, char *event), + + TP_ARGS(state, event), + + TP_STRUCT__entry( + __string( state, state ) + __string( event, event ) + ), + + TP_fast_assign( + __assign_str(state); + __assign_str(event); + ), + + TP_printk("event %s not expected in the state %s", + __get_str(event), + __get_str(state)) +); + +#include <monitors/wip/wip_trace.h> +#include <monitors/sco/sco_trace.h> +#include <monitors/scpd/scpd_trace.h> +#include <monitors/snep/snep_trace.h> +#include <monitors/sts/sts_trace.h> +#include <monitors/opid/opid_trace.h> +// Add new monitors based on CONFIG_DA_MON_EVENTS_IMPLICIT here + +#endif /* CONFIG_DA_MON_EVENTS_IMPLICIT */ + +#ifdef CONFIG_DA_MON_EVENTS_ID +DECLARE_EVENT_CLASS(event_da_monitor_id, + + TP_PROTO(int id, char *state, char *event, char *next_state, bool final_state), + + TP_ARGS(id, state, event, next_state, final_state), + + TP_STRUCT__entry( + __field( int, id ) + __string( state, state ) + __string( event, event ) + __string( next_state, next_state ) + __field( bool, final_state ) + ), + + TP_fast_assign( + __assign_str(state); + __assign_str(event); + __assign_str(next_state); + __entry->id = id; + __entry->final_state = final_state; + ), + + TP_printk("%d: %s x %s -> %s%s", + __entry->id, + __get_str(state), + __get_str(event), + __get_str(next_state), + __entry->final_state ? " (final)" : "") +); + +DECLARE_EVENT_CLASS(error_da_monitor_id, + + TP_PROTO(int id, char *state, char *event), + + TP_ARGS(id, state, event), + + TP_STRUCT__entry( + __field( int, id ) + __string( state, state ) + __string( event, event ) + ), + + TP_fast_assign( + __assign_str(state); + __assign_str(event); + __entry->id = id; + ), + + TP_printk("%d: event %s not expected in the state %s", + __entry->id, + __get_str(event), + __get_str(state)) +); + +#include <monitors/wwnr/wwnr_trace.h> +#include <monitors/snroc/snroc_trace.h> +#include <monitors/nrp/nrp_trace.h> +#include <monitors/sssw/sssw_trace.h> +// Add new monitors based on CONFIG_DA_MON_EVENTS_ID here + +#endif /* CONFIG_DA_MON_EVENTS_ID */ +#ifdef CONFIG_LTL_MON_EVENTS_ID +DECLARE_EVENT_CLASS(event_ltl_monitor_id, + + TP_PROTO(struct task_struct *task, char *states, char *atoms, char *next), + + TP_ARGS(task, states, atoms, next), + + TP_STRUCT__entry( + __string(comm, task->comm) + __field(pid_t, pid) + __string(states, states) + __string(atoms, atoms) + __string(next, next) + ), + + TP_fast_assign( + __assign_str(comm); + __entry->pid = task->pid; + __assign_str(states); + __assign_str(atoms); + __assign_str(next); + ), + + TP_printk("%s[%d]: (%s) x (%s) -> (%s)", __get_str(comm), __entry->pid, + __get_str(states), __get_str(atoms), __get_str(next)) +); + +DECLARE_EVENT_CLASS(error_ltl_monitor_id, + + TP_PROTO(struct task_struct *task), + + TP_ARGS(task), + + TP_STRUCT__entry( + __string(comm, task->comm) + __field(pid_t, pid) + ), + + TP_fast_assign( + __assign_str(comm); + __entry->pid = task->pid; + ), + + TP_printk("%s[%d]: violation detected", __get_str(comm), __entry->pid) +); +#include <monitors/pagefault/pagefault_trace.h> +#include <monitors/sleep/sleep_trace.h> +// Add new monitors based on CONFIG_LTL_MON_EVENTS_ID here +#endif /* CONFIG_LTL_MON_EVENTS_ID */ + +#ifdef CONFIG_RV_MON_MAINTENANCE_EVENTS +/* Tracepoint useful for monitors development, currenly only used in DA */ +TRACE_EVENT(rv_retries_error, + + TP_PROTO(char *name, char *event), + + TP_ARGS(name, event), + + TP_STRUCT__entry( + __string( name, name ) + __string( event, event ) + ), + + TP_fast_assign( + __assign_str(name); + __assign_str(event); + ), + + TP_printk(__stringify(MAX_DA_RETRY_RACING_EVENTS) + " retries reached for event %s, resetting monitor %s", + __get_str(event), __get_str(name)) +); +#endif /* CONFIG_RV_MON_MAINTENANCE_EVENTS */ +#endif /* _TRACE_RV_H */ + +/* This part must be outside protection */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE rv_trace +#include <trace/define_trace.h> diff --git a/kernel/trace/synth_event_gen_test.c b/kernel/trace/synth_event_gen_test.c index 8d77526892f4..354c2117be43 100644 --- a/kernel/trace/synth_event_gen_test.c +++ b/kernel/trace/synth_event_gen_test.c @@ -22,7 +22,7 @@ * Then: * * # insmod kernel/trace/synth_event_gen_test.ko - * # cat /sys/kernel/debug/tracing/trace + * # cat /sys/kernel/tracing/trace * * You should see several events in the trace buffer - * "create_synth_test", "empty_synth_test", and several instances of @@ -477,6 +477,17 @@ static int __init synth_event_gen_test_init(void) ret = test_trace_synth_event(); WARN_ON(ret); + + /* Disable when done */ + trace_array_set_clr_event(gen_synth_test->tr, + "synthetic", + "gen_synth_test", false); + trace_array_set_clr_event(empty_synth_test->tr, + "synthetic", + "empty_synth_test", false); + trace_array_set_clr_event(create_synth_test->tr, + "synthetic", + "create_synth_test", false); out: return ret; } diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 78ed5f1baa8c..e575956ef9b5 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -13,19 +13,21 @@ * Copyright (C) 2004 Nadia Yvette Chambers */ #include <linux/ring_buffer.h> -#include <generated/utsrelease.h> +#include <linux/utsname.h> #include <linux/stacktrace.h> #include <linux/writeback.h> #include <linux/kallsyms.h> #include <linux/security.h> #include <linux/seq_file.h> #include <linux/irqflags.h> +#include <linux/syscalls.h> #include <linux/debugfs.h> #include <linux/tracefs.h> #include <linux/pagemap.h> #include <linux/hardirq.h> #include <linux/linkage.h> #include <linux/uaccess.h> +#include <linux/cleanup.h> #include <linux/vmalloc.h> #include <linux/ftrace.h> #include <linux/module.h> @@ -48,16 +50,16 @@ #include <linux/fsnotify.h> #include <linux/irq_work.h> #include <linux/workqueue.h> +#include <linux/sort.h> +#include <linux/io.h> /* vmap_page_range() */ +#include <linux/fs_context.h> + +#include <asm/setup.h> /* COMMAND_LINE_SIZE */ #include "trace.h" #include "trace_output.h" -/* - * On boot up, the ring buffer is set to the minimum size, so that - * we do not waste memory on systems that are not using tracing. - */ -bool ring_buffer_expanded; - +#ifdef CONFIG_FTRACE_STARTUP_TEST /* * We need to change this state when a selftest is running. * A selftest will lurk into the ring-buffer to count the @@ -73,7 +75,6 @@ static bool __read_mostly tracing_selftest_running; */ bool __read_mostly tracing_selftest_disabled; -#ifdef CONFIG_FTRACE_STARTUP_TEST void __init disable_tracing_selftest(const char *reason) { if (!tracing_selftest_disabled) { @@ -81,31 +82,31 @@ void __init disable_tracing_selftest(const char *reason) pr_info("Ftrace startup test is disabled due to %s\n", reason); } } +#else +#define tracing_selftest_running 0 +#define tracing_selftest_disabled 0 #endif /* Pipe tracepoints to printk */ static struct trace_iterator *tracepoint_print_iter; int tracepoint_printk; static bool tracepoint_printk_stop_on_boot __initdata; +static bool traceoff_after_boot __initdata; static DEFINE_STATIC_KEY_FALSE(tracepoint_printk_key); -/* For tracers that don't implement custom flags */ -static struct tracer_opt dummy_tracer_opt[] = { - { } +/* Store tracers and their flags per instance */ +struct tracers { + struct list_head list; + struct tracer *tracer; + struct tracer_flags *flags; }; -static int -dummy_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set) -{ - return 0; -} - /* * To prevent the comm cache from being overwritten when no * tracing is active, only save the comm when a trace event * occurred. */ -static DEFINE_PER_CPU(bool, trace_taskinfo_save); +DEFINE_PER_CPU(bool, trace_taskinfo_save); /* * Kill all tracing for good (never come back). @@ -117,13 +118,14 @@ static int tracing_disabled = 1; cpumask_var_t __read_mostly tracing_buffer_mask; +#define MAX_TRACER_SIZE 100 /* * ftrace_dump_on_oops - variable to dump ftrace buffer on oops * * If there is an oops (or kernel panic) and the ftrace_dump_on_oops * is set, then ftrace_dump is called. This will output the contents * of the ftrace buffers to the console. This is very useful for - * capturing traces that lead to crashes and outputing it to a + * capturing traces that lead to crashes and outputting it to a * serial console. * * It is default off, but you can enable it with either specifying @@ -131,12 +133,48 @@ cpumask_var_t __read_mostly tracing_buffer_mask; * /proc/sys/kernel/ftrace_dump_on_oops * Set 1 if you want to dump buffers of all CPUs * Set 2 if you want to dump the buffer of the CPU that triggered oops + * Set instance name if you want to dump the specific trace instance + * Multiple instance dump is also supported, and instances are separated + * by commas. */ - -enum ftrace_dump_mode ftrace_dump_on_oops; +/* Set to string format zero to disable by default */ +char ftrace_dump_on_oops[MAX_TRACER_SIZE] = "0"; /* When set, tracing will stop when a WARN*() is hit */ -int __disable_trace_on_warning; +static int __disable_trace_on_warning; + +int tracepoint_printk_sysctl(const struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos); +static const struct ctl_table trace_sysctl_table[] = { + { + .procname = "ftrace_dump_on_oops", + .data = &ftrace_dump_on_oops, + .maxlen = MAX_TRACER_SIZE, + .mode = 0644, + .proc_handler = proc_dostring, + }, + { + .procname = "traceoff_on_warning", + .data = &__disable_trace_on_warning, + .maxlen = sizeof(__disable_trace_on_warning), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "tracepoint_printk", + .data = &tracepoint_printk, + .maxlen = sizeof(tracepoint_printk), + .mode = 0644, + .proc_handler = tracepoint_printk_sysctl, + }, +}; + +static int __init init_trace_sysctls(void) +{ + register_sysctl_init("kernel", trace_sysctl_table); + return 0; +} +subsys_initcall(init_trace_sysctls); #ifdef CONFIG_TRACE_EVAL_MAP_FILE /* Map of enums to their values, for "eval_map" file */ @@ -179,36 +217,55 @@ static void ftrace_trace_userstack(struct trace_array *tr, struct trace_buffer *buffer, unsigned int trace_ctx); -#define MAX_TRACER_SIZE 100 static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata; static char *default_bootup_tracer; static bool allocate_snapshot; static bool snapshot_at_boot; +static char boot_instance_info[COMMAND_LINE_SIZE] __initdata; +static int boot_instance_index; + +static char boot_snapshot_info[COMMAND_LINE_SIZE] __initdata; +static int boot_snapshot_index; + static int __init set_cmdline_ftrace(char *str) { - strlcpy(bootup_tracer_buf, str, MAX_TRACER_SIZE); + strscpy(bootup_tracer_buf, str, MAX_TRACER_SIZE); default_bootup_tracer = bootup_tracer_buf; /* We are using ftrace early, expand it */ - ring_buffer_expanded = true; + trace_set_ring_buffer_expanded(NULL); return 1; } __setup("ftrace=", set_cmdline_ftrace); +int ftrace_dump_on_oops_enabled(void) +{ + if (!strcmp("0", ftrace_dump_on_oops)) + return 0; + else + return 1; +} + static int __init set_ftrace_dump_on_oops(char *str) { - if (*str++ != '=' || !*str || !strcmp("1", str)) { - ftrace_dump_on_oops = DUMP_ALL; + if (!*str) { + strscpy(ftrace_dump_on_oops, "1", MAX_TRACER_SIZE); + return 1; + } + + if (*str == ',') { + strscpy(ftrace_dump_on_oops, "1", MAX_TRACER_SIZE); + strscpy(ftrace_dump_on_oops + 1, str, MAX_TRACER_SIZE - 1); return 1; } - if (!strcmp("orig_cpu", str) || !strcmp("2", str)) { - ftrace_dump_on_oops = DUMP_ORIG; - return 1; - } + if (*str++ == '=') { + strscpy(ftrace_dump_on_oops, str, MAX_TRACER_SIZE); + return 1; + } - return 0; + return 0; } __setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops); @@ -222,9 +279,22 @@ __setup("traceoff_on_warning", stop_trace_on_warning); static int __init boot_alloc_snapshot(char *str) { - allocate_snapshot = true; - /* We also need the main ring buffer expanded */ - ring_buffer_expanded = true; + char *slot = boot_snapshot_info + boot_snapshot_index; + int left = sizeof(boot_snapshot_info) - boot_snapshot_index; + int ret; + + if (str[0] == '=') { + str++; + if (strlen(str) >= left) + return -1; + + ret = snprintf(slot, left, "%s\t", str); + boot_snapshot_index += ret; + } else { + allocate_snapshot = true; + /* We also need the main ring buffer expanded */ + trace_set_ring_buffer_expanded(NULL); + } return 1; } __setup("alloc_snapshot", boot_alloc_snapshot); @@ -239,11 +309,28 @@ static int __init boot_snapshot(char *str) __setup("ftrace_boot_snapshot", boot_snapshot); +static int __init boot_instance(char *str) +{ + char *slot = boot_instance_info + boot_instance_index; + int left = sizeof(boot_instance_info) - boot_instance_index; + int ret; + + if (strlen(str) >= left) + return -1; + + ret = snprintf(slot, left, "%s\t", str); + boot_instance_index += ret; + + return 1; +} +__setup("trace_instance=", boot_instance); + + static char trace_boot_options_buf[MAX_TRACER_SIZE] __initdata; static int __init set_trace_boot_options(char *str) { - strlcpy(trace_boot_options_buf, str, MAX_TRACER_SIZE); + strscpy(trace_boot_options_buf, str, MAX_TRACER_SIZE); return 1; } __setup("trace_options=", set_trace_boot_options); @@ -253,7 +340,7 @@ static char *trace_boot_clock __initdata; static int __init set_trace_boot_clock(char *str) { - strlcpy(trace_boot_clock_buf, str, MAX_TRACER_SIZE); + strscpy(trace_boot_clock_buf, str, MAX_TRACER_SIZE); trace_boot_clock = trace_boot_clock_buf; return 1; } @@ -278,6 +365,13 @@ static int __init set_tracepoint_printk_stop(char *str) } __setup("tp_printk_stop_on_boot", set_tracepoint_printk_stop); +static int __init set_traceoff_after_boot(char *str) +{ + traceoff_after_boot = true; + return 1; +} +__setup("traceoff_after_boot", set_traceoff_after_boot); + unsigned long long ns2usecs(u64 nsec) { nsec += 500; @@ -335,15 +429,13 @@ static void ftrace_exports(struct ring_buffer_event *event, int flag) { struct trace_export *export; - preempt_disable_notrace(); + guard(preempt_notrace)(); export = rcu_dereference_raw_check(ftrace_exports_list); while (export) { trace_process_export(export, event, flag); export = rcu_dereference_raw_check(export->next); } - - preempt_enable_notrace(); } static inline void @@ -400,46 +492,40 @@ int register_ftrace_export(struct trace_export *export) if (WARN_ON_ONCE(!export->write)) return -1; - mutex_lock(&ftrace_export_lock); + guard(mutex)(&ftrace_export_lock); add_ftrace_export(&ftrace_exports_list, export); - mutex_unlock(&ftrace_export_lock); - return 0; } EXPORT_SYMBOL_GPL(register_ftrace_export); int unregister_ftrace_export(struct trace_export *export) { - int ret; - - mutex_lock(&ftrace_export_lock); - - ret = rm_ftrace_export(&ftrace_exports_list, export); - - mutex_unlock(&ftrace_export_lock); - - return ret; + guard(mutex)(&ftrace_export_lock); + return rm_ftrace_export(&ftrace_exports_list, export); } EXPORT_SYMBOL_GPL(unregister_ftrace_export); /* trace_flags holds trace_options default values */ #define TRACE_DEFAULT_FLAGS \ - (FUNCTION_DEFAULT_FLAGS | \ - TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | \ - TRACE_ITER_ANNOTATE | TRACE_ITER_CONTEXT_INFO | \ - TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE | \ - TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS | \ - TRACE_ITER_HASH_PTR) + (FUNCTION_DEFAULT_FLAGS | FPROFILE_DEFAULT_FLAGS | \ + TRACE_ITER(PRINT_PARENT) | TRACE_ITER(PRINTK) | \ + TRACE_ITER(ANNOTATE) | TRACE_ITER(CONTEXT_INFO) | \ + TRACE_ITER(RECORD_CMD) | TRACE_ITER(OVERWRITE) | \ + TRACE_ITER(IRQ_INFO) | TRACE_ITER(MARKERS) | \ + TRACE_ITER(HASH_PTR) | TRACE_ITER(TRACE_PRINTK) | \ + TRACE_ITER(COPY_MARKER)) /* trace_options that are only supported by global_trace */ -#define TOP_LEVEL_TRACE_FLAGS (TRACE_ITER_PRINTK | \ - TRACE_ITER_PRINTK_MSGONLY | TRACE_ITER_RECORD_CMD) +#define TOP_LEVEL_TRACE_FLAGS (TRACE_ITER(PRINTK) | \ + TRACE_ITER(PRINTK_MSGONLY) | TRACE_ITER(RECORD_CMD) | \ + TRACE_ITER(PROF_TEXT_OFFSET) | FPROFILE_DEFAULT_FLAGS) /* trace_flags that are default zero for instances */ #define ZEROED_TRACE_FLAGS \ - (TRACE_ITER_EVENT_FORK | TRACE_ITER_FUNC_FORK) + (TRACE_ITER(EVENT_FORK) | TRACE_ITER(FUNC_FORK) | TRACE_ITER(TRACE_PRINTK) | \ + TRACE_ITER(COPY_MARKER)) /* * The global_trace is the descriptor that holds the top-level tracing @@ -449,24 +535,76 @@ static struct trace_array global_trace = { .trace_flags = TRACE_DEFAULT_FLAGS, }; +static struct trace_array *printk_trace = &global_trace; + +/* List of trace_arrays interested in the top level trace_marker */ +static LIST_HEAD(marker_copies); + +static __always_inline bool printk_binsafe(struct trace_array *tr) +{ + /* + * The binary format of traceprintk can cause a crash if used + * by a buffer from another boot. Force the use of the + * non binary version of trace_printk if the trace_printk + * buffer is a boot mapped ring buffer. + */ + return !(tr->flags & TRACE_ARRAY_FL_BOOT); +} + +static void update_printk_trace(struct trace_array *tr) +{ + if (printk_trace == tr) + return; + + printk_trace->trace_flags &= ~TRACE_ITER(TRACE_PRINTK); + printk_trace = tr; + tr->trace_flags |= TRACE_ITER(TRACE_PRINTK); +} + +/* Returns true if the status of tr changed */ +static bool update_marker_trace(struct trace_array *tr, int enabled) +{ + lockdep_assert_held(&event_mutex); + + if (enabled) { + if (!list_empty(&tr->marker_list)) + return false; + + list_add_rcu(&tr->marker_list, &marker_copies); + tr->trace_flags |= TRACE_ITER(COPY_MARKER); + return true; + } + + if (list_empty(&tr->marker_list)) + return false; + + list_del_init(&tr->marker_list); + tr->trace_flags &= ~TRACE_ITER(COPY_MARKER); + return true; +} + +void trace_set_ring_buffer_expanded(struct trace_array *tr) +{ + if (!tr) + tr = &global_trace; + tr->ring_buffer_expanded = true; +} + LIST_HEAD(ftrace_trace_arrays); int trace_array_get(struct trace_array *this_tr) { struct trace_array *tr; - int ret = -ENODEV; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); list_for_each_entry(tr, &ftrace_trace_arrays, list) { if (tr == this_tr) { tr->ref++; - ret = 0; - break; + return 0; } } - mutex_unlock(&trace_types_lock); - return ret; + return -ENODEV; } static void __trace_array_put(struct trace_array *this_tr) @@ -489,9 +627,8 @@ void trace_array_put(struct trace_array *this_tr) if (!this_tr) return; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); __trace_array_put(this_tr); - mutex_unlock(&trace_types_lock); } EXPORT_SYMBOL_GPL(trace_array_put); @@ -512,19 +649,6 @@ int tracing_check_open_get_tr(struct trace_array *tr) return 0; } -int call_filter_check_discard(struct trace_event_call *call, void *rec, - struct trace_buffer *buffer, - struct ring_buffer_event *event) -{ - if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) && - !filter_match_preds(call->filter, rec)) { - __trace_event_discard_commit(buffer, event); - return 1; - } - - return 0; -} - /** * trace_find_filtered_pid - check if a pid exists in a filtered_pid list * @filtered_pids: The list of pids to check @@ -708,7 +832,10 @@ int trace_pid_write(struct trace_pid_list *filtered_pids, /* copy the current bits to the new max */ ret = trace_pid_list_first(filtered_pids, &pid); while (!ret) { - trace_pid_list_set(pid_list, pid); + ret = trace_pid_list_set(pid_list, pid); + if (ret < 0) + goto out; + ret = trace_pid_list_next(filtered_pids, pid + 1, &pid); nr_pids++; } @@ -745,6 +872,7 @@ int trace_pid_write(struct trace_pid_list *filtered_pids, trace_parser_clear(&parser); ret = 0; } + out: trace_parser_put(&parser); if (ret < 0) { @@ -798,7 +926,6 @@ int tracing_is_enabled(void) * return the mirror variable of the state of the ring buffer. * It's a little racy, but we don't really care. */ - smp_rmb(); return !global_trace.buffer_disabled; } @@ -907,7 +1034,8 @@ static inline void trace_access_lock_init(void) #endif #ifdef CONFIG_STACKTRACE -static void __ftrace_trace_stack(struct trace_buffer *buffer, +static void __ftrace_trace_stack(struct trace_array *tr, + struct trace_buffer *buffer, unsigned int trace_ctx, int skip, struct pt_regs *regs); static inline void ftrace_trace_stack(struct trace_array *tr, @@ -916,7 +1044,8 @@ static inline void ftrace_trace_stack(struct trace_array *tr, int skip, struct pt_regs *regs); #else -static inline void __ftrace_trace_stack(struct trace_buffer *buffer, +static inline void __ftrace_trace_stack(struct trace_array *tr, + struct trace_buffer *buffer, unsigned int trace_ctx, int skip, struct pt_regs *regs) { @@ -967,8 +1096,6 @@ void tracer_tracing_on(struct trace_array *tr) * important to be fast than accurate. */ tr->buffer_disabled = 0; - /* Make the flag seen by readers */ - smp_wmb(); } /** @@ -1001,13 +1128,8 @@ __buffer_unlock_commit(struct trace_buffer *buffer, struct ring_buffer_event *ev ring_buffer_unlock_commit(buffer); } -/** - * __trace_puts - write a constant string into the trace buffer. - * @ip: The address of the caller - * @str: The constant string to write - * @size: The size of the string. - */ -int __trace_puts(unsigned long ip, const char *str, int size) +int __trace_array_puts(struct trace_array *tr, unsigned long ip, + const char *str, int size) { struct ring_buffer_event *event; struct trace_buffer *buffer; @@ -1015,23 +1137,24 @@ int __trace_puts(unsigned long ip, const char *str, int size) unsigned int trace_ctx; int alloc; - if (!(global_trace.trace_flags & TRACE_ITER_PRINTK)) + if (!(tr->trace_flags & TRACE_ITER(PRINTK))) return 0; - if (unlikely(tracing_selftest_running || tracing_disabled)) + if (unlikely(tracing_selftest_running && tr == &global_trace)) + return 0; + + if (unlikely(tracing_disabled)) return 0; alloc = sizeof(*entry) + size + 2; /* possible \n added */ trace_ctx = tracing_gen_ctx(); - buffer = global_trace.array_buffer.buffer; - ring_buffer_nest_start(buffer); + buffer = tr->array_buffer.buffer; + guard(ring_buffer_nest)(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, alloc, trace_ctx); - if (!event) { - size = 0; - goto out; - } + if (!event) + return 0; entry = ring_buffer_event_data(event); entry->ip = ip; @@ -1046,11 +1169,21 @@ int __trace_puts(unsigned long ip, const char *str, int size) entry->buf[size] = '\0'; __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(&global_trace, buffer, trace_ctx, 4, NULL); - out: - ring_buffer_nest_end(buffer); + ftrace_trace_stack(tr, buffer, trace_ctx, 4, NULL); return size; } +EXPORT_SYMBOL_GPL(__trace_array_puts); + +/** + * __trace_puts - write a constant string into the trace buffer. + * @ip: The address of the caller + * @str: The constant string to write + * @size: The size of the string. + */ +int __trace_puts(unsigned long ip, const char *str, int size) +{ + return __trace_array_puts(printk_trace, ip, str, size); +} EXPORT_SYMBOL_GPL(__trace_puts); /** @@ -1060,39 +1193,39 @@ EXPORT_SYMBOL_GPL(__trace_puts); */ int __trace_bputs(unsigned long ip, const char *str) { + struct trace_array *tr = READ_ONCE(printk_trace); struct ring_buffer_event *event; struct trace_buffer *buffer; struct bputs_entry *entry; unsigned int trace_ctx; int size = sizeof(struct bputs_entry); - int ret = 0; - if (!(global_trace.trace_flags & TRACE_ITER_PRINTK)) + if (!printk_binsafe(tr)) + return __trace_puts(ip, str, strlen(str)); + + if (!(tr->trace_flags & TRACE_ITER(PRINTK))) return 0; if (unlikely(tracing_selftest_running || tracing_disabled)) return 0; trace_ctx = tracing_gen_ctx(); - buffer = global_trace.array_buffer.buffer; + buffer = tr->array_buffer.buffer; - ring_buffer_nest_start(buffer); + guard(ring_buffer_nest)(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_BPUTS, size, trace_ctx); if (!event) - goto out; + return 0; entry = ring_buffer_event_data(event); entry->ip = ip; entry->str = str; __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(&global_trace, buffer, trace_ctx, 4, NULL); + ftrace_trace_stack(tr, buffer, trace_ctx, 4, NULL); - ret = 1; - out: - ring_buffer_nest_end(buffer); - return ret; + return 1; } EXPORT_SYMBOL_GPL(__trace_bputs); @@ -1104,22 +1237,28 @@ static void tracing_snapshot_instance_cond(struct trace_array *tr, unsigned long flags; if (in_nmi()) { - internal_trace_puts("*** SNAPSHOT CALLED FROM NMI CONTEXT ***\n"); - internal_trace_puts("*** snapshot is being ignored ***\n"); + trace_array_puts(tr, "*** SNAPSHOT CALLED FROM NMI CONTEXT ***\n"); + trace_array_puts(tr, "*** snapshot is being ignored ***\n"); return; } if (!tr->allocated_snapshot) { - internal_trace_puts("*** SNAPSHOT NOT ALLOCATED ***\n"); - internal_trace_puts("*** stopping trace here! ***\n"); - tracing_off(); + trace_array_puts(tr, "*** SNAPSHOT NOT ALLOCATED ***\n"); + trace_array_puts(tr, "*** stopping trace here! ***\n"); + tracer_tracing_off(tr); return; } /* Note, snapshot can not be used when the tracer uses it */ if (tracer->use_max_tr) { - internal_trace_puts("*** LATENCY TRACER ACTIVE ***\n"); - internal_trace_puts("*** Can not use snapshot (sorry) ***\n"); + trace_array_puts(tr, "*** LATENCY TRACER ACTIVE ***\n"); + trace_array_puts(tr, "*** Can not use snapshot (sorry) ***\n"); + return; + } + + if (tr->mapped) { + trace_array_puts(tr, "*** BUFFER MEMORY MAPPED ***\n"); + trace_array_puts(tr, "*** Can not use snapshot (sorry) ***\n"); return; } @@ -1142,7 +1281,7 @@ void tracing_snapshot_instance(struct trace_array *tr) * * Note, make sure to allocate the snapshot with either * a tracing_snapshot_alloc(), or by doing it manually - * with: echo 1 > /sys/kernel/debug/tracing/snapshot + * with: echo 1 > /sys/kernel/tracing/snapshot * * If the snapshot buffer is not allocated, it will stop tracing. * Basically making a permanent snapshot. @@ -1211,10 +1350,17 @@ static void set_buffer_entries(struct array_buffer *buf, unsigned long val); int tracing_alloc_snapshot_instance(struct trace_array *tr) { + int order; int ret; if (!tr->allocated_snapshot) { + /* Make the snapshot buffer have the same order as main buffer */ + order = ring_buffer_subbuf_order_get(tr->array_buffer.buffer); + ret = ring_buffer_subbuf_order_set(tr->max_buffer.buffer, order); + if (ret < 0) + return ret; + /* allocate spare buffer */ ret = resize_buffer_duplicate_size(&tr->max_buffer, &tr->array_buffer, RING_BUFFER_ALL_CPUS); @@ -1234,12 +1380,52 @@ static void free_snapshot(struct trace_array *tr) * The max_tr ring buffer has some state (e.g. ring->clock) and * we want preserve it. */ + ring_buffer_subbuf_order_set(tr->max_buffer.buffer, 0); ring_buffer_resize(tr->max_buffer.buffer, 1, RING_BUFFER_ALL_CPUS); set_buffer_entries(&tr->max_buffer, 1); tracing_reset_online_cpus(&tr->max_buffer); tr->allocated_snapshot = false; } +static int tracing_arm_snapshot_locked(struct trace_array *tr) +{ + int ret; + + lockdep_assert_held(&trace_types_lock); + + spin_lock(&tr->snapshot_trigger_lock); + if (tr->snapshot == UINT_MAX || tr->mapped) { + spin_unlock(&tr->snapshot_trigger_lock); + return -EBUSY; + } + + tr->snapshot++; + spin_unlock(&tr->snapshot_trigger_lock); + + ret = tracing_alloc_snapshot_instance(tr); + if (ret) { + spin_lock(&tr->snapshot_trigger_lock); + tr->snapshot--; + spin_unlock(&tr->snapshot_trigger_lock); + } + + return ret; +} + +int tracing_arm_snapshot(struct trace_array *tr) +{ + guard(mutex)(&trace_types_lock); + return tracing_arm_snapshot_locked(tr); +} + +void tracing_disarm_snapshot(struct trace_array *tr) +{ + spin_lock(&tr->snapshot_trigger_lock); + if (!WARN_ON(!tr->snapshot)) + tr->snapshot--; + spin_unlock(&tr->snapshot_trigger_lock); +} + /** * tracing_alloc_snapshot - allocate snapshot buffer. * @@ -1301,26 +1487,20 @@ EXPORT_SYMBOL_GPL(tracing_snapshot_alloc); int tracing_snapshot_cond_enable(struct trace_array *tr, void *cond_data, cond_update_fn_t update) { - struct cond_snapshot *cond_snapshot; - int ret = 0; + struct cond_snapshot *cond_snapshot __free(kfree) = + kzalloc(sizeof(*cond_snapshot), GFP_KERNEL); + int ret; - cond_snapshot = kzalloc(sizeof(*cond_snapshot), GFP_KERNEL); if (!cond_snapshot) return -ENOMEM; cond_snapshot->cond_data = cond_data; cond_snapshot->update = update; - mutex_lock(&trace_types_lock); - - ret = tracing_alloc_snapshot_instance(tr); - if (ret) - goto fail_unlock; + guard(mutex)(&trace_types_lock); - if (tr->current_trace->use_max_tr) { - ret = -EBUSY; - goto fail_unlock; - } + if (tr->current_trace->use_max_tr) + return -EBUSY; /* * The cond_snapshot can only change to NULL without the @@ -1330,25 +1510,20 @@ int tracing_snapshot_cond_enable(struct trace_array *tr, void *cond_data, * do safely with only holding the trace_types_lock and not * having to take the max_lock. */ - if (tr->cond_snapshot) { - ret = -EBUSY; - goto fail_unlock; - } + if (tr->cond_snapshot) + return -EBUSY; + + ret = tracing_arm_snapshot_locked(tr); + if (ret) + return ret; local_irq_disable(); arch_spin_lock(&tr->max_lock); - tr->cond_snapshot = cond_snapshot; + tr->cond_snapshot = no_free_ptr(cond_snapshot); arch_spin_unlock(&tr->max_lock); local_irq_enable(); - mutex_unlock(&trace_types_lock); - - return ret; - - fail_unlock: - mutex_unlock(&trace_types_lock); - kfree(cond_snapshot); - return ret; + return 0; } EXPORT_SYMBOL_GPL(tracing_snapshot_cond_enable); @@ -1379,6 +1554,8 @@ int tracing_snapshot_cond_disable(struct trace_array *tr) arch_spin_unlock(&tr->max_lock); local_irq_enable(); + tracing_disarm_snapshot(tr); + return ret; } EXPORT_SYMBOL_GPL(tracing_snapshot_cond_disable); @@ -1421,6 +1598,7 @@ int tracing_snapshot_cond_disable(struct trace_array *tr) } EXPORT_SYMBOL_GPL(tracing_snapshot_cond_disable); #define free_snapshot(tr) do { } while (0) +#define tracing_arm_snapshot_locked(tr) ({ -EBUSY; }) #endif /* CONFIG_TRACER_SNAPSHOT */ void tracer_tracing_off(struct trace_array *tr) @@ -1436,8 +1614,39 @@ void tracer_tracing_off(struct trace_array *tr) * important to be fast than accurate. */ tr->buffer_disabled = 1; - /* Make the flag seen by readers */ - smp_wmb(); +} + +/** + * tracer_tracing_disable() - temporary disable the buffer from write + * @tr: The trace array to disable its buffer for + * + * Expects trace_tracing_enable() to re-enable tracing. + * The difference between this and tracer_tracing_off() is that this + * is a counter and can nest, whereas, tracer_tracing_off() can + * be called multiple times and a single trace_tracing_on() will + * enable it. + */ +void tracer_tracing_disable(struct trace_array *tr) +{ + if (WARN_ON_ONCE(!tr->array_buffer.buffer)) + return; + + ring_buffer_record_disable(tr->array_buffer.buffer); +} + +/** + * tracer_tracing_enable() - counter part of tracer_tracing_disable() + * @tr: The trace array that had tracer_tracincg_disable() called on it + * + * This is called after tracer_tracing_disable() has been called on @tr, + * when it's safe to re-enable tracing. + */ +void tracer_tracing_enable(struct trace_array *tr) +{ + if (WARN_ON_ONCE(!tr->array_buffer.buffer)) + return; + + ring_buffer_record_enable(tr->array_buffer.buffer); } /** @@ -1472,7 +1681,7 @@ void disable_trace_on_warning(void) bool tracer_tracing_is_on(struct trace_array *tr) { if (tr->array_buffer.buffer) - return ring_buffer_record_is_on(tr->array_buffer.buffer); + return ring_buffer_record_is_set_on(tr->array_buffer.buffer); return !tr->buffer_disabled; } @@ -1609,7 +1818,7 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf, ret = get_user(ch, ubuf++); if (ret) - goto out; + goto fail; read++; cnt--; @@ -1623,7 +1832,7 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf, while (cnt && isspace(ch)) { ret = get_user(ch, ubuf++); if (ret) - goto out; + goto fail; read++; cnt--; } @@ -1633,8 +1842,7 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf, /* only spaces were written */ if (isspace(ch) || !ch) { *ppos += read; - ret = read; - goto out; + return read; } } @@ -1644,11 +1852,12 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf, parser->buffer[parser->idx++] = ch; else { ret = -EINVAL; - goto out; + goto fail; } + ret = get_user(ch, ubuf++); if (ret) - goto out; + goto fail; read++; cnt--; } @@ -1664,13 +1873,13 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf, parser->buffer[parser->idx] = 0; } else { ret = -EINVAL; - goto out; + goto fail; } *ppos += read; - ret = read; - -out: + return read; +fail: + trace_parser_fail(parser); return ret; } @@ -1679,15 +1888,15 @@ static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt) { int len; - if (trace_seq_used(s) <= s->seq.readpos) + if (trace_seq_used(s) <= s->readpos) return -EBUSY; - len = trace_seq_used(s) - s->seq.readpos; + len = trace_seq_used(s) - s->readpos; if (cnt > len) cnt = len; - memcpy(buf, s->buffer + s->seq.readpos, cnt); + memcpy(buf, s->buffer + s->readpos, cnt); - s->seq.readpos += cnt; + s->readpos += cnt; return cnt; } @@ -1721,7 +1930,7 @@ static void trace_create_maxlat_file(struct trace_array *tr, init_irq_work(&tr->fsnotify_irqwork, latency_fsnotify_workfn_irq); tr->d_max_latency = trace_create_file("tracing_max_latency", TRACE_MODE_WRITE, - d_tracer, &tr->max_latency, + d_tracer, tr, &tracing_max_lat_fops); } @@ -1754,7 +1963,7 @@ void latency_fsnotify(struct trace_array *tr) #define trace_create_maxlat_file(tr, d_tracer) \ trace_create_file("tracing_max_latency", TRACE_MODE_WRITE, \ - d_tracer, &tr->max_latency, &tracing_max_lat_fops) + d_tracer, tr, &tracing_max_lat_fops) #endif @@ -1778,7 +1987,7 @@ __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) max_data->critical_start = data->critical_start; max_data->critical_end = data->critical_end; - strncpy(max_data->comm, tsk->comm, TASK_COMM_LEN); + strscpy(max_data->comm, tsk->comm); max_data->pid = tsk->pid; /* * If tsk == current, then use current_uid(), as that does not use @@ -1842,6 +2051,9 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu, __update_max_tr(tr, tsk, cpu); arch_spin_unlock(&tr->max_lock); + + /* Any waiters on the old snapshot buffer need to wake up */ + ring_buffer_wake_waiters(tr->array_buffer.buffer, RING_BUFFER_ALL_CPUS); } /** @@ -1877,9 +2089,10 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) * place on this CPU. We fail to record, but we reset * the max trace buffer (no one writes directly to it) * and flag that it failed. + * Another reason is resize is in progress. */ trace_array_printk_buf(tr->max_buffer.buffer, _THIS_IP_, - "Failed to swap buffers due to commit in progress\n"); + "Failed to swap buffers due to commit or resize in progress\n"); } WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY); @@ -1890,14 +2103,46 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) #endif /* CONFIG_TRACER_MAX_TRACE */ +struct pipe_wait { + struct trace_iterator *iter; + int wait_index; +}; + +static bool wait_pipe_cond(void *data) +{ + struct pipe_wait *pwait = data; + struct trace_iterator *iter = pwait->iter; + + if (atomic_read_acquire(&iter->wait_index) != pwait->wait_index) + return true; + + return iter->closed; +} + static int wait_on_pipe(struct trace_iterator *iter, int full) { + struct pipe_wait pwait; + int ret; + /* Iterators are static, they should be filled or empty */ if (trace_buffer_iter(iter, iter->cpu_file)) return 0; - return ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, - full); + pwait.wait_index = atomic_read_acquire(&iter->wait_index); + pwait.iter = iter; + + ret = ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, full, + wait_pipe_cond, &pwait); + +#ifdef CONFIG_TRACER_MAX_TRACE + /* + * Make sure this is still the snapshot buffer, as if a snapshot were + * to happen, this would now be the main buffer. + */ + if (iter->snapshot) + iter->array_buffer = &iter->tr->max_buffer; +#endif + return ret; } #ifdef CONFIG_FTRACE_STARTUP_TEST @@ -1926,6 +2171,7 @@ static int save_selftest(struct tracer *type) static int run_tracer_selftest(struct tracer *type) { struct trace_array *tr = &global_trace; + struct tracer_flags *saved_flags = tr->current_trace_flags; struct tracer *saved_tracer = tr->current_trace; int ret; @@ -1956,11 +2202,12 @@ static int run_tracer_selftest(struct tracer *type) tracing_reset_online_cpus(&tr->array_buffer); tr->current_trace = type; + tr->current_trace_flags = type->flags ? : type->default_flags; #ifdef CONFIG_TRACER_MAX_TRACE if (type->use_max_tr) { /* If we expanded the buffers, make sure the max is expanded too */ - if (ring_buffer_expanded) + if (tr->ring_buffer_expanded) ring_buffer_resize(tr->max_buffer.buffer, trace_buf_size, RING_BUFFER_ALL_CPUS); tr->allocated_snapshot = true; @@ -1972,6 +2219,7 @@ static int run_tracer_selftest(struct tracer *type) ret = type->selftest(type, tr); /* the test is responsible for resetting too */ tr->current_trace = saved_tracer; + tr->current_trace_flags = saved_flags; if (ret) { printk(KERN_CONT "FAILED!\n"); /* Add the warning after printing 'FAILED' */ @@ -1986,7 +2234,7 @@ static int run_tracer_selftest(struct tracer *type) tr->allocated_snapshot = false; /* Shrink the max buffer again */ - if (ring_buffer_expanded) + if (tr->ring_buffer_expanded) ring_buffer_resize(tr->max_buffer.buffer, 1, RING_BUFFER_ALL_CPUS); } @@ -1996,6 +2244,24 @@ static int run_tracer_selftest(struct tracer *type) return 0; } +static int do_run_tracer_selftest(struct tracer *type) +{ + int ret; + + /* + * Tests can take a long time, especially if they are run one after the + * other, as does happen during bootup when all the tracers are + * registered. This could cause the soft lockup watchdog to trigger. + */ + cond_resched(); + + tracing_selftest_running = true; + ret = run_tracer_selftest(type); + tracing_selftest_running = false; + + return ret; +} + static __init int init_trace_selftests(void) { struct trace_selftests *p, *n; @@ -2004,10 +2270,10 @@ static __init int init_trace_selftests(void) selftests_can_run = true; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); if (list_empty(&postponed_selftests)) - goto out; + return 0; pr_info("Running postponed tracer tests:\n"); @@ -2036,23 +2302,33 @@ static __init int init_trace_selftests(void) } tracing_selftest_running = false; - out: - mutex_unlock(&trace_types_lock); - return 0; } core_initcall(init_trace_selftests); #else -static inline int run_tracer_selftest(struct tracer *type) +static inline int do_run_tracer_selftest(struct tracer *type) { return 0; } #endif /* CONFIG_FTRACE_STARTUP_TEST */ -static void add_tracer_options(struct trace_array *tr, struct tracer *t); +static int add_tracer(struct trace_array *tr, struct tracer *t); static void __init apply_trace_boot_options(void); +static void free_tracers(struct trace_array *tr) +{ + struct tracers *t, *n; + + lockdep_assert_held(&trace_types_lock); + + list_for_each_entry_safe(t, n, &tr->tracers, list) { + list_del(&t->list); + kfree(t->flags); + kfree(t); + } +} + /** * register_tracer - register a tracer with the ftrace system. * @type: the plugin for the tracer @@ -2061,6 +2337,7 @@ static void __init apply_trace_boot_options(void); */ int __init register_tracer(struct tracer *type) { + struct trace_array *tr; struct tracer *t; int ret = 0; @@ -2082,8 +2359,6 @@ int __init register_tracer(struct tracer *type) mutex_lock(&trace_types_lock); - tracing_selftest_running = true; - for (t = trace_types; t; t = t->next) { if (strcmp(type->name, t->name) == 0) { /* already found */ @@ -2094,45 +2369,38 @@ int __init register_tracer(struct tracer *type) } } - if (!type->set_flag) - type->set_flag = &dummy_set_flag; - if (!type->flags) { - /*allocate a dummy tracer_flags*/ - type->flags = kmalloc(sizeof(*type->flags), GFP_KERNEL); - if (!type->flags) { - ret = -ENOMEM; - goto out; - } - type->flags->val = 0; - type->flags->opts = dummy_tracer_opt; - } else - if (!type->flags->opts) - type->flags->opts = dummy_tracer_opt; - /* store the tracer for __set_tracer_option */ - type->flags->trace = type; + if (type->flags) + type->flags->trace = type; - ret = run_tracer_selftest(type); + ret = do_run_tracer_selftest(type); if (ret < 0) goto out; + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + ret = add_tracer(tr, type); + if (ret < 0) { + /* The tracer will still exist but without options */ + pr_warn("Failed to create tracer options for %s\n", type->name); + break; + } + } + type->next = trace_types; trace_types = type; - add_tracer_options(&global_trace, type); out: - tracing_selftest_running = false; mutex_unlock(&trace_types_lock); if (ret || !default_bootup_tracer) - goto out_unlock; + return ret; if (strncmp(default_bootup_tracer, type->name, MAX_TRACER_SIZE)) - goto out_unlock; + return 0; printk(KERN_INFO "Starting tracer '%s'\n", type->name); /* Do we want this tracer to start on bootup? */ - tracing_set_tracer(&global_trace, type->name); + WARN_ON(tracing_set_tracer(&global_trace, type->name) < 0); default_bootup_tracer = NULL; apply_trace_boot_options(); @@ -2140,8 +2408,7 @@ int __init register_tracer(struct tracer *type) /* disable other selftests, since this will break it. */ disable_tracing_selftest("running a tracer"); - out_unlock: - return ret; + return 0; } static void tracing_reset_cpu(struct array_buffer *buf, int cpu) @@ -2179,6 +2446,25 @@ void tracing_reset_online_cpus(struct array_buffer *buf) ring_buffer_record_enable(buffer); } +static void tracing_reset_all_cpus(struct array_buffer *buf) +{ + struct trace_buffer *buffer = buf->buffer; + + if (!buffer) + return; + + ring_buffer_record_disable(buffer); + + /* Make sure all commits have finished */ + synchronize_rcu(); + + buf->time_start = buffer_ftrace_now(buf, buf->cpu); + + ring_buffer_reset(buffer); + + ring_buffer_record_enable(buffer); +} + /* Must have trace_types_lock held */ void tracing_reset_all_online_cpus_unlocked(void) { @@ -2199,88 +2485,8 @@ void tracing_reset_all_online_cpus_unlocked(void) void tracing_reset_all_online_cpus(void) { - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); tracing_reset_all_online_cpus_unlocked(); - mutex_unlock(&trace_types_lock); -} - -/* - * The tgid_map array maps from pid to tgid; i.e. the value stored at index i - * is the tgid last observed corresponding to pid=i. - */ -static int *tgid_map; - -/* The maximum valid index into tgid_map. */ -static size_t tgid_map_max; - -#define SAVED_CMDLINES_DEFAULT 128 -#define NO_CMDLINE_MAP UINT_MAX -/* - * Preemption must be disabled before acquiring trace_cmdline_lock. - * The various trace_arrays' max_lock must be acquired in a context - * where interrupt is disabled. - */ -static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; -struct saved_cmdlines_buffer { - unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; - unsigned *map_cmdline_to_pid; - unsigned cmdline_num; - int cmdline_idx; - char *saved_cmdlines; -}; -static struct saved_cmdlines_buffer *savedcmd; - -static inline char *get_saved_cmdlines(int idx) -{ - return &savedcmd->saved_cmdlines[idx * TASK_COMM_LEN]; -} - -static inline void set_cmdline(int idx, const char *cmdline) -{ - strncpy(get_saved_cmdlines(idx), cmdline, TASK_COMM_LEN); -} - -static int allocate_cmdlines_buffer(unsigned int val, - struct saved_cmdlines_buffer *s) -{ - s->map_cmdline_to_pid = kmalloc_array(val, - sizeof(*s->map_cmdline_to_pid), - GFP_KERNEL); - if (!s->map_cmdline_to_pid) - return -ENOMEM; - - s->saved_cmdlines = kmalloc_array(TASK_COMM_LEN, val, GFP_KERNEL); - if (!s->saved_cmdlines) { - kfree(s->map_cmdline_to_pid); - return -ENOMEM; - } - - s->cmdline_idx = 0; - s->cmdline_num = val; - memset(&s->map_pid_to_cmdline, NO_CMDLINE_MAP, - sizeof(s->map_pid_to_cmdline)); - memset(s->map_cmdline_to_pid, NO_CMDLINE_MAP, - val * sizeof(*s->map_cmdline_to_pid)); - - return 0; -} - -static int trace_create_savedcmd(void) -{ - int ret; - - savedcmd = kmalloc(sizeof(*savedcmd), GFP_KERNEL); - if (!savedcmd) - return -ENOMEM; - - ret = allocate_cmdlines_buffer(SAVED_CMDLINES_DEFAULT, savedcmd); - if (ret < 0) { - kfree(savedcmd); - savedcmd = NULL; - return -ENOMEM; - } - - return 0; } int is_tracing_stopped(void) @@ -2288,328 +2494,83 @@ int is_tracing_stopped(void) return global_trace.stop_count; } -/** - * tracing_start - quick start of the tracer - * - * If tracing is enabled but was stopped by tracing_stop, - * this will start the tracer back up. - */ -void tracing_start(void) +static void tracing_start_tr(struct trace_array *tr) { struct trace_buffer *buffer; - unsigned long flags; if (tracing_disabled) return; - raw_spin_lock_irqsave(&global_trace.start_lock, flags); - if (--global_trace.stop_count) { - if (global_trace.stop_count < 0) { + guard(raw_spinlock_irqsave)(&tr->start_lock); + if (--tr->stop_count) { + if (WARN_ON_ONCE(tr->stop_count < 0)) { /* Someone screwed up their debugging */ - WARN_ON_ONCE(1); - global_trace.stop_count = 0; + tr->stop_count = 0; } - goto out; + return; } /* Prevent the buffers from switching */ - arch_spin_lock(&global_trace.max_lock); + arch_spin_lock(&tr->max_lock); - buffer = global_trace.array_buffer.buffer; + buffer = tr->array_buffer.buffer; if (buffer) ring_buffer_record_enable(buffer); #ifdef CONFIG_TRACER_MAX_TRACE - buffer = global_trace.max_buffer.buffer; + buffer = tr->max_buffer.buffer; if (buffer) ring_buffer_record_enable(buffer); #endif - arch_spin_unlock(&global_trace.max_lock); - - out: - raw_spin_unlock_irqrestore(&global_trace.start_lock, flags); -} - -static void tracing_start_tr(struct trace_array *tr) -{ - struct trace_buffer *buffer; - unsigned long flags; - - if (tracing_disabled) - return; - - /* If global, we need to also start the max tracer */ - if (tr->flags & TRACE_ARRAY_FL_GLOBAL) - return tracing_start(); - - raw_spin_lock_irqsave(&tr->start_lock, flags); - - if (--tr->stop_count) { - if (tr->stop_count < 0) { - /* Someone screwed up their debugging */ - WARN_ON_ONCE(1); - tr->stop_count = 0; - } - goto out; - } - - buffer = tr->array_buffer.buffer; - if (buffer) - ring_buffer_record_enable(buffer); - - out: - raw_spin_unlock_irqrestore(&tr->start_lock, flags); + arch_spin_unlock(&tr->max_lock); } /** - * tracing_stop - quick stop of the tracer + * tracing_start - quick start of the tracer * - * Light weight way to stop tracing. Use in conjunction with - * tracing_start. + * If tracing is enabled but was stopped by tracing_stop, + * this will start the tracer back up. */ -void tracing_stop(void) -{ - struct trace_buffer *buffer; - unsigned long flags; - - raw_spin_lock_irqsave(&global_trace.start_lock, flags); - if (global_trace.stop_count++) - goto out; - - /* Prevent the buffers from switching */ - arch_spin_lock(&global_trace.max_lock); - - buffer = global_trace.array_buffer.buffer; - if (buffer) - ring_buffer_record_disable(buffer); - -#ifdef CONFIG_TRACER_MAX_TRACE - buffer = global_trace.max_buffer.buffer; - if (buffer) - ring_buffer_record_disable(buffer); -#endif - - arch_spin_unlock(&global_trace.max_lock); +void tracing_start(void) - out: - raw_spin_unlock_irqrestore(&global_trace.start_lock, flags); +{ + return tracing_start_tr(&global_trace); } static void tracing_stop_tr(struct trace_array *tr) { struct trace_buffer *buffer; - unsigned long flags; - - /* If global, we need to also stop the max tracer */ - if (tr->flags & TRACE_ARRAY_FL_GLOBAL) - return tracing_stop(); - raw_spin_lock_irqsave(&tr->start_lock, flags); + guard(raw_spinlock_irqsave)(&tr->start_lock); if (tr->stop_count++) - goto out; + return; + + /* Prevent the buffers from switching */ + arch_spin_lock(&tr->max_lock); buffer = tr->array_buffer.buffer; if (buffer) ring_buffer_record_disable(buffer); - out: - raw_spin_unlock_irqrestore(&tr->start_lock, flags); -} - -static int trace_save_cmdline(struct task_struct *tsk) -{ - unsigned tpid, idx; - - /* treat recording of idle task as a success */ - if (!tsk->pid) - return 1; - - tpid = tsk->pid & (PID_MAX_DEFAULT - 1); - - /* - * It's not the end of the world if we don't get - * the lock, but we also don't want to spin - * nor do we want to disable interrupts, - * so if we miss here, then better luck next time. - * - * This is called within the scheduler and wake up, so interrupts - * had better been disabled and run queue lock been held. - */ - lockdep_assert_preemption_disabled(); - if (!arch_spin_trylock(&trace_cmdline_lock)) - return 0; - - idx = savedcmd->map_pid_to_cmdline[tpid]; - if (idx == NO_CMDLINE_MAP) { - idx = (savedcmd->cmdline_idx + 1) % savedcmd->cmdline_num; - - savedcmd->map_pid_to_cmdline[tpid] = idx; - savedcmd->cmdline_idx = idx; - } - - savedcmd->map_cmdline_to_pid[idx] = tsk->pid; - set_cmdline(idx, tsk->comm); - - arch_spin_unlock(&trace_cmdline_lock); - - return 1; -} - -static void __trace_find_cmdline(int pid, char comm[]) -{ - unsigned map; - int tpid; - - if (!pid) { - strcpy(comm, "<idle>"); - return; - } - - if (WARN_ON_ONCE(pid < 0)) { - strcpy(comm, "<XXX>"); - return; - } - - tpid = pid & (PID_MAX_DEFAULT - 1); - map = savedcmd->map_pid_to_cmdline[tpid]; - if (map != NO_CMDLINE_MAP) { - tpid = savedcmd->map_cmdline_to_pid[map]; - if (tpid == pid) { - strlcpy(comm, get_saved_cmdlines(map), TASK_COMM_LEN); - return; - } - } - strcpy(comm, "<...>"); -} - -void trace_find_cmdline(int pid, char comm[]) -{ - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - - __trace_find_cmdline(pid, comm); - - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); -} - -static int *trace_find_tgid_ptr(int pid) -{ - /* - * Pairs with the smp_store_release in set_tracer_flag() to ensure that - * if we observe a non-NULL tgid_map then we also observe the correct - * tgid_map_max. - */ - int *map = smp_load_acquire(&tgid_map); - - if (unlikely(!map || pid > tgid_map_max)) - return NULL; - - return &map[pid]; -} - -int trace_find_tgid(int pid) -{ - int *ptr = trace_find_tgid_ptr(pid); - - return ptr ? *ptr : 0; -} - -static int trace_save_tgid(struct task_struct *tsk) -{ - int *ptr; - - /* treat recording of idle task as a success */ - if (!tsk->pid) - return 1; - - ptr = trace_find_tgid_ptr(tsk->pid); - if (!ptr) - return 0; - - *ptr = tsk->tgid; - return 1; -} - -static bool tracing_record_taskinfo_skip(int flags) -{ - if (unlikely(!(flags & (TRACE_RECORD_CMDLINE | TRACE_RECORD_TGID)))) - return true; - if (!__this_cpu_read(trace_taskinfo_save)) - return true; - return false; -} - -/** - * tracing_record_taskinfo - record the task info of a task - * - * @task: task to record - * @flags: TRACE_RECORD_CMDLINE for recording comm - * TRACE_RECORD_TGID for recording tgid - */ -void tracing_record_taskinfo(struct task_struct *task, int flags) -{ - bool done; - - if (tracing_record_taskinfo_skip(flags)) - return; - - /* - * Record as much task information as possible. If some fail, continue - * to try to record the others. - */ - done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(task); - done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(task); - - /* If recording any information failed, retry again soon. */ - if (!done) - return; +#ifdef CONFIG_TRACER_MAX_TRACE + buffer = tr->max_buffer.buffer; + if (buffer) + ring_buffer_record_disable(buffer); +#endif - __this_cpu_write(trace_taskinfo_save, false); + arch_spin_unlock(&tr->max_lock); } /** - * tracing_record_taskinfo_sched_switch - record task info for sched_switch + * tracing_stop - quick stop of the tracer * - * @prev: previous task during sched_switch - * @next: next task during sched_switch - * @flags: TRACE_RECORD_CMDLINE for recording comm - * TRACE_RECORD_TGID for recording tgid + * Light weight way to stop tracing. Use in conjunction with + * tracing_start. */ -void tracing_record_taskinfo_sched_switch(struct task_struct *prev, - struct task_struct *next, int flags) -{ - bool done; - - if (tracing_record_taskinfo_skip(flags)) - return; - - /* - * Record as much task information as possible. If some fail, continue - * to try to record the others. - */ - done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(prev); - done &= !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(next); - done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(prev); - done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(next); - - /* If recording any information failed, retry again soon. */ - if (!done) - return; - - __this_cpu_write(trace_taskinfo_save, false); -} - -/* Helpers to record a specific task information */ -void tracing_record_cmdline(struct task_struct *task) -{ - tracing_record_taskinfo(task, TRACE_RECORD_CMDLINE); -} - -void tracing_record_tgid(struct task_struct *task) +void tracing_stop(void) { - tracing_record_taskinfo(task, TRACE_RECORD_TGID); + return tracing_stop_tr(&global_trace); } /* @@ -2653,6 +2614,8 @@ unsigned int tracing_gen_ctx_irq_test(unsigned int irqs_status) trace_flags |= TRACE_FLAG_NEED_RESCHED; if (test_preempt_need_resched()) trace_flags |= TRACE_FLAG_PREEMPT_RESCHED; + if (IS_ENABLED(CONFIG_ARCH_HAS_PREEMPT_LAZY) && tif_test_bit(TIF_NEED_RESCHED_LAZY)) + trace_flags |= TRACE_FLAG_NEED_RESCHED_LAZY; return (trace_flags << 16) | (min_t(unsigned int, pc & 0xff, 0xf)) | (min_t(unsigned int, migration_disable_value(), 0xf)) << 4; } @@ -2698,31 +2661,28 @@ void trace_buffered_event_enable(void) for_each_tracing_cpu(cpu) { page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); - if (!page) - goto failed; + /* This is just an optimization and can handle failures */ + if (!page) { + pr_err("Failed to allocate event buffer\n"); + break; + } event = page_address(page); memset(event, 0, sizeof(*event)); per_cpu(trace_buffered_event, cpu) = event; - preempt_disable(); - if (cpu == smp_processor_id() && - __this_cpu_read(trace_buffered_event) != - per_cpu(trace_buffered_event, cpu)) - WARN_ON_ONCE(1); - preempt_enable(); + scoped_guard(preempt,) { + if (cpu == smp_processor_id() && + __this_cpu_read(trace_buffered_event) != + per_cpu(trace_buffered_event, cpu)) + WARN_ON_ONCE(1); + } } - - return; - failed: - trace_buffered_event_disable(); } static void enable_trace_buffered_event(void *data) { - /* Probably not needed, but do it anyway */ - smp_rmb(); this_cpu_dec(trace_buffered_event_cnt); } @@ -2751,11 +2711,9 @@ void trace_buffered_event_disable(void) if (--trace_buffered_event_ref) return; - preempt_disable(); /* For each CPU, set the buffer as used. */ - smp_call_function_many(tracing_buffer_mask, - disable_trace_buffered_event, NULL, 1); - preempt_enable(); + on_each_cpu_mask(tracing_buffer_mask, disable_trace_buffered_event, + NULL, true); /* Wait for all current users to finish */ synchronize_rcu(); @@ -2764,17 +2722,19 @@ void trace_buffered_event_disable(void) free_page((unsigned long)per_cpu(trace_buffered_event, cpu)); per_cpu(trace_buffered_event, cpu) = NULL; } + /* - * Make sure trace_buffered_event is NULL before clearing - * trace_buffered_event_cnt. + * Wait for all CPUs that potentially started checking if they can use + * their event buffer only after the previous synchronize_rcu() call and + * they still read a valid pointer from trace_buffered_event. It must be + * ensured they don't see cleared trace_buffered_event_cnt else they + * could wrongly decide to use the pointed-to buffer which is now freed. */ - smp_wmb(); + synchronize_rcu(); - preempt_disable(); - /* Do the work on each cpu */ - smp_call_function_many(tracing_buffer_mask, - enable_trace_buffered_event, NULL, 1); - preempt_enable(); + /* For each CPU, relinquish the buffer */ + on_each_cpu_mask(tracing_buffer_mask, enable_trace_buffered_event, NULL, + true); } static struct trace_buffer *temp_buffer; @@ -2900,14 +2860,14 @@ static void output_printk(struct trace_event_buffer *fbuffer) raw_spin_unlock_irqrestore(&tracepoint_iter_lock, flags); } -int tracepoint_printk_sysctl(struct ctl_table *table, int write, +int tracepoint_printk_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int save_tracepoint_printk; int ret; - mutex_lock(&tracepoint_printk_mutex); + guard(mutex)(&tracepoint_printk_mutex); save_tracepoint_printk = tracepoint_printk; ret = proc_dointvec(table, write, buffer, lenp, ppos); @@ -2920,16 +2880,13 @@ int tracepoint_printk_sysctl(struct ctl_table *table, int write, tracepoint_printk = 0; if (save_tracepoint_printk == tracepoint_printk) - goto out; + return ret; if (tracepoint_printk) static_key_enable(&tracepoint_printk_key.key); else static_key_disable(&tracepoint_printk_key.key); - out: - mutex_unlock(&tracepoint_printk_mutex); - return ret; } @@ -2997,14 +2954,16 @@ trace_buffer_unlock_commit_nostack(struct trace_buffer *buffer, void trace_function(struct trace_array *tr, unsigned long ip, unsigned long - parent_ip, unsigned int trace_ctx) + parent_ip, unsigned int trace_ctx, struct ftrace_regs *fregs) { - struct trace_event_call *call = &event_function; struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct ftrace_entry *entry; + int size = sizeof(*entry); + + size += FTRACE_REGS_MAX_ARGS * !!fregs * sizeof(long); - event = __trace_buffer_lock_reserve(buffer, TRACE_FN, sizeof(*entry), + event = __trace_buffer_lock_reserve(buffer, TRACE_FN, size, trace_ctx); if (!event) return; @@ -3012,11 +2971,16 @@ trace_function(struct trace_array *tr, unsigned long ip, unsigned long entry->ip = ip; entry->parent_ip = parent_ip; - if (!call_filter_check_discard(call, entry, buffer, event)) { - if (static_branch_unlikely(&trace_function_exports_enabled)) - ftrace_exports(event, TRACE_EXPORT_FUNCTION); - __buffer_unlock_commit(buffer, event); +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + if (fregs) { + for (int i = 0; i < FTRACE_REGS_MAX_ARGS; i++) + entry->args[i] = ftrace_regs_get_argument(fregs, i); } +#endif + + if (static_branch_unlikely(&trace_function_exports_enabled)) + ftrace_exports(event, TRACE_EXPORT_FUNCTION); + __buffer_unlock_commit(buffer, event); } #ifdef CONFIG_STACKTRACE @@ -3024,7 +2988,7 @@ trace_function(struct trace_array *tr, unsigned long ip, unsigned long /* Allow 4 levels of nesting: normal, softirq, irq, NMI */ #define FTRACE_KSTACK_NESTING 4 -#define FTRACE_KSTACK_ENTRIES (PAGE_SIZE / FTRACE_KSTACK_NESTING) +#define FTRACE_KSTACK_ENTRIES (SZ_4K / FTRACE_KSTACK_NESTING) struct ftrace_stack { unsigned long calls[FTRACE_KSTACK_ENTRIES]; @@ -3038,11 +3002,11 @@ struct ftrace_stacks { static DEFINE_PER_CPU(struct ftrace_stacks, ftrace_stacks); static DEFINE_PER_CPU(int, ftrace_stack_reserve); -static void __ftrace_trace_stack(struct trace_buffer *buffer, +static void __ftrace_trace_stack(struct trace_array *tr, + struct trace_buffer *buffer, unsigned int trace_ctx, int skip, struct pt_regs *regs) { - struct trace_event_call *call = &event_kernel_stack; struct ring_buffer_event *event; unsigned int size, nr_entries; struct ftrace_stack *fstack; @@ -3058,7 +3022,7 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, skip++; #endif - preempt_disable_notrace(); + guard(preempt_notrace)(); stackidx = __this_cpu_inc_return(ftrace_stack_reserve) - 1; @@ -3085,26 +3049,37 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, nr_entries = stack_trace_save(fstack->calls, size, skip); } - size = nr_entries * sizeof(unsigned long); +#ifdef CONFIG_DYNAMIC_FTRACE + /* Mark entry of stack trace as trampoline code */ + if (tr->ops && tr->ops->trampoline) { + unsigned long tramp_start = tr->ops->trampoline; + unsigned long tramp_end = tramp_start + tr->ops->trampoline_size; + unsigned long *calls = fstack->calls; + + for (int i = 0; i < nr_entries; i++) { + if (calls[i] >= tramp_start && calls[i] < tramp_end) + calls[i] = FTRACE_TRAMPOLINE_MARKER; + } + } +#endif + event = __trace_buffer_lock_reserve(buffer, TRACE_STACK, - (sizeof(*entry) - sizeof(entry->caller)) + size, + struct_size(entry, caller, nr_entries), trace_ctx); if (!event) goto out; entry = ring_buffer_event_data(event); - memcpy(&entry->caller, fstack->calls, size); entry->size = nr_entries; + memcpy(&entry->caller, fstack->calls, + flex_array_size(entry, caller, nr_entries)); - if (!call_filter_check_discard(call, entry, buffer, event)) - __buffer_unlock_commit(buffer, event); + __buffer_unlock_commit(buffer, event); out: /* Again, don't let gcc optimize things here */ barrier(); __this_cpu_dec(ftrace_stack_reserve); - preempt_enable_notrace(); - } static inline void ftrace_trace_stack(struct trace_array *tr, @@ -3112,10 +3087,10 @@ static inline void ftrace_trace_stack(struct trace_array *tr, unsigned int trace_ctx, int skip, struct pt_regs *regs) { - if (!(tr->trace_flags & TRACE_ITER_STACKTRACE)) + if (!(tr->trace_flags & TRACE_ITER(STACKTRACE))) return; - __ftrace_trace_stack(buffer, trace_ctx, skip, regs); + __ftrace_trace_stack(tr, buffer, trace_ctx, skip, regs); } void __trace_stack(struct trace_array *tr, unsigned int trace_ctx, @@ -3124,10 +3099,13 @@ void __trace_stack(struct trace_array *tr, unsigned int trace_ctx, struct trace_buffer *buffer = tr->array_buffer.buffer; if (rcu_is_watching()) { - __ftrace_trace_stack(buffer, trace_ctx, skip, NULL); + __ftrace_trace_stack(tr, buffer, trace_ctx, skip, NULL); return; } + if (WARN_ON_ONCE(IS_ENABLED(CONFIG_GENERIC_ENTRY))) + return; + /* * When an NMI triggers, RCU is enabled via ct_nmi_enter(), * but if the above rcu_is_watching() failed, then the NMI @@ -3138,7 +3116,7 @@ void __trace_stack(struct trace_array *tr, unsigned int trace_ctx, return; ct_irq_enter_irqson(); - __ftrace_trace_stack(buffer, trace_ctx, skip, NULL); + __ftrace_trace_stack(tr, buffer, trace_ctx, skip, NULL); ct_irq_exit_irqson(); } @@ -3155,8 +3133,8 @@ void trace_dump_stack(int skip) /* Skip 1 to skip this function. */ skip++; #endif - __ftrace_trace_stack(global_trace.array_buffer.buffer, - tracing_gen_ctx(), skip, NULL); + __ftrace_trace_stack(printk_trace, printk_trace->array_buffer.buffer, + tracing_gen_ctx(), skip, NULL); } EXPORT_SYMBOL_GPL(trace_dump_stack); @@ -3167,11 +3145,10 @@ static void ftrace_trace_userstack(struct trace_array *tr, struct trace_buffer *buffer, unsigned int trace_ctx) { - struct trace_event_call *call = &event_user_stack; struct ring_buffer_event *event; struct userstack_entry *entry; - if (!(tr->trace_flags & TRACE_ITER_USERSTACKTRACE)) + if (!(tr->trace_flags & TRACE_ITER(USERSTACKTRACE))) return; /* @@ -3185,9 +3162,9 @@ ftrace_trace_userstack(struct trace_array *tr, * prevent recursion, since the user stack tracing may * trigger other kernel events. */ - preempt_disable(); + guard(preempt)(); if (__this_cpu_read(user_stack_count)) - goto out; + return; __this_cpu_inc(user_stack_count); @@ -3201,13 +3178,10 @@ ftrace_trace_userstack(struct trace_array *tr, memset(&entry->caller, 0, sizeof(entry->caller)); stack_trace_save_user(entry->caller, FTRACE_STACK_ENTRIES); - if (!call_filter_check_discard(call, entry, buffer, event)) - __buffer_unlock_commit(buffer, event); + __buffer_unlock_commit(buffer, event); out_drop_count: __this_cpu_dec(user_stack_count); - out: - preempt_enable(); } #else /* CONFIG_USER_STACKTRACE_SUPPORT */ static void ftrace_trace_userstack(struct trace_array *tr, @@ -3329,7 +3303,7 @@ void trace_printk_init_buffers(void) pr_warn("**********************************************************\n"); /* Expand the buffers to set size */ - tracing_update_buffers(); + tracing_update_buffers(&global_trace); buffers_allocated = 1; @@ -3371,15 +3345,17 @@ static void trace_printk_start_stop_comm(int enabled) */ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) { - struct trace_event_call *call = &event_bprint; struct ring_buffer_event *event; struct trace_buffer *buffer; - struct trace_array *tr = &global_trace; + struct trace_array *tr = READ_ONCE(printk_trace); struct bprint_entry *entry; unsigned int trace_ctx; char *tbuffer; int len = 0, size; + if (!printk_binsafe(tr)) + return trace_vprintk(ip, fmt, args); + if (unlikely(tracing_selftest_running || tracing_disabled)) return 0; @@ -3387,7 +3363,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) pause_graph_tracing(); trace_ctx = tracing_gen_ctx(); - preempt_disable_notrace(); + guard(preempt_notrace)(); tbuffer = get_trace_buf(); if (!tbuffer) { @@ -3402,54 +3378,47 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) size = sizeof(*entry) + sizeof(u32) * len; buffer = tr->array_buffer.buffer; - ring_buffer_nest_start(buffer); - event = __trace_buffer_lock_reserve(buffer, TRACE_BPRINT, size, - trace_ctx); - if (!event) - goto out; - entry = ring_buffer_event_data(event); - entry->ip = ip; - entry->fmt = fmt; + scoped_guard(ring_buffer_nest, buffer) { + event = __trace_buffer_lock_reserve(buffer, TRACE_BPRINT, size, + trace_ctx); + if (!event) + goto out_put; + entry = ring_buffer_event_data(event); + entry->ip = ip; + entry->fmt = fmt; - memcpy(entry->buf, tbuffer, sizeof(u32) * len); - if (!call_filter_check_discard(call, entry, buffer, event)) { + memcpy(entry->buf, tbuffer, sizeof(u32) * len); __buffer_unlock_commit(buffer, event); ftrace_trace_stack(tr, buffer, trace_ctx, 6, NULL); } - -out: - ring_buffer_nest_end(buffer); out_put: put_trace_buf(); out_nobuffer: - preempt_enable_notrace(); unpause_graph_tracing(); return len; } EXPORT_SYMBOL_GPL(trace_vbprintk); -__printf(3, 0) -static int -__trace_array_vprintk(struct trace_buffer *buffer, - unsigned long ip, const char *fmt, va_list args) +static __printf(3, 0) +int __trace_array_vprintk(struct trace_buffer *buffer, + unsigned long ip, const char *fmt, va_list args) { - struct trace_event_call *call = &event_print; struct ring_buffer_event *event; int len = 0, size; struct print_entry *entry; unsigned int trace_ctx; char *tbuffer; - if (tracing_disabled || tracing_selftest_running) + if (tracing_disabled) return 0; /* Don't pollute graph traces with trace_vprintk internals */ pause_graph_tracing(); trace_ctx = tracing_gen_ctx(); - preempt_disable_notrace(); + guard(preempt_notrace)(); tbuffer = get_trace_buf(); @@ -3461,35 +3430,33 @@ __trace_array_vprintk(struct trace_buffer *buffer, len = vscnprintf(tbuffer, TRACE_BUF_SIZE, fmt, args); size = sizeof(*entry) + len + 1; - ring_buffer_nest_start(buffer); - event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, - trace_ctx); - if (!event) - goto out; - entry = ring_buffer_event_data(event); - entry->ip = ip; + scoped_guard(ring_buffer_nest, buffer) { + event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, + trace_ctx); + if (!event) + goto out; + entry = ring_buffer_event_data(event); + entry->ip = ip; - memcpy(&entry->buf, tbuffer, len + 1); - if (!call_filter_check_discard(call, entry, buffer, event)) { + memcpy(&entry->buf, tbuffer, len + 1); __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(&global_trace, buffer, trace_ctx, 6, NULL); + ftrace_trace_stack(printk_trace, buffer, trace_ctx, 6, NULL); } - out: - ring_buffer_nest_end(buffer); put_trace_buf(); out_nobuffer: - preempt_enable_notrace(); unpause_graph_tracing(); return len; } -__printf(3, 0) int trace_array_vprintk(struct trace_array *tr, unsigned long ip, const char *fmt, va_list args) { + if (tracing_selftest_running && tr == &global_trace) + return 0; + return __trace_array_vprintk(tr->array_buffer.buffer, ip, fmt, args); } @@ -3513,7 +3480,6 @@ int trace_array_vprintk(struct trace_array *tr, * Note, trace_array_init_printk() must be called on @tr before this * can be used. */ -__printf(3, 0) int trace_array_printk(struct trace_array *tr, unsigned long ip, const char *fmt, ...) { @@ -3527,7 +3493,7 @@ int trace_array_printk(struct trace_array *tr, if (tr == &global_trace) return 0; - if (!(tr->trace_flags & TRACE_ITER_PRINTK)) + if (!(tr->trace_flags & TRACE_ITER(PRINTK))) return 0; va_start(ap, fmt); @@ -3558,14 +3524,13 @@ int trace_array_init_printk(struct trace_array *tr) } EXPORT_SYMBOL_GPL(trace_array_init_printk); -__printf(3, 4) int trace_array_printk_buf(struct trace_buffer *buffer, unsigned long ip, const char *fmt, ...) { int ret; va_list ap; - if (!(global_trace.trace_flags & TRACE_ITER_PRINTK)) + if (!(printk_trace->trace_flags & TRACE_ITER(PRINTK))) return 0; va_start(ap, fmt); @@ -3574,10 +3539,9 @@ int trace_array_printk_buf(struct trace_buffer *buffer, return ret; } -__printf(2, 0) int trace_vprintk(unsigned long ip, const char *fmt, va_list args) { - return trace_array_vprintk(&global_trace, ip, fmt, args); + return trace_array_vprintk(printk_trace, ip, fmt, args); } EXPORT_SYMBOL_GPL(trace_vprintk); @@ -3678,7 +3642,7 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, #define STATIC_FMT_BUF_SIZE 128 static char static_fmt_buf[STATIC_FMT_BUF_SIZE]; -static char *trace_iter_expand_format(struct trace_iterator *iter) +char *trace_iter_expand_format(struct trace_iterator *iter) { char *tmp; @@ -3700,17 +3664,12 @@ static char *trace_iter_expand_format(struct trace_iterator *iter) } /* Returns true if the string is safe to dereference from an event */ -static bool trace_safe_str(struct trace_iterator *iter, const char *str, - bool star, int len) +static bool trace_safe_str(struct trace_iterator *iter, const char *str) { unsigned long addr = (unsigned long)str; struct trace_event *trace_event; struct trace_event_call *event; - /* Ignore strings with no length */ - if (star && !len) - return true; - /* OK if part of the event data */ if ((addr >= (unsigned long)iter->ent) && (addr < (unsigned long)iter->ent + iter->ent_size)) @@ -3718,7 +3677,7 @@ static bool trace_safe_str(struct trace_iterator *iter, const char *str, /* OK if part of the temp seq buffer */ if ((addr >= (unsigned long)iter->tmp_seq.buffer) && - (addr < (unsigned long)iter->tmp_seq.buffer + PAGE_SIZE)) + (addr < (unsigned long)iter->tmp_seq.buffer + TRACE_SEQ_BUFFER_SIZE)) return true; /* Core rodata can not be freed */ @@ -3750,142 +3709,69 @@ static bool trace_safe_str(struct trace_iterator *iter, const char *str, return false; } -static const char *show_buffer(struct trace_seq *s) -{ - struct seq_buf *seq = &s->seq; - - seq_buf_terminate(seq); - - return seq->buffer; -} - -static DEFINE_STATIC_KEY_FALSE(trace_no_verify); - -static int test_can_verify_check(const char *fmt, ...) -{ - char buf[16]; - va_list ap; - int ret; - - /* - * The verifier is dependent on vsnprintf() modifies the va_list - * passed to it, where it is sent as a reference. Some architectures - * (like x86_32) passes it by value, which means that vsnprintf() - * does not modify the va_list passed to it, and the verifier - * would then need to be able to understand all the values that - * vsnprintf can use. If it is passed by value, then the verifier - * is disabled. - */ - va_start(ap, fmt); - vsnprintf(buf, 16, "%d", ap); - ret = va_arg(ap, int); - va_end(ap); - - return ret; -} - -static void test_can_verify(void) -{ - if (!test_can_verify_check("%d %d", 0, 1)) { - pr_info("trace event string verifier disabled\n"); - static_branch_inc(&trace_no_verify); - } -} - /** - * trace_check_vprintf - Check dereferenced strings while writing to the seq buffer + * ignore_event - Check dereferenced fields while writing to the seq buffer * @iter: The iterator that holds the seq buffer and the event being printed - * @fmt: The format used to print the event - * @ap: The va_list holding the data to print from @fmt. * - * This writes the data into the @iter->seq buffer using the data from - * @fmt and @ap. If the format has a %s, then the source of the string - * is examined to make sure it is safe to print, otherwise it will - * warn and print "[UNSAFE MEMORY]" in place of the dereferenced string - * pointer. + * At boot up, test_event_printk() will flag any event that dereferences + * a string with "%s" that does exist in the ring buffer. It may still + * be valid, as the string may point to a static string in the kernel + * rodata that never gets freed. But if the string pointer is pointing + * to something that was allocated, there's a chance that it can be freed + * by the time the user reads the trace. This would cause a bad memory + * access by the kernel and possibly crash the system. + * + * This function will check if the event has any fields flagged as needing + * to be checked at runtime and perform those checks. + * + * If it is found that a field is unsafe, it will write into the @iter->seq + * a message stating what was found to be unsafe. + * + * @return: true if the event is unsafe and should be ignored, + * false otherwise. */ -void trace_check_vprintf(struct trace_iterator *iter, const char *fmt, - va_list ap) +bool ignore_event(struct trace_iterator *iter) { - const char *p = fmt; - const char *str; - int i, j; + struct ftrace_event_field *field; + struct trace_event *trace_event; + struct trace_event_call *event; + struct list_head *head; + struct trace_seq *seq; + const void *ptr; - if (WARN_ON_ONCE(!fmt)) - return; + trace_event = ftrace_find_event(iter->ent->type); - if (static_branch_unlikely(&trace_no_verify)) - goto print; + seq = &iter->seq; - /* Don't bother checking when doing a ftrace_dump() */ - if (iter->fmt == static_fmt_buf) - goto print; + if (!trace_event) { + trace_seq_printf(seq, "EVENT ID %d NOT FOUND?\n", iter->ent->type); + return true; + } - while (*p) { - bool star = false; - int len = 0; - - j = 0; - - /* We only care about %s and variants */ - for (i = 0; p[i]; i++) { - if (i + 1 >= iter->fmt_size) { - /* - * If we can't expand the copy buffer, - * just print it. - */ - if (!trace_iter_expand_format(iter)) - goto print; - } + event = container_of(trace_event, struct trace_event_call, event); + if (!(event->flags & TRACE_EVENT_FL_TEST_STR)) + return false; - if (p[i] == '\\' && p[i+1]) { - i++; - continue; - } - if (p[i] == '%') { - /* Need to test cases like %08.*s */ - for (j = 1; p[i+j]; j++) { - if (isdigit(p[i+j]) || - p[i+j] == '.') - continue; - if (p[i+j] == '*') { - star = true; - continue; - } - break; - } - if (p[i+j] == 's') - break; - star = false; - } - j = 0; - } - /* If no %s found then just print normally */ - if (!p[i]) - break; + head = trace_get_fields(event); + if (!head) { + trace_seq_printf(seq, "FIELDS FOR EVENT '%s' NOT FOUND?\n", + trace_event_name(event)); + return true; + } - /* Copy up to the %s, and print that */ - strncpy(iter->fmt, p, i); - iter->fmt[i] = '\0'; - trace_seq_vprintf(&iter->seq, iter->fmt, ap); + /* Offsets are from the iter->ent that points to the raw event */ + ptr = iter->ent; - /* - * If iter->seq is full, the above call no longer guarantees - * that ap is in sync with fmt processing, and further calls - * to va_arg() can return wrong positional arguments. - * - * Ensure that ap is no longer used in this case. - */ - if (iter->seq.full) { - p = ""; - break; - } + list_for_each_entry(field, head, link) { + const char *str; + bool good; + + if (!field->needs_test) + continue; - if (star) - len = va_arg(ap, int); + str = *(const char **)(ptr + field->offset); - /* The ap now points to the string data of the %s */ - str = va_arg(ap, const char *); + good = trace_safe_str(iter, str); /* * If you hit this warning, it is likely that the @@ -3896,45 +3782,14 @@ void trace_check_vprintf(struct trace_iterator *iter, const char *fmt, * instead. See samples/trace_events/trace-events-sample.h * for reference. */ - if (WARN_ONCE(!trace_safe_str(iter, str, star, len), - "fmt: '%s' current_buffer: '%s'", - fmt, show_buffer(&iter->seq))) { - int ret; - - /* Try to safely read the string */ - if (star) { - if (len + 1 > iter->fmt_size) - len = iter->fmt_size - 1; - if (len < 0) - len = 0; - ret = copy_from_kernel_nofault(iter->fmt, str, len); - iter->fmt[len] = 0; - star = false; - } else { - ret = strncpy_from_kernel_nofault(iter->fmt, str, - iter->fmt_size); - } - if (ret < 0) - trace_seq_printf(&iter->seq, "(0x%px)", str); - else - trace_seq_printf(&iter->seq, "(0x%px:%s)", - str, iter->fmt); - str = "[UNSAFE-MEMORY]"; - strcpy(iter->fmt, "%s"); - } else { - strncpy(iter->fmt, p + i, j + 1); - iter->fmt[j+1] = '\0'; + if (WARN_ONCE(!good, "event '%s' has unsafe pointer field '%s'", + trace_event_name(event), field->name)) { + trace_seq_printf(seq, "EVENT %s: HAS UNSAFE POINTER FIELD '%s'\n", + trace_event_name(event), field->name); + return true; } - if (star) - trace_seq_printf(&iter->seq, iter->fmt, len, str); - else - trace_seq_printf(&iter->seq, iter->fmt, str); - - p += i + j + 1; } - print: - if (*p) - trace_seq_vprintf(&iter->seq, p, ap); + return false; } const char *trace_event_format(struct trace_iterator *iter, const char *fmt) @@ -3945,7 +3800,7 @@ const char *trace_event_format(struct trace_iterator *iter, const char *fmt) if (WARN_ON_ONCE(!fmt)) return fmt; - if (!iter->tr || iter->tr->trace_flags & TRACE_ITER_HASH_PTR) + if (!iter->tr || iter->tr->trace_flags & TRACE_ITER(HASH_PTR)) return fmt; p = fmt; @@ -4094,6 +3949,8 @@ void tracing_iter_reset(struct trace_iterator *iter, int cpu) break; entries++; ring_buffer_iter_advance(buf_iter); + /* This could be a big loop */ + cond_resched(); } per_cpu_ptr(iter->array_buffer->data, cpu)->skipped_entries = entries; @@ -4112,15 +3969,16 @@ static void *s_start(struct seq_file *m, loff_t *pos) loff_t l = 0; int cpu; - /* - * copy the tracer to avoid using a global lock all around. - * iter->trace is a copy of current_trace, the pointer to the - * name may be used instead of a strcmp(), as iter->trace->name - * will point to the same string as current_trace->name. - */ mutex_lock(&trace_types_lock); - if (unlikely(tr->current_trace && iter->trace->name != tr->current_trace->name)) - *iter->trace = *tr->current_trace; + if (unlikely(tr->current_trace != iter->trace)) { + /* Close iter->trace before switching to the new current tracer */ + if (iter->trace->close) + iter->trace->close(iter); + iter->trace = tr->current_trace; + /* Reopen the new current tracer */ + if (iter->trace->open) + iter->trace->open(iter); + } mutex_unlock(&trace_types_lock); #ifdef CONFIG_TRACER_MAX_TRACE @@ -4264,7 +4122,7 @@ static void print_event_info(struct array_buffer *buf, struct seq_file *m) static void print_func_help_header(struct array_buffer *buf, struct seq_file *m, unsigned int flags) { - bool tgid = flags & TRACE_ITER_RECORD_TGID; + bool tgid = flags & TRACE_ITER(RECORD_TGID); print_event_info(buf, m); @@ -4275,7 +4133,7 @@ static void print_func_help_header(struct array_buffer *buf, struct seq_file *m, static void print_func_help_header_irq(struct array_buffer *buf, struct seq_file *m, unsigned int flags) { - bool tgid = flags & TRACE_ITER_RECORD_TGID; + bool tgid = flags & TRACE_ITER(RECORD_TGID); static const char space[] = " "; int prec = tgid ? 12 : 2; @@ -4305,7 +4163,7 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter) get_total_entries(buf, &total, &entries); seq_printf(m, "# %s latency trace v1.1.5 on %s\n", - name, UTS_RELEASE); + name, init_utsname()->release); seq_puts(m, "# -----------------------------------" "---------------------------------\n"); seq_printf(m, "# latency: %lu us, #%lu/%lu, CPU#%d |" @@ -4314,11 +4172,7 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter) entries, total, buf->cpu, - preempt_model_none() ? "server" : - preempt_model_voluntary() ? "desktop" : - preempt_model_full() ? "preempt" : - preempt_model_rt() ? "preempt_rt" : - "unknown", + preempt_model_str(), /* These are reserved for later use */ 0, 0, 0, 0); #ifdef CONFIG_SMP @@ -4352,7 +4206,7 @@ static void test_cpu_buff_start(struct trace_iterator *iter) struct trace_seq *s = &iter->seq; struct trace_array *tr = iter->tr; - if (!(tr->trace_flags & TRACE_ITER_ANNOTATE)) + if (!(tr->trace_flags & TRACE_ITER(ANNOTATE))) return; if (!(iter->iter_flags & TRACE_FILE_ANNOTATE)) @@ -4374,6 +4228,22 @@ static void test_cpu_buff_start(struct trace_iterator *iter) iter->cpu); } +#ifdef CONFIG_FTRACE_SYSCALLS +static bool is_syscall_event(struct trace_event *event) +{ + return (event->funcs == &enter_syscall_print_funcs) || + (event->funcs == &exit_syscall_print_funcs); + +} +#define syscall_buf_size CONFIG_TRACE_SYSCALL_BUF_SIZE_DEFAULT +#else +static inline bool is_syscall_event(struct trace_event *event) +{ + return false; +} +#define syscall_buf_size 0 +#endif /* CONFIG_FTRACE_SYSCALLS */ + static enum print_line_t print_trace_fmt(struct trace_iterator *iter) { struct trace_array *tr = iter->tr; @@ -4388,7 +4258,7 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) event = ftrace_find_event(entry->type); - if (tr->trace_flags & TRACE_ITER_CONTEXT_INFO) { + if (tr->trace_flags & TRACE_ITER(CONTEXT_INFO)) { if (iter->iter_flags & TRACE_FILE_LAT_FMT) trace_print_lat_context(iter); else @@ -4398,8 +4268,22 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) if (trace_seq_has_overflowed(s)) return TRACE_TYPE_PARTIAL_LINE; - if (event) + if (event) { + if (tr->trace_flags & TRACE_ITER(FIELDS)) + return print_event_fields(iter, event); + /* + * For TRACE_EVENT() events, the print_fmt is not + * safe to use if the array has delta offsets + * Force printing via the fields. + */ + if ((tr->text_delta)) { + /* ftrace and system call events are still OK */ + if ((event->type > __TRACE_LAST_TYPE) && + !is_syscall_event(event)) + return print_event_fields(iter, event); + } return event->funcs->trace(iter, sym_flags, event); + } trace_seq_printf(s, "Unknown type %d\n", entry->type); @@ -4415,7 +4299,7 @@ static enum print_line_t print_raw_fmt(struct trace_iterator *iter) entry = iter->ent; - if (tr->trace_flags & TRACE_ITER_CONTEXT_INFO) + if (tr->trace_flags & TRACE_ITER(CONTEXT_INFO)) trace_seq_printf(s, "%d %d %llu ", entry->pid, iter->cpu, iter->ts); @@ -4441,7 +4325,7 @@ static enum print_line_t print_hex_fmt(struct trace_iterator *iter) entry = iter->ent; - if (tr->trace_flags & TRACE_ITER_CONTEXT_INFO) { + if (tr->trace_flags & TRACE_ITER(CONTEXT_INFO)) { SEQ_PUT_HEX_FIELD(s, entry->pid); SEQ_PUT_HEX_FIELD(s, iter->cpu); SEQ_PUT_HEX_FIELD(s, iter->ts); @@ -4470,7 +4354,7 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter) entry = iter->ent; - if (tr->trace_flags & TRACE_ITER_CONTEXT_INFO) { + if (tr->trace_flags & TRACE_ITER(CONTEXT_INFO)) { SEQ_PUT_FIELD(s, entry->pid); SEQ_PUT_FIELD(s, iter->cpu); SEQ_PUT_FIELD(s, iter->ts); @@ -4541,27 +4425,27 @@ enum print_line_t print_trace_line(struct trace_iterator *iter) } if (iter->ent->type == TRACE_BPUTS && - trace_flags & TRACE_ITER_PRINTK && - trace_flags & TRACE_ITER_PRINTK_MSGONLY) + trace_flags & TRACE_ITER(PRINTK) && + trace_flags & TRACE_ITER(PRINTK_MSGONLY)) return trace_print_bputs_msg_only(iter); if (iter->ent->type == TRACE_BPRINT && - trace_flags & TRACE_ITER_PRINTK && - trace_flags & TRACE_ITER_PRINTK_MSGONLY) + trace_flags & TRACE_ITER(PRINTK) && + trace_flags & TRACE_ITER(PRINTK_MSGONLY)) return trace_print_bprintk_msg_only(iter); if (iter->ent->type == TRACE_PRINT && - trace_flags & TRACE_ITER_PRINTK && - trace_flags & TRACE_ITER_PRINTK_MSGONLY) + trace_flags & TRACE_ITER(PRINTK) && + trace_flags & TRACE_ITER(PRINTK_MSGONLY)) return trace_print_printk_msg_only(iter); - if (trace_flags & TRACE_ITER_BIN) + if (trace_flags & TRACE_ITER(BIN)) return print_bin_fmt(iter); - if (trace_flags & TRACE_ITER_HEX) + if (trace_flags & TRACE_ITER(HEX)) return print_hex_fmt(iter); - if (trace_flags & TRACE_ITER_RAW) + if (trace_flags & TRACE_ITER(RAW)) return print_raw_fmt(iter); return print_trace_fmt(iter); @@ -4579,7 +4463,7 @@ void trace_latency_header(struct seq_file *m) if (iter->iter_flags & TRACE_FILE_LAT_FMT) print_trace_header(m, iter); - if (!(tr->trace_flags & TRACE_ITER_VERBOSE)) + if (!(tr->trace_flags & TRACE_ITER(VERBOSE))) print_lat_help_header(m); } @@ -4589,7 +4473,7 @@ void trace_default_header(struct seq_file *m) struct trace_array *tr = iter->tr; unsigned long trace_flags = tr->trace_flags; - if (!(trace_flags & TRACE_ITER_CONTEXT_INFO)) + if (!(trace_flags & TRACE_ITER(CONTEXT_INFO))) return; if (iter->iter_flags & TRACE_FILE_LAT_FMT) { @@ -4597,11 +4481,11 @@ void trace_default_header(struct seq_file *m) if (trace_empty(iter)) return; print_trace_header(m, iter); - if (!(trace_flags & TRACE_ITER_VERBOSE)) + if (!(trace_flags & TRACE_ITER(VERBOSE))) print_lat_help_header(m); } else { - if (!(trace_flags & TRACE_ITER_VERBOSE)) { - if (trace_flags & TRACE_ITER_IRQ_INFO) + if (!(trace_flags & TRACE_ITER(VERBOSE))) { + if (trace_flags & TRACE_ITER(IRQ_INFO)) print_func_help_header_irq(iter->array_buffer, m, trace_flags); else @@ -4692,7 +4576,11 @@ static int s_show(struct seq_file *m, void *v) iter->leftover = ret; } else { - print_trace_line(iter); + ret = print_trace_line(iter); + if (ret == TRACE_TYPE_PARTIAL_LINE) { + iter->seq.full = 0; + trace_seq_puts(&iter->seq, "[LINE TOO BIG]\n"); + } ret = trace_print_seq(m, &iter->seq); /* * If we overflow the seq_file buffer, then it will @@ -4725,6 +4613,25 @@ static const struct seq_operations tracer_seq_ops = { .show = s_show, }; +/* + * Note, as iter itself can be allocated and freed in different + * ways, this function is only used to free its content, and not + * the iterator itself. The only requirement to all the allocations + * is that it must zero all fields (kzalloc), as freeing works with + * ethier allocated content or NULL. + */ +static void free_trace_iter_content(struct trace_iterator *iter) +{ + /* The fmt is either NULL, allocated or points to static_fmt_buf */ + if (iter->fmt != static_fmt_buf) + kfree(iter->fmt); + + kfree(iter->temp); + kfree(iter->buffer_iter); + mutex_destroy(&iter->mutex); + free_cpumask_var(iter->started); +} + static struct trace_iterator * __tracing_open(struct inode *inode, struct file *file, bool snapshot) { @@ -4766,16 +4673,8 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) iter->fmt = NULL; iter->fmt_size = 0; - /* - * We make a copy of the current tracer to avoid concurrent - * changes on it while we are reading. - */ mutex_lock(&trace_types_lock); - iter->trace = kzalloc(sizeof(*iter->trace), GFP_KERNEL); - if (!iter->trace) - goto fail; - - *iter->trace = *tr->current_trace; + iter->trace = tr->current_trace; if (!zalloc_cpumask_var(&iter->started, GFP_KERNEL)) goto fail; @@ -4810,27 +4709,23 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) * If pause-on-trace is enabled, then stop the trace while * dumping, unless this is the "snapshot" file */ - if (!iter->snapshot && (tr->trace_flags & TRACE_ITER_PAUSE_ON_TRACE)) + if (!iter->snapshot && (tr->trace_flags & TRACE_ITER(PAUSE_ON_TRACE))) { + iter->iter_flags |= TRACE_FILE_PAUSE; tracing_stop_tr(tr); + } if (iter->cpu_file == RING_BUFFER_ALL_CPUS) { for_each_tracing_cpu(cpu) { iter->buffer_iter[cpu] = - ring_buffer_read_prepare(iter->array_buffer->buffer, - cpu, GFP_KERNEL); - } - ring_buffer_read_prepare_sync(); - for_each_tracing_cpu(cpu) { - ring_buffer_read_start(iter->buffer_iter[cpu]); + ring_buffer_read_start(iter->array_buffer->buffer, + cpu, GFP_KERNEL); tracing_iter_reset(iter, cpu); } } else { cpu = iter->cpu_file; iter->buffer_iter[cpu] = - ring_buffer_read_prepare(iter->array_buffer->buffer, - cpu, GFP_KERNEL); - ring_buffer_read_prepare_sync(); - ring_buffer_read_start(iter->buffer_iter[cpu]); + ring_buffer_read_start(iter->array_buffer->buffer, + cpu, GFP_KERNEL); tracing_iter_reset(iter, cpu); } @@ -4840,9 +4735,7 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) fail: mutex_unlock(&trace_types_lock); - kfree(iter->trace); - kfree(iter->temp); - kfree(iter->buffer_iter); + free_trace_iter_content(iter); release: seq_release_private(inode, file); return ERR_PTR(-ENOMEM); @@ -4883,10 +4776,48 @@ int tracing_open_generic_tr(struct inode *inode, struct file *filp) return 0; } -static int tracing_mark_open(struct inode *inode, struct file *filp) +/* + * The private pointer of the inode is the trace_event_file. + * Update the tr ref count associated to it. + */ +int tracing_open_file_tr(struct inode *inode, struct file *filp) { - stream_open(inode, filp); - return tracing_open_generic_tr(inode, filp); + struct trace_event_file *file = inode->i_private; + int ret; + + ret = tracing_check_open_get_tr(file->tr); + if (ret) + return ret; + + guard(mutex)(&event_mutex); + + /* Fail if the file is marked for removal */ + if (file->flags & EVENT_FILE_FL_FREED) { + trace_array_put(file->tr); + return -ENODEV; + } else { + event_file_get(file); + } + + filp->private_data = inode->i_private; + + return 0; +} + +int tracing_release_file_tr(struct inode *inode, struct file *filp) +{ + struct trace_event_file *file = inode->i_private; + + trace_array_put(file->tr); + event_file_put(file); + + return 0; +} + +int tracing_single_release_file_tr(struct inode *inode, struct file *filp) +{ + tracing_release_file_tr(inode, filp); + return single_release(inode, filp); } static int tracing_release(struct inode *inode, struct file *file) @@ -4913,7 +4844,7 @@ static int tracing_release(struct inode *inode, struct file *file) if (iter->trace && iter->trace->close) iter->trace->close(iter); - if (!iter->snapshot && tr->stop_count) + if (iter->iter_flags & TRACE_FILE_PAUSE) /* reenable tracing if it was previously enabled */ tracing_start_tr(tr); @@ -4921,18 +4852,13 @@ static int tracing_release(struct inode *inode, struct file *file) mutex_unlock(&trace_types_lock); - mutex_destroy(&iter->mutex); - free_cpumask_var(iter->started); - kfree(iter->fmt); - kfree(iter->temp); - kfree(iter->trace); - kfree(iter->buffer_iter); + free_trace_iter_content(iter); seq_release_private(inode, file); return 0; } -static int tracing_release_generic_tr(struct inode *inode, struct file *file) +int tracing_release_generic_tr(struct inode *inode, struct file *file) { struct trace_array *tr = inode->i_private; @@ -4979,7 +4905,7 @@ static int tracing_open(struct inode *inode, struct file *file) iter = __tracing_open(inode, file, false); if (IS_ERR(iter)) ret = PTR_ERR(iter); - else if (tr->trace_flags & TRACE_ITER_LATENCY_FMT) + else if (tr->trace_flags & TRACE_ITER(LATENCY_FMT)) iter->iter_flags |= TRACE_FILE_LAT_FMT; } @@ -4997,6 +4923,11 @@ static int tracing_open(struct inode *inode, struct file *file) static bool trace_ok_for_array(struct tracer *t, struct trace_array *tr) { +#ifdef CONFIG_TRACER_SNAPSHOT + /* arrays with mapped buffer range do not have snapshots */ + if (tr->range_addr_start && t->use_max_tr) + return false; +#endif return (tr->flags & TRACE_ARRAY_FL_GLOBAL) || t->allow_instances; } @@ -5089,7 +5020,7 @@ static int show_traces_open(struct inode *inode, struct file *file) return 0; } -static int show_traces_release(struct inode *inode, struct file *file) +static int tracing_seq_release(struct inode *inode, struct file *file) { struct trace_array *tr = inode->i_private; @@ -5119,6 +5050,8 @@ loff_t tracing_lseek(struct file *file, loff_t offset, int whence) static const struct file_operations tracing_fops = { .open = tracing_open, .read = seq_read, + .read_iter = seq_read_iter, + .splice_read = copy_splice_read, .write = tracing_write_stub, .llseek = tracing_lseek, .release = tracing_release, @@ -5128,7 +5061,7 @@ static const struct file_operations show_traces_fops = { .open = show_traces_open, .read = seq_read, .llseek = seq_lseek, - .release = show_traces_release, + .release = tracing_seq_release, }; static ssize_t @@ -5136,7 +5069,7 @@ tracing_cpumask_read(struct file *filp, char __user *ubuf, size_t count, loff_t *ppos) { struct trace_array *tr = file_inode(filp)->i_private; - char *mask_str; + char *mask_str __free(kfree) = NULL; int len; len = snprintf(NULL, 0, "%*pb\n", @@ -5147,16 +5080,10 @@ tracing_cpumask_read(struct file *filp, char __user *ubuf, len = snprintf(mask_str, len, "%*pb\n", cpumask_pr_args(tr->tracing_cpumask)); - if (len >= count) { - count = -EINVAL; - goto out_err; - } - count = simple_read_from_buffer(ubuf, count, ppos, mask_str, len); - -out_err: - kfree(mask_str); + if (len >= count) + return -EINVAL; - return count; + return simple_read_from_buffer(ubuf, count, ppos, mask_str, len); } int tracing_set_cpumask(struct trace_array *tr, @@ -5176,13 +5103,17 @@ int tracing_set_cpumask(struct trace_array *tr, */ if (cpumask_test_cpu(cpu, tr->tracing_cpumask) && !cpumask_test_cpu(cpu, tracing_cpumask_new)) { - atomic_inc(&per_cpu_ptr(tr->array_buffer.data, cpu)->disabled); ring_buffer_record_disable_cpu(tr->array_buffer.buffer, cpu); +#ifdef CONFIG_TRACER_MAX_TRACE + ring_buffer_record_disable_cpu(tr->max_buffer.buffer, cpu); +#endif } if (!cpumask_test_cpu(cpu, tr->tracing_cpumask) && cpumask_test_cpu(cpu, tracing_cpumask_new)) { - atomic_dec(&per_cpu_ptr(tr->array_buffer.data, cpu)->disabled); ring_buffer_record_enable_cpu(tr->array_buffer.buffer, cpu); +#ifdef CONFIG_TRACER_MAX_TRACE + ring_buffer_record_enable_cpu(tr->max_buffer.buffer, cpu); +#endif } } arch_spin_unlock(&tr->max_lock); @@ -5201,6 +5132,9 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, cpumask_var_t tracing_cpumask_new; int err; + if (count == 0 || count > KMALLOC_MAX_SIZE) + return -EINVAL; + if (!zalloc_cpumask_var(&tracing_cpumask_new, GFP_KERNEL)) return -ENOMEM; @@ -5234,27 +5168,32 @@ static int tracing_trace_options_show(struct seq_file *m, void *v) { struct tracer_opt *trace_opts; struct trace_array *tr = m->private; + struct tracer_flags *flags; u32 tracer_flags; int i; - mutex_lock(&trace_types_lock); - tracer_flags = tr->current_trace->flags->val; - trace_opts = tr->current_trace->flags->opts; + guard(mutex)(&trace_types_lock); for (i = 0; trace_options[i]; i++) { - if (tr->trace_flags & (1 << i)) + if (tr->trace_flags & (1ULL << i)) seq_printf(m, "%s\n", trace_options[i]); else seq_printf(m, "no%s\n", trace_options[i]); } + flags = tr->current_trace_flags; + if (!flags || !flags->opts) + return 0; + + tracer_flags = flags->val; + trace_opts = flags->opts; + for (i = 0; trace_opts[i].name; i++) { if (tracer_flags & trace_opts[i].bit) seq_printf(m, "%s\n", trace_opts[i].name); else seq_printf(m, "no%s\n", trace_opts[i].name); } - mutex_unlock(&trace_types_lock); return 0; } @@ -5264,9 +5203,10 @@ static int __set_tracer_option(struct trace_array *tr, struct tracer_opt *opts, int neg) { struct tracer *trace = tracer_flags->trace; - int ret; + int ret = 0; - ret = trace->set_flag(tr, tracer_flags->val, opts->bit, !neg); + if (trace->set_flag) + ret = trace->set_flag(tr, tracer_flags->val, opts->bit, !neg); if (ret) return ret; @@ -5280,37 +5220,41 @@ static int __set_tracer_option(struct trace_array *tr, /* Try to assign a tracer specific option */ static int set_tracer_option(struct trace_array *tr, char *cmp, int neg) { - struct tracer *trace = tr->current_trace; - struct tracer_flags *tracer_flags = trace->flags; + struct tracer_flags *tracer_flags = tr->current_trace_flags; struct tracer_opt *opts = NULL; int i; + if (!tracer_flags || !tracer_flags->opts) + return 0; + for (i = 0; tracer_flags->opts[i].name; i++) { opts = &tracer_flags->opts[i]; if (strcmp(cmp, opts->name) == 0) - return __set_tracer_option(tr, trace->flags, opts, neg); + return __set_tracer_option(tr, tracer_flags, opts, neg); } return -EINVAL; } /* Some tracers require overwrite to stay enabled */ -int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set) +int trace_keep_overwrite(struct tracer *tracer, u64 mask, int set) { - if (tracer->enabled && (mask & TRACE_ITER_OVERWRITE) && !set) + if (tracer->enabled && (mask & TRACE_ITER(OVERWRITE)) && !set) return -1; return 0; } -int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) +int set_tracer_flag(struct trace_array *tr, u64 mask, int enabled) { - int *map; - - if ((mask == TRACE_ITER_RECORD_TGID) || - (mask == TRACE_ITER_RECORD_CMD)) + switch (mask) { + case TRACE_ITER(RECORD_TGID): + case TRACE_ITER(RECORD_CMD): + case TRACE_ITER(TRACE_PRINTK): + case TRACE_ITER(COPY_MARKER): lockdep_assert_held(&event_mutex); + } /* do nothing if flag is already set */ if (!!(tr->trace_flags & mask) == !!enabled) @@ -5321,52 +5265,77 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) if (tr->current_trace->flag_changed(tr, mask, !!enabled)) return -EINVAL; + switch (mask) { + case TRACE_ITER(TRACE_PRINTK): + if (enabled) { + update_printk_trace(tr); + } else { + /* + * The global_trace cannot clear this. + * It's flag only gets cleared if another instance sets it. + */ + if (printk_trace == &global_trace) + return -EINVAL; + /* + * An instance must always have it set. + * by default, that's the global_trace instance. + */ + if (printk_trace == tr) + update_printk_trace(&global_trace); + } + break; + + case TRACE_ITER(COPY_MARKER): + update_marker_trace(tr, enabled); + /* update_marker_trace updates the tr->trace_flags */ + return 0; + } + if (enabled) tr->trace_flags |= mask; else tr->trace_flags &= ~mask; - if (mask == TRACE_ITER_RECORD_CMD) + switch (mask) { + case TRACE_ITER(RECORD_CMD): trace_event_enable_cmd_record(enabled); + break; - if (mask == TRACE_ITER_RECORD_TGID) { - if (!tgid_map) { - tgid_map_max = pid_max; - map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), - GFP_KERNEL); + case TRACE_ITER(RECORD_TGID): - /* - * Pairs with smp_load_acquire() in - * trace_find_tgid_ptr() to ensure that if it observes - * the tgid_map we just allocated then it also observes - * the corresponding tgid_map_max value. - */ - smp_store_release(&tgid_map, map); - } - if (!tgid_map) { - tr->trace_flags &= ~TRACE_ITER_RECORD_TGID; + if (trace_alloc_tgid_map() < 0) { + tr->trace_flags &= ~TRACE_ITER(RECORD_TGID); return -ENOMEM; } trace_event_enable_tgid_record(enabled); - } + break; - if (mask == TRACE_ITER_EVENT_FORK) + case TRACE_ITER(EVENT_FORK): trace_event_follow_fork(tr, enabled); + break; - if (mask == TRACE_ITER_FUNC_FORK) + case TRACE_ITER(FUNC_FORK): ftrace_pid_follow_fork(tr, enabled); + break; - if (mask == TRACE_ITER_OVERWRITE) { + case TRACE_ITER(OVERWRITE): ring_buffer_change_overwrite(tr->array_buffer.buffer, enabled); #ifdef CONFIG_TRACER_MAX_TRACE ring_buffer_change_overwrite(tr->max_buffer.buffer, enabled); #endif - } + break; - if (mask == TRACE_ITER_PRINTK) { + case TRACE_ITER(PRINTK): trace_printk_start_stop_comm(enabled); trace_printk_control(enabled); + break; + +#if defined(CONFIG_FUNCTION_PROFILER) && defined(CONFIG_FUNCTION_GRAPH_TRACER) + case TRACE_GRAPH_GRAPH_TIME: + ftrace_graph_graph_time_control(enabled); + break; +#endif } return 0; @@ -5396,7 +5365,7 @@ int trace_set_options(struct trace_array *tr, char *option) if (ret < 0) ret = set_tracer_option(tr, cmp, neg); else - ret = set_tracer_flag(tr, 1 << ret, !neg); + ret = set_tracer_flag(tr, 1ULL << ret, !neg); mutex_unlock(&trace_types_lock); mutex_unlock(&event_mutex); @@ -5483,6 +5452,10 @@ static const struct file_operations tracing_iter_fops = { static const char readme_msg[] = "tracing mini-HOWTO:\n\n" + "By default tracefs removes all OTH file permission bits.\n" + "When mounting tracefs an optional group id can be specified\n" + "which adds the group to every directory and file in tracefs:\n\n" + "\t e.g. mount -t tracefs [-o [gid=<gid>]] nodev /sys/kernel/tracing\n\n" "# echo 0 > tracing_on : quick way to disable tracing\n" "# echo 1 > tracing_on : quick way to re-enable tracing\n\n" " Important files:\n" @@ -5591,14 +5564,21 @@ static const char readme_msg[] = " uprobe_events\t\t- Create/append/remove/show the userspace dynamic events\n" "\t\t\t Write into this file to define/undefine new trace events.\n" #endif -#if defined(CONFIG_KPROBE_EVENTS) || defined(CONFIG_UPROBE_EVENTS) +#if defined(CONFIG_KPROBE_EVENTS) || defined(CONFIG_UPROBE_EVENTS) || \ + defined(CONFIG_FPROBE_EVENTS) "\t accepts: event-definitions (one definition per line)\n" +#if defined(CONFIG_KPROBE_EVENTS) || defined(CONFIG_UPROBE_EVENTS) "\t Format: p[:[<group>/][<event>]] <place> [<args>]\n" "\t r[maxactive][:[<group>/][<event>]] <place> [<args>]\n" +#endif +#ifdef CONFIG_FPROBE_EVENTS + "\t f[:[<group>/][<event>]] <func-name>[%return] [<args>]\n" + "\t t[:[<group>/][<event>]] <tracepoint> [<args>]\n" +#endif #ifdef CONFIG_HIST_TRIGGERS "\t s:[synthetic/]<event> <field> [<field>]\n" #endif - "\t e[:[<group>/][<event>]] <attached-group>.<attached-event> [<args>]\n" + "\t e[:[<group>/][<event>]] <attached-group>.<attached-event> [<args>] [if <filter>]\n" "\t -:[<group>/][<event>]\n" #ifdef CONFIG_KPROBE_EVENTS "\t place: [<module>:]<symbol>[+<offset>]|<memaddr>\n" @@ -5611,13 +5591,17 @@ static const char readme_msg[] = "\t fetcharg: (%<register>|$<efield>), @<address>, @<symbol>[+|-<offset>],\n" #ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API "\t $stack<index>, $stack, $retval, $comm, $arg<N>,\n" +#ifdef CONFIG_PROBE_EVENTS_BTF_ARGS + "\t <argname>[->field[->field|.field...]],\n" +#endif #else "\t $stack<index>, $stack, $retval, $comm,\n" #endif "\t +|-[u]<offset>(<fetcharg>), \\imm-value, \\\"imm-string\"\n" - "\t type: s8/16/32/64, u8/16/32/64, x8/16/32/64, string, symbol,\n" + "\t kernel return probes support: $retval, $arg<N>, $comm\n" + "\t type: s8/16/32/64, u8/16/32/64, x8/16/32/64, char, string, symbol,\n" "\t b<bit-width>@<bit-offset>/<container-size>, ustring,\n" - "\t symstr, <type>\\[<array-size>\\]\n" + "\t symstr, %pd/%pD, <type>\\[<array-size>\\]\n" #ifdef CONFIG_HIST_TRIGGERS "\t field: <stype> <name>;\n" "\t stype: u8/u16/u32/u64, s8/s16/s32/s64, pid_t,\n" @@ -5626,6 +5610,8 @@ static const char readme_msg[] = "\t efield: For event probes ('e' types), the field is on of the fields\n" "\t of the <attached-group>/<attached-event>.\n" #endif + " set_event\t\t- Enables events by name written into it\n" + "\t\t\t Can enable module events via: :mod:<module>\n" " events/\t\t- Directory containing all trace event subsystems:\n" " enable\t\t- Write 0/1 to enable/disable tracing of all events\n" " events/<system>/\t- Directory containing all trace events for <system>:\n" @@ -5699,7 +5685,7 @@ static const char readme_msg[] = "\t table using the key(s) and value(s) named, and the value of a\n" "\t sum called 'hitcount' is incremented. Keys and values\n" "\t correspond to fields in the event's format description. Keys\n" - "\t can be any field, or the special string 'stacktrace'.\n" + "\t can be any field, or the special string 'common_stacktrace'.\n" "\t Compound keys consisting of up to two fields can be specified\n" "\t by the 'keys' keyword. Values must correspond to numeric\n" "\t fields. Sort keys consisting of up to two fields can be\n" @@ -5757,7 +5743,7 @@ static const char readme_msg[] = #ifdef CONFIG_SYNTH_EVENTS " events/synthetic_events\t- Create/append/remove/show synthetic events\n" "\t Write into this file to define/undefine new synthetic events.\n" - "\t example: echo 'myevent u64 lat; char name[]' >> synthetic_events\n" + "\t example: echo 'myevent u64 lat; char name[]; long[] stack' >> synthetic_events\n" #endif #endif ; @@ -5776,219 +5762,6 @@ static const struct file_operations tracing_readme_fops = { .llseek = generic_file_llseek, }; -static void *saved_tgids_next(struct seq_file *m, void *v, loff_t *pos) -{ - int pid = ++(*pos); - - return trace_find_tgid_ptr(pid); -} - -static void *saved_tgids_start(struct seq_file *m, loff_t *pos) -{ - int pid = *pos; - - return trace_find_tgid_ptr(pid); -} - -static void saved_tgids_stop(struct seq_file *m, void *v) -{ -} - -static int saved_tgids_show(struct seq_file *m, void *v) -{ - int *entry = (int *)v; - int pid = entry - tgid_map; - int tgid = *entry; - - if (tgid == 0) - return SEQ_SKIP; - - seq_printf(m, "%d %d\n", pid, tgid); - return 0; -} - -static const struct seq_operations tracing_saved_tgids_seq_ops = { - .start = saved_tgids_start, - .stop = saved_tgids_stop, - .next = saved_tgids_next, - .show = saved_tgids_show, -}; - -static int tracing_saved_tgids_open(struct inode *inode, struct file *filp) -{ - int ret; - - ret = tracing_check_open_get_tr(NULL); - if (ret) - return ret; - - return seq_open(filp, &tracing_saved_tgids_seq_ops); -} - - -static const struct file_operations tracing_saved_tgids_fops = { - .open = tracing_saved_tgids_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -static void *saved_cmdlines_next(struct seq_file *m, void *v, loff_t *pos) -{ - unsigned int *ptr = v; - - if (*pos || m->count) - ptr++; - - (*pos)++; - - for (; ptr < &savedcmd->map_cmdline_to_pid[savedcmd->cmdline_num]; - ptr++) { - if (*ptr == -1 || *ptr == NO_CMDLINE_MAP) - continue; - - return ptr; - } - - return NULL; -} - -static void *saved_cmdlines_start(struct seq_file *m, loff_t *pos) -{ - void *v; - loff_t l = 0; - - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - - v = &savedcmd->map_cmdline_to_pid[0]; - while (l <= *pos) { - v = saved_cmdlines_next(m, v, &l); - if (!v) - return NULL; - } - - return v; -} - -static void saved_cmdlines_stop(struct seq_file *m, void *v) -{ - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); -} - -static int saved_cmdlines_show(struct seq_file *m, void *v) -{ - char buf[TASK_COMM_LEN]; - unsigned int *pid = v; - - __trace_find_cmdline(*pid, buf); - seq_printf(m, "%d %s\n", *pid, buf); - return 0; -} - -static const struct seq_operations tracing_saved_cmdlines_seq_ops = { - .start = saved_cmdlines_start, - .next = saved_cmdlines_next, - .stop = saved_cmdlines_stop, - .show = saved_cmdlines_show, -}; - -static int tracing_saved_cmdlines_open(struct inode *inode, struct file *filp) -{ - int ret; - - ret = tracing_check_open_get_tr(NULL); - if (ret) - return ret; - - return seq_open(filp, &tracing_saved_cmdlines_seq_ops); -} - -static const struct file_operations tracing_saved_cmdlines_fops = { - .open = tracing_saved_cmdlines_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -static ssize_t -tracing_saved_cmdlines_size_read(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char buf[64]; - int r; - - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - r = scnprintf(buf, sizeof(buf), "%u\n", savedcmd->cmdline_num); - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); - - return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); -} - -static void free_saved_cmdlines_buffer(struct saved_cmdlines_buffer *s) -{ - kfree(s->saved_cmdlines); - kfree(s->map_cmdline_to_pid); - kfree(s); -} - -static int tracing_resize_saved_cmdlines(unsigned int val) -{ - struct saved_cmdlines_buffer *s, *savedcmd_temp; - - s = kmalloc(sizeof(*s), GFP_KERNEL); - if (!s) - return -ENOMEM; - - if (allocate_cmdlines_buffer(val, s) < 0) { - kfree(s); - return -ENOMEM; - } - - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - savedcmd_temp = savedcmd; - savedcmd = s; - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); - free_saved_cmdlines_buffer(savedcmd_temp); - - return 0; -} - -static ssize_t -tracing_saved_cmdlines_size_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - unsigned long val; - int ret; - - ret = kstrtoul_from_user(ubuf, cnt, 10, &val); - if (ret) - return ret; - - /* must have at least 1 entry or less than PID_MAX_DEFAULT */ - if (!val || val > PID_MAX_DEFAULT) - return -EINVAL; - - ret = tracing_resize_saved_cmdlines((unsigned int)val); - if (ret < 0) - return ret; - - *ppos += cnt; - - return cnt; -} - -static const struct file_operations tracing_saved_cmdlines_size_fops = { - .open = tracing_open_generic, - .read = tracing_saved_cmdlines_size_read, - .write = tracing_saved_cmdlines_size_write, -}; - #ifdef CONFIG_TRACE_EVAL_MAP_FILE static union trace_eval_map_item * update_eval_map(union trace_eval_map_item *ptr) @@ -6111,7 +5884,7 @@ trace_insert_eval_map_file(struct module *mod, struct trace_eval_map **start, return; } - mutex_lock(&trace_eval_mutex); + guard(mutex)(&trace_eval_mutex); if (!trace_eval_maps) trace_eval_maps = map_array; @@ -6135,8 +5908,6 @@ trace_insert_eval_map_file(struct module *mod, struct trace_eval_map **start, map_array++; } memset(map_array, 0, sizeof(*map_array)); - - mutex_unlock(&trace_eval_mutex); } static void trace_create_eval_file(struct dentry *d_tracer) @@ -6151,17 +5922,27 @@ static inline void trace_insert_eval_map_file(struct module *mod, struct trace_eval_map **start, int len) { } #endif /* !CONFIG_TRACE_EVAL_MAP_FILE */ -static void trace_insert_eval_map(struct module *mod, - struct trace_eval_map **start, int len) +static void +trace_event_update_with_eval_map(struct module *mod, + struct trace_eval_map **start, + int len) { struct trace_eval_map **map; - if (len <= 0) - return; + /* Always run sanitizer only if btf_type_tag attr exists. */ + if (len <= 0) { + if (!(IS_ENABLED(CONFIG_DEBUG_INFO_BTF) && + IS_ENABLED(CONFIG_PAHOLE_HAS_BTF_TAG) && + __has_attribute(btf_type_tag))) + return; + } map = start; - trace_event_eval_update(map, len); + trace_event_update_all(map, len); + + if (len <= 0) + return; trace_insert_eval_map_file(mod, start, len); } @@ -6174,9 +5955,9 @@ tracing_set_trace_read(struct file *filp, char __user *ubuf, char buf[MAX_TRACER_SIZE+2]; int r; - mutex_lock(&trace_types_lock); - r = sprintf(buf, "%s\n", tr->current_trace->name); - mutex_unlock(&trace_types_lock); + scoped_guard(mutex, &trace_types_lock) { + r = sprintf(buf, "%s\n", tr->current_trace->name); + } return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } @@ -6195,6 +5976,15 @@ static void set_buffer_entries(struct array_buffer *buf, unsigned long val) per_cpu_ptr(buf->data, cpu)->entries = val; } +static void update_buffer_entries(struct array_buffer *buf, int cpu) +{ + if (cpu == RING_BUFFER_ALL_CPUS) { + set_buffer_entries(buf, ring_buffer_size(buf->buffer, 0)); + } else { + per_cpu_ptr(buf->data, cpu)->entries = ring_buffer_size(buf->buffer, cpu); + } +} + #ifdef CONFIG_TRACER_MAX_TRACE /* resize @tr's buffer to the size of @size_tr's entries */ static int resize_buffer_duplicate_size(struct array_buffer *trace_buf, @@ -6233,19 +6023,21 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, * we use the size that was given, and we can forget about * expanding it later. */ - ring_buffer_expanded = true; + trace_set_ring_buffer_expanded(tr); /* May be called before buffers are initialized */ if (!tr->array_buffer.buffer) return 0; + /* Do not allow tracing while resizing ring buffer */ + tracing_stop_tr(tr); + ret = ring_buffer_resize(tr->array_buffer.buffer, size, cpu); if (ret < 0) - return ret; + goto out_start; #ifdef CONFIG_TRACER_MAX_TRACE - if (!(tr->flags & TRACE_ARRAY_FL_GLOBAL) || - !tr->current_trace->use_max_tr) + if (!tr->allocated_snapshot) goto out; ret = ring_buffer_resize(tr->max_buffer.buffer, size, cpu); @@ -6270,53 +6062,191 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, WARN_ON(1); tracing_disabled = 1; } - return ret; + goto out_start; } - if (cpu == RING_BUFFER_ALL_CPUS) - set_buffer_entries(&tr->max_buffer, size); - else - per_cpu_ptr(tr->max_buffer.data, cpu)->entries = size; + update_buffer_entries(&tr->max_buffer, cpu); out: #endif /* CONFIG_TRACER_MAX_TRACE */ - if (cpu == RING_BUFFER_ALL_CPUS) - set_buffer_entries(&tr->array_buffer, size); - else - per_cpu_ptr(tr->array_buffer.data, cpu)->entries = size; - + update_buffer_entries(&tr->array_buffer, cpu); + out_start: + tracing_start_tr(tr); return ret; } ssize_t tracing_resize_ring_buffer(struct trace_array *tr, unsigned long size, int cpu_id) { - int ret; - - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); if (cpu_id != RING_BUFFER_ALL_CPUS) { /* make sure, this cpu is enabled in the mask */ - if (!cpumask_test_cpu(cpu_id, tracing_buffer_mask)) { - ret = -EINVAL; - goto out; - } + if (!cpumask_test_cpu(cpu_id, tracing_buffer_mask)) + return -EINVAL; } - ret = __tracing_resize_ring_buffer(tr, size, cpu_id); - if (ret < 0) - ret = -ENOMEM; + return __tracing_resize_ring_buffer(tr, size, cpu_id); +} -out: - mutex_unlock(&trace_types_lock); +struct trace_mod_entry { + unsigned long mod_addr; + char mod_name[MODULE_NAME_LEN]; +}; - return ret; +struct trace_scratch { + unsigned int clock_id; + unsigned long text_addr; + unsigned long nr_entries; + struct trace_mod_entry entries[]; +}; + +static DEFINE_MUTEX(scratch_mutex); + +static int cmp_mod_entry(const void *key, const void *pivot) +{ + unsigned long addr = (unsigned long)key; + const struct trace_mod_entry *ent = pivot; + + if (addr >= ent[0].mod_addr && addr < ent[1].mod_addr) + return 0; + else + return addr - ent->mod_addr; +} + +/** + * trace_adjust_address() - Adjust prev boot address to current address. + * @tr: Persistent ring buffer's trace_array. + * @addr: Address in @tr which is adjusted. + */ +unsigned long trace_adjust_address(struct trace_array *tr, unsigned long addr) +{ + struct trace_module_delta *module_delta; + struct trace_scratch *tscratch; + struct trace_mod_entry *entry; + unsigned long raddr; + int idx = 0, nr_entries; + + /* If we don't have last boot delta, return the address */ + if (!(tr->flags & TRACE_ARRAY_FL_LAST_BOOT)) + return addr; + + /* tr->module_delta must be protected by rcu. */ + guard(rcu)(); + tscratch = tr->scratch; + /* if there is no tscrach, module_delta must be NULL. */ + module_delta = READ_ONCE(tr->module_delta); + if (!module_delta || !tscratch->nr_entries || + tscratch->entries[0].mod_addr > addr) { + raddr = addr + tr->text_delta; + return __is_kernel(raddr) || is_kernel_core_data(raddr) || + is_kernel_rodata(raddr) ? raddr : addr; + } + + /* Note that entries must be sorted. */ + nr_entries = tscratch->nr_entries; + if (nr_entries == 1 || + tscratch->entries[nr_entries - 1].mod_addr < addr) + idx = nr_entries - 1; + else { + entry = __inline_bsearch((void *)addr, + tscratch->entries, + nr_entries - 1, + sizeof(tscratch->entries[0]), + cmp_mod_entry); + if (entry) + idx = entry - tscratch->entries; + } + + return addr + module_delta->delta[idx]; +} + +#ifdef CONFIG_MODULES +static int save_mod(struct module *mod, void *data) +{ + struct trace_array *tr = data; + struct trace_scratch *tscratch; + struct trace_mod_entry *entry; + unsigned int size; + + tscratch = tr->scratch; + if (!tscratch) + return -1; + size = tr->scratch_size; + + if (struct_size(tscratch, entries, tscratch->nr_entries + 1) > size) + return -1; + + entry = &tscratch->entries[tscratch->nr_entries]; + + tscratch->nr_entries++; + + entry->mod_addr = (unsigned long)mod->mem[MOD_TEXT].base; + strscpy(entry->mod_name, mod->name); + + return 0; +} +#else +static int save_mod(struct module *mod, void *data) +{ + return 0; } +#endif + +static void update_last_data(struct trace_array *tr) +{ + struct trace_module_delta *module_delta; + struct trace_scratch *tscratch; + + if (!(tr->flags & TRACE_ARRAY_FL_BOOT)) + return; + + if (!(tr->flags & TRACE_ARRAY_FL_LAST_BOOT)) + return; + + /* Only if the buffer has previous boot data clear and update it. */ + tr->flags &= ~TRACE_ARRAY_FL_LAST_BOOT; + + /* Reset the module list and reload them */ + if (tr->scratch) { + struct trace_scratch *tscratch = tr->scratch; + + tscratch->clock_id = tr->clock_id; + memset(tscratch->entries, 0, + flex_array_size(tscratch, entries, tscratch->nr_entries)); + tscratch->nr_entries = 0; + + guard(mutex)(&scratch_mutex); + module_for_each_mod(save_mod, tr); + } + + /* + * Need to clear all CPU buffers as there cannot be events + * from the previous boot mixed with events with this boot + * as that will cause a confusing trace. Need to clear all + * CPU buffers, even for those that may currently be offline. + */ + tracing_reset_all_cpus(&tr->array_buffer); + + /* Using current data now */ + tr->text_delta = 0; + + if (!tr->scratch) + return; + tscratch = tr->scratch; + module_delta = READ_ONCE(tr->module_delta); + WRITE_ONCE(tr->module_delta, NULL); + kfree_rcu(module_delta, rcu); + + /* Set the persistent ring buffer meta data to this address */ + tscratch->text_addr = (unsigned long)_text; +} /** * tracing_update_buffers - used by tracing facility to expand ring buffers + * @tr: The tracing instance * * To save on memory when the tracing is never used on a system with it * configured in. The ring buffers are set to a minimum size. But once @@ -6325,24 +6255,20 @@ out: * * This function is to be called when a tracer is about to be used. */ -int tracing_update_buffers(void) +int tracing_update_buffers(struct trace_array *tr) { int ret = 0; - mutex_lock(&trace_types_lock); - if (!ring_buffer_expanded) - ret = __tracing_resize_ring_buffer(&global_trace, trace_buf_size, - RING_BUFFER_ALL_CPUS); - mutex_unlock(&trace_types_lock); + guard(mutex)(&trace_types_lock); + update_last_data(tr); + + if (!tr->ring_buffer_expanded) + ret = __tracing_resize_ring_buffer(tr, trace_buf_size, + RING_BUFFER_ALL_CPUS); return ret; } -struct trace_option_dentry; - -static void -create_trace_option_files(struct trace_array *tr, struct tracer *tracer); - /* * Used to clear out the tracer before deletion of an instance. * Must have trace_types_lock held. @@ -6351,89 +6277,76 @@ static void tracing_set_nop(struct trace_array *tr) { if (tr->current_trace == &nop_trace) return; - + tr->current_trace->enabled--; if (tr->current_trace->reset) tr->current_trace->reset(tr); tr->current_trace = &nop_trace; + tr->current_trace_flags = nop_trace.flags; } static bool tracer_options_updated; -static void add_tracer_options(struct trace_array *tr, struct tracer *t) -{ - /* Only enable if the directory has been created already. */ - if (!tr->dir) - return; - - /* Only create trace option files after update_tracer_options finish */ - if (!tracer_options_updated) - return; - - create_trace_option_files(tr, t); -} - int tracing_set_tracer(struct trace_array *tr, const char *buf) { - struct tracer *t; + struct tracer *trace = NULL; + struct tracers *t; #ifdef CONFIG_TRACER_MAX_TRACE bool had_max_tr; #endif - int ret = 0; + int ret; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); - if (!ring_buffer_expanded) { + update_last_data(tr); + + if (!tr->ring_buffer_expanded) { ret = __tracing_resize_ring_buffer(tr, trace_buf_size, RING_BUFFER_ALL_CPUS); if (ret < 0) - goto out; + return ret; ret = 0; } - for (t = trace_types; t; t = t->next) { - if (strcmp(t->name, buf) == 0) + list_for_each_entry(t, &tr->tracers, list) { + if (strcmp(t->tracer->name, buf) == 0) { + trace = t->tracer; break; + } } - if (!t) { - ret = -EINVAL; - goto out; - } - if (t == tr->current_trace) - goto out; + if (!trace) + return -EINVAL; + + if (trace == tr->current_trace) + return 0; #ifdef CONFIG_TRACER_SNAPSHOT - if (t->use_max_tr) { + if (trace->use_max_tr) { local_irq_disable(); arch_spin_lock(&tr->max_lock); - if (tr->cond_snapshot) - ret = -EBUSY; + ret = tr->cond_snapshot ? -EBUSY : 0; arch_spin_unlock(&tr->max_lock); local_irq_enable(); if (ret) - goto out; + return ret; } #endif /* Some tracers won't work on kernel command line */ - if (system_state < SYSTEM_RUNNING && t->noboot) { + if (system_state < SYSTEM_RUNNING && trace->noboot) { pr_warn("Tracer '%s' is not allowed on command line, ignored\n", - t->name); - goto out; + trace->name); + return -EINVAL; } /* Some tracers are only allowed for the top level buffer */ - if (!trace_ok_for_array(t, tr)) { - ret = -EINVAL; - goto out; - } + if (!trace_ok_for_array(trace, tr)) + return -EINVAL; /* If trace pipe files are being read, we can't change the tracer */ - if (tr->trace_ref) { - ret = -EBUSY; - goto out; - } + if (tr->trace_ref) + return -EBUSY; trace_branch_disable(); @@ -6447,8 +6360,9 @@ int tracing_set_tracer(struct trace_array *tr, const char *buf) /* Current trace needs to be nop_trace before synchronize_rcu */ tr->current_trace = &nop_trace; + tr->current_trace_flags = nop_trace.flags; - if (had_max_tr && !t->use_max_tr) { + if (had_max_tr && !trace->use_max_tr) { /* * We need to make sure that the update_max_tr sees that * current_trace changed to nop_trace to keep it from @@ -6458,30 +6372,37 @@ int tracing_set_tracer(struct trace_array *tr, const char *buf) */ synchronize_rcu(); free_snapshot(tr); + tracing_disarm_snapshot(tr); } - if (t->use_max_tr && !tr->allocated_snapshot) { - ret = tracing_alloc_snapshot_instance(tr); - if (ret < 0) - goto out; + if (!had_max_tr && trace->use_max_tr) { + ret = tracing_arm_snapshot_locked(tr); + if (ret) + return ret; } #else tr->current_trace = &nop_trace; #endif - if (t->init) { - ret = tracer_init(t, tr); - if (ret) - goto out; + tr->current_trace_flags = t->flags ? : t->tracer->flags; + + if (trace->init) { + ret = tracer_init(trace, tr); + if (ret) { +#ifdef CONFIG_TRACER_MAX_TRACE + if (trace->use_max_tr) + tracing_disarm_snapshot(tr); +#endif + tr->current_trace_flags = nop_trace.flags; + return ret; + } } - tr->current_trace = t; + tr->current_trace = trace; tr->current_trace->enabled++; trace_branch_enable(tr); - out: - mutex_unlock(&trace_types_lock); - return ret; + return 0; } static ssize_t @@ -6559,22 +6480,18 @@ tracing_thresh_write(struct file *filp, const char __user *ubuf, struct trace_array *tr = filp->private_data; int ret; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); ret = tracing_nsecs_write(&tracing_thresh, ubuf, cnt, ppos); if (ret < 0) - goto out; + return ret; if (tr->current_trace->update_thresh) { ret = tr->current_trace->update_thresh(tr); if (ret < 0) - goto out; + return ret; } - ret = cnt; -out: - mutex_unlock(&trace_types_lock); - - return ret; + return cnt; } #ifdef CONFIG_TRACER_MAX_TRACE @@ -6583,36 +6500,69 @@ static ssize_t tracing_max_lat_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - return tracing_nsecs_read(filp->private_data, ubuf, cnt, ppos); + struct trace_array *tr = filp->private_data; + + return tracing_nsecs_read(&tr->max_latency, ubuf, cnt, ppos); } static ssize_t tracing_max_lat_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - return tracing_nsecs_write(filp->private_data, ubuf, cnt, ppos); + struct trace_array *tr = filp->private_data; + + return tracing_nsecs_write(&tr->max_latency, ubuf, cnt, ppos); } #endif +static int open_pipe_on_cpu(struct trace_array *tr, int cpu) +{ + if (cpu == RING_BUFFER_ALL_CPUS) { + if (cpumask_empty(tr->pipe_cpumask)) { + cpumask_setall(tr->pipe_cpumask); + return 0; + } + } else if (!cpumask_test_cpu(cpu, tr->pipe_cpumask)) { + cpumask_set_cpu(cpu, tr->pipe_cpumask); + return 0; + } + return -EBUSY; +} + +static void close_pipe_on_cpu(struct trace_array *tr, int cpu) +{ + if (cpu == RING_BUFFER_ALL_CPUS) { + WARN_ON(!cpumask_full(tr->pipe_cpumask)); + cpumask_clear(tr->pipe_cpumask); + } else { + WARN_ON(!cpumask_test_cpu(cpu, tr->pipe_cpumask)); + cpumask_clear_cpu(cpu, tr->pipe_cpumask); + } +} + static int tracing_open_pipe(struct inode *inode, struct file *filp) { struct trace_array *tr = inode->i_private; struct trace_iterator *iter; + int cpu; int ret; ret = tracing_check_open_get_tr(tr); if (ret) return ret; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); + cpu = tracing_get_cpu(inode); + ret = open_pipe_on_cpu(tr, cpu); + if (ret) + goto fail_pipe_on_cpu; /* create a buffer to store the information to pass to userspace */ iter = kzalloc(sizeof(*iter), GFP_KERNEL); if (!iter) { ret = -ENOMEM; - __trace_array_put(tr); - goto out; + goto fail_alloc_iter; } trace_seq_init(&iter->seq); @@ -6626,7 +6576,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) /* trace pipe does not show start of buffer */ cpumask_setall(iter->started); - if (tr->trace_flags & TRACE_ITER_LATENCY_FMT) + if (tr->trace_flags & TRACE_ITER(LATENCY_FMT)) iter->iter_flags |= TRACE_FILE_LAT_FMT; /* Output in nanoseconds only if we are using a clock in nanoseconds. */ @@ -6635,7 +6585,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) iter->tr = tr; iter->array_buffer = &tr->array_buffer; - iter->cpu_file = tracing_get_cpu(inode); + iter->cpu_file = cpu; mutex_init(&iter->mutex); filp->private_data = iter; @@ -6645,14 +6595,15 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) nonseekable_open(inode, filp); tr->trace_ref++; -out: - mutex_unlock(&trace_types_lock); + return ret; fail: kfree(iter); +fail_alloc_iter: + close_pipe_on_cpu(tr, cpu); +fail_pipe_on_cpu: __trace_array_put(tr); - mutex_unlock(&trace_types_lock); return ret; } @@ -6661,18 +6612,15 @@ static int tracing_release_pipe(struct inode *inode, struct file *file) struct trace_iterator *iter = file->private_data; struct trace_array *tr = inode->i_private; - mutex_lock(&trace_types_lock); - - tr->trace_ref--; - - if (iter->trace->pipe_close) - iter->trace->pipe_close(iter); + scoped_guard(mutex, &trace_types_lock) { + tr->trace_ref--; - mutex_unlock(&trace_types_lock); + if (iter->trace->pipe_close) + iter->trace->pipe_close(iter); + close_pipe_on_cpu(tr, iter->cpu_file); + } - free_cpumask_var(iter->started); - kfree(iter->fmt); - mutex_destroy(&iter->mutex); + free_trace_iter_content(iter); kfree(iter); trace_array_put(tr); @@ -6689,7 +6637,7 @@ trace_poll(struct trace_iterator *iter, struct file *filp, poll_table *poll_tabl if (trace_buffer_iter(iter, iter->cpu_file)) return EPOLLIN | EPOLLRDNORM; - if (tr->trace_flags & TRACE_ITER_BLOCK) + if (tr->trace_flags & TRACE_ITER(BLOCK)) /* * Always select as readable when in blocking mode */ @@ -6744,6 +6692,22 @@ static int tracing_wait_pipe(struct file *filp) return 1; } +static bool update_last_data_if_empty(struct trace_array *tr) +{ + if (!(tr->flags & TRACE_ARRAY_FL_LAST_BOOT)) + return false; + + if (!ring_buffer_empty(tr->array_buffer.buffer)) + return false; + + /* + * If the buffer contains the last boot data and all per-cpu + * buffers are empty, reset it from the kernel side. + */ + update_last_data(tr); + return true; +} + /* * Consumer reader. */ @@ -6759,34 +6723,35 @@ tracing_read_pipe(struct file *filp, char __user *ubuf, * This is just a matter of traces coherency, the ring buffer itself * is protected. */ - mutex_lock(&iter->mutex); + guard(mutex)(&iter->mutex); /* return any leftover data */ sret = trace_seq_to_user(&iter->seq, ubuf, cnt); if (sret != -EBUSY) - goto out; + return sret; trace_seq_init(&iter->seq); if (iter->trace->read) { sret = iter->trace->read(iter, filp, ubuf, cnt, ppos); if (sret) - goto out; + return sret; } waitagain: + if (update_last_data_if_empty(iter->tr)) + return 0; + sret = tracing_wait_pipe(filp); if (sret <= 0) - goto out; + return sret; /* stop when tracing is finished */ - if (trace_empty(iter)) { - sret = 0; - goto out; - } + if (trace_empty(iter)) + return 0; - if (cnt >= PAGE_SIZE) - cnt = PAGE_SIZE - 1; + if (cnt >= TRACE_SEQ_BUFFER_SIZE) + cnt = TRACE_SEQ_BUFFER_SIZE - 1; /* reset all but tr, trace, and overruns */ trace_iterator_reset(iter); @@ -6837,7 +6802,7 @@ waitagain: /* Now copy what we have to the user */ sret = trace_seq_to_user(&iter->seq, ubuf, cnt); - if (iter->seq.seq.readpos >= trace_seq_used(&iter->seq)) + if (iter->seq.readpos >= trace_seq_used(&iter->seq)) trace_seq_init(&iter->seq); /* @@ -6847,9 +6812,6 @@ waitagain: if (sret == -EBUSY) goto waitagain; -out: - mutex_unlock(&iter->mutex); - return sret; } @@ -6962,13 +6924,14 @@ static ssize_t tracing_splice_read_pipe(struct file *filp, /* Copy the data into the page, so we can start over. */ ret = trace_seq_to_buffer(&iter->seq, page_address(spd.pages[i]), - trace_seq_used(&iter->seq)); + min((size_t)trace_seq_used(&iter->seq), + (size_t)PAGE_SIZE)); if (ret < 0) { __free_page(spd.pages[i]); break; } spd.partial[i].offset = 0; - spd.partial[i].len = trace_seq_used(&iter->seq); + spd.partial[i].len = ret; trace_seq_init(&iter->seq); } @@ -6993,6 +6956,43 @@ out_err: } static ssize_t +tracing_syscall_buf_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; + char buf[64]; + int r; + + r = snprintf(buf, 64, "%d\n", tr->syscall_buf_sz); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +static ssize_t +tracing_syscall_buf_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; + unsigned long val; + int ret; + + ret = kstrtoul_from_user(ubuf, cnt, 10, &val); + if (ret) + return ret; + + if (val > SYSCALL_FAULT_USER_MAX) + val = SYSCALL_FAULT_USER_MAX; + + tr->syscall_buf_sz = val; + + *ppos += cnt; + + return cnt; +} + +static ssize_t tracing_entries_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { @@ -7023,7 +7023,7 @@ tracing_entries_read(struct file *filp, char __user *ubuf, } if (buf_size_same) { - if (!ring_buffer_expanded) + if (!tr->ring_buffer_expanded) r = sprintf(buf, "%lu (expanded: %lu)\n", size >> 10, trace_buf_size >> 10); @@ -7080,10 +7080,10 @@ tracing_total_entries_read(struct file *filp, char __user *ubuf, mutex_lock(&trace_types_lock); for_each_tracing_cpu(cpu) { size += per_cpu_ptr(tr->array_buffer.data, cpu)->entries >> 10; - if (!ring_buffer_expanded) + if (!tr->ring_buffer_expanded) expanded_size += trace_buf_size >> 10; } - if (ring_buffer_expanded) + if (tr->ring_buffer_expanded) r = sprintf(buf, "%lu\n", size); else r = sprintf(buf, "%lu (expanded: %lu)\n", size, expanded_size); @@ -7092,6 +7092,120 @@ tracing_total_entries_read(struct file *filp, char __user *ubuf, return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } +#define LAST_BOOT_HEADER ((void *)1) + +static void *l_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct trace_array *tr = m->private; + struct trace_scratch *tscratch = tr->scratch; + unsigned int index = *pos; + + (*pos)++; + + if (*pos == 1) + return LAST_BOOT_HEADER; + + /* Only show offsets of the last boot data */ + if (!tscratch || !(tr->flags & TRACE_ARRAY_FL_LAST_BOOT)) + return NULL; + + /* *pos 0 is for the header, 1 is for the first module */ + index--; + + if (index >= tscratch->nr_entries) + return NULL; + + return &tscratch->entries[index]; +} + +static void *l_start(struct seq_file *m, loff_t *pos) +{ + mutex_lock(&scratch_mutex); + + return l_next(m, NULL, pos); +} + +static void l_stop(struct seq_file *m, void *p) +{ + mutex_unlock(&scratch_mutex); +} + +static void show_last_boot_header(struct seq_file *m, struct trace_array *tr) +{ + struct trace_scratch *tscratch = tr->scratch; + + /* + * Do not leak KASLR address. This only shows the KASLR address of + * the last boot. When the ring buffer is started, the LAST_BOOT + * flag gets cleared, and this should only report "current". + * Otherwise it shows the KASLR address from the previous boot which + * should not be the same as the current boot. + */ + if (tscratch && (tr->flags & TRACE_ARRAY_FL_LAST_BOOT)) + seq_printf(m, "%lx\t[kernel]\n", tscratch->text_addr); + else + seq_puts(m, "# Current\n"); +} + +static int l_show(struct seq_file *m, void *v) +{ + struct trace_array *tr = m->private; + struct trace_mod_entry *entry = v; + + if (v == LAST_BOOT_HEADER) { + show_last_boot_header(m, tr); + return 0; + } + + seq_printf(m, "%lx\t%s\n", entry->mod_addr, entry->mod_name); + return 0; +} + +static const struct seq_operations last_boot_seq_ops = { + .start = l_start, + .next = l_next, + .stop = l_stop, + .show = l_show, +}; + +static int tracing_last_boot_open(struct inode *inode, struct file *file) +{ + struct trace_array *tr = inode->i_private; + struct seq_file *m; + int ret; + + ret = tracing_check_open_get_tr(tr); + if (ret) + return ret; + + ret = seq_open(file, &last_boot_seq_ops); + if (ret) { + trace_array_put(tr); + return ret; + } + + m = file->private_data; + m->private = tr; + + return 0; +} + +static int tracing_buffer_meta_open(struct inode *inode, struct file *filp) +{ + struct trace_array *tr = inode->i_private; + int cpu = tracing_get_cpu(inode); + int ret; + + ret = tracing_check_open_get_tr(tr); + if (ret) + return ret; + + ret = ring_buffer_meta_seq_init(filp, tr->array_buffer.buffer, cpu); + if (ret < 0) + __trace_array_put(tr); + return ret; +} + static ssize_t tracing_free_buffer_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) @@ -7112,7 +7226,7 @@ tracing_free_buffer_release(struct inode *inode, struct file *filp) struct trace_array *tr = inode->i_private; /* disable tracing ? */ - if (tr->trace_flags & TRACE_ITER_STOP_ON_FREE) + if (tr->trace_flags & TRACE_ITER(STOP_ON_FREE)) tracer_tracing_off(tr); /* resize the ring buffer to 0 */ tracing_resize_ring_buffer(tr, 0, RING_BUFFER_ALL_CPUS); @@ -7122,57 +7236,47 @@ tracing_free_buffer_release(struct inode *inode, struct file *filp) return 0; } -static ssize_t -tracing_mark_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *fpos) +#define TRACE_MARKER_MAX_SIZE 4096 + +static ssize_t write_marker_to_buffer(struct trace_array *tr, const char *buf, + size_t cnt, unsigned long ip) { - struct trace_array *tr = filp->private_data; struct ring_buffer_event *event; enum event_trigger_type tt = ETT_NONE; struct trace_buffer *buffer; struct print_entry *entry; + int meta_size; ssize_t written; - int size; - int len; - -/* Used in tracing_mark_raw_write() as well */ -#define FAULTED_STR "<faulted>" -#define FAULTED_SIZE (sizeof(FAULTED_STR) - 1) /* '\0' is already accounted for */ - - if (tracing_disabled) - return -EINVAL; - - if (!(tr->trace_flags & TRACE_ITER_MARKERS)) - return -EINVAL; - - if (cnt > TRACE_BUF_SIZE) - cnt = TRACE_BUF_SIZE; - - BUILD_BUG_ON(TRACE_BUF_SIZE >= PAGE_SIZE); - - size = sizeof(*entry) + cnt + 2; /* add '\0' and possible '\n' */ + size_t size; - /* If less than "<faulted>", then make sure we can still add that */ - if (cnt < FAULTED_SIZE) - size += FAULTED_SIZE - cnt; + meta_size = sizeof(*entry) + 2; /* add '\0' and possible '\n' */ + again: + size = cnt + meta_size; buffer = tr->array_buffer.buffer; event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, tracing_gen_ctx()); - if (unlikely(!event)) + if (unlikely(!event)) { + /* + * If the size was greater than what was allowed, then + * make it smaller and try again. + */ + if (size > ring_buffer_max_event_size(buffer)) { + cnt = ring_buffer_max_event_size(buffer) - meta_size; + /* The above should only happen once */ + if (WARN_ON_ONCE(cnt + meta_size == size)) + return -EBADF; + goto again; + } + /* Ring buffer disabled, return as if not open for write */ return -EBADF; + } entry = ring_buffer_event_data(event); - entry->ip = _THIS_IP_; - - len = __copy_from_user_inatomic(&entry->buf, ubuf, cnt); - if (len) { - memcpy(&entry->buf, FAULTED_STR, FAULTED_SIZE); - cnt = FAULTED_SIZE; - written = -EFAULT; - } else - written = cnt; + entry->ip = ip; + memcpy(&entry->buf, buf, cnt); + written = cnt; if (tr->trace_marker_file && !list_empty(&tr->trace_marker_file->triggers)) { /* do not add \n before testing triggers, but add \0 */ @@ -7196,43 +7300,357 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, return written; } -/* Limit it for now to 3K (including tag) */ -#define RAW_DATA_MAX_SIZE (1024*3) +struct trace_user_buf { + char *buf; +}; + +static DEFINE_MUTEX(trace_user_buffer_mutex); +static struct trace_user_buf_info *trace_user_buffer; + +/** + * trace_user_fault_destroy - free up allocated memory of a trace user buffer + * @tinfo: The descriptor to free up + * + * Frees any data allocated in the trace info dsecriptor. + */ +void trace_user_fault_destroy(struct trace_user_buf_info *tinfo) +{ + char *buf; + int cpu; + + if (!tinfo || !tinfo->tbuf) + return; + + for_each_possible_cpu(cpu) { + buf = per_cpu_ptr(tinfo->tbuf, cpu)->buf; + kfree(buf); + } + free_percpu(tinfo->tbuf); +} + +static int user_fault_buffer_enable(struct trace_user_buf_info *tinfo, size_t size) +{ + char *buf; + int cpu; + + lockdep_assert_held(&trace_user_buffer_mutex); + + tinfo->tbuf = alloc_percpu(struct trace_user_buf); + if (!tinfo->tbuf) + return -ENOMEM; + + tinfo->ref = 1; + tinfo->size = size; + + /* Clear each buffer in case of error */ + for_each_possible_cpu(cpu) { + per_cpu_ptr(tinfo->tbuf, cpu)->buf = NULL; + } + + for_each_possible_cpu(cpu) { + buf = kmalloc_node(size, GFP_KERNEL, + cpu_to_node(cpu)); + if (!buf) + return -ENOMEM; + per_cpu_ptr(tinfo->tbuf, cpu)->buf = buf; + } + + return 0; +} + +/* For internal use. Free and reinitialize */ +static void user_buffer_free(struct trace_user_buf_info **tinfo) +{ + lockdep_assert_held(&trace_user_buffer_mutex); + + trace_user_fault_destroy(*tinfo); + kfree(*tinfo); + *tinfo = NULL; +} + +/* For internal use. Initialize and allocate */ +static int user_buffer_init(struct trace_user_buf_info **tinfo, size_t size) +{ + bool alloc = false; + int ret; + + lockdep_assert_held(&trace_user_buffer_mutex); + + if (!*tinfo) { + alloc = true; + *tinfo = kzalloc(sizeof(**tinfo), GFP_KERNEL); + if (!*tinfo) + return -ENOMEM; + } + + ret = user_fault_buffer_enable(*tinfo, size); + if (ret < 0 && alloc) + user_buffer_free(tinfo); + + return ret; +} + +/* For internal use, derefrence and free if necessary */ +static void user_buffer_put(struct trace_user_buf_info **tinfo) +{ + guard(mutex)(&trace_user_buffer_mutex); + + if (WARN_ON_ONCE(!*tinfo || !(*tinfo)->ref)) + return; + + if (--(*tinfo)->ref) + return; + + user_buffer_free(tinfo); +} + +/** + * trace_user_fault_init - Allocated or reference a per CPU buffer + * @tinfo: A pointer to the trace buffer descriptor + * @size: The size to allocate each per CPU buffer + * + * Create a per CPU buffer that can be used to copy from user space + * in a task context. When calling trace_user_fault_read(), preemption + * must be disabled, and it will enable preemption and copy user + * space data to the buffer. If any schedule switches occur, it will + * retry until it succeeds without a schedule switch knowing the buffer + * is still valid. + * + * Returns 0 on success, negative on failure. + */ +int trace_user_fault_init(struct trace_user_buf_info *tinfo, size_t size) +{ + int ret; + + if (!tinfo) + return -EINVAL; + + guard(mutex)(&trace_user_buffer_mutex); + + ret = user_buffer_init(&tinfo, size); + if (ret < 0) + trace_user_fault_destroy(tinfo); + + return ret; +} + +/** + * trace_user_fault_get - up the ref count for the user buffer + * @tinfo: A pointer to a pointer to the trace buffer descriptor + * + * Ups the ref count of the trace buffer. + * + * Returns the new ref count. + */ +int trace_user_fault_get(struct trace_user_buf_info *tinfo) +{ + if (!tinfo) + return -1; + + guard(mutex)(&trace_user_buffer_mutex); + + tinfo->ref++; + return tinfo->ref; +} + +/** + * trace_user_fault_put - dereference a per cpu trace buffer + * @tinfo: The @tinfo that was passed to trace_user_fault_get() + * + * Decrement the ref count of @tinfo. + * + * Returns the new refcount (negative on error). + */ +int trace_user_fault_put(struct trace_user_buf_info *tinfo) +{ + guard(mutex)(&trace_user_buffer_mutex); + + if (WARN_ON_ONCE(!tinfo || !tinfo->ref)) + return -1; + + --tinfo->ref; + return tinfo->ref; +} + +/** + * trace_user_fault_read - Read user space into a per CPU buffer + * @tinfo: The @tinfo allocated by trace_user_fault_get() + * @ptr: The user space pointer to read + * @size: The size of user space to read. + * @copy_func: Optional function to use to copy from user space + * @data: Data to pass to copy_func if it was supplied + * + * Preemption must be disabled when this is called, and must not + * be enabled while using the returned buffer. + * This does the copying from user space into a per CPU buffer. + * + * The @size must not be greater than the size passed in to + * trace_user_fault_init(). + * + * If @copy_func is NULL, trace_user_fault_read() will use copy_from_user(), + * otherwise it will call @copy_func. It will call @copy_func with: + * + * buffer: the per CPU buffer of the @tinfo. + * ptr: The pointer @ptr to user space to read + * size: The @size of the ptr to read + * data: The @data parameter + * + * It is expected that @copy_func will return 0 on success and non zero + * if there was a fault. + * + * Returns a pointer to the buffer with the content read from @ptr. + * Preemption must remain disabled while the caller accesses the + * buffer returned by this function. + * Returns NULL if there was a fault, or the size passed in is + * greater than the size passed to trace_user_fault_init(). + */ +char *trace_user_fault_read(struct trace_user_buf_info *tinfo, + const char __user *ptr, size_t size, + trace_user_buf_copy copy_func, void *data) +{ + int cpu = smp_processor_id(); + char *buffer = per_cpu_ptr(tinfo->tbuf, cpu)->buf; + unsigned int cnt; + int trys = 0; + int ret; + + lockdep_assert_preemption_disabled(); + + /* + * It's up to the caller to not try to copy more than it said + * it would. + */ + if (size > tinfo->size) + return NULL; + + /* + * This acts similar to a seqcount. The per CPU context switches are + * recorded, migration is disabled and preemption is enabled. The + * read of the user space memory is copied into the per CPU buffer. + * Preemption is disabled again, and if the per CPU context switches count + * is still the same, it means the buffer has not been corrupted. + * If the count is different, it is assumed the buffer is corrupted + * and reading must be tried again. + */ + + do { + /* + * If for some reason, copy_from_user() always causes a context + * switch, this would then cause an infinite loop. + * If this task is preempted by another user space task, it + * will cause this task to try again. But just in case something + * changes where the copying from user space causes another task + * to run, prevent this from going into an infinite loop. + * 100 tries should be plenty. + */ + if (WARN_ONCE(trys++ > 100, "Error: Too many tries to read user space")) + return NULL; + + /* Read the current CPU context switch counter */ + cnt = nr_context_switches_cpu(cpu); + + /* + * Preemption is going to be enabled, but this task must + * remain on this CPU. + */ + migrate_disable(); + + /* + * Now preemption is being enabled and another task can come in + * and use the same buffer and corrupt our data. + */ + preempt_enable_notrace(); + + /* Make sure preemption is enabled here */ + lockdep_assert_preemption_enabled(); + + if (copy_func) { + ret = copy_func(buffer, ptr, size, data); + } else { + ret = __copy_from_user(buffer, ptr, size); + } + + preempt_disable_notrace(); + migrate_enable(); + + /* if it faulted, no need to test if the buffer was corrupted */ + if (ret) + return NULL; + + /* + * Preemption is disabled again, now check the per CPU context + * switch counter. If it doesn't match, then another user space + * process may have schedule in and corrupted our buffer. In that + * case the copying must be retried. + */ + } while (nr_context_switches_cpu(cpu) != cnt); + + return buffer; +} static ssize_t -tracing_mark_raw_write(struct file *filp, const char __user *ubuf, +tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) { struct trace_array *tr = filp->private_data; - struct ring_buffer_event *event; - struct trace_buffer *buffer; - struct raw_data_entry *entry; - ssize_t written; - int size; - int len; - -#define FAULT_SIZE_ID (FAULTED_SIZE + sizeof(int)) + ssize_t written = -ENODEV; + unsigned long ip; + char *buf; if (tracing_disabled) return -EINVAL; - if (!(tr->trace_flags & TRACE_ITER_MARKERS)) + if (!(tr->trace_flags & TRACE_ITER(MARKERS))) return -EINVAL; - /* The marker must at least have a tag id */ - if (cnt < sizeof(unsigned int) || cnt > RAW_DATA_MAX_SIZE) + if ((ssize_t)cnt < 0) return -EINVAL; - if (cnt > TRACE_BUF_SIZE) - cnt = TRACE_BUF_SIZE; + if (cnt > TRACE_MARKER_MAX_SIZE) + cnt = TRACE_MARKER_MAX_SIZE; + + /* Must have preemption disabled while having access to the buffer */ + guard(preempt_notrace)(); + + buf = trace_user_fault_read(trace_user_buffer, ubuf, cnt, NULL, NULL); + if (!buf) + return -EFAULT; + + /* The selftests expect this function to be the IP address */ + ip = _THIS_IP_; - BUILD_BUG_ON(TRACE_BUF_SIZE >= PAGE_SIZE); + /* The global trace_marker can go to multiple instances */ + if (tr == &global_trace) { + guard(rcu)(); + list_for_each_entry_rcu(tr, &marker_copies, marker_list) { + written = write_marker_to_buffer(tr, buf, cnt, ip); + if (written < 0) + break; + } + } else { + written = write_marker_to_buffer(tr, buf, cnt, ip); + } - size = sizeof(*entry) + cnt; - if (cnt < FAULT_SIZE_ID) - size += FAULT_SIZE_ID - cnt; + return written; +} + +static ssize_t write_raw_marker_to_buffer(struct trace_array *tr, + const char *buf, size_t cnt) +{ + struct ring_buffer_event *event; + struct trace_buffer *buffer; + struct raw_data_entry *entry; + ssize_t written; + size_t size; + + /* cnt includes both the entry->id and the data behind it. */ + size = struct_offset(entry, id) + cnt; buffer = tr->array_buffer.buffer; + + if (size > ring_buffer_max_event_size(buffer)) + return -EINVAL; + event = __trace_buffer_lock_reserve(buffer, TRACE_RAW_DATA, size, tracing_gen_ctx()); if (!event) @@ -7240,20 +7658,88 @@ tracing_mark_raw_write(struct file *filp, const char __user *ubuf, return -EBADF; entry = ring_buffer_event_data(event); - - len = __copy_from_user_inatomic(&entry->id, ubuf, cnt); - if (len) { - entry->id = -1; - memcpy(&entry->buf, FAULTED_STR, FAULTED_SIZE); - written = -EFAULT; - } else - written = cnt; + unsafe_memcpy(&entry->id, buf, cnt, + "id and content already reserved on ring buffer" + "'buf' includes the 'id' and the data." + "'entry' was allocated with cnt from 'id'."); + written = cnt; __buffer_unlock_commit(buffer, event); return written; } +static ssize_t +tracing_mark_raw_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *fpos) +{ + struct trace_array *tr = filp->private_data; + ssize_t written = -ENODEV; + char *buf; + + if (tracing_disabled) + return -EINVAL; + + if (!(tr->trace_flags & TRACE_ITER(MARKERS))) + return -EINVAL; + + /* The marker must at least have a tag id */ + if (cnt < sizeof(unsigned int)) + return -EINVAL; + + /* raw write is all or nothing */ + if (cnt > TRACE_MARKER_MAX_SIZE) + return -EINVAL; + + /* Must have preemption disabled while having access to the buffer */ + guard(preempt_notrace)(); + + buf = trace_user_fault_read(trace_user_buffer, ubuf, cnt, NULL, NULL); + if (!buf) + return -EFAULT; + + /* The global trace_marker_raw can go to multiple instances */ + if (tr == &global_trace) { + guard(rcu)(); + list_for_each_entry_rcu(tr, &marker_copies, marker_list) { + written = write_raw_marker_to_buffer(tr, buf, cnt); + if (written < 0) + break; + } + } else { + written = write_raw_marker_to_buffer(tr, buf, cnt); + } + + return written; +} + +static int tracing_mark_open(struct inode *inode, struct file *filp) +{ + int ret; + + scoped_guard(mutex, &trace_user_buffer_mutex) { + if (!trace_user_buffer) { + ret = user_buffer_init(&trace_user_buffer, TRACE_MARKER_MAX_SIZE); + if (ret < 0) + return ret; + } else { + trace_user_buffer->ref++; + } + } + + stream_open(inode, filp); + ret = tracing_open_generic_tr(inode, filp); + if (ret < 0) + user_buffer_put(&trace_user_buffer); + return ret; +} + +static int tracing_mark_release(struct inode *inode, struct file *file) +{ + user_buffer_put(&trace_user_buffer); + return tracing_release_generic_tr(inode, file); +} + static int tracing_clock_show(struct seq_file *m, void *v) { struct trace_array *tr = m->private; @@ -7280,7 +7766,7 @@ int tracing_set_clock(struct trace_array *tr, const char *clockstr) if (i == ARRAY_SIZE(trace_clocks)) return -EINVAL; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); tr->clock_id = i; @@ -7298,7 +7784,11 @@ int tracing_set_clock(struct trace_array *tr, const char *clockstr) tracing_reset_online_cpus(&tr->max_buffer); #endif - mutex_unlock(&trace_types_lock); + if (tr->scratch && !(tr->flags & TRACE_ARRAY_FL_LAST_BOOT)) { + struct trace_scratch *tscratch = tr->scratch; + + tscratch->clock_id = i; + } return 0; } @@ -7351,15 +7841,13 @@ static int tracing_time_stamp_mode_show(struct seq_file *m, void *v) { struct trace_array *tr = m->private; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); if (ring_buffer_time_stamp_abs(tr->array_buffer.buffer)) seq_puts(m, "delta [absolute]\n"); else seq_puts(m, "[delta] absolute\n"); - mutex_unlock(&trace_types_lock); - return 0; } @@ -7392,31 +7880,26 @@ u64 tracing_event_time_stamp(struct trace_buffer *buffer, struct ring_buffer_eve */ int tracing_set_filter_buffering(struct trace_array *tr, bool set) { - int ret = 0; - - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); if (set && tr->no_filter_buffering_ref++) - goto out; + return 0; if (!set) { - if (WARN_ON_ONCE(!tr->no_filter_buffering_ref)) { - ret = -EINVAL; - goto out; - } + if (WARN_ON_ONCE(!tr->no_filter_buffering_ref)) + return -EINVAL; --tr->no_filter_buffering_ref; } - out: - mutex_unlock(&trace_types_lock); - return ret; + return 0; } struct ftrace_buffer_info { struct trace_iterator iter; void *spare; unsigned int spare_cpu; + unsigned int spare_size; unsigned int read; }; @@ -7462,6 +7945,11 @@ out: return ret; } +static void tracing_swap_cpu_buffer(void *tr) +{ + update_max_tr_single((struct trace_array *)tr, current, smp_processor_id()); +} + static ssize_t tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) @@ -7472,7 +7960,7 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, unsigned long val; int ret; - ret = tracing_update_buffers(); + ret = tracing_update_buffers(tr); if (ret < 0) return ret; @@ -7480,12 +7968,10 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, if (ret) return ret; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); - if (tr->current_trace->use_max_tr) { - ret = -EBUSY; - goto out; - } + if (tr->current_trace->use_max_tr) + return -EBUSY; local_irq_disable(); arch_spin_lock(&tr->max_lock); @@ -7494,39 +7980,39 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, arch_spin_unlock(&tr->max_lock); local_irq_enable(); if (ret) - goto out; + return ret; switch (val) { case 0: - if (iter->cpu_file != RING_BUFFER_ALL_CPUS) { - ret = -EINVAL; - break; - } + if (iter->cpu_file != RING_BUFFER_ALL_CPUS) + return -EINVAL; if (tr->allocated_snapshot) free_snapshot(tr); break; case 1: /* Only allow per-cpu swap if the ring buffer supports it */ #ifndef CONFIG_RING_BUFFER_ALLOW_SWAP - if (iter->cpu_file != RING_BUFFER_ALL_CPUS) { - ret = -EINVAL; - break; - } + if (iter->cpu_file != RING_BUFFER_ALL_CPUS) + return -EINVAL; #endif if (tr->allocated_snapshot) ret = resize_buffer_duplicate_size(&tr->max_buffer, &tr->array_buffer, iter->cpu_file); - else - ret = tracing_alloc_snapshot_instance(tr); - if (ret < 0) - break; - local_irq_disable(); + + ret = tracing_arm_snapshot_locked(tr); + if (ret) + return ret; + /* Now, we're going to swap */ - if (iter->cpu_file == RING_BUFFER_ALL_CPUS) + if (iter->cpu_file == RING_BUFFER_ALL_CPUS) { + local_irq_disable(); update_max_tr(tr, current, smp_processor_id(), NULL); - else - update_max_tr_single(tr, current, iter->cpu_file); - local_irq_enable(); + local_irq_enable(); + } else { + smp_call_function_single(iter->cpu_file, tracing_swap_cpu_buffer, + (void *)tr, 1); + } + tracing_disarm_snapshot(tr); break; default: if (tr->allocated_snapshot) { @@ -7542,8 +8028,7 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, *ppos += cnt; ret = cnt; } -out: - mutex_unlock(&trace_types_lock); + return ret; } @@ -7607,18 +8092,20 @@ static const struct file_operations tracing_thresh_fops = { #ifdef CONFIG_TRACER_MAX_TRACE static const struct file_operations tracing_max_lat_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = tracing_max_lat_read, .write = tracing_max_lat_write, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; #endif static const struct file_operations set_tracer_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = tracing_set_trace_read, .write = tracing_set_trace_write, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; static const struct file_operations tracing_pipe_fops = { @@ -7627,7 +8114,6 @@ static const struct file_operations tracing_pipe_fops = { .read = tracing_read_pipe, .splice_read = tracing_splice_read_pipe, .release = tracing_release_pipe, - .llseek = no_llseek, }; static const struct file_operations tracing_entries_fops = { @@ -7638,6 +8124,21 @@ static const struct file_operations tracing_entries_fops = { .release = tracing_release_generic_tr, }; +static const struct file_operations tracing_syscall_buf_fops = { + .open = tracing_open_generic_tr, + .read = tracing_syscall_buf_read, + .write = tracing_syscall_buf_write, + .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, +}; + +static const struct file_operations tracing_buffer_meta_fops = { + .open = tracing_buffer_meta_open, + .read = seq_read, + .llseek = seq_lseek, + .release = tracing_seq_release, +}; + static const struct file_operations tracing_total_entries_fops = { .open = tracing_open_generic_tr, .read = tracing_total_entries_read, @@ -7654,13 +8155,13 @@ static const struct file_operations tracing_free_buffer_fops = { static const struct file_operations tracing_mark_fops = { .open = tracing_mark_open, .write = tracing_mark_write, - .release = tracing_release_generic_tr, + .release = tracing_mark_release, }; static const struct file_operations tracing_mark_raw_fops = { .open = tracing_mark_open, .write = tracing_mark_raw_write, - .release = tracing_release_generic_tr, + .release = tracing_mark_release, }; static const struct file_operations trace_clock_fops = { @@ -7678,6 +8179,13 @@ static const struct file_operations trace_time_stamp_mode_fops = { .release = tracing_single_release_tr, }; +static const struct file_operations last_boot_fops = { + .open = tracing_last_boot_open, + .read = seq_read, + .llseek = seq_lseek, + .release = tracing_seq_release, +}; + #ifdef CONFIG_TRACER_SNAPSHOT static const struct file_operations snapshot_fops = { .open = tracing_snapshot_open, @@ -7692,7 +8200,6 @@ static const struct file_operations snapshot_raw_fops = { .read = tracing_buffers_read, .release = tracing_buffers_release, .splice_read = tracing_buffers_splice_read, - .llseek = no_llseek, }; #endif /* CONFIG_TRACER_SNAPSHOT */ @@ -7915,12 +8422,11 @@ void tracing_log_err(struct trace_array *tr, len += sizeof(CMD_PREFIX) + 2 * sizeof("\n") + strlen(cmd) + 1; - mutex_lock(&tracing_err_log_lock); + guard(mutex)(&tracing_err_log_lock); + err = get_tracing_log_err(tr, len); - if (PTR_ERR(err) == -ENOMEM) { - mutex_unlock(&tracing_err_log_lock); + if (PTR_ERR(err) == -ENOMEM) return; - } snprintf(err->loc, TRACING_LOG_LOC_MAX, "%s: error: ", loc); snprintf(err->cmd, len, "\n" CMD_PREFIX "%s\n", cmd); @@ -7931,21 +8437,20 @@ void tracing_log_err(struct trace_array *tr, err->info.ts = local_clock(); list_add_tail(&err->list, &tr->err_log); - mutex_unlock(&tracing_err_log_lock); } static void clear_tracing_err_log(struct trace_array *tr) { struct tracing_log_err *err, *next; - mutex_lock(&tracing_err_log_lock); + guard(mutex)(&tracing_err_log_lock); + list_for_each_entry_safe(err, next, &tr->err_log, list) { list_del(&err->list); free_tracing_log_err(err); } tr->n_err_log_entries = 0; - mutex_unlock(&tracing_err_log_lock); } static void *tracing_err_log_seq_start(struct seq_file *m, loff_t *pos) @@ -8054,7 +8559,7 @@ static const struct file_operations tracing_err_log_fops = { .open = tracing_err_log_open, .write = tracing_err_log_write, .read = seq_read, - .llseek = seq_lseek, + .llseek = tracing_lseek, .release = tracing_err_log_release, }; @@ -8112,6 +8617,8 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, { struct ftrace_buffer_info *info = filp->private_data; struct trace_iterator *iter = &info->iter; + void *trace_data; + int page_size; ssize_t ret = 0; ssize_t size; @@ -8123,6 +8630,17 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, return -EBUSY; #endif + page_size = ring_buffer_subbuf_size_get(iter->array_buffer->buffer); + + /* Make sure the spare matches the current sub buffer size */ + if (info->spare) { + if (page_size != info->spare_size) { + ring_buffer_free_read_page(iter->array_buffer->buffer, + info->spare_cpu, info->spare); + info->spare = NULL; + } + } + if (!info->spare) { info->spare = ring_buffer_alloc_read_page(iter->array_buffer->buffer, iter->cpu_file); @@ -8131,25 +8649,29 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, info->spare = NULL; } else { info->spare_cpu = iter->cpu_file; + info->spare_size = page_size; } } if (!info->spare) return ret; /* Do we have previous read data to read? */ - if (info->read < PAGE_SIZE) + if (info->read < page_size) goto read; again: trace_access_lock(iter->cpu_file); ret = ring_buffer_read_page(iter->array_buffer->buffer, - &info->spare, + info->spare, count, iter->cpu_file, 0); trace_access_unlock(iter->cpu_file); if (ret < 0) { - if (trace_empty(iter)) { + if (trace_empty(iter) && !iter->closed) { + if (update_last_data_if_empty(iter->tr)) + return 0; + if ((filp->f_flags & O_NONBLOCK)) return -EAGAIN; @@ -8164,11 +8686,11 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, info->read = 0; read: - size = PAGE_SIZE - info->read; + size = page_size - info->read; if (size > count) size = count; - - ret = copy_to_user(ubuf, info->spare + info->read, size); + trace_data = ring_buffer_read_page_data(info->spare); + ret = copy_to_user(ubuf, trace_data + info->read, size); if (ret == size) return -EFAULT; @@ -8180,30 +8702,36 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, return size; } +static int tracing_buffers_flush(struct file *file, fl_owner_t id) +{ + struct ftrace_buffer_info *info = file->private_data; + struct trace_iterator *iter = &info->iter; + + iter->closed = true; + /* Make sure the waiters see the new wait_index */ + (void)atomic_fetch_inc_release(&iter->wait_index); + + ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); + + return 0; +} + static int tracing_buffers_release(struct inode *inode, struct file *file) { struct ftrace_buffer_info *info = file->private_data; struct trace_iterator *iter = &info->iter; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); iter->tr->trace_ref--; __trace_array_put(iter->tr); - iter->wait_index++; - /* Make sure the waiters see the new wait_index */ - smp_wmb(); - - ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); - if (info->spare) ring_buffer_free_read_page(iter->array_buffer->buffer, info->spare_cpu, info->spare); kvfree(info); - mutex_unlock(&trace_types_lock); - return 0; } @@ -8279,6 +8807,8 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, .spd_release = buffer_spd_release, }; struct buffer_ref *ref; + bool woken = false; + int page_size; int entries, i; ssize_t ret = 0; @@ -8287,13 +8817,14 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, return -EBUSY; #endif - if (*ppos & (PAGE_SIZE - 1)) + page_size = ring_buffer_subbuf_size_get(iter->array_buffer->buffer); + if (*ppos & (page_size - 1)) return -EINVAL; - if (len & (PAGE_SIZE - 1)) { - if (len < PAGE_SIZE) + if (len & (page_size - 1)) { + if (len < page_size) return -EINVAL; - len &= PAGE_MASK; + len &= (~(page_size - 1)); } if (splice_grow_spd(pipe, &spd)) @@ -8303,7 +8834,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, trace_access_lock(iter->cpu_file); entries = ring_buffer_entries_cpu(iter->array_buffer->buffer, iter->cpu_file); - for (i = 0; i < spd.nr_pages_max && len && entries; i++, len -= PAGE_SIZE) { + for (i = 0; i < spd.nr_pages_max && len && entries; i++, len -= page_size) { struct page *page; int r; @@ -8324,7 +8855,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, } ref->cpu = iter->cpu_file; - r = ring_buffer_read_page(ref->buffer, &ref->page, + r = ring_buffer_read_page(ref->buffer, ref->page, len, iter->cpu_file, 1); if (r < 0) { ring_buffer_free_read_page(ref->buffer, ref->cpu, @@ -8333,14 +8864,14 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, break; } - page = virt_to_page(ref->page); + page = virt_to_page(ring_buffer_read_page_data(ref->page)); spd.pages[i] = page; - spd.partial[i].len = PAGE_SIZE; + spd.partial[i].len = page_size; spd.partial[i].offset = 0; spd.partial[i].private = (unsigned long)ref; spd.nr_pages++; - *ppos += PAGE_SIZE; + *ppos += page_size; entries = ring_buffer_entries_cpu(iter->array_buffer->buffer, iter->cpu_file); } @@ -8350,18 +8881,18 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, /* did we read anything? */ if (!spd.nr_pages) { - long wait_index; if (ret) goto out; + if (woken) + goto out; + ret = -EAGAIN; if ((file->f_flags & O_NONBLOCK) || (flags & SPLICE_F_NONBLOCK)) goto out; - wait_index = READ_ONCE(iter->wait_index); - - ret = wait_on_pipe(iter, iter->tr->buffer_percent); + ret = wait_on_pipe(iter, iter->snapshot ? 0 : iter->tr->buffer_percent); if (ret) goto out; @@ -8369,10 +8900,8 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, if (!tracer_tracing_is_on(iter->tr)) goto out; - /* Make sure we see the new wait_index */ - smp_rmb(); - if (wait_index != iter->wait_index) - goto out; + /* Iterate one more time to collect any new data then exit */ + woken = true; goto again; } @@ -8384,35 +8913,135 @@ out: return ret; } -/* An ioctl call with cmd 0 to the ring buffer file will wake up all waiters */ static long tracing_buffers_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct ftrace_buffer_info *info = file->private_data; struct trace_iterator *iter = &info->iter; + int err; - if (cmd) - return -ENOIOCTLCMD; + if (cmd == TRACE_MMAP_IOCTL_GET_READER) { + if (!(file->f_flags & O_NONBLOCK)) { + err = ring_buffer_wait(iter->array_buffer->buffer, + iter->cpu_file, + iter->tr->buffer_percent, + NULL, NULL); + if (err) + return err; + } - mutex_lock(&trace_types_lock); + return ring_buffer_map_get_reader(iter->array_buffer->buffer, + iter->cpu_file); + } else if (cmd) { + return -ENOTTY; + } + + /* + * An ioctl call with cmd 0 to the ring buffer file will wake up all + * waiters + */ + guard(mutex)(&trace_types_lock); - iter->wait_index++; /* Make sure the waiters see the new wait_index */ - smp_wmb(); + (void)atomic_fetch_inc_release(&iter->wait_index); ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); - mutex_unlock(&trace_types_lock); return 0; } +#ifdef CONFIG_TRACER_MAX_TRACE +static int get_snapshot_map(struct trace_array *tr) +{ + int err = 0; + + /* + * Called with mmap_lock held. lockdep would be unhappy if we would now + * take trace_types_lock. Instead use the specific + * snapshot_trigger_lock. + */ + spin_lock(&tr->snapshot_trigger_lock); + + if (tr->snapshot || tr->mapped == UINT_MAX) + err = -EBUSY; + else + tr->mapped++; + + spin_unlock(&tr->snapshot_trigger_lock); + + /* Wait for update_max_tr() to observe iter->tr->mapped */ + if (tr->mapped == 1) + synchronize_rcu(); + + return err; + +} +static void put_snapshot_map(struct trace_array *tr) +{ + spin_lock(&tr->snapshot_trigger_lock); + if (!WARN_ON(!tr->mapped)) + tr->mapped--; + spin_unlock(&tr->snapshot_trigger_lock); +} +#else +static inline int get_snapshot_map(struct trace_array *tr) { return 0; } +static inline void put_snapshot_map(struct trace_array *tr) { } +#endif + +static void tracing_buffers_mmap_close(struct vm_area_struct *vma) +{ + struct ftrace_buffer_info *info = vma->vm_file->private_data; + struct trace_iterator *iter = &info->iter; + + WARN_ON(ring_buffer_unmap(iter->array_buffer->buffer, iter->cpu_file)); + put_snapshot_map(iter->tr); +} + +static int tracing_buffers_may_split(struct vm_area_struct *vma, unsigned long addr) +{ + /* + * Trace buffer mappings require the complete buffer including + * the meta page. Partial mappings are not supported. + */ + return -EINVAL; +} + +static const struct vm_operations_struct tracing_buffers_vmops = { + .close = tracing_buffers_mmap_close, + .may_split = tracing_buffers_may_split, +}; + +static int tracing_buffers_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct ftrace_buffer_info *info = filp->private_data; + struct trace_iterator *iter = &info->iter; + int ret = 0; + + /* A memmap'ed and backup buffers are not supported for user space mmap */ + if (iter->tr->flags & (TRACE_ARRAY_FL_MEMMAP | TRACE_ARRAY_FL_VMALLOC)) + return -ENODEV; + + ret = get_snapshot_map(iter->tr); + if (ret) + return ret; + + ret = ring_buffer_map(iter->array_buffer->buffer, iter->cpu_file, vma); + if (ret) + put_snapshot_map(iter->tr); + + vma->vm_ops = &tracing_buffers_vmops; + + return ret; +} + static const struct file_operations tracing_buffers_fops = { .open = tracing_buffers_open, .read = tracing_buffers_read, .poll = tracing_buffers_poll, .release = tracing_buffers_release, + .flush = tracing_buffers_flush, .splice_read = tracing_buffers_splice_read, .unlocked_ioctl = tracing_buffers_ioctl, - .llseek = no_llseek, + .mmap = tracing_buffers_mmap, }; static ssize_t @@ -8496,15 +9125,22 @@ tracing_read_dyn_info(struct file *filp, char __user *ubuf, char *buf; int r; - /* 256 should be plenty to hold the amount needed */ - buf = kmalloc(256, GFP_KERNEL); + /* 512 should be plenty to hold the amount needed */ +#define DYN_INFO_BUF_SIZE 512 + + buf = kmalloc(DYN_INFO_BUF_SIZE, GFP_KERNEL); if (!buf) return -ENOMEM; - r = scnprintf(buf, 256, "%ld pages:%ld groups: %ld\n", + r = scnprintf(buf, DYN_INFO_BUF_SIZE, + "%ld pages:%ld groups: %ld\n" + "ftrace boot update time = %llu (ns)\n" + "ftrace module total update time = %llu (ns)\n", ftrace_update_tot_cnt, ftrace_number_of_pages, - ftrace_number_of_groups); + ftrace_number_of_groups, + ftrace_update_time, + ftrace_total_mod_time); ret = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); kfree(buf); @@ -8633,8 +9269,13 @@ ftrace_trace_snapshot_callback(struct trace_array *tr, struct ftrace_hash *hash, ops = param ? &snapshot_count_probe_ops : &snapshot_probe_ops; - if (glob[0] == '!') - return unregister_ftrace_function_probe_func(glob+1, tr, ops); + if (glob[0] == '!') { + ret = unregister_ftrace_function_probe_func(glob+1, tr, ops); + if (!ret) + tracing_disarm_snapshot(tr); + + return ret; + } if (!param) goto out_reg; @@ -8653,13 +9294,14 @@ ftrace_trace_snapshot_callback(struct trace_array *tr, struct ftrace_hash *hash, return ret; out_reg: - ret = tracing_alloc_snapshot_instance(tr); + ret = tracing_arm_snapshot(tr); if (ret < 0) - goto out; + return ret; ret = register_ftrace_function_probe(glob, tr, ops, count); + if (ret < 0) + tracing_disarm_snapshot(tr); - out: return ret < 0 ? ret : 0; } @@ -8678,13 +9320,13 @@ static inline __init int register_snapshot_cmd(void) { return 0; } static struct dentry *tracing_get_dentry(struct trace_array *tr) { - if (WARN_ON(!tr->dir)) - return ERR_PTR(-ENODEV); - /* Top directory uses NULL as the parent */ if (tr->flags & TRACE_ARRAY_FL_GLOBAL) return NULL; + if (WARN_ON(!tr->dir)) + return ERR_PTR(-ENODEV); + /* All sub buffers have a descriptor */ return tr->dir; } @@ -8753,12 +9395,17 @@ tracing_init_tracefs_percpu(struct trace_array *tr, long cpu) trace_create_cpu_file("buffer_size_kb", TRACE_MODE_READ, d_cpu, tr, cpu, &tracing_entries_fops); + if (tr->range_addr_start) + trace_create_cpu_file("buffer_meta", TRACE_MODE_READ, d_cpu, + tr, cpu, &tracing_buffer_meta_fops); #ifdef CONFIG_TRACER_SNAPSHOT - trace_create_cpu_file("snapshot", TRACE_MODE_WRITE, d_cpu, - tr, cpu, &snapshot_fops); + if (!tr->range_addr_start) { + trace_create_cpu_file("snapshot", TRACE_MODE_WRITE, d_cpu, + tr, cpu, &snapshot_fops); - trace_create_cpu_file("snapshot_raw", TRACE_MODE_READ, d_cpu, - tr, cpu, &snapshot_raw_fops); + trace_create_cpu_file("snapshot_raw", TRACE_MODE_READ, d_cpu, + tr, cpu, &snapshot_raw_fops); + } #endif } @@ -8798,10 +9445,9 @@ trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, return -EINVAL; if (!!(topt->flags->val & topt->opt->bit) != val) { - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); ret = __set_tracer_option(topt->tr, topt->flags, topt->opt, !val); - mutex_unlock(&trace_types_lock); if (ret) return ret; } @@ -8811,12 +9457,33 @@ trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, return cnt; } +static int tracing_open_options(struct inode *inode, struct file *filp) +{ + struct trace_option_dentry *topt = inode->i_private; + int ret; + + ret = tracing_check_open_get_tr(topt->tr); + if (ret) + return ret; + + filp->private_data = inode->i_private; + return 0; +} + +static int tracing_release_options(struct inode *inode, struct file *file) +{ + struct trace_option_dentry *topt = file->private_data; + + trace_array_put(topt->tr); + return 0; +} static const struct file_operations trace_options_fops = { - .open = tracing_open_generic, + .open = tracing_open_options, .read = trace_options_read, .write = trace_options_write, .llseek = generic_file_llseek, + .release = tracing_release_options, }; /* @@ -8863,7 +9530,7 @@ trace_options_core_read(struct file *filp, char __user *ubuf, size_t cnt, get_tr_index(tr_index, &tr, &index); - if (tr->trace_flags & (1 << index)) + if (tr->trace_flags & (1ULL << index)) buf = "1\n"; else buf = "0\n"; @@ -8892,7 +9559,7 @@ trace_options_core_write(struct file *filp, const char __user *ubuf, size_t cnt, mutex_lock(&event_mutex); mutex_lock(&trace_types_lock); - ret = set_tracer_flag(tr, 1 << index, val); + ret = set_tracer_flag(tr, 1ULL << index, val); mutex_unlock(&trace_types_lock); mutex_unlock(&event_mutex); @@ -8965,39 +9632,19 @@ create_trace_option_file(struct trace_array *tr, topt->entry = trace_create_file(opt->name, TRACE_MODE_WRITE, t_options, topt, &trace_options_fops); - } -static void -create_trace_option_files(struct trace_array *tr, struct tracer *tracer) +static int +create_trace_option_files(struct trace_array *tr, struct tracer *tracer, + struct tracer_flags *flags) { struct trace_option_dentry *topts; struct trace_options *tr_topts; - struct tracer_flags *flags; struct tracer_opt *opts; int cnt; - int i; - - if (!tracer) - return; - - flags = tracer->flags; if (!flags || !flags->opts) - return; - - /* - * If this is an instance, only create flags for tracers - * the instance may have. - */ - if (!trace_ok_for_array(tracer, tr)) - return; - - for (i = 0; i < tr->nr_topts; i++) { - /* Make sure there's no duplicate flags. */ - if (WARN_ON_ONCE(tr->topts[i].tracer->flags == tracer->flags)) - return; - } + return 0; opts = flags->opts; @@ -9006,13 +9653,13 @@ create_trace_option_files(struct trace_array *tr, struct tracer *tracer) topts = kcalloc(cnt + 1, sizeof(*topts), GFP_KERNEL); if (!topts) - return; + return 0; tr_topts = krealloc(tr->topts, sizeof(*tr->topts) * (tr->nr_topts + 1), GFP_KERNEL); if (!tr_topts) { kfree(topts); - return; + return -ENOMEM; } tr->topts = tr_topts; @@ -9027,6 +9674,97 @@ create_trace_option_files(struct trace_array *tr, struct tracer *tracer) "Failed to create trace option: %s", opts[cnt].name); } + return 0; +} + +static int get_global_flags_val(struct tracer *tracer) +{ + struct tracers *t; + + list_for_each_entry(t, &global_trace.tracers, list) { + if (t->tracer != tracer) + continue; + if (!t->flags) + return -1; + return t->flags->val; + } + return -1; +} + +static int add_tracer_options(struct trace_array *tr, struct tracers *t) +{ + struct tracer *tracer = t->tracer; + struct tracer_flags *flags = t->flags ?: tracer->flags; + + if (!flags) + return 0; + + /* Only add tracer options after update_tracer_options finish */ + if (!tracer_options_updated) + return 0; + + return create_trace_option_files(tr, tracer, flags); +} + +static int add_tracer(struct trace_array *tr, struct tracer *tracer) +{ + struct tracer_flags *flags; + struct tracers *t; + int ret; + + /* Only enable if the directory has been created already. */ + if (!tr->dir && !(tr->flags & TRACE_ARRAY_FL_GLOBAL)) + return 0; + + /* + * If this is an instance, only create flags for tracers + * the instance may have. + */ + if (!trace_ok_for_array(tracer, tr)) + return 0; + + t = kmalloc(sizeof(*t), GFP_KERNEL); + if (!t) + return -ENOMEM; + + t->tracer = tracer; + t->flags = NULL; + list_add(&t->list, &tr->tracers); + + flags = tracer->flags; + if (!flags) { + if (!tracer->default_flags) + return 0; + + /* + * If the tracer defines default flags, it means the flags are + * per trace instance. + */ + flags = kmalloc(sizeof(*flags), GFP_KERNEL); + if (!flags) + return -ENOMEM; + + *flags = *tracer->default_flags; + flags->trace = tracer; + + t->flags = flags; + + /* If this is an instance, inherit the global_trace flags */ + if (!(tr->flags & TRACE_ARRAY_FL_GLOBAL)) { + int val = get_global_flags_val(tracer); + if (!WARN_ON_ONCE(val < 0)) + flags->val = val; + } + } + + ret = add_tracer_options(tr, t); + if (ret < 0) { + list_del(&t->list); + kfree(t->flags); + kfree(t); + } + + return ret; } static struct dentry * @@ -9056,8 +9794,9 @@ static void create_trace_options_dir(struct trace_array *tr) for (i = 0; trace_options[i]; i++) { if (top_level || - !((1 << i) & TOP_LEVEL_TRACE_FLAGS)) + !((1ULL << i) & TOP_LEVEL_TRACE_FLAGS)) { create_trace_option_core_file(tr, trace_options[i], i); + } } } @@ -9089,7 +9828,7 @@ rb_simple_write(struct file *filp, const char __user *ubuf, return ret; if (buffer) { - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); if (!!val == tracer_tracing_is_on(tr)) { val = 0; /* do nothing */ } else if (val) { @@ -9103,7 +9842,6 @@ rb_simple_write(struct file *filp, const char __user *ubuf, /* Wake up any waiters */ ring_buffer_wake_waiters(buffer, RING_BUFFER_ALL_CPUS); } - mutex_unlock(&trace_types_lock); } (*ppos)++; @@ -9148,9 +9886,6 @@ buffer_percent_write(struct file *filp, const char __user *ubuf, if (val > 100) return -EINVAL; - if (!val) - val = 1; - tr->buffer_percent = val; (*ppos)++; @@ -9166,21 +9901,244 @@ static const struct file_operations buffer_percent_fops = { .llseek = default_llseek, }; +static ssize_t +buffer_subbuf_size_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) +{ + struct trace_array *tr = filp->private_data; + size_t size; + char buf[64]; + int order; + int r; + + order = ring_buffer_subbuf_order_get(tr->array_buffer.buffer); + size = (PAGE_SIZE << order) / 1024; + + r = sprintf(buf, "%zd\n", size); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +static ssize_t +buffer_subbuf_size_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct trace_array *tr = filp->private_data; + unsigned long val; + int old_order; + int order; + int pages; + int ret; + + ret = kstrtoul_from_user(ubuf, cnt, 10, &val); + if (ret) + return ret; + + val *= 1024; /* value passed in is in KB */ + + pages = DIV_ROUND_UP(val, PAGE_SIZE); + order = fls(pages - 1); + + /* limit between 1 and 128 system pages */ + if (order < 0 || order > 7) + return -EINVAL; + + /* Do not allow tracing while changing the order of the ring buffer */ + tracing_stop_tr(tr); + + old_order = ring_buffer_subbuf_order_get(tr->array_buffer.buffer); + if (old_order == order) + goto out; + + ret = ring_buffer_subbuf_order_set(tr->array_buffer.buffer, order); + if (ret) + goto out; + +#ifdef CONFIG_TRACER_MAX_TRACE + + if (!tr->allocated_snapshot) + goto out_max; + + ret = ring_buffer_subbuf_order_set(tr->max_buffer.buffer, order); + if (ret) { + /* Put back the old order */ + cnt = ring_buffer_subbuf_order_set(tr->array_buffer.buffer, old_order); + if (WARN_ON_ONCE(cnt)) { + /* + * AARGH! We are left with different orders! + * The max buffer is our "snapshot" buffer. + * When a tracer needs a snapshot (one of the + * latency tracers), it swaps the max buffer + * with the saved snap shot. We succeeded to + * update the order of the main buffer, but failed to + * update the order of the max buffer. But when we tried + * to reset the main buffer to the original size, we + * failed there too. This is very unlikely to + * happen, but if it does, warn and kill all + * tracing. + */ + tracing_disabled = 1; + } + goto out; + } + out_max: +#endif + (*ppos)++; + out: + if (ret) + cnt = ret; + tracing_start_tr(tr); + return cnt; +} + +static const struct file_operations buffer_subbuf_size_fops = { + .open = tracing_open_generic_tr, + .read = buffer_subbuf_size_read, + .write = buffer_subbuf_size_write, + .release = tracing_release_generic_tr, + .llseek = default_llseek, +}; + static struct dentry *trace_instance_dir; static void init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer); +#ifdef CONFIG_MODULES +static int make_mod_delta(struct module *mod, void *data) +{ + struct trace_module_delta *module_delta; + struct trace_scratch *tscratch; + struct trace_mod_entry *entry; + struct trace_array *tr = data; + int i; + + tscratch = tr->scratch; + module_delta = READ_ONCE(tr->module_delta); + for (i = 0; i < tscratch->nr_entries; i++) { + entry = &tscratch->entries[i]; + if (strcmp(mod->name, entry->mod_name)) + continue; + if (mod->state == MODULE_STATE_GOING) + module_delta->delta[i] = 0; + else + module_delta->delta[i] = (unsigned long)mod->mem[MOD_TEXT].base + - entry->mod_addr; + break; + } + return 0; +} +#else +static int make_mod_delta(struct module *mod, void *data) +{ + return 0; +} +#endif + +static int mod_addr_comp(const void *a, const void *b, const void *data) +{ + const struct trace_mod_entry *e1 = a; + const struct trace_mod_entry *e2 = b; + + return e1->mod_addr > e2->mod_addr ? 1 : -1; +} + +static void setup_trace_scratch(struct trace_array *tr, + struct trace_scratch *tscratch, unsigned int size) +{ + struct trace_module_delta *module_delta; + struct trace_mod_entry *entry; + int i, nr_entries; + + if (!tscratch) + return; + + tr->scratch = tscratch; + tr->scratch_size = size; + + if (tscratch->text_addr) + tr->text_delta = (unsigned long)_text - tscratch->text_addr; + + if (struct_size(tscratch, entries, tscratch->nr_entries) > size) + goto reset; + + /* Check if each module name is a valid string */ + for (i = 0; i < tscratch->nr_entries; i++) { + int n; + + entry = &tscratch->entries[i]; + + for (n = 0; n < MODULE_NAME_LEN; n++) { + if (entry->mod_name[n] == '\0') + break; + if (!isprint(entry->mod_name[n])) + goto reset; + } + if (n == MODULE_NAME_LEN) + goto reset; + } + + /* Sort the entries so that we can find appropriate module from address. */ + nr_entries = tscratch->nr_entries; + sort_r(tscratch->entries, nr_entries, sizeof(struct trace_mod_entry), + mod_addr_comp, NULL, NULL); + + if (IS_ENABLED(CONFIG_MODULES)) { + module_delta = kzalloc(struct_size(module_delta, delta, nr_entries), GFP_KERNEL); + if (!module_delta) { + pr_info("module_delta allocation failed. Not able to decode module address."); + goto reset; + } + init_rcu_head(&module_delta->rcu); + } else + module_delta = NULL; + WRITE_ONCE(tr->module_delta, module_delta); + + /* Scan modules to make text delta for modules. */ + module_for_each_mod(make_mod_delta, tr); + + /* Set trace_clock as the same of the previous boot. */ + if (tscratch->clock_id != tr->clock_id) { + if (tscratch->clock_id >= ARRAY_SIZE(trace_clocks) || + tracing_set_clock(tr, trace_clocks[tscratch->clock_id].name) < 0) { + pr_info("the previous trace_clock info is not valid."); + goto reset; + } + } + return; + reset: + /* Invalid trace modules */ + memset(tscratch, 0, size); +} + static int allocate_trace_buffer(struct trace_array *tr, struct array_buffer *buf, int size) { enum ring_buffer_flags rb_flags; + struct trace_scratch *tscratch; + unsigned int scratch_size = 0; - rb_flags = tr->trace_flags & TRACE_ITER_OVERWRITE ? RB_FL_OVERWRITE : 0; + rb_flags = tr->trace_flags & TRACE_ITER(OVERWRITE) ? RB_FL_OVERWRITE : 0; buf->tr = tr; - buf->buffer = ring_buffer_alloc(size, rb_flags); + if (tr->range_addr_start && tr->range_addr_size) { + /* Add scratch buffer to handle 128 modules */ + buf->buffer = ring_buffer_alloc_range(size, rb_flags, 0, + tr->range_addr_start, + tr->range_addr_size, + struct_size(tscratch, entries, 128)); + + tscratch = ring_buffer_meta_scratch(buf->buffer, &scratch_size); + setup_trace_scratch(tr, tscratch, scratch_size); + + /* + * This is basically the same as a mapped buffer, + * with the same restrictions. + */ + tr->mapped++; + } else { + buf->buffer = ring_buffer_alloc(size, rb_flags); + } if (!buf->buffer) return -ENOMEM; @@ -9217,6 +10175,10 @@ static int allocate_trace_buffers(struct trace_array *tr, int size) return ret; #ifdef CONFIG_TRACER_MAX_TRACE + /* Fix mapped buffer trace arrays do not have snapshot buffers */ + if (tr->range_addr_start) + return 0; + ret = allocate_trace_buffer(tr, &tr->max_buffer, allocate_snapshot ? size : 1); if (MEM_FAIL(ret, "Failed to allocate trace buffer\n")) { @@ -9225,10 +10187,6 @@ static int allocate_trace_buffers(struct trace_array *tr, int size) } tr->allocated_snapshot = allocate_snapshot; - /* - * Only the top level trace array gets its snapshot allocated - * from the kernel command line. - */ allocate_snapshot = false; #endif @@ -9241,6 +10199,7 @@ static void free_trace_buffers(struct trace_array *tr) return; free_trace_buffer(&tr->array_buffer); + kfree(tr->module_delta); #ifdef CONFIG_TRACER_MAX_TRACE free_trace_buffer(&tr->max_buffer); @@ -9256,20 +10215,39 @@ static void init_trace_flags_index(struct trace_array *tr) tr->trace_flags_index[i] = i; } -static void __update_tracer_options(struct trace_array *tr) +static int __update_tracer(struct trace_array *tr) { struct tracer *t; + int ret = 0; - for (t = trace_types; t; t = t->next) - add_tracer_options(tr, t); + for (t = trace_types; t && !ret; t = t->next) + ret = add_tracer(tr, t); + + return ret; } -static void update_tracer_options(struct trace_array *tr) +static __init int __update_tracer_options(struct trace_array *tr) { - mutex_lock(&trace_types_lock); + struct tracers *t; + int ret = 0; + + list_for_each_entry(t, &tr->tracers, list) { + ret = add_tracer_options(tr, t); + if (ret < 0) + break; + } + + return ret; +} + +static __init void update_tracer_options(void) +{ + struct trace_array *tr; + + guard(mutex)(&trace_types_lock); tracer_options_updated = true; - __update_tracer_options(tr); - mutex_unlock(&trace_types_lock); + list_for_each_entry(tr, &ftrace_trace_arrays, list) + __update_tracer_options(tr); } /* Must have trace_types_lock held */ @@ -9291,11 +10269,10 @@ struct trace_array *trace_array_find_get(const char *instance) { struct trace_array *tr; - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); tr = trace_array_find(instance); if (tr) tr->ref++; - mutex_unlock(&trace_types_lock); return tr; } @@ -9315,12 +10292,19 @@ static int trace_array_create_dir(struct trace_array *tr) } init_tracer_tracefs(tr, tr->dir); - __update_tracer_options(tr); - - return ret; + ret = __update_tracer(tr); + if (ret) { + event_trace_del_tracer(tr); + tracefs_remove(tr->dir); + return ret; + } + return 0; } -static struct trace_array *trace_array_create(const char *name) +static struct trace_array * +trace_array_create_systems(const char *name, const char *systems, + unsigned long range_addr_start, + unsigned long range_addr_size) { struct trace_array *tr; int ret; @@ -9337,24 +10321,51 @@ static struct trace_array *trace_array_create(const char *name) if (!alloc_cpumask_var(&tr->tracing_cpumask, GFP_KERNEL)) goto out_free_tr; + if (!zalloc_cpumask_var(&tr->pipe_cpumask, GFP_KERNEL)) + goto out_free_tr; + + if (systems) { + tr->system_names = kstrdup_const(systems, GFP_KERNEL); + if (!tr->system_names) + goto out_free_tr; + } + + /* Only for boot up memory mapped ring buffers */ + tr->range_addr_start = range_addr_start; + tr->range_addr_size = range_addr_size; + tr->trace_flags = global_trace.trace_flags & ~ZEROED_TRACE_FLAGS; cpumask_copy(tr->tracing_cpumask, cpu_all_mask); raw_spin_lock_init(&tr->start_lock); - tr->max_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; + tr->syscall_buf_sz = global_trace.syscall_buf_sz; + tr->max_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; +#ifdef CONFIG_TRACER_MAX_TRACE + spin_lock_init(&tr->snapshot_trigger_lock); +#endif tr->current_trace = &nop_trace; + tr->current_trace_flags = nop_trace.flags; INIT_LIST_HEAD(&tr->systems); INIT_LIST_HEAD(&tr->events); INIT_LIST_HEAD(&tr->hist_vars); INIT_LIST_HEAD(&tr->err_log); + INIT_LIST_HEAD(&tr->tracers); + INIT_LIST_HEAD(&tr->marker_list); + +#ifdef CONFIG_MODULES + INIT_LIST_HEAD(&tr->mod_events); +#endif if (allocate_trace_buffers(tr, trace_buf_size) < 0) goto out_free_tr; + /* The ring buffer is defaultly expanded */ + trace_set_ring_buffer_expanded(tr); + if (ftrace_allocate_ftrace_ops(tr) < 0) goto out_free_tr; @@ -9378,38 +10389,74 @@ static struct trace_array *trace_array_create(const char *name) out_free_tr: ftrace_free_ftrace_ops(tr); free_trace_buffers(tr); + free_cpumask_var(tr->pipe_cpumask); free_cpumask_var(tr->tracing_cpumask); + kfree_const(tr->system_names); + kfree(tr->range_name); kfree(tr->name); kfree(tr); return ERR_PTR(ret); } +static struct trace_array *trace_array_create(const char *name) +{ + return trace_array_create_systems(name, NULL, 0, 0); +} + static int instance_mkdir(const char *name) { struct trace_array *tr; int ret; - mutex_lock(&event_mutex); - mutex_lock(&trace_types_lock); + guard(mutex)(&event_mutex); + guard(mutex)(&trace_types_lock); ret = -EEXIST; if (trace_array_find(name)) - goto out_unlock; + return -EEXIST; tr = trace_array_create(name); ret = PTR_ERR_OR_ZERO(tr); -out_unlock: - mutex_unlock(&trace_types_lock); - mutex_unlock(&event_mutex); return ret; } +#ifdef CONFIG_MMU +static u64 map_pages(unsigned long start, unsigned long size) +{ + unsigned long vmap_start, vmap_end; + struct vm_struct *area; + int ret; + + area = get_vm_area(size, VM_IOREMAP); + if (!area) + return 0; + + vmap_start = (unsigned long) area->addr; + vmap_end = vmap_start + size; + + ret = vmap_page_range(vmap_start, vmap_end, + start, pgprot_nx(PAGE_KERNEL)); + if (ret < 0) { + free_vm_area(area); + return 0; + } + + return (u64)vmap_start; +} +#else +static inline u64 map_pages(unsigned long start, unsigned long size) +{ + return 0; +} +#endif + /** * trace_array_get_by_name - Create/Lookup a trace array, given its name. * @name: The name of the trace array to be looked up/created. + * @systems: A list of systems to create event directories for (NULL for all) * * Returns pointer to trace array with given name. * NULL, if it cannot be created. @@ -9423,28 +10470,27 @@ out_unlock: * trace_array_put() is called, user space can not delete it. * */ -struct trace_array *trace_array_get_by_name(const char *name) +struct trace_array *trace_array_get_by_name(const char *name, const char *systems) { struct trace_array *tr; - mutex_lock(&event_mutex); - mutex_lock(&trace_types_lock); + guard(mutex)(&event_mutex); + guard(mutex)(&trace_types_lock); list_for_each_entry(tr, &ftrace_trace_arrays, list) { - if (tr->name && strcmp(tr->name, name) == 0) - goto out_unlock; + if (tr->name && strcmp(tr->name, name) == 0) { + tr->ref++; + return tr; + } } - tr = trace_array_create(name); + tr = trace_array_create_systems(name, systems, 0, 0); if (IS_ERR(tr)) tr = NULL; -out_unlock: - if (tr) + else tr->ref++; - mutex_unlock(&trace_types_lock); - mutex_unlock(&event_mutex); return tr; } EXPORT_SYMBOL_GPL(trace_array_get_by_name); @@ -9462,9 +10508,15 @@ static int __remove_instance(struct trace_array *tr) /* Disable all the flags that were enabled coming in */ for (i = 0; i < TRACE_FLAGS_MAX_SIZE; i++) { if ((1 << i) & ZEROED_TRACE_FLAGS) - set_tracer_flag(tr, 1 << i, 0); + set_tracer_flag(tr, 1ULL << i, 0); } + if (printk_trace == tr) + update_printk_trace(&global_trace); + + if (update_marker_trace(tr, 0)) + synchronize_rcu(); + tracing_set_nop(tr); clear_ftrace_function_probes(tr); event_trace_del_tracer(tr); @@ -9473,13 +10525,24 @@ static int __remove_instance(struct trace_array *tr) tracefs_remove(tr->dir); free_percpu(tr->last_func_repeats); free_trace_buffers(tr); + clear_tracing_err_log(tr); + free_tracers(tr); + + if (tr->range_name) { + reserve_mem_release_by_name(tr->range_name); + kfree(tr->range_name); + } + if (tr->flags & TRACE_ARRAY_FL_VMALLOC) + vfree((void *)tr->range_addr_start); for (i = 0; i < tr->nr_topts; i++) { kfree(tr->topts[i].topts); } kfree(tr->topts); + free_cpumask_var(tr->pipe_cpumask); free_cpumask_var(tr->tracing_cpumask); + kfree_const(tr->system_names); kfree(tr->name); kfree(tr); @@ -9489,48 +10552,36 @@ static int __remove_instance(struct trace_array *tr) int trace_array_destroy(struct trace_array *this_tr) { struct trace_array *tr; - int ret; if (!this_tr) return -EINVAL; - mutex_lock(&event_mutex); - mutex_lock(&trace_types_lock); + guard(mutex)(&event_mutex); + guard(mutex)(&trace_types_lock); - ret = -ENODEV; /* Making sure trace array exists before destroying it. */ list_for_each_entry(tr, &ftrace_trace_arrays, list) { - if (tr == this_tr) { - ret = __remove_instance(tr); - break; - } + if (tr == this_tr) + return __remove_instance(tr); } - mutex_unlock(&trace_types_lock); - mutex_unlock(&event_mutex); - - return ret; + return -ENODEV; } EXPORT_SYMBOL_GPL(trace_array_destroy); static int instance_rmdir(const char *name) { struct trace_array *tr; - int ret; - mutex_lock(&event_mutex); - mutex_lock(&trace_types_lock); + guard(mutex)(&event_mutex); + guard(mutex)(&trace_types_lock); - ret = -ENODEV; tr = trace_array_find(name); - if (tr) - ret = __remove_instance(tr); - - mutex_unlock(&trace_types_lock); - mutex_unlock(&event_mutex); + if (!tr) + return -ENODEV; - return ret; + return __remove_instance(tr); } static __init void create_trace_instances(struct dentry *d_tracer) @@ -9543,25 +10594,21 @@ static __init void create_trace_instances(struct dentry *d_tracer) if (MEM_FAIL(!trace_instance_dir, "Failed to create instances directory\n")) return; - mutex_lock(&event_mutex); - mutex_lock(&trace_types_lock); + guard(mutex)(&event_mutex); + guard(mutex)(&trace_types_lock); list_for_each_entry(tr, &ftrace_trace_arrays, list) { if (!tr->name) continue; if (MEM_FAIL(trace_array_create_dir(tr) < 0, "Failed to create instance directory\n")) - break; + return; } - - mutex_unlock(&trace_types_lock); - mutex_unlock(&event_mutex); } static void init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) { - struct trace_event_file *file; int cpu; trace_create_file("available_tracers", TRACE_MODE_READ, d_tracer, @@ -9594,11 +10641,7 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) trace_create_file("trace_marker", 0220, d_tracer, tr, &tracing_mark_fops); - file = __find_event_file(tr, "ftrace", "print"); - if (file && file->dir) - trace_create_file("trigger", TRACE_MODE_WRITE, file->dir, - file, &event_trigger_fops); - tr->trace_marker_file = file; + tr->trace_marker_file = __find_event_file(tr, "ftrace", "print"); trace_create_file("trace_marker_raw", 0220, d_tracer, tr, &tracing_mark_raw_fops); @@ -9614,9 +10657,15 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) tr->buffer_percent = 50; - trace_create_file("buffer_percent", TRACE_MODE_READ, d_tracer, + trace_create_file("buffer_percent", TRACE_MODE_WRITE, d_tracer, tr, &buffer_percent_fops); + trace_create_file("buffer_subbuf_size_kb", TRACE_MODE_WRITE, d_tracer, + tr, &buffer_subbuf_size_fops); + + trace_create_file("syscall_user_buf_size", TRACE_MODE_WRITE, d_tracer, + tr, &tracing_syscall_buf_fops); + create_trace_options_dir(tr); #ifdef CONFIG_TRACER_MAX_TRACE @@ -9626,10 +10675,15 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) if (ftrace_create_function_files(tr, d_tracer)) MEM_FAIL(1, "Could not allocate function filter files"); + if (tr->range_addr_start) { + trace_create_file("last_boot_info", TRACE_MODE_READ, d_tracer, + tr, &last_boot_fops); #ifdef CONFIG_TRACER_SNAPSHOT - trace_create_file("snapshot", TRACE_MODE_WRITE, d_tracer, - tr, &snapshot_fops); + } else { + trace_create_file("snapshot", TRACE_MODE_WRITE, d_tracer, + tr, &snapshot_fops); #endif + } trace_create_file("error_log", TRACE_MODE_WRITE, d_tracer, tr, &tracing_err_log_fops); @@ -9640,10 +10694,13 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) ftrace_init_tracefs(tr, d_tracer); } +#ifdef CONFIG_TRACEFS_AUTOMOUNT_DEPRECATED static struct vfsmount *trace_automount(struct dentry *mntpt, void *ingore) { struct vfsmount *mnt; struct file_system_type *type; + struct fs_context *fc; + int ret; /* * To maintain backward compatibility for tools that mount @@ -9653,14 +10710,24 @@ static struct vfsmount *trace_automount(struct dentry *mntpt, void *ingore) type = get_fs_type("tracefs"); if (!type) return NULL; - mnt = vfs_submount(mntpt, type, "tracefs", NULL); + + fc = fs_context_for_submount(type, mntpt); put_filesystem(type); - if (IS_ERR(mnt)) - return NULL; - mntget(mnt); + if (IS_ERR(fc)) + return ERR_CAST(fc); + pr_warn("NOTICE: Automounting of tracing to debugfs is deprecated and will be removed in 2030\n"); + + ret = vfs_parse_fs_string(fc, "source", "tracefs"); + if (!ret) + mnt = fc_mount(fc); + else + mnt = ERR_PTR(ret); + + put_fs_context(fc); return mnt; } +#endif /** * tracing_init_dentry - initialize top level trace array @@ -9685,6 +10752,7 @@ int tracing_init_dentry(void) if (WARN_ON(!tracefs_initialized())) return -ENODEV; +#ifdef CONFIG_TRACEFS_AUTOMOUNT_DEPRECATED /* * As there may still be users that expect the tracing * files to exist in debugfs/tracing, we must automount @@ -9693,6 +10761,7 @@ int tracing_init_dentry(void) */ tr->dir = debugfs_create_automount("tracing", NULL, trace_automount, NULL); +#endif return 0; } @@ -9709,7 +10778,7 @@ static void __init eval_map_work_func(struct work_struct *work) int len; len = __stop_ftrace_eval_maps - __start_ftrace_eval_maps; - trace_insert_eval_map(NULL, __start_ftrace_eval_maps, len); + trace_event_update_with_eval_map(NULL, __start_ftrace_eval_maps, len); } static int __init trace_eval_init(void) @@ -9742,11 +10811,26 @@ late_initcall_sync(trace_eval_sync); #ifdef CONFIG_MODULES -static void trace_module_add_evals(struct module *mod) + +bool module_exists(const char *module) { - if (!mod->num_trace_evals) - return; + /* All modules have the symbol __this_module */ + static const char this_mod[] = "__this_module"; + char modname[MODULE_NAME_LEN + sizeof(this_mod) + 2]; + unsigned long val; + int n; + n = snprintf(modname, sizeof(modname), "%s:%s", module, this_mod); + + if (n > sizeof(modname) - 1) + return false; + + val = module_kallsyms_lookup_name(modname); + return val != 0; +} + +static void trace_module_add_evals(struct module *mod) +{ /* * Modules with bad taint do not have events created, do * not bother with enums either. @@ -9754,7 +10838,8 @@ static void trace_module_add_evals(struct module *mod) if (trace_module_has_bad_taint(mod)) return; - trace_insert_eval_map(mod, mod->trace_evals, mod->num_trace_evals); + /* Even if no trace_evals, this need to sanitize field types. */ + trace_event_update_with_eval_map(mod, mod->trace_evals, mod->num_trace_evals); } #ifdef CONFIG_TRACE_EVAL_MAP_FILE @@ -9766,7 +10851,7 @@ static void trace_module_remove_evals(struct module *mod) if (!mod->num_trace_evals) return; - mutex_lock(&trace_eval_mutex); + guard(mutex)(&trace_eval_mutex); map = trace_eval_maps; @@ -9778,17 +10863,33 @@ static void trace_module_remove_evals(struct module *mod) map = map->tail.next; } if (!map) - goto out; + return; *last = trace_eval_jmp_to_tail(map)->tail.next; kfree(map); - out: - mutex_unlock(&trace_eval_mutex); } #else static inline void trace_module_remove_evals(struct module *mod) { } #endif /* CONFIG_TRACE_EVAL_MAP_FILE */ +static void trace_module_record(struct module *mod, bool add) +{ + struct trace_array *tr; + unsigned long flags; + + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + flags = tr->flags & (TRACE_ARRAY_FL_BOOT | TRACE_ARRAY_FL_LAST_BOOT); + /* Update any persistent trace array that has already been started */ + if (flags == TRACE_ARRAY_FL_BOOT && add) { + guard(mutex)(&scratch_mutex); + save_mod(mod, tr); + } else if (flags & TRACE_ARRAY_FL_LAST_BOOT) { + /* Update delta if the module loaded in previous boot */ + make_mod_delta(mod, tr); + } + } +} + static int trace_module_notify(struct notifier_block *self, unsigned long val, void *data) { @@ -9797,9 +10898,11 @@ static int trace_module_notify(struct notifier_block *self, switch (val) { case MODULE_STATE_COMING: trace_module_add_evals(mod); + trace_module_record(mod, true); break; case MODULE_STATE_GOING: trace_module_remove_evals(mod); + trace_module_record(mod, false); break; } @@ -9848,7 +10951,7 @@ static __init void tracer_init_tracefs_work_func(struct work_struct *work) create_trace_instances(NULL); - update_tracer_options(&global_trace); + update_tracer_options(); } static __init int tracer_init_tracefs(void) @@ -9868,7 +10971,8 @@ static __init int tracer_init_tracefs(void) tracer_init_tracefs_work_func(NULL); } - rv_init_interface(); + if (rv_init_interface()) + pr_err("RV: Error while creating the RV interface\n"); return 0; } @@ -9897,14 +11001,14 @@ static struct notifier_block trace_die_notifier = { static int trace_die_panic_handler(struct notifier_block *self, unsigned long ev, void *unused) { - if (!ftrace_dump_on_oops) + if (!ftrace_dump_on_oops_enabled()) return NOTIFY_DONE; /* The die notifier requires DIE_OOPS to trigger */ if (self == &trace_die_notifier && ev != DIE_OOPS) return NOTIFY_DONE; - ftrace_dump(ftrace_dump_on_oops); + ftrace_dump(DUMP_PARAM); return NOTIFY_DONE; } @@ -9945,12 +11049,12 @@ trace_printk_seq(struct trace_seq *s) trace_seq_init(s); } -void trace_init_global_iter(struct trace_iterator *iter) +static void trace_init_iter(struct trace_iterator *iter, struct trace_array *tr) { - iter->tr = &global_trace; + iter->tr = tr; iter->trace = iter->tr->current_trace; iter->cpu_file = RING_BUFFER_ALL_CPUS; - iter->array_buffer = &global_trace.array_buffer; + iter->array_buffer = &tr->array_buffer; if (iter->trace && iter->trace->open) iter->trace->open(iter); @@ -9970,21 +11074,18 @@ void trace_init_global_iter(struct trace_iterator *iter) iter->fmt_size = STATIC_FMT_BUF_SIZE; } -void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) +void trace_init_global_iter(struct trace_iterator *iter) +{ + trace_init_iter(iter, &global_trace); +} + +static void ftrace_dump_one(struct trace_array *tr, enum ftrace_dump_mode dump_mode) { /* use static because iter can be a bit big for the stack */ static struct trace_iterator iter; - static atomic_t dump_running; - struct trace_array *tr = &global_trace; unsigned int old_userobj; unsigned long flags; - int cnt = 0, cpu; - - /* Only allow one dump user at a time. */ - if (atomic_inc_return(&dump_running) != 1) { - atomic_dec(&dump_running); - return; - } + int cnt = 0; /* * Always turn off tracing when we dump. @@ -9994,37 +11095,30 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) * If the user does a sysrq-z, then they can re-enable * tracing with echo 1 > tracing_on. */ - tracing_off(); + tracer_tracing_off(tr); local_irq_save(flags); /* Simulate the iterator */ - trace_init_global_iter(&iter); + trace_init_iter(&iter, tr); - for_each_tracing_cpu(cpu) { - atomic_inc(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); - } + /* While dumping, do not allow the buffer to be enable */ + tracer_tracing_disable(tr); - old_userobj = tr->trace_flags & TRACE_ITER_SYM_USEROBJ; + old_userobj = tr->trace_flags & TRACE_ITER(SYM_USEROBJ); /* don't look at user memory in panic mode */ - tr->trace_flags &= ~TRACE_ITER_SYM_USEROBJ; + tr->trace_flags &= ~TRACE_ITER(SYM_USEROBJ); - switch (oops_dump_mode) { - case DUMP_ALL: - iter.cpu_file = RING_BUFFER_ALL_CPUS; - break; - case DUMP_ORIG: + if (dump_mode == DUMP_ORIG) iter.cpu_file = raw_smp_processor_id(); - break; - case DUMP_NONE: - goto out_enable; - default: - printk(KERN_TRACE "Bad dumping mode, switching to all CPUs dump\n"); + else iter.cpu_file = RING_BUFFER_ALL_CPUS; - } - printk(KERN_TRACE "Dumping ftrace buffer:\n"); + if (tr == &global_trace) + printk(KERN_TRACE "Dumping ftrace buffer:\n"); + else + printk(KERN_TRACE "Dumping ftrace instance %s buffer:\n", tr->name); /* Did function tracer already get disabled? */ if (ftrace_is_dead()) { @@ -10055,10 +11149,10 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) ret = print_trace_line(&iter); if (ret != TRACE_TYPE_NO_CONSUME) trace_consume(&iter); + + trace_printk_seq(&iter.seq); } touch_nmi_watchdog(); - - trace_printk_seq(&iter.seq); } if (!cnt) @@ -10066,14 +11160,81 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) else printk(KERN_TRACE "---------------------------------\n"); - out_enable: tr->trace_flags |= old_userobj; - for_each_tracing_cpu(cpu) { - atomic_dec(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); + tracer_tracing_enable(tr); + local_irq_restore(flags); +} + +static void ftrace_dump_by_param(void) +{ + bool first_param = true; + char dump_param[MAX_TRACER_SIZE]; + char *buf, *token, *inst_name; + struct trace_array *tr; + + strscpy(dump_param, ftrace_dump_on_oops, MAX_TRACER_SIZE); + buf = dump_param; + + while ((token = strsep(&buf, ",")) != NULL) { + if (first_param) { + first_param = false; + if (!strcmp("0", token)) + continue; + else if (!strcmp("1", token)) { + ftrace_dump_one(&global_trace, DUMP_ALL); + continue; + } + else if (!strcmp("2", token) || + !strcmp("orig_cpu", token)) { + ftrace_dump_one(&global_trace, DUMP_ORIG); + continue; + } + } + + inst_name = strsep(&token, "="); + tr = trace_array_find(inst_name); + if (!tr) { + printk(KERN_TRACE "Instance %s not found\n", inst_name); + continue; + } + + if (token && (!strcmp("2", token) || + !strcmp("orig_cpu", token))) + ftrace_dump_one(tr, DUMP_ORIG); + else + ftrace_dump_one(tr, DUMP_ALL); + } +} + +void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) +{ + static atomic_t dump_running; + + /* Only allow one dump user at a time. */ + if (atomic_inc_return(&dump_running) != 1) { + atomic_dec(&dump_running); + return; } + + switch (oops_dump_mode) { + case DUMP_ALL: + ftrace_dump_one(&global_trace, DUMP_ALL); + break; + case DUMP_ORIG: + ftrace_dump_one(&global_trace, DUMP_ORIG); + break; + case DUMP_PARAM: + ftrace_dump_by_param(); + break; + case DUMP_NONE: + break; + default: + printk(KERN_TRACE "Bad dumping mode, switching to all CPUs dump\n"); + ftrace_dump_one(&global_trace, DUMP_ALL); + } + atomic_dec(&dump_running); - local_irq_restore(flags); } EXPORT_SYMBOL_GPL(ftrace_dump); @@ -10083,7 +11244,8 @@ ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, size_t count, loff_t *ppos, int (*createfn)(const char *)) { - char *kbuf, *buf, *tmp; + char *kbuf __free(kfree) = NULL; + char *buf, *tmp; int ret = 0; size_t done = 0; size_t size; @@ -10098,10 +11260,9 @@ ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, if (size >= WRITE_BUFSIZE) size = WRITE_BUFSIZE - 1; - if (copy_from_user(kbuf, buffer + done, size)) { - ret = -EFAULT; - goto out; - } + if (copy_from_user(kbuf, buffer + done, size)) + return -EFAULT; + kbuf[size] = '\0'; buf = kbuf; do { @@ -10117,8 +11278,7 @@ ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, /* This can accept WRITE_BUFSIZE - 2 ('\n' + '\0') */ pr_warn("Line length is too long: Should be less than %d\n", WRITE_BUFSIZE - 2); - ret = -EINVAL; - goto out; + return -EINVAL; } } done += size; @@ -10131,19 +11291,245 @@ ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, ret = createfn(buf); if (ret) - goto out; + return ret; buf += size; } while (done < count); } - ret = done; + return done; +} -out: - kfree(kbuf); +#ifdef CONFIG_TRACER_MAX_TRACE +__init static bool tr_needs_alloc_snapshot(const char *name) +{ + char *test; + int len = strlen(name); + bool ret; + + if (!boot_snapshot_index) + return false; + + if (strncmp(name, boot_snapshot_info, len) == 0 && + boot_snapshot_info[len] == '\t') + return true; + + test = kmalloc(strlen(name) + 3, GFP_KERNEL); + if (!test) + return false; + sprintf(test, "\t%s\t", name); + ret = strstr(boot_snapshot_info, test) == NULL; + kfree(test); return ret; } +__init static void do_allocate_snapshot(const char *name) +{ + if (!tr_needs_alloc_snapshot(name)) + return; + + /* + * When allocate_snapshot is set, the next call to + * allocate_trace_buffers() (called by trace_array_get_by_name()) + * will allocate the snapshot buffer. That will also clear + * this flag. + */ + allocate_snapshot = true; +} +#else +static inline void do_allocate_snapshot(const char *name) { } +#endif + +__init static int backup_instance_area(const char *backup, + unsigned long *addr, phys_addr_t *size) +{ + struct trace_array *backup_tr; + void *allocated_vaddr = NULL; + + backup_tr = trace_array_get_by_name(backup, NULL); + if (!backup_tr) { + pr_warn("Tracing: Instance %s is not found.\n", backup); + return -ENOENT; + } + + if (!(backup_tr->flags & TRACE_ARRAY_FL_BOOT)) { + pr_warn("Tracing: Instance %s is not boot mapped.\n", backup); + trace_array_put(backup_tr); + return -EINVAL; + } + + *size = backup_tr->range_addr_size; + + allocated_vaddr = vzalloc(*size); + if (!allocated_vaddr) { + pr_warn("Tracing: Failed to allocate memory for copying instance %s (size 0x%lx)\n", + backup, (unsigned long)*size); + trace_array_put(backup_tr); + return -ENOMEM; + } + + memcpy(allocated_vaddr, + (void *)backup_tr->range_addr_start, (size_t)*size); + *addr = (unsigned long)allocated_vaddr; + + trace_array_put(backup_tr); + return 0; +} + +__init static void enable_instances(void) +{ + struct trace_array *tr; + bool memmap_area = false; + char *curr_str; + char *name; + char *str; + char *tok; + + /* A tab is always appended */ + boot_instance_info[boot_instance_index - 1] = '\0'; + str = boot_instance_info; + + while ((curr_str = strsep(&str, "\t"))) { + phys_addr_t start = 0; + phys_addr_t size = 0; + unsigned long addr = 0; + bool traceprintk = false; + bool traceoff = false; + char *flag_delim; + char *addr_delim; + char *rname __free(kfree) = NULL; + char *backup; + + tok = strsep(&curr_str, ","); + + name = strsep(&tok, "="); + backup = tok; + + flag_delim = strchr(name, '^'); + addr_delim = strchr(name, '@'); + + if (addr_delim) + *addr_delim++ = '\0'; + + if (flag_delim) + *flag_delim++ = '\0'; + + if (backup) { + if (backup_instance_area(backup, &addr, &size) < 0) + continue; + } + + if (flag_delim) { + char *flag; + + while ((flag = strsep(&flag_delim, "^"))) { + if (strcmp(flag, "traceoff") == 0) { + traceoff = true; + } else if ((strcmp(flag, "printk") == 0) || + (strcmp(flag, "traceprintk") == 0) || + (strcmp(flag, "trace_printk") == 0)) { + traceprintk = true; + } else { + pr_info("Tracing: Invalid instance flag '%s' for %s\n", + flag, name); + } + } + } + + tok = addr_delim; + if (tok && isdigit(*tok)) { + start = memparse(tok, &tok); + if (!start) { + pr_warn("Tracing: Invalid boot instance address for %s\n", + name); + continue; + } + if (*tok != ':') { + pr_warn("Tracing: No size specified for instance %s\n", name); + continue; + } + tok++; + size = memparse(tok, &tok); + if (!size) { + pr_warn("Tracing: Invalid boot instance size for %s\n", + name); + continue; + } + memmap_area = true; + } else if (tok) { + if (!reserve_mem_find_by_name(tok, &start, &size)) { + start = 0; + pr_warn("Failed to map boot instance %s to %s\n", name, tok); + continue; + } + rname = kstrdup(tok, GFP_KERNEL); + } + + if (start) { + /* Start and size must be page aligned */ + if (start & ~PAGE_MASK) { + pr_warn("Tracing: mapping start addr %pa is not page aligned\n", &start); + continue; + } + if (size & ~PAGE_MASK) { + pr_warn("Tracing: mapping size %pa is not page aligned\n", &size); + continue; + } + + if (memmap_area) + addr = map_pages(start, size); + else + addr = (unsigned long)phys_to_virt(start); + if (addr) { + pr_info("Tracing: mapped boot instance %s at physical memory %pa of size 0x%lx\n", + name, &start, (unsigned long)size); + } else { + pr_warn("Tracing: Failed to map boot instance %s\n", name); + continue; + } + } else { + /* Only non mapped buffers have snapshot buffers */ + if (IS_ENABLED(CONFIG_TRACER_MAX_TRACE)) + do_allocate_snapshot(name); + } + + tr = trace_array_create_systems(name, NULL, addr, size); + if (IS_ERR(tr)) { + pr_warn("Tracing: Failed to create instance buffer %s\n", curr_str); + continue; + } + + if (traceoff) + tracer_tracing_off(tr); + + if (traceprintk) + update_printk_trace(tr); + + /* + * memmap'd buffers can not be freed. + */ + if (memmap_area) { + tr->flags |= TRACE_ARRAY_FL_MEMMAP; + tr->ref++; + } + + /* + * Backup buffers can be freed but need vfree(). + */ + if (backup) + tr->flags |= TRACE_ARRAY_FL_VMALLOC; + + if (start || backup) { + tr->flags |= TRACE_ARRAY_FL_BOOT | TRACE_ARRAY_FL_LAST_BOOT; + tr->range_name = no_free_ptr(rname); + } + + while ((tok = strsep(&curr_str, ","))) { + early_enable_events(tr, tok, true); + } + } +} + __init static int tracer_alloc_buffers(void) { int ring_buf_size; @@ -10162,7 +11548,7 @@ __init static int tracer_alloc_buffers(void) BUILD_BUG_ON(TRACE_ITER_LAST_BIT > TRACE_FLAGS_MAX_SIZE); if (!alloc_cpumask_var(&tracing_buffer_mask, GFP_KERNEL)) - goto out; + return -ENOMEM; if (!alloc_cpumask_var(&global_trace.tracing_cpumask, GFP_KERNEL)) goto out_free_buffer_mask; @@ -10173,7 +11559,7 @@ __init static int tracer_alloc_buffers(void) trace_printk_init_buffers(); /* To save memory, keep the ring buffer size to its minimum */ - if (ring_buffer_expanded) + if (global_trace.ring_buffer_expanded) ring_buf_size = trace_buf_size; else ring_buf_size = 1; @@ -10203,12 +11589,14 @@ __init static int tracer_alloc_buffers(void) if (trace_create_savedcmd() < 0) goto out_free_temp_buffer; + if (!zalloc_cpumask_var(&global_trace.pipe_cpumask, GFP_KERNEL)) + goto out_free_savedcmd; + /* TODO: make the number of buffers hot pluggable with CPUS */ if (allocate_trace_buffers(&global_trace, ring_buf_size) < 0) { MEM_FAIL(1, "tracer: failed to allocate ring buffer!\n"); - goto out_free_savedcmd; + goto out_free_pipe_cpumask; } - if (global_trace.buffer_disabled) tracing_off(); @@ -10225,17 +11613,21 @@ __init static int tracer_alloc_buffers(void) * just a bootstrap of current_trace anyway. */ global_trace.current_trace = &nop_trace; + global_trace.current_trace_flags = nop_trace.flags; global_trace.max_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; - +#ifdef CONFIG_TRACER_MAX_TRACE + spin_lock_init(&global_trace.snapshot_trigger_lock); +#endif ftrace_init_global_array_ops(&global_trace); - init_trace_flags_index(&global_trace); +#ifdef CONFIG_MODULES + INIT_LIST_HEAD(&global_trace.mod_events); +#endif - register_tracer(&nop_trace); + init_trace_flags_index(&global_trace); - /* Function tracing may start here (via kernel command line) */ - init_function_trace(); + INIT_LIST_HEAD(&global_trace.tracers); /* All seems OK, enable tracing */ tracing_disabled = 0; @@ -10247,22 +11639,30 @@ __init static int tracer_alloc_buffers(void) global_trace.flags = TRACE_ARRAY_FL_GLOBAL; + global_trace.syscall_buf_sz = syscall_buf_size; + INIT_LIST_HEAD(&global_trace.systems); INIT_LIST_HEAD(&global_trace.events); INIT_LIST_HEAD(&global_trace.hist_vars); INIT_LIST_HEAD(&global_trace.err_log); + list_add(&global_trace.marker_list, &marker_copies); list_add(&global_trace.list, &ftrace_trace_arrays); + register_tracer(&nop_trace); + + /* Function tracing may start here (via kernel command line) */ + init_function_trace(); + apply_trace_boot_options(); register_snapshot_cmd(); - test_can_verify(); - return 0; +out_free_pipe_cpumask: + free_cpumask_var(global_trace.pipe_cpumask); out_free_savedcmd: - free_saved_cmdlines_buffer(savedcmd); + trace_free_saved_cmdlines_buffer(); out_free_temp_buffer: ring_buffer_free(temp_buffer); out_rm_hp_state: @@ -10271,16 +11671,33 @@ out_free_cpumask: free_cpumask_var(global_trace.tracing_cpumask); out_free_buffer_mask: free_cpumask_var(tracing_buffer_mask); -out: return ret; } +#ifdef CONFIG_FUNCTION_TRACER +/* Used to set module cached ftrace filtering at boot up */ +struct trace_array *trace_get_global_array(void) +{ + return &global_trace; +} +#endif + void __init ftrace_boot_snapshot(void) { - if (snapshot_at_boot) { - tracing_snapshot(); - internal_trace_puts("** Boot snapshot taken **\n"); +#ifdef CONFIG_TRACER_MAX_TRACE + struct trace_array *tr; + + if (!snapshot_at_boot) + return; + + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (!tr->allocated_snapshot) + continue; + + tracing_snapshot_instance(tr); + trace_array_puts(tr, "** Boot snapshot taken **\n"); } +#endif } void __init early_trace_init(void) @@ -10302,6 +11719,9 @@ void __init early_trace_init(void) void __init trace_init(void) { trace_event_init(); + + if (boot_instance_index) + enable_instances(); } __init static void clear_boot_tracer(void) @@ -10350,6 +11770,9 @@ __init static int late_trace_init(void) tracepoint_printk = 0; } + if (traceoff_after_boot) + tracing_off(); + tracing_set_default_clock(); clear_boot_tracer(); return 0; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 4eb6d6b97a9f..b6d42fe06115 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -21,11 +21,13 @@ #include <linux/workqueue.h> #include <linux/ctype.h> #include <linux/once_lite.h> +#include <linux/ftrace_regs.h> +#include <linux/llist.h> #include "pid_list.h" #ifdef CONFIG_FTRACE_SYSCALLS -#include <asm/unistd.h> /* For NR_SYSCALLS */ +#include <asm/unistd.h> /* For NR_syscalls */ #include <asm/syscall.h> /* some archs define it here */ #endif @@ -46,6 +48,7 @@ enum trace_type { TRACE_BRANCH, TRACE_GRAPH_RET, TRACE_GRAPH_ENT, + TRACE_GRAPH_RETADDR_ENT, TRACE_USER_STACK, TRACE_BLK, TRACE_BPUTS, @@ -77,6 +80,16 @@ enum trace_type { #undef __array #define __array(type, item, size) type item[size]; +/* + * For backward compatibility, older user space expects to see the + * kernel_stack event with a fixed size caller field. But today the fix + * size is ignored by the kernel, and the real structure is dynamic. + * Expose to user space: "unsigned long caller[8];" but the real structure + * will be "unsigned long caller[] __counted_by(size)" + */ +#undef __stack_array +#define __stack_array(type, item, size, field) type item[] __counted_by(field); + #undef __array_desc #define __array_desc(type, container, item, size) @@ -113,6 +126,14 @@ enum trace_type { #define MEM_FAIL(condition, fmt, ...) \ DO_ONCE_LITE_IF(condition, pr_err, "ERROR: " fmt, ##__VA_ARGS__) +#define FAULT_STRING "(fault)" + +#define HIST_STACKTRACE_DEPTH 16 +#define HIST_STACKTRACE_SIZE (HIST_STACKTRACE_DEPTH * sizeof(unsigned long)) +#define HIST_STACKTRACE_SKIP 5 + +#define SYSCALL_FAULT_USER_MAX 165 + /* * syscalls are special, and need special handling, this is why * they are not included in trace_entries.h @@ -144,6 +165,17 @@ struct kretprobe_trace_entry_head { unsigned long ret_ip; }; +struct fentry_trace_entry_head { + struct trace_entry ent; + unsigned long ip; +}; + +struct fexit_trace_entry_head { + struct trace_entry ent; + unsigned long func; + unsigned long ret_ip; +}; + #define TRACE_BUF_SIZE 1024 struct trace_array; @@ -154,8 +186,7 @@ struct trace_array; * the trace, etc.) */ struct trace_array_cpu { - atomic_t disabled; - void *buffer_page; /* ring buffer spare */ + local_t disabled; unsigned long entries; unsigned long saved_latency; @@ -188,7 +219,7 @@ struct array_buffer { int cpu; }; -#define TRACE_FLAGS_MAX_SIZE 32 +#define TRACE_FLAGS_MAX_SIZE 64 struct trace_options { struct tracer *tracer; @@ -284,6 +315,11 @@ struct trace_func_repeats { u64 ts_last_call; }; +struct trace_module_delta { + struct rcu_head rcu; + long delta[]; +}; + /* * The trace array - an array of per-CPU trace arrays. This is the * highest level data structure that individual tracers deal with. @@ -307,8 +343,8 @@ struct trace_array { */ struct array_buffer max_buffer; bool allocated_snapshot; -#endif -#ifdef CONFIG_TRACER_MAX_TRACE + spinlock_t snapshot_trigger_lock; + unsigned int snapshot; unsigned long max_latency; #ifdef CONFIG_FSNOTIFY struct dentry *d_max_latency; @@ -316,6 +352,18 @@ struct trace_array { struct irq_work fsnotify_irqwork; #endif #endif + /* The below is for memory mapped ring buffer */ + unsigned int mapped; + unsigned long range_addr_start; + unsigned long range_addr_size; + char *range_name; + long text_delta; + struct trace_module_delta *module_delta; + void *scratch; /* pointer in persistent memory */ + int scratch_size; + + int buffer_disabled; + struct trace_pid_list __rcu *filtered_pids; struct trace_pid_list __rcu *filtered_no_pids; /* @@ -332,12 +380,11 @@ struct trace_array { * CONFIG_TRACER_MAX_TRACE. */ arch_spinlock_t max_lock; - int buffer_disabled; #ifdef CONFIG_FTRACE_SYSCALLS int sys_refcount_enter; int sys_refcount_exit; - struct trace_event_file __rcu *enter_syscall_files[NR_syscalls]; - struct trace_event_file __rcu *exit_syscall_files[NR_syscalls]; + struct trace_event_file *enter_syscall_files[NR_syscalls]; + struct trace_event_file *exit_syscall_files[NR_syscalls]; #endif int stop_count; int clock_id; @@ -346,26 +393,38 @@ struct trace_array { int buffer_percent; unsigned int n_err_log_entries; struct tracer *current_trace; - unsigned int trace_flags; + struct tracer_flags *current_trace_flags; + u64 trace_flags; unsigned char trace_flags_index[TRACE_FLAGS_MAX_SIZE]; unsigned int flags; raw_spinlock_t start_lock; + const char *system_names; struct list_head err_log; struct dentry *dir; struct dentry *options; struct dentry *percpu_dir; - struct dentry *event_dir; + struct eventfs_inode *event_dir; struct trace_options *topts; struct list_head systems; struct list_head events; + struct list_head marker_list; + struct list_head tracers; struct trace_event_file *trace_marker_file; cpumask_var_t tracing_cpumask; /* only trace on set CPUs */ + /* one per_cpu trace_pipe can be opened by only one user */ + cpumask_var_t pipe_cpumask; int ref; int trace_ref; +#ifdef CONFIG_MODULES + struct list_head mod_events; +#endif #ifdef CONFIG_FUNCTION_TRACER struct ftrace_ops *ops; struct trace_pid_list __rcu *function_pids; struct trace_pid_list __rcu *function_no_pids; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + struct fgraph_ops *gops; +#endif #ifdef CONFIG_DYNAMIC_FTRACE /* All of these are protected by the ftrace_lock */ struct list_head func_probes; @@ -376,17 +435,37 @@ struct trace_array { int function_enabled; #endif int no_filter_buffering_ref; + unsigned int syscall_buf_sz; struct list_head hist_vars; #ifdef CONFIG_TRACER_SNAPSHOT struct cond_snapshot *cond_snapshot; #endif struct trace_func_repeats __percpu *last_func_repeats; + /* + * On boot up, the ring buffer is set to the minimum size, so that + * we do not waste memory on systems that are not using tracing. + */ + bool ring_buffer_expanded; }; enum { - TRACE_ARRAY_FL_GLOBAL = (1 << 0) + TRACE_ARRAY_FL_GLOBAL = BIT(0), + TRACE_ARRAY_FL_BOOT = BIT(1), + TRACE_ARRAY_FL_LAST_BOOT = BIT(2), + TRACE_ARRAY_FL_MOD_INIT = BIT(3), + TRACE_ARRAY_FL_MEMMAP = BIT(4), + TRACE_ARRAY_FL_VMALLOC = BIT(5), }; +#ifdef CONFIG_MODULES +bool module_exists(const char *module); +#else +static inline bool module_exists(const char *module) +{ + return false; +} +#endif + extern struct list_head ftrace_trace_arrays; extern struct mutex trace_types_lock; @@ -402,6 +481,8 @@ extern int tracing_set_clock(struct trace_array *tr, const char *clockstr); extern bool trace_clock_in_ns(struct trace_array *tr); +extern unsigned long trace_adjust_address(struct trace_array *tr, unsigned long addr); + /* * The global tracer (top) should be the first trace array added, * but we check the flag anyway. @@ -466,6 +547,8 @@ extern void __ftrace_bad_type(void); IF_ASSIGN(var, ent, struct trace_branch, TRACE_BRANCH); \ IF_ASSIGN(var, ent, struct ftrace_graph_ent_entry, \ TRACE_GRAPH_ENT); \ + IF_ASSIGN(var, ent, struct fgraph_retaddr_ent_entry,\ + TRACE_GRAPH_RETADDR_ENT); \ IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry, \ TRACE_GRAPH_RET); \ IF_ASSIGN(var, ent, struct func_repeats_entry, \ @@ -555,9 +638,10 @@ struct tracer { u32 old_flags, u32 bit, int set); /* Return 0 if OK with change, else return non-zero */ int (*flag_changed)(struct trace_array *tr, - u32 mask, int set); + u64 mask, int set); struct tracer *next; struct tracer_flags *flags; + struct tracer_flags *default_flags; int enabled; bool print_max; bool allow_instances; @@ -577,21 +661,41 @@ trace_buffer_iter(struct trace_iterator *iter, int cpu) int tracer_init(struct tracer *t, struct trace_array *tr); int tracing_is_enabled(void); void tracing_reset_online_cpus(struct array_buffer *buf); -void tracing_reset_current(int cpu); void tracing_reset_all_online_cpus(void); void tracing_reset_all_online_cpus_unlocked(void); int tracing_open_generic(struct inode *inode, struct file *filp); int tracing_open_generic_tr(struct inode *inode, struct file *filp); +int tracing_release_generic_tr(struct inode *inode, struct file *file); +int tracing_open_file_tr(struct inode *inode, struct file *filp); +int tracing_release_file_tr(struct inode *inode, struct file *filp); +int tracing_single_release_file_tr(struct inode *inode, struct file *filp); bool tracing_is_disabled(void); bool tracer_tracing_is_on(struct trace_array *tr); void tracer_tracing_on(struct trace_array *tr); void tracer_tracing_off(struct trace_array *tr); +void tracer_tracing_disable(struct trace_array *tr); +void tracer_tracing_enable(struct trace_array *tr); struct dentry *trace_create_file(const char *name, umode_t mode, struct dentry *parent, void *data, const struct file_operations *fops); + +/** + * tracer_tracing_is_on_cpu - show real state of ring buffer enabled on for a cpu + * @tr : the trace array to know if ring buffer is enabled + * @cpu: The cpu buffer to check if enabled + * + * Shows real state of the per CPU buffer if it is enabled or not. + */ +static inline bool tracer_tracing_is_on_cpu(struct trace_array *tr, int cpu) +{ + if (tr->array_buffer.buffer) + return ring_buffer_record_is_on_cpu(tr->array_buffer.buffer, cpu); + return false; +} + int tracing_init_dentry(void); struct ring_buffer_event; @@ -602,6 +706,8 @@ trace_buffer_lock_reserve(struct trace_buffer *buffer, unsigned long len, unsigned int trace_ctx); +int ring_buffer_meta_seq_init(struct file *file, struct trace_buffer *buffer, int cpu); + struct trace_entry *tracing_get_trace_entry(struct trace_array *tr, struct trace_array_cpu *data); @@ -613,8 +719,8 @@ void trace_buffer_unlock_commit_nostack(struct trace_buffer *buffer, bool trace_is_tracepoint_string(const char *str); const char *trace_event_format(struct trace_iterator *iter, const char *fmt); -void trace_check_vprintf(struct trace_iterator *iter, const char *fmt, - va_list ap) __printf(2, 0); +char *trace_iter_expand_format(struct trace_iterator *iter); +bool ignore_event(struct trace_iterator *iter); int trace_empty(struct trace_iterator *iter); @@ -630,7 +736,8 @@ unsigned long trace_total_entries(struct trace_array *tr); void trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, - unsigned int trace_ctx); + unsigned int trace_ctx, + struct ftrace_regs *regs); void trace_graph_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, @@ -639,9 +746,10 @@ void trace_latency_header(struct seq_file *m); void trace_default_header(struct seq_file *m); void print_trace_header(struct seq_file *m, struct trace_iterator *iter); -void trace_graph_return(struct ftrace_graph_ret *trace); -int trace_graph_entry(struct ftrace_graph_ent *trace); -void set_graph_array(struct trace_array *tr); +void trace_graph_return(struct ftrace_graph_ret *trace, struct fgraph_ops *gops, + struct ftrace_regs *fregs); +int trace_graph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops *gops, + struct ftrace_regs *fregs); void tracing_start_cmdline_record(void); void tracing_stop_cmdline_record(void); @@ -664,8 +772,6 @@ extern unsigned long tracing_thresh; /* PID filtering */ -extern int pid_max; - bool trace_find_filtered_pid(struct trace_pid_list *filtered_pids, pid_t search_pid); bool trace_ignore_this_task(struct trace_pid_list *filtered_pids, @@ -677,7 +783,6 @@ void trace_filter_add_remove_task(struct trace_pid_list *pid_list, void *trace_pid_next(struct trace_pid_list *pid_list, void *v, loff_t *pos); void *trace_pid_start(struct trace_pid_list *pid_list, loff_t *pos); int trace_pid_show(struct seq_file *m, void *v); -void trace_free_pid_list(struct trace_pid_list *pid_list); int trace_pid_write(struct trace_pid_list *filtered_pids, struct trace_pid_list **new_pid_list, const char __user *ubuf, size_t cnt); @@ -718,10 +823,14 @@ extern void trace_find_cmdline(int pid, char comm[]); extern int trace_find_tgid(int pid); extern void trace_event_follow_fork(struct trace_array *tr, bool enable); +extern int trace_events_enabled(struct trace_array *tr, const char *system); + #ifdef CONFIG_DYNAMIC_FTRACE extern unsigned long ftrace_update_tot_cnt; extern unsigned long ftrace_number_of_pages; extern unsigned long ftrace_number_of_groups; +extern u64 ftrace_update_time; +extern u64 ftrace_total_mod_time; void ftrace_init_trace_array(struct trace_array *tr); #else static inline void ftrace_init_trace_array(struct trace_array *tr) { } @@ -731,7 +840,7 @@ extern int DYN_FTRACE_TEST_NAME(void); #define DYN_FTRACE_TEST_NAME2 trace_selftest_dynamic_test_func2 extern int DYN_FTRACE_TEST_NAME2(void); -extern bool ring_buffer_expanded; +extern void trace_set_ring_buffer_expanded(struct trace_array *tr); extern bool tracing_selftest_disabled; #ifdef CONFIG_FTRACE_STARTUP_TEST @@ -769,13 +878,15 @@ static inline void __init disable_tracing_selftest(const char *reason) extern void *head_page(struct trace_array_cpu *data); extern unsigned long long ns2usecs(u64 nsec); -extern int -trace_vbprintk(unsigned long ip, const char *fmt, va_list args); -extern int -trace_vprintk(unsigned long ip, const char *fmt, va_list args); -extern int -trace_array_vprintk(struct trace_array *tr, - unsigned long ip, const char *fmt, va_list args); + +__printf(2, 0) +int trace_vbprintk(unsigned long ip, const char *fmt, va_list args); +__printf(2, 0) +int trace_vprintk(unsigned long ip, const char *fmt, va_list args); +__printf(3, 0) +int trace_array_vprintk(struct trace_array *tr, + unsigned long ip, const char *fmt, va_list args); +__printf(3, 4) int trace_array_printk_buf(struct trace_buffer *buffer, unsigned long ip, const char *fmt, ...); void trace_printk_seq(struct trace_seq *s); @@ -827,11 +938,13 @@ static __always_inline bool ftrace_hash_empty(struct ftrace_hash *hash) #define TRACE_GRAPH_PRINT_TAIL 0x100 #define TRACE_GRAPH_SLEEP_TIME 0x200 #define TRACE_GRAPH_GRAPH_TIME 0x400 +#define TRACE_GRAPH_PRINT_RETVAL 0x800 +#define TRACE_GRAPH_PRINT_RETVAL_HEX 0x1000 +#define TRACE_GRAPH_PRINT_RETADDR 0x2000 +#define TRACE_GRAPH_ARGS 0x4000 #define TRACE_GRAPH_PRINT_FILL_SHIFT 28 #define TRACE_GRAPH_PRINT_FILL_MASK (0x3 << TRACE_GRAPH_PRINT_FILL_SHIFT) -extern void ftrace_graph_sleep_time_control(bool enable); - #ifdef CONFIG_FUNCTION_PROFILER extern void ftrace_graph_graph_time_control(bool enable); #else @@ -848,15 +961,69 @@ extern void graph_trace_close(struct trace_iterator *iter); extern int __trace_graph_entry(struct trace_array *tr, struct ftrace_graph_ent *trace, unsigned int trace_ctx); +extern int __trace_graph_retaddr_entry(struct trace_array *tr, + struct ftrace_graph_ent *trace, + unsigned int trace_ctx, + unsigned long retaddr, + struct ftrace_regs *fregs); extern void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, - unsigned int trace_ctx); + unsigned int trace_ctx, + u64 calltime, u64 rettime); + +extern void init_array_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops); +extern int allocate_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops); +extern void free_fgraph_ops(struct trace_array *tr); + +enum { + TRACE_GRAPH_FL = 1, + + /* + * In the very unlikely case that an interrupt came in + * at a start of graph tracing, and we want to trace + * the function in that interrupt, the depth can be greater + * than zero, because of the preempted start of a previous + * trace. In an even more unlikely case, depth could be 2 + * if a softirq interrupted the start of graph tracing, + * followed by an interrupt preempting a start of graph + * tracing in the softirq, and depth can even be 3 + * if an NMI came in at the start of an interrupt function + * that preempted a softirq start of a function that + * preempted normal context!!!! Luckily, it can't be + * greater than 3, so the next two bits are a mask + * of what the depth is when we set TRACE_GRAPH_FL + */ + + TRACE_GRAPH_DEPTH_START_BIT, + TRACE_GRAPH_DEPTH_END_BIT, + + /* + * To implement set_graph_notrace, if this bit is set, we ignore + * function graph tracing of called functions, until the return + * function is called to clear it. + */ + TRACE_GRAPH_NOTRACE_BIT, +}; + +#define TRACE_GRAPH_NOTRACE (1 << TRACE_GRAPH_NOTRACE_BIT) + +static inline unsigned long ftrace_graph_depth(unsigned long *task_var) +{ + return (*task_var >> TRACE_GRAPH_DEPTH_START_BIT) & 3; +} + +static inline void ftrace_graph_set_depth(unsigned long *task_var, int depth) +{ + *task_var &= ~(3 << TRACE_GRAPH_DEPTH_START_BIT); + *task_var |= (depth & 3) << TRACE_GRAPH_DEPTH_START_BIT; +} #ifdef CONFIG_DYNAMIC_FTRACE extern struct ftrace_hash __rcu *ftrace_graph_hash; extern struct ftrace_hash __rcu *ftrace_graph_notrace_hash; -static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) +static inline int +ftrace_graph_addr(unsigned long *task_var, struct ftrace_graph_ent *trace) { unsigned long addr = trace->func; int ret = 0; @@ -878,13 +1045,12 @@ static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) } if (ftrace_lookup_ip(hash, addr)) { - /* * This needs to be cleared on the return functions * when the depth is zero. */ - trace_recursion_set(TRACE_GRAPH_BIT); - trace_recursion_set_depth(trace->depth); + *task_var |= TRACE_GRAPH_FL; + ftrace_graph_set_depth(task_var, trace->depth); /* * If no irqs are to be traced, but a set_graph_function @@ -903,11 +1069,14 @@ out: return ret; } -static inline void ftrace_graph_addr_finish(struct ftrace_graph_ret *trace) +static inline void +ftrace_graph_addr_finish(struct fgraph_ops *gops, struct ftrace_graph_ret *trace) { - if (trace_recursion_test(TRACE_GRAPH_BIT) && - trace->depth == trace_recursion_depth()) - trace_recursion_clear(TRACE_GRAPH_BIT); + unsigned long *task_var = fgraph_get_task_var(gops); + + if ((*task_var & TRACE_GRAPH_FL) && + trace->depth == ftrace_graph_depth(task_var)) + *task_var &= ~TRACE_GRAPH_FL; } static inline int ftrace_graph_notrace_addr(unsigned long addr) @@ -933,7 +1102,7 @@ static inline int ftrace_graph_notrace_addr(unsigned long addr) return ret; } #else -static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) +static inline int ftrace_graph_addr(unsigned long *task_var, struct ftrace_graph_ent *trace) { return 1; } @@ -942,27 +1111,39 @@ static inline int ftrace_graph_notrace_addr(unsigned long addr) { return 0; } -static inline void ftrace_graph_addr_finish(struct ftrace_graph_ret *trace) +static inline void ftrace_graph_addr_finish(struct fgraph_ops *gops, struct ftrace_graph_ret *trace) { } #endif /* CONFIG_DYNAMIC_FTRACE */ extern unsigned int fgraph_max_depth; +extern int fgraph_no_sleep_time; +extern bool fprofile_no_sleep_time; -static inline bool ftrace_graph_ignore_func(struct ftrace_graph_ent *trace) +static inline bool +ftrace_graph_ignore_func(struct fgraph_ops *gops, struct ftrace_graph_ent *trace) { + unsigned long *task_var = fgraph_get_task_var(gops); + /* trace it when it is-nested-in or is a function enabled. */ - return !(trace_recursion_test(TRACE_GRAPH_BIT) || - ftrace_graph_addr(trace)) || + return !((*task_var & TRACE_GRAPH_FL) || + ftrace_graph_addr(task_var, trace)) || (trace->depth < 0) || (fgraph_max_depth && trace->depth >= fgraph_max_depth); } +void fgraph_init_ops(struct ftrace_ops *dst_ops, + struct ftrace_ops *src_ops); + #else /* CONFIG_FUNCTION_GRAPH_TRACER */ static inline enum print_line_t print_graph_function_flags(struct trace_iterator *iter, u32 flags) { return TRACE_TYPE_UNHANDLED; } +static inline void free_fgraph_ops(struct trace_array *tr) { } +/* ftrace_ops may not be defined */ +#define init_array_fgraph_ops(tr, ops) do { } while (0) +#define allocate_fgraph_ops(tr, ops) ({ 0; }) #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ extern struct list_head ftrace_pids; @@ -981,11 +1162,6 @@ struct ftrace_func_command { char *params, int enable); }; extern bool ftrace_filter_param __initdata; -static inline int ftrace_trace_task(struct trace_array *tr) -{ - return this_cpu_read(tr->array_buffer.data->ftrace_ignore_pid) != - FTRACE_PID_IGNORE; -} extern int ftrace_is_dead(void); int ftrace_create_function_files(struct trace_array *tr, struct dentry *parent); @@ -993,6 +1169,7 @@ void ftrace_destroy_function_files(struct trace_array *tr); int ftrace_allocate_ftrace_ops(struct trace_array *tr); void ftrace_free_ftrace_ops(struct trace_array *tr); void ftrace_init_global_array_ops(struct trace_array *tr); +struct trace_array *trace_get_global_array(void); void ftrace_init_array_ops(struct trace_array *tr, ftrace_func_t func); void ftrace_reset_array_ops(struct trace_array *tr); void ftrace_init_tracefs(struct trace_array *tr, struct dentry *d_tracer); @@ -1002,10 +1179,6 @@ void ftrace_clear_pids(struct trace_array *tr); int init_function_trace(void); void ftrace_pid_follow_fork(struct trace_array *tr, bool enable); #else -static inline int ftrace_trace_task(struct trace_array *tr) -{ - return 1; -} static inline int ftrace_is_dead(void) { return 0; } static inline int ftrace_create_function_files(struct trace_array *tr, @@ -1118,6 +1291,7 @@ bool ftrace_event_is_function(struct trace_event_call *call); */ struct trace_parser { bool cont; + bool fail; char *buffer; unsigned idx; unsigned size; @@ -1125,7 +1299,7 @@ struct trace_parser { static inline bool trace_parser_loaded(struct trace_parser *parser) { - return (parser->idx != 0); + return !parser->fail && parser->idx != 0; } static inline bool trace_parser_cont(struct trace_parser *parser) @@ -1139,6 +1313,11 @@ static inline void trace_parser_clear(struct trace_parser *parser) parser->idx = 0; } +static inline void trace_parser_fail(struct trace_parser *parser) +{ + parser->fail = true; +} + extern int trace_parser_get_init(struct trace_parser *parser, int size); extern void trace_parser_put(struct trace_parser *parser); extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, @@ -1165,11 +1344,11 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, # define FUNCTION_FLAGS \ C(FUNCTION, "function-trace"), \ C(FUNC_FORK, "function-fork"), -# define FUNCTION_DEFAULT_FLAGS TRACE_ITER_FUNCTION +# define FUNCTION_DEFAULT_FLAGS TRACE_ITER(FUNCTION) #else # define FUNCTION_FLAGS # define FUNCTION_DEFAULT_FLAGS 0UL -# define TRACE_ITER_FUNC_FORK 0UL +# define TRACE_ITER_FUNC_FORK_BIT -1 #endif #ifdef CONFIG_STACKTRACE @@ -1179,6 +1358,24 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, # define STACK_FLAGS #endif +#ifdef CONFIG_FUNCTION_PROFILER +# define PROFILER_FLAGS \ + C(PROF_TEXT_OFFSET, "prof-text-offset"), +# ifdef CONFIG_FUNCTION_GRAPH_TRACER +# define FPROFILE_FLAGS \ + C(GRAPH_TIME, "graph-time"), +# define FPROFILE_DEFAULT_FLAGS TRACE_ITER(GRAPH_TIME) +# else +# define FPROFILE_FLAGS +# define FPROFILE_DEFAULT_FLAGS 0UL +# endif +#else +# define PROFILER_FLAGS +# define FPROFILE_FLAGS +# define FPROFILE_DEFAULT_FLAGS 0UL +# define TRACE_ITER_PROF_TEXT_OFFSET_BIT -1 +#endif + /* * trace_iterator_flags is an enumeration that defines bit * positions into trace_flags that controls the output. @@ -1195,6 +1392,7 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, C(HEX, "hex"), \ C(BIN, "bin"), \ C(BLOCK, "block"), \ + C(FIELDS, "fields"), \ C(PRINTK, "trace_printk"), \ C(ANNOTATE, "annotate"), \ C(USERSTACKTRACE, "userstacktrace"), \ @@ -1209,12 +1407,16 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, C(IRQ_INFO, "irq-info"), \ C(MARKERS, "markers"), \ C(EVENT_FORK, "event-fork"), \ + C(TRACE_PRINTK, "trace_printk_dest"), \ + C(COPY_MARKER, "copy_trace_marker"), \ C(PAUSE_ON_TRACE, "pause-on-trace"), \ C(HASH_PTR, "hash-ptr"), /* Print hashed pointer */ \ FUNCTION_FLAGS \ FGRAPH_FLAGS \ STACK_FLAGS \ - BRANCH_FLAGS + BRANCH_FLAGS \ + PROFILER_FLAGS \ + FPROFILE_FLAGS /* * By defining C, we can make TRACE_FLAGS a list of bit names @@ -1230,20 +1432,17 @@ enum trace_iterator_bits { }; /* - * By redefining C, we can make TRACE_FLAGS a list of masks that - * use the bits as defined above. + * And use TRACE_ITER(flag) to define the bit masks. */ -#undef C -#define C(a, b) TRACE_ITER_##a = (1 << TRACE_ITER_##a##_BIT) - -enum trace_iterator_flags { TRACE_FLAGS }; +#define TRACE_ITER(flag) \ + (TRACE_ITER_##flag##_BIT < 0 ? 0 : 1ULL << (TRACE_ITER_##flag##_BIT)) /* * TRACE_ITER_SYM_MASK masks the options in trace_flags that * control the output of kernel symbols. */ #define TRACE_ITER_SYM_MASK \ - (TRACE_ITER_PRINT_PARENT|TRACE_ITER_SYM_OFFSET|TRACE_ITER_SYM_ADDR) + (TRACE_ITER(PRINT_PARENT)|TRACE_ITER(SYM_OFFSET)|TRACE_ITER(SYM_ADDR)) extern struct tracer nop_trace; @@ -1252,7 +1451,7 @@ extern int enable_branch_tracing(struct trace_array *tr); extern void disable_branch_tracing(void); static inline int trace_branch_enable(struct trace_array *tr) { - if (tr->trace_flags & TRACE_ITER_BRANCH) + if (tr->trace_flags & TRACE_ITER(BRANCH)) return enable_branch_tracing(tr); return 0; } @@ -1272,7 +1471,15 @@ static inline void trace_branch_disable(void) #endif /* CONFIG_BRANCH_TRACER */ /* set ring buffers to default size if not already done so */ -int tracing_update_buffers(void); +int tracing_update_buffers(struct trace_array *tr); + +union trace_synth_field { + u8 as_u8; + u16 as_u16; + u32 as_u32; + u64 as_u64; + struct trace_dynamic_info as_dynamic; +}; struct ftrace_event_field { struct list_head link; @@ -1281,7 +1488,9 @@ struct ftrace_event_field { int filter_type; int offset; int size; - int is_signed; + unsigned int is_signed:1; + unsigned int needs_test:1; + int len; }; struct prog_entry; @@ -1302,15 +1511,11 @@ struct trace_subsystem_dir { struct list_head list; struct event_subsystem *subsystem; struct trace_array *tr; - struct dentry *entry; + struct eventfs_inode *ei; int ref_count; int nr_events; }; -extern int call_filter_check_discard(struct trace_event_call *call, void *rec, - struct trace_buffer *buffer, - struct ring_buffer_event *event); - void trace_buffer_unlock_commit_regs(struct trace_array *tr, struct trace_buffer *buffer, struct ring_buffer_event *event, @@ -1325,11 +1530,40 @@ static inline void trace_buffer_unlock_commit(struct trace_array *tr, trace_buffer_unlock_commit_regs(tr, buffer, event, trace_ctx, NULL); } +DECLARE_PER_CPU(bool, trace_taskinfo_save); +int trace_save_cmdline(struct task_struct *tsk); +int trace_create_savedcmd(void); +int trace_alloc_tgid_map(void); +void trace_free_saved_cmdlines_buffer(void); + +extern const struct file_operations tracing_saved_cmdlines_fops; +extern const struct file_operations tracing_saved_tgids_fops; +extern const struct file_operations tracing_saved_cmdlines_size_fops; + DECLARE_PER_CPU(struct ring_buffer_event *, trace_buffered_event); DECLARE_PER_CPU(int, trace_buffered_event_cnt); void trace_buffered_event_disable(void); void trace_buffered_event_enable(void); +void early_enable_events(struct trace_array *tr, char *buf, bool disable_first); + +struct trace_user_buf; +struct trace_user_buf_info { + struct trace_user_buf __percpu *tbuf; + size_t size; + int ref; +}; + +typedef int (*trace_user_buf_copy)(char *dst, const char __user *src, + size_t size, void *data); +int trace_user_fault_init(struct trace_user_buf_info *tinfo, size_t size); +int trace_user_fault_get(struct trace_user_buf_info *tinfo); +int trace_user_fault_put(struct trace_user_buf_info *tinfo); +void trace_user_fault_destroy(struct trace_user_buf_info *tinfo); +char *trace_user_fault_read(struct trace_user_buf_info *tinfo, + const char __user *ptr, size_t size, + trace_user_buf_copy copy_func, void *data); + static inline void __trace_event_discard_commit(struct trace_buffer *buffer, struct ring_buffer_event *event) @@ -1510,6 +1744,29 @@ static inline void *event_file_data(struct file *filp) extern struct mutex event_mutex; extern struct list_head ftrace_events; +/* + * When the trace_event_file is the filp->i_private pointer, + * it must be taken under the event_mutex lock, and then checked + * if the EVENT_FILE_FL_FREED flag is set. If it is, then the + * data pointed to by the trace_event_file can not be trusted. + * + * Use the event_file_file() to access the trace_event_file from + * the filp the first time under the event_mutex and check for + * NULL. If it is needed to be retrieved again and the event_mutex + * is still held, then the event_file_data() can be used and it + * is guaranteed to be valid. + */ +static inline struct trace_event_file *event_file_file(struct file *filp) +{ + struct trace_event_file *file; + + lockdep_assert_held(&event_mutex); + file = READ_ONCE(file_inode(filp)->i_private); + if (!file || file->flags & EVENT_FILE_FL_FREED) + return NULL; + return file; +} + extern const struct file_operations event_trigger_fops; extern const struct file_operations event_hist_fops; extern const struct file_operations event_hist_debug_fops; @@ -1528,13 +1785,13 @@ extern void clear_event_triggers(struct trace_array *tr); enum { EVENT_TRIGGER_FL_PROBE = BIT(0), + EVENT_TRIGGER_FL_COUNT = BIT(1), }; struct event_trigger_data { unsigned long count; int ref; int flags; - struct event_trigger_ops *ops; struct event_command *cmd_ops; struct event_filter __rcu *filter; char *filter_str; @@ -1545,6 +1802,7 @@ struct event_trigger_data { char *name; struct list_head named_list; struct event_trigger_data *named_data; + struct llist_node llist; }; /* Avoid typos */ @@ -1559,6 +1817,10 @@ struct enable_trigger_data { bool hist; }; +bool event_trigger_count(struct event_trigger_data *data, + struct trace_buffer *buffer, void *rec, + struct ring_buffer_event *event); + extern int event_enable_trigger_print(struct seq_file *m, struct event_trigger_data *data); extern void event_enable_trigger_free(struct event_trigger_data *data); @@ -1572,6 +1834,9 @@ extern int event_enable_register_trigger(char *glob, extern void event_enable_unregister_trigger(char *glob, struct event_trigger_data *test, struct trace_event_file *file); +extern struct event_trigger_data * +trigger_data_alloc(struct event_command *cmd_ops, char *cmd, char *param, + void *private_data); extern void trigger_data_free(struct event_trigger_data *data); extern int event_trigger_init(struct event_trigger_data *data); extern int trace_event_trigger_enable_disable(struct trace_event_file *file, @@ -1598,11 +1863,6 @@ extern bool event_trigger_check_remove(const char *glob); extern bool event_trigger_empty_param(const char *param); extern int event_trigger_separate_filter(char *param_and_filter, char **param, char **filter, bool param_required); -extern struct event_trigger_data * -event_trigger_alloc(struct event_command *cmd_ops, - char *cmd, - char *param, - void *private_data); extern int event_trigger_parse_num(char *trigger, struct event_trigger_data *trigger_data); extern int event_trigger_set_filter(struct event_command *cmd_ops, @@ -1620,63 +1880,8 @@ extern void event_trigger_unregister(struct event_command *cmd_ops, char *glob, struct event_trigger_data *trigger_data); -/** - * struct event_trigger_ops - callbacks for trace event triggers - * - * The methods in this structure provide per-event trigger hooks for - * various trigger operations. - * - * The @init and @free methods are used during trigger setup and - * teardown, typically called from an event_command's @parse() - * function implementation. - * - * The @print method is used to print the trigger spec. - * - * The @trigger method is the function that actually implements the - * trigger and is called in the context of the triggering event - * whenever that event occurs. - * - * All the methods below, except for @init() and @free(), must be - * implemented. - * - * @trigger: The trigger 'probe' function called when the triggering - * event occurs. The data passed into this callback is the data - * that was supplied to the event_command @reg() function that - * registered the trigger (see struct event_command) along with - * the trace record, rec. - * - * @init: An optional initialization function called for the trigger - * when the trigger is registered (via the event_command reg() - * function). This can be used to perform per-trigger - * initialization such as incrementing a per-trigger reference - * count, for instance. This is usually implemented by the - * generic utility function @event_trigger_init() (see - * trace_event_triggers.c). - * - * @free: An optional de-initialization function called for the - * trigger when the trigger is unregistered (via the - * event_command @reg() function). This can be used to perform - * per-trigger de-initialization such as decrementing a - * per-trigger reference count and freeing corresponding trigger - * data, for instance. This is usually implemented by the - * generic utility function @event_trigger_free() (see - * trace_event_triggers.c). - * - * @print: The callback function invoked to have the trigger print - * itself. This is usually implemented by a wrapper function - * that calls the generic utility function @event_trigger_print() - * (see trace_event_triggers.c). - */ -struct event_trigger_ops { - void (*trigger)(struct event_trigger_data *data, - struct trace_buffer *buffer, - void *rec, - struct ring_buffer_event *rbe); - int (*init)(struct event_trigger_data *data); - void (*free)(struct event_trigger_data *data); - int (*print)(struct seq_file *m, - struct event_trigger_data *data); -}; +extern void event_file_get(struct trace_event_file *file); +extern void event_file_put(struct trace_event_file *file); /** * struct event_command - callbacks and data members for event commands @@ -1727,7 +1932,7 @@ struct event_trigger_ops { * * @reg: Adds the trigger to the list of triggers associated with the * event, and enables the event trigger itself, after - * initializing it (via the event_trigger_ops @init() function). + * initializing it (via the event_command @init() function). * This is also where commands can use the @trigger_type value to * make the decision as to whether or not multiple instances of * the trigger should be allowed. This is usually implemented by @@ -1736,7 +1941,7 @@ struct event_trigger_ops { * * @unreg: Removes the trigger from the list of triggers associated * with the event, and disables the event trigger itself, after - * initializing it (via the event_trigger_ops @free() function). + * initializing it (via the event_command @free() function). * This is usually implemented by the generic utility function * @unregister_trigger() (see trace_event_triggers.c). * @@ -1750,12 +1955,41 @@ struct event_trigger_ops { * ignored. This is usually implemented by the generic utility * function @set_trigger_filter() (see trace_event_triggers.c). * - * @get_trigger_ops: The callback function invoked to retrieve the - * event_trigger_ops implementation associated with the command. - * This callback function allows a single event_command to - * support multiple trigger implementations via different sets of - * event_trigger_ops, depending on the value of the @param - * string. + * All the methods below, except for @init() and @free(), must be + * implemented. + * + * @trigger: The trigger 'probe' function called when the triggering + * event occurs. The data passed into this callback is the data + * that was supplied to the event_command @reg() function that + * registered the trigger (see struct event_command) along with + * the trace record, rec. + * + * @count_func: If defined and a numeric parameter is passed to the + * trigger, then this function will be called before @trigger + * is called. If this function returns false, then @trigger is not + * executed. + * + * @init: An optional initialization function called for the trigger + * when the trigger is registered (via the event_command reg() + * function). This can be used to perform per-trigger + * initialization such as incrementing a per-trigger reference + * count, for instance. This is usually implemented by the + * generic utility function @event_trigger_init() (see + * trace_event_triggers.c). + * + * @free: An optional de-initialization function called for the + * trigger when the trigger is unregistered (via the + * event_command @reg() function). This can be used to perform + * per-trigger de-initialization such as decrementing a + * per-trigger reference count and freeing corresponding trigger + * data, for instance. This is usually implemented by the + * generic utility function @event_trigger_free() (see + * trace_event_triggers.c). + * + * @print: The callback function invoked to have the trigger print + * itself. This is usually implemented by a wrapper function + * that calls the generic utility function @event_trigger_print() + * (see trace_event_triggers.c). */ struct event_command { struct list_head list; @@ -1776,7 +2010,18 @@ struct event_command { int (*set_filter)(char *filter_str, struct event_trigger_data *data, struct trace_event_file *file); - struct event_trigger_ops *(*get_trigger_ops)(char *cmd, char *param); + void (*trigger)(struct event_trigger_data *data, + struct trace_buffer *buffer, + void *rec, + struct ring_buffer_event *rbe); + bool (*count_func)(struct event_trigger_data *data, + struct trace_buffer *buffer, + void *rec, + struct ring_buffer_event *rbe); + int (*init)(struct event_trigger_data *data); + void (*free)(struct event_trigger_data *data); + int (*print)(struct seq_file *m, + struct event_trigger_data *data); }; /** @@ -1797,7 +2042,7 @@ struct event_command { * either committed or discarded. At that point, if any commands * have deferred their triggers, those commands are finally * invoked following the close of the current event. In other - * words, if the event_trigger_ops @func() probe implementation + * words, if the event_command @func() probe implementation * itself logs to the trace buffer, this flag should be set, * otherwise it can be left unspecified. * @@ -1839,8 +2084,8 @@ extern const char *__stop___tracepoint_str[]; void trace_printk_control(bool enabled); void trace_printk_start_comm(void); -int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set); -int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled); +int trace_keep_overwrite(struct tracer *tracer, u64 mask, int set); +int set_tracer_flag(struct trace_array *tr, u64 mask, int enabled); /* Used from boot time tracer */ extern int trace_set_options(struct trace_array *tr, char *option); @@ -1906,24 +2151,28 @@ static inline const char *get_syscall_name(int syscall) #ifdef CONFIG_EVENT_TRACING void trace_event_init(void); -void trace_event_eval_update(struct trace_eval_map **map, int len); +void trace_event_update_all(struct trace_eval_map **map, int len); /* Used from boot time tracer */ extern int ftrace_set_clr_event(struct trace_array *tr, char *buf, int set); extern int trigger_process_regex(struct trace_event_file *file, char *buff); #else static inline void __init trace_event_init(void) { } -static inline void trace_event_eval_update(struct trace_eval_map **map, int len) { } +static inline void trace_event_update_all(struct trace_eval_map **map, int len) { } #endif #ifdef CONFIG_TRACER_SNAPSHOT void tracing_snapshot_instance(struct trace_array *tr); int tracing_alloc_snapshot_instance(struct trace_array *tr); +int tracing_arm_snapshot(struct trace_array *tr); +void tracing_disarm_snapshot(struct trace_array *tr); #else static inline void tracing_snapshot_instance(struct trace_array *tr) { } static inline int tracing_alloc_snapshot_instance(struct trace_array *tr) { return 0; } +static inline int tracing_arm_snapshot(struct trace_array *tr) { return 0; } +static inline void tracing_disarm_snapshot(struct trace_array *tr) { } #endif #ifdef CONFIG_PREEMPT_TRACER @@ -1981,7 +2230,7 @@ static inline bool is_good_system_name(const char *name) static inline void sanitize_event_name(char *name) { while (*name++ != '\0') - if (*name == ':' || *name == '.') + if (*name == ':' || *name == '.' || *name == '*') *name = '_'; } @@ -2012,4 +2261,32 @@ static inline int rv_init_interface(void) } #endif +/* + * This is used only to distinguish + * function address from trampoline code. + * So this value has no meaning. + */ +#define FTRACE_TRAMPOLINE_MARKER ((unsigned long) INT_MAX) + +/* + * This is used to get the address of the args array based on + * the type of the entry. + */ +#define FGRAPH_ENTRY_ARGS(e) \ + ({ \ + unsigned long *_args; \ + struct ftrace_graph_ent_entry *_e = e; \ + \ + if (IS_ENABLED(CONFIG_FUNCTION_GRAPH_RETADDR) && \ + e->ent.type == TRACE_GRAPH_RETADDR_ENT) { \ + struct fgraph_retaddr_ent_entry *_re; \ + \ + _re = (typeof(_re))_e; \ + _args = _re->args; \ + } else { \ + _args = _e->args; \ + } \ + _args; \ + }) + #endif /* _LINUX_KERNEL_TRACE_H */ diff --git a/kernel/trace/trace_benchmark.c b/kernel/trace/trace_benchmark.c index 54d5fa35c90a..e19c32f2a938 100644 --- a/kernel/trace/trace_benchmark.c +++ b/kernel/trace/trace_benchmark.c @@ -92,7 +92,6 @@ static void trace_do_benchmark(void) bm_total += delta; bm_totalsq += delta * delta; - if (bm_cnt > 1) { /* * Apply Welford's method to calculate standard deviation: @@ -105,7 +104,7 @@ static void trace_do_benchmark(void) stddev = 0; delta = bm_total; - do_div(delta, bm_cnt); + do_div(delta, (u32)bm_cnt); avg = delta; if (stddev > 0) { @@ -127,7 +126,7 @@ static void trace_do_benchmark(void) seed = stddev; if (!last_seed) break; - do_div(seed, last_seed); + seed = div64_u64(seed, last_seed); seed += last_seed; do_div(seed, 2); } while (i++ < 10 && last_seed != seed); diff --git a/kernel/trace/trace_boot.c b/kernel/trace/trace_boot.c index 778200dd8ede..dbe29b4c6a7a 100644 --- a/kernel/trace/trace_boot.c +++ b/kernel/trace/trace_boot.c @@ -31,7 +31,7 @@ trace_boot_set_instance_options(struct trace_array *tr, struct xbc_node *node) /* Common ftrace options */ xbc_node_for_each_array_value(node, "options", anode, p) { - if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) { + if (strscpy(buf, p, ARRAY_SIZE(buf)) < 0) { pr_err("String is too long: %s\n", p); continue; } @@ -87,7 +87,7 @@ trace_boot_enable_events(struct trace_array *tr, struct xbc_node *node) const char *p; xbc_node_for_each_array_value(node, "events", anode, p) { - if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) { + if (strscpy(buf, p, ARRAY_SIZE(buf)) < 0) { pr_err("String is too long: %s\n", p); continue; } @@ -486,7 +486,7 @@ trace_boot_init_one_event(struct trace_array *tr, struct xbc_node *gnode, p = xbc_node_find_value(enode, "filter", NULL); if (p && *p != '\0') { - if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) + if (strscpy(buf, p, ARRAY_SIZE(buf)) < 0) pr_err("filter string is too long: %s\n", p); else if (apply_event_filter(file, buf) < 0) pr_err("Failed to apply filter: %s\n", buf); @@ -494,7 +494,7 @@ trace_boot_init_one_event(struct trace_array *tr, struct xbc_node *gnode, if (IS_ENABLED(CONFIG_HIST_TRIGGERS)) { xbc_node_for_each_array_value(enode, "actions", anode, p) { - if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) + if (strscpy(buf, p, ARRAY_SIZE(buf)) < 0) pr_err("action string is too long: %s\n", p); else if (trigger_process_regex(file, buf) < 0) pr_err("Failed to apply an action: %s\n", p); @@ -633,7 +633,7 @@ trace_boot_init_instances(struct xbc_node *node) if (!p || *p == '\0') continue; - tr = trace_array_get_by_name(p); + tr = trace_array_get_by_name(p, NULL); if (!tr) { pr_err("Failed to get trace instance %s\n", p); continue; diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index e47fdb4c92fb..6809b370e991 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -30,10 +30,8 @@ static struct trace_array *branch_tracer; static void probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) { - struct trace_event_call *call = &event_branch; struct trace_array *tr = branch_tracer; struct trace_buffer *buffer; - struct trace_array_cpu *data; struct ring_buffer_event *event; struct trace_branch *entry; unsigned long flags; @@ -55,8 +53,7 @@ probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) raw_local_irq_save(flags); current->trace_recursion |= TRACE_BRANCH_BIT; - data = this_cpu_ptr(tr->array_buffer.data); - if (atomic_read(&data->disabled)) + if (!tracer_tracing_is_on_cpu(tr, raw_smp_processor_id())) goto out; trace_ctx = tracing_gen_ctx_flags(flags); @@ -74,16 +71,13 @@ probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) p--; p++; - strncpy(entry->func, f->data.func, TRACE_FUNC_SIZE); - strncpy(entry->file, p, TRACE_FILE_SIZE); - entry->func[TRACE_FUNC_SIZE] = 0; - entry->file[TRACE_FILE_SIZE] = 0; + strscpy(entry->func, f->data.func); + strscpy(entry->file, p); entry->constant = f->constant; entry->line = f->data.line; entry->correct = val == expect; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + trace_buffer_unlock_commit_nostack(buffer, event); out: current->trace_recursion &= ~TRACE_BRANCH_BIT; diff --git a/kernel/trace/trace_btf.c b/kernel/trace/trace_btf.c new file mode 100644 index 000000000000..5bbdbcbbde3c --- /dev/null +++ b/kernel/trace/trace_btf.c @@ -0,0 +1,122 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/btf.h> +#include <linux/kernel.h> +#include <linux/slab.h> + +#include "trace_btf.h" + +/* + * Find a function proto type by name, and return the btf_type with its btf + * in *@btf_p. Return NULL if not found. + * Note that caller has to call btf_put(*@btf_p) after using the btf_type. + */ +const struct btf_type *btf_find_func_proto(const char *func_name, struct btf **btf_p) +{ + const struct btf_type *t; + s32 id; + + id = bpf_find_btf_id(func_name, BTF_KIND_FUNC, btf_p); + if (id < 0) + return NULL; + + /* Get BTF_KIND_FUNC type */ + t = btf_type_by_id(*btf_p, id); + if (!t || !btf_type_is_func(t)) + goto err; + + /* The type of BTF_KIND_FUNC is BTF_KIND_FUNC_PROTO */ + t = btf_type_by_id(*btf_p, t->type); + if (!t || !btf_type_is_func_proto(t)) + goto err; + + return t; +err: + btf_put(*btf_p); + return NULL; +} + +/* + * Get function parameter with the number of parameters. + * This can return NULL if the function has no parameters. + * It can return -EINVAL if the @func_proto is not a function proto type. + */ +const struct btf_param *btf_get_func_param(const struct btf_type *func_proto, s32 *nr) +{ + if (!btf_type_is_func_proto(func_proto)) + return ERR_PTR(-EINVAL); + + *nr = btf_type_vlen(func_proto); + if (*nr > 0) + return (const struct btf_param *)(func_proto + 1); + else + return NULL; +} + +#define BTF_ANON_STACK_MAX 16 + +struct btf_anon_stack { + u32 tid; + u32 offset; +}; + +/* + * Find a member of data structure/union by name and return it. + * Return NULL if not found, or -EINVAL if parameter is invalid. + * If the member is an member of anonymous union/structure, the offset + * of that anonymous union/structure is stored into @anon_offset. Caller + * can calculate the correct offset from the root data structure by + * adding anon_offset to the member's offset. + */ +const struct btf_member *btf_find_struct_member(struct btf *btf, + const struct btf_type *type, + const char *member_name, + u32 *anon_offset) +{ + struct btf_anon_stack *anon_stack; + const struct btf_member *member; + u32 tid, cur_offset = 0; + const char *name; + int i, top = 0; + + anon_stack = kcalloc(BTF_ANON_STACK_MAX, sizeof(*anon_stack), GFP_KERNEL); + if (!anon_stack) + return ERR_PTR(-ENOMEM); + +retry: + if (!btf_type_is_struct(type)) { + member = ERR_PTR(-EINVAL); + goto out; + } + + for_each_member(i, type, member) { + if (!member->name_off) { + /* Anonymous union/struct: push it for later use */ + if (btf_type_skip_modifiers(btf, member->type, &tid) && + top < BTF_ANON_STACK_MAX) { + anon_stack[top].tid = tid; + anon_stack[top++].offset = + cur_offset + member->offset; + } + } else { + name = btf_name_by_offset(btf, member->name_off); + if (name && !strcmp(member_name, name)) { + if (anon_offset) + *anon_offset = cur_offset; + goto out; + } + } + } + if (top > 0) { + /* Pop from the anonymous stack and retry */ + tid = anon_stack[--top].tid; + cur_offset = anon_stack[top].offset; + type = btf_type_by_id(btf, tid); + goto retry; + } + member = NULL; + +out: + kfree(anon_stack); + return member; +} + diff --git a/kernel/trace/trace_btf.h b/kernel/trace/trace_btf.h new file mode 100644 index 000000000000..4bc44bc261e6 --- /dev/null +++ b/kernel/trace/trace_btf.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/btf.h> + +const struct btf_type *btf_find_func_proto(const char *func_name, + struct btf **btf_p); +const struct btf_param *btf_get_func_param(const struct btf_type *func_proto, + s32 *nr); +const struct btf_member *btf_find_struct_member(struct btf *btf, + const struct btf_type *type, + const char *member_name, + u32 *anon_offset); diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 4702efb00ff2..4cb2ebc439be 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -154,5 +154,5 @@ static atomic64_t trace_counter; */ u64 notrace trace_clock_counter(void) { - return atomic64_add_return(1, &trace_counter); + return atomic64_inc_return(&trace_counter); } diff --git a/kernel/trace/trace_dynevent.c b/kernel/trace/trace_dynevent.c index 4376887e0d8a..c4dfbc293bae 100644 --- a/kernel/trace/trace_dynevent.c +++ b/kernel/trace/trace_dynevent.c @@ -16,7 +16,7 @@ #include "trace_output.h" /* for trace_event_sem */ #include "trace_dynevent.h" -static DEFINE_MUTEX(dyn_event_ops_mutex); +DEFINE_MUTEX(dyn_event_ops_mutex); static LIST_HEAD(dyn_event_ops_list); bool trace_event_dyn_try_get_ref(struct trace_event_call *dyn_call) @@ -74,24 +74,19 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type struct dyn_event *pos, *n; char *system = NULL, *event, *p; int argc, ret = -ENOENT; - char **argv; + char **argv __free(argv_free) = argv_split(GFP_KERNEL, raw_command, &argc); - argv = argv_split(GFP_KERNEL, raw_command, &argc); if (!argv) return -ENOMEM; if (argv[0][0] == '-') { - if (argv[0][1] != ':') { - ret = -EINVAL; - goto out; - } + if (argv[0][1] != ':') + return -EINVAL; event = &argv[0][2]; } else { event = strchr(argv[0], ':'); - if (!event) { - ret = -EINVAL; - goto out; - } + if (!event) + return -EINVAL; event++; } @@ -101,10 +96,8 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type event = p + 1; *p = '\0'; } - if (!system && event[0] == '\0') { - ret = -EINVAL; - goto out; - } + if (!system && event[0] == '\0') + return -EINVAL; mutex_lock(&event_mutex); for_each_dyn_event_safe(pos, n) { @@ -120,8 +113,20 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type } tracing_reset_all_online_cpus(); mutex_unlock(&event_mutex); -out: - argv_free(argv); + return ret; +} + +/* + * Locked version of event creation. The event creation must be protected by + * dyn_event_ops_mutex because of protecting trace_probe_log. + */ +int dyn_event_create(const char *raw_command, struct dyn_event_operations *type) +{ + int ret; + + mutex_lock(&dyn_event_ops_mutex); + ret = type->create(raw_command); + mutex_unlock(&dyn_event_ops_mutex); return ret; } @@ -139,9 +144,16 @@ static int create_dyn_event(const char *raw_command) if (!ret || ret != -ECANCELED) break; } - mutex_unlock(&dyn_event_ops_mutex); - if (ret == -ECANCELED) + if (ret == -ECANCELED) { + static const char *err_msg[] = {"No matching dynamic event type"}; + + /* Wrong dynamic event. Leave an error message. */ + tracing_log_err(NULL, "dynevent", raw_command, err_msg, + 0, 0); ret = -EINVAL; + } + + mutex_unlock(&dyn_event_ops_mutex); return ret; } @@ -225,6 +237,10 @@ static int dyn_event_open(struct inode *inode, struct file *file) { int ret; + ret = security_locked_down(LOCKDOWN_TRACEFS); + if (ret) + return ret; + ret = tracing_check_open_get_tr(NULL); if (ret) return ret; diff --git a/kernel/trace/trace_dynevent.h b/kernel/trace/trace_dynevent.h index 936477a111d3..beee3f8d7544 100644 --- a/kernel/trace/trace_dynevent.h +++ b/kernel/trace/trace_dynevent.h @@ -100,6 +100,7 @@ void *dyn_event_seq_next(struct seq_file *m, void *v, loff_t *pos); void dyn_event_seq_stop(struct seq_file *m, void *v); int dyn_events_release_all(struct dyn_event_operations *type); int dyn_event_release(const char *raw_command, struct dyn_event_operations *type); +int dyn_event_create(const char *raw_command, struct dyn_event_operations *type); /* * for_each_dyn_event - iterate over the dyn_event list diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h index cd41e863b51c..f6a8d29c0d76 100644 --- a/kernel/trace/trace_entries.h +++ b/kernel/trace/trace_entries.h @@ -61,8 +61,9 @@ FTRACE_ENTRY_REG(function, ftrace_entry, TRACE_FN, F_STRUCT( - __field_fn( unsigned long, ip ) - __field_fn( unsigned long, parent_ip ) + __field_fn( unsigned long, ip ) + __field_fn( unsigned long, parent_ip ) + __dynamic_array( unsigned long, args ) ), F_printk(" %ps <-- %ps", @@ -72,19 +73,49 @@ FTRACE_ENTRY_REG(function, ftrace_entry, ); /* Function call entry */ -FTRACE_ENTRY_PACKED(funcgraph_entry, ftrace_graph_ent_entry, +FTRACE_ENTRY(funcgraph_entry, ftrace_graph_ent_entry, TRACE_GRAPH_ENT, F_STRUCT( __field_struct( struct ftrace_graph_ent, graph_ent ) __field_packed( unsigned long, graph_ent, func ) - __field_packed( int, graph_ent, depth ) + __field_packed( unsigned long, graph_ent, depth ) + __dynamic_array(unsigned long, args ) ), - F_printk("--> %ps (%d)", (void *)__entry->func, __entry->depth) + F_printk("--> %ps (%lu)", (void *)__entry->func, __entry->depth) ); +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR + +/* Function call entry with a return address */ +FTRACE_ENTRY_PACKED(fgraph_retaddr_entry, fgraph_retaddr_ent_entry, + + TRACE_GRAPH_RETADDR_ENT, + + F_STRUCT( + __field_struct( struct fgraph_retaddr_ent, graph_rent ) + __field_packed( unsigned long, graph_rent.ent, func ) + __field_packed( unsigned long, graph_rent.ent, depth ) + __field_packed( unsigned long, graph_rent, retaddr ) + __dynamic_array(unsigned long, args ) + ), + + F_printk("--> %ps (%lu) <- %ps", (void *)__entry->func, __entry->depth, + (void *)__entry->retaddr) +); + +#else + +#ifndef fgraph_retaddr_ent_entry +#define fgraph_retaddr_ent_entry ftrace_graph_ent_entry +#endif + +#endif + +#ifdef CONFIG_FUNCTION_GRAPH_RETVAL + /* Function return entry */ FTRACE_ENTRY_PACKED(funcgraph_exit, ftrace_graph_ret_entry, @@ -93,18 +124,43 @@ FTRACE_ENTRY_PACKED(funcgraph_exit, ftrace_graph_ret_entry, F_STRUCT( __field_struct( struct ftrace_graph_ret, ret ) __field_packed( unsigned long, ret, func ) - __field_packed( int, ret, depth ) + __field_packed( unsigned long, ret, retval ) + __field_packed( unsigned int, ret, depth ) __field_packed( unsigned int, ret, overrun ) - __field_packed( unsigned long long, ret, calltime) - __field_packed( unsigned long long, ret, rettime ) + __field(unsigned long long, calltime ) + __field(unsigned long long, rettime ) ), - F_printk("<-- %ps (%d) (start: %llx end: %llx) over: %d", + F_printk("<-- %ps (%u) (start: %llx end: %llx) over: %u retval: %lx", + (void *)__entry->func, __entry->depth, + __entry->calltime, __entry->rettime, + __entry->depth, __entry->retval) +); + +#else + +/* Function return entry */ +FTRACE_ENTRY_PACKED(funcgraph_exit, ftrace_graph_ret_entry, + + TRACE_GRAPH_RET, + + F_STRUCT( + __field_struct( struct ftrace_graph_ret, ret ) + __field_packed( unsigned long, ret, func ) + __field_packed( unsigned int, ret, depth ) + __field_packed( unsigned int, ret, overrun ) + __field(unsigned long long, calltime ) + __field(unsigned long long, rettime ) + ), + + F_printk("<-- %ps (%u) (start: %llx end: %llx) over: %u", (void *)__entry->func, __entry->depth, __entry->calltime, __entry->rettime, __entry->depth) ); +#endif + /* * Context switch trace entry - which task (and prio) we switched from/to: * @@ -164,7 +220,7 @@ FTRACE_ENTRY(kernel_stack, stack_entry, F_STRUCT( __field( int, size ) - __array( unsigned long, caller, FTRACE_STACK_ENTRIES ) + __stack_array( unsigned long, caller, FTRACE_STACK_ENTRIES, size) ), F_printk("\t=> %ps\n\t=> %ps\n\t=> %ps\n" diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c index 352b65e2b910..3ee39715d5e4 100644 --- a/kernel/trace/trace_eprobe.c +++ b/kernel/trace/trace_eprobe.c @@ -9,14 +9,15 @@ * Copyright (C) 2021, VMware Inc, Tzvetomir Stoyanov tz.stoyanov@gmail.com> * */ +#include <linux/cleanup.h> +#include <linux/ftrace.h> #include <linux/module.h> #include <linux/mutex.h> -#include <linux/ftrace.h> #include "trace_dynevent.h" #include "trace_probe.h" -#include "trace_probe_tmpl.h" #include "trace_probe_kernel.h" +#include "trace_probe_tmpl.h" #define EPROBE_EVENT_SYSTEM "eprobes" @@ -41,6 +42,10 @@ struct eprobe_data { struct trace_eprobe *ep; }; + +#define for_each_trace_eprobe_tp(ep, _tp) \ + list_for_each_entry(ep, trace_probe_probe_list(_tp), tp.list) + static int __trace_eprobe_create(int argc, const char *argv[]); static void trace_event_probe_cleanup(struct trace_eprobe *ep) @@ -56,6 +61,9 @@ static void trace_event_probe_cleanup(struct trace_eprobe *ep) kfree(ep); } +DEFINE_FREE(trace_event_probe_cleanup, struct trace_eprobe *, + if (!IS_ERR_OR_NULL(_T)) trace_event_probe_cleanup(_T)) + static struct trace_eprobe *to_trace_eprobe(struct dyn_event *ev) { return container_of(ev, struct trace_eprobe, devent); @@ -192,10 +200,10 @@ static struct trace_eprobe *alloc_event_probe(const char *group, struct trace_event_call *event, int nargs) { - struct trace_eprobe *ep; + struct trace_eprobe *ep __free(trace_event_probe_cleanup) = NULL; const char *event_name; const char *sys_name; - int ret = -ENOMEM; + int ret; if (!event) return ERR_PTR(-ENODEV); @@ -206,56 +214,22 @@ static struct trace_eprobe *alloc_event_probe(const char *group, ep = kzalloc(struct_size(ep, tp.args, nargs), GFP_KERNEL); if (!ep) { trace_event_put_ref(event); - goto error; + return ERR_PTR(-ENOMEM); } ep->event = event; ep->event_name = kstrdup(event_name, GFP_KERNEL); if (!ep->event_name) - goto error; + return ERR_PTR(-ENOMEM); ep->event_system = kstrdup(sys_name, GFP_KERNEL); if (!ep->event_system) - goto error; + return ERR_PTR(-ENOMEM); - ret = trace_probe_init(&ep->tp, this_event, group, false); + ret = trace_probe_init(&ep->tp, this_event, group, false, nargs); if (ret < 0) - goto error; + return ERR_PTR(ret); dyn_event_init(&ep->devent, &eprobe_dyn_event_ops); - return ep; -error: - trace_event_probe_cleanup(ep); - return ERR_PTR(ret); -} - -static int trace_eprobe_tp_arg_update(struct trace_eprobe *ep, int i) -{ - struct probe_arg *parg = &ep->tp.args[i]; - struct ftrace_event_field *field; - struct list_head *head; - int ret = -ENOENT; - - head = trace_get_fields(ep->event); - list_for_each_entry(field, head, link) { - if (!strcmp(parg->code->data, field->name)) { - kfree(parg->code->data); - parg->code->data = field; - return 0; - } - } - - /* - * Argument not found on event. But allow for comm and COMM - * to be used to get the current->comm. - */ - if (strcmp(parg->code->data, "COMM") == 0 || - strcmp(parg->code->data, "comm") == 0) { - parg->code->op = FETCH_OP_COMM; - ret = 0; - } - - kfree(parg->code->data); - parg->code->data = NULL; - return ret; + return_ptr(ep); } static int eprobe_event_define_fields(struct trace_event_call *event_call) @@ -311,7 +285,7 @@ print_eprobe_event(struct trace_iterator *iter, int flags, trace_seq_putc(s, ')'); - if (print_probe_args(s, tp->args, tp->nr_args, + if (trace_probe_print_args(s, tp->args, tp->nr_args, (u8 *)&field[1], field) < 0) goto out; @@ -320,7 +294,8 @@ print_eprobe_event(struct trace_iterator *iter, int flags, return trace_handle_return(s); } -static unsigned long get_event_field(struct fetch_insn *code, void *rec) +static nokprobe_inline unsigned long +get_event_field(struct fetch_insn *code, void *rec) { struct ftrace_event_field *field = code->data; unsigned long val; @@ -369,10 +344,15 @@ static unsigned long get_event_field(struct fetch_insn *code, void *rec) val = *(unsigned int *)addr; break; default: - if (field->is_signed) - val = *(long *)addr; - else - val = *(unsigned long *)addr; + if (field->size == sizeof(long)) { + if (field->is_signed) + val = *(long *)addr; + else + val = *(unsigned long *)addr; + break; + } + /* This is an array, point to the addr itself */ + val = (unsigned long)addr; break; } return val; @@ -395,20 +375,12 @@ static int get_eprobe_size(struct trace_probe *tp, void *rec) case FETCH_OP_TP_ARG: val = get_event_field(code, rec); break; - case FETCH_OP_IMM: - val = code->immediate; - break; - case FETCH_OP_COMM: - val = (unsigned long)current->comm; - break; - case FETCH_OP_DATA: - val = (unsigned long)code->data; - break; case FETCH_NOP_SYMBOL: /* Ignore a place holder */ code++; goto retry; default: - continue; + if (process_common_fetch_insn(code, &val) < 0) + continue; } code++; len = process_fetch_insn_bottom(code, val, NULL, NULL); @@ -424,88 +396,30 @@ static int get_eprobe_size(struct trace_probe *tp, void *rec) /* Note that we don't verify it, since the code does not come from user space */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { unsigned long val; + int ret; retry: switch (code->op) { case FETCH_OP_TP_ARG: val = get_event_field(code, rec); break; - case FETCH_OP_IMM: - val = code->immediate; - break; - case FETCH_OP_COMM: - val = (unsigned long)current->comm; - break; - case FETCH_OP_DATA: - val = (unsigned long)code->data; - break; case FETCH_NOP_SYMBOL: /* Ignore a place holder */ code++; goto retry; default: - return -EILSEQ; + ret = process_common_fetch_insn(code, &val); + if (ret < 0) + return ret; } code++; return process_fetch_insn_bottom(code, val, dest, base); } NOKPROBE_SYMBOL(process_fetch_insn) -/* Return the length of string -- including null terminal byte */ -static nokprobe_inline int -fetch_store_strlen_user(unsigned long addr) -{ - return kern_fetch_store_strlen_user(addr); -} - -/* Return the length of string -- including null terminal byte */ -static nokprobe_inline int -fetch_store_strlen(unsigned long addr) -{ - return kern_fetch_store_strlen(addr); -} - -/* - * Fetch a null-terminated string from user. Caller MUST set *(u32 *)buf - * with max length and relative data location. - */ -static nokprobe_inline int -fetch_store_string_user(unsigned long addr, void *dest, void *base) -{ - return kern_fetch_store_string_user(addr, dest, base); -} - -/* - * Fetch a null-terminated string. Caller MUST set *(u32 *)buf with max - * length and relative data location. - */ -static nokprobe_inline int -fetch_store_string(unsigned long addr, void *dest, void *base) -{ - return kern_fetch_store_string(addr, dest, base); -} - -static nokprobe_inline int -probe_mem_read_user(void *dest, void *src, size_t size) -{ - const void __user *uaddr = (__force const void __user *)src; - - return copy_from_user_nofault(dest, uaddr, size); -} - -static nokprobe_inline int -probe_mem_read(void *dest, void *src, size_t size) -{ -#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE - if ((unsigned long)src < TASK_SIZE) - return probe_mem_read_user(dest, src, size); -#endif - return copy_from_kernel_nofault(dest, src, size); -} - /* eprobe handler */ static inline void __eprobe_trace_func(struct eprobe_data *edata, void *rec) @@ -530,7 +444,7 @@ __eprobe_trace_func(struct eprobe_data *edata, void *rec) return; entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); - store_trace_args(&entry[1], &edata->ep->tp, rec, sizeof(*entry), dsize); + store_trace_args(&entry[1], &edata->ep->tp, rec, NULL, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } @@ -570,13 +484,6 @@ static void eprobe_trigger_func(struct event_trigger_data *data, __eprobe_trace_func(edata, rec); } -static struct event_trigger_ops eprobe_trigger_ops = { - .trigger = eprobe_trigger_func, - .print = eprobe_trigger_print, - .init = eprobe_trigger_init, - .free = eprobe_trigger_free, -}; - static int eprobe_trigger_cmd_parse(struct event_command *cmd_ops, struct trace_event_file *file, char *glob, char *cmd, @@ -599,12 +506,6 @@ static void eprobe_trigger_unreg_func(char *glob, } -static struct event_trigger_ops *eprobe_trigger_get_ops(char *cmd, - char *param) -{ - return &eprobe_trigger_ops; -} - static struct event_command event_trigger_cmd = { .name = "eprobe", .trigger_type = ETT_EVENT_EPROBE, @@ -613,8 +514,11 @@ static struct event_command event_trigger_cmd = { .reg = eprobe_trigger_reg_func, .unreg = eprobe_trigger_unreg_func, .unreg_all = NULL, - .get_trigger_ops = eprobe_trigger_get_ops, .set_filter = NULL, + .trigger = eprobe_trigger_func, + .print = eprobe_trigger_print, + .init = eprobe_trigger_init, + .free = eprobe_trigger_free, }; static struct event_trigger_data * @@ -634,7 +538,6 @@ new_eprobe_trigger(struct trace_eprobe *ep, struct trace_event_file *file) trigger->flags = EVENT_TRIGGER_FL_PROBE; trigger->count = -1; - trigger->ops = &eprobe_trigger_ops; /* * EVENT PROBE triggers are not registered as commands with @@ -736,10 +639,11 @@ static int disable_eprobe(struct trace_eprobe *ep, static int enable_trace_eprobe(struct trace_event_call *call, struct trace_event_file *file) { - struct trace_probe *pos, *tp; + struct trace_probe *tp; struct trace_eprobe *ep; bool enabled; int ret = 0; + int cnt = 0; tp = trace_probe_primary_from_call(call); if (WARN_ON_ONCE(!tp)) @@ -757,18 +661,29 @@ static int enable_trace_eprobe(struct trace_event_call *call, if (enabled) return 0; - list_for_each_entry(pos, trace_probe_probe_list(tp), list) { - ep = container_of(pos, struct trace_eprobe, tp); + for_each_trace_eprobe_tp(ep, tp) { ret = enable_eprobe(ep, file); if (ret) break; enabled = true; + cnt++; } if (ret) { /* Failed to enable one of them. Roll back all */ - if (enabled) - disable_eprobe(ep, file->tr); + if (enabled) { + /* + * It's a bug if one failed for something other than memory + * not being available but another eprobe succeeded. + */ + WARN_ON_ONCE(ret != -ENOMEM); + + for_each_trace_eprobe_tp(ep, tp) { + disable_eprobe(ep, file->tr); + if (!--cnt) + break; + } + } if (file) trace_probe_remove_file(tp, file); else @@ -781,7 +696,7 @@ static int enable_trace_eprobe(struct trace_event_call *call, static int disable_trace_eprobe(struct trace_event_call *call, struct trace_event_file *file) { - struct trace_probe *pos, *tp; + struct trace_probe *tp; struct trace_eprobe *ep; tp = trace_probe_primary_from_call(call); @@ -798,10 +713,8 @@ static int disable_trace_eprobe(struct trace_event_call *call, trace_probe_clear_flag(tp, TP_FLAG_PROFILE); if (!trace_probe_is_enabled(tp)) { - list_for_each_entry(pos, trace_probe_probe_list(tp), list) { - ep = container_of(pos, struct trace_eprobe, tp); + for_each_trace_eprobe_tp(ep, tp) disable_eprobe(ep, file->tr); - } } out: @@ -870,38 +783,13 @@ find_and_get_event(const char *system, const char *event_name) name = trace_event_name(tp_event); if (!name || strcmp(event_name, name)) continue; - if (!trace_event_try_get_ref(tp_event)) { + if (!trace_event_try_get_ref(tp_event)) return NULL; - break; - } return tp_event; - break; } return NULL; } -static int trace_eprobe_tp_update_arg(struct trace_eprobe *ep, const char *argv[], int i) -{ - unsigned int flags = TPARG_FL_KERNEL | TPARG_FL_TPOINT; - int ret; - - ret = traceprobe_parse_probe_arg(&ep->tp, i, argv[i], flags); - if (ret) - return ret; - - if (ep->tp.args[i].code->op == FETCH_OP_TP_ARG) { - ret = trace_eprobe_tp_arg_update(ep, i); - if (ret) - trace_probe_log_err(0, BAD_ATTACH_ARG); - } - - /* Handle symbols "@" */ - if (!ret) - ret = traceprobe_update_arg(&ep->tp.args[i]); - - return ret; -} - static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const char *argv[]) { struct event_filter *dummy = NULL; @@ -923,17 +811,13 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const ch p = ep->filter_str; for (i = 0; i < argc; i++) { - ret = snprintf(p, len, "%s ", argv[i]); - if (ret < 0) - goto error; - if (ret > len) { - ret = -E2BIG; - goto error; - } + if (i) + ret = snprintf(p, len, " %s", argv[i]); + else + ret = snprintf(p, len, "%s", argv[i]); p += ret; len -= ret; } - p[-1] = '\0'; /* * Ensure the filter string can be parsed correctly. Note, this @@ -942,13 +826,10 @@ static int trace_eprobe_parse_filter(struct trace_eprobe *ep, int argc, const ch ret = create_event_filter(top_trace_array(), ep->event, ep->filter_str, true, &dummy); free_event_filter(dummy); - if (ret) - goto error; - - return 0; -error: - kfree(ep->filter_str); - ep->filter_str = NULL; + if (ret) { + kfree(ep->filter_str); + ep->filter_str = NULL; + } return ret; } @@ -960,40 +841,52 @@ static int __trace_eprobe_create(int argc, const char *argv[]) * Fetch args (no space): * <name>=$<field>[:TYPE] */ + struct traceprobe_parse_context *ctx __free(traceprobe_parse_context) = NULL; + struct trace_eprobe *ep __free(trace_event_probe_cleanup) = NULL; + const char *trlog __free(trace_probe_log_clear) = NULL; const char *event = NULL, *group = EPROBE_EVENT_SYSTEM; const char *sys_event = NULL, *sys_name = NULL; struct trace_event_call *event_call; - struct trace_eprobe *ep = NULL; - char buf1[MAX_EVENT_NAME_LEN]; - char buf2[MAX_EVENT_NAME_LEN]; - char gbuf[MAX_EVENT_NAME_LEN]; + char *buf1 __free(kfree) = NULL; + char *buf2 __free(kfree) = NULL; + char *gbuf __free(kfree) = NULL; int ret = 0, filter_idx = 0; int i, filter_cnt; if (argc < 2 || argv[0][0] != 'e') return -ECANCELED; - trace_probe_log_init("event_probe", argc, argv); + trlog = trace_probe_log_init("event_probe", argc, argv); event = strchr(&argv[0][1], ':'); if (event) { + gbuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!gbuf) + return -ENOMEM; event++; ret = traceprobe_parse_event_name(&event, &group, gbuf, event - argv[0]); if (ret) - goto parse_error; + return -EINVAL; } trace_probe_log_set_index(1); sys_event = argv[1]; + + buf2 = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!buf2) + return -ENOMEM; + ret = traceprobe_parse_event_name(&sys_event, &sys_name, buf2, 0); if (ret || !sys_event || !sys_name) { trace_probe_log_err(0, NO_EVENT_INFO); - goto parse_error; + return -EINVAL; } if (!event) { - strscpy(buf1, sys_event, MAX_EVENT_NAME_LEN); + buf1 = kstrdup(sys_event, GFP_KERNEL); + if (!buf1) + return -ENOMEM; event = buf1; } @@ -1006,10 +899,16 @@ static int __trace_eprobe_create(int argc, const char *argv[]) } } - mutex_lock(&event_mutex); - event_call = find_and_get_event(sys_name, sys_event); - ep = alloc_event_probe(group, event, event_call, argc - 2); - mutex_unlock(&event_mutex); + if (argc - 2 > MAX_TRACE_ARGS) { + trace_probe_log_set_index(2); + trace_probe_log_err(0, TOO_MANY_ARGS); + return -E2BIG; + } + + scoped_guard(mutex, &event_mutex) { + event_call = find_and_get_event(sys_name, sys_event); + ep = alloc_event_probe(group, event, event_call, argc - 2); + } if (IS_ERR(ep)) { ret = PTR_ERR(ep); @@ -1017,47 +916,57 @@ static int __trace_eprobe_create(int argc, const char *argv[]) trace_probe_log_err(0, BAD_ATTACH_EVENT); /* This must return -ENOMEM or missing event, else there is a bug */ WARN_ON_ONCE(ret != -ENOMEM && ret != -ENODEV); - ep = NULL; - goto error; + return ret; } if (filter_idx) { trace_probe_log_set_index(filter_idx); ret = trace_eprobe_parse_filter(ep, filter_cnt, argv + filter_idx); if (ret) - goto parse_error; + return -EINVAL; } else ep->filter_str = NULL; + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + ctx->event = ep->event; + ctx->flags = TPARG_FL_KERNEL | TPARG_FL_TEVENT; + argc -= 2; argv += 2; /* parse arguments */ - for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + for (i = 0; i < argc; i++) { trace_probe_log_set_index(i + 2); - ret = trace_eprobe_tp_update_arg(ep, argv, i); + + ret = traceprobe_parse_probe_arg(&ep->tp, i, argv[i], ctx); + /* Handle symbols "@" */ + if (!ret) + ret = traceprobe_update_arg(&ep->tp.args[i]); if (ret) - goto error; + return ret; } ret = traceprobe_set_print_fmt(&ep->tp, PROBE_PRINT_EVENT); if (ret < 0) - goto error; + return ret; + init_trace_eprobe_call(ep); - mutex_lock(&event_mutex); - ret = trace_probe_register_event_call(&ep->tp); - if (ret) { - if (ret == -EEXIST) { - trace_probe_log_set_index(0); - trace_probe_log_err(0, EVENT_EXIST); + scoped_guard(mutex, &event_mutex) { + ret = trace_probe_register_event_call(&ep->tp); + if (ret) { + if (ret == -EEXIST) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, EVENT_EXIST); + } + return ret; } - mutex_unlock(&event_mutex); - goto error; + ret = dyn_event_add(&ep->devent, &ep->tp.event->call); + if (ret < 0) { + trace_probe_unregister_event_call(&ep->tp); + return ret; + } + /* To avoid freeing registered eprobe event, clear ep. */ + ep = NULL; } - ret = dyn_event_add(&ep->devent, &ep->tp.event->call); - mutex_unlock(&event_mutex); - return ret; -parse_error: - ret = -EINVAL; -error: - trace_event_probe_cleanup(ep); return ret; } diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 05e791241812..a6bb7577e8c5 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -49,7 +49,7 @@ static int perf_trace_event_perm(struct trace_event_call *tp_event, /* The ftrace function trace is allowed only for root. */ if (ftrace_event_is_function(tp_event)) { - ret = perf_allow_tracepoint(&p_event->attr); + ret = perf_allow_tracepoint(); if (ret) return ret; @@ -86,7 +86,7 @@ static int perf_trace_event_perm(struct trace_event_call *tp_event, * ...otherwise raw tracepoint data can be a severe data leak, * only allow root to have these. */ - ret = perf_allow_tracepoint(&p_event->attr); + ret = perf_allow_tracepoint(); if (ret) return ret; @@ -352,10 +352,16 @@ void perf_uprobe_destroy(struct perf_event *p_event) int perf_trace_add(struct perf_event *p_event, int flags) { struct trace_event_call *tp_event = p_event->tp_event; + struct hw_perf_event *hwc = &p_event->hw; if (!(flags & PERF_EF_START)) p_event->hw.state = PERF_HES_STOPPED; + if (is_sampling_event(p_event)) { + hwc->last_period = hwc->sample_period; + perf_swevent_set_period(p_event); + } + /* * If TRACE_REG_PERF_ADD returns false; no custom action was performed * and we need to take the default action of enqueueing our event on diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 33e0b4f8ebe6..b16a5a158040 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -82,7 +82,7 @@ static int system_refcount_dec(struct event_subsystem *system) } static struct ftrace_event_field * -__find_event_field(struct list_head *head, char *name) +__find_event_field(struct list_head *head, const char *name) { struct ftrace_event_field *field; @@ -114,7 +114,8 @@ trace_find_event_field(struct trace_event_call *call, char *name) static int __trace_define_field(struct list_head *head, const char *type, const char *name, int offset, int size, - int is_signed, int filter_type) + int is_signed, int filter_type, int len, + int need_test) { struct ftrace_event_field *field; @@ -133,6 +134,8 @@ static int __trace_define_field(struct list_head *head, const char *type, field->offset = offset; field->size = size; field->is_signed = is_signed; + field->needs_test = need_test; + field->len = len; list_add(&field->link, head); @@ -150,14 +153,28 @@ int trace_define_field(struct trace_event_call *call, const char *type, head = trace_get_fields(call); return __trace_define_field(head, type, name, offset, size, - is_signed, filter_type); + is_signed, filter_type, 0, 0); } EXPORT_SYMBOL_GPL(trace_define_field); +static int trace_define_field_ext(struct trace_event_call *call, const char *type, + const char *name, int offset, int size, int is_signed, + int filter_type, int len, int need_test) +{ + struct list_head *head; + + if (WARN_ON(!call->class)) + return 0; + + head = trace_get_fields(call); + return __trace_define_field(head, type, name, offset, size, + is_signed, filter_type, len, need_test); +} + #define __generic_field(type, item, filter_type) \ ret = __trace_define_field(&ftrace_generic_fields, #type, \ #item, 0, 0, is_signed_type(type), \ - filter_type); \ + filter_type, 0, 0); \ if (ret) \ return ret; @@ -166,7 +183,8 @@ EXPORT_SYMBOL_GPL(trace_define_field); "common_" #item, \ offsetof(typeof(ent), item), \ sizeof(ent.item), \ - is_signed_type(type), FILTER_OTHER); \ + is_signed_type(type), FILTER_OTHER, \ + 0, 0); \ if (ret) \ return ret; @@ -179,6 +197,8 @@ static int trace_define_generic_fields(void) __generic_field(int, common_cpu, FILTER_CPU); __generic_field(char *, COMM, FILTER_COMM); __generic_field(char *, comm, FILTER_COMM); + __generic_field(char *, stacktrace, FILTER_STACKTRACE); + __generic_field(char *, STACKTRACE, FILTER_STACKTRACE); return ret; } @@ -227,19 +247,16 @@ int trace_event_get_offsets(struct trace_event_call *call) return tail->offset + tail->size; } -/* - * Check if the referenced field is an array and return true, - * as arrays are OK to dereference. - */ -static bool test_field(const char *fmt, struct trace_event_call *call) + +static struct trace_event_fields *find_event_field(const char *fmt, + struct trace_event_call *call) { struct trace_event_fields *field = call->class->fields_array; - const char *array_descriptor; const char *p = fmt; int len; if (!(len = str_has_prefix(fmt, "REC->"))) - return false; + return NULL; fmt += len; for (p = fmt; *p; p++) { if (!isalnum(*p) && *p != '_') @@ -248,16 +265,155 @@ static bool test_field(const char *fmt, struct trace_event_call *call) len = p - fmt; for (; field->type; field++) { - if (strncmp(field->name, fmt, len) || - field->name[len]) + if (strncmp(field->name, fmt, len) || field->name[len]) continue; - array_descriptor = strchr(field->type, '['); - /* This is an array and is OK to dereference. */ - return array_descriptor != NULL; + + return field; + } + return NULL; +} + +/* + * Check if the referenced field is an array and return true, + * as arrays are OK to dereference. + */ +static bool test_field(const char *fmt, struct trace_event_call *call) +{ + struct trace_event_fields *field; + + field = find_event_field(fmt, call); + if (!field) + return false; + + /* This is an array and is OK to dereference. */ + return strchr(field->type, '[') != NULL; +} + +/* Look for a string within an argument */ +static bool find_print_string(const char *arg, const char *str, const char *end) +{ + const char *r; + + r = strstr(arg, str); + return r && r < end; +} + +/* Return true if the argument pointer is safe */ +static bool process_pointer(const char *fmt, int len, struct trace_event_call *call) +{ + const char *r, *e, *a; + + e = fmt + len; + + /* Find the REC-> in the argument */ + r = strstr(fmt, "REC->"); + if (r && r < e) { + /* + * Addresses of events on the buffer, or an array on the buffer is + * OK to dereference. There's ways to fool this, but + * this is to catch common mistakes, not malicious code. + */ + a = strchr(fmt, '&'); + if ((a && (a < r)) || test_field(r, call)) + return true; + } else if (find_print_string(fmt, "__get_dynamic_array(", e)) { + return true; + } else if (find_print_string(fmt, "__get_rel_dynamic_array(", e)) { + return true; + } else if (find_print_string(fmt, "__get_dynamic_array_len(", e)) { + return true; + } else if (find_print_string(fmt, "__get_rel_dynamic_array_len(", e)) { + return true; + } else if (find_print_string(fmt, "__get_sockaddr(", e)) { + return true; + } else if (find_print_string(fmt, "__get_rel_sockaddr(", e)) { + return true; } return false; } +/* Return true if the string is safe */ +static bool process_string(const char *fmt, int len, struct trace_event_call *call) +{ + struct trace_event_fields *field; + const char *r, *e, *s; + + e = fmt + len; + + /* + * There are several helper functions that return strings. + * If the argument contains a function, then assume its field is valid. + * It is considered that the argument has a function if it has: + * alphanumeric or '_' before a parenthesis. + */ + s = fmt; + do { + r = strstr(s, "("); + if (!r || r >= e) + break; + for (int i = 1; r - i >= s; i++) { + char ch = *(r - i); + if (isspace(ch)) + continue; + if (isalnum(ch) || ch == '_') + return true; + /* Anything else, this isn't a function */ + break; + } + /* A function could be wrapped in parenthesis, try the next one */ + s = r + 1; + } while (s < e); + + /* + * Check for arrays. If the argument has: foo[REC->val] + * then it is very likely that foo is an array of strings + * that are safe to use. + */ + r = strstr(s, "["); + if (r && r < e) { + r = strstr(r, "REC->"); + if (r && r < e) + return true; + } + + /* + * If there's any strings in the argument consider this arg OK as it + * could be: REC->field ? "foo" : "bar" and we don't want to get into + * verifying that logic here. + */ + if (find_print_string(fmt, "\"", e)) + return true; + + /* Dereferenced strings are also valid like any other pointer */ + if (process_pointer(fmt, len, call)) + return true; + + /* Make sure the field is found */ + field = find_event_field(fmt, call); + if (!field) + return false; + + /* Test this field's string before printing the event */ + call->flags |= TRACE_EVENT_FL_TEST_STR; + field->needs_test = 1; + + return true; +} + +static void handle_dereference_arg(const char *arg_str, u64 string_flags, int len, + u64 *dereference_flags, int arg, + struct trace_event_call *call) +{ + if (string_flags & (1ULL << arg)) { + if (process_string(arg_str, len, call)) + *dereference_flags &= ~(1ULL << arg); + } else if (process_pointer(arg_str, len, call)) + *dereference_flags &= ~(1ULL << arg); + else + pr_warn("TRACE EVENT ERROR: Bad dereference argument: '%.*s'\n", + len, arg_str); +} + /* * Examine the print fmt of the event looking for unsafe dereference * pointers using %p* that could be recorded in the trace event and @@ -267,13 +423,14 @@ static bool test_field(const char *fmt, struct trace_event_call *call) static void test_event_printk(struct trace_event_call *call) { u64 dereference_flags = 0; + u64 string_flags = 0; bool first = true; - const char *fmt, *c, *r, *a; + const char *fmt; int parens = 0; char in_quote = 0; int start_arg = 0; int arg = 0; - int i; + int i, e; fmt = call->print_fmt; @@ -327,6 +484,7 @@ static void test_event_printk(struct trace_event_call *call) case '%': continue; case 'p': + do_pointer: /* Find dereferencing fields */ switch (fmt[i + 1]) { case 'B': case 'R': case 'r': @@ -355,10 +513,24 @@ static void test_event_printk(struct trace_event_call *call) continue; if (fmt[i + j] == '*') { star = true; + /* Handle %*pbl case */ + if (!j && fmt[i + 1] == 'p') { + arg++; + i++; + goto do_pointer; + } continue; } - if ((fmt[i + j] == 's') && star) - arg++; + if ((fmt[i + j] == 's')) { + if (star) + arg++; + if (WARN_ONCE(arg == 63, + "Too many args for event: %s", + trace_event_name(call))) + return; + dereference_flags |= 1ULL << arg; + string_flags |= 1ULL << arg; + } break; } break; @@ -386,42 +558,43 @@ static void test_event_printk(struct trace_event_call *call) case ',': if (in_quote || parens) continue; + e = i; i++; while (isspace(fmt[i])) i++; - start_arg = i; - if (!(dereference_flags & (1ULL << arg))) - goto next_arg; - /* Find the REC-> in the argument */ - c = strchr(fmt + i, ','); - r = strstr(fmt + i, "REC->"); - if (r && (!c || r < c)) { - /* - * Addresses of events on the buffer, - * or an array on the buffer is - * OK to dereference. - * There's ways to fool this, but - * this is to catch common mistakes, - * not malicious code. - */ - a = strchr(fmt + i, '&'); - if ((a && (a < r)) || test_field(r, call)) - dereference_flags &= ~(1ULL << arg); - } else if ((r = strstr(fmt + i, "__get_dynamic_array(")) && - (!c || r < c)) { - dereference_flags &= ~(1ULL << arg); - } else if ((r = strstr(fmt + i, "__get_sockaddr(")) && - (!c || r < c)) { - dereference_flags &= ~(1ULL << arg); + /* + * If start_arg is zero, then this is the start of the + * first argument. The processing of the argument happens + * when the end of the argument is found, as it needs to + * handle parenthesis and such. + */ + if (!start_arg) { + start_arg = i; + /* Balance out the i++ in the for loop */ + i--; + continue; } - next_arg: - i--; + if (dereference_flags & (1ULL << arg)) { + handle_dereference_arg(fmt + start_arg, string_flags, + e - start_arg, + &dereference_flags, arg, call); + } + + start_arg = i; arg++; + /* Balance out the i++ in the for loop */ + i--; } } + if (dereference_flags & (1ULL << arg)) { + handle_dereference_arg(fmt + start_arg, string_flags, + i - start_arg, + &dereference_flags, arg, call); + } + /* * If you triggered the below warning, the trace event reported * uses an unsafe dereference pointer %p*. As the data stored @@ -459,7 +632,6 @@ EXPORT_SYMBOL_GPL(trace_event_raw_init); bool trace_event_ignore_this_pid(struct trace_event_file *trace_file) { struct trace_array *tr = trace_file->tr; - struct trace_array_cpu *data; struct trace_pid_list *no_pid_list; struct trace_pid_list *pid_list; @@ -469,9 +641,11 @@ bool trace_event_ignore_this_pid(struct trace_event_file *trace_file) if (!pid_list && !no_pid_list) return false; - data = this_cpu_ptr(tr->array_buffer.data); - - return data->ignore_pid; + /* + * This is recorded at every sched_switch for this task. + * Thus, even if the task migrates the ignore value will be the same. + */ + return this_cpu_read(tr->array_buffer.data->ignore_pid) != 0; } EXPORT_SYMBOL_GPL(trace_event_ignore_this_pid); @@ -594,7 +768,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, { struct trace_event_call *call = file->event_call; struct trace_array *tr = file->tr; - unsigned long file_flags = file->flags; + bool soft_mode = atomic_read(&file->sm_ref) != 0; int ret = 0; int disable; @@ -609,17 +783,19 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, * is set we do not want the event to be enabled before we * clear the bit. * - * When soft_disable is not set but the SOFT_MODE flag is, + * When soft_disable is not set but the soft_mode is, * we do nothing. Do not disable the tracepoint, otherwise - * "soft enable"s (clearing the SOFT_DISABLED bit) wont work. + * "soft enable"s (clearing the SOFT_DISABLED bit) won't work. */ if (soft_disable) { if (atomic_dec_return(&file->sm_ref) > 0) break; disable = file->flags & EVENT_FILE_FL_SOFT_DISABLED; - clear_bit(EVENT_FILE_FL_SOFT_MODE_BIT, &file->flags); + soft_mode = false; + /* Disable use of trace_buffered_event */ + trace_buffered_event_disable(); } else - disable = !(file->flags & EVENT_FILE_FL_SOFT_MODE); + disable = !soft_mode; if (disable && (file->flags & EVENT_FILE_FL_ENABLED)) { clear_bit(EVENT_FILE_FL_ENABLED_BIT, &file->flags); @@ -633,10 +809,12 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, clear_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, &file->flags); } - call->class->reg(call, TRACE_REG_UNREGISTER, file); + ret = call->class->reg(call, TRACE_REG_UNREGISTER, file); + + WARN_ON_ONCE(ret); } - /* If in SOFT_MODE, just set the SOFT_DISABLE_BIT, else clear it */ - if (file->flags & EVENT_FILE_FL_SOFT_MODE) + /* If in soft mode, just set the SOFT_DISABLE_BIT, else clear it */ + if (soft_mode) set_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags); else clear_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags); @@ -646,7 +824,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, * When soft_disable is set and enable is set, we want to * register the tracepoint for the event, but leave the event * as is. That means, if the event was already enabled, we do - * nothing (but set SOFT_MODE). If the event is disabled, we + * nothing (but set soft_mode). If the event is disabled, we * set SOFT_DISABLED before enabling the event tracepoint, so * it still seems to be disabled. */ @@ -655,23 +833,25 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, else { if (atomic_inc_return(&file->sm_ref) > 1) break; - set_bit(EVENT_FILE_FL_SOFT_MODE_BIT, &file->flags); + soft_mode = true; + /* Enable use of trace_buffered_event */ + trace_buffered_event_enable(); } if (!(file->flags & EVENT_FILE_FL_ENABLED)) { bool cmd = false, tgid = false; - /* Keep the event disabled, when going to SOFT_MODE. */ + /* Keep the event disabled, when going to soft mode. */ if (soft_disable) set_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags); - if (tr->trace_flags & TRACE_ITER_RECORD_CMD) { + if (tr->trace_flags & TRACE_ITER(RECORD_CMD)) { cmd = true; tracing_start_cmdline_record(); set_bit(EVENT_FILE_FL_RECORDED_CMD_BIT, &file->flags); } - if (tr->trace_flags & TRACE_ITER_RECORD_TGID) { + if (tr->trace_flags & TRACE_ITER(RECORD_TGID)) { tgid = true; tracing_start_tgid_record(); set_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, &file->flags); @@ -695,15 +875,6 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, break; } - /* Enable or disable use of trace_buffered_event */ - if ((file_flags & EVENT_FILE_FL_SOFT_DISABLED) != - (file->flags & EVENT_FILE_FL_SOFT_DISABLED)) { - if (file->flags & EVENT_FILE_FL_SOFT_DISABLED) - trace_buffered_event_enable(); - else - trace_buffered_event_disable(); - } - return ret; } @@ -719,6 +890,120 @@ static int ftrace_event_enable_disable(struct trace_event_file *file, return __ftrace_event_enable_disable(file, enable, 0); } +#ifdef CONFIG_MODULES +struct event_mod_load { + struct list_head list; + char *module; + char *match; + char *system; + char *event; +}; + +static void free_event_mod(struct event_mod_load *event_mod) +{ + list_del(&event_mod->list); + kfree(event_mod->module); + kfree(event_mod->match); + kfree(event_mod->system); + kfree(event_mod->event); + kfree(event_mod); +} + +static void clear_mod_events(struct trace_array *tr) +{ + struct event_mod_load *event_mod, *n; + + list_for_each_entry_safe(event_mod, n, &tr->mod_events, list) { + free_event_mod(event_mod); + } +} + +static int remove_cache_mod(struct trace_array *tr, const char *mod, + const char *match, const char *system, const char *event) +{ + struct event_mod_load *event_mod, *n; + int ret = -EINVAL; + + list_for_each_entry_safe(event_mod, n, &tr->mod_events, list) { + if (strcmp(event_mod->module, mod) != 0) + continue; + + if (match && strcmp(event_mod->match, match) != 0) + continue; + + if (system && + (!event_mod->system || strcmp(event_mod->system, system) != 0)) + continue; + + if (event && + (!event_mod->event || strcmp(event_mod->event, event) != 0)) + continue; + + free_event_mod(event_mod); + ret = 0; + } + + return ret; +} + +static int cache_mod(struct trace_array *tr, const char *mod, int set, + const char *match, const char *system, const char *event) +{ + struct event_mod_load *event_mod; + + /* If the module exists, then this just failed to find an event */ + if (module_exists(mod)) + return -EINVAL; + + /* See if this is to remove a cached filter */ + if (!set) + return remove_cache_mod(tr, mod, match, system, event); + + event_mod = kzalloc(sizeof(*event_mod), GFP_KERNEL); + if (!event_mod) + return -ENOMEM; + + INIT_LIST_HEAD(&event_mod->list); + event_mod->module = kstrdup(mod, GFP_KERNEL); + if (!event_mod->module) + goto out_free; + + if (match) { + event_mod->match = kstrdup(match, GFP_KERNEL); + if (!event_mod->match) + goto out_free; + } + + if (system) { + event_mod->system = kstrdup(system, GFP_KERNEL); + if (!event_mod->system) + goto out_free; + } + + if (event) { + event_mod->event = kstrdup(event, GFP_KERNEL); + if (!event_mod->event) + goto out_free; + } + + list_add(&event_mod->list, &tr->mod_events); + + return 0; + + out_free: + free_event_mod(event_mod); + + return -ENOMEM; +} +#else /* CONFIG_MODULES */ +static inline void clear_mod_events(struct trace_array *tr) { } +static int cache_mod(struct trace_array *tr, const char *mod, int set, + const char *match, const char *system, const char *event) +{ + return -EINVAL; +} +#endif + static void ftrace_clear_events(struct trace_array *tr) { struct trace_event_file *file; @@ -727,6 +1012,7 @@ static void ftrace_clear_events(struct trace_array *tr) list_for_each_entry(file, &tr->events, list) { ftrace_event_enable_disable(file, 0); } + clear_mod_events(tr); mutex_unlock(&event_mutex); } @@ -973,32 +1259,41 @@ static void remove_subsystem(struct trace_subsystem_dir *dir) return; if (!--dir->nr_events) { - tracefs_remove(dir->entry); + eventfs_remove_dir(dir->ei); list_del(&dir->list); __put_system_dir(dir); } } -static void remove_event_file_dir(struct trace_event_file *file) +void event_file_get(struct trace_event_file *file) { - struct dentry *dir = file->dir; - struct dentry *child; + refcount_inc(&file->ref); +} - if (dir) { - spin_lock(&dir->d_lock); /* probably unneeded */ - list_for_each_entry(child, &dir->d_subdirs, d_child) { - if (d_really_is_positive(child)) /* probably unneeded */ - d_inode(child)->i_private = NULL; - } - spin_unlock(&dir->d_lock); +void event_file_put(struct trace_event_file *file) +{ + if (WARN_ON_ONCE(!refcount_read(&file->ref))) { + if (file->flags & EVENT_FILE_FL_FREED) + kmem_cache_free(file_cachep, file); + return; + } - tracefs_remove(dir); + if (refcount_dec_and_test(&file->ref)) { + /* Count should only go to zero when it is freed */ + if (WARN_ON_ONCE(!(file->flags & EVENT_FILE_FL_FREED))) + return; + kmem_cache_free(file_cachep, file); } +} +static void remove_event_file_dir(struct trace_event_file *file) +{ + eventfs_remove_dir(file->ei); list_del(&file->list); remove_subsystem(file->system); free_event_filter(file->filter); - kmem_cache_free(file_cachep, file); + file->flags |= EVENT_FILE_FL_FREED; + event_file_put(file); } /* @@ -1006,17 +1301,36 @@ static void remove_event_file_dir(struct trace_event_file *file) */ static int __ftrace_set_clr_event_nolock(struct trace_array *tr, const char *match, - const char *sub, const char *event, int set) + const char *sub, const char *event, int set, + const char *mod) { struct trace_event_file *file; struct trace_event_call *call; + char *module __free(kfree) = NULL; const char *name; int ret = -EINVAL; int eret = 0; + if (mod) { + char *p; + + module = kstrdup(mod, GFP_KERNEL); + if (!module) + return -ENOMEM; + + /* Replace all '-' with '_' as that's what modules do */ + for (p = strchr(module, '-'); p; p = strchr(p + 1, '-')) + *p = '_'; + } + list_for_each_entry(file, &tr->events, list) { call = file->event_call; + + /* If a module is specified, skip events that are not that module */ + if (module && (!call->module || strcmp(module_name(call->module), module))) + continue; + name = trace_event_name(call); if (!name || !call->class || !call->class->reg) @@ -1049,16 +1363,24 @@ __ftrace_set_clr_event_nolock(struct trace_array *tr, const char *match, ret = eret; } + /* + * If this is a module setting and nothing was found, + * check if the module was loaded. If it wasn't cache it. + */ + if (module && ret == -EINVAL && !eret) + ret = cache_mod(tr, module, set, match, sub, event); + return ret; } static int __ftrace_set_clr_event(struct trace_array *tr, const char *match, - const char *sub, const char *event, int set) + const char *sub, const char *event, int set, + const char *mod) { int ret; mutex_lock(&event_mutex); - ret = __ftrace_set_clr_event_nolock(tr, match, sub, event, set); + ret = __ftrace_set_clr_event_nolock(tr, match, sub, event, set, mod); mutex_unlock(&event_mutex); return ret; @@ -1066,11 +1388,20 @@ static int __ftrace_set_clr_event(struct trace_array *tr, const char *match, int ftrace_set_clr_event(struct trace_array *tr, char *buf, int set) { - char *event = NULL, *sub = NULL, *match; + char *event = NULL, *sub = NULL, *match, *mod; int ret; if (!tr) return -ENOENT; + + /* Modules events can be appended with :mod:<module> */ + mod = strstr(buf, ":mod:"); + if (mod) { + *mod = '\0'; + /* move to the module name */ + mod += 5; + } + /* * The buf format can be <subsystem>:<event-name> * *:<event-name> means any event by that name. @@ -1093,9 +1424,13 @@ int ftrace_set_clr_event(struct trace_array *tr, char *buf, int set) sub = NULL; if (!strlen(event) || strcmp(event, "*") == 0) event = NULL; + } else if (mod) { + /* Allow wildcard for no length or star */ + if (!strlen(match) || strcmp(match, "*") == 0) + match = NULL; } - ret = __ftrace_set_clr_event(tr, match, sub, event, set); + ret = __ftrace_set_clr_event(tr, match, sub, event, set, mod); /* Put back the colon to allow this to be called again */ if (buf) @@ -1123,7 +1458,7 @@ int trace_set_clr_event(const char *system, const char *event, int set) if (!tr) return -ENODEV; - return __ftrace_set_clr_event(tr, NULL, system, event, set); + return __ftrace_set_clr_event(tr, NULL, system, event, set, NULL); } EXPORT_SYMBOL_GPL(trace_set_clr_event); @@ -1149,7 +1484,7 @@ int trace_array_set_clr_event(struct trace_array *tr, const char *system, return -ENOENT; set = (enable == true) ? 1 : 0; - return __ftrace_set_clr_event(tr, NULL, system, event, set); + return __ftrace_set_clr_event(tr, NULL, system, event, set, NULL); } EXPORT_SYMBOL_GPL(trace_array_set_clr_event); @@ -1168,7 +1503,7 @@ ftrace_event_write(struct file *file, const char __user *ubuf, if (!cnt) return 0; - ret = tracing_update_buffers(); + ret = tracing_update_buffers(tr); if (ret < 0) return ret; @@ -1236,37 +1571,77 @@ static void *t_start(struct seq_file *m, loff_t *pos) return file; } +enum set_event_iter_type { + SET_EVENT_FILE, + SET_EVENT_MOD, +}; + +struct set_event_iter { + enum set_event_iter_type type; + union { + struct trace_event_file *file; + struct event_mod_load *event_mod; + }; +}; + static void * s_next(struct seq_file *m, void *v, loff_t *pos) { - struct trace_event_file *file = v; + struct set_event_iter *iter = v; + struct trace_event_file *file; struct trace_array *tr = m->private; (*pos)++; - list_for_each_entry_continue(file, &tr->events, list) { - if (file->flags & EVENT_FILE_FL_ENABLED) - return file; + if (iter->type == SET_EVENT_FILE) { + file = iter->file; + list_for_each_entry_continue(file, &tr->events, list) { + if (file->flags & EVENT_FILE_FL_ENABLED) { + iter->file = file; + return iter; + } + } +#ifdef CONFIG_MODULES + iter->type = SET_EVENT_MOD; + iter->event_mod = list_entry(&tr->mod_events, struct event_mod_load, list); +#endif } +#ifdef CONFIG_MODULES + list_for_each_entry_continue(iter->event_mod, &tr->mod_events, list) + return iter; +#endif + + /* + * The iter is allocated in s_start() and passed via the 'v' + * parameter. To stop the iterator, NULL must be returned. But + * the return value is what the 'v' parameter in s_stop() receives + * and frees. Free iter here as it will no longer be used. + */ + kfree(iter); return NULL; } static void *s_start(struct seq_file *m, loff_t *pos) { - struct trace_event_file *file; struct trace_array *tr = m->private; + struct set_event_iter *iter; loff_t l; + iter = kzalloc(sizeof(*iter), GFP_KERNEL); mutex_lock(&event_mutex); + if (!iter) + return NULL; + + iter->type = SET_EVENT_FILE; + iter->file = list_entry(&tr->events, struct trace_event_file, list); - file = list_entry(&tr->events, struct trace_event_file, list); for (l = 0; l <= *pos; ) { - file = s_next(m, file, &l); - if (!file) + iter = s_next(m, iter, &l); + if (!iter) break; } - return file; + return iter; } static int t_show(struct seq_file *m, void *v) @@ -1286,6 +1661,45 @@ static void t_stop(struct seq_file *m, void *p) mutex_unlock(&event_mutex); } +#ifdef CONFIG_MODULES +static int s_show(struct seq_file *m, void *v) +{ + struct set_event_iter *iter = v; + const char *system; + const char *event; + + if (iter->type == SET_EVENT_FILE) + return t_show(m, iter->file); + + /* When match is set, system and event are not */ + if (iter->event_mod->match) { + seq_printf(m, "%s:mod:%s\n", iter->event_mod->match, + iter->event_mod->module); + return 0; + } + + system = iter->event_mod->system ? : "*"; + event = iter->event_mod->event ? : "*"; + + seq_printf(m, "%s:%s:mod:%s\n", system, event, iter->event_mod->module); + + return 0; +} +#else /* CONFIG_MODULES */ +static int s_show(struct seq_file *m, void *v) +{ + struct set_event_iter *iter = v; + + return t_show(m, iter->file); +} +#endif + +static void s_stop(struct seq_file *m, void *v) +{ + kfree(v); + t_stop(m, NULL); +} + static void * __next(struct seq_file *m, void *v, loff_t *pos, int type) { @@ -1366,7 +1780,7 @@ event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, char buf[4] = "0"; mutex_lock(&event_mutex); - file = event_file_data(filp); + file = event_file_file(filp); if (likely(file)) flags = file->flags; mutex_unlock(&event_mutex); @@ -1378,8 +1792,7 @@ event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, !(flags & EVENT_FILE_FL_SOFT_DISABLED)) strcpy(buf, "1"); - if (flags & EVENT_FILE_FL_SOFT_DISABLED || - flags & EVENT_FILE_FL_SOFT_MODE) + if (atomic_read(&file->sm_ref) != 0) strcat(buf, "*"); strcat(buf, "\n"); @@ -1399,19 +1812,20 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, if (ret) return ret; - ret = tracing_update_buffers(); - if (ret < 0) - return ret; + guard(mutex)(&event_mutex); switch (val) { case 0: case 1: - ret = -ENODEV; - mutex_lock(&event_mutex); - file = event_file_data(filp); - if (likely(file)) - ret = ftrace_event_enable_disable(file, val); - mutex_unlock(&event_mutex); + file = event_file_file(filp); + if (!file) + return -ENODEV; + ret = tracing_update_buffers(file->tr); + if (ret < 0) + return ret; + ret = ftrace_event_enable_disable(file, val); + if (ret < 0) + return ret; break; default: @@ -1420,31 +1834,31 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, *ppos += cnt; - return ret ? ret : cnt; + return cnt; } -static ssize_t -system_enable_read(struct file *filp, char __user *ubuf, size_t cnt, - loff_t *ppos) +/* + * Returns: + * 0 : no events exist? + * 1 : all events are disabled + * 2 : all events are enabled + * 3 : some events are enabled and some are enabled + */ +int trace_events_enabled(struct trace_array *tr, const char *system) { - const char set_to_char[4] = { '?', '0', '1', 'X' }; - struct trace_subsystem_dir *dir = filp->private_data; - struct event_subsystem *system = dir->subsystem; struct trace_event_call *call; struct trace_event_file *file; - struct trace_array *tr = dir->tr; - char buf[2]; int set = 0; - int ret; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); + list_for_each_entry(file, &tr->events, list) { call = file->event_call; if ((call->flags & TRACE_EVENT_FL_IGNORE_ENABLE) || !trace_event_name(call) || !call->class || !call->class->reg) continue; - if (system && strcmp(call->class->system, system->name) != 0) + if (system && strcmp(call->class->system, system) != 0) continue; /* @@ -1460,7 +1874,23 @@ system_enable_read(struct file *filp, char __user *ubuf, size_t cnt, if (set == 3) break; } - mutex_unlock(&event_mutex); + + return set; +} + +static ssize_t +system_enable_read(struct file *filp, char __user *ubuf, size_t cnt, + loff_t *ppos) +{ + const char set_to_char[4] = { '?', '0', '1', 'X' }; + struct trace_subsystem_dir *dir = filp->private_data; + struct event_subsystem *system = dir->subsystem; + struct trace_array *tr = dir->tr; + char buf[2]; + int set; + int ret; + + set = trace_events_enabled(tr, system ? system->name : NULL); buf[0] = set_to_char[set]; buf[1] = '\n'; @@ -1484,7 +1914,7 @@ system_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, if (ret) return ret; - ret = tracing_update_buffers(); + ret = tracing_update_buffers(dir->tr); if (ret < 0) return ret; @@ -1498,7 +1928,7 @@ system_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, if (system) name = system->name; - ret = __ftrace_set_clr_event(dir->tr, NULL, name, NULL, val); + ret = __ftrace_set_clr_event(dir->tr, NULL, name, NULL, val, NULL); if (ret) goto out; @@ -1518,7 +1948,8 @@ enum { static void *f_next(struct seq_file *m, void *v, loff_t *pos) { - struct trace_event_call *call = event_file_data(m->private); + struct trace_event_file *file = event_file_data(m->private); + struct trace_event_call *call = file->event_call; struct list_head *common_head = &ftrace_common_fields; struct list_head *head = trace_get_fields(call); struct list_head *node = v; @@ -1550,7 +1981,8 @@ static void *f_next(struct seq_file *m, void *v, loff_t *pos) static int f_show(struct seq_file *m, void *v) { - struct trace_event_call *call = event_file_data(m->private); + struct trace_event_file *file = event_file_data(m->private); + struct trace_event_call *call = file->event_call; struct ftrace_event_field *field; const char *array_descriptor; @@ -1588,24 +2020,31 @@ static int f_show(struct seq_file *m, void *v) seq_printf(m, "\tfield:%s %s;\toffset:%u;\tsize:%u;\tsigned:%d;\n", field->type, field->name, field->offset, field->size, !!field->is_signed); - else - seq_printf(m, "\tfield:%.*s %s%s;\toffset:%u;\tsize:%u;\tsigned:%d;\n", + else if (field->len) + seq_printf(m, "\tfield:%.*s %s[%d];\toffset:%u;\tsize:%u;\tsigned:%d;\n", (int)(array_descriptor - field->type), field->type, field->name, - array_descriptor, field->offset, + field->len, field->offset, field->size, !!field->is_signed); + else + seq_printf(m, "\tfield:%.*s %s[];\toffset:%u;\tsize:%u;\tsigned:%d;\n", + (int)(array_descriptor - field->type), + field->type, field->name, + field->offset, field->size, !!field->is_signed); return 0; } static void *f_start(struct seq_file *m, loff_t *pos) { + struct trace_event_file *file; void *p = (void *)FORMAT_HEADER; loff_t l = 0; /* ->stop() is called even if ->start() fails */ mutex_lock(&event_mutex); - if (!event_file_data(m->private)) + file = event_file_file(m->private); + if (!file) return ERR_PTR(-ENODEV); while (l < *pos && p) @@ -1643,6 +2082,7 @@ static int trace_format_open(struct inode *inode, struct file *file) return 0; } +#ifdef CONFIG_PERF_EVENTS static ssize_t event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { @@ -1657,6 +2097,7 @@ event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) return simple_read_from_buffer(ubuf, cnt, ppos, buf, len); } +#endif static ssize_t event_filter_read(struct file *filp, char __user *ubuf, size_t cnt, @@ -1677,7 +2118,7 @@ event_filter_read(struct file *filp, char __user *ubuf, size_t cnt, trace_seq_init(s); mutex_lock(&event_mutex); - file = event_file_data(filp); + file = event_file_file(filp); if (file) print_event_filter(file, s); mutex_unlock(&event_mutex); @@ -1707,9 +2148,13 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, return PTR_ERR(buf); mutex_lock(&event_mutex); - file = event_file_data(filp); - if (file) - err = apply_event_filter(file, buf); + file = event_file_file(filp); + if (file) { + if (file->flags & EVENT_FILE_FL_FREED) + err = -ENODEV; + else + err = apply_event_filter(file, buf); + } mutex_unlock(&event_mutex); kfree(buf); @@ -1866,9 +2311,9 @@ subsystem_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, } static ssize_t -show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) +show_header_page_file(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - int (*func)(struct trace_seq *s) = filp->private_data; + struct trace_array *tr = filp->private_data; struct trace_seq *s; int r; @@ -1881,7 +2326,31 @@ show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) trace_seq_init(s); - func(s); + ring_buffer_print_page_header(tr->array_buffer.buffer, s); + r = simple_read_from_buffer(ubuf, cnt, ppos, + s->buffer, trace_seq_used(s)); + + kfree(s); + + return r; +} + +static ssize_t +show_header_event_file(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) +{ + struct trace_seq *s; + int r; + + if (*ppos) + return 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) + return -ENOMEM; + + trace_seq_init(s); + + ring_buffer_print_entry_header(s); r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, trace_seq_used(s)); @@ -1953,11 +2422,11 @@ event_pid_write(struct file *filp, const char __user *ubuf, if (!cnt) return 0; - ret = tracing_update_buffers(); + ret = tracing_update_buffers(tr); if (ret < 0) return ret; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); if (type == TRACE_PIDS) { filtered_pids = rcu_dereference_protected(tr->filtered_pids, @@ -1973,7 +2442,7 @@ event_pid_write(struct file *filp, const char __user *ubuf, ret = trace_pid_write(filtered_pids, &pid_list, ubuf, cnt); if (ret < 0) - goto out; + return ret; if (type == TRACE_PIDS) rcu_assign_pointer(tr->filtered_pids, pid_list); @@ -1998,11 +2467,7 @@ event_pid_write(struct file *filp, const char __user *ubuf, */ on_each_cpu(ignore_task_cpu, tr, 1); - out: - mutex_unlock(&event_mutex); - - if (ret > 0) - *ppos += ret; + *ppos += ret; return ret; } @@ -2037,8 +2502,8 @@ static const struct seq_operations show_event_seq_ops = { static const struct seq_operations show_set_event_seq_ops = { .start = s_start, .next = s_next, - .show = t_show, - .stop = t_stop, + .show = s_show, + .stop = s_stop, }; static const struct seq_operations show_set_pid_seq_ops = { @@ -2087,9 +2552,10 @@ static const struct file_operations ftrace_set_event_notrace_pid_fops = { }; static const struct file_operations ftrace_enable_fops = { - .open = tracing_open_generic, + .open = tracing_open_file_tr, .read = event_enable_read, .write = event_enable_write, + .release = tracing_release_file_tr, .llseek = default_llseek, }; @@ -2100,15 +2566,18 @@ static const struct file_operations ftrace_event_format_fops = { .release = seq_release, }; +#ifdef CONFIG_PERF_EVENTS static const struct file_operations ftrace_event_id_fops = { .read = event_id_read, .llseek = default_llseek, }; +#endif static const struct file_operations ftrace_event_filter_fops = { - .open = tracing_open_generic, + .open = tracing_open_file_tr, .read = event_filter_read, .write = event_filter_write, + .release = tracing_release_file_tr, .llseek = default_llseek, }; @@ -2136,10 +2605,18 @@ static const struct file_operations ftrace_tr_enable_fops = { .release = subsystem_release, }; -static const struct file_operations ftrace_show_header_fops = { - .open = tracing_open_generic, - .read = show_header, +static const struct file_operations ftrace_show_header_page_fops = { + .open = tracing_open_generic_tr, + .read = show_header_page_file, .llseek = default_llseek, + .release = tracing_release_generic_tr, +}; + +static const struct file_operations ftrace_show_header_event_fops = { + .open = tracing_open_generic_tr, + .read = show_header_event_file, + .llseek = default_llseek, + .release = tracing_release_generic_tr, }; static int @@ -2261,8 +2738,6 @@ create_new_subsystem(const char *name) if (!system->name) goto out_free; - system->filter = NULL; - system->filter = kzalloc(sizeof(struct event_filter), GFP_KERNEL); if (!system->filter) goto out_free; @@ -2277,13 +2752,40 @@ create_new_subsystem(const char *name) return NULL; } -static struct dentry * +static int system_callback(const char *name, umode_t *mode, void **data, + const struct file_operations **fops) +{ + if (strcmp(name, "filter") == 0) + *fops = &ftrace_subsystem_filter_fops; + + else if (strcmp(name, "enable") == 0) + *fops = &ftrace_system_enable_fops; + + else + return 0; + + *mode = TRACE_MODE_WRITE; + return 1; +} + +static struct eventfs_inode * event_subsystem_dir(struct trace_array *tr, const char *name, - struct trace_event_file *file, struct dentry *parent) + struct trace_event_file *file, struct eventfs_inode *parent) { struct event_subsystem *system, *iter; struct trace_subsystem_dir *dir; - struct dentry *entry; + struct eventfs_inode *ei; + int nr_entries; + static struct eventfs_entry system_entries[] = { + { + .name = "filter", + .callback = system_callback, + }, + { + .name = "enable", + .callback = system_callback, + } + }; /* First see if we did not already create this dir */ list_for_each_entry(dir, &tr->systems, list) { @@ -2291,7 +2793,7 @@ event_subsystem_dir(struct trace_array *tr, const char *name, if (strcmp(system->name, name) == 0) { dir->nr_events++; file->system = dir; - return dir->entry; + return dir->ei; } } @@ -2315,38 +2817,29 @@ event_subsystem_dir(struct trace_array *tr, const char *name, } else __get_system(system); - dir->entry = tracefs_create_dir(name, parent); - if (!dir->entry) { + /* ftrace only has directories no files */ + if (strcmp(name, "ftrace") == 0) + nr_entries = 0; + else + nr_entries = ARRAY_SIZE(system_entries); + + ei = eventfs_create_dir(name, parent, system_entries, nr_entries, dir); + if (IS_ERR(ei)) { pr_warn("Failed to create system directory %s\n", name); __put_system(system); goto out_free; } + dir->ei = ei; dir->tr = tr; dir->ref_count = 1; dir->nr_events = 1; dir->subsystem = system; file->system = dir; - /* the ftrace system is special, do not create enable or filter files */ - if (strcmp(name, "ftrace") != 0) { - - entry = tracefs_create_file("filter", TRACE_MODE_WRITE, - dir->entry, dir, - &ftrace_subsystem_filter_fops); - if (!entry) { - kfree(system->filter); - system->filter = NULL; - pr_warn("Could not create tracefs '%s/filter' entry\n", name); - } - - trace_create_file("enable", TRACE_MODE_WRITE, dir->entry, dir, - &ftrace_system_enable_fops); - } - list_add(&dir->list, &tr->systems); - return dir->entry; + return dir->ei; out_free: kfree(dir); @@ -2379,9 +2872,10 @@ event_define_fields(struct trace_event_call *call) } offset = ALIGN(offset, field->align); - ret = trace_define_field(call, field->type, field->name, + ret = trace_define_field_ext(call, field->type, field->name, offset, field->size, - field->is_signed, field->filter_type); + field->is_signed, field->filter_type, + field->len, field->needs_test); if (WARN_ON_ONCE(ret)) { pr_err("error code is %d\n", ret); break; @@ -2394,43 +2888,165 @@ event_define_fields(struct trace_event_call *call) return ret; } +static int event_callback(const char *name, umode_t *mode, void **data, + const struct file_operations **fops) +{ + struct trace_event_file *file = *data; + struct trace_event_call *call = file->event_call; + + if (strcmp(name, "format") == 0) { + *mode = TRACE_MODE_READ; + *fops = &ftrace_event_format_fops; + return 1; + } + + /* + * Only event directories that can be enabled should have + * triggers or filters, with the exception of the "print" + * event that can have a "trigger" file. + */ + if (!(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE)) { + if (call->class->reg && strcmp(name, "enable") == 0) { + *mode = TRACE_MODE_WRITE; + *fops = &ftrace_enable_fops; + return 1; + } + + if (strcmp(name, "filter") == 0) { + *mode = TRACE_MODE_WRITE; + *fops = &ftrace_event_filter_fops; + return 1; + } + } + + if (!(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE) || + strcmp(trace_event_name(call), "print") == 0) { + if (strcmp(name, "trigger") == 0) { + *mode = TRACE_MODE_WRITE; + *fops = &event_trigger_fops; + return 1; + } + } + +#ifdef CONFIG_PERF_EVENTS + if (call->event.type && call->class->reg && + strcmp(name, "id") == 0) { + *mode = TRACE_MODE_READ; + *data = (void *)(long)call->event.type; + *fops = &ftrace_event_id_fops; + return 1; + } +#endif + +#ifdef CONFIG_HIST_TRIGGERS + if (strcmp(name, "hist") == 0) { + *mode = TRACE_MODE_READ; + *fops = &event_hist_fops; + return 1; + } +#endif +#ifdef CONFIG_HIST_TRIGGERS_DEBUG + if (strcmp(name, "hist_debug") == 0) { + *mode = TRACE_MODE_READ; + *fops = &event_hist_debug_fops; + return 1; + } +#endif +#ifdef CONFIG_TRACE_EVENT_INJECT + if (call->event.type && call->class->reg && + strcmp(name, "inject") == 0) { + *mode = 0200; + *fops = &event_inject_fops; + return 1; + } +#endif + return 0; +} + +/* The file is incremented on creation and freeing the enable file decrements it */ +static void event_release(const char *name, void *data) +{ + struct trace_event_file *file = data; + + event_file_put(file); +} + static int -event_create_dir(struct dentry *parent, struct trace_event_file *file) +event_create_dir(struct eventfs_inode *parent, struct trace_event_file *file) { struct trace_event_call *call = file->event_call; struct trace_array *tr = file->tr; - struct dentry *d_events; + struct eventfs_inode *e_events; + struct eventfs_inode *ei; const char *name; + int nr_entries; int ret; + static struct eventfs_entry event_entries[] = { + { + .name = "enable", + .callback = event_callback, + .release = event_release, + }, + { + .name = "filter", + .callback = event_callback, + }, + { + .name = "trigger", + .callback = event_callback, + }, + { + .name = "format", + .callback = event_callback, + }, +#ifdef CONFIG_PERF_EVENTS + { + .name = "id", + .callback = event_callback, + }, +#endif +#ifdef CONFIG_HIST_TRIGGERS + { + .name = "hist", + .callback = event_callback, + }, +#endif +#ifdef CONFIG_HIST_TRIGGERS_DEBUG + { + .name = "hist_debug", + .callback = event_callback, + }, +#endif +#ifdef CONFIG_TRACE_EVENT_INJECT + { + .name = "inject", + .callback = event_callback, + }, +#endif + }; /* * If the trace point header did not define TRACE_SYSTEM - * then the system would be called "TRACE_SYSTEM". + * then the system would be called "TRACE_SYSTEM". This should + * never happen. */ - if (strcmp(call->class->system, TRACE_SYSTEM) != 0) { - d_events = event_subsystem_dir(tr, call->class->system, file, parent); - if (!d_events) - return -ENOMEM; - } else - d_events = parent; + if (WARN_ON_ONCE(strcmp(call->class->system, TRACE_SYSTEM) == 0)) + return -ENODEV; + + e_events = event_subsystem_dir(tr, call->class->system, file, parent); + if (!e_events) + return -ENOMEM; + + nr_entries = ARRAY_SIZE(event_entries); name = trace_event_name(call); - file->dir = tracefs_create_dir(name, d_events); - if (!file->dir) { + ei = eventfs_create_dir(name, e_events, event_entries, nr_entries, file); + if (IS_ERR(ei)) { pr_warn("Could not create tracefs '%s' directory\n", name); return -1; } - if (call->class->reg && !(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE)) - trace_create_file("enable", TRACE_MODE_WRITE, file->dir, file, - &ftrace_enable_fops); - -#ifdef CONFIG_PERF_EVENTS - if (call->event.type && call->class->reg) - trace_create_file("id", TRACE_MODE_READ, file->dir, - (void *)(long)call->event.type, - &ftrace_event_id_fops); -#endif + file->ei = ei; ret = event_define_fields(call); if (ret < 0) { @@ -2438,34 +3054,8 @@ event_create_dir(struct dentry *parent, struct trace_event_file *file) return ret; } - /* - * Only event directories that can be enabled should have - * triggers or filters. - */ - if (!(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE)) { - trace_create_file("filter", TRACE_MODE_WRITE, file->dir, - file, &ftrace_event_filter_fops); - - trace_create_file("trigger", TRACE_MODE_WRITE, file->dir, - file, &event_trigger_fops); - } - -#ifdef CONFIG_HIST_TRIGGERS - trace_create_file("hist", TRACE_MODE_READ, file->dir, file, - &event_hist_fops); -#endif -#ifdef CONFIG_HIST_TRIGGERS_DEBUG - trace_create_file("hist_debug", TRACE_MODE_READ, file->dir, file, - &event_hist_debug_fops); -#endif - trace_create_file("format", TRACE_MODE_READ, file->dir, call, - &ftrace_event_format_fops); - -#ifdef CONFIG_TRACE_EVENT_INJECT - if (call->event.type && call->class->reg) - trace_create_file("inject", 0200, file->dir, file, - &event_inject_fops); -#endif + /* Gets decremented on freeing of the "enable" file */ + event_file_get(file); return 0; } @@ -2545,7 +3135,10 @@ __register_event(struct trace_event_call *call, struct module *mod) if (ret < 0) return ret; + down_write(&trace_event_sem); list_add(&call->list, &ftrace_events); + up_write(&trace_event_sem); + if (call->flags & TRACE_EVENT_FL_DYNAMIC) atomic_set(&call->refcnt, 0); else @@ -2673,43 +3266,120 @@ static void add_str_to_module(struct module *module, char *str) list_add(&modstr->next, &module_strings); } +#define ATTRIBUTE_STR "__attribute__(" +#define ATTRIBUTE_STR_LEN (sizeof(ATTRIBUTE_STR) - 1) + +/* Remove all __attribute__() from @type. Return allocated string or @type. */ +static char *sanitize_field_type(const char *type) +{ + char *attr, *tmp, *next, *ret = (char *)type; + int depth; + + next = (char *)type; + while ((attr = strstr(next, ATTRIBUTE_STR))) { + /* Retry if "__attribute__(" is a part of another word. */ + if (attr != next && !isspace(attr[-1])) { + next = attr + ATTRIBUTE_STR_LEN; + continue; + } + + if (ret == type) { + ret = kstrdup(type, GFP_KERNEL); + if (WARN_ON_ONCE(!ret)) + return NULL; + attr = ret + (attr - type); + } + + /* the ATTRIBUTE_STR already has the first '(' */ + depth = 1; + next = attr + ATTRIBUTE_STR_LEN; + do { + tmp = strpbrk(next, "()"); + /* There is unbalanced parentheses */ + if (WARN_ON_ONCE(!tmp)) { + kfree(ret); + return (char *)type; + } + + if (*tmp == '(') + depth++; + else + depth--; + next = tmp + 1; + } while (depth > 0); + next = skip_spaces(next); + strcpy(attr, next); + next = attr; + } + return ret; +} + +static char *find_replacable_eval(const char *type, const char *eval_string, + int len) +{ + char *ptr; + + if (!eval_string) + return NULL; + + ptr = strchr(type, '['); + if (!ptr) + return NULL; + ptr++; + + if (!isalpha(*ptr) && *ptr != '_') + return NULL; + + if (strncmp(eval_string, ptr, len) != 0) + return NULL; + + return ptr; +} + static void update_event_fields(struct trace_event_call *call, struct trace_eval_map *map) { struct ftrace_event_field *field; + const char *eval_string = NULL; struct list_head *head; + int len = 0; char *ptr; char *str; - int len = strlen(map->eval_string); /* Dynamic events should never have field maps */ - if (WARN_ON_ONCE(call->flags & TRACE_EVENT_FL_DYNAMIC)) + if (call->flags & TRACE_EVENT_FL_DYNAMIC) return; + if (map) { + eval_string = map->eval_string; + len = strlen(map->eval_string); + } + head = trace_get_fields(call); list_for_each_entry(field, head, link) { - ptr = strchr(field->type, '['); - if (!ptr) - continue; - ptr++; - - if (!isalpha(*ptr) && *ptr != '_') - continue; + str = sanitize_field_type(field->type); + if (!str) + return; - if (strncmp(map->eval_string, ptr, len) != 0) - continue; + ptr = find_replacable_eval(str, eval_string, len); + if (ptr) { + if (str == field->type) { + str = kstrdup(field->type, GFP_KERNEL); + if (WARN_ON_ONCE(!str)) + return; + ptr = str + (ptr - field->type); + } - str = kstrdup(field->type, GFP_KERNEL); - if (WARN_ON_ONCE(!str)) - return; - ptr = str + (ptr - field->type); - ptr = eval_replace(ptr, map, len); - /* enum/sizeof string smaller than value */ - if (WARN_ON_ONCE(!ptr)) { - kfree(str); - continue; + ptr = eval_replace(ptr, map, len); + /* enum/sizeof string smaller than value */ + if (WARN_ON_ONCE(!ptr)) { + kfree(str); + continue; + } } + if (str == field->type) + continue; /* * If the event is part of a module, then we need to free the string * when the module is removed. Otherwise, it will stay allocated @@ -2719,14 +3389,18 @@ static void update_event_fields(struct trace_event_call *call, add_str_to_module(call->module, str); field->type = str; + if (field->filter_type == FILTER_OTHER) + field->filter_type = filter_assign_type(field->type); } } -void trace_event_eval_update(struct trace_eval_map **map, int len) +/* Update all events for replacing eval and sanitizing */ +void trace_event_update_all(struct trace_eval_map **map, int len) { struct trace_event_call *call, *p; const char *last_system = NULL; bool first = false; + bool updated; int last_i; int i; @@ -2739,6 +3413,7 @@ void trace_event_eval_update(struct trace_eval_map **map, int len) last_system = call->class->system; } + updated = false; /* * Since calls are grouped by systems, the likelihood that the * next call in the iteration belongs to the same system as the @@ -2758,12 +3433,52 @@ void trace_event_eval_update(struct trace_eval_map **map, int len) } update_event_printk(call, map[i]); update_event_fields(call, map[i]); + updated = true; } } + /* If not updated yet, update field for sanitizing. */ + if (!updated) + update_event_fields(call, NULL); + cond_resched(); } up_write(&trace_event_sem); } +static bool event_in_systems(struct trace_event_call *call, + const char *systems) +{ + const char *system; + const char *p; + + if (!systems) + return true; + + system = call->class->system; + p = strstr(systems, system); + if (!p) + return false; + + if (p != systems && !isspace(*(p - 1)) && *(p - 1) != ',') + return false; + + p += strlen(system); + return !*p || isspace(*p) || *p == ','; +} + +#ifdef CONFIG_HIST_TRIGGERS +/* + * Wake up waiter on the hist_poll_wq from irq_work because the hist trigger + * may happen in any context. + */ +static void hist_poll_event_irq_work(struct irq_work *work) +{ + wake_up_all(&hist_poll_wq); +} + +DEFINE_IRQ_WORK(hist_poll_work, hist_poll_event_irq_work); +DECLARE_WAIT_QUEUE_HEAD(hist_poll_wq); +#endif + static struct trace_event_file * trace_create_new_event(struct trace_event_call *call, struct trace_array *tr) @@ -2773,9 +3488,12 @@ trace_create_new_event(struct trace_event_call *call, struct trace_event_file *file; unsigned int first; + if (!event_in_systems(call, tr->system_names)) + return NULL; + file = kmem_cache_alloc(file_cachep, GFP_TRACE); if (!file) - return NULL; + return ERR_PTR(-ENOMEM); pid_list = rcu_dereference_protected(tr->filtered_pids, lockdep_is_held(&event_mutex)); @@ -2792,6 +3510,7 @@ trace_create_new_event(struct trace_event_call *call, atomic_set(&file->tm_ref, 0); INIT_LIST_HEAD(&file->triggers); list_add(&file->list, &tr->events); + refcount_set(&file->ref, 1); return file; } @@ -2812,8 +3531,8 @@ static __init int setup_trace_triggers(char *str) char *buf; int i; - strlcpy(bootup_trigger_buf, str, COMMAND_LINE_SIZE); - ring_buffer_expanded = true; + strscpy(bootup_trigger_buf, str, COMMAND_LINE_SIZE); + trace_set_ring_buffer_expanded(NULL); disable_tracing_selftest("running event triggers"); buf = bootup_trigger_buf; @@ -2822,7 +3541,7 @@ static __init int setup_trace_triggers(char *str) if (!trigger) break; bootup_triggers[i].event = strsep(&trigger, "."); - bootup_triggers[i].trigger = strsep(&trigger, "."); + bootup_triggers[i].trigger = trigger; if (!bootup_triggers[i].trigger) break; } @@ -2839,8 +3558,17 @@ __trace_add_new_event(struct trace_event_call *call, struct trace_array *tr) struct trace_event_file *file; file = trace_create_new_event(call, tr); + /* + * trace_create_new_event() returns ERR_PTR(-ENOMEM) if failed + * allocation, or NULL if the event is not part of the tr->system_names. + * When the event is not part of the tr->system_names, return zero, not + * an error. + */ if (!file) - return -ENOMEM; + return 0; + + if (IS_ERR(file)) + return PTR_ERR(file); if (eventdir_initialized) return event_create_dir(tr->event_dir, file); @@ -2879,8 +3607,17 @@ __trace_early_add_new_event(struct trace_event_call *call, int ret; file = trace_create_new_event(call, tr); + /* + * trace_create_new_event() returns ERR_PTR(-ENOMEM) if failed + * allocation, or NULL if the event is not part of the tr->system_names. + * When the event is not part of the tr->system_names, return zero, not + * an error. + */ if (!file) - return -ENOMEM; + return 0; + + if (IS_ERR(file)) + return PTR_ERR(file); ret = event_define_fields(call); if (ret) @@ -2900,13 +3637,13 @@ int trace_add_event_call(struct trace_event_call *call) int ret; lockdep_assert_held(&event_mutex); - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); ret = __register_event(call, NULL); - if (ret >= 0) - __add_event_to_tracers(call); + if (ret < 0) + return ret; - mutex_unlock(&trace_types_lock); + __add_event_to_tracers(call); return ret; } EXPORT_SYMBOL_GPL(trace_add_event_call); @@ -2919,8 +3656,6 @@ static void __trace_remove_event_call(struct trace_event_call *call) { event_remove(call); trace_destroy_fields(call); - free_event_filter(call->filter); - call->filter = NULL; } static int probe_remove_event_call(struct trace_event_call *call) @@ -2937,7 +3672,7 @@ static int probe_remove_event_call(struct trace_event_call *call) continue; /* * We can't rely on ftrace_event_enable_disable(enable => 0) - * we are going to do, EVENT_FILE_FL_SOFT_MODE can suppress + * we are going to do, soft mode can suppress * TRACE_REG_UNREGISTER. */ if (file->flags & EVENT_FILE_FL_ENABLED) @@ -2988,6 +3723,28 @@ EXPORT_SYMBOL_GPL(trace_remove_event_call); event++) #ifdef CONFIG_MODULES +static void update_mod_cache(struct trace_array *tr, struct module *mod) +{ + struct event_mod_load *event_mod, *n; + + list_for_each_entry_safe(event_mod, n, &tr->mod_events, list) { + if (strcmp(event_mod->module, mod->name) != 0) + continue; + + __ftrace_set_clr_event_nolock(tr, event_mod->match, + event_mod->system, + event_mod->event, 1, mod->name); + free_event_mod(event_mod); + } +} + +static void update_cache_events(struct module *mod) +{ + struct trace_array *tr; + + list_for_each_entry(tr, &ftrace_trace_arrays, list) + update_mod_cache(tr, mod); +} static void trace_module_add_events(struct module *mod) { @@ -3010,6 +3767,8 @@ static void trace_module_add_events(struct module *mod) __register_event(*call, mod); __add_event_to_tracers(*call); } + + update_cache_events(mod); } static void trace_module_remove_events(struct module *mod) @@ -3024,7 +3783,7 @@ static void trace_module_remove_events(struct module *mod) if (call->module == mod) __trace_remove_event_call(call); } - /* Check for any strings allocade for this module */ + /* Check for any strings allocated for this module */ list_for_each_entry_safe(modstr, m, &module_strings, next) { if (modstr->module != mod) continue; @@ -3079,6 +3838,8 @@ __trace_add_event_dirs(struct trace_array *tr) struct trace_event_call *call; int ret; + lockdep_assert_held(&trace_event_sem); + list_for_each_entry(call, &ftrace_events, list) { ret = __trace_add_new_event(call, tr); if (ret < 0) @@ -3162,30 +3923,21 @@ struct trace_event_file *trace_get_event_file(const char *instance, return ERR_PTR(ret); } - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); file = find_event_file(tr, system, event); if (!file) { trace_array_put(tr); - ret = -EINVAL; - goto out; + return ERR_PTR(-EINVAL); } /* Don't let event modules unload while in use */ ret = trace_event_try_get_ref(file->event_call); if (!ret) { trace_array_put(tr); - ret = -EBUSY; - goto out; + return ERR_PTR(-EBUSY); } - ret = 0; - out: - mutex_unlock(&event_mutex); - - if (ret) - file = ERR_PTR(ret); - return file; } EXPORT_SYMBOL_GPL(trace_get_event_file); @@ -3335,7 +4087,7 @@ static int free_probe_data(void *data) edata->ref--; if (!edata->ref) { - /* Remove the SOFT_MODE flag */ + /* Remove soft mode */ __ftrace_event_enable_disable(edata->file, 0, 1); trace_event_put_ref(edata->file->event_call); kfree(edata); @@ -3403,6 +4155,7 @@ event_enable_func(struct trace_array *tr, struct ftrace_hash *hash, struct trace_event_file *file; struct ftrace_probe_ops *ops; struct event_probe_data *data; + unsigned long count = -1; const char *system; const char *event; char *number; @@ -3422,12 +4175,11 @@ event_enable_func(struct trace_array *tr, struct ftrace_hash *hash, event = strsep(¶m, ":"); - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); - ret = -EINVAL; file = find_event_file(tr, system, event); if (!file) - goto out; + return -EINVAL; enable = strcmp(cmd, ENABLE_EVENT_STR) == 0; @@ -3436,74 +4188,62 @@ event_enable_func(struct trace_array *tr, struct ftrace_hash *hash, else ops = param ? &event_disable_count_probe_ops : &event_disable_probe_ops; - if (glob[0] == '!') { - ret = unregister_ftrace_function_probe_func(glob+1, tr, ops); - goto out; - } - - ret = -ENOMEM; - - data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) - goto out; - - data->enable = enable; - data->count = -1; - data->file = file; + if (glob[0] == '!') + return unregister_ftrace_function_probe_func(glob+1, tr, ops); - if (!param) - goto out_reg; + if (param) { + number = strsep(¶m, ":"); - number = strsep(¶m, ":"); - - ret = -EINVAL; - if (!strlen(number)) - goto out_free; + if (!strlen(number)) + return -EINVAL; - /* - * We use the callback data field (which is a pointer) - * as our counter. - */ - ret = kstrtoul(number, 0, &data->count); - if (ret) - goto out_free; + /* + * We use the callback data field (which is a pointer) + * as our counter. + */ + ret = kstrtoul(number, 0, &count); + if (ret) + return ret; + } - out_reg: /* Don't let event modules unload while probe registered */ ret = trace_event_try_get_ref(file->event_call); - if (!ret) { - ret = -EBUSY; - goto out_free; - } + if (!ret) + return -EBUSY; ret = __ftrace_event_enable_disable(file, 1, 1); if (ret < 0) goto out_put; + ret = -ENOMEM; + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + goto out_put; + + data->enable = enable; + data->count = count; + data->file = file; + ret = register_ftrace_function_probe(glob, tr, ops, data); /* * The above returns on success the # of functions enabled, * but if it didn't find any functions it returns zero. * Consider no functions a failure too. */ - if (!ret) { - ret = -ENOENT; - goto out_disable; - } else if (ret < 0) - goto out_disable; + /* Just return zero, not the number of enabled functions */ - ret = 0; - out: - mutex_unlock(&event_mutex); - return ret; + if (ret > 0) + return 0; + + kfree(data); + + if (!ret) + ret = -ENOENT; - out_disable: __ftrace_event_enable_disable(file, 0, 1); out_put: trace_event_put_ref(file->event_call); - out_free: - kfree(data); - goto out; + return ret; } static struct ftrace_func_command event_enable_cmd = { @@ -3602,37 +4342,72 @@ static char bootup_event_buf[COMMAND_LINE_SIZE] __initdata; static __init int setup_trace_event(char *str) { - strlcpy(bootup_event_buf, str, COMMAND_LINE_SIZE); - ring_buffer_expanded = true; + strscpy(bootup_event_buf, str, COMMAND_LINE_SIZE); + trace_set_ring_buffer_expanded(NULL); disable_tracing_selftest("running event tracing"); return 1; } __setup("trace_event=", setup_trace_event); +static int events_callback(const char *name, umode_t *mode, void **data, + const struct file_operations **fops) +{ + if (strcmp(name, "enable") == 0) { + *mode = TRACE_MODE_WRITE; + *fops = &ftrace_tr_enable_fops; + return 1; + } + + if (strcmp(name, "header_page") == 0) { + *mode = TRACE_MODE_READ; + *fops = &ftrace_show_header_page_fops; + + } else if (strcmp(name, "header_event") == 0) { + *mode = TRACE_MODE_READ; + *fops = &ftrace_show_header_event_fops; + } else + return 0; + + return 1; +} + /* Expects to have event_mutex held when called */ static int create_event_toplevel_files(struct dentry *parent, struct trace_array *tr) { - struct dentry *d_events; + struct eventfs_inode *e_events; struct dentry *entry; + int nr_entries; + static struct eventfs_entry events_entries[] = { + { + .name = "enable", + .callback = events_callback, + }, + { + .name = "header_page", + .callback = events_callback, + }, + { + .name = "header_event", + .callback = events_callback, + }, + }; entry = trace_create_file("set_event", TRACE_MODE_WRITE, parent, tr, &ftrace_set_event_fops); if (!entry) return -ENOMEM; - d_events = tracefs_create_dir("events", parent); - if (!d_events) { + nr_entries = ARRAY_SIZE(events_entries); + + e_events = eventfs_create_events_dir("events", parent, events_entries, + nr_entries, tr); + if (IS_ERR(e_events)) { pr_warn("Could not create tracefs 'events' directory\n"); return -ENOMEM; } - entry = trace_create_file("enable", TRACE_MODE_WRITE, d_events, - tr, &ftrace_tr_enable_fops); - if (!entry) - return -ENOMEM; - /* There are not as crucial, just warn if they are not created */ trace_create_file("set_event_pid", TRACE_MODE_WRITE, parent, @@ -3642,16 +4417,7 @@ create_event_toplevel_files(struct dentry *parent, struct trace_array *tr) TRACE_MODE_WRITE, parent, tr, &ftrace_set_event_notrace_pid_fops); - /* ring buffer internal formats */ - trace_create_file("header_page", TRACE_MODE_READ, d_events, - ring_buffer_print_page_header, - &ftrace_show_header_fops); - - trace_create_file("header_event", TRACE_MODE_READ, d_events, - ring_buffer_print_entry_header, - &ftrace_show_header_fops); - - tr->event_dir = d_events; + tr->event_dir = e_events; return 0; } @@ -3700,20 +4466,17 @@ early_event_add_tracer(struct dentry *parent, struct trace_array *tr) { int ret; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); ret = create_event_toplevel_files(parent, tr); if (ret) - goto out_unlock; + return ret; down_write(&trace_event_sem); __trace_early_add_event_dirs(tr); up_write(&trace_event_sem); - out_unlock: - mutex_unlock(&event_mutex); - - return ret; + return 0; } /* Must be called with event_mutex held */ @@ -3728,14 +4491,14 @@ int event_trace_del_tracer(struct trace_array *tr) __ftrace_clear_event_pids(tr, TRACE_PIDS | TRACE_NO_PIDS); /* Disable any running events */ - __ftrace_set_clr_event_nolock(tr, NULL, NULL, NULL, 0); + __ftrace_set_clr_event_nolock(tr, NULL, NULL, NULL, 0, NULL); /* Make sure no more events are being executed */ tracepoint_synchronize_unregister(); down_write(&trace_event_sem); __trace_remove_event_dirs(tr); - tracefs_remove(tr->event_dir); + eventfs_remove_events_dir(tr->event_dir); up_write(&trace_event_sem); tr->event_dir = NULL; @@ -3750,10 +4513,9 @@ static __init int event_trace_memsetup(void) return 0; } -static __init void -early_enable_events(struct trace_array *tr, bool disable_first) +__init void +early_enable_events(struct trace_array *tr, char *buf, bool disable_first) { - char *buf = bootup_event_buf; char *token; int ret; @@ -3806,7 +4568,7 @@ static __init int event_trace_enable(void) */ __trace_early_add_events(tr); - early_enable_events(tr, false); + early_enable_events(tr, bootup_event_buf, false); trace_printk_start_comm(); @@ -3834,7 +4596,7 @@ static __init int event_trace_enable_again(void) if (!tr) return -ENODEV; - early_enable_events(tr, true); + early_enable_events(tr, bootup_event_buf, true); return 0; } @@ -4013,7 +4775,7 @@ static __init void event_trace_self_tests(void) pr_info("Testing event system %s: ", system->name); - ret = __ftrace_set_clr_event(tr, NULL, system->name, NULL, 1); + ret = __ftrace_set_clr_event(tr, NULL, system->name, NULL, 1, NULL); if (WARN_ON_ONCE(ret)) { pr_warn("error enabling system %s\n", system->name); @@ -4022,7 +4784,7 @@ static __init void event_trace_self_tests(void) event_test_stuff(); - ret = __ftrace_set_clr_event(tr, NULL, system->name, NULL, 0); + ret = __ftrace_set_clr_event(tr, NULL, system->name, NULL, 0, NULL); if (WARN_ON_ONCE(ret)) { pr_warn("error disabling system %s\n", system->name); @@ -4037,7 +4799,7 @@ static __init void event_trace_self_tests(void) pr_info("Running tests on all trace events:\n"); pr_info("Testing all events: "); - ret = __ftrace_set_clr_event(tr, NULL, NULL, NULL, 1); + ret = __ftrace_set_clr_event(tr, NULL, NULL, NULL, 1, NULL); if (WARN_ON_ONCE(ret)) { pr_warn("error enabling all events\n"); return; @@ -4046,7 +4808,7 @@ static __init void event_trace_self_tests(void) event_test_stuff(); /* reset sysname */ - ret = __ftrace_set_clr_event(tr, NULL, NULL, NULL, 0); + ret = __ftrace_set_clr_event(tr, NULL, NULL, NULL, 0, NULL); if (WARN_ON_ONCE(ret)) { pr_warn("error disabling all events\n"); return; diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index e095c3b3a50d..385af8405392 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -46,15 +46,19 @@ static const char * ops[] = { OPS }; enum filter_pred_fn { FILTER_PRED_FN_NOP, FILTER_PRED_FN_64, + FILTER_PRED_FN_64_CPUMASK, FILTER_PRED_FN_S64, FILTER_PRED_FN_U64, FILTER_PRED_FN_32, + FILTER_PRED_FN_32_CPUMASK, FILTER_PRED_FN_S32, FILTER_PRED_FN_U32, FILTER_PRED_FN_16, + FILTER_PRED_FN_16_CPUMASK, FILTER_PRED_FN_S16, FILTER_PRED_FN_U16, FILTER_PRED_FN_8, + FILTER_PRED_FN_8_CPUMASK, FILTER_PRED_FN_S8, FILTER_PRED_FN_U8, FILTER_PRED_FN_COMM, @@ -64,19 +68,25 @@ enum filter_pred_fn { FILTER_PRED_FN_PCHAR_USER, FILTER_PRED_FN_PCHAR, FILTER_PRED_FN_CPU, + FILTER_PRED_FN_CPU_CPUMASK, + FILTER_PRED_FN_CPUMASK, + FILTER_PRED_FN_CPUMASK_CPU, + FILTER_PRED_FN_FUNCTION, FILTER_PRED_FN_, FILTER_PRED_TEST_VISITED, }; struct filter_pred { - enum filter_pred_fn fn_num; - u64 val; - struct regex regex; + struct regex *regex; + struct cpumask *mask; unsigned short *ops; struct ftrace_event_field *field; - int offset; + u64 val; + u64 val2; + enum filter_pred_fn fn_num; + int offset; int not; - int op; + int op; }; /* @@ -92,6 +102,8 @@ struct filter_pred { C(TOO_MANY_OPEN, "Too many '('"), \ C(TOO_MANY_CLOSE, "Too few '('"), \ C(MISSING_QUOTE, "Missing matching quote"), \ + C(MISSING_BRACE_OPEN, "Missing '{'"), \ + C(MISSING_BRACE_CLOSE, "Missing '}'"), \ C(OPERAND_TOO_LONG, "Operand too long"), \ C(EXPECT_STRING, "Expecting string field"), \ C(EXPECT_DIGIT, "Expecting numeric field"), \ @@ -101,8 +113,10 @@ struct filter_pred { C(BAD_SUBSYS_FILTER, "Couldn't find or set field in one of a subsystem's events"), \ C(TOO_MANY_PREDS, "Too many terms in predicate expression"), \ C(INVALID_FILTER, "Meaningless filter expression"), \ + C(INVALID_CPULIST, "Invalid cpulist"), \ C(IP_FIELD_ONLY, "Only 'ip' field is supported for function trace"), \ C(INVALID_VALUE, "Invalid value (did you forget quotes)?"), \ + C(NO_FUNCTION, "Function not found"), \ C(ERRNO, "Error"), \ C(NO_FILTER, "No filter found") @@ -128,7 +142,7 @@ static bool is_not(const char *str) } /** - * struct prog_entry - a singe entry in the filter program + * struct prog_entry - a single entry in the filter program * @target: Index to jump to on a branch (actually one minus the index) * @when_to_branch: The value of the result of the predicate to do a branch * @pred: The predicate to execute. @@ -183,6 +197,15 @@ enum { PROCESS_OR = 4, }; +static void free_predicate(struct filter_pred *pred) +{ + if (pred) { + kfree(pred->regex); + kfree(pred->mask); + kfree(pred); + } +} + /* * Without going into a formal proof, this explains the method that is used in * parsing the logical expressions. @@ -620,12 +643,64 @@ out_free: kfree(inverts); if (prog_stack) { for (i = 0; prog_stack[i].pred; i++) - kfree(prog_stack[i].pred); + free_predicate(prog_stack[i].pred); kfree(prog_stack); } return ERR_PTR(ret); } +static inline int +do_filter_cpumask(int op, const struct cpumask *mask, const struct cpumask *cmp) +{ + switch (op) { + case OP_EQ: + return cpumask_equal(mask, cmp); + case OP_NE: + return !cpumask_equal(mask, cmp); + case OP_BAND: + return cpumask_intersects(mask, cmp); + default: + return 0; + } +} + +/* Optimisation of do_filter_cpumask() for scalar fields */ +static inline int +do_filter_scalar_cpumask(int op, unsigned int cpu, const struct cpumask *mask) +{ + /* + * Per the weight-of-one cpumask optimisations, the mask passed in this + * function has a weight >= 2, so it is never equal to a single scalar. + */ + switch (op) { + case OP_EQ: + return false; + case OP_NE: + return true; + case OP_BAND: + return cpumask_test_cpu(cpu, mask); + default: + return 0; + } +} + +static inline int +do_filter_cpumask_scalar(int op, const struct cpumask *mask, unsigned int cpu) +{ + switch (op) { + case OP_EQ: + return cpumask_test_cpu(cpu, mask) && + cpumask_nth(1, mask) >= nr_cpu_ids; + case OP_NE: + return !cpumask_test_cpu(cpu, mask) || + cpumask_nth(1, mask) < nr_cpu_ids; + case OP_BAND: + return cpumask_test_cpu(cpu, mask); + default: + return 0; + } +} + enum pred_cmp_types { PRED_CMP_TYPE_NOP, PRED_CMP_TYPE_LT, @@ -669,6 +744,18 @@ static int filter_pred_##type(struct filter_pred *pred, void *event) \ } \ } +#define DEFINE_CPUMASK_COMPARISON_PRED(size) \ +static int filter_pred_##size##_cpumask(struct filter_pred *pred, void *event) \ +{ \ + u##size *addr = (u##size *)(event + pred->offset); \ + unsigned int cpu = *addr; \ + \ + if (cpu >= nr_cpu_ids) \ + return 0; \ + \ + return do_filter_scalar_cpumask(pred->op, cpu, pred->mask); \ +} + #define DEFINE_EQUALITY_PRED(size) \ static int filter_pred_##size(struct filter_pred *pred, void *event) \ { \ @@ -690,6 +777,11 @@ DEFINE_COMPARISON_PRED(u16); DEFINE_COMPARISON_PRED(s8); DEFINE_COMPARISON_PRED(u8); +DEFINE_CPUMASK_COMPARISON_PRED(64); +DEFINE_CPUMASK_COMPARISON_PRED(32); +DEFINE_CPUMASK_COMPARISON_PRED(16); +DEFINE_CPUMASK_COMPARISON_PRED(8); + DEFINE_EQUALITY_PRED(64); DEFINE_EQUALITY_PRED(32); DEFINE_EQUALITY_PRED(16); @@ -716,7 +808,7 @@ static __always_inline char *test_string(char *str) kstr = ubuf->buffer; /* For safety, do not trust the string pointer */ - if (!strncpy_from_kernel_nofault(kstr, str, USTRING_BUF_SIZE)) + if (strncpy_from_kernel_nofault(kstr, str, USTRING_BUF_SIZE) < 0) return NULL; return kstr; } @@ -735,7 +827,7 @@ static __always_inline char *test_ustring(char *str) /* user space address? */ ustr = (char __user *)str; - if (!strncpy_from_user_nofault(kstr, ustr, USTRING_BUF_SIZE)) + if (strncpy_from_user_nofault(kstr, ustr, USTRING_BUF_SIZE) < 0) return NULL; return kstr; @@ -747,7 +839,7 @@ static int filter_pred_string(struct filter_pred *pred, void *event) char *addr = (char *)(event + pred->offset); int cmp, match; - cmp = pred->regex.match(addr, &pred->regex, pred->regex.field_len); + cmp = pred->regex->match(addr, pred->regex, pred->regex->field_len); match = cmp ^ pred->not; @@ -760,7 +852,7 @@ static __always_inline int filter_pchar(struct filter_pred *pred, char *str) int len; len = strlen(str) + 1; /* including tailing '\0' */ - cmp = pred->regex.match(str, &pred->regex, len); + cmp = pred->regex->match(str, pred->regex, len); match = cmp ^ pred->not; @@ -810,7 +902,7 @@ static int filter_pred_strloc(struct filter_pred *pred, void *event) char *addr = (char *)(event + str_loc); int cmp, match; - cmp = pred->regex.match(addr, &pred->regex, str_len); + cmp = pred->regex->match(addr, pred->regex, str_len); match = cmp ^ pred->not; @@ -833,7 +925,7 @@ static int filter_pred_strrelloc(struct filter_pred *pred, void *event) char *addr = (char *)(&item[1]) + str_loc; int cmp, match; - cmp = pred->regex.match(addr, &pred->regex, str_len); + cmp = pred->regex->match(addr, pred->regex, str_len); match = cmp ^ pred->not; @@ -866,16 +958,57 @@ static int filter_pred_cpu(struct filter_pred *pred, void *event) } } +/* Filter predicate for current CPU vs user-provided cpumask */ +static int filter_pred_cpu_cpumask(struct filter_pred *pred, void *event) +{ + int cpu = raw_smp_processor_id(); + + return do_filter_scalar_cpumask(pred->op, cpu, pred->mask); +} + +/* Filter predicate for cpumask field vs user-provided cpumask */ +static int filter_pred_cpumask(struct filter_pred *pred, void *event) +{ + u32 item = *(u32 *)(event + pred->offset); + int loc = item & 0xffff; + const struct cpumask *mask = (event + loc); + const struct cpumask *cmp = pred->mask; + + return do_filter_cpumask(pred->op, mask, cmp); +} + +/* Filter predicate for cpumask field vs user-provided scalar */ +static int filter_pred_cpumask_cpu(struct filter_pred *pred, void *event) +{ + u32 item = *(u32 *)(event + pred->offset); + int loc = item & 0xffff; + const struct cpumask *mask = (event + loc); + unsigned int cpu = pred->val; + + return do_filter_cpumask_scalar(pred->op, mask, cpu); +} + /* Filter predicate for COMM. */ static int filter_pred_comm(struct filter_pred *pred, void *event) { int cmp; - cmp = pred->regex.match(current->comm, &pred->regex, + cmp = pred->regex->match(current->comm, pred->regex, TASK_COMM_LEN); return cmp ^ pred->not; } +/* Filter predicate for functions. */ +static int filter_pred_function(struct filter_pred *pred, void *event) +{ + unsigned long *addr = (unsigned long *)(event + pred->offset); + unsigned long start = (unsigned long)pred->val; + unsigned long end = (unsigned long)pred->val2; + int ret = *addr >= start && *addr < end; + + return pred->op == OP_EQ ? ret : !ret; +} + /* * regex_match_foo - Basic regex callbacks * @@ -990,7 +1123,7 @@ enum regex_type filter_parse_regex(char *buff, int len, char **search, int *not) static void filter_build_regex(struct filter_pred *pred) { - struct regex *r = &pred->regex; + struct regex *r = pred->regex; char *search; enum regex_type type = MATCH_FULL; @@ -1117,7 +1250,9 @@ static void append_filter_err(struct trace_array *tr, static inline struct event_filter *event_filter(struct trace_event_file *file) { - return file->filter; + return rcu_dereference_protected(file->filter, + lockdep_is_held(&event_mutex)); + } /* caller must hold event_mutex */ @@ -1155,7 +1290,7 @@ static void free_prog(struct event_filter *filter) return; for (i = 0; prog[i].pred; i++) - kfree(prog[i].pred); + free_predicate(prog[i].pred); kfree(prog); } @@ -1187,7 +1322,7 @@ void free_event_filter(struct event_filter *filter) static inline void __remove_filter(struct trace_event_file *file) { filter_disable(file); - remove_filter_string(file->filter); + remove_filter_string(event_filter(file)); } static void filter_free_subsystem_preds(struct trace_subsystem_dir *dir, @@ -1202,28 +1337,159 @@ static void filter_free_subsystem_preds(struct trace_subsystem_dir *dir, } } +struct filter_list { + struct list_head list; + struct event_filter *filter; +}; + +struct filter_head { + struct list_head list; + union { + struct rcu_head rcu; + struct rcu_work rwork; + }; +}; + +static void free_filter_list(struct filter_head *filter_list) +{ + struct filter_list *filter_item, *tmp; + + list_for_each_entry_safe(filter_item, tmp, &filter_list->list, list) { + __free_filter(filter_item->filter); + list_del(&filter_item->list); + kfree(filter_item); + } + kfree(filter_list); +} + +static void free_filter_list_work(struct work_struct *work) +{ + struct filter_head *filter_list; + + filter_list = container_of(to_rcu_work(work), struct filter_head, rwork); + free_filter_list(filter_list); +} + +static void free_filter_list_tasks(struct rcu_head *rhp) +{ + struct filter_head *filter_list = container_of(rhp, struct filter_head, rcu); + + INIT_RCU_WORK(&filter_list->rwork, free_filter_list_work); + queue_rcu_work(system_wq, &filter_list->rwork); +} + +/* + * The tracepoint_synchronize_unregister() is a double rcu call. + * It calls synchronize_rcu_tasks_trace() followed by synchronize_rcu(). + * Instead of waiting for it, simply call these via the call_rcu*() + * variants. + */ +static void delay_free_filter(struct filter_head *head) +{ + call_rcu_tasks_trace(&head->rcu, free_filter_list_tasks); +} + +static void try_delay_free_filter(struct event_filter *filter) +{ + struct filter_head *head; + struct filter_list *item; + + head = kmalloc(sizeof(*head), GFP_KERNEL); + if (!head) + goto free_now; + + INIT_LIST_HEAD(&head->list); + + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) { + kfree(head); + goto free_now; + } + + item->filter = filter; + list_add_tail(&item->list, &head->list); + delay_free_filter(head); + return; + + free_now: + /* Make sure the filter is not being used */ + tracepoint_synchronize_unregister(); + __free_filter(filter); +} + static inline void __free_subsystem_filter(struct trace_event_file *file) { - __free_filter(file->filter); + __free_filter(event_filter(file)); file->filter = NULL; } +static inline void event_set_filter(struct trace_event_file *file, + struct event_filter *filter) +{ + rcu_assign_pointer(file->filter, filter); +} + +static inline void event_clear_filter(struct trace_event_file *file) +{ + RCU_INIT_POINTER(file->filter, NULL); +} + static void filter_free_subsystem_filters(struct trace_subsystem_dir *dir, - struct trace_array *tr) + struct trace_array *tr, + struct event_filter *filter) { struct trace_event_file *file; + struct filter_head *head; + struct filter_list *item; + + head = kmalloc(sizeof(*head), GFP_KERNEL); + if (!head) + goto free_now; + + INIT_LIST_HEAD(&head->list); list_for_each_entry(file, &tr->events, list) { if (file->system != dir) continue; + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) + goto free_now; + item->filter = event_filter(file); + list_add_tail(&item->list, &head->list); + event_clear_filter(file); + } + + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) + goto free_now; + + item->filter = filter; + list_add_tail(&item->list, &head->list); + + delay_free_filter(head); + return; + free_now: + tracepoint_synchronize_unregister(); + + if (head) + free_filter_list(head); + + list_for_each_entry(file, &tr->events, list) { + if (file->system != dir || !file->filter) + continue; __free_subsystem_filter(file); } + __free_filter(filter); } int filter_assign_type(const char *type) { - if (strstr(type, "__data_loc") && strstr(type, "char")) - return FILTER_DYN_STRING; + if (strstr(type, "__data_loc")) { + if (strstr(type, "char")) + return FILTER_DYN_STRING; + if (strstr(type, "cpumask_t")) + return FILTER_CPUMASK; + } if (strstr(type, "__rel_loc") && strstr(type, "char")) return FILTER_RDYN_STRING; @@ -1299,24 +1565,32 @@ static int filter_pred_fn_call(struct filter_pred *pred, void *event) switch (pred->fn_num) { case FILTER_PRED_FN_64: return filter_pred_64(pred, event); + case FILTER_PRED_FN_64_CPUMASK: + return filter_pred_64_cpumask(pred, event); case FILTER_PRED_FN_S64: return filter_pred_s64(pred, event); case FILTER_PRED_FN_U64: return filter_pred_u64(pred, event); case FILTER_PRED_FN_32: return filter_pred_32(pred, event); + case FILTER_PRED_FN_32_CPUMASK: + return filter_pred_32_cpumask(pred, event); case FILTER_PRED_FN_S32: return filter_pred_s32(pred, event); case FILTER_PRED_FN_U32: return filter_pred_u32(pred, event); case FILTER_PRED_FN_16: return filter_pred_16(pred, event); + case FILTER_PRED_FN_16_CPUMASK: + return filter_pred_16_cpumask(pred, event); case FILTER_PRED_FN_S16: return filter_pred_s16(pred, event); case FILTER_PRED_FN_U16: return filter_pred_u16(pred, event); case FILTER_PRED_FN_8: return filter_pred_8(pred, event); + case FILTER_PRED_FN_8_CPUMASK: + return filter_pred_8_cpumask(pred, event); case FILTER_PRED_FN_S8: return filter_pred_s8(pred, event); case FILTER_PRED_FN_U8: @@ -1335,6 +1609,14 @@ static int filter_pred_fn_call(struct filter_pred *pred, void *event) return filter_pred_pchar(pred, event); case FILTER_PRED_FN_CPU: return filter_pred_cpu(pred, event); + case FILTER_PRED_FN_CPU_CPUMASK: + return filter_pred_cpu_cpumask(pred, event); + case FILTER_PRED_FN_CPUMASK: + return filter_pred_cpumask(pred, event); + case FILTER_PRED_FN_CPUMASK_CPU: + return filter_pred_cpumask_cpu(pred, event); + case FILTER_PRED_FN_FUNCTION: + return filter_pred_function(pred, event); case FILTER_PRED_TEST_VISITED: return test_pred_visited_fn(pred, event); default: @@ -1350,8 +1632,13 @@ static int parse_pred(const char *str, void *data, struct trace_event_call *call = data; struct ftrace_event_field *field; struct filter_pred *pred = NULL; + unsigned long offset; + unsigned long size; + unsigned long ip; char num_buf[24]; /* Big enough to hold an address */ char *field_name; + char *name; + bool function = false; bool ustring = false; char q; u64 val; @@ -1393,6 +1680,12 @@ static int parse_pred(const char *str, void *data, i += len; } + /* See if the field is a kernel function name */ + if ((len = str_has_prefix(str + i, ".function"))) { + function = true; + i += len; + } + while (isspace(str[i])) i++; @@ -1423,7 +1716,71 @@ static int parse_pred(const char *str, void *data, pred->offset = field->offset; pred->op = op; - if (ftrace_event_is_function(call)) { + if (function) { + /* The field must be the same size as long */ + if (field->size != sizeof(long)) { + parse_error(pe, FILT_ERR_ILLEGAL_FIELD_OP, pos + i); + goto err_free; + } + + /* Function only works with '==' or '!=' and an unquoted string */ + switch (op) { + case OP_NE: + case OP_EQ: + break; + default: + parse_error(pe, FILT_ERR_INVALID_OP, pos + i); + goto err_free; + } + + if (isdigit(str[i])) { + /* We allow 0xDEADBEEF */ + while (isalnum(str[i])) + i++; + + len = i - s; + /* 0xfeedfacedeadbeef is 18 chars max */ + if (len >= sizeof(num_buf)) { + parse_error(pe, FILT_ERR_OPERAND_TOO_LONG, pos + i); + goto err_free; + } + + memcpy(num_buf, str + s, len); + num_buf[len] = 0; + + ret = kstrtoul(num_buf, 0, &ip); + if (ret) { + parse_error(pe, FILT_ERR_INVALID_VALUE, pos + i); + goto err_free; + } + } else { + s = i; + for (; str[i] && !isspace(str[i]); i++) + ; + + len = i - s; + name = kmemdup_nul(str + s, len, GFP_KERNEL); + if (!name) + goto err_mem; + ip = kallsyms_lookup_name(name); + kfree(name); + if (!ip) { + parse_error(pe, FILT_ERR_NO_FUNCTION, pos + i); + goto err_free; + } + } + + /* Now find the function start and end address */ + if (!kallsyms_lookup_size_offset(ip, &size, &offset)) { + parse_error(pe, FILT_ERR_NO_FUNCTION, pos + i); + goto err_free; + } + + pred->fn_num = FILTER_PRED_FN_FUNCTION; + pred->val = ip - offset; + pred->val2 = pred->val + size; + + } else if (ftrace_event_is_function(call)) { /* * Perf does things different with function events. * It only allows an "ip" field, and expects a string. @@ -1462,9 +1819,130 @@ static int parse_pred(const char *str, void *data, goto err_free; } - pred->regex.len = len; - strncpy(pred->regex.pattern, str + s, len); - pred->regex.pattern[len] = 0; + pred->regex = kzalloc(sizeof(*pred->regex), GFP_KERNEL); + if (!pred->regex) + goto err_mem; + pred->regex->len = len; + memcpy(pred->regex->pattern, str + s, len); + pred->regex->pattern[len] = 0; + + } else if (!strncmp(str + i, "CPUS", 4)) { + unsigned int maskstart; + bool single; + char *tmp; + + switch (field->filter_type) { + case FILTER_CPUMASK: + case FILTER_CPU: + case FILTER_OTHER: + break; + default: + parse_error(pe, FILT_ERR_ILLEGAL_FIELD_OP, pos + i); + goto err_free; + } + + switch (op) { + case OP_EQ: + case OP_NE: + case OP_BAND: + break; + default: + parse_error(pe, FILT_ERR_ILLEGAL_FIELD_OP, pos + i); + goto err_free; + } + + /* Skip CPUS */ + i += 4; + if (str[i++] != '{') { + parse_error(pe, FILT_ERR_MISSING_BRACE_OPEN, pos + i); + goto err_free; + } + maskstart = i; + + /* Walk the cpulist until closing } */ + for (; str[i] && str[i] != '}'; i++) + ; + + if (str[i] != '}') { + parse_error(pe, FILT_ERR_MISSING_BRACE_CLOSE, pos + i); + goto err_free; + } + + if (maskstart == i) { + parse_error(pe, FILT_ERR_INVALID_CPULIST, pos + i); + goto err_free; + } + + /* Copy the cpulist between { and } */ + tmp = kmalloc((i - maskstart) + 1, GFP_KERNEL); + if (!tmp) + goto err_mem; + + strscpy(tmp, str + maskstart, (i - maskstart) + 1); + pred->mask = kzalloc(cpumask_size(), GFP_KERNEL); + if (!pred->mask) { + kfree(tmp); + goto err_mem; + } + + /* Now parse it */ + if (cpulist_parse(tmp, pred->mask)) { + kfree(tmp); + parse_error(pe, FILT_ERR_INVALID_CPULIST, pos + i); + goto err_free; + } + kfree(tmp); + + /* Move along */ + i++; + + /* + * Optimisation: if the user-provided mask has a weight of one + * then we can treat it as a scalar input. + */ + single = cpumask_weight(pred->mask) == 1; + if (single) { + pred->val = cpumask_first(pred->mask); + kfree(pred->mask); + pred->mask = NULL; + } + + if (field->filter_type == FILTER_CPUMASK) { + pred->fn_num = single ? + FILTER_PRED_FN_CPUMASK_CPU : + FILTER_PRED_FN_CPUMASK; + } else if (field->filter_type == FILTER_CPU) { + if (single) { + if (pred->op == OP_BAND) + pred->op = OP_EQ; + + pred->fn_num = FILTER_PRED_FN_CPU; + } else { + pred->fn_num = FILTER_PRED_FN_CPU_CPUMASK; + } + } else if (single) { + if (pred->op == OP_BAND) + pred->op = OP_EQ; + + pred->fn_num = select_comparison_fn(pred->op, field->size, false); + if (pred->op == OP_NE) + pred->not = 1; + } else { + switch (field->size) { + case 8: + pred->fn_num = FILTER_PRED_FN_64_CPUMASK; + break; + case 4: + pred->fn_num = FILTER_PRED_FN_32_CPUMASK; + break; + case 2: + pred->fn_num = FILTER_PRED_FN_16_CPUMASK; + break; + case 1: + pred->fn_num = FILTER_PRED_FN_8_CPUMASK; + break; + } + } /* This is either a string, or an integer */ } else if (str[i] == '\'' || str[i] == '"') { @@ -1506,9 +1984,12 @@ static int parse_pred(const char *str, void *data, goto err_free; } - pred->regex.len = len; - strncpy(pred->regex.pattern, str + s, len); - pred->regex.pattern[len] = 0; + pred->regex = kzalloc(sizeof(*pred->regex), GFP_KERNEL); + if (!pred->regex) + goto err_mem; + pred->regex->len = len; + memcpy(pred->regex->pattern, str + s, len); + pred->regex->pattern[len] = 0; filter_build_regex(pred); @@ -1517,7 +1998,7 @@ static int parse_pred(const char *str, void *data, } else if (field->filter_type == FILTER_STATIC_STRING) { pred->fn_num = FILTER_PRED_FN_STRING; - pred->regex.field_len = field->size; + pred->regex->field_len = field->size; } else if (field->filter_type == FILTER_DYN_STRING) { pred->fn_num = FILTER_PRED_FN_STRLOC; @@ -1567,7 +2048,7 @@ static int parse_pred(const char *str, void *data, goto err_free; } - strncpy(num_buf, str + s, len); + memcpy(num_buf, str + s, len); num_buf[len] = 0; /* Make sure it is a value */ @@ -1600,10 +2081,10 @@ static int parse_pred(const char *str, void *data, return i; err_free: - kfree(pred); + free_predicate(pred); return -EINVAL; err_mem: - kfree(pred); + free_predicate(pred); return -ENOMEM; } @@ -1768,22 +2249,6 @@ static inline void event_set_filtered_flag(struct trace_event_file *file) trace_buffered_event_enable(); } -static inline void event_set_filter(struct trace_event_file *file, - struct event_filter *filter) -{ - rcu_assign_pointer(file->filter, filter); -} - -static inline void event_clear_filter(struct trace_event_file *file) -{ - RCU_INIT_POINTER(file->filter, NULL); -} - -struct filter_list { - struct list_head list; - struct event_filter *filter; -}; - static int process_system_preds(struct trace_subsystem_dir *dir, struct trace_array *tr, struct filter_parse_error *pe, @@ -1792,11 +2257,16 @@ static int process_system_preds(struct trace_subsystem_dir *dir, struct trace_event_file *file; struct filter_list *filter_item; struct event_filter *filter = NULL; - struct filter_list *tmp; - LIST_HEAD(filter_list); + struct filter_head *filter_list; bool fail = true; int err; + filter_list = kmalloc(sizeof(*filter_list), GFP_KERNEL); + if (!filter_list) + return -ENOMEM; + + INIT_LIST_HEAD(&filter_list->list); + list_for_each_entry(file, &tr->events, list) { if (file->system != dir) @@ -1823,7 +2293,7 @@ static int process_system_preds(struct trace_subsystem_dir *dir, if (!filter_item) goto fail_mem; - list_add_tail(&filter_item->list, &filter_list); + list_add_tail(&filter_item->list, &filter_list->list); /* * Regardless of if this returned an error, we still * replace the filter for the call. @@ -1843,31 +2313,22 @@ static int process_system_preds(struct trace_subsystem_dir *dir, * Do a synchronize_rcu() and to ensure all calls are * done with them before we free them. */ - tracepoint_synchronize_unregister(); - list_for_each_entry_safe(filter_item, tmp, &filter_list, list) { - __free_filter(filter_item->filter); - list_del(&filter_item->list); - kfree(filter_item); - } + delay_free_filter(filter_list); return 0; fail: /* No call succeeded */ - list_for_each_entry_safe(filter_item, tmp, &filter_list, list) { - list_del(&filter_item->list); - kfree(filter_item); - } + free_filter_list(filter_list); parse_error(pe, FILT_ERR_BAD_SUBSYS_FILTER, 0); return -EINVAL; fail_mem: __free_filter(filter); + /* If any call succeeded, we still need to sync */ if (!fail) - tracepoint_synchronize_unregister(); - list_for_each_entry_safe(filter_item, tmp, &filter_list, list) { - __free_filter(filter_item->filter); - list_del(&filter_item->list); - kfree(filter_item); - } + delay_free_filter(filter_list); + else + free_filter_list(filter_list); + return -ENOMEM; } @@ -1997,6 +2458,9 @@ int apply_event_filter(struct trace_event_file *file, char *filter_string) struct event_filter *filter = NULL; int err; + if (file->flags & EVENT_FILE_FL_FREED) + return -ENODEV; + if (!strcmp(strstrip(filter_string), "0")) { filter_disable(file); filter = event_filter(file); @@ -2006,9 +2470,7 @@ int apply_event_filter(struct trace_event_file *file, char *filter_string) event_clear_filter(file); - /* Make sure the filter is not being used */ - tracepoint_synchronize_unregister(); - __free_filter(filter); + try_delay_free_filter(filter); return 0; } @@ -2032,11 +2494,8 @@ int apply_event_filter(struct trace_event_file *file, char *filter_string) event_set_filter(file, filter); - if (tmp) { - /* Make sure the call is done with the filter */ - tracepoint_synchronize_unregister(); - __free_filter(tmp); - } + if (tmp) + try_delay_free_filter(tmp); } return err; @@ -2050,13 +2509,11 @@ int apply_subsystem_event_filter(struct trace_subsystem_dir *dir, struct event_filter *filter = NULL; int err = 0; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); /* Make sure the system still has events */ - if (!dir->nr_events) { - err = -ENODEV; - goto out_unlock; - } + if (!dir->nr_events) + return -ENODEV; if (!strcmp(strstrip(filter_string), "0")) { filter_free_subsystem_preds(dir, tr); @@ -2064,10 +2521,8 @@ int apply_subsystem_event_filter(struct trace_subsystem_dir *dir, filter = system->filter; system->filter = NULL; /* Ensure all filters are no longer used */ - tracepoint_synchronize_unregister(); - filter_free_subsystem_filters(dir, tr); - __free_filter(filter); - goto out_unlock; + filter_free_subsystem_filters(dir, tr, filter); + return 0; } err = create_system_filter(dir, filter_string, &filter); @@ -2079,8 +2534,6 @@ int apply_subsystem_event_filter(struct trace_subsystem_dir *dir, __free_filter(system->filter); system->filter = filter; } -out_unlock: - mutex_unlock(&event_mutex); return err; } @@ -2196,8 +2649,8 @@ static int ftrace_function_set_filter_pred(struct filter_pred *pred, return ret; return __ftrace_function_set_filter(pred->op == OP_EQ, - pred->regex.pattern, - pred->regex.len, + pred->regex->pattern, + pred->regex->len, data); } @@ -2257,17 +2710,15 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id, struct event_filter *filter = NULL; struct trace_event_call *call; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); call = event->tp_event; - err = -EINVAL; if (!call) - goto out_unlock; + return -EINVAL; - err = -EEXIST; if (event->filter) - goto out_unlock; + return -EEXIST; err = create_filter(NULL, call, filter_str, false, &filter); if (err) @@ -2282,9 +2733,6 @@ free_filter: if (err || ftrace_event_is_function(call)) __free_filter(filter); -out_unlock: - mutex_unlock(&event_mutex); - return err; } @@ -2464,6 +2912,10 @@ static __init int ftrace_test_event_filter(void) if (i == DATA_CNT) printk(KERN_CONT "OK\n"); + /* Need to call ftrace_test_filter to prevent a warning */ + if (!trace_ftrace_test_filter_enabled()) + trace_ftrace_test_filter(1, 2, 3, 4, 5, 6, 7, 8); + return 0; } diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 5edbf6b1da3f..5e6e70540eef 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -114,6 +114,7 @@ enum hist_field_fn { HIST_FIELD_FN_BUCKET, HIST_FIELD_FN_TIMESTAMP, HIST_FIELD_FN_CPU, + HIST_FIELD_FN_COMM, HIST_FIELD_FN_STRING, HIST_FIELD_FN_DYNSTRING, HIST_FIELD_FN_RELDYNSTRING, @@ -135,6 +136,7 @@ enum hist_field_fn { HIST_FIELD_FN_DIV_NOT_POWER2, HIST_FIELD_FN_DIV_MULT_SHIFT, HIST_FIELD_FN_EXECNAME, + HIST_FIELD_FN_STACK, }; /* @@ -480,10 +482,6 @@ DEFINE_HIST_FIELD_FN(u8); #define for_each_hist_key_field(i, hist_data) \ for ((i) = (hist_data)->n_vals; (i) < (hist_data)->n_fields; (i)++) -#define HIST_STACKTRACE_DEPTH 16 -#define HIST_STACKTRACE_SIZE (HIST_STACKTRACE_DEPTH * sizeof(unsigned long)) -#define HIST_STACKTRACE_SKIP 5 - #define HITCOUNT_IDX 0 #define HIST_KEY_SIZE_MAX (MAX_FILTER_STR_VAL + HIST_STACKTRACE_SIZE) @@ -509,6 +507,7 @@ enum hist_field_flags { HIST_FIELD_FL_CONST = 1 << 18, HIST_FIELD_FL_PERCENT = 1 << 19, HIST_FIELD_FL_GRAPH = 1 << 20, + HIST_FIELD_FL_COMM = 1 << 21, }; struct var_defs { @@ -777,23 +776,16 @@ static void last_cmd_set(struct trace_event_file *file, char *str) { const char *system = NULL, *name = NULL; struct trace_event_call *call; - int len; if (!str) return; - /* sizeof() contains the nul byte */ - len = sizeof(HIST_PREFIX) + strlen(str); kfree(last_cmd); - last_cmd = kzalloc(len, GFP_KERNEL); + + last_cmd = kasprintf(GFP_KERNEL, HIST_PREFIX "%s", str); if (!last_cmd) return; - strcpy(last_cmd, HIST_PREFIX); - /* Again, sizeof() contains the nul byte */ - len -= sizeof(HIST_PREFIX); - strncat(last_cmd, str, len); - if (file) { call = file->event_call; system = call->class->system; @@ -832,7 +824,7 @@ static inline void trace_synth(struct synth_event *event, u64 *var_ref_vals, { struct tracepoint *tp = event->tp; - if (unlikely(atomic_read(&tp->key.enabled) > 0)) { + if (unlikely(static_key_enabled(&tp->key))) { struct tracepoint_func *probe_func_ptr; synth_probe_func_t probe_func; void *__data; @@ -895,6 +887,15 @@ static u64 hist_field_cpu(struct hist_field *hist_field, return cpu; } +static u64 hist_field_comm(struct hist_field *hist_field, + struct tracing_map_elt *elt, + struct trace_buffer *buffer, + struct ring_buffer_event *rbe, + void *event) +{ + return (u64)(unsigned long)current->comm; +} + /** * check_field_for_var_ref - Check if a VAR_REF field references a variable * @hist_field: The VAR_REF field to check @@ -1334,6 +1335,9 @@ static const char *hist_field_name(struct hist_field *field, { const char *field_name = ""; + if (WARN_ON_ONCE(!field)) + return field_name; + if (level > 1) return field_name; @@ -1345,6 +1349,8 @@ static const char *hist_field_name(struct hist_field *field, field_name = hist_field_name(field->operands[0], ++level); else if (field->flags & HIST_FIELD_FL_CPU) field_name = "common_cpu"; + else if (field->flags & HIST_FIELD_FL_COMM) + field_name = "common_comm"; else if (field->flags & HIST_FIELD_FL_EXPR || field->flags & HIST_FIELD_FL_VAR_REF) { if (field->system) { @@ -1360,7 +1366,9 @@ static const char *hist_field_name(struct hist_field *field, field_name = field->name; } else if (field->flags & HIST_FIELD_FL_TIMESTAMP) field_name = "common_timestamp"; - else if (field->flags & HIST_FIELD_FL_HITCOUNT) + else if (field->flags & HIST_FIELD_FL_STACKTRACE) { + field_name = "common_stacktrace"; + } else if (field->flags & HIST_FIELD_FL_HITCOUNT) field_name = "hitcount"; if (field_name == NULL) @@ -1601,7 +1609,7 @@ static inline void save_comm(char *comm, struct task_struct *task) return; } - strncpy(comm, task->comm, TASK_COMM_LEN); + strscpy(comm, task->comm, TASK_COMM_LEN); } static void hist_elt_data_free(struct hist_elt_data *elt_data) @@ -1718,6 +1726,8 @@ static const char *get_hist_field_flags(struct hist_field *hist_field) flags_str = "percent"; else if (hist_field->flags & HIST_FIELD_FL_GRAPH) flags_str = "graph"; + else if (hist_field->flags & HIST_FIELD_FL_STACKTRACE) + flags_str = "stacktrace"; return flags_str; } @@ -1979,7 +1989,14 @@ static struct hist_field *create_hist_field(struct hist_trigger_data *hist_data, } if (flags & HIST_FIELD_FL_STACKTRACE) { - hist_field->fn_num = HIST_FIELD_FN_NOP; + if (field) + hist_field->fn_num = HIST_FIELD_FN_STACK; + else + hist_field->fn_num = HIST_FIELD_FN_NOP; + hist_field->size = HIST_STACKTRACE_SIZE; + hist_field->type = kstrdup_const("unsigned long[]", GFP_KERNEL); + if (!hist_field->type) + goto free; goto out; } @@ -2011,6 +2028,13 @@ static struct hist_field *create_hist_field(struct hist_trigger_data *hist_data, goto out; } + if (flags & HIST_FIELD_FL_COMM) { + hist_field->fn_num = HIST_FIELD_FN_COMM; + hist_field->size = MAX_FILTER_STR_VAL; + hist_field->type = "char[]"; + goto out; + } + if (WARN_ON_ONCE(!field)) goto out; @@ -2312,6 +2336,8 @@ parse_field(struct hist_trigger_data *hist_data, struct trace_event_file *file, *flags |= HIST_FIELD_FL_EXECNAME; else if (strcmp(modifier, "syscall") == 0) *flags |= HIST_FIELD_FL_SYSCALL; + else if (strcmp(modifier, "stacktrace") == 0) + *flags |= HIST_FIELD_FL_STACKTRACE; else if (strcmp(modifier, "log2") == 0) *flags |= HIST_FIELD_FL_LOG2; else if (strcmp(modifier, "usecs") == 0) @@ -2351,20 +2377,30 @@ parse_field(struct hist_trigger_data *hist_data, struct trace_event_file *file, hist_data->enable_timestamps = true; if (*flags & HIST_FIELD_FL_TIMESTAMP_USECS) hist_data->attrs->ts_in_usecs = true; - } else if (strcmp(field_name, "common_cpu") == 0) + } else if (strcmp(field_name, "common_stacktrace") == 0) { + *flags |= HIST_FIELD_FL_STACKTRACE; + } else if (strcmp(field_name, "common_cpu") == 0) { *flags |= HIST_FIELD_FL_CPU; - else if (strcmp(field_name, "hitcount") == 0) + } else if (strcmp(field_name, "common_comm") == 0) { + *flags |= HIST_FIELD_FL_COMM | HIST_FIELD_FL_STRING; + } else if (strcmp(field_name, "hitcount") == 0) *flags |= HIST_FIELD_FL_HITCOUNT; else { field = trace_find_event_field(file->event_call, field_name); if (!field || !field->size) { /* * For backward compatibility, if field_name - * was "cpu", then we treat this the same as - * common_cpu. This also works for "CPU". + * was "cpu" or "stacktrace", then we treat this + * the same as common_cpu and common_stacktrace + * respectively. This also works for "CPU", and + * "STACKTRACE". */ if (field && field->filter_type == FILTER_CPU) { *flags |= HIST_FIELD_FL_CPU; + } else if (field && field->filter_type == FILTER_STACKTRACE) { + *flags |= HIST_FIELD_FL_STACKTRACE; + } else if (field && field->filter_type == FILTER_COMM) { + *flags |= HIST_FIELD_FL_COMM | HIST_FIELD_FL_STRING; } else { hist_err(tr, HIST_ERR_FIELD_NOT_FOUND, errpos(field_name)); @@ -3111,6 +3147,9 @@ static inline void __update_field_vars(struct tracing_map_elt *elt, unsigned int i, j, var_idx; u64 var_val; + /* Make sure stacktrace can fit in the string variable length */ + BUILD_BUG_ON((HIST_STACKTRACE_DEPTH + 1) * sizeof(long) >= STR_VAR_LEN_MAX); + for (i = 0, j = field_var_str_start; i < n_field_vars; i++) { struct field_var *field_var = field_vars[i]; struct hist_field *var = field_var->var; @@ -3119,13 +3158,26 @@ static inline void __update_field_vars(struct tracing_map_elt *elt, var_val = hist_fn_call(val, elt, buffer, rbe, rec); var_idx = var->var.idx; - if (val->flags & HIST_FIELD_FL_STRING) { + if (val->flags & (HIST_FIELD_FL_STRING | + HIST_FIELD_FL_STACKTRACE)) { char *str = elt_data->field_var_str[j++]; char *val_str = (char *)(uintptr_t)var_val; unsigned int size; - size = min(val->size, STR_VAR_LEN_MAX); - strscpy(str, val_str, size); + if (val->flags & HIST_FIELD_FL_STRING) { + size = min(val->size, STR_VAR_LEN_MAX); + strscpy(str, val_str, size); + } else { + char *stack_start = str + sizeof(unsigned long); + int e; + + e = stack_trace_save((void *)stack_start, + HIST_STACKTRACE_DEPTH, + HIST_STACKTRACE_SKIP); + if (e < HIST_STACKTRACE_DEPTH - 1) + ((unsigned long *)stack_start)[e] = 0; + *((unsigned long *)str) = e; + } var_val = (u64)(uintptr_t)str; } tracing_map_set_var(elt, var_idx, var_val); @@ -3220,14 +3272,16 @@ static struct field_var *create_field_var(struct hist_trigger_data *hist_data, var = create_var(hist_data, file, field_name, val->size, val->type); if (IS_ERR(var)) { hist_err(tr, HIST_ERR_VAR_CREATE_FIND_FAIL, errpos(field_name)); - kfree(val); + destroy_hist_field(val, 0); ret = PTR_ERR(var); goto err; } field_var = kzalloc(sizeof(struct field_var), GFP_KERNEL); if (!field_var) { - kfree(val); + destroy_hist_field(val, 0); + kfree_const(var->type); + kfree(var->var.name); kfree(var); ret = -ENOMEM; goto err; @@ -3374,7 +3428,7 @@ static bool cond_snapshot_update(struct trace_array *tr, void *cond_data) elt_data = context->elt->private_data; track_elt_data = track_data->elt.private_data; if (elt_data->comm) - strncpy(track_elt_data->comm, elt_data->comm, TASK_COMM_LEN); + strscpy(track_elt_data->comm, elt_data->comm, TASK_COMM_LEN); track_data->updated = true; @@ -3824,7 +3878,8 @@ static void save_field_var(struct hist_trigger_data *hist_data, { hist_data->field_vars[hist_data->n_field_vars++] = field_var; - if (field_var->val->flags & HIST_FIELD_FL_STRING) + /* Stack traces are saved in the string storage too */ + if (field_var->val->flags & (HIST_FIELD_FL_STRING | HIST_FIELD_FL_STACKTRACE)) hist_data->n_field_var_str++; } @@ -3849,6 +3904,9 @@ static int check_synth_field(struct synth_event *event, && field->is_dynamic) return 0; + if (strstr(hist_field->type, "long[") && field->is_stack) + return 0; + if (strcmp(field->type, hist_field->type) != 0) { if (field->size != hist_field->size || (!field->is_string && field->is_signed != hist_field->is_signed)) @@ -4103,7 +4161,8 @@ static int action_create(struct hist_trigger_data *hist_data, } hist_data->save_vars[hist_data->n_save_vars++] = field_var; - if (field_var->val->flags & HIST_FIELD_FL_STRING) + if (field_var->val->flags & + (HIST_FIELD_FL_STRING | HIST_FIELD_FL_STACKTRACE)) hist_data->n_save_var_str++; kfree(param); } @@ -4199,6 +4258,21 @@ static int __create_val_field(struct hist_trigger_data *hist_data, goto out; } + /* values and variables should not have some modifiers */ + if (hist_field->flags & HIST_FIELD_FL_VAR) { + /* Variable */ + if (hist_field->flags & (HIST_FIELD_FL_GRAPH | HIST_FIELD_FL_PERCENT | + HIST_FIELD_FL_BUCKET | HIST_FIELD_FL_LOG2)) + goto err; + } else { + /* Value */ + if (hist_field->flags & (HIST_FIELD_FL_GRAPH | HIST_FIELD_FL_PERCENT | + HIST_FIELD_FL_BUCKET | HIST_FIELD_FL_LOG2 | + HIST_FIELD_FL_SYM | HIST_FIELD_FL_SYM_OFFSET | + HIST_FIELD_FL_SYSCALL | HIST_FIELD_FL_STACKTRACE)) + goto err; + } + hist_data->fields[val_idx] = hist_field; ++hist_data->n_vals; @@ -4208,6 +4282,9 @@ static int __create_val_field(struct hist_trigger_data *hist_data, ret = -EINVAL; out: return ret; + err: + hist_err(file->tr, HIST_ERR_BAD_FIELD_MODIFIER, errpos(field_str)); + return -EINVAL; } static int create_val_field(struct hist_trigger_data *hist_data, @@ -4242,6 +4319,19 @@ static u64 hist_field_execname(struct hist_field *hist_field, return (u64)(unsigned long)(elt_data->comm); } +static u64 hist_field_stack(struct hist_field *hist_field, + struct tracing_map_elt *elt, + struct trace_buffer *buffer, + struct ring_buffer_event *rbe, + void *event) +{ + u32 str_item = *(u32 *)(event + hist_field->field->offset); + int str_loc = str_item & 0xffff; + char *addr = (char *)(event + str_loc); + + return (u64)(unsigned long)addr; +} + static u64 hist_fn_call(struct hist_field *hist_field, struct tracing_map_elt *elt, struct trace_buffer *buffer, @@ -4263,6 +4353,8 @@ static u64 hist_fn_call(struct hist_field *hist_field, return hist_field_timestamp(hist_field, elt, buffer, rbe, event); case HIST_FIELD_FN_CPU: return hist_field_cpu(hist_field, elt, buffer, rbe, event); + case HIST_FIELD_FN_COMM: + return hist_field_comm(hist_field, elt, buffer, rbe, event); case HIST_FIELD_FN_STRING: return hist_field_string(hist_field, elt, buffer, rbe, event); case HIST_FIELD_FN_DYNSTRING: @@ -4305,6 +4397,8 @@ static u64 hist_fn_call(struct hist_field *hist_field, return div_by_mult_and_shift(hist_field, elt, buffer, rbe, event); case HIST_FIELD_FN_EXECNAME: return hist_field_execname(hist_field, elt, buffer, rbe, event); + case HIST_FIELD_FN_STACK: + return hist_field_stack(hist_field, elt, buffer, rbe, event); default: return 0; } @@ -4351,7 +4445,8 @@ static int create_var_field(struct hist_trigger_data *hist_data, if (!ret && hist_data->fields[val_idx]->flags & HIST_FIELD_FL_EXECNAME) update_var_execname(hist_data->fields[val_idx]); - if (!ret && hist_data->fields[val_idx]->flags & HIST_FIELD_FL_STRING) + if (!ret && hist_data->fields[val_idx]->flags & + (HIST_FIELD_FL_STRING | HIST_FIELD_FL_STACKTRACE)) hist_data->fields[val_idx]->var_str_idx = hist_data->n_var_str++; return ret; @@ -4735,36 +4830,35 @@ static int parse_actions(struct hist_trigger_data *hist_data) int len; for (i = 0; i < hist_data->attrs->n_actions; i++) { + enum handler_id hid = 0; + char *action_str; + str = hist_data->attrs->action_str[i]; - if ((len = str_has_prefix(str, "onmatch("))) { - char *action_str = str + len; + if ((len = str_has_prefix(str, "onmatch("))) + hid = HANDLER_ONMATCH; + else if ((len = str_has_prefix(str, "onmax("))) + hid = HANDLER_ONMAX; + else if ((len = str_has_prefix(str, "onchange("))) + hid = HANDLER_ONCHANGE; - data = onmatch_parse(tr, action_str); - if (IS_ERR(data)) { - ret = PTR_ERR(data); - break; - } - } else if ((len = str_has_prefix(str, "onmax("))) { - char *action_str = str + len; + action_str = str + len; - data = track_data_parse(hist_data, action_str, - HANDLER_ONMAX); - if (IS_ERR(data)) { - ret = PTR_ERR(data); - break; - } - } else if ((len = str_has_prefix(str, "onchange("))) { - char *action_str = str + len; + switch (hid) { + case HANDLER_ONMATCH: + data = onmatch_parse(tr, action_str); + break; + case HANDLER_ONMAX: + case HANDLER_ONCHANGE: + data = track_data_parse(hist_data, action_str, hid); + break; + default: + data = ERR_PTR(-EINVAL); + break; + } - data = track_data_parse(hist_data, action_str, - HANDLER_ONCHANGE); - if (IS_ERR(data)) { - ret = PTR_ERR(data); - break; - } - } else { - ret = -EINVAL; + if (IS_ERR(data)) { + ret = PTR_ERR(data); break; } @@ -5092,7 +5186,8 @@ static void hist_trigger_elt_update(struct hist_trigger_data *hist_data, if (hist_field->flags & HIST_FIELD_FL_VAR) { var_idx = hist_field->var.idx; - if (hist_field->flags & HIST_FIELD_FL_STRING) { + if (hist_field->flags & + (HIST_FIELD_FL_STRING | HIST_FIELD_FL_STACKTRACE)) { unsigned int str_start, var_str_idx, idx; char *str, *val_str; unsigned int size; @@ -5105,9 +5200,20 @@ static void hist_trigger_elt_update(struct hist_trigger_data *hist_data, str = elt_data->field_var_str[idx]; val_str = (char *)(uintptr_t)hist_val; - size = min(hist_field->size, STR_VAR_LEN_MAX); - strscpy(str, val_str, size); - + if (hist_field->flags & HIST_FIELD_FL_STRING) { + size = min(hist_field->size, STR_VAR_LEN_MAX); + strscpy(str, val_str, size); + } else { + char *stack_start = str + sizeof(unsigned long); + int e; + + e = stack_trace_save((void *)stack_start, + HIST_STACKTRACE_DEPTH, + HIST_STACKTRACE_SKIP); + if (e < HIST_STACKTRACE_DEPTH - 1) + ((unsigned long *)stack_start)[e] = 0; + *((unsigned long *)str) = e; + } hist_val = (u64)(uintptr_t)str; } tracing_map_set_var(elt, var_idx, hist_val); @@ -5134,22 +5240,25 @@ static inline void add_to_key(char *compound_key, void *key, size_t size = key_field->size; if (key_field->flags & HIST_FIELD_FL_STRING) { - struct ftrace_event_field *field; - field = key_field->field; - if (field->filter_type == FILTER_DYN_STRING || - field->filter_type == FILTER_RDYN_STRING) - size = *(u32 *)(rec + field->offset) >> 16; - else if (field->filter_type == FILTER_STATIC_STRING) - size = field->size; + if (key_field->flags & HIST_FIELD_FL_COMM) { + size = strlen((char *)key); + } else { + struct ftrace_event_field *field; + + field = key_field->field; + if (field->filter_type == FILTER_DYN_STRING || + field->filter_type == FILTER_RDYN_STRING) + size = *(u32 *)(rec + field->offset) >> 16; + else if (field->filter_type == FILTER_STATIC_STRING) + size = field->size; + } /* ensure NULL-termination */ if (size > key_field->size - 1) size = key_field->size - 1; - - strncpy(compound_key + key_field->offset, (char *)key, size); - } else - memcpy(compound_key + key_field->offset, key, size); + } + memcpy(compound_key + key_field->offset, key, size); } static void @@ -5168,17 +5277,94 @@ hist_trigger_actions(struct hist_trigger_data *hist_data, } } +/* + * The hist_pad structure is used to save information to create + * a histogram from the histogram trigger. It's too big to store + * on the stack, so when the histogram trigger is initialized + * a percpu array of 4 hist_pad structures is allocated. + * This will cover every context from normal, softirq, irq and NMI + * in the very unlikely event that a trigger happens at each of + * these contexts and interrupts a currently active trigger. + */ +struct hist_pad { + unsigned long entries[HIST_STACKTRACE_DEPTH]; + u64 var_ref_vals[TRACING_MAP_VARS_MAX]; + char compound_key[HIST_KEY_SIZE_MAX]; +}; + +static struct hist_pad __percpu *hist_pads; +static DEFINE_PER_CPU(int, hist_pad_cnt); +static refcount_t hist_pad_ref; + +/* One hist_pad for every context (normal, softirq, irq, NMI) */ +#define MAX_HIST_CNT 4 + +static int alloc_hist_pad(void) +{ + lockdep_assert_held(&event_mutex); + + if (refcount_read(&hist_pad_ref)) { + refcount_inc(&hist_pad_ref); + return 0; + } + + hist_pads = __alloc_percpu(sizeof(struct hist_pad) * MAX_HIST_CNT, + __alignof__(struct hist_pad)); + if (!hist_pads) + return -ENOMEM; + + refcount_set(&hist_pad_ref, 1); + return 0; +} + +static void free_hist_pad(void) +{ + lockdep_assert_held(&event_mutex); + + if (!refcount_dec_and_test(&hist_pad_ref)) + return; + + free_percpu(hist_pads); + hist_pads = NULL; +} + +static struct hist_pad *get_hist_pad(void) +{ + struct hist_pad *hist_pad; + int cnt; + + if (WARN_ON_ONCE(!hist_pads)) + return NULL; + + preempt_disable(); + + hist_pad = per_cpu_ptr(hist_pads, smp_processor_id()); + + if (this_cpu_read(hist_pad_cnt) == MAX_HIST_CNT) { + preempt_enable(); + return NULL; + } + + cnt = this_cpu_inc_return(hist_pad_cnt) - 1; + + return &hist_pad[cnt]; +} + +static void put_hist_pad(void) +{ + this_cpu_dec(hist_pad_cnt); + preempt_enable(); +} + static void event_hist_trigger(struct event_trigger_data *data, struct trace_buffer *buffer, void *rec, struct ring_buffer_event *rbe) { struct hist_trigger_data *hist_data = data->private_data; bool use_compound_key = (hist_data->n_keys > 1); - unsigned long entries[HIST_STACKTRACE_DEPTH]; - u64 var_ref_vals[TRACING_MAP_VARS_MAX]; - char compound_key[HIST_KEY_SIZE_MAX]; struct tracing_map_elt *elt = NULL; struct hist_field *key_field; + struct hist_pad *hist_pad; u64 field_contents; void *key = NULL; unsigned int i; @@ -5186,15 +5372,30 @@ static void event_hist_trigger(struct event_trigger_data *data, if (unlikely(!rbe)) return; - memset(compound_key, 0, hist_data->key_size); + hist_pad = get_hist_pad(); + if (!hist_pad) + return; + + memset(hist_pad->compound_key, 0, hist_data->key_size); for_each_hist_key_field(i, hist_data) { key_field = hist_data->fields[i]; if (key_field->flags & HIST_FIELD_FL_STACKTRACE) { + unsigned long *entries = hist_pad->entries; + memset(entries, 0, HIST_STACKTRACE_SIZE); - stack_trace_save(entries, HIST_STACKTRACE_DEPTH, - HIST_STACKTRACE_SKIP); + if (key_field->field) { + unsigned long *stack, n_entries; + + field_contents = hist_fn_call(key_field, elt, buffer, rbe, rec); + stack = (unsigned long *)(long)field_contents; + n_entries = *stack; + memcpy(entries, ++stack, n_entries * sizeof(unsigned long)); + } else { + stack_trace_save(entries, HIST_STACKTRACE_DEPTH, + HIST_STACKTRACE_SKIP); + } key = entries; } else { field_contents = hist_fn_call(key_field, elt, buffer, rbe, rec); @@ -5206,24 +5407,31 @@ static void event_hist_trigger(struct event_trigger_data *data, } if (use_compound_key) - add_to_key(compound_key, key, key_field, rec); + add_to_key(hist_pad->compound_key, key, key_field, rec); } if (use_compound_key) - key = compound_key; + key = hist_pad->compound_key; if (hist_data->n_var_refs && - !resolve_var_refs(hist_data, key, var_ref_vals, false)) - return; + !resolve_var_refs(hist_data, key, hist_pad->var_ref_vals, false)) + goto out; elt = tracing_map_insert(hist_data->map, key); if (!elt) - return; + goto out; + + hist_trigger_elt_update(hist_data, elt, buffer, rec, rbe, hist_pad->var_ref_vals); - hist_trigger_elt_update(hist_data, elt, buffer, rec, rbe, var_ref_vals); + if (resolve_var_refs(hist_data, key, hist_pad->var_ref_vals, true)) { + hist_trigger_actions(hist_data, elt, buffer, rec, rbe, + key, hist_pad->var_ref_vals); + } + + hist_poll_wakeup(); - if (resolve_var_refs(hist_data, key, var_ref_vals, true)) - hist_trigger_actions(hist_data, elt, buffer, rec, rbe, key, var_ref_vals); + out: + put_hist_pad(); } static void hist_trigger_stacktrace_print(struct seq_file *m, @@ -5297,7 +5505,10 @@ static void hist_trigger_print_key(struct seq_file *m, seq_printf(m, "%s: %-30s[%3llu]", field_name, syscall_name, uval); } else if (key_field->flags & HIST_FIELD_FL_STACKTRACE) { - seq_puts(m, "stacktrace:\n"); + if (key_field->field) + seq_printf(m, "%s.stacktrace", key_field->field->name); + else + seq_puts(m, "common_stacktrace:\n"); hist_trigger_stacktrace_print(m, key + key_field->offset, HIST_STACKTRACE_DEPTH); @@ -5485,7 +5696,7 @@ static void hist_trigger_show(struct seq_file *m, seq_puts(m, "\n\n"); seq_puts(m, "# event histogram\n#\n# trigger info: "); - data->ops->print(m, data); + data->cmd_ops->print(m, data); seq_puts(m, "#\n\n"); hist_data = data->private_data; @@ -5500,47 +5711,137 @@ static void hist_trigger_show(struct seq_file *m, n_entries, (u64)atomic64_read(&hist_data->map->drops)); } +struct hist_file_data { + struct file *file; + u64 last_read; + u64 last_act; +}; + +static u64 get_hist_hit_count(struct trace_event_file *event_file) +{ + struct hist_trigger_data *hist_data; + struct event_trigger_data *data; + u64 ret = 0; + + list_for_each_entry(data, &event_file->triggers, list) { + if (data->cmd_ops->trigger_type == ETT_EVENT_HIST) { + hist_data = data->private_data; + ret += atomic64_read(&hist_data->map->hits); + } + } + return ret; +} + static int hist_show(struct seq_file *m, void *v) { + struct hist_file_data *hist_file = m->private; struct event_trigger_data *data; struct trace_event_file *event_file; - int n = 0, ret = 0; + int n = 0; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); - event_file = event_file_data(m->private); - if (unlikely(!event_file)) { - ret = -ENODEV; - goto out_unlock; - } + event_file = event_file_file(hist_file->file); + if (unlikely(!event_file)) + return -ENODEV; list_for_each_entry(data, &event_file->triggers, list) { if (data->cmd_ops->trigger_type == ETT_EVENT_HIST) hist_trigger_show(m, data, n++); } + hist_file->last_read = get_hist_hit_count(event_file); + /* + * Update last_act too so that poll()/POLLPRI can wait for the next + * event after any syscall on hist file. + */ + hist_file->last_act = hist_file->last_read; + + return 0; +} + +static __poll_t event_hist_poll(struct file *file, struct poll_table_struct *wait) +{ + struct trace_event_file *event_file; + struct seq_file *m = file->private_data; + struct hist_file_data *hist_file = m->private; + __poll_t ret = 0; + u64 cnt; + + guard(mutex)(&event_mutex); + + event_file = event_file_data(file); + if (!event_file) + return EPOLLERR; - out_unlock: - mutex_unlock(&event_mutex); + hist_poll_wait(file, wait); + + cnt = get_hist_hit_count(event_file); + if (hist_file->last_read != cnt) + ret |= EPOLLIN | EPOLLRDNORM; + if (hist_file->last_act != cnt) { + hist_file->last_act = cnt; + ret |= EPOLLPRI; + } return ret; } +static int event_hist_release(struct inode *inode, struct file *file) +{ + struct seq_file *m = file->private_data; + struct hist_file_data *hist_file = m->private; + + kfree(hist_file); + return tracing_single_release_file_tr(inode, file); +} + static int event_hist_open(struct inode *inode, struct file *file) { + struct trace_event_file *event_file; + struct hist_file_data *hist_file; int ret; - ret = security_locked_down(LOCKDOWN_TRACEFS); + ret = tracing_open_file_tr(inode, file); if (ret) return ret; - return single_open(file, hist_show, file); + guard(mutex)(&event_mutex); + + event_file = event_file_data(file); + if (!event_file) { + ret = -ENODEV; + goto err; + } + + hist_file = kzalloc(sizeof(*hist_file), GFP_KERNEL); + if (!hist_file) { + ret = -ENOMEM; + goto err; + } + + hist_file->file = file; + hist_file->last_act = get_hist_hit_count(event_file); + + /* Clear private_data to avoid warning in single_open() */ + file->private_data = NULL; + ret = single_open(file, hist_show, hist_file); + if (ret) { + kfree(hist_file); + goto err; + } + + return 0; +err: + tracing_release_file_tr(inode, file); + return ret; } const struct file_operations event_hist_fops = { .open = event_hist_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = event_hist_release, + .poll = event_hist_poll, }; #ifdef CONFIG_HIST_TRIGGERS_DEBUG @@ -5717,7 +6018,7 @@ static void hist_trigger_debug_show(struct seq_file *m, seq_puts(m, "\n\n"); seq_puts(m, "# event histogram\n#\n# trigger info: "); - data->ops->print(m, data); + data->cmd_ops->print(m, data); seq_puts(m, "#\n\n"); hist_data = data->private_data; @@ -5781,43 +6082,42 @@ static int hist_debug_show(struct seq_file *m, void *v) { struct event_trigger_data *data; struct trace_event_file *event_file; - int n = 0, ret = 0; + int n = 0; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); - event_file = event_file_data(m->private); - if (unlikely(!event_file)) { - ret = -ENODEV; - goto out_unlock; - } + event_file = event_file_file(m->private); + if (unlikely(!event_file)) + return -ENODEV; list_for_each_entry(data, &event_file->triggers, list) { if (data->cmd_ops->trigger_type == ETT_EVENT_HIST) hist_trigger_debug_show(m, data, n++); } - - out_unlock: - mutex_unlock(&event_mutex); - - return ret; + return 0; } static int event_hist_debug_open(struct inode *inode, struct file *file) { int ret; - ret = security_locked_down(LOCKDOWN_TRACEFS); + ret = tracing_open_file_tr(inode, file); if (ret) return ret; - return single_open(file, hist_debug_show, file); + /* Clear private_data to avoid warning in single_open() */ + file->private_data = NULL; + ret = single_open(file, hist_debug_show, file); + if (ret) + tracing_release_file_tr(inode, file); + return ret; } const struct file_operations event_hist_debug_fops = { .open = event_hist_debug_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = tracing_single_release_file_tr, }; #endif @@ -5830,6 +6130,8 @@ static void hist_field_print(struct seq_file *m, struct hist_field *hist_field) if (hist_field->flags & HIST_FIELD_FL_CPU) seq_puts(m, "common_cpu"); + if (hist_field->flags & HIST_FIELD_FL_COMM) + seq_puts(m, "common_comm"); else if (hist_field->flags & HIST_FIELD_FL_CONST) seq_printf(m, "%llu", hist_field->constant); else if (field_name) { @@ -5842,7 +6144,8 @@ static void hist_field_print(struct seq_file *m, struct hist_field *hist_field) if (hist_field->flags) { if (!(hist_field->flags & HIST_FIELD_FL_VAR_REF) && - !(hist_field->flags & HIST_FIELD_FL_EXPR)) { + !(hist_field->flags & HIST_FIELD_FL_EXPR) && + !(hist_field->flags & HIST_FIELD_FL_STACKTRACE)) { const char *flags = get_hist_field_flags(hist_field); if (flags) @@ -5875,9 +6178,12 @@ static int event_hist_trigger_print(struct seq_file *m, if (i > hist_data->n_vals) seq_puts(m, ","); - if (field->flags & HIST_FIELD_FL_STACKTRACE) - seq_puts(m, "stacktrace"); - else + if (field->flags & HIST_FIELD_FL_STACKTRACE) { + if (field->field) + seq_printf(m, "%s.stacktrace", field->field->name); + else + seq_puts(m, "common_stacktrace"); + } else hist_field_print(m, field); } @@ -5972,6 +6278,9 @@ static int event_hist_trigger_init(struct event_trigger_data *data) { struct hist_trigger_data *hist_data = data->private_data; + if (alloc_hist_pad() < 0) + return -ENOMEM; + if (!data->ref && hist_data->attrs->name) save_named_trigger(hist_data->attrs->name, data); @@ -6016,24 +6325,24 @@ static void event_hist_trigger_free(struct event_trigger_data *data) destroy_hist_data(hist_data); } + free_hist_pad(); } -static struct event_trigger_ops event_hist_trigger_ops = { - .trigger = event_hist_trigger, - .print = event_hist_trigger_print, - .init = event_hist_trigger_init, - .free = event_hist_trigger_free, -}; - static int event_hist_trigger_named_init(struct event_trigger_data *data) { + int ret; + data->ref++; save_named_trigger(data->named_data->name, data); - event_hist_trigger_init(data->named_data); + ret = event_hist_trigger_init(data->named_data); + if (ret < 0) { + kfree(data->cmd_ops); + data->cmd_ops = &trigger_hist_cmd; + } - return 0; + return ret; } static void event_hist_trigger_named_free(struct event_trigger_data *data) @@ -6045,24 +6354,14 @@ static void event_hist_trigger_named_free(struct event_trigger_data *data) data->ref--; if (!data->ref) { + struct event_command *cmd_ops = data->cmd_ops; + del_named_trigger(data); trigger_data_free(data); + kfree(cmd_ops); } } -static struct event_trigger_ops event_hist_trigger_named_ops = { - .trigger = event_hist_trigger, - .print = event_hist_trigger_print, - .init = event_hist_trigger_named_init, - .free = event_hist_trigger_named_free, -}; - -static struct event_trigger_ops *event_hist_get_trigger_ops(char *cmd, - char *param) -{ - return &event_hist_trigger_ops; -} - static void hist_clear(struct event_trigger_data *data) { struct hist_trigger_data *hist_data = data->private_data; @@ -6256,13 +6555,24 @@ static int hist_register_trigger(char *glob, data->paused = true; if (named_data) { + struct event_command *cmd_ops; + data->private_data = named_data->private_data; set_named_trigger_data(data, named_data); - data->ops = &event_hist_trigger_named_ops; + /* Copy the command ops and update some of the functions */ + cmd_ops = kmalloc(sizeof(*cmd_ops), GFP_KERNEL); + if (!cmd_ops) { + ret = -ENOMEM; + goto out; + } + *cmd_ops = *data->cmd_ops; + cmd_ops->init = event_hist_trigger_named_init; + cmd_ops->free = event_hist_trigger_named_free; + data->cmd_ops = cmd_ops; } - if (data->ops->init) { - ret = data->ops->init(data); + if (data->cmd_ops->init) { + ret = data->cmd_ops->init(data); if (ret < 0) goto out; } @@ -6376,8 +6686,8 @@ static void hist_unregister_trigger(char *glob, } } - if (test && test->ops->free) - test->ops->free(test); + if (test && test->cmd_ops->free) + test->cmd_ops->free(test); if (hist_data->enable_timestamps) { if (!hist_data->remove || test) @@ -6429,8 +6739,8 @@ static void hist_unreg_all(struct trace_event_file *file) update_cond_flag(file); if (hist_data->enable_timestamps) tracing_set_filter_buffering(file->tr, false); - if (test->ops->free) - test->ops->free(test); + if (test->cmd_ops->free) + test->cmd_ops->free(test); } } } @@ -6520,7 +6830,7 @@ static int event_hist_trigger_parse(struct event_command *cmd_ops, return PTR_ERR(hist_data); } - trigger_data = event_trigger_alloc(cmd_ops, cmd, param, hist_data); + trigger_data = trigger_data_alloc(cmd_ops, cmd, param, hist_data); if (!trigger_data) { ret = -ENOMEM; goto out_free; @@ -6551,24 +6861,27 @@ static int event_hist_trigger_parse(struct event_command *cmd_ops, if (existing_hist_update_only(glob, trigger_data, file)) goto out_free; - ret = event_trigger_register(cmd_ops, file, glob, trigger_data); - if (ret < 0) - goto out_free; + if (!get_named_trigger_data(trigger_data)) { - if (get_named_trigger_data(trigger_data)) - goto enable; + ret = create_actions(hist_data); + if (ret) + goto out_free; - if (has_hist_vars(hist_data)) - save_hist_vars(hist_data); + if (has_hist_vars(hist_data) || hist_data->n_var_refs) { + ret = save_hist_vars(hist_data); + if (ret) + goto out_free; + } - ret = create_actions(hist_data); - if (ret) - goto out_unreg; + ret = tracing_map_init(hist_data->map); + if (ret) + goto out_free; + } + + ret = event_trigger_register(cmd_ops, file, glob, trigger_data); + if (ret < 0) + goto out_free; - ret = tracing_map_init(hist_data->map); - if (ret) - goto out_unreg; -enable: ret = hist_trigger_enable(trigger_data, file); if (ret) goto out_unreg; @@ -6603,8 +6916,11 @@ static struct event_command trigger_hist_cmd = { .reg = hist_register_trigger, .unreg = hist_unregister_trigger, .unreg_all = hist_unreg_all, - .get_trigger_ops = event_hist_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = event_hist_trigger, + .print = event_hist_trigger_print, + .init = event_hist_trigger_init, + .free = event_hist_trigger_free, }; __init int register_trigger_hist_cmd(void) @@ -6636,66 +6952,6 @@ hist_enable_trigger(struct event_trigger_data *data, } } -static void -hist_enable_count_trigger(struct event_trigger_data *data, - struct trace_buffer *buffer, void *rec, - struct ring_buffer_event *event) -{ - if (!data->count) - return; - - if (data->count != -1) - (data->count)--; - - hist_enable_trigger(data, buffer, rec, event); -} - -static struct event_trigger_ops hist_enable_trigger_ops = { - .trigger = hist_enable_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops hist_enable_count_trigger_ops = { - .trigger = hist_enable_count_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops hist_disable_trigger_ops = { - .trigger = hist_enable_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops hist_disable_count_trigger_ops = { - .trigger = hist_enable_count_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops * -hist_enable_get_trigger_ops(char *cmd, char *param) -{ - struct event_trigger_ops *ops; - bool enable; - - enable = (strcmp(cmd, ENABLE_HIST_STR) == 0); - - if (enable) - ops = param ? &hist_enable_count_trigger_ops : - &hist_enable_trigger_ops; - else - ops = param ? &hist_disable_count_trigger_ops : - &hist_disable_trigger_ops; - - return ops; -} - static void hist_enable_unreg_all(struct trace_event_file *file) { struct event_trigger_data *test, *n; @@ -6705,8 +6961,8 @@ static void hist_enable_unreg_all(struct trace_event_file *file) list_del_rcu(&test->list); update_cond_flag(file); trace_event_trigger_enable_disable(file, 0); - if (test->ops->free) - test->ops->free(test); + if (test->cmd_ops->free) + test->cmd_ops->free(test); } } } @@ -6718,8 +6974,12 @@ static struct event_command trigger_hist_enable_cmd = { .reg = event_enable_register_trigger, .unreg = event_enable_unregister_trigger, .unreg_all = hist_enable_unreg_all, - .get_trigger_ops = hist_enable_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = hist_enable_trigger, + .count_func = event_trigger_count, + .print = event_enable_trigger_print, + .init = event_trigger_init, + .free = event_enable_trigger_free, }; static struct event_command trigger_hist_disable_cmd = { @@ -6729,8 +6989,12 @@ static struct event_command trigger_hist_disable_cmd = { .reg = event_enable_register_trigger, .unreg = event_enable_unregister_trigger, .unreg_all = hist_enable_unreg_all, - .get_trigger_ops = hist_enable_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = hist_enable_trigger, + .count_func = event_trigger_count, + .print = event_enable_trigger_print, + .init = event_trigger_init, + .free = event_enable_trigger_free, }; static __init void unregister_trigger_hist_enable_disable_cmds(void) diff --git a/kernel/trace/trace_events_inject.c b/kernel/trace/trace_events_inject.c index d6b4935a78c0..a8f076809db4 100644 --- a/kernel/trace/trace_events_inject.c +++ b/kernel/trace/trace_events_inject.c @@ -217,7 +217,7 @@ static int parse_entry(char *str, struct trace_event_call *call, void **pentry) char *addr = (char *)(unsigned long) val; if (field->filter_type == FILTER_STATIC_STRING) { - strlcpy(entry + field->offset, addr, field->size); + strscpy(entry + field->offset, addr, field->size); } else if (field->filter_type == FILTER_DYN_STRING || field->filter_type == FILTER_RDYN_STRING) { int str_len = strlen(addr) + 1; @@ -232,7 +232,7 @@ static int parse_entry(char *str, struct trace_event_call *call, void **pentry) } entry = *pentry; - strlcpy(entry + (entry_size - str_len), addr, str_len); + strscpy(entry + (entry_size - str_len), addr, str_len); str_item = (u32 *)(entry + field->offset); if (field->filter_type == FILTER_RDYN_STRING) str_loc -= field->offset + field->size; @@ -299,7 +299,7 @@ event_inject_write(struct file *filp, const char __user *ubuf, size_t cnt, strim(buf); mutex_lock(&event_mutex); - file = event_file_data(filp); + file = event_file_file(filp); if (file) { call = file->event_call; size = parse_entry(buf, call, &entry); @@ -328,7 +328,8 @@ event_inject_read(struct file *file, char __user *buf, size_t size, } const struct file_operations event_inject_fops = { - .open = tracing_open_generic, + .open = tracing_open_file_tr, .read = event_inject_read, .write = event_inject_write, + .release = tracing_release_file_tr, }; diff --git a/kernel/trace/trace_events_synth.c b/kernel/trace/trace_events_synth.c index 67592eed0be8..4554c458b78c 100644 --- a/kernel/trace/trace_events_synth.c +++ b/kernel/trace/trace_events_synth.c @@ -44,10 +44,12 @@ enum { ERRORS }; static const char *err_text[] = { ERRORS }; +static DEFINE_MUTEX(lastcmd_mutex); static char *last_cmd; static int errpos(const char *str) { + guard(mutex)(&lastcmd_mutex); if (!str || !last_cmd) return 0; @@ -59,13 +61,15 @@ static void last_cmd_set(const char *str) if (!str) return; + mutex_lock(&lastcmd_mutex); kfree(last_cmd); - last_cmd = kstrdup(str, GFP_KERNEL); + mutex_unlock(&lastcmd_mutex); } static void synth_err(u8 err_type, u16 err_pos) { + guard(mutex)(&lastcmd_mutex); if (!last_cmd) return; @@ -116,7 +120,7 @@ static bool synth_event_match(const char *system, const char *event, struct synth_trace_event { struct trace_entry ent; - u64 fields[]; + union trace_synth_field fields[]; }; static int synth_event_define_fields(struct trace_event_call *call) @@ -173,6 +177,14 @@ static int synth_field_is_string(char *type) return false; } +static int synth_field_is_stack(char *type) +{ + if (strstr(type, "long[") != NULL) + return true; + + return false; +} + static int synth_field_string_size(char *type) { char buf[4], *end, *start; @@ -195,7 +207,7 @@ static int synth_field_string_size(char *type) if (len == 0) return 0; /* variable-length string */ - strncpy(buf, start, len); + memcpy(buf, start, len); buf[len] = '\0'; err = kstrtouint(buf, 0, &size); @@ -248,6 +260,8 @@ static int synth_field_size(char *type) size = sizeof(gfp_t); else if (synth_field_is_string(type)) size = synth_field_string_size(type); + else if (synth_field_is_stack(type)) + size = 0; return size; } @@ -291,30 +305,32 @@ static const char *synth_field_fmt(char *type) else if (strcmp(type, "gfp_t") == 0) fmt = "%x"; else if (synth_field_is_string(type)) - fmt = "%.*s"; + fmt = "%s"; + else if (synth_field_is_stack(type)) + fmt = "%s"; return fmt; } static void print_synth_event_num_val(struct trace_seq *s, char *print_fmt, char *name, - int size, u64 val, char *space) + int size, union trace_synth_field *val, char *space) { switch (size) { case 1: - trace_seq_printf(s, print_fmt, name, (u8)val, space); + trace_seq_printf(s, print_fmt, name, val->as_u8, space); break; case 2: - trace_seq_printf(s, print_fmt, name, (u16)val, space); + trace_seq_printf(s, print_fmt, name, val->as_u16, space); break; case 4: - trace_seq_printf(s, print_fmt, name, (u32)val, space); + trace_seq_printf(s, print_fmt, name, val->as_u32, space); break; default: - trace_seq_printf(s, print_fmt, name, val, space); + trace_seq_printf(s, print_fmt, name, val->as_u64, space); break; } } @@ -327,7 +343,7 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, struct trace_seq *s = &iter->seq; struct synth_trace_event *entry; struct synth_event *se; - unsigned int i, n_u64; + unsigned int i, j, n_u64; char print_fmt[32]; const char *fmt; @@ -343,7 +359,7 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, fmt = synth_field_fmt(se->fields[i]->type); /* parameter types */ - if (tr && tr->trace_flags & TRACE_ITER_VERBOSE) + if (tr && tr->trace_flags & TRACE_ITER(VERBOSE)) trace_seq_printf(s, "%s ", fmt); snprintf(print_fmt, sizeof(print_fmt), "%%s=%s%%s", fmt); @@ -351,26 +367,26 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, /* parameter values */ if (se->fields[i]->is_string) { if (se->fields[i]->is_dynamic) { - u32 offset, data_offset; - char *str_field; - - offset = (u32)entry->fields[n_u64]; - data_offset = offset & 0xffff; - - str_field = (char *)entry + data_offset; + union trace_synth_field *data = &entry->fields[n_u64]; trace_seq_printf(s, print_fmt, se->fields[i]->name, - STR_VAR_LEN_MAX, - str_field, + (char *)entry + data->as_dynamic.offset, i == se->n_fields - 1 ? "" : " "); n_u64++; } else { trace_seq_printf(s, print_fmt, se->fields[i]->name, - STR_VAR_LEN_MAX, - (char *)&entry->fields[n_u64], + (char *)&entry->fields[n_u64].as_u64, i == se->n_fields - 1 ? "" : " "); n_u64 += STR_VAR_LEN_MAX / sizeof(u64); } + } else if (se->fields[i]->is_stack) { + union trace_synth_field *data = &entry->fields[n_u64]; + unsigned long *p = (void *)entry + data->as_dynamic.offset; + + trace_seq_printf(s, "%s=STACK:\n", se->fields[i]->name); + for (j = 1; j < data->as_dynamic.len / sizeof(long); j++) + trace_seq_printf(s, "=> %pS\n", (void *)p[j]); + n_u64++; } else { struct trace_print_flags __flags[] = { __def_gfpflag_names, {-1, NULL} }; @@ -379,13 +395,13 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, print_synth_event_num_val(s, print_fmt, se->fields[i]->name, se->fields[i]->size, - entry->fields[n_u64], + &entry->fields[n_u64], space); if (strcmp(se->fields[i]->type, "gfp_t") == 0) { trace_seq_puts(s, " ("); trace_print_flags_seq(s, "|", - entry->fields[n_u64], + entry->fields[n_u64].as_u64, __flags); trace_seq_putc(s, ')'); } @@ -414,26 +430,21 @@ static unsigned int trace_string(struct synth_trace_event *entry, int ret; if (is_dynamic) { - u32 data_offset; - - data_offset = offsetof(typeof(*entry), fields); - data_offset += event->n_u64 * sizeof(u64); - data_offset += data_size; + union trace_synth_field *data = &entry->fields[*n_u64]; - len = kern_fetch_store_strlen((unsigned long)str_val); + len = fetch_store_strlen((unsigned long)str_val); + data->as_dynamic.offset = struct_size(entry, fields, event->n_u64) + data_size; + data->as_dynamic.len = len; - data_offset |= len << 16; - *(u32 *)&entry->fields[*n_u64] = data_offset; - - ret = kern_fetch_store_string((unsigned long)str_val, &entry->fields[*n_u64], entry); + ret = fetch_store_string((unsigned long)str_val, &entry->fields[*n_u64], entry); (*n_u64)++; } else { - str_field = (char *)&entry->fields[*n_u64]; + str_field = (char *)&entry->fields[*n_u64].as_u64; #ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE if ((unsigned long)str_val < TASK_SIZE) - ret = strncpy_from_user_nofault(str_field, str_val, STR_VAR_LEN_MAX); + ret = strncpy_from_user_nofault(str_field, (const void __user *)str_val, STR_VAR_LEN_MAX); else #endif ret = strncpy_from_kernel_nofault(str_field, str_val, STR_VAR_LEN_MAX); @@ -447,6 +458,41 @@ static unsigned int trace_string(struct synth_trace_event *entry, return len; } +static unsigned int trace_stack(struct synth_trace_event *entry, + struct synth_event *event, + long *stack, + unsigned int data_size, + unsigned int *n_u64) +{ + union trace_synth_field *data = &entry->fields[*n_u64]; + unsigned int len; + u32 data_offset; + void *data_loc; + + data_offset = struct_size(entry, fields, event->n_u64); + data_offset += data_size; + + for (len = 0; len < HIST_STACKTRACE_DEPTH; len++) { + if (!stack[len]) + break; + } + + len *= sizeof(long); + + /* Find the dynamic section to copy the stack into. */ + data_loc = (void *)entry + data_offset; + memcpy(data_loc, stack, len); + + /* Fill in the field that holds the offset/len combo */ + + data->as_dynamic.offset = data_offset; + data->as_dynamic.len = len; + + (*n_u64)++; + + return len; +} + static notrace void trace_event_raw_event_synth(void *__data, u64 *var_ref_vals, unsigned int *var_ref_idx) @@ -473,7 +519,13 @@ static notrace void trace_event_raw_event_synth(void *__data, val_idx = var_ref_idx[field_pos]; str_val = (char *)(long)var_ref_vals[val_idx]; - len = kern_fetch_store_strlen((unsigned long)str_val); + if (event->dynamic_fields[i]->is_stack) { + /* reserve one extra element for size */ + len = *((unsigned long *)str_val) + 1; + len *= sizeof(unsigned long); + } else { + len = fetch_store_strlen((unsigned long)str_val); + } fields_size += len; } @@ -483,12 +535,12 @@ static notrace void trace_event_raw_event_synth(void *__data, * is being performed within another event. */ buffer = trace_file->tr->array_buffer.buffer; - ring_buffer_nest_start(buffer); + guard(ring_buffer_nest)(buffer); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + fields_size); if (!entry) - goto out; + return; for (i = 0, n_u64 = 0; i < event->n_fields; i++) { val_idx = var_ref_idx[i]; @@ -499,25 +551,31 @@ static notrace void trace_event_raw_event_synth(void *__data, event->fields[i]->is_dynamic, data_size, &n_u64); data_size += len; /* only dynamic string increments */ + } else if (event->fields[i]->is_stack) { + long *stack = (long *)(long)var_ref_vals[val_idx]; + + len = trace_stack(entry, event, stack, + data_size, &n_u64); + data_size += len; } else { struct synth_field *field = event->fields[i]; u64 val = var_ref_vals[val_idx]; switch (field->size) { case 1: - *(u8 *)&entry->fields[n_u64] = (u8)val; + entry->fields[n_u64].as_u8 = (u8)val; break; case 2: - *(u16 *)&entry->fields[n_u64] = (u16)val; + entry->fields[n_u64].as_u16 = (u16)val; break; case 4: - *(u32 *)&entry->fields[n_u64] = (u32)val; + entry->fields[n_u64].as_u32 = (u32)val; break; default: - entry->fields[n_u64] = val; + entry->fields[n_u64].as_u64 = val; break; } n_u64++; @@ -525,8 +583,6 @@ static notrace void trace_event_raw_event_synth(void *__data, } trace_event_buffer_commit(&fbuffer); -out: - ring_buffer_nest_end(buffer); } static void free_synth_event_print_fmt(struct trace_event_call *call) @@ -552,7 +608,7 @@ static int __set_synth_event_print_fmt(struct synth_event *event, fmt = synth_field_fmt(event->fields[i]->type); pos += snprintf(buf + pos, LEN_OR_ZERO, "%s=%s%s", event->fields[i]->name, fmt, - i == event->n_fields - 1 ? "" : ", "); + i == event->n_fields - 1 ? "" : " "); } pos += snprintf(buf + pos, LEN_OR_ZERO, "\""); @@ -561,6 +617,9 @@ static int __set_synth_event_print_fmt(struct synth_event *event, event->fields[i]->is_dynamic) pos += snprintf(buf + pos, LEN_OR_ZERO, ", __get_str(%s)", event->fields[i]->name); + else if (event->fields[i]->is_stack) + pos += snprintf(buf + pos, LEN_OR_ZERO, + ", __get_stacktrace(%s)", event->fields[i]->name); else pos += snprintf(buf + pos, LEN_OR_ZERO, ", REC->%s", event->fields[i]->name); @@ -697,7 +756,8 @@ static struct synth_field *parse_synth_field(int argc, char **argv, ret = -EINVAL; goto free; } else if (size == 0) { - if (synth_field_is_string(field->type)) { + if (synth_field_is_string(field->type) || + synth_field_is_stack(field->type)) { char *type; len = sizeof("__data_loc ") + strlen(field->type) + 1; @@ -728,6 +788,8 @@ static struct synth_field *parse_synth_field(int argc, char **argv, if (synth_field_is_string(field->type)) field->is_string = true; + else if (synth_field_is_stack(field->type)) + field->is_stack = true; field->is_signed = synth_field_signed(field->type); out: @@ -786,6 +848,38 @@ static struct trace_event_fields synth_event_fields_array[] = { {} }; +static int synth_event_reg(struct trace_event_call *call, + enum trace_reg type, void *data) +{ + struct synth_event *event = container_of(call, struct synth_event, call); + + switch (type) { +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_REGISTER: +#endif + case TRACE_REG_REGISTER: + if (!try_module_get(event->mod)) + return -EBUSY; + break; + default: + break; + } + + int ret = trace_event_reg(call, type, data); + + switch (type) { +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_UNREGISTER: +#endif + case TRACE_REG_UNREGISTER: + module_put(event->mod); + break; + default: + break; + } + return ret; +} + static int register_synth_event(struct synth_event *event) { struct trace_event_call *call = &event->call; @@ -815,7 +909,7 @@ static int register_synth_event(struct synth_event *event) goto out; } call->flags = TRACE_EVENT_FL_TRACEPOINT; - call->class->reg = trace_event_reg; + call->class->reg = synth_event_reg; call->class->probe = trace_event_raw_event_synth; call->data = event; call->tp = event->tp; @@ -1065,7 +1159,7 @@ EXPORT_SYMBOL_GPL(synth_event_add_fields); * @cmd: A pointer to the dynevent_cmd struct representing the new event * @name: The name of the synthetic event * @mod: The module creating the event, NULL if not created from a module - * @args: Variable number of arg (pairs), one pair for each field + * @...: Variable number of arg (pairs), one pair for each field * * NOTE: Users normally won't want to call this function directly, but * rather use the synth_event_gen_cmd_start() wrapper, which @@ -1137,6 +1231,7 @@ EXPORT_SYMBOL_GPL(__synth_event_gen_cmd_start); * synth_event_gen_cmd_array_start - Start synthetic event command from an array * @cmd: A pointer to the dynevent_cmd struct representing the new event * @name: The name of the synthetic event + * @mod: The module creating the event, NULL if not created from a module * @fields: An array of type/name field descriptions * @n_fields: The number of field descriptions contained in the fields array * @@ -1622,7 +1717,7 @@ __synth_event_trace_end(struct synth_event_trace_state *trace_state) * synth_event_trace - Trace a synthetic event * @file: The trace_event_file representing the synthetic event * @n_vals: The number of values in vals - * @args: Variable number of args containing the event values + * @...: Variable number of args containing the event values * * Trace a synthetic event using the values passed in the variable * argument list. @@ -1697,19 +1792,19 @@ int synth_event_trace(struct trace_event_file *file, unsigned int n_vals, ...) switch (field->size) { case 1: - *(u8 *)&state.entry->fields[n_u64] = (u8)val; + state.entry->fields[n_u64].as_u8 = (u8)val; break; case 2: - *(u16 *)&state.entry->fields[n_u64] = (u16)val; + state.entry->fields[n_u64].as_u16 = (u16)val; break; case 4: - *(u32 *)&state.entry->fields[n_u64] = (u32)val; + state.entry->fields[n_u64].as_u32 = (u32)val; break; default: - state.entry->fields[n_u64] = val; + state.entry->fields[n_u64].as_u64 = val; break; } n_u64++; @@ -1790,19 +1885,19 @@ int synth_event_trace_array(struct trace_event_file *file, u64 *vals, switch (field->size) { case 1: - *(u8 *)&state.entry->fields[n_u64] = (u8)val; + state.entry->fields[n_u64].as_u8 = (u8)val; break; case 2: - *(u16 *)&state.entry->fields[n_u64] = (u16)val; + state.entry->fields[n_u64].as_u16 = (u16)val; break; case 4: - *(u32 *)&state.entry->fields[n_u64] = (u32)val; + state.entry->fields[n_u64].as_u32 = (u32)val; break; default: - state.entry->fields[n_u64] = val; + state.entry->fields[n_u64].as_u64 = val; break; } n_u64++; @@ -1937,19 +2032,19 @@ static int __synth_event_add_val(const char *field_name, u64 val, } else { switch (field->size) { case 1: - *(u8 *)&trace_state->entry->fields[field->offset] = (u8)val; + trace_state->entry->fields[field->offset].as_u8 = (u8)val; break; case 2: - *(u16 *)&trace_state->entry->fields[field->offset] = (u16)val; + trace_state->entry->fields[field->offset].as_u16 = (u16)val; break; case 4: - *(u32 *)&trace_state->entry->fields[field->offset] = (u32)val; + trace_state->entry->fields[field->offset].as_u32 = (u32)val; break; default: - trace_state->entry->fields[field->offset] = val; + trace_state->entry->fields[field->offset].as_u64 = val; break; } } diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index e535959939d3..06b75bcfc7b8 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -6,6 +6,7 @@ */ #include <linux/security.h> +#include <linux/kthread.h> #include <linux/module.h> #include <linux/ctype.h> #include <linux/mutex.h> @@ -17,21 +18,85 @@ static LIST_HEAD(trigger_commands); static DEFINE_MUTEX(trigger_cmd_mutex); +static struct task_struct *trigger_kthread; +static struct llist_head trigger_data_free_list; +static DEFINE_MUTEX(trigger_data_kthread_mutex); + +/* Bulk garbage collection of event_trigger_data elements */ +static int trigger_kthread_fn(void *ignore) +{ + struct event_trigger_data *data, *tmp; + struct llist_node *llnodes; + + /* Once this task starts, it lives forever */ + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (llist_empty(&trigger_data_free_list)) + schedule(); + + __set_current_state(TASK_RUNNING); + + llnodes = llist_del_all(&trigger_data_free_list); + + /* make sure current triggers exit before free */ + tracepoint_synchronize_unregister(); + + llist_for_each_entry_safe(data, tmp, llnodes, llist) + kfree(data); + } + + return 0; +} + void trigger_data_free(struct event_trigger_data *data) { if (data->cmd_ops->set_filter) data->cmd_ops->set_filter(NULL, data, NULL); - /* make sure current triggers exit before free */ - tracepoint_synchronize_unregister(); + if (unlikely(!trigger_kthread)) { + guard(mutex)(&trigger_data_kthread_mutex); + /* Check again after taking mutex */ + if (!trigger_kthread) { + struct task_struct *kthread; + + kthread = kthread_create(trigger_kthread_fn, NULL, + "trigger_data_free"); + if (!IS_ERR(kthread)) + WRITE_ONCE(trigger_kthread, kthread); + } + } - kfree(data); + if (!trigger_kthread) { + /* Do it the slow way */ + tracepoint_synchronize_unregister(); + kfree(data); + return; + } + + llist_add(&data->llist, &trigger_data_free_list); + wake_up_process(trigger_kthread); +} + +static inline void data_ops_trigger(struct event_trigger_data *data, + struct trace_buffer *buffer, void *rec, + struct ring_buffer_event *event) +{ + const struct event_command *cmd_ops = data->cmd_ops; + + if (data->flags & EVENT_TRIGGER_FL_COUNT) { + if (!cmd_ops->count_func(data, buffer, rec, event)) + return; + } + + cmd_ops->trigger(data, buffer, rec, event); } /** * event_triggers_call - Call triggers associated with a trace event * @file: The trace_event_file associated with the event + * @buffer: The ring buffer that the event is being written to * @rec: The trace entry for the event, NULL for unconditional invocation + * @event: The event meta data in the ring buffer * * For each trigger associated with an event, invoke the trigger * function registered with the associated trigger command. If rec is @@ -68,7 +133,7 @@ event_triggers_call(struct trace_event_file *file, if (data->paused) continue; if (!rec) { - data->ops->trigger(data, buffer, rec, event); + data_ops_trigger(data, buffer, rec, event); continue; } filter = rcu_dereference_sched(data->filter); @@ -78,7 +143,7 @@ event_triggers_call(struct trace_event_file *file, tt |= data->cmd_ops->trigger_type; continue; } - data->ops->trigger(data, buffer, rec, event); + data_ops_trigger(data, buffer, rec, event); } return tt; } @@ -120,7 +185,7 @@ event_triggers_post_call(struct trace_event_file *file, if (data->paused) continue; if (data->cmd_ops->trigger_type & tt) - data->ops->trigger(data, NULL, NULL, NULL); + data_ops_trigger(data, NULL, NULL, NULL); } } EXPORT_SYMBOL_GPL(event_triggers_post_call); @@ -157,7 +222,7 @@ static void *trigger_start(struct seq_file *m, loff_t *pos) /* ->stop() is called even if ->start() fails */ mutex_lock(&event_mutex); - event_file = event_file_data(m->private); + event_file = event_file_file(m->private); if (unlikely(!event_file)) return ERR_PTR(-ENODEV); @@ -189,7 +254,7 @@ static int trigger_show(struct seq_file *m, void *v) } data = list_entry(v, struct event_trigger_data, list); - data->ops->print(m, data); + data->cmd_ops->print(m, data); return 0; } @@ -209,12 +274,10 @@ static int event_trigger_regex_open(struct inode *inode, struct file *file) if (ret) return ret; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); - if (unlikely(!event_file_data(file))) { - mutex_unlock(&event_mutex); + if (unlikely(!event_file_file(file))) return -ENODEV; - } if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { @@ -237,8 +300,6 @@ static int event_trigger_regex_open(struct inode *inode, struct file *file) } } - mutex_unlock(&event_mutex); - return ret; } @@ -246,9 +307,9 @@ int trigger_process_regex(struct trace_event_file *file, char *buff) { char *command, *next; struct event_command *p; - int ret = -EINVAL; - next = buff = skip_spaces(buff); + next = buff = strim(buff); + command = strsep(&next, ": \t"); if (next) { next = skip_spaces(next); @@ -257,17 +318,14 @@ int trigger_process_regex(struct trace_event_file *file, char *buff) } command = (command[0] != '!') ? command : command + 1; - mutex_lock(&trigger_cmd_mutex); + guard(mutex)(&trigger_cmd_mutex); + list_for_each_entry(p, &trigger_commands, list) { - if (strcmp(p->name, command) == 0) { - ret = p->parse(p, file, buff, command, next); - goto out_unlock; - } + if (strcmp(p->name, command) == 0) + return p->parse(p, file, buff, command, next); } - out_unlock: - mutex_unlock(&trigger_cmd_mutex); - return ret; + return -EINVAL; } static ssize_t event_trigger_regex_write(struct file *file, @@ -276,7 +334,7 @@ static ssize_t event_trigger_regex_write(struct file *file, { struct trace_event_file *event_file; ssize_t ret; - char *buf; + char *buf __free(kfree) = NULL; if (!cnt) return 0; @@ -288,37 +346,25 @@ static ssize_t event_trigger_regex_write(struct file *file, if (IS_ERR(buf)) return PTR_ERR(buf); - strim(buf); + guard(mutex)(&event_mutex); - mutex_lock(&event_mutex); - event_file = event_file_data(file); - if (unlikely(!event_file)) { - mutex_unlock(&event_mutex); - kfree(buf); + event_file = event_file_file(file); + if (unlikely(!event_file)) return -ENODEV; - } - ret = trigger_process_regex(event_file, buf); - mutex_unlock(&event_mutex); - kfree(buf); + ret = trigger_process_regex(event_file, buf); if (ret < 0) - goto out; + return ret; *ppos += cnt; - ret = cnt; - out: - return ret; + return cnt; } static int event_trigger_regex_release(struct inode *inode, struct file *file) { - mutex_lock(&event_mutex); - if (file->f_mode & FMODE_READ) seq_release(inode, file); - mutex_unlock(&event_mutex); - return 0; } @@ -357,20 +403,16 @@ const struct file_operations event_trigger_fops = { __init int register_event_command(struct event_command *cmd) { struct event_command *p; - int ret = 0; - mutex_lock(&trigger_cmd_mutex); + guard(mutex)(&trigger_cmd_mutex); + list_for_each_entry(p, &trigger_commands, list) { - if (strcmp(cmd->name, p->name) == 0) { - ret = -EBUSY; - goto out_unlock; - } + if (strcmp(cmd->name, p->name) == 0) + return -EBUSY; } list_add(&cmd->list, &trigger_commands); - out_unlock: - mutex_unlock(&trigger_cmd_mutex); - return ret; + return 0; } /* @@ -380,24 +422,51 @@ __init int register_event_command(struct event_command *cmd) __init int unregister_event_command(struct event_command *cmd) { struct event_command *p, *n; - int ret = -ENODEV; - mutex_lock(&trigger_cmd_mutex); + guard(mutex)(&trigger_cmd_mutex); + list_for_each_entry_safe(p, n, &trigger_commands, list) { if (strcmp(cmd->name, p->name) == 0) { - ret = 0; list_del_init(&p->list); - goto out_unlock; + return 0; } } - out_unlock: - mutex_unlock(&trigger_cmd_mutex); - return ret; + return -ENODEV; +} + +/** + * event_trigger_count - Optional count function for event triggers + * @data: Trigger-specific data + * @buffer: The ring buffer that the event is being written to + * @rec: The trace entry for the event, NULL for unconditional invocation + * @event: The event meta data in the ring buffer + * + * For triggers that can take a count parameter that doesn't do anything + * special, they can use this function to assign to their .count_func + * field. + * + * This simply does a count down of the @data->count field. + * + * If the @data->count is greater than zero, it will decrement it. + * + * Returns false if @data->count is zero, otherwise true. + */ +bool event_trigger_count(struct event_trigger_data *data, + struct trace_buffer *buffer, void *rec, + struct ring_buffer_event *event) +{ + if (!data->count) + return false; + + if (data->count != -1) + (data->count)--; + + return true; } /** - * event_trigger_print - Generic event_trigger_ops @print implementation + * event_trigger_print - Generic event_command @print implementation * @name: The name of the event trigger * @m: The seq_file being printed to * @data: Trigger-specific data @@ -432,7 +501,7 @@ event_trigger_print(const char *name, struct seq_file *m, } /** - * event_trigger_init - Generic event_trigger_ops @init implementation + * event_trigger_init - Generic event_command @init implementation * @data: Trigger-specific data * * Common implementation of event trigger initialization. @@ -449,7 +518,7 @@ int event_trigger_init(struct event_trigger_data *data) } /** - * event_trigger_free - Generic event_trigger_ops @free implementation + * event_trigger_free - Generic event_command @free implementation * @data: Trigger-specific data * * Common implementation of event trigger de-initialization. @@ -511,8 +580,8 @@ clear_event_triggers(struct trace_array *tr) list_for_each_entry_safe(data, n, &file->triggers, list) { trace_event_trigger_enable_disable(file, 0); list_del_rcu(&data->list); - if (data->ops->free) - data->ops->free(data); + if (data->cmd_ops->free) + data->cmd_ops->free(data); } } } @@ -571,16 +640,14 @@ static int register_trigger(char *glob, lockdep_assert_held(&event_mutex); list_for_each_entry(test, &file->triggers, list) { - if (test->cmd_ops->trigger_type == data->cmd_ops->trigger_type) { - ret = -EEXIST; - goto out; - } + if (test->cmd_ops->trigger_type == data->cmd_ops->trigger_type) + return -EEXIST; } - if (data->ops->init) { - ret = data->ops->init(data); + if (data->cmd_ops->init) { + ret = data->cmd_ops->init(data); if (ret < 0) - goto out; + return ret; } list_add_rcu(&data->list, &file->triggers); @@ -591,24 +658,15 @@ static int register_trigger(char *glob, list_del_rcu(&data->list); update_cond_flag(file); } -out: return ret; } -/** - * unregister_trigger - Generic event_command @unreg implementation - * @glob: The raw string used to register the trigger - * @test: Trigger-specific data used to find the trigger to remove - * @file: The trace_event_file associated with the event - * - * Common implementation for event trigger unregistration. - * - * Usually used directly as the @unreg method in event command - * implementations. +/* + * True if the trigger was found and unregistered, else false. */ -static void unregister_trigger(char *glob, - struct event_trigger_data *test, - struct trace_event_file *file) +static bool try_unregister_trigger(char *glob, + struct event_trigger_data *test, + struct trace_event_file *file) { struct event_trigger_data *data = NULL, *iter; @@ -624,8 +682,32 @@ static void unregister_trigger(char *glob, } } - if (data && data->ops->free) - data->ops->free(data); + if (data) { + if (data->cmd_ops->free) + data->cmd_ops->free(data); + + return true; + } + + return false; +} + +/** + * unregister_trigger - Generic event_command @unreg implementation + * @glob: The raw string used to register the trigger + * @test: Trigger-specific data used to find the trigger to remove + * @file: The trace_event_file associated with the event + * + * Common implementation for event trigger unregistration. + * + * Usually used directly as the @unreg method in event command + * implementations. + */ +static void unregister_trigger(char *glob, + struct event_trigger_data *test, + struct trace_event_file *file) +{ + try_unregister_trigger(glob, test, file); } /* @@ -650,7 +732,7 @@ static void unregister_trigger(char *glob, * param - text following cmd and ':' and stripped of filter * filter - the optional filter text following (and including) 'if' * - * To illustrate the use of these componenents, here are some concrete + * To illustrate the use of these components, here are some concrete * examples. For the following triggers: * * echo 'traceon:5 if pid == 0' > trigger @@ -773,7 +855,7 @@ int event_trigger_separate_filter(char *param_and_filter, char **param, if (!param_and_filter) { if (param_required) ret = -EINVAL; - goto out; + return ret; } /* @@ -784,7 +866,7 @@ int event_trigger_separate_filter(char *param_and_filter, char **param, */ if (!param_required && param_and_filter && !isdigit(param_and_filter[0])) { *filter = param_and_filter; - goto out; + return ret; } /* @@ -802,44 +884,45 @@ int event_trigger_separate_filter(char *param_and_filter, char **param, if (!**filter) *filter = NULL; } -out: return ret; } /** - * event_trigger_alloc - allocate and init event_trigger_data for a trigger + * trigger_data_alloc - allocate and init event_trigger_data for a trigger * @cmd_ops: The event_command operations for the trigger * @cmd: The cmd string * @param: The param string * @private_data: User data to associate with the event trigger * * Allocate an event_trigger_data instance and initialize it. The - * @cmd_ops are used along with the @cmd and @param to get the - * trigger_ops to assign to the event_trigger_data. @private_data can - * also be passed in and associated with the event_trigger_data. + * @cmd_ops defines how the trigger will operate. If @param is set, + * and @cmd_ops->trigger_ops->count_func is non NULL, then the + * data->count is set to @param and before the trigger is executed, the + * @cmd_ops->trigger_ops->count_func() is called. If that function returns + * false, the @cmd_ops->trigger_ops->trigger() function will not be called. + * @private_data can also be passed in and associated with the + * event_trigger_data. * - * Use event_trigger_free() to free an event_trigger_data object. + * Use trigger_data_free() to free an event_trigger_data object. * * Return: The trigger_data object success, NULL otherwise */ -struct event_trigger_data *event_trigger_alloc(struct event_command *cmd_ops, - char *cmd, - char *param, - void *private_data) +struct event_trigger_data *trigger_data_alloc(struct event_command *cmd_ops, + char *cmd, + char *param, + void *private_data) { struct event_trigger_data *trigger_data; - struct event_trigger_ops *trigger_ops; - - trigger_ops = cmd_ops->get_trigger_ops(cmd, param); trigger_data = kzalloc(sizeof(*trigger_data), GFP_KERNEL); if (!trigger_data) return NULL; trigger_data->count = -1; - trigger_data->ops = trigger_ops; trigger_data->cmd_ops = cmd_ops; trigger_data->private_data = private_data; + if (param && cmd_ops->count_func) + trigger_data->flags |= EVENT_TRIGGER_FL_COUNT; INIT_LIST_HEAD(&trigger_data->list); INIT_LIST_HEAD(&trigger_data->named_list); @@ -992,15 +1075,14 @@ event_trigger_parse(struct event_command *cmd_ops, return ret; ret = -ENOMEM; - trigger_data = event_trigger_alloc(cmd_ops, cmd, param, file); + trigger_data = trigger_data_alloc(cmd_ops, cmd, param, file); if (!trigger_data) - goto out; + return ret; if (remove) { event_trigger_unregister(cmd_ops, file, glob+1, trigger_data); - kfree(trigger_data); - ret = 0; - goto out; + trigger_data_free(trigger_data); + return 0; } ret = event_trigger_parse_num(param, trigger_data); @@ -1020,13 +1102,12 @@ event_trigger_parse(struct event_command *cmd_ops, /* Down the counter of trigger_data or free it if not used anymore */ event_trigger_free(trigger_data); - out: return ret; out_free: event_trigger_reset_filter(cmd_ops, trigger_data); - kfree(trigger_data); - goto out; + trigger_data_free(trigger_data); + return ret; } /** @@ -1060,10 +1141,10 @@ int set_trigger_filter(char *filter_str, s = strsep(&filter_str, " \t"); if (!strlen(s) || strcmp(s, "if") != 0) - goto out; + return ret; if (!filter_str) - goto out; + return ret; /* The filter is for the 'trigger' event, not the triggered event */ ret = create_event_filter(file->tr, file->event_call, @@ -1107,7 +1188,6 @@ int set_trigger_filter(char *filter_str, ret = -ENOMEM; } } - out: return ret; } @@ -1281,31 +1361,28 @@ traceon_trigger(struct event_trigger_data *data, tracing_on(); } -static void -traceon_count_trigger(struct event_trigger_data *data, - struct trace_buffer *buffer, void *rec, - struct ring_buffer_event *event) +static bool +traceon_count_func(struct event_trigger_data *data, + struct trace_buffer *buffer, void *rec, + struct ring_buffer_event *event) { struct trace_event_file *file = data->private_data; if (file) { if (tracer_tracing_is_on(file->tr)) - return; + return false; } else { if (tracing_is_on()) - return; + return false; } if (!data->count) - return; + return false; if (data->count != -1) (data->count)--; - if (file) - tracer_tracing_on(file->tr); - else - tracing_on(); + return true; } static void @@ -1329,31 +1406,28 @@ traceoff_trigger(struct event_trigger_data *data, tracing_off(); } -static void -traceoff_count_trigger(struct event_trigger_data *data, - struct trace_buffer *buffer, void *rec, - struct ring_buffer_event *event) +static bool +traceoff_count_func(struct event_trigger_data *data, + struct trace_buffer *buffer, void *rec, + struct ring_buffer_event *event) { struct trace_event_file *file = data->private_data; if (file) { if (!tracer_tracing_is_on(file->tr)) - return; + return false; } else { if (!tracing_is_on()) - return; + return false; } if (!data->count) - return; + return false; if (data->count != -1) (data->count)--; - if (file) - tracer_tracing_off(file->tr); - else - tracing_off(); + return true; } static int @@ -1370,58 +1444,18 @@ traceoff_trigger_print(struct seq_file *m, struct event_trigger_data *data) data->filter_str); } -static struct event_trigger_ops traceon_trigger_ops = { - .trigger = traceon_trigger, - .print = traceon_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops traceon_count_trigger_ops = { - .trigger = traceon_count_trigger, - .print = traceon_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops traceoff_trigger_ops = { - .trigger = traceoff_trigger, - .print = traceoff_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops traceoff_count_trigger_ops = { - .trigger = traceoff_count_trigger, - .print = traceoff_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops * -onoff_get_trigger_ops(char *cmd, char *param) -{ - struct event_trigger_ops *ops; - - /* we register both traceon and traceoff to this callback */ - if (strcmp(cmd, "traceon") == 0) - ops = param ? &traceon_count_trigger_ops : - &traceon_trigger_ops; - else - ops = param ? &traceoff_count_trigger_ops : - &traceoff_trigger_ops; - - return ops; -} - static struct event_command trigger_traceon_cmd = { .name = "traceon", .trigger_type = ETT_TRACE_ONOFF, .parse = event_trigger_parse, .reg = register_trigger, .unreg = unregister_trigger, - .get_trigger_ops = onoff_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = traceon_trigger, + .count_func = traceon_count_func, + .print = traceon_trigger_print, + .init = event_trigger_init, + .free = event_trigger_free, }; static struct event_command trigger_traceoff_cmd = { @@ -1431,8 +1465,12 @@ static struct event_command trigger_traceoff_cmd = { .parse = event_trigger_parse, .reg = register_trigger, .unreg = unregister_trigger, - .get_trigger_ops = onoff_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = traceoff_trigger, + .count_func = traceoff_count_func, + .print = traceoff_trigger_print, + .init = event_trigger_init, + .free = event_trigger_free, }; #ifdef CONFIG_TRACER_SNAPSHOT @@ -1449,29 +1487,28 @@ snapshot_trigger(struct event_trigger_data *data, tracing_snapshot(); } -static void -snapshot_count_trigger(struct event_trigger_data *data, - struct trace_buffer *buffer, void *rec, - struct ring_buffer_event *event) -{ - if (!data->count) - return; - - if (data->count != -1) - (data->count)--; - - snapshot_trigger(data, buffer, rec, event); -} - static int register_snapshot_trigger(char *glob, struct event_trigger_data *data, struct trace_event_file *file) { - if (tracing_alloc_snapshot_instance(file->tr) != 0) - return 0; + int ret = tracing_arm_snapshot(file->tr); - return register_trigger(glob, data, file); + if (ret < 0) + return ret; + + ret = register_trigger(glob, data, file); + if (ret < 0) + tracing_disarm_snapshot(file->tr); + return ret; +} + +static void unregister_snapshot_trigger(char *glob, + struct event_trigger_data *data, + struct trace_event_file *file) +{ + if (try_unregister_trigger(glob, data, file)) + tracing_disarm_snapshot(file->tr); } static int @@ -1481,34 +1518,18 @@ snapshot_trigger_print(struct seq_file *m, struct event_trigger_data *data) data->filter_str); } -static struct event_trigger_ops snapshot_trigger_ops = { - .trigger = snapshot_trigger, - .print = snapshot_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops snapshot_count_trigger_ops = { - .trigger = snapshot_count_trigger, - .print = snapshot_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops * -snapshot_get_trigger_ops(char *cmd, char *param) -{ - return param ? &snapshot_count_trigger_ops : &snapshot_trigger_ops; -} - static struct event_command trigger_snapshot_cmd = { .name = "snapshot", .trigger_type = ETT_SNAPSHOT, .parse = event_trigger_parse, .reg = register_snapshot_trigger, - .unreg = unregister_trigger, - .get_trigger_ops = snapshot_get_trigger_ops, + .unreg = unregister_snapshot_trigger, .set_filter = set_trigger_filter, + .trigger = snapshot_trigger, + .count_func = event_trigger_count, + .print = snapshot_trigger_print, + .init = event_trigger_init, + .free = event_trigger_free, }; static __init int register_trigger_snapshot_cmd(void) @@ -1550,25 +1571,11 @@ stacktrace_trigger(struct event_trigger_data *data, struct trace_event_file *file = data->private_data; if (file) - __trace_stack(file->tr, tracing_gen_ctx(), STACK_SKIP); + __trace_stack(file->tr, tracing_gen_ctx_dec(), STACK_SKIP); else trace_dump_stack(STACK_SKIP); } -static void -stacktrace_count_trigger(struct event_trigger_data *data, - struct trace_buffer *buffer, void *rec, - struct ring_buffer_event *event) -{ - if (!data->count) - return; - - if (data->count != -1) - (data->count)--; - - stacktrace_trigger(data, buffer, rec, event); -} - static int stacktrace_trigger_print(struct seq_file *m, struct event_trigger_data *data) { @@ -1576,26 +1583,6 @@ stacktrace_trigger_print(struct seq_file *m, struct event_trigger_data *data) data->filter_str); } -static struct event_trigger_ops stacktrace_trigger_ops = { - .trigger = stacktrace_trigger, - .print = stacktrace_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops stacktrace_count_trigger_ops = { - .trigger = stacktrace_count_trigger, - .print = stacktrace_trigger_print, - .init = event_trigger_init, - .free = event_trigger_free, -}; - -static struct event_trigger_ops * -stacktrace_get_trigger_ops(char *cmd, char *param) -{ - return param ? &stacktrace_count_trigger_ops : &stacktrace_trigger_ops; -} - static struct event_command trigger_stacktrace_cmd = { .name = "stacktrace", .trigger_type = ETT_STACKTRACE, @@ -1603,8 +1590,12 @@ static struct event_command trigger_stacktrace_cmd = { .parse = event_trigger_parse, .reg = register_trigger, .unreg = unregister_trigger, - .get_trigger_ops = stacktrace_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = stacktrace_trigger, + .count_func = event_trigger_count, + .print = stacktrace_trigger_print, + .init = event_trigger_init, + .free = event_trigger_free, }; static __init int register_trigger_stacktrace_cmd(void) @@ -1639,24 +1630,24 @@ event_enable_trigger(struct event_trigger_data *data, set_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &enable_data->file->flags); } -static void -event_enable_count_trigger(struct event_trigger_data *data, - struct trace_buffer *buffer, void *rec, - struct ring_buffer_event *event) +static bool +event_enable_count_func(struct event_trigger_data *data, + struct trace_buffer *buffer, void *rec, + struct ring_buffer_event *event) { struct enable_trigger_data *enable_data = data->private_data; if (!data->count) - return; + return false; /* Skip if the event is in a state we want to switch to */ if (enable_data->enable == !(enable_data->file->flags & EVENT_FILE_FL_SOFT_DISABLED)) - return; + return false; if (data->count != -1) (data->count)--; - event_enable_trigger(data, buffer, rec, event); + return true; } int event_enable_trigger_print(struct seq_file *m, @@ -1701,34 +1692,6 @@ void event_enable_trigger_free(struct event_trigger_data *data) } } -static struct event_trigger_ops event_enable_trigger_ops = { - .trigger = event_enable_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops event_enable_count_trigger_ops = { - .trigger = event_enable_count_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops event_disable_trigger_ops = { - .trigger = event_enable_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - -static struct event_trigger_ops event_disable_count_trigger_ops = { - .trigger = event_enable_count_trigger, - .print = event_enable_trigger_print, - .init = event_trigger_init, - .free = event_enable_trigger_free, -}; - int event_enable_trigger_parse(struct event_command *cmd_ops, struct trace_event_file *file, char *glob, char *cmd, char *param_and_filter) @@ -1762,7 +1725,7 @@ int event_enable_trigger_parse(struct event_command *cmd_ops, ret = -EINVAL; event_enable_file = find_event_file(tr, system, event); if (!event_enable_file) - goto out; + return ret; #ifdef CONFIG_HIST_TRIGGERS hist = ((strcmp(cmd, ENABLE_HIST_STR) == 0) || @@ -1777,16 +1740,16 @@ int event_enable_trigger_parse(struct event_command *cmd_ops, enable_data = kzalloc(sizeof(*enable_data), GFP_KERNEL); if (!enable_data) - goto out; + return ret; enable_data->hist = hist; enable_data->enable = enable; enable_data->file = event_enable_file; - trigger_data = event_trigger_alloc(cmd_ops, cmd, param, enable_data); + trigger_data = trigger_data_alloc(cmd_ops, cmd, param, enable_data); if (!trigger_data) { kfree(enable_data); - goto out; + return ret; } if (remove) { @@ -1794,7 +1757,7 @@ int event_enable_trigger_parse(struct event_command *cmd_ops, kfree(trigger_data); kfree(enable_data); ret = 0; - goto out; + return ret; } /* Up the trigger_data count to make sure nothing frees it on failure */ @@ -1824,7 +1787,6 @@ int event_enable_trigger_parse(struct event_command *cmd_ops, goto out_disable; event_trigger_free(trigger_data); - out: return ret; out_disable: trace_event_enable_disable(event_enable_file, 0, 1); @@ -1835,7 +1797,7 @@ int event_enable_trigger_parse(struct event_command *cmd_ops, event_trigger_free(trigger_data); kfree(enable_data); - goto out; + return ret; } int event_enable_register_trigger(char *glob, @@ -1855,15 +1817,14 @@ int event_enable_register_trigger(char *glob, (test->cmd_ops->trigger_type == data->cmd_ops->trigger_type) && (test_enable_data->file == enable_data->file)) { - ret = -EEXIST; - goto out; + return -EEXIST; } } - if (data->ops->init) { - ret = data->ops->init(data); + if (data->cmd_ops->init) { + ret = data->cmd_ops->init(data); if (ret < 0) - goto out; + return ret; } list_add_rcu(&data->list, &file->triggers); @@ -1874,7 +1835,6 @@ int event_enable_register_trigger(char *glob, list_del_rcu(&data->list); update_cond_flag(file); } -out: return ret; } @@ -1902,30 +1862,8 @@ void event_enable_unregister_trigger(char *glob, } } - if (data && data->ops->free) - data->ops->free(data); -} - -static struct event_trigger_ops * -event_enable_get_trigger_ops(char *cmd, char *param) -{ - struct event_trigger_ops *ops; - bool enable; - -#ifdef CONFIG_HIST_TRIGGERS - enable = ((strcmp(cmd, ENABLE_EVENT_STR) == 0) || - (strcmp(cmd, ENABLE_HIST_STR) == 0)); -#else - enable = strcmp(cmd, ENABLE_EVENT_STR) == 0; -#endif - if (enable) - ops = param ? &event_enable_count_trigger_ops : - &event_enable_trigger_ops; - else - ops = param ? &event_disable_count_trigger_ops : - &event_disable_trigger_ops; - - return ops; + if (data && data->cmd_ops->free) + data->cmd_ops->free(data); } static struct event_command trigger_enable_cmd = { @@ -1934,8 +1872,12 @@ static struct event_command trigger_enable_cmd = { .parse = event_enable_trigger_parse, .reg = event_enable_register_trigger, .unreg = event_enable_unregister_trigger, - .get_trigger_ops = event_enable_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = event_enable_trigger, + .count_func = event_enable_count_func, + .print = event_enable_trigger_print, + .init = event_trigger_init, + .free = event_enable_trigger_free, }; static struct event_command trigger_disable_cmd = { @@ -1944,8 +1886,12 @@ static struct event_command trigger_disable_cmd = { .parse = event_enable_trigger_parse, .reg = event_enable_register_trigger, .unreg = event_enable_unregister_trigger, - .get_trigger_ops = event_enable_get_trigger_ops, .set_filter = set_trigger_filter, + .trigger = event_enable_trigger, + .count_func = event_enable_count_func, + .print = event_enable_trigger_print, + .init = event_trigger_init, + .free = event_enable_trigger_free, }; static __init void unregister_trigger_enable_disable_cmds(void) diff --git a/kernel/trace/trace_events_user.c b/kernel/trace/trace_events_user.c index 908e8a13c675..dca6e50b3b21 100644 --- a/kernel/trace/trace_events_user.c +++ b/kernel/trace/trace_events_user.c @@ -19,14 +19,12 @@ #include <linux/tracefs.h> #include <linux/types.h> #include <linux/uaccess.h> -/* Reminder to move to uapi when everything works */ -#ifdef CONFIG_COMPILE_TEST +#include <linux/highmem.h> +#include <linux/init.h> #include <linux/user_events.h> -#else -#include <uapi/linux/user_events.h> -#endif -#include "trace.h" #include "trace_dynevent.h" +#include "trace_output.h" +#include "trace.h" #define USER_EVENTS_PREFIX_LEN (sizeof(USER_EVENTS_PREFIX)-1) @@ -34,35 +32,13 @@ #define FIELD_DEPTH_NAME 1 #define FIELD_DEPTH_SIZE 2 -/* - * Limits how many trace_event calls user processes can create: - * Must be a power of two of PAGE_SIZE. - */ -#define MAX_PAGE_ORDER 0 -#define MAX_PAGES (1 << MAX_PAGE_ORDER) -#define MAX_BYTES (MAX_PAGES * PAGE_SIZE) -#define MAX_EVENTS (MAX_BYTES * 8) - /* Limit how long of an event name plus args within the subsystem. */ #define MAX_EVENT_DESC 512 -#define EVENT_NAME(user_event) ((user_event)->tracepoint.name) +#define EVENT_NAME(user_event) ((user_event)->reg_name) +#define EVENT_TP_NAME(user_event) ((user_event)->tracepoint.name) #define MAX_FIELD_ARRAY_SIZE 1024 /* - * The MAP_STATUS_* macros are used for taking a index and determining the - * appropriate byte and the bit in the byte to set/reset for an event. - * - * The lower 3 bits of the index decide which bit to set. - * The remaining upper bits of the index decide which byte to use for the bit. - * - * This is used when an event has a probe attached/removed to reflect live - * status of the event wanting tracing or not to user-programs via shared - * memory maps. - */ -#define MAP_STATUS_BYTE(index) ((index) >> 3) -#define MAP_STATUS_MASK(index) BIT((index) & 7) - -/* * Internal bits (kernel side only) to keep track of connected probes: * These are used when status is requested in text form about an event. These * bits are compared against an internal byte on the event to determine which @@ -75,25 +51,28 @@ #define EVENT_STATUS_OTHER BIT(7) /* - * Stores the pages, tables, and locks for a group of events. - * Each logical grouping of events has its own group, with a - * matching page for status checks within user programs. This - * allows for isolation of events to user programs by various - * means. + * Stores the system name, tables, and locks for a group of events. This + * allows isolation for events by various means. */ struct user_event_group { - struct page *pages; - char *register_page_data; - char *system_name; - struct hlist_node node; - struct mutex reg_mutex; + char *system_name; + char *system_multi_name; + struct hlist_node node; + struct mutex reg_mutex; DECLARE_HASHTABLE(register_table, 8); - DECLARE_BITMAP(page_bitmap, MAX_EVENTS); + /* ID that moves forward within the group for multi-event names */ + u64 multi_id; }; /* Group for init_user_ns mapping, top-most group */ static struct user_event_group *init_group; +/* Max allowed events for the whole system */ +static unsigned int max_user_events = 32768; + +/* Current number of events on the whole system */ +static unsigned int current_user_events; + /* * Stores per-event properties, as users register events * within a file a user_event might be created if it does not @@ -102,21 +81,73 @@ static struct user_event_group *init_group; * refcnt reaches one. */ struct user_event { - struct user_event_group *group; - struct tracepoint tracepoint; - struct trace_event_call call; - struct trace_event_class class; - struct dyn_event devent; - struct hlist_node node; - struct list_head fields; - struct list_head validators; - refcount_t refcnt; - int index; - int flags; - int min_size; - char status; + struct user_event_group *group; + char *reg_name; + struct tracepoint tracepoint; + struct trace_event_call call; + struct trace_event_class class; + struct dyn_event devent; + struct hlist_node node; + struct list_head fields; + struct list_head validators; + struct work_struct put_work; + refcount_t refcnt; + int min_size; + int reg_flags; + char status; +}; + +/* + * Stores per-mm/event properties that enable an address to be + * updated properly for each task. As tasks are forked, we use + * these to track enablement sites that are tied to an event. + */ +struct user_event_enabler { + struct list_head mm_enablers_link; + struct user_event *event; + unsigned long addr; + + /* Track enable bit, flags, etc. Aligned for bitops. */ + unsigned long values; }; +/* Bits 0-5 are for the bit to update upon enable/disable (0-63 allowed) */ +#define ENABLE_VAL_BIT_MASK 0x3F + +/* Bit 6 is for faulting status of enablement */ +#define ENABLE_VAL_FAULTING_BIT 6 + +/* Bit 7 is for freeing status of enablement */ +#define ENABLE_VAL_FREEING_BIT 7 + +/* Bit 8 is for marking 32-bit on 64-bit */ +#define ENABLE_VAL_32_ON_64_BIT 8 + +#define ENABLE_VAL_COMPAT_MASK (1 << ENABLE_VAL_32_ON_64_BIT) + +/* Only duplicate the bit and compat values */ +#define ENABLE_VAL_DUP_MASK (ENABLE_VAL_BIT_MASK | ENABLE_VAL_COMPAT_MASK) + +#define ENABLE_BITOPS(e) (&(e)->values) + +#define ENABLE_BIT(e) ((int)((e)->values & ENABLE_VAL_BIT_MASK)) + +#define EVENT_MULTI_FORMAT(f) ((f) & USER_EVENT_REG_MULTI_FORMAT) + +/* Used for asynchronous faulting in of pages */ +struct user_event_enabler_fault { + struct work_struct work; + struct user_event_mm *mm; + struct user_event_enabler *enabler; + int attempt; +}; + +static struct kmem_cache *fault_cache; + +/* Global list of memory descriptors using user_events */ +static LIST_HEAD(user_event_mms); +static DEFINE_SPINLOCK(user_event_mms_lock); + /* * Stores per-file events references, as users register events * within a file this structure is modified and freed via RCU. @@ -124,78 +155,197 @@ struct user_event { * These are not shared and only accessible by the file that created it. */ struct user_event_refs { - struct rcu_head rcu; - int count; - struct user_event *events[]; + struct rcu_head rcu; + int count; + struct user_event *events[]; }; struct user_event_file_info { - struct user_event_group *group; - struct user_event_refs *refs; + struct user_event_group *group; + struct user_event_refs *refs; }; #define VALIDATOR_ENSURE_NULL (1 << 0) #define VALIDATOR_REL (1 << 1) struct user_event_validator { - struct list_head link; - int offset; - int flags; + struct list_head user_event_link; + int offset; + int flags; }; +static inline void align_addr_bit(unsigned long *addr, int *bit, + unsigned long *flags) +{ + if (IS_ALIGNED(*addr, sizeof(long))) { +#ifdef __BIG_ENDIAN + /* 32 bit on BE 64 bit requires a 32 bit offset when aligned. */ + if (test_bit(ENABLE_VAL_32_ON_64_BIT, flags)) + *bit += 32; +#endif + return; + } + + *addr = ALIGN_DOWN(*addr, sizeof(long)); + + /* + * We only support 32 and 64 bit values. The only time we need + * to align is a 32 bit value on a 64 bit kernel, which on LE + * is always 32 bits, and on BE requires no change when unaligned. + */ +#ifdef __LITTLE_ENDIAN + *bit += 32; +#endif +} + typedef void (*user_event_func_t) (struct user_event *user, struct iov_iter *i, void *tpdata, bool *faulted); static int user_event_parse(struct user_event_group *group, char *name, char *args, char *flags, - struct user_event **newuser); + struct user_event **newuser, int reg_flags); + +static struct user_event_mm *user_event_mm_get(struct user_event_mm *mm); +static struct user_event_mm *user_event_mm_get_all(struct user_event *user); +static void user_event_mm_put(struct user_event_mm *mm); +static int destroy_user_event(struct user_event *user); +static bool user_fields_match(struct user_event *user, int argc, + const char **argv); static u32 user_event_key(char *name) { return jhash(name, strlen(name), 0); } -static void set_page_reservations(char *pages, bool set) +static bool user_event_capable(u16 reg_flags) { - int page; + /* Persistent events require CAP_PERFMON / CAP_SYS_ADMIN */ + if (reg_flags & USER_EVENT_REG_PERSIST) { + if (!perfmon_capable()) + return false; + } - for (page = 0; page < MAX_PAGES; ++page) { - void *addr = pages + (PAGE_SIZE * page); + return true; +} - if (set) - SetPageReserved(virt_to_page(addr)); - else - ClearPageReserved(virt_to_page(addr)); +static struct user_event *user_event_get(struct user_event *user) +{ + refcount_inc(&user->refcnt); + + return user; +} + +static void delayed_destroy_user_event(struct work_struct *work) +{ + struct user_event *user = container_of( + work, struct user_event, put_work); + + mutex_lock(&event_mutex); + + if (!refcount_dec_and_test(&user->refcnt)) + goto out; + + if (destroy_user_event(user)) { + /* + * The only reason this would fail here is if we cannot + * update the visibility of the event. In this case the + * event stays in the hashtable, waiting for someone to + * attempt to delete it later. + */ + pr_warn("user_events: Unable to delete event\n"); + refcount_set(&user->refcnt, 1); } +out: + mutex_unlock(&event_mutex); } -static void user_event_group_destroy(struct user_event_group *group) +static void user_event_put(struct user_event *user, bool locked) { - if (group->register_page_data) - set_page_reservations(group->register_page_data, false); + bool delete; - if (group->pages) - __free_pages(group->pages, MAX_PAGE_ORDER); + if (unlikely(!user)) + return; + + /* + * When the event is not enabled for auto-delete there will always + * be at least 1 reference to the event. During the event creation + * we initially set the refcnt to 2 to achieve this. In those cases + * the caller must acquire event_mutex and after decrement check if + * the refcnt is 1, meaning this is the last reference. When auto + * delete is enabled, there will only be 1 ref, IE: refcnt will be + * only set to 1 during creation to allow the below checks to go + * through upon the last put. The last put must always be done with + * the event mutex held. + */ + if (!locked) { + lockdep_assert_not_held(&event_mutex); + delete = refcount_dec_and_mutex_lock(&user->refcnt, &event_mutex); + } else { + lockdep_assert_held(&event_mutex); + delete = refcount_dec_and_test(&user->refcnt); + } + if (!delete) + return; + + /* + * We now have the event_mutex in all cases, which ensures that + * no new references will be taken until event_mutex is released. + * New references come through find_user_event(), which requires + * the event_mutex to be held. + */ + + if (user->reg_flags & USER_EVENT_REG_PERSIST) { + /* We should not get here when persist flag is set */ + pr_alert("BUG: Auto-delete engaged on persistent event\n"); + goto out; + } + + /* + * Unfortunately we have to attempt the actual destroy in a work + * queue. This is because not all cases handle a trace_event_call + * being removed within the class->reg() operation for unregister. + */ + INIT_WORK(&user->put_work, delayed_destroy_user_event); + + /* + * Since the event is still in the hashtable, we have to re-inc + * the ref count to 1. This count will be decremented and checked + * in the work queue to ensure it's still the last ref. This is + * needed because a user-process could register the same event in + * between the time of event_mutex release and the work queue + * running the delayed destroy. If we removed the item now from + * the hashtable, this would result in a timing window where a + * user process would fail a register because the trace_event_call + * register would fail in the tracing layers. + */ + refcount_set(&user->refcnt, 1); + + if (WARN_ON_ONCE(!schedule_work(&user->put_work))) { + /* + * If we fail we must wait for an admin to attempt delete or + * another register/close of the event, whichever is first. + */ + pr_warn("user_events: Unable to queue delayed destroy\n"); + } +out: + /* Ensure if we didn't have event_mutex before we unlock it */ + if (!locked) + mutex_unlock(&event_mutex); +} + +static void user_event_group_destroy(struct user_event_group *group) +{ kfree(group->system_name); + kfree(group->system_multi_name); kfree(group); } -static char *user_event_group_system_name(struct user_namespace *user_ns) +static char *user_event_group_system_name(void) { char *system_name; int len = sizeof(USER_EVENTS_SYSTEM) + 1; - if (user_ns != &init_user_ns) { - /* - * Unexpected at this point: - * We only currently support init_user_ns. - * When we enable more, this will trigger a failure so log. - */ - pr_warn("user_events: Namespace other than init_user_ns!\n"); - return NULL; - } - system_name = kmalloc(len, GFP_KERNEL); if (!system_name) @@ -206,34 +356,17 @@ static char *user_event_group_system_name(struct user_namespace *user_ns) return system_name; } -static inline struct user_event_group -*user_event_group_from_user_ns(struct user_namespace *user_ns) +static char *user_event_group_system_multi_name(void) { - if (user_ns == &init_user_ns) - return init_group; - - return NULL; + return kstrdup(USER_EVENTS_MULTI_SYSTEM, GFP_KERNEL); } static struct user_event_group *current_user_event_group(void) { - struct user_namespace *user_ns = current_user_ns(); - struct user_event_group *group = NULL; - - while (user_ns) { - group = user_event_group_from_user_ns(user_ns); - - if (group) - break; - - user_ns = user_ns->parent; - } - - return group; + return init_group; } -static struct user_event_group -*user_event_group_create(struct user_namespace *user_ns) +static struct user_event_group *user_event_group_create(void) { struct user_event_group *group; @@ -242,24 +375,16 @@ static struct user_event_group if (!group) return NULL; - group->system_name = user_event_group_system_name(user_ns); + group->system_name = user_event_group_system_name(); if (!group->system_name) goto error; - group->pages = alloc_pages(GFP_KERNEL | __GFP_ZERO, MAX_PAGE_ORDER); + group->system_multi_name = user_event_group_system_multi_name(); - if (!group->pages) + if (!group->system_multi_name) goto error; - group->register_page_data = page_address(group->pages); - - set_page_reservations(group->register_page_data, true); - - /* Zero all bits beside 0 (which is reserved for failures) */ - bitmap_zero(group->page_bitmap, MAX_EVENTS); - set_bit(0, group->page_bitmap); - mutex_init(&group->reg_mutex); hash_init(group->register_table); @@ -271,26 +396,564 @@ error: return NULL; }; -static __always_inline -void user_event_register_set(struct user_event *user) +static void user_event_enabler_destroy(struct user_event_enabler *enabler, + bool locked) { - int i = user->index; + list_del_rcu(&enabler->mm_enablers_link); + + /* No longer tracking the event via the enabler */ + user_event_put(enabler->event, locked); - user->group->register_page_data[MAP_STATUS_BYTE(i)] |= MAP_STATUS_MASK(i); + kfree(enabler); } -static __always_inline -void user_event_register_clear(struct user_event *user) +static int user_event_mm_fault_in(struct user_event_mm *mm, unsigned long uaddr, + int attempt) +{ + bool unlocked; + int ret; + + /* + * Normally this is low, ensure that it cannot be taken advantage of by + * bad user processes to cause excessive looping. + */ + if (attempt > 10) + return -EFAULT; + + mmap_read_lock(mm->mm); + + /* Ensure MM has tasks, cannot use after exit_mm() */ + if (refcount_read(&mm->tasks) == 0) { + ret = -ENOENT; + goto out; + } + + ret = fixup_user_fault(mm->mm, uaddr, FAULT_FLAG_WRITE | FAULT_FLAG_REMOTE, + &unlocked); +out: + mmap_read_unlock(mm->mm); + + return ret; +} + +static int user_event_enabler_write(struct user_event_mm *mm, + struct user_event_enabler *enabler, + bool fixup_fault, int *attempt); + +static void user_event_enabler_fault_fixup(struct work_struct *work) { - int i = user->index; + struct user_event_enabler_fault *fault = container_of( + work, struct user_event_enabler_fault, work); + struct user_event_enabler *enabler = fault->enabler; + struct user_event_mm *mm = fault->mm; + unsigned long uaddr = enabler->addr; + int attempt = fault->attempt; + int ret; + + ret = user_event_mm_fault_in(mm, uaddr, attempt); + + if (ret && ret != -ENOENT) { + struct user_event *user = enabler->event; + + pr_warn("user_events: Fault for mm: 0x%p @ 0x%llx event: %s\n", + mm->mm, (unsigned long long)uaddr, EVENT_NAME(user)); + } + + /* Prevent state changes from racing */ + mutex_lock(&event_mutex); + + /* User asked for enabler to be removed during fault */ + if (test_bit(ENABLE_VAL_FREEING_BIT, ENABLE_BITOPS(enabler))) { + user_event_enabler_destroy(enabler, true); + goto out; + } + + /* + * If we managed to get the page, re-issue the write. We do not + * want to get into a possible infinite loop, which is why we only + * attempt again directly if the page came in. If we couldn't get + * the page here, then we will try again the next time the event is + * enabled/disabled. + */ + clear_bit(ENABLE_VAL_FAULTING_BIT, ENABLE_BITOPS(enabler)); + + if (!ret) { + mmap_read_lock(mm->mm); + user_event_enabler_write(mm, enabler, true, &attempt); + mmap_read_unlock(mm->mm); + } +out: + mutex_unlock(&event_mutex); - user->group->register_page_data[MAP_STATUS_BYTE(i)] &= ~MAP_STATUS_MASK(i); + /* In all cases we no longer need the mm or fault */ + user_event_mm_put(mm); + kmem_cache_free(fault_cache, fault); +} + +static bool user_event_enabler_queue_fault(struct user_event_mm *mm, + struct user_event_enabler *enabler, + int attempt) +{ + struct user_event_enabler_fault *fault; + + fault = kmem_cache_zalloc(fault_cache, GFP_NOWAIT); + + if (!fault) + return false; + + INIT_WORK(&fault->work, user_event_enabler_fault_fixup); + fault->mm = user_event_mm_get(mm); + fault->enabler = enabler; + fault->attempt = attempt; + + /* Don't try to queue in again while we have a pending fault */ + set_bit(ENABLE_VAL_FAULTING_BIT, ENABLE_BITOPS(enabler)); + + if (!schedule_work(&fault->work)) { + /* Allow another attempt later */ + clear_bit(ENABLE_VAL_FAULTING_BIT, ENABLE_BITOPS(enabler)); + + user_event_mm_put(mm); + kmem_cache_free(fault_cache, fault); + + return false; + } + + return true; +} + +static int user_event_enabler_write(struct user_event_mm *mm, + struct user_event_enabler *enabler, + bool fixup_fault, int *attempt) +{ + unsigned long uaddr = enabler->addr; + unsigned long *ptr; + struct page *page; + void *kaddr; + int bit = ENABLE_BIT(enabler); + int ret; + + lockdep_assert_held(&event_mutex); + mmap_assert_locked(mm->mm); + + *attempt += 1; + + /* Ensure MM has tasks, cannot use after exit_mm() */ + if (refcount_read(&mm->tasks) == 0) + return -ENOENT; + + if (unlikely(test_bit(ENABLE_VAL_FAULTING_BIT, ENABLE_BITOPS(enabler)) || + test_bit(ENABLE_VAL_FREEING_BIT, ENABLE_BITOPS(enabler)))) + return -EBUSY; + + align_addr_bit(&uaddr, &bit, ENABLE_BITOPS(enabler)); + + ret = pin_user_pages_remote(mm->mm, uaddr, 1, FOLL_WRITE | FOLL_NOFAULT, + &page, NULL); + + if (unlikely(ret <= 0)) { + if (!fixup_fault) + return -EFAULT; + + if (!user_event_enabler_queue_fault(mm, enabler, *attempt)) + pr_warn("user_events: Unable to queue fault handler\n"); + + return -EFAULT; + } + + kaddr = kmap_local_page(page); + ptr = kaddr + (uaddr & ~PAGE_MASK); + + /* Update bit atomically, user tracers must be atomic as well */ + if (enabler->event && enabler->event->status) + set_bit(bit, ptr); + else + clear_bit(bit, ptr); + + kunmap_local(kaddr); + unpin_user_pages_dirty_lock(&page, 1, true); + + return 0; +} + +static bool user_event_enabler_exists(struct user_event_mm *mm, + unsigned long uaddr, unsigned char bit) +{ + struct user_event_enabler *enabler; + + list_for_each_entry(enabler, &mm->enablers, mm_enablers_link) { + if (enabler->addr == uaddr && ENABLE_BIT(enabler) == bit) + return true; + } + + return false; +} + +static void user_event_enabler_update(struct user_event *user) +{ + struct user_event_enabler *enabler; + struct user_event_mm *next; + struct user_event_mm *mm; + int attempt; + + lockdep_assert_held(&event_mutex); + + /* + * We need to build a one-shot list of all the mms that have an + * enabler for the user_event passed in. This list is only valid + * while holding the event_mutex. The only reason for this is due + * to the global mm list being RCU protected and we use methods + * which can wait (mmap_read_lock and pin_user_pages_remote). + * + * NOTE: user_event_mm_get_all() increments the ref count of each + * mm that is added to the list to prevent removal timing windows. + * We must always put each mm after they are used, which may wait. + */ + mm = user_event_mm_get_all(user); + + while (mm) { + next = mm->next; + mmap_read_lock(mm->mm); + + list_for_each_entry(enabler, &mm->enablers, mm_enablers_link) { + if (enabler->event == user) { + attempt = 0; + user_event_enabler_write(mm, enabler, true, &attempt); + } + } + + mmap_read_unlock(mm->mm); + user_event_mm_put(mm); + mm = next; + } +} + +static bool user_event_enabler_dup(struct user_event_enabler *orig, + struct user_event_mm *mm) +{ + struct user_event_enabler *enabler; + + /* Skip pending frees */ + if (unlikely(test_bit(ENABLE_VAL_FREEING_BIT, ENABLE_BITOPS(orig)))) + return true; + + enabler = kzalloc(sizeof(*enabler), GFP_NOWAIT | __GFP_ACCOUNT); + + if (!enabler) + return false; + + enabler->event = user_event_get(orig->event); + enabler->addr = orig->addr; + + /* Only dup part of value (ignore future flags, etc) */ + enabler->values = orig->values & ENABLE_VAL_DUP_MASK; + + /* Enablers not exposed yet, RCU not required */ + list_add(&enabler->mm_enablers_link, &mm->enablers); + + return true; +} + +static struct user_event_mm *user_event_mm_get(struct user_event_mm *mm) +{ + refcount_inc(&mm->refcnt); + + return mm; +} + +static struct user_event_mm *user_event_mm_get_all(struct user_event *user) +{ + struct user_event_mm *found = NULL; + struct user_event_enabler *enabler; + struct user_event_mm *mm; + + /* + * We use the mm->next field to build a one-shot list from the global + * RCU protected list. To build this list the event_mutex must be held. + * This lets us build a list without requiring allocs that could fail + * when user based events are most wanted for diagnostics. + */ + lockdep_assert_held(&event_mutex); + + /* + * We do not want to block fork/exec while enablements are being + * updated, so we use RCU to walk the current tasks that have used + * user_events ABI for 1 or more events. Each enabler found in each + * task that matches the event being updated has a write to reflect + * the kernel state back into the process. Waits/faults must not occur + * during this. So we scan the list under RCU for all the mm that have + * the event within it. This is needed because mm_read_lock() can wait. + * Each user mm returned has a ref inc to handle remove RCU races. + */ + rcu_read_lock(); + + list_for_each_entry_rcu(mm, &user_event_mms, mms_link) { + list_for_each_entry_rcu(enabler, &mm->enablers, mm_enablers_link) { + if (enabler->event == user) { + mm->next = found; + found = user_event_mm_get(mm); + break; + } + } + } + + rcu_read_unlock(); + + return found; +} + +static struct user_event_mm *user_event_mm_alloc(struct task_struct *t) +{ + struct user_event_mm *user_mm; + + user_mm = kzalloc(sizeof(*user_mm), GFP_KERNEL_ACCOUNT); + + if (!user_mm) + return NULL; + + user_mm->mm = t->mm; + INIT_LIST_HEAD(&user_mm->enablers); + refcount_set(&user_mm->refcnt, 1); + refcount_set(&user_mm->tasks, 1); + + /* + * The lifetime of the memory descriptor can slightly outlast + * the task lifetime if a ref to the user_event_mm is taken + * between list_del_rcu() and call_rcu(). Therefore we need + * to take a reference to it to ensure it can live this long + * under this corner case. This can also occur in clones that + * outlast the parent. + */ + mmgrab(user_mm->mm); + + return user_mm; +} + +static void user_event_mm_attach(struct user_event_mm *user_mm, struct task_struct *t) +{ + unsigned long flags; + + spin_lock_irqsave(&user_event_mms_lock, flags); + list_add_rcu(&user_mm->mms_link, &user_event_mms); + spin_unlock_irqrestore(&user_event_mms_lock, flags); + + t->user_event_mm = user_mm; +} + +static struct user_event_mm *current_user_event_mm(void) +{ + struct user_event_mm *user_mm = current->user_event_mm; + + if (user_mm) + goto inc; + + user_mm = user_event_mm_alloc(current); + + if (!user_mm) + goto error; + + user_event_mm_attach(user_mm, current); +inc: + refcount_inc(&user_mm->refcnt); +error: + return user_mm; +} + +static void user_event_mm_destroy(struct user_event_mm *mm) +{ + struct user_event_enabler *enabler, *next; + + list_for_each_entry_safe(enabler, next, &mm->enablers, mm_enablers_link) + user_event_enabler_destroy(enabler, false); + + mmdrop(mm->mm); + kfree(mm); +} + +static void user_event_mm_put(struct user_event_mm *mm) +{ + if (mm && refcount_dec_and_test(&mm->refcnt)) + user_event_mm_destroy(mm); +} + +static void delayed_user_event_mm_put(struct work_struct *work) +{ + struct user_event_mm *mm; + + mm = container_of(to_rcu_work(work), struct user_event_mm, put_rwork); + user_event_mm_put(mm); +} + +void user_event_mm_remove(struct task_struct *t) +{ + struct user_event_mm *mm; + unsigned long flags; + + might_sleep(); + + mm = t->user_event_mm; + t->user_event_mm = NULL; + + /* Clone will increment the tasks, only remove if last clone */ + if (!refcount_dec_and_test(&mm->tasks)) + return; + + /* Remove the mm from the list, so it can no longer be enabled */ + spin_lock_irqsave(&user_event_mms_lock, flags); + list_del_rcu(&mm->mms_link); + spin_unlock_irqrestore(&user_event_mms_lock, flags); + + /* + * We need to wait for currently occurring writes to stop within + * the mm. This is required since exit_mm() snaps the current rss + * stats and clears them. On the final mmdrop(), check_mm() will + * report a bug if these increment. + * + * All writes/pins are done under mmap_read lock, take the write + * lock to ensure in-progress faults have completed. Faults that + * are pending but yet to run will check the task count and skip + * the fault since the mm is going away. + */ + mmap_write_lock(mm->mm); + mmap_write_unlock(mm->mm); + + /* + * Put for mm must be done after RCU delay to handle new refs in + * between the list_del_rcu() and now. This ensures any get refs + * during rcu_read_lock() are accounted for during list removal. + * + * CPU A | CPU B + * --------------------------------------------------------------- + * user_event_mm_remove() | rcu_read_lock(); + * list_del_rcu() | list_for_each_entry_rcu(); + * call_rcu() | refcount_inc(); + * . | rcu_read_unlock(); + * schedule_work() | . + * user_event_mm_put() | . + * + * mmdrop() cannot be called in the softirq context of call_rcu() + * so we use a work queue after call_rcu() to run within. + */ + INIT_RCU_WORK(&mm->put_rwork, delayed_user_event_mm_put); + queue_rcu_work(system_percpu_wq, &mm->put_rwork); +} + +void user_event_mm_dup(struct task_struct *t, struct user_event_mm *old_mm) +{ + struct user_event_mm *mm = user_event_mm_alloc(t); + struct user_event_enabler *enabler; + + if (!mm) + return; + + rcu_read_lock(); + + list_for_each_entry_rcu(enabler, &old_mm->enablers, mm_enablers_link) { + if (!user_event_enabler_dup(enabler, mm)) + goto error; + } + + rcu_read_unlock(); + + user_event_mm_attach(mm, t); + return; +error: + rcu_read_unlock(); + user_event_mm_destroy(mm); +} + +static bool current_user_event_enabler_exists(unsigned long uaddr, + unsigned char bit) +{ + struct user_event_mm *user_mm = current_user_event_mm(); + bool exists; + + if (!user_mm) + return false; + + exists = user_event_enabler_exists(user_mm, uaddr, bit); + + user_event_mm_put(user_mm); + + return exists; +} + +static struct user_event_enabler +*user_event_enabler_create(struct user_reg *reg, struct user_event *user, + int *write_result) +{ + struct user_event_enabler *enabler; + struct user_event_mm *user_mm; + unsigned long uaddr = (unsigned long)reg->enable_addr; + int attempt = 0; + + user_mm = current_user_event_mm(); + + if (!user_mm) + return NULL; + + enabler = kzalloc(sizeof(*enabler), GFP_KERNEL_ACCOUNT); + + if (!enabler) + goto out; + + enabler->event = user; + enabler->addr = uaddr; + enabler->values = reg->enable_bit; + +#if BITS_PER_LONG >= 64 + if (reg->enable_size == 4) + set_bit(ENABLE_VAL_32_ON_64_BIT, ENABLE_BITOPS(enabler)); +#endif + +retry: + /* Prevents state changes from racing with new enablers */ + mutex_lock(&event_mutex); + + /* Attempt to reflect the current state within the process */ + mmap_read_lock(user_mm->mm); + *write_result = user_event_enabler_write(user_mm, enabler, false, + &attempt); + mmap_read_unlock(user_mm->mm); + + /* + * If the write works, then we will track the enabler. A ref to the + * underlying user_event is held by the enabler to prevent it going + * away while the enabler is still in use by a process. The ref is + * removed when the enabler is destroyed. This means a event cannot + * be forcefully deleted from the system until all tasks using it + * exit or run exec(), which includes forks and clones. + */ + if (!*write_result) { + user_event_get(user); + list_add_rcu(&enabler->mm_enablers_link, &user_mm->enablers); + } + + mutex_unlock(&event_mutex); + + if (*write_result) { + /* Attempt to fault-in and retry if it worked */ + if (!user_event_mm_fault_in(user_mm, uaddr, attempt)) + goto retry; + + kfree(enabler); + enabler = NULL; + } +out: + user_event_mm_put(user_mm); + + return enabler; } static __always_inline __must_check bool user_event_last_ref(struct user_event *user) { - return refcount_read(&user->refcnt) == 1; + int last = 0; + + if (user->reg_flags & USER_EVENT_REG_PERSIST) + last = 1; + + return refcount_read(&user->refcnt) == last; } static __always_inline __must_check @@ -329,7 +992,8 @@ static struct list_head *user_event_get_fields(struct trace_event_call *call) * Upon success user_event has its ref count increased by 1. */ static int user_event_parse_cmd(struct user_event_group *group, - char *raw_command, struct user_event **newuser) + char *raw_command, struct user_event **newuser, + int reg_flags) { char *name = raw_command; char *args = strpbrk(name, " "); @@ -343,7 +1007,7 @@ static int user_event_parse_cmd(struct user_event_group *group, if (flags) *flags++ = '\0'; - return user_event_parse(group, name, args, flags, newuser); + return user_event_parse(group, name, args, flags, newuser, reg_flags); } static int user_field_array_size(const char *type) @@ -377,7 +1041,7 @@ static int user_field_array_size(const char *type) static int user_field_size(const char *type) { - /* long is not allowed from a user, since it's ambigious in size */ + /* long is not allowed from a user, since it's ambiguous in size */ if (strcmp(type, "s64") == 0) return sizeof(s64); if (strcmp(type, "u64") == 0) @@ -415,7 +1079,7 @@ static int user_field_size(const char *type) if (str_has_prefix(type, "__rel_loc ")) return sizeof(u32); - /* Uknown basic type, error */ + /* Unknown basic type, error */ return -EINVAL; } @@ -424,8 +1088,8 @@ static void user_event_destroy_validators(struct user_event *user) struct user_event_validator *validator, *next; struct list_head *head = &user->validators; - list_for_each_entry_safe(validator, next, head, link) { - list_del(&validator->link); + list_for_each_entry_safe(validator, next, head, user_event_link) { + list_del(&validator->user_event_link); kfree(validator); } } @@ -449,7 +1113,7 @@ static int user_event_add_field(struct user_event *user, const char *type, struct ftrace_event_field *field; int validator_flags = 0; - field = kmalloc(sizeof(*field), GFP_KERNEL); + field = kmalloc(sizeof(*field), GFP_KERNEL_ACCOUNT); if (!field) return -ENOMEM; @@ -468,7 +1132,7 @@ add_validator: if (strstr(type, "char") != NULL) validator_flags |= VALIDATOR_ENSURE_NULL; - validator = kmalloc(sizeof(*validator), GFP_KERNEL); + validator = kmalloc(sizeof(*validator), GFP_KERNEL_ACCOUNT); if (!validator) { kfree(field); @@ -479,7 +1143,7 @@ add_validator: validator->offset = offset; /* Want sequential access when validating */ - list_add_tail(&validator->link, &user->validators); + list_add_tail(&validator->user_event_link, &user->validators); add_field: field->type = type; @@ -489,6 +1153,9 @@ add_field: field->is_signed = is_signed; field->filter_type = filter_type; + if (filter_type == FILTER_OTHER) + field->filter_type = filter_assign_type(type); + list_add(&field->link, &user->fields); /* @@ -707,6 +1374,9 @@ static int user_field_set_string(struct ftrace_event_field *field, pos += snprintf(buf + pos, LEN_OR_ZERO, " "); pos += snprintf(buf + pos, LEN_OR_ZERO, "%s", field->name); + if (str_has_prefix(field->type, "struct ")) + pos += snprintf(buf + pos, LEN_OR_ZERO, " %d", field->size); + if (colon) pos += snprintf(buf + pos, LEN_OR_ZERO, ";"); @@ -715,14 +1385,14 @@ static int user_field_set_string(struct ftrace_event_field *field, static int user_event_set_print_fmt(struct user_event *user, char *buf, int len) { - struct ftrace_event_field *field, *next; + struct ftrace_event_field *field; struct list_head *head = &user->fields; int pos = 0, depth = 0; const char *str_func; pos += snprintf(buf + pos, LEN_OR_ZERO, "\""); - list_for_each_entry_safe_reverse(field, next, head, link) { + list_for_each_entry_reverse(field, head, link) { if (depth != 0) pos += snprintf(buf + pos, LEN_OR_ZERO, " "); @@ -734,7 +1404,7 @@ static int user_event_set_print_fmt(struct user_event *user, char *buf, int len) pos += snprintf(buf + pos, LEN_OR_ZERO, "\""); - list_for_each_entry_safe_reverse(field, next, head, link) { + list_for_each_entry_reverse(field, head, link) { if (user_field_is_dyn_string(field->type, &str_func)) pos += snprintf(buf + pos, LEN_OR_ZERO, ", %s(%s)", str_func, field->name); @@ -754,7 +1424,7 @@ static int user_event_create_print_fmt(struct user_event *user) len = user_event_set_print_fmt(user, NULL, 0); - print_fmt = kmalloc(len, GFP_KERNEL); + print_fmt = kmalloc(len, GFP_KERNEL_ACCOUNT); if (!print_fmt) return -ENOMEM; @@ -770,11 +1440,7 @@ static enum print_line_t user_event_print_trace(struct trace_iterator *iter, int flags, struct trace_event *event) { - /* Unsafe to try to decode user provided print_fmt, use hex */ - trace_print_hex_dump_seq(&iter->seq, "", DUMP_PREFIX_OFFSET, 16, - 1, iter->ent, iter->ent_size, true); - - return trace_handle_return(&iter->seq); + return print_event_fields(iter, event); } static struct trace_event_functions user_event_funcs = { @@ -783,12 +1449,7 @@ static struct trace_event_functions user_event_funcs = { static int user_event_set_call_visible(struct user_event *user, bool visible) { - int ret; - const struct cred *old_cred; - struct cred *cred; - - cred = prepare_creds(); - + CLASS(prepare_creds, cred)(); if (!cred) return -ENOMEM; @@ -803,23 +1464,20 @@ static int user_event_set_call_visible(struct user_event *user, bool visible) */ cred->fsuid = GLOBAL_ROOT_UID; - old_cred = override_creds(cred); - - if (visible) - ret = trace_add_event_call(&user->call); - else - ret = trace_remove_event_call(&user->call); + scoped_with_creds(cred) { + if (visible) + return trace_add_event_call(&user->call); - revert_creds(old_cred); - put_cred(cred); - - return ret; + return trace_remove_event_call(&user->call); + } } static int destroy_user_event(struct user_event *user) { int ret = 0; + lockdep_assert_held(&event_mutex); + /* Must destroy fields before call removal */ user_event_destroy_fields(user); @@ -829,32 +1487,56 @@ static int destroy_user_event(struct user_event *user) return ret; dyn_event_remove(&user->devent); - - user_event_register_clear(user); - clear_bit(user->index, user->group->page_bitmap); hash_del(&user->node); user_event_destroy_validators(user); + + /* If we have different names, both must be freed */ + if (EVENT_NAME(user) != EVENT_TP_NAME(user)) + kfree(EVENT_TP_NAME(user)); + kfree(user->call.print_fmt); kfree(EVENT_NAME(user)); kfree(user); + if (current_user_events > 0) + current_user_events--; + else + pr_alert("BUG: Bad current_user_events\n"); + return ret; } static struct user_event *find_user_event(struct user_event_group *group, - char *name, u32 *outkey) + char *name, int argc, const char **argv, + u32 flags, u32 *outkey) { struct user_event *user; u32 key = user_event_key(name); *outkey = key; - hash_for_each_possible(group->register_table, user, node, key) - if (!strcmp(EVENT_NAME(user), name)) { - refcount_inc(&user->refcnt); - return user; - } + hash_for_each_possible(group->register_table, user, node, key) { + /* + * Single-format events shouldn't return multi-format + * events. Callers expect the underlying tracepoint to match + * the name exactly in these cases. Only check like-formats. + */ + if (EVENT_MULTI_FORMAT(flags) != EVENT_MULTI_FORMAT(user->reg_flags)) + continue; + + if (strcmp(EVENT_NAME(user), name)) + continue; + + if (user_fields_match(user, argc, argv)) + return user_event_get(user); + + /* Scan others if this is a multi-format event */ + if (EVENT_MULTI_FORMAT(flags)) + continue; + + return ERR_PTR(-EADDRINUSE); + } return NULL; } @@ -866,7 +1548,7 @@ static int user_event_validate(struct user_event *user, void *data, int len) void *pos, *end = data + len; u32 loc, offset, size; - list_for_each_entry(validator, head, link) { + list_for_each_entry(validator, head, user_event_link) { pos = data + validator->offset; /* Already done min_size check, no bounds check here */ @@ -916,7 +1598,7 @@ static void user_event_ftrace(struct user_event *user, struct iov_iter *i, if (unlikely(!entry)) return; - if (unlikely(!copy_nofault(entry + 1, i->count, i))) + if (unlikely(i->count != 0 && !copy_nofault(entry + 1, i->count, i))) goto discard; if (!list_empty(&user->validators) && @@ -957,7 +1639,7 @@ static void user_event_perf(struct user_event *user, struct iov_iter *i, perf_fetch_caller_regs(regs); - if (unlikely(!copy_nofault(perf_entry + 1, i->count, i))) + if (unlikely(i->count != 0 && !copy_nofault(perf_entry + 1, i->count, i))) goto discard; if (!list_empty(&user->validators) && @@ -977,14 +1659,14 @@ discard: #endif /* - * Update the register page that is shared between user processes. + * Update the enabled bit among all user processes. */ -static void update_reg_page_for(struct user_event *user) +static void update_enable_bit_for(struct user_event *user) { struct tracepoint *tp = &user->tracepoint; char status = 0; - if (atomic_read(&tp->key.enabled) > 0) { + if (static_key_enabled(&tp->key)) { struct tracepoint_func *probe_func_ptr; user_event_func_t probe_func; @@ -1010,12 +1692,9 @@ static void update_reg_page_for(struct user_event *user) rcu_read_unlock_sched(); } - if (status) - user_event_register_set(user); - else - user_event_register_clear(user); - user->status = status; + + user_event_enabler_update(user); } /* @@ -1071,12 +1750,12 @@ static int user_event_reg(struct trace_event_call *call, return ret; inc: - refcount_inc(&user->refcnt); - update_reg_page_for(user); + user_event_get(user); + update_enable_bit_for(user); return 0; dec: - update_reg_page_for(user); - refcount_dec(&user->refcnt); + update_enable_bit_for(user); + user_event_put(user, true); return 0; } @@ -1093,7 +1772,7 @@ static int user_event_create(const char *raw_command) raw_command += USER_EVENTS_PREFIX_LEN; raw_command = skip_spaces(raw_command); - name = kstrdup(raw_command, GFP_KERNEL); + name = kstrdup(raw_command, GFP_KERNEL_ACCOUNT); if (!name) return -ENOMEM; @@ -1107,10 +1786,11 @@ static int user_event_create(const char *raw_command) mutex_lock(&group->reg_mutex); - ret = user_event_parse_cmd(group, name, &user); + /* Dyn events persist, otherwise they would cleanup immediately */ + ret = user_event_parse_cmd(group, name, &user, USER_EVENT_REG_PERSIST); if (!ret) - refcount_dec(&user->refcnt); + user_event_put(user, false); mutex_unlock(&group->reg_mutex); @@ -1123,7 +1803,7 @@ static int user_event_create(const char *raw_command) static int user_event_show(struct seq_file *m, struct dyn_event *ev) { struct user_event *user = container_of(ev, struct user_event, devent); - struct ftrace_event_field *field, *next; + struct ftrace_event_field *field; struct list_head *head; int depth = 0; @@ -1131,7 +1811,7 @@ static int user_event_show(struct seq_file *m, struct dyn_event *ev) head = trace_get_fields(&user->call); - list_for_each_entry_safe_reverse(field, next, head, link) { + list_for_each_entry_reverse(field, head, link) { if (depth == 0) seq_puts(m, " "); else @@ -1164,6 +1844,9 @@ static int user_event_free(struct dyn_event *ev) if (!user_event_last_ref(user)) return -EBUSY; + if (!user_event_capable(user->reg_flags)) + return -EPERM; + return destroy_user_event(user); } @@ -1207,13 +1890,17 @@ out: static bool user_fields_match(struct user_event *user, int argc, const char **argv) { - struct ftrace_event_field *field, *next; + struct ftrace_event_field *field; struct list_head *head = &user->fields; int i = 0; - list_for_each_entry_safe_reverse(field, next, head, link) + if (argc == 0) + return list_empty(head); + + list_for_each_entry_reverse(field, head, link) { if (!user_field_match(field, argc, argv, &i)) return false; + } if (i != argc) return false; @@ -1227,10 +1914,14 @@ static bool user_event_match(const char *system, const char *event, struct user_event *user = container_of(ev, struct user_event, devent); bool match; - match = strcmp(EVENT_NAME(user), event) == 0 && - (!system || strcmp(system, USER_EVENTS_SYSTEM) == 0); + match = strcmp(EVENT_NAME(user), event) == 0; + + if (match && system) { + match = strcmp(system, user->group->system_name) == 0 || + strcmp(system, user->group->system_multi_name) == 0; + } - if (match && argc > 0) + if (match) match = user_fields_match(user, argc, argv); return match; @@ -1261,6 +1952,107 @@ static int user_event_trace_register(struct user_event *user) return ret; } +static int user_event_set_tp_name(struct user_event *user) +{ + lockdep_assert_held(&user->group->reg_mutex); + + if (EVENT_MULTI_FORMAT(user->reg_flags)) { + char *multi_name; + + multi_name = kasprintf(GFP_KERNEL_ACCOUNT, "%s.%llx", + user->reg_name, user->group->multi_id); + + if (!multi_name) + return -ENOMEM; + + user->call.name = multi_name; + user->tracepoint.name = multi_name; + + /* Inc to ensure unique multi-event name next time */ + user->group->multi_id++; + } else { + /* Non Multi-format uses register name */ + user->call.name = user->reg_name; + user->tracepoint.name = user->reg_name; + } + + return 0; +} + +/* + * Counts how many ';' without a trailing space are in the args. + */ +static int count_semis_no_space(char *args) +{ + int count = 0; + + while ((args = strchr(args, ';'))) { + args++; + + if (!isspace(*args)) + count++; + } + + return count; +} + +/* + * Copies the arguments while ensuring all ';' have a trailing space. + */ +static char *insert_space_after_semis(char *args, int count) +{ + char *fixed, *pos; + int len; + + len = strlen(args) + count; + fixed = kmalloc(len + 1, GFP_KERNEL); + + if (!fixed) + return NULL; + + pos = fixed; + + /* Insert a space after ';' if there is no trailing space. */ + while (*args) { + *pos = *args++; + + if (*pos++ == ';' && !isspace(*args)) + *pos++ = ' '; + } + + *pos = '\0'; + + return fixed; +} + +static char **user_event_argv_split(char *args, int *argc) +{ + char **split; + char *fixed; + int count; + + /* Count how many ';' without a trailing space */ + count = count_semis_no_space(args); + + /* No fixup is required */ + if (!count) + return argv_split(GFP_KERNEL, args, argc); + + /* We must fixup 'field;field' to 'field; field' */ + fixed = insert_space_after_semis(args, count); + + if (!fixed) + return NULL; + + /* We do a normal split afterwards */ + split = argv_split(GFP_KERNEL, fixed, argc); + + /* We can free since argv_split makes a copy */ + kfree(fixed); + + return split; +} + /* * Parses the event name, arguments and flags then registers if successful. * The name buffer lifetime is owned by this method for success cases only. @@ -1268,18 +2060,40 @@ static int user_event_trace_register(struct user_event *user) */ static int user_event_parse(struct user_event_group *group, char *name, char *args, char *flags, - struct user_event **newuser) + struct user_event **newuser, int reg_flags) { + struct user_event *user; + char **argv = NULL; + int argc = 0; int ret; - int index; u32 key; - struct user_event *user; + + /* Currently don't support any text based flags */ + if (flags != NULL) + return -EINVAL; + + if (!user_event_capable(reg_flags)) + return -EPERM; + + if (args) { + argv = user_event_argv_split(args, &argc); + + if (!argv) + return -ENOMEM; + } /* Prevent dyn_event from racing */ mutex_lock(&event_mutex); - user = find_user_event(group, name, &key); + user = find_user_event(group, name, argc, (const char **)argv, + reg_flags, &key); mutex_unlock(&event_mutex); + if (argv) + argv_free(argv); + + if (IS_ERR(user)) + return PTR_ERR(user); + if (user) { *newuser = user; /* @@ -1287,15 +2101,11 @@ static int user_event_parse(struct user_event_group *group, char *name, * Caller only worries about failure cases for freeing. */ kfree(name); + return 0; } - index = find_first_zero_bit(group->page_bitmap, MAX_EVENTS); - - if (index == MAX_EVENTS) - return -EMFILE; - - user = kzalloc(sizeof(*user), GFP_KERNEL); + user = kzalloc(sizeof(*user), GFP_KERNEL_ACCOUNT); if (!user) return -ENOMEM; @@ -1305,7 +2115,13 @@ static int user_event_parse(struct user_event_group *group, char *name, INIT_LIST_HEAD(&user->validators); user->group = group; - user->tracepoint.name = name; + user->reg_name = name; + user->reg_flags = reg_flags; + + ret = user_event_set_tp_name(user); + + if (ret) + goto put_user; ret = user_event_parse_fields(user, args); @@ -1319,11 +2135,14 @@ static int user_event_parse(struct user_event_group *group, char *name, user->call.data = user; user->call.class = &user->class; - user->call.name = name; user->call.flags = TRACE_EVENT_FL_TRACEPOINT; user->call.tp = &user->tracepoint; user->call.event.funcs = &user_event_funcs; - user->class.system = group->system_name; + + if (EVENT_MULTI_FORMAT(user->reg_flags)) + user->class.system = group->system_multi_name; + else + user->class.system = group->system_name; user->class.fields_array = user_event_fields_array; user->class.get_fields = user_event_get_fields; @@ -1335,20 +2154,28 @@ static int user_event_parse(struct user_event_group *group, char *name, mutex_lock(&event_mutex); + if (current_user_events >= max_user_events) { + ret = -EMFILE; + goto put_user_lock; + } + ret = user_event_trace_register(user); if (ret) goto put_user_lock; - user->index = index; - - /* Ensure we track self ref and caller ref (2) */ - refcount_set(&user->refcnt, 2); + if (user->reg_flags & USER_EVENT_REG_PERSIST) { + /* Ensure we track self ref and caller ref (2) */ + refcount_set(&user->refcnt, 2); + } else { + /* Ensure we track only caller ref (1) */ + refcount_set(&user->refcnt, 1); + } dyn_event_init(&user->devent, &user_event_dops); dyn_event_add(&user->devent, &user->call); - set_bit(user->index, group->page_bitmap); hash_add(group->register_table, &user->node, key); + current_user_events++; mutex_unlock(&event_mutex); @@ -1360,27 +2187,43 @@ put_user: user_event_destroy_fields(user); user_event_destroy_validators(user); kfree(user->call.print_fmt); + + /* Caller frees reg_name on error, but not multi-name */ + if (EVENT_NAME(user) != EVENT_TP_NAME(user)) + kfree(EVENT_TP_NAME(user)); + kfree(user); return ret; } /* - * Deletes a previously created event if it is no longer being used. + * Deletes previously created events if they are no longer being used. */ static int delete_user_event(struct user_event_group *group, char *name) { - u32 key; - struct user_event *user = find_user_event(group, name, &key); + struct user_event *user; + struct hlist_node *tmp; + u32 key = user_event_key(name); + int ret = -ENOENT; - if (!user) - return -ENOENT; + /* Attempt to delete all event(s) with the name passed in */ + hash_for_each_possible_safe(group->register_table, user, tmp, node, key) { + if (strcmp(EVENT_NAME(user), name)) + continue; - refcount_dec(&user->refcnt); + if (!user_event_last_ref(user)) + return -EBUSY; - if (!user_event_last_ref(user)) - return -EBUSY; + if (!user_event_capable(user->reg_flags)) + return -EPERM; - return destroy_user_event(user); + ret = destroy_user_event(user); + + if (ret) + goto out; + } +out: + return ret; } /* @@ -1398,6 +2241,9 @@ static ssize_t user_events_write_core(struct file *file, struct iov_iter *i) if (unlikely(copy_from_iter(&idx, sizeof(idx), i) != sizeof(idx))) return -EFAULT; + if (idx < 0) + return -EINVAL; + rcu_read_lock_sched(); refs = rcu_dereference_sched(info->refs); @@ -1424,7 +2270,7 @@ static ssize_t user_events_write_core(struct file *file, struct iov_iter *i) * It's possible key.enabled disables after this check, however * we don't mind if a few events are included in this condition. */ - if (likely(atomic_read(&tp->key.enabled) > 0)) { + if (likely(static_key_enabled(&tp->key))) { struct tracepoint_func *probe_func_ptr; user_event_func_t probe_func; struct iov_iter copy; @@ -1453,7 +2299,8 @@ static ssize_t user_events_write_core(struct file *file, struct iov_iter *i) if (unlikely(faulted)) return -EFAULT; - } + } else + return -EBADF; return ret; } @@ -1468,7 +2315,7 @@ static int user_events_open(struct inode *node, struct file *file) if (!group) return -ENOENT; - info = kzalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL_ACCOUNT); if (!info) return -ENOMEM; @@ -1483,14 +2330,12 @@ static int user_events_open(struct inode *node, struct file *file) static ssize_t user_events_write(struct file *file, const char __user *ubuf, size_t count, loff_t *ppos) { - struct iovec iov; struct iov_iter i; if (unlikely(*ppos != 0)) return -EFAULT; - if (unlikely(import_single_range(ITER_SOURCE, (char __user *)ubuf, - count, &iov, &i))) + if (unlikely(import_ubuf(ITER_SOURCE, (char __user *)ubuf, count, &i))) return -EFAULT; return user_events_write_core(file, &i); @@ -1521,7 +2366,7 @@ static int user_events_ref_add(struct user_event_file_info *info, size = struct_size(refs, events, count + 1); - new_refs = kzalloc(size, GFP_KERNEL); + new_refs = kzalloc(size, GFP_KERNEL_ACCOUNT); if (!new_refs) return -ENOMEM; @@ -1531,9 +2376,7 @@ static int user_events_ref_add(struct user_event_file_info *info, for (i = 0; i < count; ++i) new_refs->events[i] = refs->events[i]; - new_refs->events[i] = user; - - refcount_inc(&user->refcnt); + new_refs->events[i] = user_event_get(user); rcu_assign_pointer(info->refs, new_refs); @@ -1564,6 +2407,37 @@ static long user_reg_get(struct user_reg __user *ureg, struct user_reg *kreg) if (ret) return ret; + /* Ensure only valid flags */ + if (kreg->flags & ~(USER_EVENT_REG_MAX-1)) + return -EINVAL; + + /* Ensure supported size */ + switch (kreg->enable_size) { + case 4: + /* 32-bit */ + break; +#if BITS_PER_LONG >= 64 + case 8: + /* 64-bit */ + break; +#endif + default: + return -EINVAL; + } + + /* Ensure natural alignment */ + if (kreg->enable_addr % kreg->enable_size) + return -EINVAL; + + /* Ensure bit range for size */ + if (kreg->enable_bit > (kreg->enable_size * BITS_PER_BYTE) - 1) + return -EINVAL; + + /* Ensure accessible */ + if (!access_ok((const void __user *)(uintptr_t)kreg->enable_addr, + kreg->enable_size)) + return -EFAULT; + kreg->size = size; return 0; @@ -1578,14 +2452,26 @@ static long user_events_ioctl_reg(struct user_event_file_info *info, struct user_reg __user *ureg = (struct user_reg __user *)uarg; struct user_reg reg; struct user_event *user; + struct user_event_enabler *enabler; char *name; long ret; + int write_result; ret = user_reg_get(ureg, ®); if (ret) return ret; + /* + * Prevent users from using the same address and bit multiple times + * within the same mm address space. This can cause unexpected behavior + * for user processes that is far easier to debug if this is explicitly + * an error upon registering. + */ + if (current_user_event_enabler_exists((unsigned long)reg.enable_addr, + reg.enable_bit)) + return -EADDRINUSE; + name = strndup_user((const char __user *)(uintptr_t)reg.name_args, MAX_EVENT_DESC); @@ -1594,7 +2480,7 @@ static long user_events_ioctl_reg(struct user_event_file_info *info, return ret; } - ret = user_event_parse_cmd(info->group, name, &user); + ret = user_event_parse_cmd(info->group, name, &user, reg.flags); if (ret) { kfree(name); @@ -1604,14 +2490,34 @@ static long user_events_ioctl_reg(struct user_event_file_info *info, ret = user_events_ref_add(info, user); /* No longer need parse ref, ref_add either worked or not */ - refcount_dec(&user->refcnt); + user_event_put(user, false); /* Positive number is index and valid */ if (ret < 0) return ret; + /* + * user_events_ref_add succeeded: + * At this point we have a user_event, it's lifetime is bound by the + * reference count, not this file. If anything fails, the user_event + * still has a reference until the file is released. During release + * any remaining references (from user_events_ref_add) are decremented. + * + * Attempt to create an enabler, which too has a lifetime tied in the + * same way for the event. Once the task that caused the enabler to be + * created exits or issues exec() then the enablers it has created + * will be destroyed and the ref to the event will be decremented. + */ + enabler = user_event_enabler_create(®, user, &write_result); + + if (!enabler) + return -ENOMEM; + + /* Write failed/faulted, give error back to caller */ + if (write_result) + return write_result; + put_user((u32)ret, &ureg->write_index); - put_user(user->index, &ureg->status_bit); return 0; } @@ -1641,6 +2547,121 @@ static long user_events_ioctl_del(struct user_event_file_info *info, return ret; } +static long user_unreg_get(struct user_unreg __user *ureg, + struct user_unreg *kreg) +{ + u32 size; + long ret; + + ret = get_user(size, &ureg->size); + + if (ret) + return ret; + + if (size > PAGE_SIZE) + return -E2BIG; + + if (size < offsetofend(struct user_unreg, disable_addr)) + return -EINVAL; + + ret = copy_struct_from_user(kreg, sizeof(*kreg), ureg, size); + + /* Ensure no reserved values, since we don't support any yet */ + if (kreg->__reserved || kreg->__reserved2) + return -EINVAL; + + return ret; +} + +static int user_event_mm_clear_bit(struct user_event_mm *user_mm, + unsigned long uaddr, unsigned char bit, + unsigned long flags) +{ + struct user_event_enabler enabler; + int result; + int attempt = 0; + + memset(&enabler, 0, sizeof(enabler)); + enabler.addr = uaddr; + enabler.values = bit | flags; +retry: + /* Prevents state changes from racing with new enablers */ + mutex_lock(&event_mutex); + + /* Force the bit to be cleared, since no event is attached */ + mmap_read_lock(user_mm->mm); + result = user_event_enabler_write(user_mm, &enabler, false, &attempt); + mmap_read_unlock(user_mm->mm); + + mutex_unlock(&event_mutex); + + if (result) { + /* Attempt to fault-in and retry if it worked */ + if (!user_event_mm_fault_in(user_mm, uaddr, attempt)) + goto retry; + } + + return result; +} + +/* + * Unregisters an enablement address/bit within a task/user mm. + */ +static long user_events_ioctl_unreg(unsigned long uarg) +{ + struct user_unreg __user *ureg = (struct user_unreg __user *)uarg; + struct user_event_mm *mm = current->user_event_mm; + struct user_event_enabler *enabler, *next; + struct user_unreg reg; + unsigned long flags; + long ret; + + ret = user_unreg_get(ureg, ®); + + if (ret) + return ret; + + if (!mm) + return -ENOENT; + + flags = 0; + ret = -ENOENT; + + /* + * Flags freeing and faulting are used to indicate if the enabler is in + * use at all. When faulting is set a page-fault is occurring asyncly. + * During async fault if freeing is set, the enabler will be destroyed. + * If no async fault is happening, we can destroy it now since we hold + * the event_mutex during these checks. + */ + mutex_lock(&event_mutex); + + list_for_each_entry_safe(enabler, next, &mm->enablers, mm_enablers_link) { + if (enabler->addr == reg.disable_addr && + ENABLE_BIT(enabler) == reg.disable_bit) { + set_bit(ENABLE_VAL_FREEING_BIT, ENABLE_BITOPS(enabler)); + + /* We must keep compat flags for the clear */ + flags |= enabler->values & ENABLE_VAL_COMPAT_MASK; + + if (!test_bit(ENABLE_VAL_FAULTING_BIT, ENABLE_BITOPS(enabler))) + user_event_enabler_destroy(enabler, true); + + /* Removed at least one */ + ret = 0; + } + } + + mutex_unlock(&event_mutex); + + /* Ensure bit is now cleared for user, regardless of event status */ + if (!ret) + ret = user_event_mm_clear_bit(mm, reg.disable_addr, + reg.disable_bit, flags); + + return ret; +} + /* * Handles the ioctl from user mode to register or alter operations. */ @@ -1663,6 +2684,12 @@ static long user_events_ioctl(struct file *file, unsigned int cmd, ret = user_events_ioctl_del(info, uarg); mutex_unlock(&group->reg_mutex); break; + + case DIAG_IOCSUNREG: + mutex_lock(&group->reg_mutex); + ret = user_events_ioctl_unreg(uarg); + mutex_unlock(&group->reg_mutex); + break; } return ret; @@ -1676,7 +2703,6 @@ static int user_events_release(struct inode *node, struct file *file) struct user_event_file_info *info = file->private_data; struct user_event_group *group; struct user_event_refs *refs; - struct user_event *user; int i; if (!info) @@ -1700,12 +2726,9 @@ static int user_events_release(struct inode *node, struct file *file) * The underlying user_events are ref counted, and cannot be freed. * After this decrement, the user_events may be freed elsewhere. */ - for (i = 0; i < refs->count; ++i) { - user = refs->events[i]; + for (i = 0; i < refs->count; ++i) + user_event_put(refs->events[i], false); - if (user) - refcount_dec(&user->refcnt); - } out: file->private_data = NULL; @@ -1718,45 +2741,13 @@ out: } static const struct file_operations user_data_fops = { - .open = user_events_open, - .write = user_events_write, - .write_iter = user_events_write_iter, + .open = user_events_open, + .write = user_events_write, + .write_iter = user_events_write_iter, .unlocked_ioctl = user_events_ioctl, - .release = user_events_release, + .release = user_events_release, }; -static struct user_event_group *user_status_group(struct file *file) -{ - struct seq_file *m = file->private_data; - - if (!m) - return NULL; - - return m->private; -} - -/* - * Maps the shared page into the user process for checking if event is enabled. - */ -static int user_status_mmap(struct file *file, struct vm_area_struct *vma) -{ - char *pages; - struct user_event_group *group = user_status_group(file); - unsigned long size = vma->vm_end - vma->vm_start; - - if (size != MAX_BYTES) - return -EINVAL; - - if (!group) - return -EINVAL; - - pages = group->register_page_data; - - return remap_pfn_range(vma, vma->vm_start, - virt_to_phys(pages) >> PAGE_SHIFT, - size, vm_get_page_prot(VM_READ)); -} - static void *user_seq_start(struct seq_file *m, loff_t *pos) { if (*pos) @@ -1780,7 +2771,7 @@ static int user_seq_show(struct seq_file *m, void *p) struct user_event_group *group = m->private; struct user_event *user; char status; - int i, active = 0, busy = 0, flags; + int i, active = 0, busy = 0; if (!group) return -EINVAL; @@ -1789,15 +2780,11 @@ static int user_seq_show(struct seq_file *m, void *p) hash_for_each(group->register_table, i, user, node) { status = user->status; - flags = user->flags; - - seq_printf(m, "%d:%s", user->index, EVENT_NAME(user)); - if (flags != 0 || status != 0) - seq_puts(m, " #"); + seq_printf(m, "%s", EVENT_TP_NAME(user)); if (status != 0) { - seq_puts(m, " Used by"); + seq_puts(m, " # Used by"); if (status & EVENT_STATUS_FTRACE) seq_puts(m, " ftrace"); if (status & EVENT_STATUS_PERF) @@ -1816,16 +2803,15 @@ static int user_seq_show(struct seq_file *m, void *p) seq_puts(m, "\n"); seq_printf(m, "Active: %d\n", active); seq_printf(m, "Busy: %d\n", busy); - seq_printf(m, "Max: %ld\n", MAX_EVENTS); return 0; } static const struct seq_operations user_seq_ops = { - .start = user_seq_start, - .next = user_seq_next, - .stop = user_seq_stop, - .show = user_seq_show, + .start = user_seq_start, + .next = user_seq_next, + .stop = user_seq_stop, + .show = user_seq_show, }; static int user_status_open(struct inode *node, struct file *file) @@ -1851,11 +2837,10 @@ static int user_status_open(struct inode *node, struct file *file) } static const struct file_operations user_status_fops = { - .open = user_status_open, - .mmap = user_status_mmap, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, + .open = user_status_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, }; /* @@ -1873,8 +2858,7 @@ static int create_user_tracefs(void) goto err; } - /* mmap with MAP_SHARED requires writable fd */ - emmap = tracefs_create_file("user_events_status", TRACE_MODE_WRITE, + emmap = tracefs_create_file("user_events_status", TRACE_MODE_READ, NULL, NULL, &user_status_fops); if (!emmap) { @@ -1888,20 +2872,52 @@ err: return -ENODEV; } +static int set_max_user_events_sysctl(const struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + int ret; + + mutex_lock(&event_mutex); + + ret = proc_douintvec(table, write, buffer, lenp, ppos); + + mutex_unlock(&event_mutex); + + return ret; +} + +static const struct ctl_table user_event_sysctls[] = { + { + .procname = "user_events_max", + .data = &max_user_events, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = set_max_user_events_sysctl, + }, +}; + static int __init trace_events_user_init(void) { int ret; - init_group = user_event_group_create(&init_user_ns); + fault_cache = KMEM_CACHE(user_event_enabler_fault, 0); - if (!init_group) + if (!fault_cache) return -ENOMEM; + init_group = user_event_group_create(); + + if (!init_group) { + kmem_cache_destroy(fault_cache); + return -ENOMEM; + } + ret = create_user_tracefs(); if (ret) { pr_warn("user_events could not register with tracefs\n"); user_event_group_destroy(init_group); + kmem_cache_destroy(fault_cache); init_group = NULL; return ret; } @@ -1909,6 +2925,8 @@ static int __init trace_events_user_init(void) if (dyn_event_register(&user_event_dops)) pr_warn("user_events could not register with dyn_events\n"); + register_sysctl_init("kernel", user_event_sysctls); + return 0; } diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index d960f6b11b5e..1698fc22afa0 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -51,6 +51,9 @@ static int ftrace_event_register(struct trace_event_call *call, #undef __array #define __array(type, item, size) type item[size]; +#undef __stack_array +#define __stack_array(type, item, size, field) __array(type, item, size) + #undef __array_desc #define __array_desc(type, container, item, size) type item[size]; @@ -111,7 +114,11 @@ static void __always_unused ____ftrace_check_##name(void) \ #define __array(_type, _item, _len) { \ .type = #_type"["__stringify(_len)"]", .name = #_item, \ .size = sizeof(_type[_len]), .align = __alignof__(_type), \ - is_signed_type(_type), .filter_type = FILTER_OTHER }, + is_signed_type(_type), .filter_type = FILTER_OTHER, \ + .len = _len }, + +#undef __stack_array +#define __stack_array(_type, _item, _len, _field) __array(_type, _item, _len) #undef __array_desc #define __array_desc(_type, _container, _item, _len) __array(_type, _item, _len) @@ -148,6 +155,9 @@ static struct trace_event_fields ftrace_event_fields_##name[] = { \ #undef __array #define __array(type, item, len) +#undef __stack_array +#define __stack_array(type, item, len, field) + #undef __array_desc #define __array_desc(type, container, item, len) diff --git a/kernel/trace/trace_fprobe.c b/kernel/trace/trace_fprobe.c new file mode 100644 index 000000000000..262c0556e4af --- /dev/null +++ b/kernel/trace/trace_fprobe.c @@ -0,0 +1,1589 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Fprobe-based tracing events + * Copyright (C) 2022 Google LLC. + */ +#define pr_fmt(fmt) "trace_fprobe: " fmt + +#include <linux/fprobe.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/rculist.h> +#include <linux/security.h> +#include <linux/tracepoint.h> +#include <linux/uaccess.h> + +#include <asm/ptrace.h> + +#include "trace_dynevent.h" +#include "trace_probe.h" +#include "trace_probe_kernel.h" +#include "trace_probe_tmpl.h" + +#define FPROBE_EVENT_SYSTEM "fprobes" +#define TRACEPOINT_EVENT_SYSTEM "tracepoints" +#define RETHOOK_MAXACTIVE_MAX 4096 + +static int trace_fprobe_create(const char *raw_command); +static int trace_fprobe_show(struct seq_file *m, struct dyn_event *ev); +static int trace_fprobe_release(struct dyn_event *ev); +static bool trace_fprobe_is_busy(struct dyn_event *ev); +static bool trace_fprobe_match(const char *system, const char *event, + int argc, const char **argv, struct dyn_event *ev); + +static struct dyn_event_operations trace_fprobe_ops = { + .create = trace_fprobe_create, + .show = trace_fprobe_show, + .is_busy = trace_fprobe_is_busy, + .free = trace_fprobe_release, + .match = trace_fprobe_match, +}; + +/* List of tracepoint_user */ +static LIST_HEAD(tracepoint_user_list); +static DEFINE_MUTEX(tracepoint_user_mutex); + +/* While living tracepoint_user, @tpoint can be NULL and @refcount != 0. */ +struct tracepoint_user { + struct list_head list; + const char *name; + struct tracepoint *tpoint; + unsigned int refcount; +}; + +/* NOTE: you must lock tracepoint_user_mutex. */ +#define for_each_tracepoint_user(tuser) \ + list_for_each_entry(tuser, &tracepoint_user_list, list) + +static int tracepoint_user_register(struct tracepoint_user *tuser) +{ + struct tracepoint *tpoint = tuser->tpoint; + + if (!tpoint) + return 0; + + return tracepoint_probe_register_prio_may_exist(tpoint, + tpoint->probestub, NULL, 0); +} + +static void tracepoint_user_unregister(struct tracepoint_user *tuser) +{ + if (!tuser->tpoint) + return; + + WARN_ON_ONCE(tracepoint_probe_unregister(tuser->tpoint, tuser->tpoint->probestub, NULL)); + tuser->tpoint = NULL; +} + +static unsigned long tracepoint_user_ip(struct tracepoint_user *tuser) +{ + if (!tuser->tpoint) + return 0UL; + + return (unsigned long)tuser->tpoint->probestub; +} + +static void __tracepoint_user_free(struct tracepoint_user *tuser) +{ + if (!tuser) + return; + kfree(tuser->name); + kfree(tuser); +} + +DEFINE_FREE(tuser_free, struct tracepoint_user *, __tracepoint_user_free(_T)) + +static struct tracepoint_user *__tracepoint_user_init(const char *name, struct tracepoint *tpoint) +{ + struct tracepoint_user *tuser __free(tuser_free) = NULL; + int ret; + + tuser = kzalloc(sizeof(*tuser), GFP_KERNEL); + if (!tuser) + return NULL; + tuser->name = kstrdup(name, GFP_KERNEL); + if (!tuser->name) + return NULL; + + /* Register tracepoint if it is loaded. */ + if (tpoint) { + tuser->tpoint = tpoint; + ret = tracepoint_user_register(tuser); + if (ret) + return ERR_PTR(ret); + } + + tuser->refcount = 1; + INIT_LIST_HEAD(&tuser->list); + list_add(&tuser->list, &tracepoint_user_list); + + return_ptr(tuser); +} + +static struct tracepoint *find_tracepoint(const char *tp_name, + struct module **tp_mod); + +/* + * Get tracepoint_user if exist, or allocate new one and register it. + * If tracepoint is on a module, get its refcounter too. + * This returns errno or NULL (not loaded yet) or tracepoint_user. + */ +static struct tracepoint_user *tracepoint_user_find_get(const char *name, struct module **pmod) +{ + struct module *mod __free(module_put) = NULL; + struct tracepoint_user *tuser; + struct tracepoint *tpoint; + + if (!name || !pmod) + return ERR_PTR(-EINVAL); + + /* Get and lock the module which has tracepoint. */ + tpoint = find_tracepoint(name, &mod); + + guard(mutex)(&tracepoint_user_mutex); + /* Search existing tracepoint_user */ + for_each_tracepoint_user(tuser) { + if (!strcmp(tuser->name, name)) { + tuser->refcount++; + *pmod = no_free_ptr(mod); + return tuser; + } + } + + /* The corresponding tracepoint_user is not found. */ + tuser = __tracepoint_user_init(name, tpoint); + if (!IS_ERR_OR_NULL(tuser)) + *pmod = no_free_ptr(mod); + + return tuser; +} + +static void tracepoint_user_put(struct tracepoint_user *tuser) +{ + scoped_guard(mutex, &tracepoint_user_mutex) { + if (--tuser->refcount > 0) + return; + + list_del(&tuser->list); + tracepoint_user_unregister(tuser); + } + + __tracepoint_user_free(tuser); +} + +DEFINE_FREE(tuser_put, struct tracepoint_user *, + if (!IS_ERR_OR_NULL(_T)) + tracepoint_user_put(_T)) + +/* + * Fprobe event core functions + */ + +/* + * @tprobe is true for tracepoint probe. + * @tuser can be NULL if the trace_fprobe is disabled or the tracepoint is not + * loaded with a module. If @tuser != NULL, this trace_fprobe is enabled. + */ +struct trace_fprobe { + struct dyn_event devent; + struct fprobe fp; + const char *symbol; + bool tprobe; + struct tracepoint_user *tuser; + struct trace_probe tp; +}; + +static bool is_trace_fprobe(struct dyn_event *ev) +{ + return ev->ops == &trace_fprobe_ops; +} + +static struct trace_fprobe *to_trace_fprobe(struct dyn_event *ev) +{ + return container_of(ev, struct trace_fprobe, devent); +} + +/** + * for_each_trace_fprobe - iterate over the trace_fprobe list + * @pos: the struct trace_fprobe * for each entry + * @dpos: the struct dyn_event * to use as a loop cursor + */ +#define for_each_trace_fprobe(pos, dpos) \ + for_each_dyn_event(dpos) \ + if (is_trace_fprobe(dpos) && (pos = to_trace_fprobe(dpos))) + +static bool trace_fprobe_is_return(struct trace_fprobe *tf) +{ + return tf->fp.exit_handler != NULL; +} + +static bool trace_fprobe_is_tracepoint(struct trace_fprobe *tf) +{ + return tf->tprobe; +} + +static const char *trace_fprobe_symbol(struct trace_fprobe *tf) +{ + return tf->symbol ? tf->symbol : "unknown"; +} + +static bool trace_fprobe_is_busy(struct dyn_event *ev) +{ + struct trace_fprobe *tf = to_trace_fprobe(ev); + + return trace_probe_is_enabled(&tf->tp); +} + +static bool trace_fprobe_match_command_head(struct trace_fprobe *tf, + int argc, const char **argv) +{ + char buf[MAX_ARGSTR_LEN + 1]; + + if (!argc) + return true; + + snprintf(buf, sizeof(buf), "%s", trace_fprobe_symbol(tf)); + if (strcmp(buf, argv[0])) + return false; + argc--; argv++; + + return trace_probe_match_command_args(&tf->tp, argc, argv); +} + +static bool trace_fprobe_match(const char *system, const char *event, + int argc, const char **argv, struct dyn_event *ev) +{ + struct trace_fprobe *tf = to_trace_fprobe(ev); + + if (event[0] != '\0' && strcmp(trace_probe_name(&tf->tp), event)) + return false; + + if (system && strcmp(trace_probe_group_name(&tf->tp), system)) + return false; + + return trace_fprobe_match_command_head(tf, argc, argv); +} + +static bool trace_fprobe_is_registered(struct trace_fprobe *tf) +{ + return fprobe_is_registered(&tf->fp); +} + +/* + * Note that we don't verify the fetch_insn code, since it does not come + * from user space. + */ +static int +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) +{ + struct ftrace_regs *fregs = rec; + unsigned long val; + int ret; + +retry: + /* 1st stage: get value from context */ + switch (code->op) { + case FETCH_OP_STACK: + val = ftrace_regs_get_kernel_stack_nth(fregs, code->param); + break; + case FETCH_OP_STACKP: + val = ftrace_regs_get_stack_pointer(fregs); + break; + case FETCH_OP_RETVAL: + val = ftrace_regs_get_return_value(fregs); + break; +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + case FETCH_OP_ARG: + val = ftrace_regs_get_argument(fregs, code->param); + break; + case FETCH_OP_EDATA: + val = *(unsigned long *)((unsigned long)edata + code->offset); + break; +#endif + case FETCH_NOP_SYMBOL: /* Ignore a place holder */ + code++; + goto retry; + default: + ret = process_common_fetch_insn(code, &val); + if (ret < 0) + return ret; + } + code++; + + return process_fetch_insn_bottom(code, val, dest, base); +} +NOKPROBE_SYMBOL(process_fetch_insn) + +/* function entry handler */ +static nokprobe_inline void +__fentry_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, + struct ftrace_regs *fregs, + struct trace_event_file *trace_file) +{ + struct fentry_trace_entry_head *entry; + struct trace_event_call *call = trace_probe_event_call(&tf->tp); + struct trace_event_buffer fbuffer; + int dsize; + + if (WARN_ON_ONCE(call != trace_file->event_call)) + return; + + if (trace_trigger_soft_disabled(trace_file)) + return; + + dsize = __get_data_size(&tf->tp, fregs, NULL); + + entry = trace_event_buffer_reserve(&fbuffer, trace_file, + sizeof(*entry) + tf->tp.size + dsize); + if (!entry) + return; + + fbuffer.regs = ftrace_get_regs(fregs); + entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); + entry->ip = entry_ip; + store_trace_args(&entry[1], &tf->tp, fregs, NULL, sizeof(*entry), dsize); + + trace_event_buffer_commit(&fbuffer); +} + +static void +fentry_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, + struct ftrace_regs *fregs) +{ + struct event_file_link *link; + + trace_probe_for_each_link_rcu(link, &tf->tp) + __fentry_trace_func(tf, entry_ip, fregs, link->file); +} +NOKPROBE_SYMBOL(fentry_trace_func); + +static nokprobe_inline +void store_fprobe_entry_data(void *edata, struct trace_probe *tp, struct ftrace_regs *fregs) +{ + struct probe_entry_arg *earg = tp->entry_arg; + unsigned long val = 0; + int i; + + if (!earg) + return; + + for (i = 0; i < earg->size; i++) { + struct fetch_insn *code = &earg->code[i]; + + switch (code->op) { + case FETCH_OP_ARG: + val = ftrace_regs_get_argument(fregs, code->param); + break; + case FETCH_OP_ST_EDATA: + *(unsigned long *)((unsigned long)edata + code->offset) = val; + break; + case FETCH_OP_END: + goto end; + default: + break; + } + } +end: + return; +} + +/* function exit handler */ +static int trace_fprobe_entry_handler(struct fprobe *fp, unsigned long entry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, + void *entry_data) +{ + struct trace_fprobe *tf = container_of(fp, struct trace_fprobe, fp); + + if (tf->tp.entry_arg) + store_fprobe_entry_data(entry_data, &tf->tp, fregs); + + return 0; +} +NOKPROBE_SYMBOL(trace_fprobe_entry_handler) + +static nokprobe_inline void +__fexit_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, + void *entry_data, struct trace_event_file *trace_file) +{ + struct fexit_trace_entry_head *entry; + struct trace_event_buffer fbuffer; + struct trace_event_call *call = trace_probe_event_call(&tf->tp); + int dsize; + + if (WARN_ON_ONCE(call != trace_file->event_call)) + return; + + if (trace_trigger_soft_disabled(trace_file)) + return; + + dsize = __get_data_size(&tf->tp, fregs, entry_data); + + entry = trace_event_buffer_reserve(&fbuffer, trace_file, + sizeof(*entry) + tf->tp.size + dsize); + if (!entry) + return; + + fbuffer.regs = ftrace_get_regs(fregs); + entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); + entry->func = entry_ip; + entry->ret_ip = ret_ip; + store_trace_args(&entry[1], &tf->tp, fregs, entry_data, sizeof(*entry), dsize); + + trace_event_buffer_commit(&fbuffer); +} + +static void +fexit_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, void *entry_data) +{ + struct event_file_link *link; + + trace_probe_for_each_link_rcu(link, &tf->tp) + __fexit_trace_func(tf, entry_ip, ret_ip, fregs, entry_data, link->file); +} +NOKPROBE_SYMBOL(fexit_trace_func); + +#ifdef CONFIG_PERF_EVENTS + +static int fentry_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, + struct ftrace_regs *fregs) +{ + struct trace_event_call *call = trace_probe_event_call(&tf->tp); + struct fentry_trace_entry_head *entry; + struct hlist_head *head; + int size, __size, dsize; + struct pt_regs *regs; + int rctx; + + head = this_cpu_ptr(call->perf_events); + if (hlist_empty(head)) + return 0; + + dsize = __get_data_size(&tf->tp, fregs, NULL); + __size = sizeof(*entry) + tf->tp.size + dsize; + size = ALIGN(__size + sizeof(u32), sizeof(u64)); + size -= sizeof(u32); + + entry = perf_trace_buf_alloc(size, ®s, &rctx); + if (!entry) + return 0; + + regs = ftrace_fill_perf_regs(fregs, regs); + + entry->ip = entry_ip; + memset(&entry[1], 0, dsize); + store_trace_args(&entry[1], &tf->tp, fregs, NULL, sizeof(*entry), dsize); + perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, + head, NULL); + return 0; +} +NOKPROBE_SYMBOL(fentry_perf_func); + +static void +fexit_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, + void *entry_data) +{ + struct trace_event_call *call = trace_probe_event_call(&tf->tp); + struct fexit_trace_entry_head *entry; + struct hlist_head *head; + int size, __size, dsize; + struct pt_regs *regs; + int rctx; + + head = this_cpu_ptr(call->perf_events); + if (hlist_empty(head)) + return; + + dsize = __get_data_size(&tf->tp, fregs, entry_data); + __size = sizeof(*entry) + tf->tp.size + dsize; + size = ALIGN(__size + sizeof(u32), sizeof(u64)); + size -= sizeof(u32); + + entry = perf_trace_buf_alloc(size, ®s, &rctx); + if (!entry) + return; + + regs = ftrace_fill_perf_regs(fregs, regs); + + entry->func = entry_ip; + entry->ret_ip = ret_ip; + store_trace_args(&entry[1], &tf->tp, fregs, entry_data, sizeof(*entry), dsize); + perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, + head, NULL); +} +NOKPROBE_SYMBOL(fexit_perf_func); +#endif /* CONFIG_PERF_EVENTS */ + +static int fentry_dispatcher(struct fprobe *fp, unsigned long entry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, + void *entry_data) +{ + struct trace_fprobe *tf = container_of(fp, struct trace_fprobe, fp); + unsigned int flags = trace_probe_load_flag(&tf->tp); + int ret = 0; + + if (flags & TP_FLAG_TRACE) + fentry_trace_func(tf, entry_ip, fregs); + +#ifdef CONFIG_PERF_EVENTS + if (flags & TP_FLAG_PROFILE) + ret = fentry_perf_func(tf, entry_ip, fregs); +#endif + return ret; +} +NOKPROBE_SYMBOL(fentry_dispatcher); + +static void fexit_dispatcher(struct fprobe *fp, unsigned long entry_ip, + unsigned long ret_ip, struct ftrace_regs *fregs, + void *entry_data) +{ + struct trace_fprobe *tf = container_of(fp, struct trace_fprobe, fp); + unsigned int flags = trace_probe_load_flag(&tf->tp); + + if (flags & TP_FLAG_TRACE) + fexit_trace_func(tf, entry_ip, ret_ip, fregs, entry_data); +#ifdef CONFIG_PERF_EVENTS + if (flags & TP_FLAG_PROFILE) + fexit_perf_func(tf, entry_ip, ret_ip, fregs, entry_data); +#endif +} +NOKPROBE_SYMBOL(fexit_dispatcher); + +static void free_trace_fprobe(struct trace_fprobe *tf) +{ + if (tf) { + trace_probe_cleanup(&tf->tp); + if (tf->tuser) + tracepoint_user_put(tf->tuser); + kfree(tf->symbol); + kfree(tf); + } +} + +/* Since alloc_trace_fprobe() can return error, check the pointer is ERR too. */ +DEFINE_FREE(free_trace_fprobe, struct trace_fprobe *, if (!IS_ERR_OR_NULL(_T)) free_trace_fprobe(_T)) + +/* + * Allocate new trace_probe and initialize it (including fprobe). + */ +static struct trace_fprobe *alloc_trace_fprobe(const char *group, + const char *event, + const char *symbol, + int nargs, bool is_return, + bool is_tracepoint) +{ + struct trace_fprobe *tf __free(free_trace_fprobe) = NULL; + int ret = -ENOMEM; + + tf = kzalloc(struct_size(tf, tp.args, nargs), GFP_KERNEL); + if (!tf) + return ERR_PTR(ret); + + tf->symbol = kstrdup(symbol, GFP_KERNEL); + if (!tf->symbol) + return ERR_PTR(-ENOMEM); + + if (is_return) + tf->fp.exit_handler = fexit_dispatcher; + else + tf->fp.entry_handler = fentry_dispatcher; + + tf->tprobe = is_tracepoint; + + ret = trace_probe_init(&tf->tp, event, group, false, nargs); + if (ret < 0) + return ERR_PTR(ret); + + dyn_event_init(&tf->devent, &trace_fprobe_ops); + return_ptr(tf); +} + +static struct trace_fprobe *find_trace_fprobe(const char *event, + const char *group) +{ + struct dyn_event *pos; + struct trace_fprobe *tf; + + for_each_trace_fprobe(tf, pos) + if (strcmp(trace_probe_name(&tf->tp), event) == 0 && + strcmp(trace_probe_group_name(&tf->tp), group) == 0) + return tf; + return NULL; +} + +/* Event entry printers */ +static enum print_line_t +print_fentry_event(struct trace_iterator *iter, int flags, + struct trace_event *event) +{ + struct fentry_trace_entry_head *field; + struct trace_seq *s = &iter->seq; + struct trace_probe *tp; + + field = (struct fentry_trace_entry_head *)iter->ent; + tp = trace_probe_primary_from_call( + container_of(event, struct trace_event_call, event)); + if (WARN_ON_ONCE(!tp)) + goto out; + + trace_seq_printf(s, "%s: (", trace_probe_name(tp)); + + if (!seq_print_ip_sym_offset(s, field->ip, flags)) + goto out; + + trace_seq_putc(s, ')'); + + if (trace_probe_print_args(s, tp->args, tp->nr_args, + (u8 *)&field[1], field) < 0) + goto out; + + trace_seq_putc(s, '\n'); + out: + return trace_handle_return(s); +} + +static enum print_line_t +print_fexit_event(struct trace_iterator *iter, int flags, + struct trace_event *event) +{ + struct fexit_trace_entry_head *field; + struct trace_seq *s = &iter->seq; + struct trace_probe *tp; + + field = (struct fexit_trace_entry_head *)iter->ent; + tp = trace_probe_primary_from_call( + container_of(event, struct trace_event_call, event)); + if (WARN_ON_ONCE(!tp)) + goto out; + + trace_seq_printf(s, "%s: (", trace_probe_name(tp)); + + if (!seq_print_ip_sym_offset(s, field->ret_ip, flags)) + goto out; + + trace_seq_puts(s, " <- "); + + if (!seq_print_ip_sym_no_offset(s, field->func, flags)) + goto out; + + trace_seq_putc(s, ')'); + + if (trace_probe_print_args(s, tp->args, tp->nr_args, + (u8 *)&field[1], field) < 0) + goto out; + + trace_seq_putc(s, '\n'); + + out: + return trace_handle_return(s); +} + +static int fentry_event_define_fields(struct trace_event_call *event_call) +{ + int ret; + struct fentry_trace_entry_head field; + struct trace_probe *tp; + + tp = trace_probe_primary_from_call(event_call); + if (WARN_ON_ONCE(!tp)) + return -ENOENT; + + DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0); + + return traceprobe_define_arg_fields(event_call, sizeof(field), tp); +} + +static int fexit_event_define_fields(struct trace_event_call *event_call) +{ + int ret; + struct fexit_trace_entry_head field; + struct trace_probe *tp; + + tp = trace_probe_primary_from_call(event_call); + if (WARN_ON_ONCE(!tp)) + return -ENOENT; + + DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0); + DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0); + + return traceprobe_define_arg_fields(event_call, sizeof(field), tp); +} + +static struct trace_event_functions fentry_funcs = { + .trace = print_fentry_event +}; + +static struct trace_event_functions fexit_funcs = { + .trace = print_fexit_event +}; + +static struct trace_event_fields fentry_fields_array[] = { + { .type = TRACE_FUNCTION_TYPE, + .define_fields = fentry_event_define_fields }, + {} +}; + +static struct trace_event_fields fexit_fields_array[] = { + { .type = TRACE_FUNCTION_TYPE, + .define_fields = fexit_event_define_fields }, + {} +}; + +static int fprobe_register(struct trace_event_call *event, + enum trace_reg type, void *data); + +static inline void init_trace_event_call(struct trace_fprobe *tf) +{ + struct trace_event_call *call = trace_probe_event_call(&tf->tp); + + if (trace_fprobe_is_return(tf)) { + call->event.funcs = &fexit_funcs; + call->class->fields_array = fexit_fields_array; + } else { + call->event.funcs = &fentry_funcs; + call->class->fields_array = fentry_fields_array; + } + + call->flags = TRACE_EVENT_FL_FPROBE; + call->class->reg = fprobe_register; +} + +static int register_fprobe_event(struct trace_fprobe *tf) +{ + init_trace_event_call(tf); + + return trace_probe_register_event_call(&tf->tp); +} + +static int unregister_fprobe_event(struct trace_fprobe *tf) +{ + return trace_probe_unregister_event_call(&tf->tp); +} + +static int __regsiter_tracepoint_fprobe(struct trace_fprobe *tf) +{ + struct tracepoint_user *tuser __free(tuser_put) = NULL; + struct module *mod __free(module_put) = NULL; + unsigned long ip; + int ret; + + if (WARN_ON_ONCE(tf->tuser)) + return -EINVAL; + + /* If the tracepoint is in a module, it must be locked in this function. */ + tuser = tracepoint_user_find_get(tf->symbol, &mod); + /* This tracepoint is not loaded yet */ + if (IS_ERR(tuser)) + return PTR_ERR(tuser); + if (!tuser) + return -ENOMEM; + + /* Register fprobe only if the tracepoint is loaded. */ + if (tuser->tpoint) { + ip = tracepoint_user_ip(tuser); + if (WARN_ON_ONCE(!ip)) + return -ENOENT; + + ret = register_fprobe_ips(&tf->fp, &ip, 1); + if (ret < 0) + return ret; + } + + tf->tuser = no_free_ptr(tuser); + return 0; +} + +/* Returns an error if the target function is not available, or 0 */ +static int trace_fprobe_verify_target(struct trace_fprobe *tf) +{ + int ret; + + /* Tracepoint should have a stub function. */ + if (trace_fprobe_is_tracepoint(tf)) + return 0; + + /* + * Note: since we don't lock the module, even if this succeeded, + * register_fprobe() later can fail. + */ + ret = fprobe_count_ips_from_filter(tf->symbol, NULL); + return (ret < 0) ? ret : 0; +} + +/* Internal register function - just handle fprobe and flags */ +static int __register_trace_fprobe(struct trace_fprobe *tf) +{ + int i, ret; + + /* Should we need new LOCKDOWN flag for fprobe? */ + ret = security_locked_down(LOCKDOWN_KPROBES); + if (ret) + return ret; + + if (trace_fprobe_is_registered(tf)) + return -EINVAL; + + for (i = 0; i < tf->tp.nr_args; i++) { + ret = traceprobe_update_arg(&tf->tp.args[i]); + if (ret) + return ret; + } + + tf->fp.flags &= ~FPROBE_FL_DISABLED; + + if (trace_fprobe_is_tracepoint(tf)) + return __regsiter_tracepoint_fprobe(tf); + + /* TODO: handle filter, nofilter or symbol list */ + return register_fprobe(&tf->fp, tf->symbol, NULL); +} + +/* Internal unregister function - just handle fprobe and flags */ +static void __unregister_trace_fprobe(struct trace_fprobe *tf) +{ + if (trace_fprobe_is_registered(tf)) + unregister_fprobe(&tf->fp); + if (tf->tuser) { + tracepoint_user_put(tf->tuser); + tf->tuser = NULL; + } +} + +/* TODO: make this trace_*probe common function */ +/* Unregister a trace_probe and probe_event */ +static int unregister_trace_fprobe(struct trace_fprobe *tf) +{ + /* If other probes are on the event, just unregister fprobe */ + if (trace_probe_has_sibling(&tf->tp)) + goto unreg; + + /* Enabled event can not be unregistered */ + if (trace_probe_is_enabled(&tf->tp)) + return -EBUSY; + + /* If there's a reference to the dynamic event */ + if (trace_event_dyn_busy(trace_probe_event_call(&tf->tp))) + return -EBUSY; + + /* Will fail if probe is being used by ftrace or perf */ + if (unregister_fprobe_event(tf)) + return -EBUSY; + +unreg: + __unregister_trace_fprobe(tf); + dyn_event_remove(&tf->devent); + trace_probe_unlink(&tf->tp); + + return 0; +} + +static bool trace_fprobe_has_same_fprobe(struct trace_fprobe *orig, + struct trace_fprobe *comp) +{ + struct trace_probe_event *tpe = orig->tp.event; + int i; + + list_for_each_entry(orig, &tpe->probes, tp.list) { + if (strcmp(trace_fprobe_symbol(orig), + trace_fprobe_symbol(comp))) + continue; + + /* + * trace_probe_compare_arg_type() ensured that nr_args and + * each argument name and type are same. Let's compare comm. + */ + for (i = 0; i < orig->tp.nr_args; i++) { + if (strcmp(orig->tp.args[i].comm, + comp->tp.args[i].comm)) + break; + } + + if (i == orig->tp.nr_args) + return true; + } + + return false; +} + +static int append_trace_fprobe_event(struct trace_fprobe *tf, struct trace_fprobe *to) +{ + int ret; + + if (trace_fprobe_is_return(tf) != trace_fprobe_is_return(to) || + trace_fprobe_is_tracepoint(tf) != trace_fprobe_is_tracepoint(to)) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, DIFF_PROBE_TYPE); + return -EEXIST; + } + ret = trace_probe_compare_arg_type(&tf->tp, &to->tp); + if (ret) { + /* Note that argument starts index = 2 */ + trace_probe_log_set_index(ret + 1); + trace_probe_log_err(0, DIFF_ARG_TYPE); + return -EEXIST; + } + if (trace_fprobe_has_same_fprobe(to, tf)) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, SAME_PROBE); + return -EEXIST; + } + + /* Append to existing event */ + ret = trace_probe_append(&tf->tp, &to->tp); + if (ret) + return ret; + + ret = trace_fprobe_verify_target(tf); + if (ret) + trace_probe_unlink(&tf->tp); + else + dyn_event_add(&tf->devent, trace_probe_event_call(&tf->tp)); + + return ret; +} + +/* Register a trace_probe and probe_event, and check the fprobe is available. */ +static int register_trace_fprobe_event(struct trace_fprobe *tf) +{ + struct trace_fprobe *old_tf; + int ret; + + guard(mutex)(&event_mutex); + + old_tf = find_trace_fprobe(trace_probe_name(&tf->tp), + trace_probe_group_name(&tf->tp)); + if (old_tf) + return append_trace_fprobe_event(tf, old_tf); + + /* Register new event */ + ret = register_fprobe_event(tf); + if (ret) { + if (ret == -EEXIST) { + trace_probe_log_set_index(0); + trace_probe_log_err(0, EVENT_EXIST); + } else + pr_warn("Failed to register probe event(%d)\n", ret); + return ret; + } + + /* Verify fprobe is sane. */ + ret = trace_fprobe_verify_target(tf); + if (ret < 0) + unregister_fprobe_event(tf); + else + dyn_event_add(&tf->devent, trace_probe_event_call(&tf->tp)); + + return ret; +} + +struct __find_tracepoint_cb_data { + const char *tp_name; + struct tracepoint *tpoint; + struct module *mod; +}; + +static void __find_tracepoint_module_cb(struct tracepoint *tp, struct module *mod, void *priv) +{ + struct __find_tracepoint_cb_data *data = priv; + + if (!data->tpoint && !strcmp(data->tp_name, tp->name)) { + /* If module is not specified, try getting module refcount. */ + if (!data->mod && mod) { + /* If failed to get refcount, ignore this tracepoint. */ + if (!try_module_get(mod)) + return; + + data->mod = mod; + } + data->tpoint = tp; + } +} + +static void __find_tracepoint_cb(struct tracepoint *tp, void *priv) +{ + struct __find_tracepoint_cb_data *data = priv; + + if (!data->tpoint && !strcmp(data->tp_name, tp->name)) + data->tpoint = tp; +} + +/* + * Find a tracepoint from kernel and module. If the tracepoint is on the module, + * the module's refcount is incremented and returned as *@tp_mod. Thus, if it is + * not NULL, caller must call module_put(*tp_mod) after used the tracepoint. + */ +static struct tracepoint *find_tracepoint(const char *tp_name, + struct module **tp_mod) +{ + struct __find_tracepoint_cb_data data = { + .tp_name = tp_name, + .mod = NULL, + }; + + for_each_kernel_tracepoint(__find_tracepoint_cb, &data); + + if (!data.tpoint && IS_ENABLED(CONFIG_MODULES)) { + for_each_module_tracepoint(__find_tracepoint_module_cb, &data); + *tp_mod = data.mod; + } + + return data.tpoint; +} + +#ifdef CONFIG_MODULES +/* + * Find a tracepoint from specified module. In this case, this does not get the + * module's refcount. The caller must ensure the module is not freed. + */ +static struct tracepoint *find_tracepoint_in_module(struct module *mod, + const char *tp_name) +{ + struct __find_tracepoint_cb_data data = { + .tp_name = tp_name, + .mod = mod, + }; + + for_each_tracepoint_in_module(mod, __find_tracepoint_module_cb, &data); + return data.tpoint; +} + +/* These are CONFIG_MODULES=y specific functions. */ +static bool tracepoint_user_within_module(struct tracepoint_user *tuser, + struct module *mod) +{ + return within_module(tracepoint_user_ip(tuser), mod); +} + +static int tracepoint_user_register_again(struct tracepoint_user *tuser, + struct tracepoint *tpoint) +{ + tuser->tpoint = tpoint; + return tracepoint_user_register(tuser); +} + +static void tracepoint_user_unregister_clear(struct tracepoint_user *tuser) +{ + tracepoint_user_unregister(tuser); + tuser->tpoint = NULL; +} + +/* module callback for tracepoint_user */ +static int __tracepoint_probe_module_cb(struct notifier_block *self, + unsigned long val, void *data) +{ + struct tp_module *tp_mod = data; + struct tracepoint_user *tuser; + struct tracepoint *tpoint; + + if (val != MODULE_STATE_GOING && val != MODULE_STATE_COMING) + return NOTIFY_DONE; + + mutex_lock(&tracepoint_user_mutex); + for_each_tracepoint_user(tuser) { + if (val == MODULE_STATE_COMING) { + /* This is not a tracepoint in this module. Skip it. */ + tpoint = find_tracepoint_in_module(tp_mod->mod, tuser->name); + if (!tpoint) + continue; + WARN_ON_ONCE(tracepoint_user_register_again(tuser, tpoint)); + } else if (val == MODULE_STATE_GOING && + tracepoint_user_within_module(tuser, tp_mod->mod)) { + /* Unregister all tracepoint_user in this module. */ + tracepoint_user_unregister_clear(tuser); + } + } + mutex_unlock(&tracepoint_user_mutex); + + return NOTIFY_DONE; +} + +static struct notifier_block tracepoint_module_nb = { + .notifier_call = __tracepoint_probe_module_cb, +}; + +/* module callback for tprobe events */ +static int __tprobe_event_module_cb(struct notifier_block *self, + unsigned long val, void *data) +{ + struct trace_fprobe *tf; + struct dyn_event *pos; + struct module *mod = data; + + if (val != MODULE_STATE_GOING && val != MODULE_STATE_COMING) + return NOTIFY_DONE; + + mutex_lock(&event_mutex); + for_each_trace_fprobe(tf, pos) { + /* Skip fprobe and disabled tprobe events. */ + if (!trace_fprobe_is_tracepoint(tf) || !tf->tuser) + continue; + + /* Before this notification, tracepoint notifier has already done. */ + if (val == MODULE_STATE_COMING && + tracepoint_user_within_module(tf->tuser, mod)) { + unsigned long ip = tracepoint_user_ip(tf->tuser); + + WARN_ON_ONCE(register_fprobe_ips(&tf->fp, &ip, 1)); + } else if (val == MODULE_STATE_GOING && + /* + * tracepoint_user_within_module() does not work here because + * tracepoint_user is already unregistered and cleared tpoint. + * Instead, checking whether the fprobe is registered but + * tpoint is cleared(unregistered). Such unbalance probes + * must be adjusted anyway. + */ + trace_fprobe_is_registered(tf) && + !tf->tuser->tpoint) { + unregister_fprobe(&tf->fp); + } + } + mutex_unlock(&event_mutex); + + return NOTIFY_DONE; +} + +/* NOTE: this must be called after tracepoint callback */ +static struct notifier_block tprobe_event_module_nb = { + .notifier_call = __tprobe_event_module_cb, + /* Make sure this is later than tracepoint module notifier. */ + .priority = -10, +}; +#endif /* CONFIG_MODULES */ + +static int parse_symbol_and_return(int argc, const char *argv[], + char **symbol, bool *is_return, + bool is_tracepoint) +{ + char *tmp = strchr(argv[1], '%'); + int i; + + if (tmp) { + int len = tmp - argv[1]; + + if (!is_tracepoint && !strcmp(tmp, "%return")) { + *is_return = true; + } else { + trace_probe_log_err(len, BAD_ADDR_SUFFIX); + return -EINVAL; + } + *symbol = kmemdup_nul(argv[1], len, GFP_KERNEL); + } else + *symbol = kstrdup(argv[1], GFP_KERNEL); + if (!*symbol) + return -ENOMEM; + + if (*is_return) + return 0; + + if (is_tracepoint) { + tmp = *symbol; + while (*tmp && (isalnum(*tmp) || *tmp == '_')) + tmp++; + if (*tmp) { + /* find a wrong character. */ + trace_probe_log_err(tmp - *symbol, BAD_TP_NAME); + kfree(*symbol); + *symbol = NULL; + return -EINVAL; + } + } + + /* If there is $retval, this should be a return fprobe. */ + for (i = 2; i < argc; i++) { + tmp = strstr(argv[i], "$retval"); + if (tmp && !isalnum(tmp[7]) && tmp[7] != '_') { + if (is_tracepoint) { + trace_probe_log_set_index(i); + trace_probe_log_err(tmp - argv[i], RETVAL_ON_PROBE); + kfree(*symbol); + *symbol = NULL; + return -EINVAL; + } + *is_return = true; + break; + } + } + return 0; +} + +static int trace_fprobe_create_internal(int argc, const char *argv[], + struct traceprobe_parse_context *ctx) +{ + /* + * Argument syntax: + * - Add fentry probe: + * f[:[GRP/][EVENT]] [MOD:]KSYM [FETCHARGS] + * - Add fexit probe: + * f[N][:[GRP/][EVENT]] [MOD:]KSYM%return [FETCHARGS] + * - Add tracepoint probe: + * t[:[GRP/][EVENT]] TRACEPOINT [FETCHARGS] + * + * Fetch args: + * $retval : fetch return value + * $stack : fetch stack address + * $stackN : fetch Nth entry of stack (N:0-) + * $argN : fetch Nth argument (N:1-) + * $comm : fetch current task comm + * @ADDR : fetch memory at ADDR (ADDR should be in kernel) + * @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol) + * Dereferencing memory fetch: + * +|-offs(ARG) : fetch memory at ARG +|- offs address. + * Alias name of args: + * NAME=FETCHARG : set NAME as alias of FETCHARG. + * Type of args: + * FETCHARG:TYPE : use TYPE instead of unsigned long. + */ + struct trace_fprobe *tf __free(free_trace_fprobe) = NULL; + const char *event = NULL, *group = FPROBE_EVENT_SYSTEM; + struct module *mod __free(module_put) = NULL; + const char **new_argv __free(kfree) = NULL; + char *symbol __free(kfree) = NULL; + char *ebuf __free(kfree) = NULL; + char *gbuf __free(kfree) = NULL; + char *sbuf __free(kfree) = NULL; + char *abuf __free(kfree) = NULL; + char *dbuf __free(kfree) = NULL; + int i, new_argc = 0, ret = 0; + bool is_tracepoint = false; + bool is_return = false; + + if ((argv[0][0] != 'f' && argv[0][0] != 't') || argc < 2) + return -ECANCELED; + + if (argv[0][0] == 't') { + is_tracepoint = true; + group = TRACEPOINT_EVENT_SYSTEM; + } + + if (argv[0][1] != '\0') { + if (argv[0][1] != ':') { + trace_probe_log_set_index(0); + trace_probe_log_err(1, BAD_MAXACT); + return -EINVAL; + } + event = &argv[0][2]; + } + + trace_probe_log_set_index(1); + + /* a symbol(or tracepoint) must be specified */ + ret = parse_symbol_and_return(argc, argv, &symbol, &is_return, is_tracepoint); + if (ret < 0) + return -EINVAL; + + trace_probe_log_set_index(0); + if (event) { + gbuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!gbuf) + return -ENOMEM; + ret = traceprobe_parse_event_name(&event, &group, gbuf, + event - argv[0]); + if (ret) + return -EINVAL; + } + + if (!event) { + ebuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!ebuf) + return -ENOMEM; + /* Make a new event name */ + if (is_tracepoint) + snprintf(ebuf, MAX_EVENT_NAME_LEN, "%s%s", + isdigit(*symbol) ? "_" : "", symbol); + else + snprintf(ebuf, MAX_EVENT_NAME_LEN, "%s__%s", symbol, + is_return ? "exit" : "entry"); + sanitize_event_name(ebuf); + event = ebuf; + } + + if (is_return) + ctx->flags |= TPARG_FL_RETURN; + else + ctx->flags |= TPARG_FL_FENTRY; + + ctx->funcname = NULL; + if (is_tracepoint) { + /* Get tracepoint and lock its module until the end of the registration. */ + struct tracepoint *tpoint; + + ctx->flags |= TPARG_FL_TPOINT; + mod = NULL; + tpoint = find_tracepoint(symbol, &mod); + if (tpoint) { + sbuf = kmalloc(KSYM_NAME_LEN, GFP_KERNEL); + if (!sbuf) + return -ENOMEM; + ctx->funcname = kallsyms_lookup((unsigned long)tpoint->probestub, + NULL, NULL, NULL, sbuf); + } + } + if (!ctx->funcname) + ctx->funcname = symbol; + + abuf = kmalloc(MAX_BTF_ARGS_LEN, GFP_KERNEL); + if (!abuf) + return -ENOMEM; + argc -= 2; argv += 2; + new_argv = traceprobe_expand_meta_args(argc, argv, &new_argc, + abuf, MAX_BTF_ARGS_LEN, ctx); + if (IS_ERR(new_argv)) + return PTR_ERR(new_argv); + if (new_argv) { + argc = new_argc; + argv = new_argv; + } + if (argc > MAX_TRACE_ARGS) { + trace_probe_log_set_index(2); + trace_probe_log_err(0, TOO_MANY_ARGS); + return -E2BIG; + } + + ret = traceprobe_expand_dentry_args(argc, argv, &dbuf); + if (ret) + return ret; + + /* setup a probe */ + tf = alloc_trace_fprobe(group, event, symbol, argc, is_return, is_tracepoint); + if (IS_ERR(tf)) { + ret = PTR_ERR(tf); + /* This must return -ENOMEM, else there is a bug */ + WARN_ON_ONCE(ret != -ENOMEM); + return ret; + } + + /* parse arguments */ + for (i = 0; i < argc; i++) { + trace_probe_log_set_index(i + 2); + ctx->offset = 0; + ret = traceprobe_parse_probe_arg(&tf->tp, i, argv[i], ctx); + if (ret) + return ret; /* This can be -ENOMEM */ + } + + if (is_return && tf->tp.entry_arg) { + tf->fp.entry_handler = trace_fprobe_entry_handler; + tf->fp.entry_data_size = traceprobe_get_entry_data_size(&tf->tp); + if (ALIGN(tf->fp.entry_data_size, sizeof(long)) > MAX_FPROBE_DATA_SIZE) { + trace_probe_log_set_index(2); + trace_probe_log_err(0, TOO_MANY_EARGS); + return -E2BIG; + } + } + + ret = traceprobe_set_print_fmt(&tf->tp, + is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL); + if (ret < 0) + return ret; + + ret = register_trace_fprobe_event(tf); + if (ret) { + trace_probe_log_set_index(1); + if (ret == -EILSEQ) + trace_probe_log_err(0, BAD_INSN_BNDRY); + else if (ret == -ENOENT) + trace_probe_log_err(0, BAD_PROBE_ADDR); + else if (ret != -ENOMEM && ret != -EEXIST) + trace_probe_log_err(0, FAIL_REG_PROBE); + return -EINVAL; + } + + /* 'tf' is successfully registered. To avoid freeing, assign NULL. */ + tf = NULL; + + return 0; +} + +static int trace_fprobe_create_cb(int argc, const char *argv[]) +{ + struct traceprobe_parse_context *ctx __free(traceprobe_parse_context) = NULL; + int ret; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->flags = TPARG_FL_KERNEL | TPARG_FL_FPROBE; + + trace_probe_log_init("trace_fprobe", argc, argv); + ret = trace_fprobe_create_internal(argc, argv, ctx); + trace_probe_log_clear(); + return ret; +} + +static int trace_fprobe_create(const char *raw_command) +{ + return trace_probe_create(raw_command, trace_fprobe_create_cb); +} + +static int trace_fprobe_release(struct dyn_event *ev) +{ + struct trace_fprobe *tf = to_trace_fprobe(ev); + int ret = unregister_trace_fprobe(tf); + + if (!ret) + free_trace_fprobe(tf); + return ret; +} + +static int trace_fprobe_show(struct seq_file *m, struct dyn_event *ev) +{ + struct trace_fprobe *tf = to_trace_fprobe(ev); + int i; + + if (trace_fprobe_is_tracepoint(tf)) + seq_putc(m, 't'); + else + seq_putc(m, 'f'); + seq_printf(m, ":%s/%s", trace_probe_group_name(&tf->tp), + trace_probe_name(&tf->tp)); + + seq_printf(m, " %s%s", trace_fprobe_symbol(tf), + trace_fprobe_is_return(tf) ? "%return" : ""); + + for (i = 0; i < tf->tp.nr_args; i++) + seq_printf(m, " %s=%s", tf->tp.args[i].name, tf->tp.args[i].comm); + seq_putc(m, '\n'); + + return 0; +} + +/* + * Enable trace_probe + * if the file is NULL, enable "perf" handler, or enable "trace" handler. + */ +static int enable_trace_fprobe(struct trace_event_call *call, + struct trace_event_file *file) +{ + struct trace_probe *tp; + struct trace_fprobe *tf; + bool enabled; + int ret = 0; + + tp = trace_probe_primary_from_call(call); + if (WARN_ON_ONCE(!tp)) + return -ENODEV; + enabled = trace_probe_is_enabled(tp); + + /* This also changes "enabled" state */ + if (file) { + ret = trace_probe_add_file(tp, file); + if (ret) + return ret; + } else + trace_probe_set_flag(tp, TP_FLAG_PROFILE); + + if (!enabled) { + list_for_each_entry(tf, trace_probe_probe_list(tp), tp.list) { + ret = __register_trace_fprobe(tf); + if (ret < 0) + return ret; + } + } + + return 0; +} + +/* + * Disable trace_probe + * if the file is NULL, disable "perf" handler, or disable "trace" handler. + */ +static int disable_trace_fprobe(struct trace_event_call *call, + struct trace_event_file *file) +{ + struct trace_fprobe *tf; + struct trace_probe *tp; + + tp = trace_probe_primary_from_call(call); + if (WARN_ON_ONCE(!tp)) + return -ENODEV; + + if (file) { + if (!trace_probe_get_file_link(tp, file)) + return -ENOENT; + if (!trace_probe_has_single_file(tp)) + goto out; + trace_probe_clear_flag(tp, TP_FLAG_TRACE); + } else + trace_probe_clear_flag(tp, TP_FLAG_PROFILE); + + if (!trace_probe_is_enabled(tp)) { + list_for_each_entry(tf, trace_probe_probe_list(tp), tp.list) { + unregister_fprobe(&tf->fp); + if (tf->tuser) { + tracepoint_user_put(tf->tuser); + tf->tuser = NULL; + } + } + } + + out: + if (file) + /* + * Synchronization is done in below function. For perf event, + * file == NULL and perf_trace_event_unreg() calls + * tracepoint_synchronize_unregister() to ensure synchronize + * event. We don't need to care about it. + */ + trace_probe_remove_file(tp, file); + + return 0; +} + +/* + * called by perf_trace_init() or __ftrace_set_clr_event() under event_mutex. + */ +static int fprobe_register(struct trace_event_call *event, + enum trace_reg type, void *data) +{ + struct trace_event_file *file = data; + + switch (type) { + case TRACE_REG_REGISTER: + return enable_trace_fprobe(event, file); + case TRACE_REG_UNREGISTER: + return disable_trace_fprobe(event, file); + +#ifdef CONFIG_PERF_EVENTS + case TRACE_REG_PERF_REGISTER: + return enable_trace_fprobe(event, NULL); + case TRACE_REG_PERF_UNREGISTER: + return disable_trace_fprobe(event, NULL); + case TRACE_REG_PERF_OPEN: + case TRACE_REG_PERF_CLOSE: + case TRACE_REG_PERF_ADD: + case TRACE_REG_PERF_DEL: + return 0; +#endif + } + return 0; +} + +/* + * Register dynevent at core_initcall. This allows kernel to setup fprobe + * events in postcore_initcall without tracefs. + */ +static __init int init_fprobe_trace_early(void) +{ + int ret; + + ret = dyn_event_register(&trace_fprobe_ops); + if (ret) + return ret; + +#ifdef CONFIG_MODULES + ret = register_tracepoint_module_notifier(&tracepoint_module_nb); + if (ret) + return ret; + ret = register_module_notifier(&tprobe_event_module_nb); + if (ret) + return ret; +#endif + + return 0; +} +core_initcall(init_fprobe_trace_early); diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 9f1bfbe105e8..c12795c2fb39 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -25,6 +25,9 @@ static void function_trace_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs); static void +function_args_trace_call(unsigned long ip, unsigned long parent_ip, + struct ftrace_ops *op, struct ftrace_regs *fregs); +static void function_stack_trace_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs); static void @@ -42,9 +45,10 @@ enum { TRACE_FUNC_NO_OPTS = 0x0, /* No flags set. */ TRACE_FUNC_OPT_STACK = 0x1, TRACE_FUNC_OPT_NO_REPEATS = 0x2, + TRACE_FUNC_OPT_ARGS = 0x4, /* Update this to next highest bit. */ - TRACE_FUNC_OPT_HIGHEST_BIT = 0x4 + TRACE_FUNC_OPT_HIGHEST_BIT = 0x8 }; #define TRACE_FUNC_OPT_MASK (TRACE_FUNC_OPT_HIGHEST_BIT - 1) @@ -80,6 +84,7 @@ void ftrace_free_ftrace_ops(struct trace_array *tr) int ftrace_create_function_files(struct trace_array *tr, struct dentry *parent) { + int ret; /* * The top level array uses the "global_ops", and the files are * created on boot up. @@ -90,6 +95,12 @@ int ftrace_create_function_files(struct trace_array *tr, if (!tr->ops) return -EINVAL; + ret = allocate_fgraph_ops(tr, tr->ops); + if (ret) { + kfree(tr->ops); + return ret; + } + ftrace_create_filter_files(tr->ops, parent); return 0; @@ -99,6 +110,7 @@ void ftrace_destroy_function_files(struct trace_array *tr) { ftrace_destroy_filter_files(tr->ops); ftrace_free_ftrace_ops(tr); + free_fgraph_ops(tr); } static ftrace_func_t select_trace_function(u32 flags_val) @@ -106,6 +118,8 @@ static ftrace_func_t select_trace_function(u32 flags_val) switch (flags_val & TRACE_FUNC_OPT_MASK) { case TRACE_FUNC_NO_OPTS: return function_trace_call; + case TRACE_FUNC_OPT_ARGS: + return function_args_trace_call; case TRACE_FUNC_OPT_STACK: return function_stack_trace_call; case TRACE_FUNC_OPT_NO_REPEATS: @@ -140,11 +154,11 @@ static int function_trace_init(struct trace_array *tr) if (!tr->ops) return -ENOMEM; - func = select_trace_function(func_flags.val); + func = select_trace_function(tr->current_trace_flags->val); if (!func) return -EINVAL; - if (!handle_func_repeats(tr, func_flags.val)) + if (!handle_func_repeats(tr, tr->current_trace_flags->val)) return -ENOMEM; ftrace_init_array_ops(tr, func); @@ -168,15 +182,59 @@ static void function_trace_start(struct trace_array *tr) tracing_reset_online_cpus(&tr->array_buffer); } +/* fregs are guaranteed not to be NULL if HAVE_DYNAMIC_FTRACE_WITH_ARGS is set */ +#if defined(CONFIG_FUNCTION_GRAPH_TRACER) && defined(CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS) +static __always_inline unsigned long +function_get_true_parent_ip(unsigned long parent_ip, struct ftrace_regs *fregs) +{ + unsigned long true_parent_ip; + int idx = 0; + + true_parent_ip = parent_ip; + if (unlikely(parent_ip == (unsigned long)&return_to_handler) && fregs) + true_parent_ip = ftrace_graph_ret_addr(current, &idx, parent_ip, + (unsigned long *)ftrace_regs_get_stack_pointer(fregs)); + return true_parent_ip; +} +#else +static __always_inline unsigned long +function_get_true_parent_ip(unsigned long parent_ip, struct ftrace_regs *fregs) +{ + return parent_ip; +} +#endif + static void function_trace_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs) { struct trace_array *tr = op->private; - struct trace_array_cpu *data; unsigned int trace_ctx; int bit; - int cpu; + + if (unlikely(!tr->function_enabled)) + return; + + bit = ftrace_test_recursion_trylock(ip, parent_ip); + if (bit < 0) + return; + + parent_ip = function_get_true_parent_ip(parent_ip, fregs); + + trace_ctx = tracing_gen_ctx_dec(); + + trace_function(tr, ip, parent_ip, trace_ctx, NULL); + + ftrace_test_recursion_unlock(bit); +} + +static void +function_args_trace_call(unsigned long ip, unsigned long parent_ip, + struct ftrace_ops *op, struct ftrace_regs *fregs) +{ + struct trace_array *tr = op->private; + unsigned int trace_ctx; + int bit; if (unlikely(!tr->function_enabled)) return; @@ -187,10 +245,7 @@ function_trace_call(unsigned long ip, unsigned long parent_ip, trace_ctx = tracing_gen_ctx(); - cpu = smp_processor_id(); - data = per_cpu_ptr(tr->array_buffer.data, cpu); - if (!atomic_read(&data->disabled)) - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, fregs); ftrace_test_recursion_unlock(bit); } @@ -223,6 +278,7 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, long disabled; int cpu; unsigned int trace_ctx; + int skip = STACK_SKIP; if (unlikely(!tr->function_enabled)) return; @@ -232,17 +288,22 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, * recursive protection is performed. */ local_irq_save(flags); + parent_ip = function_get_true_parent_ip(parent_ip, fregs); cpu = raw_smp_processor_id(); data = per_cpu_ptr(tr->array_buffer.data, cpu); - disabled = atomic_inc_return(&data->disabled); + disabled = local_inc_return(&data->disabled); if (likely(disabled == 1)) { trace_ctx = tracing_gen_ctx_flags(flags); - trace_function(tr, ip, parent_ip, trace_ctx); - __trace_stack(tr, trace_ctx, STACK_SKIP); + trace_function(tr, ip, parent_ip, trace_ctx, NULL); +#ifdef CONFIG_UNWINDER_FRAME_POINTER + if (ftrace_pids_enabled(op)) + skip++; +#endif + __trace_stack(tr, trace_ctx, skip); } - atomic_dec(&data->disabled); + local_dec(&data->disabled); local_irq_restore(flags); } @@ -283,11 +344,8 @@ function_no_repeats_trace_call(unsigned long ip, unsigned long parent_ip, { struct trace_func_repeats *last_info; struct trace_array *tr = op->private; - struct trace_array_cpu *data; unsigned int trace_ctx; - unsigned long flags; int bit; - int cpu; if (unlikely(!tr->function_enabled)) return; @@ -296,9 +354,8 @@ function_no_repeats_trace_call(unsigned long ip, unsigned long parent_ip, if (bit < 0) return; - cpu = smp_processor_id(); - data = per_cpu_ptr(tr->array_buffer.data, cpu); - if (atomic_read(&data->disabled)) + parent_ip = function_get_true_parent_ip(parent_ip, fregs); + if (!tracer_tracing_is_on(tr)) goto out; /* @@ -308,15 +365,14 @@ function_no_repeats_trace_call(unsigned long ip, unsigned long parent_ip, * TODO: think about a solution that is better than just hoping to be * lucky. */ - last_info = per_cpu_ptr(tr->last_func_repeats, cpu); + last_info = this_cpu_ptr(tr->last_func_repeats); if (is_repeat_check(tr, last_info, ip, parent_ip)) goto out; - local_save_flags(flags); - trace_ctx = tracing_gen_ctx_flags(flags); + trace_ctx = tracing_gen_ctx_dec(); process_repeats(tr, ip, parent_ip, last_info, trace_ctx); - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, NULL); out: ftrace_test_recursion_unlock(bit); @@ -343,9 +399,10 @@ function_stack_no_repeats_trace_call(unsigned long ip, unsigned long parent_ip, * recursive protection is performed. */ local_irq_save(flags); + parent_ip = function_get_true_parent_ip(parent_ip, fregs); cpu = raw_smp_processor_id(); data = per_cpu_ptr(tr->array_buffer.data, cpu); - disabled = atomic_inc_return(&data->disabled); + disabled = local_inc_return(&data->disabled); if (likely(disabled == 1)) { last_info = per_cpu_ptr(tr->last_func_repeats, cpu); @@ -355,12 +412,12 @@ function_stack_no_repeats_trace_call(unsigned long ip, unsigned long parent_ip, trace_ctx = tracing_gen_ctx_flags(flags); process_repeats(tr, ip, parent_ip, last_info, trace_ctx); - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, NULL); __trace_stack(tr, trace_ctx, STACK_SKIP); } out: - atomic_dec(&data->disabled); + local_dec(&data->disabled); local_irq_restore(flags); } @@ -369,6 +426,9 @@ static struct tracer_opt func_opts[] = { { TRACER_OPT(func_stack_trace, TRACE_FUNC_OPT_STACK) }, #endif { TRACER_OPT(func-no-repeats, TRACE_FUNC_OPT_NO_REPEATS) }, +#ifdef CONFIG_FUNCTION_TRACE_ARGS + { TRACER_OPT(func-args, TRACE_FUNC_OPT_ARGS) }, +#endif { } /* Always set a last empty entry */ }; @@ -399,14 +459,14 @@ func_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set) u32 new_flags; /* Do nothing if already set. */ - if (!!set == !!(func_flags.val & bit)) + if (!!set == !!(tr->current_trace_flags->val & bit)) return 0; /* We can change this flag only when not running. */ if (tr->current_trace != &function_trace) return 0; - new_flags = (func_flags.val & ~bit) | (set ? bit : 0); + new_flags = (tr->current_trace_flags->val & ~bit) | (set ? bit : 0); func = select_trace_function(new_flags); if (!func) return -EINVAL; @@ -431,7 +491,7 @@ static struct tracer function_trace __tracer_data = .init = function_trace_init, .reset = function_trace_reset, .start = function_trace_start, - .flags = &func_flags, + .default_flags = &func_flags, .set_flag = func_set_flag, .allow_instances = true, #ifdef CONFIG_FTRACE_SELFTEST @@ -563,11 +623,7 @@ ftrace_traceoff(unsigned long ip, unsigned long parent_ip, static __always_inline void trace_stack(struct trace_array *tr) { - unsigned int trace_ctx; - - trace_ctx = tracing_gen_ctx(); - - __trace_stack(tr, trace_ctx, FTRACE_STACK_SKIP); + __trace_stack(tr, tracing_gen_ctx_dec(), FTRACE_STACK_SKIP); } static void diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 203204cadf92..b1e9c9913309 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -16,9 +16,12 @@ #include "trace.h" #include "trace_output.h" -/* When set, irq functions will be ignored */ +/* When set, irq functions might be ignored */ static int ftrace_graph_skip_irqs; +/* Do not record function time when task is sleeping */ +int fgraph_no_sleep_time; + struct fgraph_cpu_data { pid_t last_pid; int depth; @@ -27,11 +30,26 @@ struct fgraph_cpu_data { unsigned long enter_funcs[FTRACE_RETFUNC_DEPTH]; }; +struct fgraph_ent_args { + struct ftrace_graph_ent_entry ent; + /* Force the sizeof of args[] to have FTRACE_REGS_MAX_ARGS entries */ + unsigned long args[FTRACE_REGS_MAX_ARGS]; +}; + +struct fgraph_retaddr_ent_args { + struct fgraph_retaddr_ent_entry ent; + /* Force the sizeof of args[] to have FTRACE_REGS_MAX_ARGS entries */ + unsigned long args[FTRACE_REGS_MAX_ARGS]; +}; + struct fgraph_data { struct fgraph_cpu_data __percpu *cpu_data; /* Place to preserve last processed entry. */ - struct ftrace_graph_ent_entry ent; + union { + struct fgraph_ent_args ent; + struct fgraph_retaddr_ent_args rent; + }; struct ftrace_graph_ret_entry ret; int failed; int cpu; @@ -58,14 +76,23 @@ static struct tracer_opt trace_opts[] = { { TRACER_OPT(funcgraph-irqs, TRACE_GRAPH_PRINT_IRQS) }, /* Display function name after trailing } */ { TRACER_OPT(funcgraph-tail, TRACE_GRAPH_PRINT_TAIL) }, +#ifdef CONFIG_FUNCTION_GRAPH_RETVAL + /* Display function return value ? */ + { TRACER_OPT(funcgraph-retval, TRACE_GRAPH_PRINT_RETVAL) }, + /* Display function return value in hexadecimal format ? */ + { TRACER_OPT(funcgraph-retval-hex, TRACE_GRAPH_PRINT_RETVAL_HEX) }, +#endif +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR + /* Display function return address ? */ + { TRACER_OPT(funcgraph-retaddr, TRACE_GRAPH_PRINT_RETADDR) }, +#endif +#ifdef CONFIG_FUNCTION_TRACE_ARGS + /* Display function arguments ? */ + { TRACER_OPT(funcgraph-args, TRACE_GRAPH_ARGS) }, +#endif /* Include sleep time (scheduled out) between entry and return */ { TRACER_OPT(sleep-time, TRACE_GRAPH_SLEEP_TIME) }, -#ifdef CONFIG_FUNCTION_PROFILER - /* Include time within nested functions */ - { TRACER_OPT(graph-time, TRACE_GRAPH_GRAPH_TIME) }, -#endif - { } /* Empty entry */ }; @@ -73,11 +100,14 @@ static struct tracer_flags tracer_flags = { /* Don't display overruns, proc, or tail by default */ .val = TRACE_GRAPH_PRINT_CPU | TRACE_GRAPH_PRINT_OVERHEAD | TRACE_GRAPH_PRINT_DURATION | TRACE_GRAPH_PRINT_IRQS | - TRACE_GRAPH_SLEEP_TIME | TRACE_GRAPH_GRAPH_TIME, + TRACE_GRAPH_SLEEP_TIME, .opts = trace_opts }; -static struct trace_array *graph_array; +static bool tracer_flags_is_set(struct trace_array *tr, u32 flags) +{ + return (tr->current_trace_flags->val & flags) == flags; +} /* * DURATION column is being also used to display IRQ signs, @@ -94,46 +124,115 @@ static void print_graph_duration(struct trace_array *tr, unsigned long long duration, struct trace_seq *s, u32 flags); +static int __graph_entry(struct trace_array *tr, struct ftrace_graph_ent *trace, + unsigned int trace_ctx, struct ftrace_regs *fregs) +{ + struct ring_buffer_event *event; + struct trace_buffer *buffer = tr->array_buffer.buffer; + struct ftrace_graph_ent_entry *entry; + int size; + + /* If fregs is defined, add FTRACE_REGS_MAX_ARGS long size words */ + size = sizeof(*entry) + (FTRACE_REGS_MAX_ARGS * !!fregs * sizeof(long)); + + event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT, size, trace_ctx); + if (!event) + return 0; + + entry = ring_buffer_event_data(event); + entry->graph_ent = *trace; + +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + if (fregs) { + for (int i = 0; i < FTRACE_REGS_MAX_ARGS; i++) + entry->args[i] = ftrace_regs_get_argument(fregs, i); + } +#endif + + trace_buffer_unlock_commit_nostack(buffer, event); + + return 1; +} + int __trace_graph_entry(struct trace_array *tr, struct ftrace_graph_ent *trace, unsigned int trace_ctx) { - struct trace_event_call *call = &event_funcgraph_entry; + return __graph_entry(tr, trace, trace_ctx, NULL); +} + +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR +int __trace_graph_retaddr_entry(struct trace_array *tr, + struct ftrace_graph_ent *trace, + unsigned int trace_ctx, + unsigned long retaddr, + struct ftrace_regs *fregs) +{ struct ring_buffer_event *event; struct trace_buffer *buffer = tr->array_buffer.buffer; - struct ftrace_graph_ent_entry *entry; + struct fgraph_retaddr_ent_entry *entry; + int size; - event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT, - sizeof(*entry), trace_ctx); + /* If fregs is defined, add FTRACE_REGS_MAX_ARGS long size words */ + size = sizeof(*entry) + (FTRACE_REGS_MAX_ARGS * !!fregs * sizeof(long)); + + event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RETADDR_ENT, + size, trace_ctx); if (!event) return 0; entry = ring_buffer_event_data(event); - entry->graph_ent = *trace; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + entry->graph_rent.ent = *trace; + entry->graph_rent.retaddr = retaddr; + +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + if (fregs) { + for (int i = 0; i < FTRACE_REGS_MAX_ARGS; i++) + entry->args[i] = ftrace_regs_get_argument(fregs, i); + } +#endif + + trace_buffer_unlock_commit_nostack(buffer, event); return 1; } +#else +int __trace_graph_retaddr_entry(struct trace_array *tr, + struct ftrace_graph_ent *trace, + unsigned int trace_ctx, + unsigned long retaddr, + struct ftrace_regs *fregs) +{ + return 1; +} +#endif -static inline int ftrace_graph_ignore_irqs(void) +static inline int ftrace_graph_ignore_irqs(struct trace_array *tr) { if (!ftrace_graph_skip_irqs || trace_recursion_test(TRACE_IRQ_BIT)) return 0; + if (tracer_flags_is_set(tr, TRACE_GRAPH_PRINT_IRQS)) + return 0; + return in_hardirq(); } -int trace_graph_entry(struct ftrace_graph_ent *trace) +struct fgraph_times { + unsigned long long calltime; + unsigned long long sleeptime; /* may be optional! */ +}; + +static int graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { - struct trace_array *tr = graph_array; - struct trace_array_cpu *data; - unsigned long flags; + unsigned long *task_var = fgraph_get_task_var(gops); + struct trace_array *tr = gops->private; + struct fgraph_times *ftimes; unsigned int trace_ctx; - long disabled; - int ret; - int cpu; + int ret = 0; - if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) + if (*task_var & TRACE_GRAPH_NOTRACE) return 0; /* @@ -144,7 +243,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) * returning from the function. */ if (ftrace_graph_notrace_addr(trace->func)) { - trace_recursion_set(TRACE_GRAPH_NOTRACE_BIT); + *task_var |= TRACE_GRAPH_NOTRACE; /* * Need to return 1 to have the return called * that will clear the NOTRACE bit. @@ -152,15 +251,26 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) return 1; } - if (!ftrace_trace_task(tr)) + if (ftrace_graph_ignore_func(gops, trace)) return 0; - if (ftrace_graph_ignore_func(trace)) + if (ftrace_graph_ignore_irqs(tr)) return 0; - if (ftrace_graph_ignore_irqs()) + if (fgraph_no_sleep_time && + !tracer_flags_is_set(tr, TRACE_GRAPH_SLEEP_TIME)) { + ftimes = fgraph_reserve_data(gops->idx, sizeof(*ftimes)); + if (ftimes) + ftimes->sleeptime = current->ftrace_sleeptime; + } else { + /* Only need to record the calltime */ + ftimes = fgraph_reserve_data(gops->idx, sizeof(ftimes->calltime)); + } + if (!ftimes) return 0; + ftimes->calltime = trace_clock_local(); + /* * Stop here if tracing_threshold is set. We only write function return * events to the ring buffer. @@ -168,23 +278,33 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) if (tracing_thresh) return 1; - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = per_cpu_ptr(tr->array_buffer.data, cpu); - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) { - trace_ctx = tracing_gen_ctx_flags(flags); - ret = __trace_graph_entry(tr, trace, trace_ctx); + trace_ctx = tracing_gen_ctx(); + if (IS_ENABLED(CONFIG_FUNCTION_GRAPH_RETADDR) && + tracer_flags_is_set(tr, TRACE_GRAPH_PRINT_RETADDR)) { + unsigned long retaddr = ftrace_graph_top_ret_addr(current); + ret = __trace_graph_retaddr_entry(tr, trace, trace_ctx, + retaddr, fregs); } else { - ret = 0; + ret = __graph_entry(tr, trace, trace_ctx, fregs); } - atomic_dec(&data->disabled); - local_irq_restore(flags); - return ret; } +int trace_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + return graph_entry(trace, gops, NULL); +} + +static int trace_graph_entry_args(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + return graph_entry(trace, gops, fregs); +} + static void __trace_graph_function(struct trace_array *tr, unsigned long ip, unsigned int trace_ctx) @@ -197,12 +317,10 @@ __trace_graph_function(struct trace_array *tr, struct ftrace_graph_ret ret = { .func = ip, .depth = 0, - .calltime = time, - .rettime = time, }; __trace_graph_entry(tr, &ent, trace_ctx); - __trace_graph_return(tr, &ret, trace_ctx); + __trace_graph_return(tr, &ret, trace_ctx, time, time); } void @@ -214,10 +332,10 @@ trace_graph_function(struct trace_array *tr, } void __trace_graph_return(struct trace_array *tr, - struct ftrace_graph_ret *trace, - unsigned int trace_ctx) + struct ftrace_graph_ret *trace, + unsigned int trace_ctx, + u64 calltime, u64 rettime) { - struct trace_event_call *call = &event_funcgraph_exit; struct ring_buffer_event *event; struct trace_buffer *buffer = tr->array_buffer.buffer; struct ftrace_graph_ret_entry *entry; @@ -228,82 +346,145 @@ void __trace_graph_return(struct trace_array *tr, return; entry = ring_buffer_event_data(event); entry->ret = *trace; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + entry->calltime = calltime; + entry->rettime = rettime; + trace_buffer_unlock_commit_nostack(buffer, event); } -void trace_graph_return(struct ftrace_graph_ret *trace) +static void handle_nosleeptime(struct trace_array *tr, + struct ftrace_graph_ret *trace, + struct fgraph_times *ftimes, + int size) { - struct trace_array *tr = graph_array; - struct trace_array_cpu *data; - unsigned long flags; + if (size < sizeof(*ftimes)) + return; + + if (!fgraph_no_sleep_time || tracer_flags_is_set(tr, TRACE_GRAPH_SLEEP_TIME)) + return; + + ftimes->calltime += current->ftrace_sleeptime - ftimes->sleeptime; +} + +void trace_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, struct ftrace_regs *fregs) +{ + unsigned long *task_var = fgraph_get_task_var(gops); + struct trace_array *tr = gops->private; + struct fgraph_times *ftimes; unsigned int trace_ctx; - long disabled; - int cpu; + u64 calltime, rettime; + int size; - ftrace_graph_addr_finish(trace); + rettime = trace_clock_local(); - if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) { - trace_recursion_clear(TRACE_GRAPH_NOTRACE_BIT); + ftrace_graph_addr_finish(gops, trace); + + if (*task_var & TRACE_GRAPH_NOTRACE) { + *task_var &= ~TRACE_GRAPH_NOTRACE; return; } - local_irq_save(flags); - cpu = raw_smp_processor_id(); - data = per_cpu_ptr(tr->array_buffer.data, cpu); - disabled = atomic_inc_return(&data->disabled); - if (likely(disabled == 1)) { - trace_ctx = tracing_gen_ctx_flags(flags); - __trace_graph_return(tr, trace, trace_ctx); - } - atomic_dec(&data->disabled); - local_irq_restore(flags); -} + ftimes = fgraph_retrieve_data(gops->idx, &size); + if (!ftimes) + return; -void set_graph_array(struct trace_array *tr) -{ - graph_array = tr; + handle_nosleeptime(tr, trace, ftimes, size); - /* Make graph_array visible before we start tracing */ + calltime = ftimes->calltime; - smp_mb(); + trace_ctx = tracing_gen_ctx(); + __trace_graph_return(tr, trace, trace_ctx, calltime, rettime); } -static void trace_graph_thresh_return(struct ftrace_graph_ret *trace) +static void trace_graph_thresh_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { - ftrace_graph_addr_finish(trace); + struct fgraph_times *ftimes; + struct trace_array *tr; + int size; + + ftrace_graph_addr_finish(gops, trace); if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) { trace_recursion_clear(TRACE_GRAPH_NOTRACE_BIT); return; } + ftimes = fgraph_retrieve_data(gops->idx, &size); + if (!ftimes) + return; + + tr = gops->private; + handle_nosleeptime(tr, trace, ftimes, size); + if (tracing_thresh && - (trace->rettime - trace->calltime < tracing_thresh)) + (trace_clock_local() - ftimes->calltime < tracing_thresh)) return; else - trace_graph_return(trace); + trace_graph_return(trace, gops, fregs); } -static struct fgraph_ops funcgraph_thresh_ops = { - .entryfunc = &trace_graph_entry, - .retfunc = &trace_graph_thresh_return, -}; - static struct fgraph_ops funcgraph_ops = { .entryfunc = &trace_graph_entry, .retfunc = &trace_graph_return, }; +int allocate_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops) +{ + struct fgraph_ops *gops; + + gops = kzalloc(sizeof(*gops), GFP_KERNEL); + if (!gops) + return -ENOMEM; + + gops->entryfunc = &trace_graph_entry; + gops->retfunc = &trace_graph_return; + + tr->gops = gops; + gops->private = tr; + + fgraph_init_ops(&gops->ops, ops); + + return 0; +} + +void free_fgraph_ops(struct trace_array *tr) +{ + kfree(tr->gops); +} + +__init void init_array_fgraph_ops(struct trace_array *tr, struct ftrace_ops *ops) +{ + tr->gops = &funcgraph_ops; + funcgraph_ops.private = tr; + fgraph_init_ops(&tr->gops->ops, ops); +} + static int graph_trace_init(struct trace_array *tr) { int ret; - set_graph_array(tr); + if (tracer_flags_is_set(tr, TRACE_GRAPH_ARGS)) + tr->gops->entryfunc = trace_graph_entry_args; + else + tr->gops->entryfunc = trace_graph_entry; + if (tracing_thresh) - ret = register_ftrace_graph(&funcgraph_thresh_ops); + tr->gops->retfunc = trace_graph_thresh_return; else - ret = register_ftrace_graph(&funcgraph_ops); + tr->gops->retfunc = trace_graph_return; + + if (!tracer_flags_is_set(tr, TRACE_GRAPH_PRINT_IRQS)) + ftrace_graph_skip_irqs++; + + if (!tracer_flags_is_set(tr, TRACE_GRAPH_SLEEP_TIME)) + fgraph_no_sleep_time++; + + /* Make gops functions visible before we start tracing */ + smp_mb(); + + ret = register_ftrace_graph(tr->gops); if (ret) return ret; tracing_start_cmdline_record(); @@ -311,13 +492,44 @@ static int graph_trace_init(struct trace_array *tr) return 0; } +static struct tracer graph_trace; + +static int ftrace_graph_trace_args(struct trace_array *tr, int set) +{ + trace_func_graph_ent_t entry; + + if (set) + entry = trace_graph_entry_args; + else + entry = trace_graph_entry; + + /* See if there's any changes */ + if (tr->gops->entryfunc == entry) + return 0; + + unregister_ftrace_graph(tr->gops); + + tr->gops->entryfunc = entry; + + /* Make gops functions visible before we start tracing */ + smp_mb(); + return register_ftrace_graph(tr->gops); +} + static void graph_trace_reset(struct trace_array *tr) { + if (!tracer_flags_is_set(tr, TRACE_GRAPH_PRINT_IRQS)) + ftrace_graph_skip_irqs--; + if (WARN_ON_ONCE(ftrace_graph_skip_irqs < 0)) + ftrace_graph_skip_irqs = 0; + + if (!tracer_flags_is_set(tr, TRACE_GRAPH_SLEEP_TIME)) + fgraph_no_sleep_time--; + if (WARN_ON_ONCE(fgraph_no_sleep_time < 0)) + fgraph_no_sleep_time = 0; + tracing_stop_cmdline_record(); - if (tracing_thresh) - unregister_ftrace_graph(&funcgraph_thresh_ops); - else - unregister_ftrace_graph(&funcgraph_ops); + unregister_ftrace_graph(tr->gops); } static int graph_trace_update_thresh(struct trace_array *tr) @@ -344,7 +556,7 @@ static void print_graph_proc(struct trace_seq *s, pid_t pid) { char comm[TASK_COMM_LEN]; /* sign + log10(MAX_INT) + '\0' */ - char pid_str[11]; + char pid_str[12]; int spaces = 0; int len; int i; @@ -428,7 +640,7 @@ get_return_for_leaf(struct trace_iterator *iter, * then we just reuse the data from before. */ if (data && data->failed) { - curr = &data->ent; + curr = &data->ent.ent; next = &data->ret; } else { @@ -458,7 +670,9 @@ get_return_for_leaf(struct trace_iterator *iter, * Save current and next entries for later reference * if the output fails. */ - data->ent = *curr; + int size = min_t(int, sizeof(data->rent), iter->ent_size); + + memcpy(&data->rent, curr, size); /* * If the next event is not a return type, then * we only care about what type it is. Otherwise we can @@ -515,11 +729,13 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr, struct trace_seq *s = &iter->seq; struct trace_entry *ent = iter->ent; + addr += iter->tr->text_delta; + if (addr < (unsigned long)__irqentry_text_start || addr >= (unsigned long)__irqentry_text_end) return; - if (tr->trace_flags & TRACE_ITER_CONTEXT_INFO) { + if (tr->trace_flags & TRACE_ITER(CONTEXT_INFO)) { /* Absolute time */ if (flags & TRACE_GRAPH_PRINT_ABS_TIME) print_graph_abs_time(iter->ts, s); @@ -539,7 +755,7 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr, } /* Latency format */ - if (tr->trace_flags & TRACE_ITER_LATENCY_FMT) + if (tr->trace_flags & TRACE_ITER(LATENCY_FMT)) print_graph_lat_fmt(s, ent); } @@ -593,7 +809,7 @@ print_graph_duration(struct trace_array *tr, unsigned long long duration, struct trace_seq *s, u32 flags) { if (!(flags & TRACE_GRAPH_PRINT_DURATION) || - !(tr->trace_flags & TRACE_ITER_CONTEXT_INFO)) + !(tr->trace_flags & TRACE_ITER(CONTEXT_INFO))) return; /* No real adata, just filling the column with spaces */ @@ -619,6 +835,106 @@ print_graph_duration(struct trace_array *tr, unsigned long long duration, trace_seq_puts(s, "| "); } +#ifdef CONFIG_FUNCTION_GRAPH_RETVAL +#define __TRACE_GRAPH_PRINT_RETVAL TRACE_GRAPH_PRINT_RETVAL +#else +#define __TRACE_GRAPH_PRINT_RETVAL 0 +#endif + +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR +#define __TRACE_GRAPH_PRINT_RETADDR TRACE_GRAPH_PRINT_RETADDR +static void print_graph_retaddr(struct trace_seq *s, struct fgraph_retaddr_ent_entry *entry, + u32 trace_flags, bool comment) +{ + if (comment) + trace_seq_puts(s, " /*"); + + trace_seq_puts(s, " <-"); + seq_print_ip_sym_offset(s, entry->graph_rent.retaddr, trace_flags); + + if (comment) + trace_seq_puts(s, " */"); +} +#else +#define __TRACE_GRAPH_PRINT_RETADDR 0 +#define print_graph_retaddr(_seq, _entry, _tflags, _comment) do { } while (0) +#endif + +#if defined(CONFIG_FUNCTION_GRAPH_RETVAL) || defined(CONFIG_FUNCTION_GRAPH_RETADDR) + +static void print_graph_retval(struct trace_seq *s, struct ftrace_graph_ent_entry *entry, + struct ftrace_graph_ret *graph_ret, void *func, + u32 opt_flags, u32 trace_flags, int args_size) +{ + unsigned long err_code = 0; + unsigned long retval = 0; + bool print_retaddr = false; + bool print_retval = false; + bool hex_format = !!(opt_flags & TRACE_GRAPH_PRINT_RETVAL_HEX); + +#ifdef CONFIG_FUNCTION_GRAPH_RETVAL + retval = graph_ret->retval; + print_retval = !!(opt_flags & TRACE_GRAPH_PRINT_RETVAL); +#endif + +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR + print_retaddr = !!(opt_flags & TRACE_GRAPH_PRINT_RETADDR); +#endif + + if (print_retval && retval && !hex_format) { + /* Check if the return value matches the negative format */ + if (IS_ENABLED(CONFIG_64BIT) && (retval & BIT(31)) && + (((u64)retval) >> 32) == 0) { + err_code = sign_extend64(retval, 31); + } else { + err_code = retval; + } + + if (!IS_ERR_VALUE(err_code)) + err_code = 0; + } + + if (entry) { + if (entry->ent.type != TRACE_GRAPH_RETADDR_ENT) + print_retaddr = false; + + trace_seq_printf(s, "%ps", func); + + if (args_size >= FTRACE_REGS_MAX_ARGS * sizeof(long)) { + print_function_args(s, entry->args, (unsigned long)func); + trace_seq_putc(s, ';'); + } else + trace_seq_puts(s, "();"); + + if (print_retval || print_retaddr) + trace_seq_puts(s, " /*"); + } else { + print_retaddr = false; + trace_seq_printf(s, "} /* %ps", func); + } + + if (print_retaddr) + print_graph_retaddr(s, (struct fgraph_retaddr_ent_entry *)entry, + trace_flags, false); + + if (print_retval) { + if (hex_format || (err_code == 0)) + trace_seq_printf(s, " ret=0x%lx", retval); + else + trace_seq_printf(s, " ret=%ld", err_code); + } + + if (!entry || print_retval || print_retaddr) + trace_seq_puts(s, " */"); +} + +#else + +#define print_graph_retval(_seq, _ent, _ret, _func, _opt_flags, _trace_flags, args_size) \ + do {} while (0) + +#endif + /* Case of a leaf function on its call entry */ static enum print_line_t print_graph_entry_leaf(struct trace_iterator *iter, @@ -631,12 +947,16 @@ print_graph_entry_leaf(struct trace_iterator *iter, struct ftrace_graph_ret *graph_ret; struct ftrace_graph_ent *call; unsigned long long duration; + unsigned long ret_func; + int args_size; int cpu = iter->cpu; int i; + args_size = iter->ent_size - offsetof(struct ftrace_graph_ent_entry, args); + graph_ret = &ret_entry->ret; call = &entry->graph_ent; - duration = graph_ret->rettime - graph_ret->calltime; + duration = ret_entry->rettime - ret_entry->calltime; if (data) { struct fgraph_cpu_data *cpu_data; @@ -663,7 +983,25 @@ print_graph_entry_leaf(struct trace_iterator *iter, for (i = 0; i < call->depth * TRACE_GRAPH_INDENT; i++) trace_seq_putc(s, ' '); - trace_seq_printf(s, "%ps();\n", (void *)call->func); + ret_func = graph_ret->func + iter->tr->text_delta; + + /* + * Write out the function return value or return address + */ + if (flags & (__TRACE_GRAPH_PRINT_RETVAL | __TRACE_GRAPH_PRINT_RETADDR)) { + print_graph_retval(s, entry, graph_ret, + (void *)graph_ret->func + iter->tr->text_delta, + flags, tr->trace_flags, args_size); + } else { + trace_seq_printf(s, "%ps", (void *)ret_func); + + if (args_size >= FTRACE_REGS_MAX_ARGS * sizeof(long)) { + print_function_args(s, FGRAPH_ENTRY_ARGS(entry), ret_func); + trace_seq_putc(s, ';'); + } else + trace_seq_puts(s, "();"); + } + trace_seq_putc(s, '\n'); print_graph_irq(iter, graph_ret->func, TRACE_GRAPH_RET, cpu, iter->ent->pid, flags); @@ -679,6 +1017,8 @@ print_graph_entry_nested(struct trace_iterator *iter, struct ftrace_graph_ent *call = &entry->graph_ent; struct fgraph_data *data = iter->private; struct trace_array *tr = iter->tr; + unsigned long func; + int args_size; int i; if (data) { @@ -701,7 +1041,24 @@ print_graph_entry_nested(struct trace_iterator *iter, for (i = 0; i < call->depth * TRACE_GRAPH_INDENT; i++) trace_seq_putc(s, ' '); - trace_seq_printf(s, "%ps() {\n", (void *)call->func); + func = call->func + iter->tr->text_delta; + + trace_seq_printf(s, "%ps", (void *)func); + + args_size = iter->ent_size - offsetof(struct ftrace_graph_ent_entry, args); + + if (args_size >= FTRACE_REGS_MAX_ARGS * sizeof(long)) + print_function_args(s, FGRAPH_ENTRY_ARGS(entry), func); + else + trace_seq_puts(s, "()"); + + trace_seq_puts(s, " {"); + + if (flags & __TRACE_GRAPH_PRINT_RETADDR && + entry->ent.type == TRACE_GRAPH_RETADDR_ENT) + print_graph_retaddr(s, (struct fgraph_retaddr_ent_entry *)entry, + tr->trace_flags, true); + trace_seq_putc(s, '\n'); if (trace_seq_has_overflowed(s)) return TRACE_TYPE_PARTIAL_LINE; @@ -729,7 +1086,7 @@ print_graph_prologue(struct trace_iterator *iter, struct trace_seq *s, /* Interrupt */ print_graph_irq(iter, addr, type, cpu, ent->pid, flags); - if (!(tr->trace_flags & TRACE_ITER_CONTEXT_INFO)) + if (!(tr->trace_flags & TRACE_ITER(CONTEXT_INFO))) return; /* Absolute time */ @@ -751,7 +1108,7 @@ print_graph_prologue(struct trace_iterator *iter, struct trace_seq *s, } /* Latency format */ - if (tr->trace_flags & TRACE_ITER_LATENCY_FMT) + if (tr->trace_flags & TRACE_ITER(LATENCY_FMT)) print_graph_lat_fmt(s, ent); return; @@ -776,6 +1133,8 @@ check_irq_entry(struct trace_iterator *iter, u32 flags, int *depth_irq; struct fgraph_data *data = iter->private; + addr += iter->tr->text_delta; + /* * If we are either displaying irqs, or we got called as * a graph event and private data does not exist, @@ -864,21 +1223,41 @@ print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, struct trace_iterator *iter, u32 flags) { struct fgraph_data *data = iter->private; - struct ftrace_graph_ent *call = &field->graph_ent; + struct ftrace_graph_ent *call; struct ftrace_graph_ret_entry *leaf_ret; static enum print_line_t ret; int cpu = iter->cpu; + /* + * print_graph_entry() may consume the current event, + * thus @field may become invalid, so we need to save it. + * This function is shared by ftrace_graph_ent_entry and + * fgraph_retaddr_ent_entry, the size of the latter one + * is larger, but it is very small and can be safely saved + * at the stack. + */ + struct ftrace_graph_ent_entry *entry; + struct fgraph_retaddr_ent_entry *rentry; + u8 save_buf[sizeof(*rentry) + FTRACE_REGS_MAX_ARGS * sizeof(long)]; + + /* The ent_size is expected to be as big as the entry */ + if (iter->ent_size > sizeof(save_buf)) + iter->ent_size = sizeof(save_buf); + + entry = (void *)save_buf; + memcpy(entry, field, iter->ent_size); + + call = &entry->graph_ent; if (check_irq_entry(iter, flags, call->func, call->depth)) return TRACE_TYPE_HANDLED; print_graph_prologue(iter, s, TRACE_GRAPH_ENT, call->func, flags); - leaf_ret = get_return_for_leaf(iter, field); + leaf_ret = get_return_for_leaf(iter, entry); if (leaf_ret) - ret = print_graph_entry_leaf(iter, field, leaf_ret, s, flags); + ret = print_graph_entry_leaf(iter, entry, leaf_ret, s, flags); else - ret = print_graph_entry_nested(iter, field, s, cpu, flags); + ret = print_graph_entry_nested(iter, entry, s, cpu, flags); if (data) { /* @@ -896,18 +1275,24 @@ print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s, } static enum print_line_t -print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, +print_graph_return(struct ftrace_graph_ret_entry *retentry, struct trace_seq *s, struct trace_entry *ent, struct trace_iterator *iter, u32 flags) { - unsigned long long duration = trace->rettime - trace->calltime; + struct ftrace_graph_ret *trace = &retentry->ret; + u64 calltime = retentry->calltime; + u64 rettime = retentry->rettime; + unsigned long long duration = rettime - calltime; struct fgraph_data *data = iter->private; struct trace_array *tr = iter->tr; + unsigned long func; pid_t pid = ent->pid; int cpu = iter->cpu; int func_match = 1; int i; + func = trace->func + iter->tr->text_delta; + if (check_irq_return(iter, flags, trace->depth)) return TRACE_TYPE_HANDLED; @@ -942,16 +1327,26 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, trace_seq_putc(s, ' '); /* - * If the return function does not have a matching entry, - * then the entry was lost. Instead of just printing - * the '}' and letting the user guess what function this - * belongs to, write out the function name. Always do - * that if the funcgraph-tail option is enabled. + * Always write out the function name and its return value if the + * funcgraph-retval option is enabled. */ - if (func_match && !(flags & TRACE_GRAPH_PRINT_TAIL)) - trace_seq_puts(s, "}\n"); - else - trace_seq_printf(s, "} /* %ps */\n", (void *)trace->func); + if (flags & __TRACE_GRAPH_PRINT_RETVAL) { + print_graph_retval(s, NULL, trace, (void *)func, flags, + tr->trace_flags, 0); + } else { + /* + * If the return function does not have a matching entry, + * then the entry was lost. Instead of just printing + * the '}' and letting the user guess what function this + * belongs to, write out the function name. Always do + * that if the funcgraph-tail option is enabled. + */ + if (func_match && !(flags & TRACE_GRAPH_PRINT_TAIL)) + trace_seq_puts(s, "}"); + else + trace_seq_printf(s, "} /* %ps */", (void *)func); + } + trace_seq_putc(s, '\n'); /* Overrun */ if (flags & TRACE_GRAPH_PRINT_OVERRUN) @@ -1053,7 +1448,7 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags) * to print out the missing entry which would never go out. */ if (data && data->failed) { - field = &data->ent; + field = &data->ent.ent; iter->cpu = data->cpu; ret = print_graph_entry(field, s, iter, flags); if (ret == TRACE_TYPE_HANDLED && iter->cpu != cpu) { @@ -1066,21 +1461,28 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags) switch (entry->type) { case TRACE_GRAPH_ENT: { + trace_assign_type(field, entry); + return print_graph_entry(field, s, iter, flags); + } +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR + case TRACE_GRAPH_RETADDR_ENT: { /* - * print_graph_entry() may consume the current event, - * thus @field may become invalid, so we need to save it. - * sizeof(struct ftrace_graph_ent_entry) is very small, - * it can be safely saved at the stack. + * ftrace_graph_ent_entry and fgraph_retaddr_ent_entry have + * similar functions and memory layouts. The only difference + * is that the latter one has an extra retaddr member, so + * they can share most of the logic. */ - struct ftrace_graph_ent_entry saved; - trace_assign_type(field, entry); - saved = *field; - return print_graph_entry(&saved, s, iter, flags); + struct fgraph_retaddr_ent_entry *rfield; + + trace_assign_type(rfield, entry); + return print_graph_entry((struct ftrace_graph_ent_entry *)rfield, + s, iter, flags); } +#endif case TRACE_GRAPH_RET: { struct ftrace_graph_ret_entry *field; trace_assign_type(field, entry); - return print_graph_return(&field->ret, s, entry, iter, flags); + return print_graph_return(field, s, entry, iter, flags); } case TRACE_STACK: case TRACE_FN: @@ -1097,7 +1499,8 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags) static enum print_line_t print_graph_function(struct trace_iterator *iter) { - return print_graph_function_flags(iter, tracer_flags.val); + struct trace_array *tr = iter->tr; + return print_graph_function_flags(iter, tr->current_trace_flags->val); } static enum print_line_t @@ -1133,7 +1536,7 @@ static void print_lat_header(struct seq_file *s, u32 flags) static void __print_graph_headers_flags(struct trace_array *tr, struct seq_file *s, u32 flags) { - int lat = tr->trace_flags & TRACE_ITER_LATENCY_FMT; + int lat = tr->trace_flags & TRACE_ITER(LATENCY_FMT); if (lat) print_lat_header(s, flags); @@ -1173,7 +1576,10 @@ static void __print_graph_headers_flags(struct trace_array *tr, static void print_graph_headers(struct seq_file *s) { - print_graph_headers_flags(s, tracer_flags.val); + struct trace_iterator *iter = s->private; + struct trace_array *tr = iter->tr; + + print_graph_headers_flags(s, tr->current_trace_flags->val); } void print_graph_headers_flags(struct seq_file *s, u32 flags) @@ -1181,10 +1587,10 @@ void print_graph_headers_flags(struct seq_file *s, u32 flags) struct trace_iterator *iter = s->private; struct trace_array *tr = iter->tr; - if (!(tr->trace_flags & TRACE_ITER_CONTEXT_INFO)) + if (!(tr->trace_flags & TRACE_ITER(CONTEXT_INFO))) return; - if (tr->trace_flags & TRACE_ITER_LATENCY_FMT) { + if (tr->trace_flags & TRACE_ITER(LATENCY_FMT)) { /* print nothing if the buffers are empty */ if (trace_empty(iter)) return; @@ -1244,20 +1650,63 @@ void graph_trace_close(struct trace_iterator *iter) if (data) { free_percpu(data->cpu_data); kfree(data); + iter->private = NULL; } } static int func_graph_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set) { - if (bit == TRACE_GRAPH_PRINT_IRQS) - ftrace_graph_skip_irqs = !set; +/* + * The function profiler gets updated even if function graph + * isn't the current tracer. Handle it separately. + */ +#ifdef CONFIG_FUNCTION_PROFILER + if (bit == TRACE_GRAPH_SLEEP_TIME && (tr->flags & TRACE_ARRAY_FL_GLOBAL) && + !!set == fprofile_no_sleep_time) { + if (set) { + fgraph_no_sleep_time--; + if (WARN_ON_ONCE(fgraph_no_sleep_time < 0)) + fgraph_no_sleep_time = 0; + fprofile_no_sleep_time = false; + } else { + fgraph_no_sleep_time++; + fprofile_no_sleep_time = true; + } + } +#endif + + /* Do nothing if the current tracer is not this tracer */ + if (tr->current_trace != &graph_trace) + return 0; + + /* Do nothing if already set. */ + if (!!set == !!(tr->current_trace_flags->val & bit)) + return 0; - if (bit == TRACE_GRAPH_SLEEP_TIME) - ftrace_graph_sleep_time_control(set); + switch (bit) { + case TRACE_GRAPH_SLEEP_TIME: + if (set) { + fgraph_no_sleep_time--; + if (WARN_ON_ONCE(fgraph_no_sleep_time < 0)) + fgraph_no_sleep_time = 0; + } else { + fgraph_no_sleep_time++; + } + break; - if (bit == TRACE_GRAPH_GRAPH_TIME) - ftrace_graph_graph_time_control(set); + case TRACE_GRAPH_PRINT_IRQS: + if (set) + ftrace_graph_skip_irqs--; + else + ftrace_graph_skip_irqs++; + if (WARN_ON_ONCE(ftrace_graph_skip_irqs < 0)) + ftrace_graph_skip_irqs = 0; + break; + + case TRACE_GRAPH_ARGS: + return ftrace_graph_trace_args(tr, set); + } return 0; } @@ -1271,6 +1720,13 @@ static struct trace_event graph_trace_entry_event = { .funcs = &graph_functions, }; +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR +static struct trace_event graph_trace_retaddr_entry_event = { + .type = TRACE_GRAPH_RETADDR_ENT, + .funcs = &graph_functions, +}; +#endif + static struct trace_event graph_trace_ret_event = { .type = TRACE_GRAPH_RET, .funcs = &graph_functions @@ -1287,8 +1743,9 @@ static struct tracer graph_trace __tracer_data = { .reset = graph_trace_reset, .print_line = print_graph_function, .print_header = print_graph_headers, - .flags = &tracer_flags, + .default_flags = &tracer_flags, .set_flag = func_graph_set_flag, + .allow_instances = true, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_function_graph, #endif @@ -1356,6 +1813,13 @@ static __init int init_graph_trace(void) return 1; } +#ifdef CONFIG_FUNCTION_GRAPH_RETADDR + if (!register_trace_event(&graph_trace_retaddr_entry_event)) { + pr_warn("Warning: could not register graph trace retaddr events\n"); + return 1; + } +#endif + if (!register_trace_event(&graph_trace_ret_event)) { pr_warn("Warning: could not register graph trace events\n"); return 1; diff --git a/kernel/trace/trace_hwlat.c b/kernel/trace/trace_hwlat.c index d440ddd5fd8b..2f7b94e98317 100644 --- a/kernel/trace/trace_hwlat.c +++ b/kernel/trace/trace_hwlat.c @@ -130,7 +130,6 @@ static bool hwlat_busy; static void trace_hwlat_sample(struct hwlat_sample *sample) { struct trace_array *tr = hwlat_trace; - struct trace_event_call *call = &event_hwlat; struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct hwlat_entry *entry; @@ -148,8 +147,7 @@ static void trace_hwlat_sample(struct hwlat_sample *sample) entry->nmi_count = sample->nmi_count; entry->count = sample->count; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + trace_buffer_unlock_commit_nostack(buffer, event); } /* Macros to encapsulate the time capturing infrastructure */ @@ -327,19 +325,16 @@ static void move_to_next_cpu(void) cpus_read_lock(); cpumask_and(current_mask, cpu_online_mask, tr->tracing_cpumask); - next_cpu = cpumask_next(raw_smp_processor_id(), current_mask); + next_cpu = cpumask_next_wrap(raw_smp_processor_id(), current_mask); cpus_read_unlock(); - if (next_cpu >= nr_cpu_ids) - next_cpu = cpumask_first(current_mask); - if (next_cpu >= nr_cpu_ids) /* Shouldn't happen! */ goto change_mode; cpumask_clear(current_mask); cpumask_set_cpu(next_cpu, current_mask); - sched_setaffinity(0, current_mask); + set_cpus_allowed_ptr(current, current_mask); return; change_mode: @@ -446,7 +441,7 @@ static int start_single_kthread(struct trace_array *tr) } - sched_setaffinity(kthread->pid, current_mask); + set_cpus_allowed_ptr(kthread, current_mask); kdata->kthread = kthread; wake_up_process(kthread); @@ -492,6 +487,10 @@ static int start_cpu_kthread(unsigned int cpu) { struct task_struct *kthread; + /* Do not start a new hwlatd thread if it is already running */ + if (per_cpu(hwlat_per_cpu_data, cpu).kthread) + return 0; + kthread = kthread_run_on_cpu(kthread_fn, NULL, cpu, "hwlatd/%u"); if (IS_ERR(kthread)) { pr_err(BANNER "could not start sampling thread\n"); @@ -516,6 +515,8 @@ static void hwlat_hotplug_workfn(struct work_struct *dummy) if (!hwlat_busy || hwlat_data.thread_mode != MODE_PER_CPU) goto out_unlock; + if (!cpu_online(cpu)) + goto out_unlock; if (!cpumask_test_cpu(cpu, tr->tracing_cpumask)) goto out_unlock; @@ -584,9 +585,6 @@ static int start_per_cpu_kthreads(struct trace_array *tr) */ cpumask_and(current_mask, cpu_online_mask, tr->tracing_cpumask); - for_each_online_cpu(cpu) - per_cpu(hwlat_per_cpu_data, cpu).kthread = NULL; - for_each_cpu(cpu, current_mask) { retval = start_cpu_kthread(cpu); if (retval) @@ -634,7 +632,7 @@ static int s_mode_show(struct seq_file *s, void *v) else seq_printf(s, "%s", thread_mode_str[mode]); - if (mode != MODE_MAX) + if (mode < MODE_MAX - 1) /* if mode is any but last */ seq_puts(s, " "); return 0; diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 590b3d51afae..17673905907c 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -63,7 +63,7 @@ irq_trace(void) #ifdef CONFIG_FUNCTION_GRAPH_TRACER static int irqsoff_display_graph(struct trace_array *tr, int set); -# define is_graph(tr) ((tr)->trace_flags & TRACE_ITER_DISPLAY_GRAPH) +# define is_graph(tr) ((tr)->trace_flags & TRACE_ITER(DISPLAY_GRAPH)) #else static inline int irqsoff_display_graph(struct trace_array *tr, int set) { @@ -123,12 +123,12 @@ static int func_prolog_dec(struct trace_array *tr, return 0; *data = per_cpu_ptr(tr->array_buffer.data, cpu); - disabled = atomic_inc_return(&(*data)->disabled); + disabled = local_inc_return(&(*data)->disabled); if (likely(disabled == 1)) return 1; - atomic_dec(&(*data)->disabled); + local_dec(&(*data)->disabled); return 0; } @@ -150,9 +150,9 @@ irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip, trace_ctx = tracing_gen_ctx_flags(flags); - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, fregs); - atomic_dec(&data->disabled); + local_dec(&data->disabled); } #endif /* CONFIG_FUNCTION_TRACER */ @@ -175,15 +175,18 @@ static int irqsoff_display_graph(struct trace_array *tr, int set) return start_irqsoff_tracer(irqsoff_trace, set); } -static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) +static int irqsoff_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; unsigned int trace_ctx; - int ret; + u64 *calltime; + int ret = 0; - if (ftrace_graph_ignore_func(trace)) + if (ftrace_graph_ignore_func(gops, trace)) return 0; /* * Do not trace a function if it's filtered by set_graph_notrace. @@ -198,28 +201,41 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) if (!func_prolog_dec(tr, &data, &flags)) return 0; - trace_ctx = tracing_gen_ctx_flags(flags); - ret = __trace_graph_entry(tr, trace, trace_ctx); - atomic_dec(&data->disabled); + calltime = fgraph_reserve_data(gops->idx, sizeof(*calltime)); + if (calltime) { + *calltime = trace_clock_local(); + trace_ctx = tracing_gen_ctx_flags(flags); + ret = __trace_graph_entry(tr, trace, trace_ctx); + } + local_dec(&data->disabled); return ret; } -static void irqsoff_graph_return(struct ftrace_graph_ret *trace) +static void irqsoff_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; unsigned int trace_ctx; + u64 *calltime; + u64 rettime; + int size; - ftrace_graph_addr_finish(trace); + ftrace_graph_addr_finish(gops, trace); if (!func_prolog_dec(tr, &data, &flags)) return; - trace_ctx = tracing_gen_ctx_flags(flags); - __trace_graph_return(tr, trace, trace_ctx); - atomic_dec(&data->disabled); + rettime = trace_clock_local(); + calltime = fgraph_retrieve_data(gops->idx, &size); + if (calltime) { + trace_ctx = tracing_gen_ctx_flags(flags); + __trace_graph_return(tr, trace, trace_ctx, *calltime, rettime); + } + local_dec(&data->disabled); } static struct fgraph_ops fgraph_ops = { @@ -231,7 +247,6 @@ static void irqsoff_trace_open(struct trace_iterator *iter) { if (is_graph(iter->tr)) graph_trace_open(iter); - } static void irqsoff_trace_close(struct trace_iterator *iter) @@ -275,11 +290,17 @@ __trace_function(struct trace_array *tr, if (is_graph(tr)) trace_graph_function(tr, ip, parent_ip, trace_ctx); else - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, NULL); } #else -#define __trace_function trace_function +static inline void +__trace_function(struct trace_array *tr, + unsigned long ip, unsigned long parent_ip, + unsigned int trace_ctx) +{ + return trace_function(tr, ip, parent_ip, trace_ctx, NULL); +} static enum print_line_t irqsoff_print_line(struct trace_iterator *iter) { @@ -373,6 +394,7 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip) int cpu; struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; + long disabled; if (!tracer_enabled || !tracing_is_enabled()) return; @@ -384,20 +406,22 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip) data = per_cpu_ptr(tr->array_buffer.data, cpu); - if (unlikely(!data) || atomic_read(&data->disabled)) + if (unlikely(!data) || local_read(&data->disabled)) return; - atomic_inc(&data->disabled); + disabled = local_inc_return(&data->disabled); - data->critical_sequence = max_sequence; - data->preempt_timestamp = ftrace_now(cpu); - data->critical_start = parent_ip ? : ip; + if (disabled == 1) { + data->critical_sequence = max_sequence; + data->preempt_timestamp = ftrace_now(cpu); + data->critical_start = parent_ip ? : ip; - __trace_function(tr, ip, parent_ip, tracing_gen_ctx()); + __trace_function(tr, ip, parent_ip, tracing_gen_ctx()); - per_cpu(tracing_cpu, cpu) = 1; + per_cpu(tracing_cpu, cpu) = 1; + } - atomic_dec(&data->disabled); + local_dec(&data->disabled); } static nokprobe_inline void @@ -407,6 +431,7 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip) struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned int trace_ctx; + long disabled; cpu = raw_smp_processor_id(); /* Always clear the tracing cpu on stopping the trace */ @@ -421,16 +446,19 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip) data = per_cpu_ptr(tr->array_buffer.data, cpu); if (unlikely(!data) || - !data->critical_start || atomic_read(&data->disabled)) + !data->critical_start || local_read(&data->disabled)) return; - atomic_inc(&data->disabled); + disabled = local_inc_return(&data->disabled); - trace_ctx = tracing_gen_ctx(); - __trace_function(tr, ip, parent_ip, trace_ctx); - check_critical_timing(tr, data, parent_ip ? : ip, cpu); - data->critical_start = 0; - atomic_dec(&data->disabled); + if (disabled == 1) { + trace_ctx = tracing_gen_ctx(); + __trace_function(tr, ip, parent_ip, trace_ctx); + check_critical_timing(tr, data, parent_ip ? : ip, cpu); + data->critical_start = 0; + } + + local_dec(&data->disabled); } /* start and stop critical timings used to for stoppage (in idle) */ @@ -457,8 +485,8 @@ static int register_irqsoff_function(struct trace_array *tr, int graph, int set) { int ret; - /* 'set' is set if TRACE_ITER_FUNCTION is about to be set */ - if (function_enabled || (!set && !(tr->trace_flags & TRACE_ITER_FUNCTION))) + /* 'set' is set if TRACE_ITER(FUNCTION) is about to be set */ + if (function_enabled || (!set && !(tr->trace_flags & TRACE_ITER(FUNCTION)))) return 0; if (graph) @@ -487,7 +515,7 @@ static void unregister_irqsoff_function(struct trace_array *tr, int graph) static int irqsoff_function_set(struct trace_array *tr, u32 mask, int set) { - if (!(mask & TRACE_ITER_FUNCTION)) + if (!(mask & TRACE_ITER(FUNCTION))) return 0; if (set) @@ -508,7 +536,7 @@ static inline int irqsoff_function_set(struct trace_array *tr, u32 mask, int set } #endif /* CONFIG_FUNCTION_TRACER */ -static int irqsoff_flag_changed(struct trace_array *tr, u32 mask, int set) +static int irqsoff_flag_changed(struct trace_array *tr, u64 mask, int set) { struct tracer *tracer = tr->current_trace; @@ -516,7 +544,7 @@ static int irqsoff_flag_changed(struct trace_array *tr, u32 mask, int set) return 0; #ifdef CONFIG_FUNCTION_GRAPH_TRACER - if (mask & TRACE_ITER_DISPLAY_GRAPH) + if (mask & TRACE_ITER(DISPLAY_GRAPH)) return irqsoff_display_graph(tr, set); #endif @@ -554,10 +582,10 @@ static int __irqsoff_tracer_init(struct trace_array *tr) save_flags = tr->trace_flags; /* non overwrite screws up the latency tracers */ - set_tracer_flag(tr, TRACE_ITER_OVERWRITE, 1); - set_tracer_flag(tr, TRACE_ITER_LATENCY_FMT, 1); + set_tracer_flag(tr, TRACE_ITER(OVERWRITE), 1); + set_tracer_flag(tr, TRACE_ITER(LATENCY_FMT), 1); /* without pause, we will produce garbage if another latency occurs */ - set_tracer_flag(tr, TRACE_ITER_PAUSE_ON_TRACE, 1); + set_tracer_flag(tr, TRACE_ITER(PAUSE_ON_TRACE), 1); tr->max_latency = 0; irqsoff_trace = tr; @@ -577,15 +605,15 @@ static int __irqsoff_tracer_init(struct trace_array *tr) static void __irqsoff_tracer_reset(struct trace_array *tr) { - int lat_flag = save_flags & TRACE_ITER_LATENCY_FMT; - int overwrite_flag = save_flags & TRACE_ITER_OVERWRITE; - int pause_flag = save_flags & TRACE_ITER_PAUSE_ON_TRACE; + int lat_flag = save_flags & TRACE_ITER(LATENCY_FMT); + int overwrite_flag = save_flags & TRACE_ITER(OVERWRITE); + int pause_flag = save_flags & TRACE_ITER(PAUSE_ON_TRACE); stop_irqsoff_tracer(tr, is_graph(tr)); - set_tracer_flag(tr, TRACE_ITER_LATENCY_FMT, lat_flag); - set_tracer_flag(tr, TRACE_ITER_OVERWRITE, overwrite_flag); - set_tracer_flag(tr, TRACE_ITER_PAUSE_ON_TRACE, pause_flag); + set_tracer_flag(tr, TRACE_ITER(LATENCY_FMT), lat_flag); + set_tracer_flag(tr, TRACE_ITER(OVERWRITE), overwrite_flag); + set_tracer_flag(tr, TRACE_ITER(PAUSE_ON_TRACE), pause_flag); ftrace_reset_array_ops(tr); irqsoff_busy = false; diff --git a/kernel/trace/trace_kdb.c b/kernel/trace/trace_kdb.c index 59857a1ee44c..b30795f34079 100644 --- a/kernel/trace/trace_kdb.c +++ b/kernel/trace/trace_kdb.c @@ -31,7 +31,7 @@ static void ftrace_dump_buf(int skip_entries, long cpu_file) old_userobj = tr->trace_flags; /* don't look at user memory in panic mode */ - tr->trace_flags &= ~TRACE_ITER_SYM_USEROBJ; + tr->trace_flags &= ~TRACE_ITER(SYM_USEROBJ); kdb_printf("Dumping ftrace buffer:\n"); if (skip_entries) @@ -43,17 +43,15 @@ static void ftrace_dump_buf(int skip_entries, long cpu_file) if (cpu_file == RING_BUFFER_ALL_CPUS) { for_each_tracing_cpu(cpu) { iter.buffer_iter[cpu] = - ring_buffer_read_prepare(iter.array_buffer->buffer, - cpu, GFP_ATOMIC); - ring_buffer_read_start(iter.buffer_iter[cpu]); + ring_buffer_read_start(iter.array_buffer->buffer, + cpu, GFP_ATOMIC); tracing_iter_reset(&iter, cpu); } } else { iter.cpu_file = cpu_file; iter.buffer_iter[cpu_file] = - ring_buffer_read_prepare(iter.array_buffer->buffer, + ring_buffer_read_start(iter.array_buffer->buffer, cpu_file, GFP_ATOMIC); - ring_buffer_read_start(iter.buffer_iter[cpu_file]); tracing_iter_reset(&iter, cpu_file); } @@ -96,22 +94,18 @@ static int kdb_ftdump(int argc, const char **argv) { int skip_entries = 0; long cpu_file; - char *cp; + int err; int cnt; - int cpu; if (argc > 2) return KDB_ARGCOUNT; - if (argc) { - skip_entries = simple_strtol(argv[1], &cp, 0); - if (*cp) - skip_entries = 0; - } + if (argc && kstrtoint(argv[1], 0, &skip_entries)) + return KDB_BADINT; if (argc == 2) { - cpu_file = simple_strtol(argv[2], &cp, 0); - if (*cp || cpu_file >= NR_CPUS || cpu_file < 0 || + err = kstrtol(argv[2], 0, &cpu_file); + if (err || cpu_file >= NR_CPUS || cpu_file < 0 || !cpu_online(cpu_file)) return KDB_BADINT; } else { @@ -123,9 +117,7 @@ static int kdb_ftdump(int argc, const char **argv) trace_init_global_iter(&iter); iter.buffer_iter = buffer_iter; - for_each_tracing_cpu(cpu) { - atomic_inc(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); - } + tracer_tracing_disable(iter.tr); /* A negative skip_entries means skip all but the last entries */ if (skip_entries < 0) { @@ -138,9 +130,7 @@ static int kdb_ftdump(int argc, const char **argv) ftrace_dump_buf(skip_entries, cpu_file); - for_each_tracing_cpu(cpu) { - atomic_dec(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); - } + tracer_tracing_enable(iter.tr); kdb_trap_printk--; diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index ee77c8203bd5..9953506370a5 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -8,19 +8,20 @@ #define pr_fmt(fmt) "trace_kprobe: " fmt #include <linux/bpf-cgroup.h> -#include <linux/security.h> +#include <linux/cleanup.h> +#include <linux/error-injection.h> #include <linux/module.h> -#include <linux/uaccess.h> #include <linux/rculist.h> -#include <linux/error-injection.h> +#include <linux/security.h> +#include <linux/uaccess.h> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ #include "trace_dynevent.h" #include "trace_kprobe_selftest.h" #include "trace_probe.h" -#include "trace_probe_tmpl.h" #include "trace_probe_kernel.h" +#include "trace_probe_tmpl.h" #define KPROBE_EVENT_SYSTEM "kprobes" #define KRETPROBE_MAXACTIVE_MAX 4096 @@ -30,7 +31,7 @@ static char kprobe_boot_events_buf[COMMAND_LINE_SIZE] __initdata; static int __init set_kprobe_boot_events(char *str) { - strlcpy(kprobe_boot_events_buf, str, COMMAND_LINE_SIZE); + strscpy(kprobe_boot_events_buf, str, COMMAND_LINE_SIZE); disable_tracing_selftest("running kprobe events"); return 1; @@ -111,6 +112,7 @@ static nokprobe_inline bool trace_kprobe_within_module(struct trace_kprobe *tk, return strncmp(module_name(mod), name, len) == 0 && name[len] == ':'; } +#ifdef CONFIG_MODULES static nokprobe_inline bool trace_kprobe_module_exist(struct trace_kprobe *tk) { char *p; @@ -122,13 +124,18 @@ static nokprobe_inline bool trace_kprobe_module_exist(struct trace_kprobe *tk) if (!p) return true; *p = '\0'; - rcu_read_lock_sched(); - ret = !!find_module(tk->symbol); - rcu_read_unlock_sched(); + scoped_guard(rcu) + ret = !!find_module(tk->symbol); *p = ':'; return ret; } +#else +static inline bool trace_kprobe_module_exist(struct trace_kprobe *tk) +{ + return false; +} +#endif static bool trace_kprobe_is_busy(struct dyn_event *ev) { @@ -250,6 +257,9 @@ static void free_trace_kprobe(struct trace_kprobe *tk) } } +DEFINE_FREE(free_trace_kprobe, struct trace_kprobe *, + if (!IS_ERR_OR_NULL(_T)) free_trace_kprobe(_T)) + /* * Allocate new trace_probe and initialize it (including kprobes). */ @@ -261,7 +271,7 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group, int maxactive, int nargs, bool is_return) { - struct trace_kprobe *tk; + struct trace_kprobe *tk __free(free_trace_kprobe) = NULL; int ret = -ENOMEM; tk = kzalloc(struct_size(tk, tp.args, nargs), GFP_KERNEL); @@ -270,12 +280,12 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group, tk->nhit = alloc_percpu(unsigned long); if (!tk->nhit) - goto error; + return ERR_PTR(ret); if (symbol) { tk->symbol = kstrdup(symbol, GFP_KERNEL); if (!tk->symbol) - goto error; + return ERR_PTR(ret); tk->rp.kp.symbol_name = tk->symbol; tk->rp.kp.offset = offs; } else @@ -290,15 +300,12 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group, INIT_HLIST_NODE(&tk->rp.kp.hlist); INIT_LIST_HEAD(&tk->rp.kp.list); - ret = trace_probe_init(&tk->tp, event, group, false); + ret = trace_probe_init(&tk->tp, event, group, false, nargs); if (ret < 0) - goto error; + return ERR_PTR(ret); dyn_event_init(&tk->devent, &trace_kprobe_ops); - return tk; -error: - free_trace_kprobe(tk); - return ERR_PTR(ret); + return_ptr(tk); } static struct trace_kprobe *find_trace_kprobe(const char *event, @@ -487,8 +494,8 @@ static int __register_trace_kprobe(struct trace_kprobe *tk) return -EINVAL; if (within_notrace_func(tk)) { - pr_warn("Could not probe notrace function %s\n", - trace_kprobe_symbol(tk)); + pr_warn("Could not probe notrace function %ps\n", + (void *)trace_kprobe_address(tk)); return -EINVAL; } @@ -627,7 +634,7 @@ static int register_trace_kprobe(struct trace_kprobe *tk) struct trace_kprobe *old_tk; int ret; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); old_tk = find_trace_kprobe(trace_probe_name(&tk->tp), trace_probe_group_name(&tk->tp)); @@ -635,11 +642,9 @@ static int register_trace_kprobe(struct trace_kprobe *tk) if (trace_kprobe_is_return(tk) != trace_kprobe_is_return(old_tk)) { trace_probe_log_set_index(0); trace_probe_log_err(0, DIFF_PROBE_TYPE); - ret = -EEXIST; - } else { - ret = append_trace_kprobe(tk, old_tk); + return -EEXIST; } - goto end; + return append_trace_kprobe(tk, old_tk); } /* Register new event */ @@ -650,7 +655,7 @@ static int register_trace_kprobe(struct trace_kprobe *tk) trace_probe_log_err(0, EVENT_EXIST); } else pr_warn("Failed to register probe event(%d)\n", ret); - goto end; + return ret; } /* Register k*probe */ @@ -665,8 +670,22 @@ static int register_trace_kprobe(struct trace_kprobe *tk) else dyn_event_add(&tk->devent, trace_probe_event_call(&tk->tp)); -end: - mutex_unlock(&event_mutex); + return ret; +} + +#ifdef CONFIG_MODULES +static int validate_module_probe_symbol(const char *modname, const char *symbol); + +static int register_module_trace_kprobe(struct module *mod, struct trace_kprobe *tk) +{ + const char *p; + int ret = 0; + + p = strchr(trace_kprobe_symbol(tk), ':'); + if (p) + ret = validate_module_probe_symbol(module_name(mod), p + 1); + if (!ret) + ret = __register_trace_kprobe(tk); return ret; } @@ -683,29 +702,139 @@ static int trace_kprobe_module_callback(struct notifier_block *nb, return NOTIFY_DONE; /* Update probes on coming module */ - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); for_each_trace_kprobe(tk, pos) { if (trace_kprobe_within_module(tk, mod)) { /* Don't need to check busy - this should have gone. */ __unregister_trace_kprobe(tk); - ret = __register_trace_kprobe(tk); + ret = register_module_trace_kprobe(mod, tk); if (ret) pr_warn("Failed to re-register probe %s on %s: %d\n", trace_probe_name(&tk->tp), module_name(mod), ret); } } - mutex_unlock(&event_mutex); return NOTIFY_DONE; } static struct notifier_block trace_kprobe_module_nb = { .notifier_call = trace_kprobe_module_callback, - .priority = 1 /* Invoked after kprobe module callback */ + .priority = 2 /* Invoked after kprobe and jump_label module callback */ +}; +static int trace_kprobe_register_module_notifier(void) +{ + return register_module_notifier(&trace_kprobe_module_nb); +} +#else +static int trace_kprobe_register_module_notifier(void) +{ + return 0; +} +#endif /* CONFIG_MODULES */ + +static int count_symbols(void *data, unsigned long unused) +{ + unsigned int *count = data; + + (*count)++; + + return 0; +} + +struct sym_count_ctx { + unsigned int count; + const char *name; }; -static int __trace_kprobe_create(int argc, const char *argv[]) +static int count_mod_symbols(void *data, const char *name, unsigned long unused) +{ + struct sym_count_ctx *ctx = data; + + if (strcmp(name, ctx->name) == 0) + ctx->count++; + + return 0; +} + +static unsigned int number_of_same_symbols(const char *mod, const char *func_name) +{ + struct sym_count_ctx ctx = { .count = 0, .name = func_name }; + + if (!mod) + kallsyms_on_each_match_symbol(count_symbols, func_name, &ctx.count); + + module_kallsyms_on_each_symbol(mod, count_mod_symbols, &ctx); + + return ctx.count; +} + +static int validate_module_probe_symbol(const char *modname, const char *symbol) +{ + unsigned int count = number_of_same_symbols(modname, symbol); + + if (count > 1) { + /* + * Users should use ADDR to remove the ambiguity of + * using KSYM only. + */ + return -EADDRNOTAVAIL; + } else if (count == 0) { + /* + * We can return ENOENT earlier than when register the + * kprobe. + */ + return -ENOENT; + } + return 0; +} + +#ifdef CONFIG_MODULES +/* Return NULL if the module is not loaded or under unloading. */ +static struct module *try_module_get_by_name(const char *name) +{ + struct module *mod; + + guard(rcu)(); + mod = find_module(name); + if (mod && !try_module_get(mod)) + mod = NULL; + return mod; +} +#else +#define try_module_get_by_name(name) (NULL) +#endif + +static int validate_probe_symbol(char *symbol) +{ + struct module *mod = NULL; + char *modname = NULL, *p; + int ret = 0; + + p = strchr(symbol, ':'); + if (p) { + modname = symbol; + symbol = p + 1; + *p = '\0'; + mod = try_module_get_by_name(modname); + if (!mod) + goto out; + } + + ret = validate_module_probe_symbol(modname, symbol); +out: + if (p) + *p = ':'; + if (mod) + module_put(mod); + return ret; +} + +static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, + struct pt_regs *regs); + +static int trace_kprobe_create_internal(int argc, const char *argv[], + struct traceprobe_parse_context *ctx) { /* * Argument syntax: @@ -731,18 +860,21 @@ static int __trace_kprobe_create(int argc, const char *argv[]) * Type of args: * FETCHARG:TYPE : use TYPE instead of unsigned long. */ - struct trace_kprobe *tk = NULL; - int i, len, ret = 0; - bool is_return = false; - char *symbol = NULL, *tmp = NULL; + struct trace_kprobe *tk __free(free_trace_kprobe) = NULL; const char *event = NULL, *group = KPROBE_EVENT_SYSTEM; + const char **new_argv __free(kfree) = NULL; + int i, len, new_argc = 0, ret = 0; + char *symbol __free(kfree) = NULL; + char *ebuf __free(kfree) = NULL; + char *gbuf __free(kfree) = NULL; + char *abuf __free(kfree) = NULL; + char *dbuf __free(kfree) = NULL; enum probe_print_type ptype; + bool is_return = false; int maxactive = 0; - long offset = 0; void *addr = NULL; - char buf[MAX_EVENT_NAME_LEN]; - char gbuf[MAX_EVENT_NAME_LEN]; - unsigned int flags = TPARG_FL_KERNEL; + char *tmp = NULL; + long offset = 0; switch (argv[0][0]) { case 'r': @@ -756,16 +888,16 @@ static int __trace_kprobe_create(int argc, const char *argv[]) if (argc < 2) return -ECANCELED; - trace_probe_log_init("trace_kprobe", argc, argv); - event = strchr(&argv[0][1], ':'); if (event) event++; if (isdigit(argv[0][1])) { + char *buf __free(kfree) = NULL; + if (!is_return) { - trace_probe_log_err(1, MAXACT_NO_KPROBE); - goto parse_error; + trace_probe_log_err(1, BAD_MAXACT_TYPE); + return -EINVAL; } if (event) len = event - &argv[0][1] - 1; @@ -773,21 +905,23 @@ static int __trace_kprobe_create(int argc, const char *argv[]) len = strlen(&argv[0][1]); if (len > MAX_EVENT_NAME_LEN - 1) { trace_probe_log_err(1, BAD_MAXACT); - goto parse_error; + return -EINVAL; } - memcpy(buf, &argv[0][1], len); + buf = kmemdup(&argv[0][1], len + 1, GFP_KERNEL); + if (!buf) + return -ENOMEM; buf[len] = '\0'; ret = kstrtouint(buf, 0, &maxactive); if (ret || !maxactive) { trace_probe_log_err(1, BAD_MAXACT); - goto parse_error; + return -EINVAL; } /* kretprobes instances are iterated over via a list. The * maximum should stay reasonable. */ if (maxactive > KRETPROBE_MAXACTIVE_MAX) { trace_probe_log_err(1, MAXACT_TOO_BIG); - goto parse_error; + return -EINVAL; } } @@ -796,10 +930,9 @@ static int __trace_kprobe_create(int argc, const char *argv[]) if (kstrtoul(argv[1], 0, (unsigned long *)&addr)) { trace_probe_log_set_index(1); /* Check whether uprobe event specified */ - if (strchr(argv[1], '/') && strchr(argv[1], ':')) { - ret = -ECANCELED; - goto error; - } + if (strchr(argv[1], '/') && strchr(argv[1], ':')) + return -ECANCELED; + /* a symbol specified */ symbol = kstrdup(argv[1], GFP_KERNEL); if (!symbol) @@ -812,7 +945,7 @@ static int __trace_kprobe_create(int argc, const char *argv[]) is_return = true; } else { trace_probe_log_err(tmp - symbol, BAD_ADDR_SUFFIX); - goto parse_error; + return -EINVAL; } } @@ -820,63 +953,108 @@ static int __trace_kprobe_create(int argc, const char *argv[]) ret = traceprobe_split_symbol_offset(symbol, &offset); if (ret || offset < 0 || offset > UINT_MAX) { trace_probe_log_err(0, BAD_PROBE_ADDR); - goto parse_error; + return -EINVAL; + } + ret = validate_probe_symbol(symbol); + if (ret) { + if (ret == -EADDRNOTAVAIL) + trace_probe_log_err(0, NON_UNIQ_SYMBOL); + else + trace_probe_log_err(0, BAD_PROBE_ADDR); + return -EINVAL; } if (is_return) - flags |= TPARG_FL_RETURN; + ctx->flags |= TPARG_FL_RETURN; ret = kprobe_on_func_entry(NULL, symbol, offset); - if (ret == 0) - flags |= TPARG_FL_FENTRY; + if (ret == 0 && !is_return) + ctx->flags |= TPARG_FL_FENTRY; /* Defer the ENOENT case until register kprobe */ if (ret == -EINVAL && is_return) { trace_probe_log_err(0, BAD_RETPROBE); - goto parse_error; + return -EINVAL; } } trace_probe_log_set_index(0); if (event) { + gbuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!gbuf) + return -ENOMEM; ret = traceprobe_parse_event_name(&event, &group, gbuf, event - argv[0]); if (ret) - goto parse_error; + return ret; } if (!event) { /* Make a new event name */ + ebuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!ebuf) + return -ENOMEM; if (symbol) - snprintf(buf, MAX_EVENT_NAME_LEN, "%c_%s_%ld", + snprintf(ebuf, MAX_EVENT_NAME_LEN, "%c_%s_%ld", is_return ? 'r' : 'p', symbol, offset); else - snprintf(buf, MAX_EVENT_NAME_LEN, "%c_0x%p", + snprintf(ebuf, MAX_EVENT_NAME_LEN, "%c_0x%p", is_return ? 'r' : 'p', addr); - sanitize_event_name(buf); - event = buf; + sanitize_event_name(ebuf); + event = ebuf; + } + + abuf = kmalloc(MAX_BTF_ARGS_LEN, GFP_KERNEL); + if (!abuf) + return -ENOMEM; + argc -= 2; argv += 2; + ctx->funcname = symbol; + new_argv = traceprobe_expand_meta_args(argc, argv, &new_argc, + abuf, MAX_BTF_ARGS_LEN, ctx); + if (IS_ERR(new_argv)) { + ret = PTR_ERR(new_argv); + new_argv = NULL; + return ret; + } + if (new_argv) { + argc = new_argc; + argv = new_argv; + } + if (argc > MAX_TRACE_ARGS) { + trace_probe_log_set_index(2); + trace_probe_log_err(0, TOO_MANY_ARGS); + return -E2BIG; } + ret = traceprobe_expand_dentry_args(argc, argv, &dbuf); + if (ret) + return ret; + /* setup a probe */ tk = alloc_trace_kprobe(group, event, addr, symbol, offset, maxactive, - argc - 2, is_return); + argc, is_return); if (IS_ERR(tk)) { ret = PTR_ERR(tk); /* This must return -ENOMEM, else there is a bug */ WARN_ON_ONCE(ret != -ENOMEM); - goto out; /* We know tk is not allocated */ + return ret; /* We know tk is not allocated */ } - argc -= 2; argv += 2; /* parse arguments */ - for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + for (i = 0; i < argc; i++) { trace_probe_log_set_index(i + 2); - ret = traceprobe_parse_probe_arg(&tk->tp, i, argv[i], flags); + ctx->offset = 0; + ret = traceprobe_parse_probe_arg(&tk->tp, i, argv[i], ctx); if (ret) - goto error; /* This can be -ENOMEM */ + return ret; /* This can be -ENOMEM */ + } + /* entry handler for kretprobe */ + if (is_return && tk->tp.entry_arg) { + tk->rp.entry_handler = trace_kprobe_entry_handler; + tk->rp.data_size = traceprobe_get_entry_data_size(&tk->tp); } ptype = is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL; ret = traceprobe_set_print_fmt(&tk->tp, ptype); if (ret < 0) - goto error; + return ret; ret = register_trace_kprobe(tk); if (ret) { @@ -887,24 +1065,38 @@ static int __trace_kprobe_create(int argc, const char *argv[]) trace_probe_log_err(0, BAD_PROBE_ADDR); else if (ret != -ENOMEM && ret != -EEXIST) trace_probe_log_err(0, FAIL_REG_PROBE); - goto error; + return ret; } + /* + * Here, 'tk' has been registered to the list successfully, + * so we don't need to free it. + */ + tk = NULL; + + return 0; +} + +static int trace_kprobe_create_cb(int argc, const char *argv[]) +{ + struct traceprobe_parse_context *ctx __free(traceprobe_parse_context) = NULL; + int ret; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + ctx->flags = TPARG_FL_KERNEL; + + trace_probe_log_init("trace_kprobe", argc, argv); + + ret = trace_kprobe_create_internal(argc, argv, ctx); -out: trace_probe_log_clear(); - kfree(symbol); return ret; - -parse_error: - ret = -EINVAL; -error: - free_trace_kprobe(tk); - goto out; } static int trace_kprobe_create(const char *raw_command) { - return trace_probe_create(raw_command, __trace_kprobe_create); + return trace_probe_create(raw_command, trace_kprobe_create_cb); } static int create_or_delete_trace_kprobe(const char *raw_command) @@ -914,7 +1106,7 @@ static int create_or_delete_trace_kprobe(const char *raw_command) if (raw_command[0] == '-') return dyn_event_release(raw_command, &trace_kprobe_ops); - ret = trace_kprobe_create(raw_command); + ret = dyn_event_create(raw_command, &trace_kprobe_ops); return ret == -ECANCELED ? -EINVAL : ret; } @@ -942,10 +1134,10 @@ EXPORT_SYMBOL_GPL(kprobe_event_cmd_init); /** * __kprobe_event_gen_cmd_start - Generate a kprobe event command from arg list * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @kretprobe: Is this a return probe? * @name: The name of the kprobe event * @loc: The location of the kprobe event - * @kretprobe: Is this a return probe? - * @args: Variable number of arg (pairs), one pair for each field + * @...: Variable number of arg (pairs), one pair for each field * * NOTE: Users normally won't want to call this function directly, but * rather use the kprobe_event_gen_cmd_start() wrapper, which automatically @@ -1018,7 +1210,7 @@ EXPORT_SYMBOL_GPL(__kprobe_event_gen_cmd_start); /** * __kprobe_event_add_fields - Add probe fields to a kprobe command from arg list * @cmd: A pointer to the dynevent_cmd struct representing the new event - * @args: Variable number of arg (pairs), one pair for each field + * @...: Variable number of arg (pairs), one pair for each field * * NOTE: Users normally won't want to call this function directly, but * rather use the kprobe_event_add_fields() wrapper, which @@ -1171,6 +1363,12 @@ static const struct file_operations kprobe_events_ops = { .write = probes_write, }; +static unsigned long trace_kprobe_missed(struct trace_kprobe *tk) +{ + return trace_kprobe_is_return(tk) ? + tk->rp.kp.nmissed + tk->rp.nmissed : tk->rp.kp.nmissed; +} + /* Probes profiling interfaces */ static int probes_profile_seq_show(struct seq_file *m, void *v) { @@ -1182,8 +1380,7 @@ static int probes_profile_seq_show(struct seq_file *m, void *v) return 0; tk = to_trace_kprobe(ev); - nmissed = trace_kprobe_is_return(tk) ? - tk->rp.kp.nmissed + tk->rp.nmissed : tk->rp.kp.nmissed; + nmissed = trace_kprobe_missed(tk); seq_printf(m, " %-44s %15lu %15lu\n", trace_probe_name(&tk->tp), trace_kprobe_nhit(tk), @@ -1218,67 +1415,14 @@ static const struct file_operations kprobe_profile_ops = { .release = seq_release, }; -/* Kprobe specific fetch functions */ - -/* Return the length of string -- including null terminal byte */ -static nokprobe_inline int -fetch_store_strlen_user(unsigned long addr) -{ - return kern_fetch_store_strlen_user(addr); -} - -/* Return the length of string -- including null terminal byte */ -static nokprobe_inline int -fetch_store_strlen(unsigned long addr) -{ - return kern_fetch_store_strlen(addr); -} - -/* - * Fetch a null-terminated string from user. Caller MUST set *(u32 *)buf - * with max length and relative data location. - */ -static nokprobe_inline int -fetch_store_string_user(unsigned long addr, void *dest, void *base) -{ - return kern_fetch_store_string_user(addr, dest, base); -} - -/* - * Fetch a null-terminated string. Caller MUST set *(u32 *)buf with max - * length and relative data location. - */ -static nokprobe_inline int -fetch_store_string(unsigned long addr, void *dest, void *base) -{ - return kern_fetch_store_string(addr, dest, base); -} - -static nokprobe_inline int -probe_mem_read_user(void *dest, void *src, size_t size) -{ - const void __user *uaddr = (__force const void __user *)src; - - return copy_from_user_nofault(dest, uaddr, size); -} - -static nokprobe_inline int -probe_mem_read(void *dest, void *src, size_t size) -{ -#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE - if ((unsigned long)src < TASK_SIZE) - return probe_mem_read_user(dest, src, size); -#endif - return copy_from_kernel_nofault(dest, src, size); -} - /* Note that we don't verify it, since the code does not come from user space */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { struct pt_regs *regs = rec; unsigned long val; + int ret; retry: /* 1st stage: get value from context */ @@ -1295,25 +1439,21 @@ retry: case FETCH_OP_RETVAL: val = regs_return_value(regs); break; - case FETCH_OP_IMM: - val = code->immediate; - break; - case FETCH_OP_COMM: - val = (unsigned long)current->comm; - break; - case FETCH_OP_DATA: - val = (unsigned long)code->data; - break; #ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API case FETCH_OP_ARG: val = regs_get_kernel_argument(regs, code->param); break; + case FETCH_OP_EDATA: + val = *(unsigned long *)((unsigned long)edata + code->offset); + break; #endif case FETCH_NOP_SYMBOL: /* Ignore a place holder */ code++; goto retry; default: - return -EILSEQ; + ret = process_common_fetch_insn(code, &val); + if (ret < 0) + return ret; } code++; @@ -1336,7 +1476,7 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, if (trace_trigger_soft_disabled(trace_file)) return; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, NULL); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + tk->tp.size + dsize); @@ -1345,7 +1485,7 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, fbuffer.regs = regs; entry->ip = (unsigned long)tk->rp.kp.addr; - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, NULL, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } @@ -1361,6 +1501,31 @@ kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs) NOKPROBE_SYMBOL(kprobe_trace_func); /* Kretprobe handler */ + +static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, + struct pt_regs *regs) +{ + struct kretprobe *rp = get_kretprobe(ri); + struct trace_kprobe *tk; + + /* + * There is a small chance that get_kretprobe(ri) returns NULL when + * the kretprobe is unregister on another CPU between kretprobe's + * trampoline_handler and this function. + */ + if (unlikely(!rp)) + return -ENOENT; + + tk = container_of(rp, struct trace_kprobe, rp); + + /* store argument values into ri->data as entry data */ + if (tk->tp.entry_arg) + store_trace_entry_data(ri->data, &tk->tp, regs); + + return 0; +} + + static nokprobe_inline void __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct pt_regs *regs, @@ -1376,7 +1541,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, if (trace_trigger_soft_disabled(trace_file)) return; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, ri->data); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + tk->tp.size + dsize); @@ -1386,7 +1551,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, fbuffer.regs = regs; entry->func = (unsigned long)tk->rp.kp.addr; entry->ret_ip = get_kretprobe_retaddr(ri); - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, ri->data, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } @@ -1419,12 +1584,12 @@ print_kprobe_event(struct trace_iterator *iter, int flags, trace_seq_printf(s, "%s: (", trace_probe_name(tp)); - if (!seq_print_ip_sym(s, field->ip, flags | TRACE_ITER_SYM_OFFSET)) + if (!seq_print_ip_sym_offset(s, field->ip, flags)) goto out; trace_seq_putc(s, ')'); - if (print_probe_args(s, tp->args, tp->nr_args, + if (trace_probe_print_args(s, tp->args, tp->nr_args, (u8 *)&field[1], field) < 0) goto out; @@ -1449,17 +1614,17 @@ print_kretprobe_event(struct trace_iterator *iter, int flags, trace_seq_printf(s, "%s: (", trace_probe_name(tp)); - if (!seq_print_ip_sym(s, field->ret_ip, flags | TRACE_ITER_SYM_OFFSET)) + if (!seq_print_ip_sym_offset(s, field->ret_ip, flags)) goto out; trace_seq_puts(s, " <- "); - if (!seq_print_ip_sym(s, field->func, flags & ~TRACE_ITER_SYM_OFFSET)) + if (!seq_print_ip_sym_no_offset(s, field->func, flags)) goto out; trace_seq_putc(s, ')'); - if (print_probe_args(s, tp->args, tp->nr_args, + if (trace_probe_print_args(s, tp->args, tp->nr_args, (u8 *)&field[1], field) < 0) goto out; @@ -1534,7 +1699,7 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) if (hlist_empty(head)) return 0; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, NULL); __size = sizeof(*entry) + tk->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); @@ -1545,7 +1710,7 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) entry->ip = (unsigned long)tk->rp.kp.addr; memset(&entry[1], 0, dsize); - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, NULL, sizeof(*entry), dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); return 0; @@ -1570,7 +1735,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, if (hlist_empty(head)) return; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, ri->data); __size = sizeof(*entry) + tk->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); @@ -1581,7 +1746,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, entry->func = (unsigned long)tk->rp.kp.addr; entry->ret_ip = get_kretprobe_retaddr(ri); - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, ri->data, sizeof(*entry), dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); } @@ -1589,7 +1754,8 @@ NOKPROBE_SYMBOL(kretprobe_perf_func); int bpf_get_kprobe_info(const struct perf_event *event, u32 *fd_type, const char **symbol, u64 *probe_offset, - u64 *probe_addr, bool perf_type_tracepoint) + u64 *probe_addr, unsigned long *missed, + bool perf_type_tracepoint) { const char *pevent = trace_event_name(event->tp_event); const char *group = event->tp_event->class->system; @@ -1604,15 +1770,12 @@ int bpf_get_kprobe_info(const struct perf_event *event, u32 *fd_type, *fd_type = trace_kprobe_is_return(tk) ? BPF_FD_TYPE_KRETPROBE : BPF_FD_TYPE_KPROBE; - if (tk->symbol) { - *symbol = tk->symbol; - *probe_offset = tk->rp.kp.offset; - *probe_addr = 0; - } else { - *symbol = NULL; - *probe_offset = 0; - *probe_addr = (unsigned long)tk->rp.kp.addr; - } + *probe_offset = tk->rp.kp.offset; + *probe_addr = kallsyms_show_value(current_cred()) ? + (unsigned long)tk->rp.kp.addr : 0; + *symbol = tk->symbol; + if (missed) + *missed = trace_kprobe_missed(tk); return 0; } #endif /* CONFIG_PERF_EVENTS */ @@ -1652,14 +1815,15 @@ static int kprobe_register(struct trace_event_call *event, static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) { struct trace_kprobe *tk = container_of(kp, struct trace_kprobe, rp.kp); + unsigned int flags = trace_probe_load_flag(&tk->tp); int ret = 0; raw_cpu_inc(*tk->nhit); - if (trace_probe_test_flag(&tk->tp, TP_FLAG_TRACE)) + if (flags & TP_FLAG_TRACE) kprobe_trace_func(tk, regs); #ifdef CONFIG_PERF_EVENTS - if (trace_probe_test_flag(&tk->tp, TP_FLAG_PROFILE)) + if (flags & TP_FLAG_PROFILE) ret = kprobe_perf_func(tk, regs); #endif return ret; @@ -1671,6 +1835,7 @@ kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs) { struct kretprobe *rp = get_kretprobe(ri); struct trace_kprobe *tk; + unsigned int flags; /* * There is a small chance that get_kretprobe(ri) returns NULL when @@ -1683,10 +1848,11 @@ kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs) tk = container_of(rp, struct trace_kprobe, rp); raw_cpu_inc(*tk->nhit); - if (trace_probe_test_flag(&tk->tp, TP_FLAG_TRACE)) + flags = trace_probe_load_flag(&tk->tp); + if (flags & TP_FLAG_TRACE) kretprobe_trace_func(tk, ri, regs); #ifdef CONFIG_PERF_EVENTS - if (trace_probe_test_flag(&tk->tp, TP_FLAG_PROFILE)) + if (flags & TP_FLAG_PROFILE) kretprobe_perf_func(tk, ri, regs); #endif return 0; /* We don't tweak kernel, so just return 0 */ @@ -1742,16 +1908,23 @@ static int unregister_kprobe_event(struct trace_kprobe *tk) } #ifdef CONFIG_PERF_EVENTS + /* create a trace_kprobe, but don't add it to global lists */ struct trace_event_call * create_local_trace_kprobe(char *func, void *addr, unsigned long offs, bool is_return) { enum probe_print_type ptype; - struct trace_kprobe *tk; + struct trace_kprobe *tk __free(free_trace_kprobe) = NULL; int ret; char *event; + if (func) { + ret = validate_probe_symbol(func); + if (ret) + return ERR_PTR(ret); + } + /* * local trace_kprobes are not added to dyn_event, so they are never * searched in find_trace_kprobe(). Therefore, there is no concern of @@ -1773,19 +1946,14 @@ create_local_trace_kprobe(char *func, void *addr, unsigned long offs, ptype = trace_kprobe_is_return(tk) ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL; - if (traceprobe_set_print_fmt(&tk->tp, ptype) < 0) { - ret = -ENOMEM; - goto error; - } + if (traceprobe_set_print_fmt(&tk->tp, ptype) < 0) + return ERR_PTR(-ENOMEM); ret = __register_trace_kprobe(tk); if (ret < 0) - goto error; + return ERR_PTR(ret); - return trace_probe_event_call(&tk->tp); -error: - free_trace_kprobe(tk); - return ERR_PTR(ret); + return trace_probe_event_call(&(no_free_ptr(tk)->tp)); } void destroy_local_trace_kprobe(struct trace_event_call *event_call) @@ -1814,13 +1982,12 @@ static __init void enable_boot_kprobe_events(void) struct trace_kprobe *tk; struct dyn_event *pos; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); for_each_trace_kprobe(tk, pos) { list_for_each_entry(file, &tr->events, list) if (file->event_call == trace_probe_event_call(&tk->tp)) trace_event_enable_disable(file, 1, 0); } - mutex_unlock(&event_mutex); } static __init void setup_boot_kprobe_events(void) @@ -1857,7 +2024,7 @@ static __init int init_kprobe_trace_early(void) if (ret) return ret; - if (register_module_notifier(&trace_kprobe_module_nb)) + if (trace_kprobe_register_module_notifier()) return -EINVAL; return 0; @@ -1923,19 +2090,16 @@ static __init int kprobe_trace_self_tests_init(void) pr_info("Testing kprobe tracing: "); ret = create_or_delete_trace_kprobe("p:testprobe kprobe_trace_selftest_target $stack $stack0 +0($stack)"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on probing function entry.\n"); + if (WARN_ONCE(ret, "error on probing function entry.")) { warn++; } else { /* Enable trace point */ tk = find_trace_kprobe("testprobe", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting new probe.\n"); + if (WARN_ONCE(tk == NULL, "error on probing function entry.")) { warn++; } else { file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else enable_trace_kprobe( @@ -1944,19 +2108,16 @@ static __init int kprobe_trace_self_tests_init(void) } ret = create_or_delete_trace_kprobe("r:testprobe2 kprobe_trace_selftest_target $retval"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on probing function return.\n"); + if (WARN_ONCE(ret, "error on probing function return.")) { warn++; } else { /* Enable trace point */ tk = find_trace_kprobe("testprobe2", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting 2nd new probe.\n"); + if (WARN_ONCE(tk == NULL, "error on getting 2nd new probe.")) { warn++; } else { file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else enable_trace_kprobe( @@ -1979,18 +2140,15 @@ static __init int kprobe_trace_self_tests_init(void) /* Disable trace points before removing it */ tk = find_trace_kprobe("testprobe", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting test probe.\n"); + if (WARN_ONCE(tk == NULL, "error on getting test probe.")) { warn++; } else { - if (trace_kprobe_nhit(tk) != 1) { - pr_warn("incorrect number of testprobe hits\n"); + if (WARN_ONCE(trace_kprobe_nhit(tk) != 1, + "incorrect number of testprobe hits.")) warn++; - } file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else disable_trace_kprobe( @@ -1998,18 +2156,15 @@ static __init int kprobe_trace_self_tests_init(void) } tk = find_trace_kprobe("testprobe2", KPROBE_EVENT_SYSTEM); - if (WARN_ON_ONCE(tk == NULL)) { - pr_warn("error on getting 2nd test probe.\n"); + if (WARN_ONCE(tk == NULL, "error on getting 2nd test probe.")) { warn++; } else { - if (trace_kprobe_nhit(tk) != 1) { - pr_warn("incorrect number of testprobe2 hits\n"); + if (WARN_ONCE(trace_kprobe_nhit(tk) != 1, + "incorrect number of testprobe2 hits.")) warn++; - } file = find_trace_probe_file(tk, top_trace_array()); - if (WARN_ON_ONCE(file == NULL)) { - pr_warn("error on getting probe file.\n"); + if (WARN_ONCE(file == NULL, "error on getting probe file.")) { warn++; } else disable_trace_kprobe( @@ -2017,23 +2172,15 @@ static __init int kprobe_trace_self_tests_init(void) } ret = create_or_delete_trace_kprobe("-:testprobe"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on deleting a probe.\n"); + if (WARN_ONCE(ret, "error on deleting a probe.")) warn++; - } ret = create_or_delete_trace_kprobe("-:testprobe2"); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on deleting a probe.\n"); + if (WARN_ONCE(ret, "error on deleting a probe.")) warn++; - } + end: - ret = dyn_events_release_all(&trace_kprobe_ops); - if (WARN_ON_ONCE(ret)) { - pr_warn("error on cleaning up probes.\n"); - warn++; - } /* * Wait for the optimizer work to finish. Otherwise it might fiddle * with probes in already freed __init text. diff --git a/kernel/trace/trace_kprobe_selftest.c b/kernel/trace/trace_kprobe_selftest.c index 16548ee4c8c6..3851cd1e6a62 100644 --- a/kernel/trace/trace_kprobe_selftest.c +++ b/kernel/trace/trace_kprobe_selftest.c @@ -1,4 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 + +#include "trace_kprobe_selftest.h" + /* * Function used during the kprobe self test. This function is in a separate * compile unit so it can be compile with CC_FLAGS_FTRACE to ensure that it diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index 64e77b513697..c706544be60c 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -291,10 +291,8 @@ __init static int init_mmio_trace(void) device_initcall(init_mmio_trace); static void __trace_mmiotrace_rw(struct trace_array *tr, - struct trace_array_cpu *data, struct mmiotrace_rw *rw) { - struct trace_event_call *call = &event_mmiotrace_rw; struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct trace_mmiotrace_rw *entry; @@ -310,22 +308,18 @@ static void __trace_mmiotrace_rw(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->rw = *rw; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } void mmio_trace_rw(struct mmiotrace_rw *rw) { struct trace_array *tr = mmio_trace_array; - struct trace_array_cpu *data = per_cpu_ptr(tr->array_buffer.data, smp_processor_id()); - __trace_mmiotrace_rw(tr, data, rw); + __trace_mmiotrace_rw(tr, rw); } static void __trace_mmiotrace_map(struct trace_array *tr, - struct trace_array_cpu *data, struct mmiotrace_map *map) { - struct trace_event_call *call = &event_mmiotrace_map; struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct trace_mmiotrace_map *entry; @@ -341,19 +335,13 @@ static void __trace_mmiotrace_map(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->map = *map; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } void mmio_trace_mapping(struct mmiotrace_map *map) { struct trace_array *tr = mmio_trace_array; - struct trace_array_cpu *data; - - preempt_disable(); - data = per_cpu_ptr(tr->array_buffer.data, smp_processor_id()); - __trace_mmiotrace_map(tr, data, map); - preempt_enable(); + __trace_mmiotrace_map(tr, map); } int mmio_trace_printk(const char *fmt, va_list args) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index 210e1f168392..827104d00bc0 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -24,6 +24,7 @@ #include <linux/sched/clock.h> #include <uapi/linux/sched/types.h> #include <linux/sched.h> +#include <linux/string.h> #include "trace.h" #ifdef CONFIG_X86_LOCAL_APIC @@ -159,7 +160,7 @@ static void osnoise_unregister_instance(struct trace_array *tr) if (!found) return; - kvfree_rcu(inst); + kvfree_rcu_mightsleep(inst); } /* @@ -181,6 +182,7 @@ struct osn_irq { #define IRQ_CONTEXT 0 #define THREAD_CONTEXT 1 +#define THREAD_URET 2 /* * sofirq runtime info. */ @@ -217,7 +219,7 @@ struct osnoise_variables { /* * Per-cpu runtime information. */ -DEFINE_PER_CPU(struct osnoise_variables, per_cpu_osnoise_var); +static DEFINE_PER_CPU(struct osnoise_variables, per_cpu_osnoise_var); /* * this_cpu_osn_var - Return the per-cpu osnoise_variables on its relative CPU @@ -227,6 +229,11 @@ static inline struct osnoise_variables *this_cpu_osn_var(void) return this_cpu_ptr(&per_cpu_osnoise_var); } +/* + * Protect the interface. + */ +static struct mutex interface_lock; + #ifdef CONFIG_TIMERLAT_TRACER /* * Runtime information for the timer mode. @@ -238,9 +245,10 @@ struct timerlat_variables { u64 abs_period; bool tracing_thread; u64 count; + bool uthread_migrate; }; -DEFINE_PER_CPU(struct timerlat_variables, per_cpu_timerlat_var); +static DEFINE_PER_CPU(struct timerlat_variables, per_cpu_timerlat_var); /* * this_cpu_tmr_var - Return the per-cpu timerlat_variables on its relative CPU @@ -257,14 +265,20 @@ static inline void tlat_var_reset(void) { struct timerlat_variables *tlat_var; int cpu; + + /* Synchronize with the timerlat interfaces */ + mutex_lock(&interface_lock); /* * So far, all the values are initialized as 0, so * zeroing the structure is perfect. */ - for_each_cpu(cpu, cpu_online_mask) { + for_each_online_cpu(cpu) { tlat_var = per_cpu_ptr(&per_cpu_timerlat_var, cpu); + if (tlat_var->kthread) + hrtimer_cancel(&tlat_var->timer); memset(tlat_var, 0, sizeof(*tlat_var)); } + mutex_unlock(&interface_lock); } #else /* CONFIG_TIMERLAT_TRACER */ #define tlat_var_reset() do {} while (0) @@ -282,7 +296,7 @@ static inline void osn_var_reset(void) * So far, all the values are initialized as 0, so * zeroing the structure is perfect. */ - for_each_cpu(cpu, cpu_online_mask) { + for_each_online_cpu(cpu) { osn_var = per_cpu_ptr(&per_cpu_osnoise_var, cpu); memset(osn_var, 0, sizeof(*osn_var)); } @@ -303,38 +317,6 @@ static inline void osn_var_reset_all(void) bool trace_osnoise_callback_enabled; /* - * osnoise sample structure definition. Used to store the statistics of a - * sample run. - */ -struct osnoise_sample { - u64 runtime; /* runtime */ - u64 noise; /* noise */ - u64 max_sample; /* max single noise sample */ - int hw_count; /* # HW (incl. hypervisor) interference */ - int nmi_count; /* # NMIs during this sample */ - int irq_count; /* # IRQs during this sample */ - int softirq_count; /* # softirqs during this sample */ - int thread_count; /* # threads during this sample */ -}; - -#ifdef CONFIG_TIMERLAT_TRACER -/* - * timerlat sample structure definition. Used to store the statistics of - * a sample run. - */ -struct timerlat_sample { - u64 timer_latency; /* timer_latency */ - unsigned int seqnum; /* unique sequence */ - int context; /* timer context */ -}; -#endif - -/* - * Protect the interface. - */ -struct mutex interface_lock; - -/* * Tracer data. */ static struct osnoise_data { @@ -347,7 +329,7 @@ static struct osnoise_data { u64 print_stack; /* print IRQ stack if total > */ int timerlat_tracer; /* timerlat tracer */ #endif - bool tainted; /* infor users and developers about a problem */ + bool tainted; /* info users and developers about a problem */ } osnoise_data = { .sample_period = DEFAULT_SAMPLE_PERIOD, .sample_runtime = DEFAULT_SAMPLE_RUNTIME, @@ -489,9 +471,8 @@ static void print_osnoise_headers(struct seq_file *s) * Record an osnoise_sample into the tracer buffer. */ static void -__trace_osnoise_sample(struct osnoise_sample *sample, struct trace_buffer *buffer) +__record_osnoise_sample(struct osnoise_sample *sample, struct trace_buffer *buffer) { - struct trace_event_call *call = &event_osnoise; struct ring_buffer_event *event; struct osnoise_entry *entry; @@ -509,22 +490,23 @@ __trace_osnoise_sample(struct osnoise_sample *sample, struct trace_buffer *buffe entry->softirq_count = sample->softirq_count; entry->thread_count = sample->thread_count; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + trace_buffer_unlock_commit_nostack(buffer, event); } /* - * Record an osnoise_sample on all osnoise instances. + * Record an osnoise_sample on all osnoise instances and fire trace event. */ -static void trace_osnoise_sample(struct osnoise_sample *sample) +static void record_osnoise_sample(struct osnoise_sample *sample) { struct osnoise_instance *inst; struct trace_buffer *buffer; + trace_osnoise_sample(sample); + rcu_read_lock(); list_for_each_entry_rcu(inst, &osnoise_instances, list) { buffer = inst->tr->array_buffer.buffer; - __trace_osnoise_sample(sample, buffer); + __record_osnoise_sample(sample, buffer); } rcu_read_unlock(); } @@ -568,9 +550,8 @@ static void print_timerlat_headers(struct seq_file *s) #endif /* CONFIG_PREEMPT_RT */ static void -__trace_timerlat_sample(struct timerlat_sample *sample, struct trace_buffer *buffer) +__record_timerlat_sample(struct timerlat_sample *sample, struct trace_buffer *buffer) { - struct trace_event_call *call = &event_osnoise; struct ring_buffer_event *event; struct timerlat_entry *entry; @@ -583,22 +564,23 @@ __trace_timerlat_sample(struct timerlat_sample *sample, struct trace_buffer *buf entry->context = sample->context; entry->timer_latency = sample->timer_latency; - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + trace_buffer_unlock_commit_nostack(buffer, event); } /* * Record an timerlat_sample into the tracer buffer. */ -static void trace_timerlat_sample(struct timerlat_sample *sample) +static void record_timerlat_sample(struct timerlat_sample *sample) { struct osnoise_instance *inst; struct trace_buffer *buffer; + trace_timerlat_sample(sample); + rcu_read_lock(); list_for_each_entry_rcu(inst, &osnoise_instances, list) { buffer = inst->tr->array_buffer.buffer; - __trace_timerlat_sample(sample, buffer); + __record_timerlat_sample(sample, buffer); } rcu_read_unlock(); } @@ -646,7 +628,6 @@ static void timerlat_save_stack(int skip) static void __timerlat_dump_stack(struct trace_buffer *buffer, struct trace_stack *fstack, unsigned int size) { - struct trace_event_call *call = &event_osnoise; struct ring_buffer_event *event; struct stack_entry *entry; @@ -657,11 +638,10 @@ __timerlat_dump_stack(struct trace_buffer *buffer, struct trace_stack *fstack, u entry = ring_buffer_event_data(event); - memcpy(&entry->caller, fstack->calls, size); entry->size = fstack->nr_entries; + memcpy(&entry->caller, fstack->calls, size); - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit_nostack(buffer, event); + trace_buffer_unlock_commit_nostack(buffer, event); } /* @@ -758,7 +738,7 @@ cond_move_thread_delta_start(struct osnoise_variables *osn_var, u64 duration) /* * get_int_safe_duration - Get the duration of a window * - * The irq, softirq and thread varaibles need to have its duration without + * The irq, softirq and thread variables need to have its duration without * the interference from higher priority interrupts. Instead of keeping a * variable to discount the interrupt interference from these variables, the * starting time of these variables are pushed forward with the interrupt's @@ -1181,6 +1161,89 @@ thread_exit(struct osnoise_variables *osn_var, struct task_struct *t) osn_var->thread.arrival_time = 0; } +#ifdef CONFIG_TIMERLAT_TRACER +/* + * osnoise_stop_exception - Stop tracing and the tracer. + */ +static __always_inline void osnoise_stop_exception(char *msg, int cpu) +{ + struct osnoise_instance *inst; + struct trace_array *tr; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + tr = inst->tr; + trace_array_printk_buf(tr->array_buffer.buffer, _THIS_IP_, + "stop tracing hit on cpu %d due to exception: %s\n", + smp_processor_id(), + msg); + + if (test_bit(OSN_PANIC_ON_STOP, &osnoise_options)) + panic("tracer hit on cpu %d due to exception: %s\n", + smp_processor_id(), + msg); + + tracer_tracing_off(tr); + } + rcu_read_unlock(); +} + +/* + * trace_sched_migrate_callback - sched:sched_migrate_task trace event handler + * + * his function is hooked to the sched:sched_migrate_task trace event, and monitors + * timerlat user-space thread migration. + */ +static void trace_sched_migrate_callback(void *data, struct task_struct *p, int dest_cpu) +{ + struct osnoise_variables *osn_var; + long cpu = task_cpu(p); + + osn_var = per_cpu_ptr(&per_cpu_osnoise_var, cpu); + if (osn_var->pid == p->pid && dest_cpu != cpu) { + per_cpu_ptr(&per_cpu_timerlat_var, cpu)->uthread_migrate = 1; + osnoise_taint("timerlat user-thread migrated\n"); + osnoise_stop_exception("timerlat user-thread migrated", cpu); + } +} + +static bool monitor_enabled; + +static int register_migration_monitor(void) +{ + int ret = 0; + + /* + * Timerlat thread migration check is only required when running timerlat in user-space. + * Thus, enable callback only if timerlat is set with no workload. + */ + if (timerlat_enabled() && !test_bit(OSN_WORKLOAD, &osnoise_options)) { + if (WARN_ON_ONCE(monitor_enabled)) + return 0; + + ret = register_trace_sched_migrate_task(trace_sched_migrate_callback, NULL); + if (!ret) + monitor_enabled = true; + } + + return ret; +} + +static void unregister_migration_monitor(void) +{ + if (!monitor_enabled) + return; + + unregister_trace_sched_migrate_task(trace_sched_migrate_callback, NULL); + monitor_enabled = false; +} +#else +static int register_migration_monitor(void) +{ + return 0; +} +static void unregister_migration_monitor(void) {} +#endif /* * trace_sched_switch - sched:sched_switch trace event handler * @@ -1204,7 +1267,7 @@ trace_sched_switch_callback(void *data, bool preempt, } /* - * hook_thread_events - Hook the insturmentation for thread noise + * hook_thread_events - Hook the instrumentation for thread noise * * Hook the osnoise tracer callbacks to handle the noise from other * threads on the necessary kernel events. @@ -1217,11 +1280,19 @@ static int hook_thread_events(void) if (ret) return -EINVAL; + ret = register_migration_monitor(); + if (ret) + goto out_unreg; + return 0; + +out_unreg: + unregister_trace_sched_switch(trace_sched_switch_callback, NULL); + return -EINVAL; } /* - * unhook_thread_events - *nhook the insturmentation for thread noise + * unhook_thread_events - unhook the instrumentation for thread noise * * Unook the osnoise tracer callbacks to handle the noise from other * threads on the necessary kernel events. @@ -1229,6 +1300,7 @@ static int hook_thread_events(void) static void unhook_thread_events(void) { unregister_trace_sched_switch(trace_sched_switch_callback, NULL); + unregister_migration_monitor(); } /* @@ -1286,6 +1358,22 @@ static __always_inline void osnoise_stop_tracing(void) } /* + * osnoise_has_tracing_on - Check if there is at least one instance on + */ +static __always_inline int osnoise_has_tracing_on(void) +{ + struct osnoise_instance *inst; + int trace_is_on = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) + trace_is_on += tracer_tracing_is_on(inst->tr); + rcu_read_unlock(); + + return trace_is_on; +} + +/* * notify_new_max_latency - Notify a new max latency via fsnotify interface. */ static void notify_new_max_latency(u64 latency) @@ -1296,7 +1384,7 @@ static void notify_new_max_latency(u64 latency) rcu_read_lock(); list_for_each_entry_rcu(inst, &osnoise_instances, list) { tr = inst->tr; - if (tr->max_latency < latency) { + if (tracer_tracing_is_on(tr) && tr->max_latency < latency) { tr->max_latency = latency; latency_fsnotify(tr); } @@ -1345,9 +1433,9 @@ static int run_osnoise(void) save_osn_sample_stats(osn_var, &s); /* - * if threshold is 0, use the default value of 5 us. + * if threshold is 0, use the default value of 1 us. */ - threshold = tracing_thresh ? : 5000; + threshold = tracing_thresh ? : 1000; /* * Apply PREEMPT and IRQ disabled options. @@ -1372,7 +1460,7 @@ static int run_osnoise(void) stop_in = osnoise_data.stop_tracing * NSEC_PER_USEC; /* - * Start timestemp + * Start timestamp */ start = time_get(); @@ -1432,27 +1520,25 @@ static int run_osnoise(void) /* * In some cases, notably when running on a nohz_full CPU with - * a stopped tick PREEMPT_RCU has no way to account for QSs. - * This will eventually cause unwarranted noise as PREEMPT_RCU - * will force preemption as the means of ending the current - * grace period. We avoid this problem by calling - * rcu_momentary_dyntick_idle(), which performs a zero duration - * EQS allowing PREEMPT_RCU to end the current grace period. - * This call shouldn't be wrapped inside an RCU critical - * section. + * a stopped tick PREEMPT_RCU or PREEMPT_LAZY have no way to + * account for QSs. This will eventually cause unwarranted + * noise as RCU forces preemption as the means of ending the + * current grace period. We avoid this by calling + * rcu_momentary_eqs(), which performs a zero duration EQS + * allowing RCU to end the current grace period. This call + * shouldn't be wrapped inside an RCU critical section. * - * Note that in non PREEMPT_RCU kernels QSs are handled through - * cond_resched() + * Normally QSs for other cases are handled through cond_resched(). + * For simplicity, however, we call rcu_momentary_eqs() for all + * configurations here. */ - if (IS_ENABLED(CONFIG_PREEMPT_RCU)) { - if (!disable_irq) - local_irq_disable(); + if (!disable_irq) + local_irq_disable(); - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); - if (!disable_irq) - local_irq_enable(); - } + if (!disable_irq) + local_irq_enable(); /* * For the non-preemptive kernel config: let threads runs, if @@ -1498,7 +1584,7 @@ static int run_osnoise(void) /* Save interference stats info */ diff_osn_sample_stats(osn_var, &s); - trace_osnoise_sample(&s); + record_osnoise_sample(&s); notify_new_max_latency(max_noise); @@ -1513,17 +1599,21 @@ out: static struct cpumask osnoise_cpumask; static struct cpumask save_cpumask; +static struct cpumask kthread_cpumask; /* * osnoise_sleep - sleep until the next period */ -static void osnoise_sleep(void) +static void osnoise_sleep(bool skip_period) { u64 interval; ktime_t wake_time; mutex_lock(&interface_lock); - interval = osnoise_data.sample_period - osnoise_data.sample_runtime; + if (skip_period) + interval = osnoise_data.sample_period; + else + interval = osnoise_data.sample_period - osnoise_data.sample_runtime; mutex_unlock(&interface_lock); /* @@ -1539,13 +1629,47 @@ static void osnoise_sleep(void) wake_time = ktime_add_us(ktime_get(), interval); __set_current_state(TASK_INTERRUPTIBLE); - while (schedule_hrtimeout_range(&wake_time, 0, HRTIMER_MODE_ABS)) { + while (schedule_hrtimeout(&wake_time, HRTIMER_MODE_ABS)) { if (kthread_should_stop()) break; } } /* + * osnoise_migration_pending - checks if the task needs to migrate + * + * osnoise/timerlat threads are per-cpu. If there is a pending request to + * migrate the thread away from the current CPU, something bad has happened. + * Play the good citizen and leave. + * + * Returns 0 if it is safe to continue, 1 otherwise. + */ +static inline int osnoise_migration_pending(void) +{ + if (!current->migration_pending) + return 0; + + /* + * If migration is pending, there is a task waiting for the + * tracer to enable migration. The tracer does not allow migration, + * thus: taint and leave to unblock the blocked thread. + */ + osnoise_taint("migration requested to osnoise threads, leaving."); + + /* + * Unset this thread from the threads managed by the interface. + * The tracers are responsible for cleaning their env before + * exiting. + */ + mutex_lock(&interface_lock); + this_cpu_osn_var()->kthread = NULL; + cpumask_clear_cpu(smp_processor_id(), &kthread_cpumask); + mutex_unlock(&interface_lock); + + return 1; +} + +/* * osnoise_main - The osnoise detection kernel thread * * Calls run_osnoise() function to measure the osnoise for the configured runtime, @@ -1553,12 +1677,35 @@ static void osnoise_sleep(void) */ static int osnoise_main(void *data) { + unsigned long flags; + + /* + * This thread was created pinned to the CPU using PF_NO_SETAFFINITY. + * The problem is that cgroup does not allow PF_NO_SETAFFINITY thread. + * + * To work around this limitation, disable migration and remove the + * flag. + */ + migrate_disable(); + raw_spin_lock_irqsave(¤t->pi_lock, flags); + current->flags &= ~(PF_NO_SETAFFINITY); + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); while (!kthread_should_stop()) { + if (osnoise_migration_pending()) + break; + + /* skip a period if tracing is off on all instances */ + if (!osnoise_has_tracing_on()) { + osnoise_sleep(true); + continue; + } + run_osnoise(); - osnoise_sleep(); + osnoise_sleep(false); } + migrate_enable(); return 0; } @@ -1632,7 +1779,7 @@ static enum hrtimer_restart timerlat_irq(struct hrtimer *timer) s.timer_latency = diff; s.context = IRQ_CONTEXT; - trace_timerlat_sample(&s); + record_timerlat_sample(&s); if (osnoise_data.stop_tracing) { if (time_to_us(diff) >= osnoise_data.stop_tracing) { @@ -1652,6 +1799,8 @@ static enum hrtimer_restart timerlat_irq(struct hrtimer *timer) osnoise_stop_tracing(); notify_new_max_latency(diff); + wake_up_process(tlat->kthread); + return HRTIMER_NORESTART; } } @@ -1704,6 +1853,7 @@ static int timerlat_main(void *data) struct timerlat_variables *tlat = this_cpu_tmr_var(); struct timerlat_sample s; struct sched_param sp; + unsigned long flags; u64 now, diff; /* @@ -1712,15 +1862,26 @@ static int timerlat_main(void *data) sp.sched_priority = DEFAULT_TIMERLAT_PRIO; sched_setscheduler_nocheck(current, SCHED_FIFO, &sp); + /* + * This thread was created pinned to the CPU using PF_NO_SETAFFINITY. + * The problem is that cgroup does not allow PF_NO_SETAFFINITY thread. + * + * To work around this limitation, disable migration and remove the + * flag. + */ + migrate_disable(); + raw_spin_lock_irqsave(¤t->pi_lock, flags); + current->flags &= ~(PF_NO_SETAFFINITY); + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); + tlat->count = 0; tlat->tracing_thread = false; - hrtimer_init(&tlat->timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD); - tlat->timer.function = timerlat_irq; + hrtimer_setup(&tlat->timer, timerlat_irq, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD); tlat->kthread = current; osn_var->pid = current->pid; /* - * Anotate the arrival time. + * Annotate the arrival time. */ tlat->abs_period = hrtimer_cb_get_time(&tlat->timer); @@ -1729,6 +1890,7 @@ static int timerlat_main(void *data) osn_var->sampling = 1; while (!kthread_should_stop()) { + now = ktime_to_ns(hrtimer_cb_get_time(&tlat->timer)); diff = now - tlat->abs_period; @@ -1736,7 +1898,9 @@ static int timerlat_main(void *data) s.timer_latency = diff; s.context = THREAD_CONTEXT; - trace_timerlat_sample(&s); + record_timerlat_sample(&s); + + notify_new_max_latency(diff); timerlat_dump_stack(time_to_us(diff)); @@ -1745,10 +1909,14 @@ static int timerlat_main(void *data) if (time_to_us(diff) >= osnoise_data.stop_tracing_total) osnoise_stop_tracing(); + if (osnoise_migration_pending()) + break; + wait_next_period(tlat); } hrtimer_cancel(&tlat->timer); + migrate_enable(); return 0; } #else /* CONFIG_TIMERLAT_TRACER */ @@ -1765,15 +1933,28 @@ static void stop_kthread(unsigned int cpu) { struct task_struct *kthread; - kthread = per_cpu(per_cpu_osnoise_var, cpu).kthread; + kthread = xchg_relaxed(&(per_cpu(per_cpu_osnoise_var, cpu).kthread), NULL); if (kthread) { - kthread_stop(kthread); - per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL; + if (cpumask_test_and_clear_cpu(cpu, &kthread_cpumask) && + !WARN_ON(!test_bit(OSN_WORKLOAD, &osnoise_options))) { + kthread_stop(kthread); + } else if (!WARN_ON(test_bit(OSN_WORKLOAD, &osnoise_options))) { + /* + * This is a user thread waiting on the timerlat_fd. We need + * to close all users, and the best way to guarantee this is + * by killing the thread. NOTE: this is a purpose specific file. + */ + kill_pid(kthread->thread_pid, SIGKILL, 1); + put_task_struct(kthread); + } } else { + /* if no workload, just return */ if (!test_bit(OSN_WORKLOAD, &osnoise_options)) { + /* + * This is set in the osnoise tracer case. + */ per_cpu(per_cpu_osnoise_var, cpu).sampling = false; barrier(); - return; } } } @@ -1797,7 +1978,7 @@ static void stop_per_cpu_kthreads(void) } /* - * start_kthread - Start a workload tread + * start_kthread - Start a workload thread */ static int start_kthread(unsigned int cpu) { @@ -1805,6 +1986,10 @@ static int start_kthread(unsigned int cpu) void *main = osnoise_main; char comm[24]; + /* Do not start a new thread if it is already running */ + if (per_cpu(per_cpu_osnoise_var, cpu).kthread) + return 0; + if (timerlat_enabled()) { snprintf(comm, 24, "timerlat/%d", cpu); main = timerlat_main; @@ -1815,7 +2000,6 @@ static int start_kthread(unsigned int cpu) barrier(); return 0; } - snprintf(comm, 24, "osnoise/%d", cpu); } @@ -1823,11 +2007,11 @@ static int start_kthread(unsigned int cpu) if (IS_ERR(kthread)) { pr_err(BANNER "could not start sampling thread\n"); - stop_per_cpu_kthreads(); return -ENOMEM; } per_cpu(per_cpu_osnoise_var, cpu).kthread = kthread; + cpumask_set_cpu(cpu, &kthread_cpumask); return 0; } @@ -1844,14 +2028,26 @@ static int start_per_cpu_kthreads(void) int retval = 0; int cpu; + if (!test_bit(OSN_WORKLOAD, &osnoise_options)) { + if (timerlat_enabled()) + return 0; + } + cpus_read_lock(); /* * Run only on online CPUs in which osnoise is allowed to run. */ cpumask_and(current_mask, cpu_online_mask, &osnoise_cpumask); - for_each_possible_cpu(cpu) - per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL; + for_each_possible_cpu(cpu) { + if (cpumask_test_and_clear_cpu(cpu, &kthread_cpumask)) { + struct task_struct *kthread; + + kthread = xchg_relaxed(&(per_cpu(per_cpu_osnoise_var, cpu).kthread), NULL); + if (!WARN_ON(!kthread)) + kthread_stop(kthread); + } + } for_each_cpu(cpu, current_mask) { retval = start_kthread(cpu); @@ -1872,24 +2068,21 @@ static void osnoise_hotplug_workfn(struct work_struct *dummy) { unsigned int cpu = smp_processor_id(); - mutex_lock(&trace_types_lock); + guard(mutex)(&trace_types_lock); if (!osnoise_has_registered_instances()) - goto out_unlock_trace; + return; - mutex_lock(&interface_lock); - cpus_read_lock(); + guard(mutex)(&interface_lock); + guard(cpus_read_lock)(); + + if (!cpu_online(cpu)) + return; if (!cpumask_test_cpu(cpu, &osnoise_cpumask)) - goto out_unlock; + return; start_kthread(cpu); - -out_unlock: - cpus_read_unlock(); - mutex_unlock(&interface_lock); -out_unlock_trace: - mutex_unlock(&trace_types_lock); } static DECLARE_WORK(osnoise_hotplug_work, osnoise_hotplug_workfn); @@ -2087,31 +2280,22 @@ static ssize_t osnoise_cpus_read(struct file *filp, char __user *ubuf, size_t count, loff_t *ppos) { - char *mask_str; + char *mask_str __free(kfree) = NULL; int len; - mutex_lock(&interface_lock); + guard(mutex)(&interface_lock); len = snprintf(NULL, 0, "%*pbl\n", cpumask_pr_args(&osnoise_cpumask)) + 1; mask_str = kmalloc(len, GFP_KERNEL); - if (!mask_str) { - count = -ENOMEM; - goto out_unlock; - } + if (!mask_str) + return -ENOMEM; len = snprintf(mask_str, len, "%*pbl\n", cpumask_pr_args(&osnoise_cpumask)); - if (len >= count) { - count = -EINVAL; - goto out_free; - } + if (len >= count) + return -EINVAL; count = simple_read_from_buffer(ubuf, count, ppos, mask_str, len); -out_free: - kfree(mask_str); -out_unlock: - mutex_unlock(&interface_lock); - return count; } @@ -2119,7 +2303,7 @@ out_unlock: * osnoise_cpus_write - Write function for "cpus" entry * @filp: The active open file structure * @ubuf: The user buffer that contains the value to write - * @cnt: The maximum number of bytes to write to "file" + * @count: The maximum number of bytes to write to "file" * @ppos: The current position in @file * * This function provides a write implementation for the "cpus" @@ -2137,13 +2321,14 @@ osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, { cpumask_var_t osnoise_cpumask_new; int running, err; - char buf[256]; + char *buf __free(kfree) = NULL; - if (count >= 256) - return -EINVAL; + if (count < 1) + return 0; - if (copy_from_user(buf, ubuf, count)) - return -EFAULT; + buf = memdup_user_nul(ubuf, count); + if (IS_ERR(buf)) + return PTR_ERR(buf); if (!zalloc_cpumask_var(&osnoise_cpumask_new, GFP_KERNEL)) return -ENOMEM; @@ -2184,6 +2369,223 @@ err_free: return err; } +#ifdef CONFIG_TIMERLAT_TRACER +static int timerlat_fd_open(struct inode *inode, struct file *file) +{ + struct osnoise_variables *osn_var; + struct timerlat_variables *tlat; + long cpu = (long) inode->i_cdev; + + mutex_lock(&interface_lock); + + /* + * This file is accessible only if timerlat is enabled, and + * NO_OSNOISE_WORKLOAD is set. + */ + if (!timerlat_enabled() || test_bit(OSN_WORKLOAD, &osnoise_options)) { + mutex_unlock(&interface_lock); + return -EINVAL; + } + + migrate_disable(); + + osn_var = this_cpu_osn_var(); + + /* + * The osn_var->pid holds the single access to this file. + */ + if (osn_var->pid) { + mutex_unlock(&interface_lock); + migrate_enable(); + return -EBUSY; + } + + /* + * timerlat tracer is a per-cpu tracer. Check if the user-space too + * is pinned to a single CPU. The tracer laters monitor if the task + * migrates and then disables tracer if it does. However, it is + * worth doing this basic acceptance test to avoid obviusly wrong + * setup. + */ + if (current->nr_cpus_allowed > 1 || cpu != smp_processor_id()) { + mutex_unlock(&interface_lock); + migrate_enable(); + return -EPERM; + } + + /* + * From now on, it is good to go. + */ + file->private_data = inode->i_cdev; + + get_task_struct(current); + + osn_var->kthread = current; + osn_var->pid = current->pid; + + /* + * Setup is done. + */ + mutex_unlock(&interface_lock); + + tlat = this_cpu_tmr_var(); + tlat->count = 0; + + hrtimer_setup(&tlat->timer, timerlat_irq, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD); + + migrate_enable(); + return 0; +}; + +/* + * timerlat_fd_read - Read function for "timerlat_fd" file + * @file: The active open file structure + * @ubuf: The userspace provided buffer to read value into + * @cnt: The maximum number of bytes to read + * @ppos: The current "file" position + * + * Prints 1 on timerlat, the number of interferences on osnoise, -1 on error. + */ +static ssize_t +timerlat_fd_read(struct file *file, char __user *ubuf, size_t count, + loff_t *ppos) +{ + long cpu = (long) file->private_data; + struct osnoise_variables *osn_var; + struct timerlat_variables *tlat; + struct timerlat_sample s; + s64 diff; + u64 now; + + migrate_disable(); + + tlat = this_cpu_tmr_var(); + + /* + * While in user-space, the thread is migratable. There is nothing + * we can do about it. + * So, if the thread is running on another CPU, stop the machinery. + */ + if (cpu == smp_processor_id()) { + if (tlat->uthread_migrate) { + migrate_enable(); + return -EINVAL; + } + } else { + per_cpu_ptr(&per_cpu_timerlat_var, cpu)->uthread_migrate = 1; + osnoise_taint("timerlat user thread migrate\n"); + osnoise_stop_tracing(); + migrate_enable(); + return -EINVAL; + } + + osn_var = this_cpu_osn_var(); + + /* + * The timerlat in user-space runs in a different order: + * the read() starts from the execution of the previous occurrence, + * sleeping for the next occurrence. + * + * So, skip if we are entering on read() before the first wakeup + * from timerlat IRQ: + */ + if (likely(osn_var->sampling)) { + now = ktime_to_ns(hrtimer_cb_get_time(&tlat->timer)); + diff = now - tlat->abs_period; + + /* + * it was not a timer firing, but some other signal? + */ + if (diff < 0) + goto out; + + s.seqnum = tlat->count; + s.timer_latency = diff; + s.context = THREAD_URET; + + record_timerlat_sample(&s); + + notify_new_max_latency(diff); + + tlat->tracing_thread = false; + if (osnoise_data.stop_tracing_total) + if (time_to_us(diff) >= osnoise_data.stop_tracing_total) + osnoise_stop_tracing(); + } else { + tlat->tracing_thread = false; + tlat->kthread = current; + + /* Annotate now to drift new period */ + tlat->abs_period = hrtimer_cb_get_time(&tlat->timer); + + osn_var->sampling = 1; + } + + /* wait for the next period */ + wait_next_period(tlat); + + /* This is the wakeup from this cycle */ + now = ktime_to_ns(hrtimer_cb_get_time(&tlat->timer)); + diff = now - tlat->abs_period; + + /* + * it was not a timer firing, but some other signal? + */ + if (diff < 0) + goto out; + + s.seqnum = tlat->count; + s.timer_latency = diff; + s.context = THREAD_CONTEXT; + + record_timerlat_sample(&s); + + if (osnoise_data.stop_tracing_total) { + if (time_to_us(diff) >= osnoise_data.stop_tracing_total) { + timerlat_dump_stack(time_to_us(diff)); + notify_new_max_latency(diff); + osnoise_stop_tracing(); + } + } + +out: + migrate_enable(); + return 0; +} + +static int timerlat_fd_release(struct inode *inode, struct file *file) +{ + struct osnoise_variables *osn_var; + struct timerlat_variables *tlat_var; + long cpu = (long) file->private_data; + + migrate_disable(); + mutex_lock(&interface_lock); + + osn_var = per_cpu_ptr(&per_cpu_osnoise_var, cpu); + tlat_var = per_cpu_ptr(&per_cpu_timerlat_var, cpu); + + if (tlat_var->kthread) + hrtimer_cancel(&tlat_var->timer); + memset(tlat_var, 0, sizeof(*tlat_var)); + + osn_var->sampling = 0; + osn_var->pid = 0; + + /* + * We are leaving, not being stopped... see stop_kthread(); + */ + if (osn_var->kthread) { + put_task_struct(osn_var->kthread); + osn_var->kthread = NULL; + } + + mutex_unlock(&interface_lock); + migrate_enable(); + return 0; +} +#endif + /* * osnoise/runtime_us: cannot be greater than the period. */ @@ -2239,14 +2641,21 @@ static struct trace_min_max_param osnoise_print_stack = { /* * osnoise/timerlat_period: min 100 us, max 1 s */ -u64 timerlat_min_period = 100; -u64 timerlat_max_period = 1000000; +static u64 timerlat_min_period = 100; +static u64 timerlat_max_period = 1000000; static struct trace_min_max_param timerlat_period = { .lock = &interface_lock, .val = &osnoise_data.timerlat_period, .max = &timerlat_max_period, .min = &timerlat_min_period, }; + +static const struct file_operations timerlat_fd_fops = { + .open = timerlat_fd_open, + .read = timerlat_fd_read, + .release = timerlat_fd_release, + .llseek = generic_file_llseek, +}; #endif static const struct file_operations cpus_fops = { @@ -2284,18 +2693,63 @@ static int init_timerlat_stack_tracefs(struct dentry *top_dir) } #endif /* CONFIG_STACKTRACE */ +static int osnoise_create_cpu_timerlat_fd(struct dentry *top_dir) +{ + struct dentry *timerlat_fd; + struct dentry *per_cpu; + struct dentry *cpu_dir; + char cpu_str[30]; /* see trace.c: tracing_init_tracefs_percpu() */ + long cpu; + + /* + * Why not using tracing instance per_cpu/ dir? + * + * Because osnoise/timerlat have a single workload, having + * multiple files like these are waste of memory. + */ + per_cpu = tracefs_create_dir("per_cpu", top_dir); + if (!per_cpu) + return -ENOMEM; + + for_each_possible_cpu(cpu) { + snprintf(cpu_str, 30, "cpu%ld", cpu); + cpu_dir = tracefs_create_dir(cpu_str, per_cpu); + if (!cpu_dir) + goto out_clean; + + timerlat_fd = trace_create_file("timerlat_fd", TRACE_MODE_READ, + cpu_dir, NULL, &timerlat_fd_fops); + if (!timerlat_fd) + goto out_clean; + + /* Record the CPU */ + d_inode(timerlat_fd)->i_cdev = (void *)(cpu); + } + + return 0; + +out_clean: + tracefs_remove(per_cpu); + return -ENOMEM; +} + /* * init_timerlat_tracefs - A function to initialize the timerlat interface files */ static int init_timerlat_tracefs(struct dentry *top_dir) { struct dentry *tmp; + int retval; tmp = tracefs_create_file("timerlat_period_us", TRACE_MODE_WRITE, top_dir, &timerlat_period, &trace_min_max_fops); if (!tmp) return -ENOMEM; + retval = osnoise_create_cpu_timerlat_fd(top_dir); + if (retval) + return retval; + return init_timerlat_stack_tracefs(top_dir); } #else /* CONFIG_TIMERLAT_TRACER */ diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index bd475a00f96d..cc2d3306bb60 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -5,6 +5,7 @@ * Copyright (C) 2008 Red Hat Inc, Steven Rostedt <srostedt@redhat.com> * */ +#include "trace.h" #include <linux/module.h> #include <linux/mutex.h> #include <linux/ftrace.h> @@ -12,15 +13,19 @@ #include <linux/sched/clock.h> #include <linux/sched/mm.h> #include <linux/idr.h> +#include <linux/btf.h> +#include <linux/bpf.h> +#include <linux/hashtable.h> #include "trace_output.h" +#include "trace_btf.h" -/* must be a power of 2 */ -#define EVENT_HASHSIZE 128 +/* 2^7 = 128 */ +#define EVENT_HASH_BITS 7 DECLARE_RWSEM(trace_event_sem); -static struct hlist_head event_hash[EVENT_HASHSIZE] __read_mostly; +static DEFINE_HASHTABLE(event_hash, EVENT_HASH_BITS); enum print_line_t trace_print_bputs_msg_only(struct trace_iterator *iter) { @@ -221,8 +226,11 @@ trace_print_hex_seq(struct trace_seq *p, const unsigned char *buf, int buf_len, const char *ret = trace_seq_buffer_ptr(p); const char *fmt = concatenate ? "%*phN" : "%*ph"; - for (i = 0; i < buf_len; i += 16) + for (i = 0; i < buf_len; i += 16) { + if (!concatenate && i != 0) + trace_seq_putc(p, ' '); trace_seq_printf(p, fmt, min(buf_len - i, 16), &buf[i]); + } trace_seq_putc(p, 0); return ret; @@ -314,10 +322,14 @@ EXPORT_SYMBOL(trace_raw_output_prep); void trace_event_printf(struct trace_iterator *iter, const char *fmt, ...) { + struct trace_seq *s = &iter->seq; va_list ap; + if (ignore_event(iter)) + return; + va_start(ap, fmt); - trace_check_vprintf(iter, trace_event_format(iter, fmt), ap); + trace_seq_vprintf(s, trace_event_format(iter, fmt), ap); va_end(ap); } EXPORT_SYMBOL(trace_event_printf); @@ -401,14 +413,14 @@ static int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm, vmstart = vma->vm_start; } if (file) { - ret = trace_seq_path(s, &file->f_path); + ret = trace_seq_path(s, file_user_path(file)); if (ret) trace_seq_printf(s, "[+0x%lx]", ip - vmstart); } mmap_read_unlock(mm); } - if (ret && ((sym_flags & TRACE_ITER_SYM_ADDR) || !file)) + if (ret && ((sym_flags & TRACE_ITER(SYM_ADDR)) || !file)) trace_seq_printf(s, " <" IP_FMT ">", ip); return !trace_seq_has_overflowed(s); } @@ -421,9 +433,9 @@ seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags) goto out; } - trace_seq_print_sym(s, ip, sym_flags & TRACE_ITER_SYM_OFFSET); + trace_seq_print_sym(s, ip, sym_flags & TRACE_ITER(SYM_OFFSET)); - if (sym_flags & TRACE_ITER_SYM_ADDR) + if (sym_flags & TRACE_ITER(SYM_ADDR)) trace_seq_printf(s, " <" IP_FMT ">", ip); out: @@ -457,20 +469,31 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry) (entry->flags & TRACE_FLAG_IRQS_OFF && bh_off) ? 'D' : (entry->flags & TRACE_FLAG_IRQS_OFF) ? 'd' : bh_off ? 'b' : - (entry->flags & TRACE_FLAG_IRQS_NOSUPPORT) ? 'X' : '.'; - switch (entry->flags & (TRACE_FLAG_NEED_RESCHED | + switch (entry->flags & (TRACE_FLAG_NEED_RESCHED | TRACE_FLAG_NEED_RESCHED_LAZY | TRACE_FLAG_PREEMPT_RESCHED)) { + case TRACE_FLAG_NEED_RESCHED | TRACE_FLAG_NEED_RESCHED_LAZY | TRACE_FLAG_PREEMPT_RESCHED: + need_resched = 'B'; + break; case TRACE_FLAG_NEED_RESCHED | TRACE_FLAG_PREEMPT_RESCHED: need_resched = 'N'; break; + case TRACE_FLAG_NEED_RESCHED_LAZY | TRACE_FLAG_PREEMPT_RESCHED: + need_resched = 'L'; + break; + case TRACE_FLAG_NEED_RESCHED | TRACE_FLAG_NEED_RESCHED_LAZY: + need_resched = 'b'; + break; case TRACE_FLAG_NEED_RESCHED: need_resched = 'n'; break; case TRACE_FLAG_PREEMPT_RESCHED: need_resched = 'p'; break; + case TRACE_FLAG_NEED_RESCHED_LAZY: + need_resched = 'l'; + break; default: need_resched = '.'; break; @@ -546,7 +569,7 @@ static int lat_print_timestamp(struct trace_iterator *iter, u64 next_ts) { struct trace_array *tr = iter->tr; - unsigned long verbose = tr->trace_flags & TRACE_ITER_VERBOSE; + unsigned long verbose = tr->trace_flags & TRACE_ITER(VERBOSE); unsigned long in_ns = iter->iter_flags & TRACE_FILE_TIME_IN_NS; unsigned long long abs_ts = iter->ts - iter->array_buffer->time_start; unsigned long long rel_ts = next_ts - iter->ts; @@ -613,7 +636,7 @@ int trace_print_context(struct trace_iterator *iter) trace_seq_printf(s, "%16s-%-7d ", comm, entry->pid); - if (tr->trace_flags & TRACE_ITER_RECORD_TGID) { + if (tr->trace_flags & TRACE_ITER(RECORD_TGID)) { unsigned int tgid = trace_find_tgid(entry->pid); if (!tgid) @@ -624,7 +647,7 @@ int trace_print_context(struct trace_iterator *iter) trace_seq_printf(s, "[%03d] ", iter->cpu); - if (tr->trace_flags & TRACE_ITER_IRQ_INFO) + if (tr->trace_flags & TRACE_ITER(IRQ_INFO)) trace_print_lat_fmt(s, entry); trace_print_time(s, iter, iter->ts); @@ -638,7 +661,7 @@ int trace_print_lat_context(struct trace_iterator *iter) struct trace_entry *entry, *next_entry; struct trace_array *tr = iter->tr; struct trace_seq *s = &iter->seq; - unsigned long verbose = (tr->trace_flags & TRACE_ITER_VERBOSE); + unsigned long verbose = (tr->trace_flags & TRACE_ITER(VERBOSE)); u64 next_ts; next_entry = trace_find_next_entry(iter, NULL, &next_ts); @@ -666,6 +689,94 @@ int trace_print_lat_context(struct trace_iterator *iter) return !trace_seq_has_overflowed(s); } +#ifdef CONFIG_FUNCTION_TRACE_ARGS +void print_function_args(struct trace_seq *s, unsigned long *args, + unsigned long func) +{ + const struct btf_param *param; + const struct btf_type *t; + const char *param_name; + char name[KSYM_NAME_LEN]; + unsigned long arg; + struct btf *btf; + s32 tid, nr = 0; + int a, p, x; + u16 encode; + + trace_seq_printf(s, "("); + + if (!args) + goto out; + if (lookup_symbol_name(func, name)) + goto out; + + /* TODO: Pass module name here too */ + t = btf_find_func_proto(name, &btf); + if (IS_ERR_OR_NULL(t)) + goto out; + + param = btf_get_func_param(t, &nr); + if (!param) + goto out_put; + + for (a = 0, p = 0; p < nr; a++, p++) { + if (p) + trace_seq_puts(s, ", "); + + /* This only prints what the arch allows (6 args by default) */ + if (a == FTRACE_REGS_MAX_ARGS) { + trace_seq_puts(s, "..."); + break; + } + + arg = args[a]; + + param_name = btf_name_by_offset(btf, param[p].name_off); + if (param_name) + trace_seq_printf(s, "%s=", param_name); + t = btf_type_skip_modifiers(btf, param[p].type, &tid); + + switch (t ? BTF_INFO_KIND(t->info) : BTF_KIND_UNKN) { + case BTF_KIND_UNKN: + trace_seq_putc(s, '?'); + /* Still print unknown type values */ + fallthrough; + case BTF_KIND_PTR: + trace_seq_printf(s, "0x%lx", arg); + break; + case BTF_KIND_INT: + encode = btf_int_encoding(t); + /* Print unsigned ints as hex */ + if (encode & BTF_INT_SIGNED) + trace_seq_printf(s, "%ld", arg); + else + trace_seq_printf(s, "0x%lx", arg); + break; + case BTF_KIND_ENUM: + trace_seq_printf(s, "%ld", arg); + break; + default: + /* This does not handle complex arguments */ + trace_seq_printf(s, "(%s)[0x%lx", btf_type_str(t), arg); + for (x = sizeof(long); x < t->size; x += sizeof(long)) { + trace_seq_putc(s, ':'); + if (++a == FTRACE_REGS_MAX_ARGS) { + trace_seq_puts(s, "...]"); + goto out_put; + } + trace_seq_printf(s, "0x%lx", args[a]); + } + trace_seq_putc(s, ']'); + break; + } + } +out_put: + btf_put(btf); +out: + trace_seq_printf(s, ")"); +} +#endif + /** * ftrace_find_event - find a registered event * @type: the type of event to look for @@ -676,11 +787,8 @@ int trace_print_lat_context(struct trace_iterator *iter) struct trace_event *ftrace_find_event(int type) { struct trace_event *event; - unsigned key; - - key = type & (EVENT_HASHSIZE - 1); - hlist_for_each_entry(event, &event_hash[key], node) { + hash_for_each_possible(event_hash, event, node, type) { if (event->type == type) return event; } @@ -735,7 +843,6 @@ void trace_event_read_unlock(void) */ int register_trace_event(struct trace_event *event) { - unsigned key; int ret = 0; down_write(&trace_event_sem); @@ -768,9 +875,7 @@ int register_trace_event(struct trace_event *event) if (event->funcs->binary == NULL) event->funcs->binary = trace_nop_print; - key = event->type & (EVENT_HASHSIZE - 1); - - hlist_add_head(&event->node, &event_hash[key]); + hash_add(event_hash, &event->node, event->type); ret = event->type; out: @@ -785,7 +890,7 @@ EXPORT_SYMBOL_GPL(register_trace_event); */ int __unregister_trace_event(struct trace_event *event) { - hlist_del(&event->node); + hash_del(&event->node); free_trace_event_type(event->type); return 0; } @@ -808,6 +913,224 @@ EXPORT_SYMBOL_GPL(unregister_trace_event); * Standard events */ +static void print_array(struct trace_iterator *iter, void *pos, + struct ftrace_event_field *field) +{ + int offset; + int len; + int i; + + offset = *(int *)pos & 0xffff; + len = *(int *)pos >> 16; + + if (field) + offset += field->offset + sizeof(int); + + if (offset + len > iter->ent_size) { + trace_seq_puts(&iter->seq, "<OVERFLOW>"); + return; + } + + pos = (void *)iter->ent + offset; + + for (i = 0; i < len; i++, pos++) { + if (i) + trace_seq_putc(&iter->seq, ','); + trace_seq_printf(&iter->seq, "%02x", *(unsigned char *)pos); + } +} + +static void print_fields(struct trace_iterator *iter, struct trace_event_call *call, + struct list_head *head) +{ + struct ftrace_event_field *field; + struct trace_array *tr = iter->tr; + unsigned long long laddr; + unsigned long addr; + int offset; + int len; + int ret; + int i; + void *pos; + char *str; + + list_for_each_entry_reverse(field, head, link) { + trace_seq_printf(&iter->seq, " %s=", field->name); + if (field->offset + field->size > iter->ent_size) { + trace_seq_puts(&iter->seq, "<OVERFLOW>"); + continue; + } + pos = (void *)iter->ent + field->offset; + + switch (field->filter_type) { + case FILTER_COMM: + case FILTER_STATIC_STRING: + trace_seq_printf(&iter->seq, "%.*s", field->size, (char *)pos); + break; + case FILTER_RDYN_STRING: + case FILTER_DYN_STRING: + offset = *(int *)pos & 0xffff; + len = *(int *)pos >> 16; + + if (field->filter_type == FILTER_RDYN_STRING) + offset += field->offset + sizeof(int); + + if (offset + len > iter->ent_size) { + trace_seq_puts(&iter->seq, "<OVERFLOW>"); + break; + } + str = (char *)iter->ent + offset; + /* Check if there's any non printable strings */ + for (i = 0; i < len; i++) { + if (str[i] && !(isascii(str[i]) && isprint(str[i]))) + break; + } + if (i < len) { + for (i = 0; i < len; i++) { + if (isascii(str[i]) && isprint(str[i])) + trace_seq_putc(&iter->seq, str[i]); + else + trace_seq_putc(&iter->seq, '.'); + } + trace_seq_puts(&iter->seq, " ("); + for (i = 0; i < len; i++) { + if (i) + trace_seq_putc(&iter->seq, ':'); + trace_seq_printf(&iter->seq, "%02x", str[i]); + } + trace_seq_putc(&iter->seq, ')'); + } else { + trace_seq_printf(&iter->seq, "%.*s", len, str); + } + break; + case FILTER_PTR_STRING: + if (!iter->fmt_size) + trace_iter_expand_format(iter); + addr = trace_adjust_address(tr, *(unsigned long *)pos); + ret = strncpy_from_kernel_nofault(iter->fmt, (void *)addr, + iter->fmt_size); + if (ret < 0) + trace_seq_printf(&iter->seq, "(0x%px)", pos); + else + trace_seq_printf(&iter->seq, "(0x%px:%s)", + pos, iter->fmt); + break; + case FILTER_TRACE_FN: + addr = trace_adjust_address(tr, *(unsigned long *)pos); + trace_seq_printf(&iter->seq, "%pS", (void *)addr); + break; + case FILTER_CPU: + case FILTER_OTHER: + switch (field->size) { + case 1: + if (isprint(*(char *)pos)) { + trace_seq_printf(&iter->seq, "'%c'", + *(unsigned char *)pos); + } + trace_seq_printf(&iter->seq, "(%d)", + *(unsigned char *)pos); + break; + case 2: + trace_seq_printf(&iter->seq, "0x%x (%d)", + *(unsigned short *)pos, + *(unsigned short *)pos); + break; + case 4: + /* dynamic array info is 4 bytes */ + if (strstr(field->type, "__data_loc")) { + print_array(iter, pos, NULL); + break; + } + + if (strstr(field->type, "__rel_loc")) { + print_array(iter, pos, field); + break; + } + + addr = *(unsigned int *)pos; + + /* Some fields reference offset from _stext. */ + if (!strcmp(field->name, "caller_offs") || + !strcmp(field->name, "parent_offs")) { + unsigned long ip; + + ip = addr + (unsigned long)_stext; + ip = trace_adjust_address(tr, ip); + trace_seq_printf(&iter->seq, "%pS ", (void *)ip); + } + + if (sizeof(long) == 4) { + addr = trace_adjust_address(tr, addr); + trace_seq_printf(&iter->seq, "%pS (%d)", + (void *)addr, (int)addr); + } else { + trace_seq_printf(&iter->seq, "0x%x (%d)", + (unsigned int)addr, (int)addr); + } + break; + case 8: + laddr = *(unsigned long long *)pos; + if (sizeof(long) == 8) { + laddr = trace_adjust_address(tr, (unsigned long)laddr); + trace_seq_printf(&iter->seq, "%pS (%lld)", + (void *)(long)laddr, laddr); + } else { + trace_seq_printf(&iter->seq, "0x%llx (%lld)", laddr, laddr); + } + break; + default: + trace_seq_puts(&iter->seq, "<INVALID-SIZE>"); + break; + } + break; + default: + trace_seq_puts(&iter->seq, "<INVALID-TYPE>"); + } + } + trace_seq_putc(&iter->seq, '\n'); +} + +enum print_line_t print_event_fields(struct trace_iterator *iter, + struct trace_event *event) +{ + struct trace_event_call *call; + struct list_head *head; + + lockdep_assert_held_read(&trace_event_sem); + + /* ftrace defined events have separate call structures */ + if (event->type <= __TRACE_LAST_TYPE) { + bool found = false; + + list_for_each_entry(call, &ftrace_events, list) { + if (call->event.type == event->type) { + found = true; + break; + } + /* No need to search all events */ + if (call->event.type > __TRACE_LAST_TYPE) + break; + } + if (!found) { + trace_seq_printf(&iter->seq, "UNKNOWN TYPE %d\n", event->type); + goto out; + } + } else { + call = container_of(event, struct trace_event_call, event); + } + head = trace_get_fields(call); + + trace_seq_printf(&iter->seq, "%s:", trace_event_name(call)); + + if (head && !list_empty(head)) + print_fields(iter, call, head); + else + trace_seq_puts(&iter->seq, "No fields found\n"); + + out: + return trace_handle_return(&iter->seq); +} + enum print_line_t trace_nop_print(struct trace_iterator *iter, int flags, struct trace_event *event) { @@ -817,11 +1140,17 @@ enum print_line_t trace_nop_print(struct trace_iterator *iter, int flags, } static void print_fn_trace(struct trace_seq *s, unsigned long ip, - unsigned long parent_ip, int flags) + unsigned long parent_ip, unsigned long *args, + struct trace_array *tr, int flags) { + ip = trace_adjust_address(tr, ip); + parent_ip = trace_adjust_address(tr, parent_ip); + seq_print_ip_sym(s, ip, flags); + if (args) + print_function_args(s, args, ip); - if ((flags & TRACE_ITER_PRINT_PARENT) && parent_ip) { + if ((flags & TRACE_ITER(PRINT_PARENT)) && parent_ip) { trace_seq_puts(s, " <-"); seq_print_ip_sym(s, parent_ip, flags); } @@ -833,10 +1162,18 @@ static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags, { struct ftrace_entry *field; struct trace_seq *s = &iter->seq; + unsigned long *args; + int args_size; trace_assign_type(field, iter->ent); - print_fn_trace(s, field->ip, field->parent_ip, flags); + args_size = iter->ent_size - offsetof(struct ftrace_entry, args); + if (args_size >= FTRACE_REGS_MAX_ARGS * sizeof(long)) + args = field->args; + else + args = NULL; + + print_fn_trace(s, field->ip, field->parent_ip, args, iter->tr, flags); trace_seq_putc(s, '\n'); return trace_handle_return(s); @@ -1069,7 +1406,11 @@ static enum print_line_t trace_stack_print(struct trace_iterator *iter, break; trace_seq_puts(s, " => "); - seq_print_ip_sym(s, *p, flags); + if ((*p) == FTRACE_TRAMPOLINE_MARKER) { + trace_seq_puts(s, "[FTRACE TRAMPOLINE]\n"); + continue; + } + seq_print_ip_sym(s, trace_adjust_address(iter->tr, *p), flags); trace_seq_putc(s, '\n'); } @@ -1099,7 +1440,7 @@ static enum print_line_t trace_user_stack_print(struct trace_iterator *iter, trace_seq_puts(s, "<user stack trace>\n"); - if (tr->trace_flags & TRACE_ITER_SYM_USEROBJ) { + if (tr->trace_flags & TRACE_ITER(SYM_USEROBJ)) { struct task_struct *task; /* * we do the lookup on the thread group leader, @@ -1149,12 +1490,12 @@ trace_hwlat_print(struct trace_iterator *iter, int flags, trace_assign_type(field, entry); - trace_seq_printf(s, "#%-5u inner/outer(us): %4llu/%-5llu ts:%lld.%09ld count:%d", + trace_seq_printf(s, "#%-5u inner/outer(us): %4llu/%-5llu ts:%ptSp count:%d", field->seqnum, field->duration, field->outer_duration, - (long long)field->timestamp.tv_sec, - field->timestamp.tv_nsec, field->count); + &field->timestamp, + field->count); if (field->nmi_count) { /* @@ -1273,6 +1614,8 @@ static struct trace_event trace_osnoise_event = { }; /* TRACE_TIMERLAT */ + +static char *timerlat_lat_context[] = {"irq", "thread", "user-ret"}; static enum print_line_t trace_timerlat_print(struct trace_iterator *iter, int flags, struct trace_event *event) @@ -1285,7 +1628,7 @@ trace_timerlat_print(struct trace_iterator *iter, int flags, trace_seq_printf(s, "#%-5u context %6s timer_latency %9llu ns\n", field->seqnum, - field->context ? "thread" : "irq", + timerlat_lat_context[field->context], field->timer_latency); return trace_handle_return(s); @@ -1412,10 +1755,13 @@ static enum print_line_t trace_print_print(struct trace_iterator *iter, { struct print_entry *field; struct trace_seq *s = &iter->seq; + unsigned long ip; trace_assign_type(field, iter->ent); - seq_print_ip_sym(s, field->ip, flags); + ip = trace_adjust_address(iter->tr, field->ip); + + seq_print_ip_sym(s, ip, flags); trace_seq_printf(s, ": %s", field->buf); return trace_handle_return(s); @@ -1499,7 +1845,7 @@ trace_func_repeats_print(struct trace_iterator *iter, int flags, trace_assign_type(field, iter->ent); - print_fn_trace(s, field->ip, field->parent_ip, flags); + print_fn_trace(s, field->ip, field->parent_ip, NULL, iter->tr, flags); trace_seq_printf(s, " (repeats: %u, last_ts:", field->count); trace_print_time(s, iter, iter->ts - FUNC_REPEATS_GET_DELTA_TS(field)); diff --git a/kernel/trace/trace_output.h b/kernel/trace/trace_output.h index 4c954636caf0..99b676733d46 100644 --- a/kernel/trace/trace_output.h +++ b/kernel/trace/trace_output.h @@ -16,9 +16,22 @@ extern int seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags); +static inline int seq_print_ip_sym_offset(struct trace_seq *s, unsigned long ip, + unsigned long sym_flags) +{ + return seq_print_ip_sym(s, ip, sym_flags | TRACE_ITER(SYM_OFFSET)); +} +static inline int seq_print_ip_sym_no_offset(struct trace_seq *s, unsigned long ip, + unsigned long sym_flags) +{ + return seq_print_ip_sym(s, ip, sym_flags & ~TRACE_ITER(SYM_OFFSET)); +} + extern void trace_seq_print_sym(struct trace_seq *s, unsigned long address, bool offset); extern int trace_print_context(struct trace_iterator *iter); extern int trace_print_lat_context(struct trace_iterator *iter); +extern enum print_line_t print_event_fields(struct trace_iterator *iter, + struct trace_event *event); extern void trace_event_read_lock(void); extern void trace_event_read_unlock(void); @@ -39,5 +52,14 @@ extern struct rw_semaphore trace_event_sem; #define SEQ_PUT_HEX_FIELD(s, x) \ trace_seq_putmem_hex(s, &(x), sizeof(x)) +#ifdef CONFIG_FUNCTION_TRACE_ARGS +void print_function_args(struct trace_seq *s, unsigned long *args, + unsigned long func); +#else +static inline void print_function_args(struct trace_seq *s, unsigned long *args, + unsigned long func) { + trace_seq_puts(s, "()"); +} +#endif #endif diff --git a/kernel/trace/trace_preemptirq.c b/kernel/trace/trace_preemptirq.c index 1e130da1b742..0c42b15c3800 100644 --- a/kernel/trace/trace_preemptirq.c +++ b/kernel/trace/trace_preemptirq.c @@ -10,11 +10,42 @@ #include <linux/module.h> #include <linux/ftrace.h> #include <linux/kprobes.h> +#include <linux/hardirq.h> #include "trace.h" #define CREATE_TRACE_POINTS #include <trace/events/preemptirq.h> +/* + * Use regular trace points on architectures that implement noinstr + * tooling: these calls will only happen with RCU enabled, which can + * use a regular tracepoint. + * + * On older architectures, RCU may not be watching in idle. In that + * case, wake up RCU to watch while calling the tracepoint. These + * aren't NMI-safe - so exclude NMI contexts: + */ +#ifdef CONFIG_ARCH_WANTS_NO_INSTR +#define trace(point, args) trace_##point(args) +#else +#define trace(point, args) \ + do { \ + if (trace_##point##_enabled()) { \ + bool exit_rcu = false; \ + if (in_nmi()) \ + break; \ + if (!IS_ENABLED(CONFIG_TINY_RCU) && \ + is_idle_task(current)) { \ + ct_irq_enter(); \ + exit_rcu = true; \ + } \ + trace_##point(args); \ + if (exit_rcu) \ + ct_irq_exit(); \ + } \ + } while (0) +#endif + #ifdef CONFIG_TRACE_IRQFLAGS /* Per-cpu variable to prevent redundant calls when IRQs already off */ static DEFINE_PER_CPU(int, tracing_irq_cpu); @@ -28,8 +59,7 @@ static DEFINE_PER_CPU(int, tracing_irq_cpu); void trace_hardirqs_on_prepare(void) { if (this_cpu_read(tracing_irq_cpu)) { - if (!in_nmi()) - trace_irq_enable(CALLER_ADDR0, CALLER_ADDR1); + trace(irq_enable, TP_ARGS(CALLER_ADDR0, CALLER_ADDR1)); tracer_hardirqs_on(CALLER_ADDR0, CALLER_ADDR1); this_cpu_write(tracing_irq_cpu, 0); } @@ -40,8 +70,7 @@ NOKPROBE_SYMBOL(trace_hardirqs_on_prepare); void trace_hardirqs_on(void) { if (this_cpu_read(tracing_irq_cpu)) { - if (!in_nmi()) - trace_irq_enable_rcuidle(CALLER_ADDR0, CALLER_ADDR1); + trace(irq_enable, TP_ARGS(CALLER_ADDR0, CALLER_ADDR1)); tracer_hardirqs_on(CALLER_ADDR0, CALLER_ADDR1); this_cpu_write(tracing_irq_cpu, 0); } @@ -63,8 +92,7 @@ void trace_hardirqs_off_finish(void) if (!this_cpu_read(tracing_irq_cpu)) { this_cpu_write(tracing_irq_cpu, 1); tracer_hardirqs_off(CALLER_ADDR0, CALLER_ADDR1); - if (!in_nmi()) - trace_irq_disable(CALLER_ADDR0, CALLER_ADDR1); + trace(irq_disable, TP_ARGS(CALLER_ADDR0, CALLER_ADDR1)); } } @@ -78,56 +106,24 @@ void trace_hardirqs_off(void) if (!this_cpu_read(tracing_irq_cpu)) { this_cpu_write(tracing_irq_cpu, 1); tracer_hardirqs_off(CALLER_ADDR0, CALLER_ADDR1); - if (!in_nmi()) - trace_irq_disable_rcuidle(CALLER_ADDR0, CALLER_ADDR1); + trace(irq_disable, TP_ARGS(CALLER_ADDR0, CALLER_ADDR1)); } } EXPORT_SYMBOL(trace_hardirqs_off); NOKPROBE_SYMBOL(trace_hardirqs_off); - -__visible void trace_hardirqs_on_caller(unsigned long caller_addr) -{ - if (this_cpu_read(tracing_irq_cpu)) { - if (!in_nmi()) - trace_irq_enable_rcuidle(CALLER_ADDR0, caller_addr); - tracer_hardirqs_on(CALLER_ADDR0, caller_addr); - this_cpu_write(tracing_irq_cpu, 0); - } - - lockdep_hardirqs_on_prepare(); - lockdep_hardirqs_on(caller_addr); -} -EXPORT_SYMBOL(trace_hardirqs_on_caller); -NOKPROBE_SYMBOL(trace_hardirqs_on_caller); - -__visible void trace_hardirqs_off_caller(unsigned long caller_addr) -{ - lockdep_hardirqs_off(caller_addr); - - if (!this_cpu_read(tracing_irq_cpu)) { - this_cpu_write(tracing_irq_cpu, 1); - tracer_hardirqs_off(CALLER_ADDR0, caller_addr); - if (!in_nmi()) - trace_irq_disable_rcuidle(CALLER_ADDR0, caller_addr); - } -} -EXPORT_SYMBOL(trace_hardirqs_off_caller); -NOKPROBE_SYMBOL(trace_hardirqs_off_caller); #endif /* CONFIG_TRACE_IRQFLAGS */ #ifdef CONFIG_TRACE_PREEMPT_TOGGLE void trace_preempt_on(unsigned long a0, unsigned long a1) { - if (!in_nmi()) - trace_preempt_enable_rcuidle(a0, a1); + trace(preempt_enable, TP_ARGS(a0, a1)); tracer_preempt_on(a0, a1); } void trace_preempt_off(unsigned long a0, unsigned long a1) { - if (!in_nmi()) - trace_preempt_disable_rcuidle(a0, a1); + trace(preempt_disable, TP_ARGS(a0, a1)); tracer_preempt_off(a0, a1); } #endif diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 01ebabbbe8c9..2f571083ce9e 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -11,6 +11,10 @@ */ #define pr_fmt(fmt) "trace_probe: " fmt +#include <linux/bpf.h> +#include <linux/fs.h> + +#include "trace_btf.h" #include "trace_probe.h" #undef C @@ -50,6 +54,7 @@ DEFINE_BASIC_PRINT_TYPE_FUNC(x8, u8, "0x%x") DEFINE_BASIC_PRINT_TYPE_FUNC(x16, u16, "0x%x") DEFINE_BASIC_PRINT_TYPE_FUNC(x32, u32, "0x%x") DEFINE_BASIC_PRINT_TYPE_FUNC(x64, u64, "0x%Lx") +DEFINE_BASIC_PRINT_TYPE_FUNC(char, u8, "'%c'") int PRINT_TYPE_FUNC_NAME(symbol)(struct trace_seq *s, void *data, void *ent) { @@ -64,7 +69,7 @@ int PRINT_TYPE_FUNC_NAME(string)(struct trace_seq *s, void *data, void *ent) int len = *(u32 *)data >> 16; if (!len) - trace_seq_puts(s, "(fault)"); + trace_seq_puts(s, FAULT_STRING); else trace_seq_printf(s, "\"%s\"", (const char *)get_loc_data(data, ent)); @@ -95,6 +100,7 @@ static const struct fetch_type probe_fetch_types[] = { ASSIGN_FETCH_TYPE_ALIAS(x16, u16, u16, 0), ASSIGN_FETCH_TYPE_ALIAS(x32, u32, u32, 0), ASSIGN_FETCH_TYPE_ALIAS(x64, u64, u64, 0), + ASSIGN_FETCH_TYPE_ALIAS(char, u8, u8, 0), ASSIGN_FETCH_TYPE_ALIAS(symbol, ADDR_FETCH_TYPE, ADDR_FETCH_TYPE, 0), ASSIGN_FETCH_TYPE_END @@ -148,22 +154,30 @@ fail: } static struct trace_probe_log trace_probe_log; +extern struct mutex dyn_event_ops_mutex; -void trace_probe_log_init(const char *subsystem, int argc, const char **argv) +const char *trace_probe_log_init(const char *subsystem, int argc, const char **argv) { + lockdep_assert_held(&dyn_event_ops_mutex); + trace_probe_log.subsystem = subsystem; trace_probe_log.argc = argc; trace_probe_log.argv = argv; trace_probe_log.index = 0; + return subsystem; } void trace_probe_log_clear(void) { + lockdep_assert_held(&dyn_event_ops_mutex); + memset(&trace_probe_log, 0, sizeof(trace_probe_log)); } void trace_probe_log_set_index(int index) { + lockdep_assert_held(&dyn_event_ops_mutex); + trace_probe_log.index = index; } @@ -172,6 +186,8 @@ void __trace_probe_log_err(int offset, int err_type) char *command, *p; int i, len = 0, pos = 0; + lockdep_assert_held(&dyn_event_ops_mutex); + if (!trace_probe_log.argv) return; @@ -199,7 +215,7 @@ void __trace_probe_log_err(int offset, int err_type) p = command; for (i = 0; i < trace_probe_log.argc; i++) { len = strlen(trace_probe_log.argv[i]); - strcpy(p, trace_probe_log.argv[i]); + memcpy(p, trace_probe_log.argv[i], len); p[len] = ' '; p += len + 1; } @@ -232,7 +248,25 @@ int traceprobe_split_symbol_offset(char *symbol, long *offset) return 0; } -/* @buf must has MAX_EVENT_NAME_LEN size */ +/** + * traceprobe_parse_event_name() - Parse a string into group and event names + * @pevent: A pointer to the string to be parsed. + * @pgroup: A pointer to the group name. + * @buf: A buffer to store the parsed group name. + * @offset: The offset of the string in the original user command, for logging. + * + * This parses a string with the format `[GROUP/][EVENT]` or `[GROUP.][EVENT]` + * (either GROUP or EVENT or both must be specified). + * Since the parsed group name is stored in @buf, the caller must ensure @buf + * is at least MAX_EVENT_NAME_LEN bytes. + * + * Return: 0 on success, or -EINVAL on failure. + * + * If success, *@pevent is updated to point to the event name part of the + * original string, or NULL if there is no event name. + * Also, *@pgroup is updated to point to the parsed group which is stored + * in @buf, or NULL if there is no group name. + */ int traceprobe_parse_event_name(const char **pevent, const char **pgroup, char *buf, int offset) { @@ -252,7 +286,7 @@ int traceprobe_parse_event_name(const char **pevent, const char **pgroup, trace_probe_log_err(offset, GROUP_TOO_LONG); return -EINVAL; } - strlcpy(buf, event, slash - event + 1); + strscpy(buf, event, slash - event + 1); if (!is_good_system_name(buf)) { trace_probe_log_err(offset, BAD_GROUP_NAME); return -EINVAL; @@ -270,7 +304,7 @@ int traceprobe_parse_event_name(const char **pevent, const char **pgroup, } trace_probe_log_err(offset, NO_EVENT_NAME); return -EINVAL; - } else if (len > MAX_EVENT_NAME_LEN) { + } else if (len >= MAX_EVENT_NAME_LEN) { trace_probe_log_err(offset, EVENT_TOO_LONG); return -EINVAL; } @@ -281,69 +315,742 @@ int traceprobe_parse_event_name(const char **pevent, const char **pgroup, return 0; } +static int parse_trace_event_arg(char *arg, struct fetch_insn *code, + struct traceprobe_parse_context *ctx) +{ + struct ftrace_event_field *field; + struct list_head *head; + + head = trace_get_fields(ctx->event); + list_for_each_entry(field, head, link) { + if (!strcmp(arg, field->name)) { + code->op = FETCH_OP_TP_ARG; + code->data = field; + return 0; + } + } + return -ENOENT; +} + +#ifdef CONFIG_PROBE_EVENTS_BTF_ARGS + +static u32 btf_type_int(const struct btf_type *t) +{ + return *(u32 *)(t + 1); +} + +static bool btf_type_is_char_ptr(struct btf *btf, const struct btf_type *type) +{ + const struct btf_type *real_type; + u32 intdata; + s32 tid; + + real_type = btf_type_skip_modifiers(btf, type->type, &tid); + if (!real_type) + return false; + + if (BTF_INFO_KIND(real_type->info) != BTF_KIND_INT) + return false; + + intdata = btf_type_int(real_type); + return !(BTF_INT_ENCODING(intdata) & BTF_INT_SIGNED) + && BTF_INT_BITS(intdata) == 8; +} + +static bool btf_type_is_char_array(struct btf *btf, const struct btf_type *type) +{ + const struct btf_type *real_type; + const struct btf_array *array; + u32 intdata; + s32 tid; + + if (BTF_INFO_KIND(type->info) != BTF_KIND_ARRAY) + return false; + + array = (const struct btf_array *)(type + 1); + + real_type = btf_type_skip_modifiers(btf, array->type, &tid); + + intdata = btf_type_int(real_type); + return !(BTF_INT_ENCODING(intdata) & BTF_INT_SIGNED) + && BTF_INT_BITS(intdata) == 8; +} + +static int check_prepare_btf_string_fetch(char *typename, + struct fetch_insn **pcode, + struct traceprobe_parse_context *ctx) +{ + struct btf *btf = ctx->btf; + + if (!btf || !ctx->last_type) + return 0; + + /* char [] does not need any change. */ + if (btf_type_is_char_array(btf, ctx->last_type)) + return 0; + + /* char * requires dereference the pointer. */ + if (btf_type_is_char_ptr(btf, ctx->last_type)) { + struct fetch_insn *code = *pcode + 1; + + if (code->op == FETCH_OP_END) { + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); + return -E2BIG; + } + if (typename[0] == 'u') + code->op = FETCH_OP_UDEREF; + else + code->op = FETCH_OP_DEREF; + code->offset = 0; + *pcode = code; + return 0; + } + /* Other types are not available for string */ + trace_probe_log_err(ctx->offset, BAD_TYPE4STR); + return -EINVAL; +} + +static const char *fetch_type_from_btf_type(struct btf *btf, + const struct btf_type *type, + struct traceprobe_parse_context *ctx) +{ + u32 intdata; + + /* TODO: const char * could be converted as a string */ + switch (BTF_INFO_KIND(type->info)) { + case BTF_KIND_ENUM: + /* enum is "int", so convert to "s32" */ + return "s32"; + case BTF_KIND_ENUM64: + return "s64"; + case BTF_KIND_PTR: + /* pointer will be converted to "x??" */ + if (IS_ENABLED(CONFIG_64BIT)) + return "x64"; + else + return "x32"; + case BTF_KIND_INT: + intdata = btf_type_int(type); + if (BTF_INT_ENCODING(intdata) & BTF_INT_SIGNED) { + switch (BTF_INT_BITS(intdata)) { + case 8: + return "s8"; + case 16: + return "s16"; + case 32: + return "s32"; + case 64: + return "s64"; + } + } else { /* unsigned */ + switch (BTF_INT_BITS(intdata)) { + case 8: + return "u8"; + case 16: + return "u16"; + case 32: + return "u32"; + case 64: + return "u64"; + } + /* bitfield, size is encoded in the type */ + ctx->last_bitsize = BTF_INT_BITS(intdata); + ctx->last_bitoffs += BTF_INT_OFFSET(intdata); + return "u64"; + } + } + /* TODO: support other types */ + + return NULL; +} + +static int query_btf_context(struct traceprobe_parse_context *ctx) +{ + const struct btf_param *param; + const struct btf_type *type; + struct btf *btf; + s32 nr; + + if (ctx->btf) + return 0; + + if (!ctx->funcname) + return -EINVAL; + + type = btf_find_func_proto(ctx->funcname, &btf); + if (!type) + return -ENOENT; + + ctx->btf = btf; + ctx->proto = type; + + /* ctx->params is optional, since func(void) will not have params. */ + nr = 0; + param = btf_get_func_param(type, &nr); + if (!IS_ERR_OR_NULL(param)) { + /* Hide the first 'data' argument of tracepoint */ + if (ctx->flags & TPARG_FL_TPOINT) { + nr--; + param++; + } + } + + if (nr > 0) { + ctx->nr_params = nr; + ctx->params = param; + } else { + ctx->nr_params = 0; + ctx->params = NULL; + } + + return 0; +} + +static void clear_btf_context(struct traceprobe_parse_context *ctx) +{ + if (ctx->btf) { + btf_put(ctx->btf); + ctx->btf = NULL; + ctx->proto = NULL; + ctx->params = NULL; + ctx->nr_params = 0; + } +} + +/* Return 1 if the field separator is arrow operator ('->') */ +static int split_next_field(char *varname, char **next_field, + struct traceprobe_parse_context *ctx) +{ + char *field; + int ret = 0; + + field = strpbrk(varname, ".-"); + if (field) { + if (field[0] == '-' && field[1] == '>') { + field[0] = '\0'; + field += 2; + ret = 1; + } else if (field[0] == '.') { + field[0] = '\0'; + field += 1; + } else { + trace_probe_log_err(ctx->offset + field - varname, BAD_HYPHEN); + return -EINVAL; + } + *next_field = field; + } + + return ret; +} + +/* + * Parse the field of data structure. The @type must be a pointer type + * pointing the target data structure type. + */ +static int parse_btf_field(char *fieldname, const struct btf_type *type, + struct fetch_insn **pcode, struct fetch_insn *end, + struct traceprobe_parse_context *ctx) +{ + struct fetch_insn *code = *pcode; + const struct btf_member *field; + u32 bitoffs, anon_offs; + char *next; + int is_ptr; + s32 tid; + + do { + /* Outer loop for solving arrow operator ('->') */ + if (BTF_INFO_KIND(type->info) != BTF_KIND_PTR) { + trace_probe_log_err(ctx->offset, NO_PTR_STRCT); + return -EINVAL; + } + /* Convert a struct pointer type to a struct type */ + type = btf_type_skip_modifiers(ctx->btf, type->type, &tid); + if (!type) { + trace_probe_log_err(ctx->offset, BAD_BTF_TID); + return -EINVAL; + } + + bitoffs = 0; + do { + /* Inner loop for solving dot operator ('.') */ + next = NULL; + is_ptr = split_next_field(fieldname, &next, ctx); + if (is_ptr < 0) + return is_ptr; + + anon_offs = 0; + field = btf_find_struct_member(ctx->btf, type, fieldname, + &anon_offs); + if (IS_ERR(field)) { + trace_probe_log_err(ctx->offset, BAD_BTF_TID); + return PTR_ERR(field); + } + if (!field) { + trace_probe_log_err(ctx->offset, NO_BTF_FIELD); + return -ENOENT; + } + /* Add anonymous structure/union offset */ + bitoffs += anon_offs; + + /* Accumulate the bit-offsets of the dot-connected fields */ + if (btf_type_kflag(type)) { + bitoffs += BTF_MEMBER_BIT_OFFSET(field->offset); + ctx->last_bitsize = BTF_MEMBER_BITFIELD_SIZE(field->offset); + } else { + bitoffs += field->offset; + ctx->last_bitsize = 0; + } + + type = btf_type_skip_modifiers(ctx->btf, field->type, &tid); + if (!type) { + trace_probe_log_err(ctx->offset, BAD_BTF_TID); + return -EINVAL; + } + + ctx->offset += next - fieldname; + fieldname = next; + } while (!is_ptr && fieldname); + + if (++code == end) { + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); + return -EINVAL; + } + code->op = FETCH_OP_DEREF; /* TODO: user deref support */ + code->offset = bitoffs / 8; + *pcode = code; + + ctx->last_bitoffs = bitoffs % 8; + ctx->last_type = type; + } while (fieldname); + + return 0; +} + +static int __store_entry_arg(struct trace_probe *tp, int argnum); + +static int parse_btf_arg(char *varname, + struct fetch_insn **pcode, struct fetch_insn *end, + struct traceprobe_parse_context *ctx) +{ + struct fetch_insn *code = *pcode; + const struct btf_param *params; + const struct btf_type *type; + char *field = NULL; + int i, is_ptr, ret; + u32 tid; + + if (WARN_ON_ONCE(!ctx->funcname)) + return -EINVAL; + + is_ptr = split_next_field(varname, &field, ctx); + if (is_ptr < 0) + return is_ptr; + if (!is_ptr && field) { + /* dot-connected field on an argument is not supported. */ + trace_probe_log_err(ctx->offset + field - varname, + NOSUP_DAT_ARG); + return -EOPNOTSUPP; + } + + if (ctx->flags & TPARG_FL_RETURN && !strcmp(varname, "$retval")) { + code->op = FETCH_OP_RETVAL; + /* Check whether the function return type is not void */ + if (query_btf_context(ctx) == 0) { + if (ctx->proto->type == 0) { + trace_probe_log_err(ctx->offset, NO_RETVAL); + return -ENOENT; + } + tid = ctx->proto->type; + goto found; + } + if (field) { + trace_probe_log_err(ctx->offset + field - varname, + NO_BTF_ENTRY); + return -ENOENT; + } + return 0; + } + + if (!ctx->btf) { + ret = query_btf_context(ctx); + if (ret < 0 || ctx->nr_params == 0) { + trace_probe_log_err(ctx->offset, NO_BTF_ENTRY); + return -ENOENT; + } + } + params = ctx->params; + + for (i = 0; i < ctx->nr_params; i++) { + const char *name = btf_name_by_offset(ctx->btf, params[i].name_off); + + if (name && !strcmp(name, varname)) { + if (tparg_is_function_entry(ctx->flags)) { + code->op = FETCH_OP_ARG; + if (ctx->flags & TPARG_FL_TPOINT) + code->param = i + 1; + else + code->param = i; + } else if (tparg_is_function_return(ctx->flags)) { + code->op = FETCH_OP_EDATA; + ret = __store_entry_arg(ctx->tp, i); + if (ret < 0) { + /* internal error */ + return ret; + } + code->offset = ret; + } + tid = params[i].type; + goto found; + } + } + trace_probe_log_err(ctx->offset, NO_BTFARG); + return -ENOENT; + +found: + type = btf_type_skip_modifiers(ctx->btf, tid, &tid); + if (!type) { + trace_probe_log_err(ctx->offset, BAD_BTF_TID); + return -EINVAL; + } + /* Initialize the last type information */ + ctx->last_type = type; + ctx->last_bitoffs = 0; + ctx->last_bitsize = 0; + if (field) { + ctx->offset += field - varname; + return parse_btf_field(field, type, pcode, end, ctx); + } + return 0; +} + +static const struct fetch_type *find_fetch_type_from_btf_type( + struct traceprobe_parse_context *ctx) +{ + struct btf *btf = ctx->btf; + const char *typestr = NULL; + + if (btf && ctx->last_type) + typestr = fetch_type_from_btf_type(btf, ctx->last_type, ctx); + + return find_fetch_type(typestr, ctx->flags); +} + +static int parse_btf_bitfield(struct fetch_insn **pcode, + struct traceprobe_parse_context *ctx) +{ + struct fetch_insn *code = *pcode; + + if ((ctx->last_bitsize % 8 == 0) && ctx->last_bitoffs == 0) + return 0; + + code++; + if (code->op != FETCH_OP_NOP) { + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); + return -EINVAL; + } + *pcode = code; + + code->op = FETCH_OP_MOD_BF; + code->lshift = 64 - (ctx->last_bitsize + ctx->last_bitoffs); + code->rshift = 64 - ctx->last_bitsize; + code->basesize = 64 / 8; + return 0; +} + +#else +static void clear_btf_context(struct traceprobe_parse_context *ctx) +{ + ctx->btf = NULL; +} + +static int query_btf_context(struct traceprobe_parse_context *ctx) +{ + return -EOPNOTSUPP; +} + +static int parse_btf_arg(char *varname, + struct fetch_insn **pcode, struct fetch_insn *end, + struct traceprobe_parse_context *ctx) +{ + trace_probe_log_err(ctx->offset, NOSUP_BTFARG); + return -EOPNOTSUPP; +} + +static int parse_btf_bitfield(struct fetch_insn **pcode, + struct traceprobe_parse_context *ctx) +{ + trace_probe_log_err(ctx->offset, NOSUP_BTFARG); + return -EOPNOTSUPP; +} + +#define find_fetch_type_from_btf_type(ctx) \ + find_fetch_type(NULL, ctx->flags) + +static int check_prepare_btf_string_fetch(char *typename, + struct fetch_insn **pcode, + struct traceprobe_parse_context *ctx) +{ + return 0; +} + +#endif + +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + +static void store_entry_arg_at(struct fetch_insn *code, int argnum, int offset) +{ + code[0].op = FETCH_OP_ARG; + code[0].param = argnum; + code[1].op = FETCH_OP_ST_EDATA; + code[1].offset = offset; +} + +static int get_entry_arg_max_offset(struct probe_entry_arg *earg) +{ + int i, max_offset = 0; + + /* + * earg->code[] array has an operation sequence which is run in + * the entry handler. + * The sequence stopped by FETCH_OP_END and each data stored in + * the entry data buffer by FETCH_OP_ST_EDATA. The FETCH_OP_ST_EDATA + * stores the data at the data buffer + its offset, and all data are + * "unsigned long" size. The offset must be increased when a data is + * stored. Thus we need to find the last FETCH_OP_ST_EDATA in the + * code array. + */ + for (i = 0; i < earg->size - 1 && earg->code[i].op != FETCH_OP_END; i++) { + if (earg->code[i].op == FETCH_OP_ST_EDATA) + if (earg->code[i].offset > max_offset) + max_offset = earg->code[i].offset; + } + return max_offset; +} + +/* + * Add the entry code to store the 'argnum'th parameter and return the offset + * in the entry data buffer where the data will be stored. + */ +static int __store_entry_arg(struct trace_probe *tp, int argnum) +{ + struct probe_entry_arg *earg = tp->entry_arg; + int i, offset, last_offset = 0; + + if (!earg) { + earg = kzalloc(sizeof(*tp->entry_arg), GFP_KERNEL); + if (!earg) + return -ENOMEM; + earg->size = 2 * tp->nr_args + 1; + earg->code = kcalloc(earg->size, sizeof(struct fetch_insn), + GFP_KERNEL); + if (!earg->code) { + kfree(earg); + return -ENOMEM; + } + /* Fill the code buffer with 'end' to simplify it */ + for (i = 0; i < earg->size; i++) + earg->code[i].op = FETCH_OP_END; + tp->entry_arg = earg; + store_entry_arg_at(earg->code, argnum, 0); + return 0; + } + + /* + * NOTE: if anyone change the following rule, please rewrite this. + * The entry code array is filled with the pair of + * + * [FETCH_OP_ARG(argnum)] + * [FETCH_OP_ST_EDATA(offset of entry data buffer)] + * + * and the rest of entries are filled with [FETCH_OP_END]. + * The offset should be incremented, thus the last pair should + * have the largest offset. + */ + + /* Search the offset for the sprcified argnum. */ + for (i = 0; i < earg->size - 1 && earg->code[i].op != FETCH_OP_END; i += 2) { + if (WARN_ON_ONCE(earg->code[i].op != FETCH_OP_ARG)) + return -EINVAL; + + if (earg->code[i].param != argnum) + continue; + + if (WARN_ON_ONCE(earg->code[i + 1].op != FETCH_OP_ST_EDATA)) + return -EINVAL; + + return earg->code[i + 1].offset; + } + /* Not found, append new entry if possible. */ + if (i >= earg->size - 1) + return -ENOSPC; + + /* The last entry must have the largest offset. */ + if (i != 0) { + if (WARN_ON_ONCE(earg->code[i - 1].op != FETCH_OP_ST_EDATA)) + return -EINVAL; + last_offset = earg->code[i - 1].offset; + } + + offset = last_offset + sizeof(unsigned long); + store_entry_arg_at(&earg->code[i], argnum, offset); + return offset; +} + +int traceprobe_get_entry_data_size(struct trace_probe *tp) +{ + struct probe_entry_arg *earg = tp->entry_arg; + + if (!earg) + return 0; + + return get_entry_arg_max_offset(earg) + sizeof(unsigned long); +} + +void store_trace_entry_data(void *edata, struct trace_probe *tp, struct pt_regs *regs) +{ + struct probe_entry_arg *earg = tp->entry_arg; + unsigned long val = 0; + int i; + + if (!earg) + return; + + for (i = 0; i < earg->size; i++) { + struct fetch_insn *code = &earg->code[i]; + + switch (code->op) { + case FETCH_OP_ARG: + val = regs_get_kernel_argument(regs, code->param); + break; + case FETCH_OP_ST_EDATA: + *(unsigned long *)((unsigned long)edata + code->offset) = val; + break; + case FETCH_OP_END: + goto end; + default: + break; + } + } +end: + return; +} +NOKPROBE_SYMBOL(store_trace_entry_data) +#endif + #define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long)) -static int parse_probe_vars(char *arg, const struct fetch_type *t, - struct fetch_insn *code, unsigned int flags, int offs) +/* Parse $vars. @orig_arg points '$', which syncs to @ctx->offset */ +static int parse_probe_vars(char *orig_arg, const struct fetch_type *t, + struct fetch_insn **pcode, + struct fetch_insn *end, + struct traceprobe_parse_context *ctx) { + struct fetch_insn *code = *pcode; + int err = TP_ERR_BAD_VAR; + char *arg = orig_arg + 1; unsigned long param; int ret = 0; int len; - if (flags & TPARG_FL_TPOINT) { + if (ctx->flags & TPARG_FL_TEVENT) { if (code->data) return -EFAULT; - code->data = kstrdup(arg, GFP_KERNEL); - if (!code->data) - return -ENOMEM; - code->op = FETCH_OP_TP_ARG; - } else if (strcmp(arg, "retval") == 0) { - if (flags & TPARG_FL_RETURN) { + ret = parse_trace_event_arg(arg, code, ctx); + if (!ret) + return 0; + if (strcmp(arg, "comm") == 0 || strcmp(arg, "COMM") == 0) { + code->op = FETCH_OP_COMM; + return 0; + } + /* backward compatibility */ + ctx->offset = 0; + goto inval; + } + + if (str_has_prefix(arg, "retval")) { + if (!(ctx->flags & TPARG_FL_RETURN)) { + err = TP_ERR_RETVAL_ON_PROBE; + goto inval; + } + if (!(ctx->flags & TPARG_FL_KERNEL) || + !IS_ENABLED(CONFIG_PROBE_EVENTS_BTF_ARGS)) { code->op = FETCH_OP_RETVAL; - } else { - trace_probe_log_err(offs, RETVAL_ON_PROBE); - ret = -EINVAL; + return 0; } - } else if ((len = str_has_prefix(arg, "stack"))) { + return parse_btf_arg(orig_arg, pcode, end, ctx); + } + + len = str_has_prefix(arg, "stack"); + if (len) { + if (arg[len] == '\0') { code->op = FETCH_OP_STACKP; - } else if (isdigit(arg[len])) { + return 0; + } + + if (isdigit(arg[len])) { ret = kstrtoul(arg + len, 10, ¶m); - if (ret) { - goto inval_var; - } else if ((flags & TPARG_FL_KERNEL) && - param > PARAM_MAX_STACK) { - trace_probe_log_err(offs, BAD_STACK_NUM); - ret = -EINVAL; - } else { - code->op = FETCH_OP_STACK; - code->param = (unsigned int)param; + if (ret) + goto inval; + + if ((ctx->flags & TPARG_FL_KERNEL) && + param > PARAM_MAX_STACK) { + err = TP_ERR_BAD_STACK_NUM; + goto inval; } - } else - goto inval_var; - } else if (strcmp(arg, "comm") == 0 || strcmp(arg, "COMM") == 0) { + code->op = FETCH_OP_STACK; + code->param = (unsigned int)param; + return 0; + } + goto inval; + } + + if (strcmp(arg, "comm") == 0 || strcmp(arg, "COMM") == 0) { code->op = FETCH_OP_COMM; + return 0; + } + #ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API - } else if (((flags & TPARG_FL_MASK) == - (TPARG_FL_KERNEL | TPARG_FL_FENTRY)) && - (len = str_has_prefix(arg, "arg"))) { + len = str_has_prefix(arg, "arg"); + if (len) { ret = kstrtoul(arg + len, 10, ¶m); - if (ret) { - goto inval_var; - } else if (!param || param > PARAM_MAX_STACK) { - trace_probe_log_err(offs, BAD_ARG_NUM); - return -EINVAL; + if (ret) + goto inval; + + if (!param || param > PARAM_MAX_STACK) { + err = TP_ERR_BAD_ARG_NUM; + goto inval; } - code->op = FETCH_OP_ARG; - code->param = (unsigned int)param - 1; -#endif - } else - goto inval_var; + param--; /* argN starts from 1, but internal arg[N] starts from 0 */ - return ret; + if (tparg_is_function_entry(ctx->flags)) { + code->op = FETCH_OP_ARG; + code->param = (unsigned int)param; + /* + * The tracepoint probe will probe a stub function, and the + * first parameter of the stub is a dummy and should be ignored. + */ + if (ctx->flags & TPARG_FL_TPOINT) + code->param++; + } else if (tparg_is_function_return(ctx->flags)) { + /* function entry argument access from return probe */ + ret = __store_entry_arg(ctx->tp, param); + if (ret < 0) /* This error should be an internal error */ + return ret; + + code->op = FETCH_OP_EDATA; + code->offset = ret; + } else { + err = TP_ERR_NOFENTRY_ARGS; + goto inval; + } + return 0; + } +#endif -inval_var: - trace_probe_log_err(offs, BAD_VAR); +inval: + __trace_probe_log_err(ctx->offset, err); return -EINVAL; } @@ -376,7 +1083,7 @@ static int __parse_imm_string(char *str, char **pbuf, int offs) static int parse_probe_arg(char *arg, const struct fetch_type *type, struct fetch_insn **pcode, struct fetch_insn *end, - unsigned int flags, int offs) + struct traceprobe_parse_context *ctx) { struct fetch_insn *code = *pcode; unsigned long param; @@ -387,13 +1094,13 @@ parse_probe_arg(char *arg, const struct fetch_type *type, switch (arg[0]) { case '$': - ret = parse_probe_vars(arg + 1, type, code, flags, offs); + ret = parse_probe_vars(arg, type, pcode, end, ctx); break; case '%': /* named register */ - if (flags & TPARG_FL_TPOINT) { - /* eprobes do not handle registers */ - trace_probe_log_err(offs, BAD_VAR); + if (ctx->flags & (TPARG_FL_TEVENT | TPARG_FL_FPROBE)) { + /* eprobe and fprobe do not handle registers */ + trace_probe_log_err(ctx->offset, BAD_VAR); break; } ret = regs_query_register_offset(arg + 1); @@ -402,14 +1109,14 @@ parse_probe_arg(char *arg, const struct fetch_type *type, code->param = (unsigned int)ret; ret = 0; } else - trace_probe_log_err(offs, BAD_REG_NAME); + trace_probe_log_err(ctx->offset, BAD_REG_NAME); break; case '@': /* memory, file-offset or symbol */ if (isdigit(arg[1])) { ret = kstrtoul(arg + 1, 0, ¶m); if (ret) { - trace_probe_log_err(offs, BAD_MEM_ADDR); + trace_probe_log_err(ctx->offset, BAD_MEM_ADDR); break; } /* load address */ @@ -417,13 +1124,13 @@ parse_probe_arg(char *arg, const struct fetch_type *type, code->immediate = param; } else if (arg[1] == '+') { /* kprobes don't support file offsets */ - if (flags & TPARG_FL_KERNEL) { - trace_probe_log_err(offs, FILE_ON_KPROBE); + if (ctx->flags & TPARG_FL_KERNEL) { + trace_probe_log_err(ctx->offset, FILE_ON_KPROBE); return -EINVAL; } ret = kstrtol(arg + 2, 0, &offset); if (ret) { - trace_probe_log_err(offs, BAD_FILE_OFFS); + trace_probe_log_err(ctx->offset, BAD_FILE_OFFS); break; } @@ -431,8 +1138,8 @@ parse_probe_arg(char *arg, const struct fetch_type *type, code->immediate = (unsigned long)offset; // imm64? } else { /* uprobes don't support symbols */ - if (!(flags & TPARG_FL_KERNEL)) { - trace_probe_log_err(offs, SYM_ON_UPROBE); + if (!(ctx->flags & TPARG_FL_KERNEL)) { + trace_probe_log_err(ctx->offset, SYM_ON_UPROBE); return -EINVAL; } /* Preserve symbol for updating */ @@ -441,7 +1148,7 @@ parse_probe_arg(char *arg, const struct fetch_type *type, if (!code->data) return -ENOMEM; if (++code == end) { - trace_probe_log_err(offs, TOO_MANY_OPS); + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); return -EINVAL; } code->op = FETCH_OP_IMM; @@ -449,7 +1156,7 @@ parse_probe_arg(char *arg, const struct fetch_type *type, } /* These are fetching from memory */ if (++code == end) { - trace_probe_log_err(offs, TOO_MANY_OPS); + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); return -EINVAL; } *pcode = code; @@ -468,47 +1175,51 @@ parse_probe_arg(char *arg, const struct fetch_type *type, arg++; /* Skip '+', because kstrtol() rejects it. */ tmp = strchr(arg, '('); if (!tmp) { - trace_probe_log_err(offs, DEREF_NEED_BRACE); + trace_probe_log_err(ctx->offset, DEREF_NEED_BRACE); return -EINVAL; } *tmp = '\0'; ret = kstrtol(arg, 0, &offset); if (ret) { - trace_probe_log_err(offs, BAD_DEREF_OFFS); + trace_probe_log_err(ctx->offset, BAD_DEREF_OFFS); break; } - offs += (tmp + 1 - arg) + (arg[0] != '-' ? 1 : 0); + ctx->offset += (tmp + 1 - arg) + (arg[0] != '-' ? 1 : 0); arg = tmp + 1; tmp = strrchr(arg, ')'); if (!tmp) { - trace_probe_log_err(offs + strlen(arg), + trace_probe_log_err(ctx->offset + strlen(arg), DEREF_OPEN_BRACE); return -EINVAL; } else { - const struct fetch_type *t2 = find_fetch_type(NULL, flags); + const struct fetch_type *t2 = find_fetch_type(NULL, ctx->flags); + int cur_offs = ctx->offset; *tmp = '\0'; - ret = parse_probe_arg(arg, t2, &code, end, flags, offs); + ret = parse_probe_arg(arg, t2, &code, end, ctx); if (ret) break; + ctx->offset = cur_offs; if (code->op == FETCH_OP_COMM || code->op == FETCH_OP_DATA) { - trace_probe_log_err(offs, COMM_CANT_DEREF); + trace_probe_log_err(ctx->offset, COMM_CANT_DEREF); return -EINVAL; } if (++code == end) { - trace_probe_log_err(offs, TOO_MANY_OPS); + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); return -EINVAL; } *pcode = code; code->op = deref; code->offset = offset; + /* Reset the last type if used */ + ctx->last_type = NULL; } break; case '\\': /* Immediate value */ if (arg[1] == '"') { /* Immediate string */ - ret = __parse_imm_string(arg + 2, &tmp, offs + 2); + ret = __parse_imm_string(arg + 2, &tmp, ctx->offset + 2); if (ret) break; code->op = FETCH_OP_DATA; @@ -516,22 +1227,30 @@ parse_probe_arg(char *arg, const struct fetch_type *type, } else { ret = str_to_immediate(arg + 1, &code->immediate); if (ret) - trace_probe_log_err(offs + 1, BAD_IMM); + trace_probe_log_err(ctx->offset + 1, BAD_IMM); else code->op = FETCH_OP_IMM; } break; + default: + if (isalpha(arg[0]) || arg[0] == '_') { /* BTF variable */ + if (!tparg_is_function_entry(ctx->flags) && + !tparg_is_function_return(ctx->flags)) { + trace_probe_log_err(ctx->offset, NOSUP_BTFARG); + return -EINVAL; + } + ret = parse_btf_arg(arg, pcode, end, ctx); + break; + } } if (!ret && code->op == FETCH_OP_NOP) { /* Parsed, but do not find fetch method */ - trace_probe_log_err(offs, BAD_FETCH_ARG); + trace_probe_log_err(ctx->offset, BAD_FETCH_ARG); ret = -EINVAL; } return ret; } -#define BYTES_TO_BITS(nb) ((BITS_PER_LONG * (nb)) / sizeof(long)) - /* Bitfield type needs to be parsed into a fetch function */ static int __parse_bitfield_probe_arg(const char *bf, const struct fetch_type *t, @@ -567,126 +1286,101 @@ static int __parse_bitfield_probe_arg(const char *bf, return (BYTES_TO_BITS(t->size) < (bw + bo)) ? -EINVAL : 0; } -/* String length checking wrapper */ -static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size, - struct probe_arg *parg, unsigned int flags, int offset) +/* Split type part from @arg and return it. */ +static char *parse_probe_arg_type(char *arg, struct probe_arg *parg, + struct traceprobe_parse_context *ctx) { - struct fetch_insn *code, *scode, *tmp = NULL; - char *t, *t2, *t3; - char *arg; - int ret, len; + char *t = NULL, *t2, *t3; + int offs; - arg = kstrdup(argv, GFP_KERNEL); - if (!arg) - return -ENOMEM; - - ret = -EINVAL; - len = strlen(arg); - if (len > MAX_ARGSTR_LEN) { - trace_probe_log_err(offset, ARG_TOO_LONG); - goto out; - } else if (len == 0) { - trace_probe_log_err(offset, NO_ARG_BODY); - goto out; - } - - ret = -ENOMEM; - parg->comm = kstrdup(arg, GFP_KERNEL); - if (!parg->comm) - goto out; - - ret = -EINVAL; t = strchr(arg, ':'); if (t) { - *t = '\0'; - t2 = strchr(++t, '['); + *t++ = '\0'; + t2 = strchr(t, '['); if (t2) { *t2++ = '\0'; t3 = strchr(t2, ']'); if (!t3) { - offset += t2 + strlen(t2) - arg; - trace_probe_log_err(offset, + offs = t2 + strlen(t2) - arg; + + trace_probe_log_err(ctx->offset + offs, ARRAY_NO_CLOSE); - goto out; + return ERR_PTR(-EINVAL); } else if (t3[1] != '\0') { - trace_probe_log_err(offset + t3 + 1 - arg, + trace_probe_log_err(ctx->offset + t3 + 1 - arg, BAD_ARRAY_SUFFIX); - goto out; + return ERR_PTR(-EINVAL); } *t3 = '\0'; if (kstrtouint(t2, 0, &parg->count) || !parg->count) { - trace_probe_log_err(offset + t2 - arg, + trace_probe_log_err(ctx->offset + t2 - arg, BAD_ARRAY_NUM); - goto out; + return ERR_PTR(-EINVAL); } if (parg->count > MAX_ARRAY_LEN) { - trace_probe_log_err(offset + t2 - arg, + trace_probe_log_err(ctx->offset + t2 - arg, ARRAY_TOO_BIG); - goto out; + return ERR_PTR(-EINVAL); } } } + offs = t ? t - arg : 0; /* * Since $comm and immediate string can not be dereferenced, * we can find those by strcmp. But ignore for eprobes. */ - if (!(flags & TPARG_FL_TPOINT) && + if (!(ctx->flags & TPARG_FL_TEVENT) && (strcmp(arg, "$comm") == 0 || strcmp(arg, "$COMM") == 0 || strncmp(arg, "\\\"", 2) == 0)) { - /* The type of $comm must be "string", and not an array. */ - if (parg->count || (t && strcmp(t, "string"))) - goto out; - parg->type = find_fetch_type("string", flags); + /* The type of $comm must be "string", and not an array type. */ + if (parg->count || (t && strcmp(t, "string"))) { + trace_probe_log_err(ctx->offset + offs, NEED_STRING_TYPE); + return ERR_PTR(-EINVAL); + } + parg->type = find_fetch_type("string", ctx->flags); } else - parg->type = find_fetch_type(t, flags); - if (!parg->type) { - trace_probe_log_err(offset + (t ? (t - arg) : 0), BAD_TYPE); - goto out; - } - parg->offset = *size; - *size += parg->type->size * (parg->count ?: 1); + parg->type = find_fetch_type(t, ctx->flags); - ret = -ENOMEM; - if (parg->count) { - len = strlen(parg->type->fmttype) + 6; - parg->fmt = kmalloc(len, GFP_KERNEL); - if (!parg->fmt) - goto out; - snprintf(parg->fmt, len, "%s[%d]", parg->type->fmttype, - parg->count); + if (!parg->type) { + trace_probe_log_err(ctx->offset + offs, BAD_TYPE); + return ERR_PTR(-EINVAL); } - code = tmp = kcalloc(FETCH_INSN_MAX, sizeof(*code), GFP_KERNEL); - if (!code) - goto out; - code[FETCH_INSN_MAX - 1].op = FETCH_OP_END; + return t; +} - ret = parse_probe_arg(arg, parg->type, &code, &code[FETCH_INSN_MAX - 1], - flags, offset); - if (ret) - goto fail; +/* After parsing, adjust the fetch_insn according to the probe_arg */ +static int finalize_fetch_insn(struct fetch_insn *code, + struct probe_arg *parg, + char *type, + int type_offset, + struct traceprobe_parse_context *ctx) +{ + struct fetch_insn *scode; + int ret; - ret = -EINVAL; /* Store operation */ if (parg->type->is_string) { + /* Check bad combination of the type and the last fetch_insn. */ if (!strcmp(parg->type->name, "symstr")) { if (code->op != FETCH_OP_REG && code->op != FETCH_OP_STACK && code->op != FETCH_OP_RETVAL && code->op != FETCH_OP_ARG && code->op != FETCH_OP_DEREF && code->op != FETCH_OP_TP_ARG) { - trace_probe_log_err(offset + (t ? (t - arg) : 0), + trace_probe_log_err(ctx->offset + type_offset, BAD_SYMSTRING); - goto fail; + return -EINVAL; } } else { if (code->op != FETCH_OP_DEREF && code->op != FETCH_OP_UDEREF && code->op != FETCH_OP_IMM && code->op != FETCH_OP_COMM && code->op != FETCH_OP_DATA && code->op != FETCH_OP_TP_ARG) { - trace_probe_log_err(offset + (t ? (t - arg) : 0), + trace_probe_log_err(ctx->offset + type_offset, BAD_STRING); - goto fail; + return -EINVAL; } } + if (!strcmp(parg->type->name, "symstr") || (code->op == FETCH_OP_IMM || code->op == FETCH_OP_COMM || code->op == FETCH_OP_DATA) || code->op == FETCH_OP_TP_ARG || @@ -701,10 +1395,11 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size, */ code++; if (code->op != FETCH_OP_NOP) { - trace_probe_log_err(offset, TOO_MANY_OPS); - goto fail; + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); + return -EINVAL; } } + /* If op == DEREF, replace it with STRING */ if (!strcmp(parg->type->name, "ustring") || code->op == FETCH_OP_UDEREF) @@ -724,43 +1419,129 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size, } else { code++; if (code->op != FETCH_OP_NOP) { - trace_probe_log_err(offset, TOO_MANY_OPS); - goto fail; + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); + return -E2BIG; } code->op = FETCH_OP_ST_RAW; code->size = parg->type->size; } + + /* Save storing fetch_insn. */ scode = code; + /* Modify operation */ - if (t != NULL) { - ret = __parse_bitfield_probe_arg(t, parg->type, &code); + if (type != NULL) { + /* Bitfield needs a special fetch_insn. */ + ret = __parse_bitfield_probe_arg(type, parg->type, &code); if (ret) { - trace_probe_log_err(offset + t - arg, BAD_BITFIELD); - goto fail; + trace_probe_log_err(ctx->offset + type_offset, BAD_BITFIELD); + return ret; } + } else if (IS_ENABLED(CONFIG_PROBE_EVENTS_BTF_ARGS) && + ctx->last_type) { + /* If user not specified the type, try parsing BTF bitfield. */ + ret = parse_btf_bitfield(&code, ctx); + if (ret) + return ret; } - ret = -EINVAL; + /* Loop(Array) operation */ if (parg->count) { if (scode->op != FETCH_OP_ST_MEM && scode->op != FETCH_OP_ST_STRING && scode->op != FETCH_OP_ST_USTRING) { - trace_probe_log_err(offset + (t ? (t - arg) : 0), - BAD_STRING); - goto fail; + trace_probe_log_err(ctx->offset + type_offset, BAD_STRING); + return -EINVAL; } code++; if (code->op != FETCH_OP_NOP) { - trace_probe_log_err(offset, TOO_MANY_OPS); - goto fail; + trace_probe_log_err(ctx->offset, TOO_MANY_OPS); + return -E2BIG; } code->op = FETCH_OP_LP_ARRAY; code->param = parg->count; } + + /* Finalize the fetch_insn array. */ code++; code->op = FETCH_OP_END; - ret = 0; + return 0; +} + +/* String length checking wrapper */ +static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size, + struct probe_arg *parg, + struct traceprobe_parse_context *ctx) +{ + struct fetch_insn *code, *tmp = NULL; + char *type, *arg __free(kfree) = NULL; + int ret, len; + + len = strlen(argv); + if (len > MAX_ARGSTR_LEN) { + trace_probe_log_err(ctx->offset, ARG_TOO_LONG); + return -E2BIG; + } else if (len == 0) { + trace_probe_log_err(ctx->offset, NO_ARG_BODY); + return -EINVAL; + } + + arg = kstrdup(argv, GFP_KERNEL); + if (!arg) + return -ENOMEM; + + parg->comm = kstrdup(arg, GFP_KERNEL); + if (!parg->comm) + return -ENOMEM; + + type = parse_probe_arg_type(arg, parg, ctx); + if (IS_ERR(type)) + return PTR_ERR(type); + + code = tmp = kcalloc(FETCH_INSN_MAX, sizeof(*code), GFP_KERNEL); + if (!code) + return -ENOMEM; + code[FETCH_INSN_MAX - 1].op = FETCH_OP_END; + + ctx->last_type = NULL; + ret = parse_probe_arg(arg, parg->type, &code, &code[FETCH_INSN_MAX - 1], + ctx); + if (ret < 0) + goto fail; + + /* Update storing type if BTF is available */ + if (IS_ENABLED(CONFIG_PROBE_EVENTS_BTF_ARGS) && + ctx->last_type) { + if (!type) { + parg->type = find_fetch_type_from_btf_type(ctx); + } else if (strstr(type, "string")) { + ret = check_prepare_btf_string_fetch(type, &code, ctx); + if (ret) + goto fail; + } + } + parg->offset = *size; + *size += parg->type->size * (parg->count ?: 1); + + if (parg->count) { + len = strlen(parg->type->fmttype) + 6; + parg->fmt = kmalloc(len, GFP_KERNEL); + if (!parg->fmt) { + ret = -ENOMEM; + goto fail; + } + snprintf(parg->fmt, len, "%s[%d]", parg->type->fmttype, + parg->count); + } + + ret = finalize_fetch_insn(code, parg, type, type ? type - arg : 0, ctx); + if (ret < 0) + goto fail; + + for (; code < tmp + FETCH_INSN_MAX; code++) + if (code->op == FETCH_OP_END) + break; /* Shrink down the code buffer */ parg->code = kcalloc(code - tmp + 1, sizeof(*code), GFP_KERNEL); if (!parg->code) @@ -769,15 +1550,13 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size, memcpy(parg->code, tmp, sizeof(*code) * (code - tmp + 1)); fail: - if (ret) { + if (ret < 0) { for (code = tmp; code < tmp + FETCH_INSN_MAX; code++) if (code->op == FETCH_NOP_SYMBOL || code->op == FETCH_OP_DATA) kfree(code->data); } kfree(tmp); -out: - kfree(arg); return ret; } @@ -799,15 +1578,40 @@ static int traceprobe_conflict_field_name(const char *name, return 0; } +static char *generate_probe_arg_name(const char *arg, int idx) +{ + char *name = NULL; + const char *end; + + /* + * If argument name is omitted, try arg as a name (BTF variable) + * or "argN". + */ + if (IS_ENABLED(CONFIG_PROBE_EVENTS_BTF_ARGS)) { + end = strchr(arg, ':'); + if (!end) + end = arg + strlen(arg); + + name = kmemdup_nul(arg, end - arg, GFP_KERNEL); + if (!name || !is_good_name(name)) { + kfree(name); + name = NULL; + } + } + + if (!name) + name = kasprintf(GFP_KERNEL, "arg%d", idx + 1); + + return name; +} + int traceprobe_parse_probe_arg(struct trace_probe *tp, int i, const char *arg, - unsigned int flags) + struct traceprobe_parse_context *ctx) { struct probe_arg *parg = &tp->args[i]; const char *body; - /* Increment count for freeing args in error case */ - tp->nr_args++; - + ctx->tp = tp; body = strchr(arg, '='); if (body) { if (body - arg > MAX_ARG_NAME_LEN) { @@ -820,8 +1624,7 @@ int traceprobe_parse_probe_arg(struct trace_probe *tp, int i, const char *arg, parg->name = kmemdup_nul(arg, body - arg, GFP_KERNEL); body++; } else { - /* If argument name is omitted, set "argN" */ - parg->name = kasprintf(GFP_KERNEL, "arg%d", i + 1); + parg->name = generate_probe_arg_name(arg, i); body = arg; } if (!parg->name) @@ -835,9 +1638,9 @@ int traceprobe_parse_probe_arg(struct trace_probe *tp, int i, const char *arg, trace_probe_log_err(0, USED_ARG_NAME); return -EINVAL; } + ctx->offset = body - arg; /* Parse fetch argument */ - return traceprobe_parse_probe_arg_body(body, &tp->size, parg, flags, - body - arg); + return traceprobe_parse_probe_arg_body(body, &tp->size, parg, ctx); } void traceprobe_free_probe_arg(struct probe_arg *arg) @@ -856,6 +1659,205 @@ void traceprobe_free_probe_arg(struct probe_arg *arg) kfree(arg->fmt); } +static int argv_has_var_arg(int argc, const char *argv[], int *args_idx, + struct traceprobe_parse_context *ctx) +{ + int i, found = 0; + + for (i = 0; i < argc; i++) + if (str_has_prefix(argv[i], "$arg")) { + trace_probe_log_set_index(i + 2); + + if (!tparg_is_function_entry(ctx->flags) && + !tparg_is_function_return(ctx->flags)) { + trace_probe_log_err(0, NOFENTRY_ARGS); + return -EINVAL; + } + + if (isdigit(argv[i][4])) { + found = 1; + continue; + } + + if (argv[i][4] != '*') { + trace_probe_log_err(0, BAD_VAR); + return -EINVAL; + } + + if (*args_idx >= 0 && *args_idx < argc) { + trace_probe_log_err(0, DOUBLE_ARGS); + return -EINVAL; + } + found = 1; + *args_idx = i; + } + + return found; +} + +static int sprint_nth_btf_arg(int idx, const char *type, + char *buf, int bufsize, + struct traceprobe_parse_context *ctx) +{ + const char *name; + int ret; + + if (idx >= ctx->nr_params) { + trace_probe_log_err(0, NO_BTFARG); + return -ENOENT; + } + name = btf_name_by_offset(ctx->btf, ctx->params[idx].name_off); + if (!name) { + trace_probe_log_err(0, NO_BTF_ENTRY); + return -ENOENT; + } + ret = snprintf(buf, bufsize, "%s%s", name, type); + if (ret >= bufsize) { + trace_probe_log_err(0, ARGS_2LONG); + return -E2BIG; + } + return ret; +} + +/* Return new_argv which must be freed after use */ +const char **traceprobe_expand_meta_args(int argc, const char *argv[], + int *new_argc, char *buf, int bufsize, + struct traceprobe_parse_context *ctx) +{ + const struct btf_param *params = NULL; + int i, j, n, used, ret, args_idx = -1; + const char **new_argv __free(kfree) = NULL; + + ret = argv_has_var_arg(argc, argv, &args_idx, ctx); + if (ret < 0) + return ERR_PTR(ret); + + if (!ret) { + *new_argc = argc; + return NULL; + } + + ret = query_btf_context(ctx); + if (ret < 0 || ctx->nr_params == 0) { + if (args_idx != -1) { + /* $arg* requires BTF info */ + trace_probe_log_err(0, NOSUP_BTFARG); + return (const char **)params; + } + *new_argc = argc; + return NULL; + } + + if (args_idx >= 0) + *new_argc = argc + ctx->nr_params - 1; + else + *new_argc = argc; + + new_argv = kcalloc(*new_argc, sizeof(char *), GFP_KERNEL); + if (!new_argv) + return ERR_PTR(-ENOMEM); + + used = 0; + for (i = 0, j = 0; i < argc; i++) { + trace_probe_log_set_index(i + 2); + if (i == args_idx) { + for (n = 0; n < ctx->nr_params; n++) { + ret = sprint_nth_btf_arg(n, "", buf + used, + bufsize - used, ctx); + if (ret < 0) + return ERR_PTR(ret); + + new_argv[j++] = buf + used; + used += ret + 1; + } + continue; + } + + if (str_has_prefix(argv[i], "$arg")) { + char *type = NULL; + + n = simple_strtoul(argv[i] + 4, &type, 10); + if (type && !(*type == ':' || *type == '\0')) { + trace_probe_log_err(0, BAD_VAR); + return ERR_PTR(-ENOENT); + } + /* Note: $argN starts from $arg1 */ + ret = sprint_nth_btf_arg(n - 1, type, buf + used, + bufsize - used, ctx); + if (ret < 0) + return ERR_PTR(ret); + new_argv[j++] = buf + used; + used += ret + 1; + } else + new_argv[j++] = argv[i]; + } + + return_ptr(new_argv); +} + +/* @buf: *buf must be equal to NULL. Caller must to free *buf */ +int traceprobe_expand_dentry_args(int argc, const char *argv[], char **buf) +{ + int i, used, ret; + const int bufsize = MAX_DENTRY_ARGS_LEN; + char *tmpbuf __free(kfree) = NULL; + + if (*buf) + return -EINVAL; + + used = 0; + for (i = 0; i < argc; i++) { + char *tmp __free(kfree) = NULL; + char *equal; + size_t arg_len; + + if (!glob_match("*:%p[dD]", argv[i])) + continue; + + if (!tmpbuf) { + tmpbuf = kmalloc(bufsize, GFP_KERNEL); + if (!tmpbuf) + return -ENOMEM; + } + + tmp = kstrdup(argv[i], GFP_KERNEL); + if (!tmp) + return -ENOMEM; + + equal = strchr(tmp, '='); + if (equal) + *equal = '\0'; + arg_len = strlen(argv[i]); + tmp[arg_len - 4] = '\0'; + if (argv[i][arg_len - 1] == 'd') + ret = snprintf(tmpbuf + used, bufsize - used, + "%s%s+0x0(+0x%zx(%s)):string", + equal ? tmp : "", equal ? "=" : "", + offsetof(struct dentry, d_name.name), + equal ? equal + 1 : tmp); + else + ret = snprintf(tmpbuf + used, bufsize - used, + "%s%s+0x0(+0x%zx(+0x%zx(%s))):string", + equal ? tmp : "", equal ? "=" : "", + offsetof(struct dentry, d_name.name), + offsetof(struct file, f_path.dentry), + equal ? equal + 1 : tmp); + + if (ret >= bufsize - used) + return -ENOMEM; + argv[i] = tmpbuf + used; + used += ret + 1; + } + + *buf = no_free_ptr(tmpbuf); + return 0; +} + +void traceprobe_finish_parse(struct traceprobe_parse_context *ctx) +{ + clear_btf_context(ctx); +} + int traceprobe_update_arg(struct probe_arg *arg) { struct fetch_insn *code = arg->code; @@ -1043,12 +2045,18 @@ void trace_probe_cleanup(struct trace_probe *tp) for (i = 0; i < tp->nr_args; i++) traceprobe_free_probe_arg(&tp->args[i]); + if (tp->entry_arg) { + kfree(tp->entry_arg->code); + kfree(tp->entry_arg); + tp->entry_arg = NULL; + } + if (tp->event) trace_probe_unlink(tp); } int trace_probe_init(struct trace_probe *tp, const char *event, - const char *group, bool alloc_filter) + const char *group, bool alloc_filter, int nargs) { struct trace_event_call *call; size_t size = sizeof(struct trace_probe_event); @@ -1084,6 +2092,11 @@ int trace_probe_init(struct trace_probe *tp, const char *event, goto error; } + tp->nr_args = nargs; + /* Make sure pointers in args[] are NULL */ + if (nargs) + memset(tp->args, 0, sizeof(tp->args[0]) * nargs); + return 0; error: @@ -1170,7 +2183,7 @@ int trace_probe_remove_file(struct trace_probe *tp, return -ENOENT; list_del_rcu(&link->list); - kvfree_rcu(link); + kvfree_rcu_mightsleep(link); if (list_empty(&tp->event->files)) trace_probe_clear_flag(tp, TP_FLAG_TRACE); @@ -1237,3 +2250,30 @@ int trace_probe_create(const char *raw_command, int (*createfn)(int, const char return ret; } + +int trace_probe_print_args(struct trace_seq *s, struct probe_arg *args, int nr_args, + u8 *data, void *field) +{ + void *p; + int i, j; + + for (i = 0; i < nr_args; i++) { + struct probe_arg *a = args + i; + + trace_seq_printf(s, " %s=", a->name); + if (likely(!a->count)) { + if (!a->type->print(s, data + a->offset, field)) + return -ENOMEM; + continue; + } + trace_seq_putc(s, '{'); + p = data + a->offset; + for (j = 0; j < a->count; j++) { + if (!a->type->print(s, p, field)) + return -ENOMEM; + trace_seq_putc(s, j == a->count - 1 ? '}' : ','); + p += a->type->size; + } + } + return 0; +} diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index 23acfd1c3812..9fc56c937130 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -10,19 +10,22 @@ * Author: Srikar Dronamraju */ +#include <linux/bitops.h> +#include <linux/btf.h> +#include <linux/cleanup.h> +#include <linux/kprobes.h> +#include <linux/limits.h> +#include <linux/perf_event.h> +#include <linux/ptrace.h> #include <linux/seq_file.h> #include <linux/slab.h> #include <linux/smp.h> -#include <linux/tracefs.h> -#include <linux/types.h> #include <linux/string.h> -#include <linux/ptrace.h> -#include <linux/perf_event.h> -#include <linux/kprobes.h> #include <linux/stringify.h> -#include <linux/limits.h> +#include <linux/tracefs.h> +#include <linux/types.h> #include <linux/uaccess.h> -#include <linux/bitops.h> + #include <asm/bitsperlong.h> #include "trace.h" @@ -32,6 +35,8 @@ #define MAX_ARGSTR_LEN 63 #define MAX_ARRAY_LEN 64 #define MAX_ARG_NAME_LEN 32 +#define MAX_BTF_ARGS_LEN 128 +#define MAX_DENTRY_ARGS_LEN 256 #define MAX_STRING_SIZE PATH_MAX /* Reserved field names */ @@ -89,6 +94,7 @@ enum fetch_op { FETCH_OP_ARG, /* Function argument : .param */ FETCH_OP_FOFFS, /* File offset: .immediate */ FETCH_OP_DATA, /* Allocated data: .data */ + FETCH_OP_EDATA, /* Entry data: .offset */ // Stage 2 (dereference) op FETCH_OP_DEREF, /* Dereference: .offset */ FETCH_OP_UDEREF, /* User-space Dereference: .offset */ @@ -99,6 +105,7 @@ enum fetch_op { FETCH_OP_ST_STRING, /* String: .offset, .size */ FETCH_OP_ST_USTRING, /* User String: .offset, .size */ FETCH_OP_ST_SYMSTR, /* Kernel Symbol String: .offset, .size */ + FETCH_OP_ST_EDATA, /* Store Entry Data: .offset */ // Stage 4 (modify) op FETCH_OP_MOD_BF, /* Bitfield: .basesize, .lshift, .rshift */ // Stage 5 (loop) op @@ -166,6 +173,7 @@ DECLARE_BASIC_PRINT_TYPE_FUNC(x16); DECLARE_BASIC_PRINT_TYPE_FUNC(x32); DECLARE_BASIC_PRINT_TYPE_FUNC(x64); +DECLARE_BASIC_PRINT_TYPE_FUNC(char); DECLARE_BASIC_PRINT_TYPE_FUNC(string); DECLARE_BASIC_PRINT_TYPE_FUNC(symbol); @@ -228,6 +236,11 @@ struct probe_arg { const struct fetch_type *type; /* Type of this argument */ }; +struct probe_entry_arg { + struct fetch_insn *code; + unsigned int size; /* The entry data size */ +}; + struct trace_uprobe_filter { rwlock_t rwlock; int nr_systemwide; @@ -249,6 +262,7 @@ struct trace_probe { struct trace_probe_event *event; ssize_t size; /* trace entry size */ unsigned int nr_args; + struct probe_entry_arg *entry_arg; /* This is only for return probe */ struct probe_arg args[]; }; @@ -257,16 +271,21 @@ struct event_file_link { struct list_head list; }; +static inline unsigned int trace_probe_load_flag(struct trace_probe *tp) +{ + return smp_load_acquire(&tp->event->flags); +} + static inline bool trace_probe_test_flag(struct trace_probe *tp, unsigned int flag) { - return !!(tp->event->flags & flag); + return !!(trace_probe_load_flag(tp) & flag); } static inline void trace_probe_set_flag(struct trace_probe *tp, unsigned int flag) { - tp->event->flags |= flag; + smp_store_release(&tp->event->flags, tp->event->flags | flag); } static inline void trace_probe_clear_flag(struct trace_probe *tp, @@ -307,7 +326,7 @@ trace_probe_primary_from_call(struct trace_event_call *call) { struct trace_probe_event *tpe = trace_probe_event_from_call(call); - return list_first_entry(&tpe->probes, struct trace_probe, list); + return list_first_entry_or_null(&tpe->probes, struct trace_probe, list); } static inline struct list_head *trace_probe_probe_list(struct trace_probe *tp) @@ -334,7 +353,7 @@ static inline bool trace_probe_has_single_file(struct trace_probe *tp) } int trace_probe_init(struct trace_probe *tp, const char *event, - const char *group, bool alloc_filter); + const char *group, bool alloc_filter, int nargs); void trace_probe_cleanup(struct trace_probe *tp); int trace_probe_append(struct trace_probe *tp, struct trace_probe *to); void trace_probe_unlink(struct trace_probe *tp); @@ -348,25 +367,93 @@ int trace_probe_compare_arg_type(struct trace_probe *a, struct trace_probe *b); bool trace_probe_match_command_args(struct trace_probe *tp, int argc, const char **argv); int trace_probe_create(const char *raw_command, int (*createfn)(int, const char **)); +int trace_probe_print_args(struct trace_seq *s, struct probe_arg *args, int nr_args, + u8 *data, void *field); + +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API +int traceprobe_get_entry_data_size(struct trace_probe *tp); +/* This is a runtime function to store entry data */ +void store_trace_entry_data(void *edata, struct trace_probe *tp, struct pt_regs *regs); +#else /* !CONFIG_HAVE_FUNCTION_ARG_ACCESS_API */ +static inline int traceprobe_get_entry_data_size(struct trace_probe *tp) +{ + return 0; +} +#define store_trace_entry_data(edata, tp, regs) do { } while (0) +#endif #define trace_probe_for_each_link(pos, tp) \ list_for_each_entry(pos, &(tp)->event->files, list) #define trace_probe_for_each_link_rcu(pos, tp) \ list_for_each_entry_rcu(pos, &(tp)->event->files, list) +/* + * The flags used for parsing trace_probe arguments. + * TPARG_FL_RETURN, TPARG_FL_FENTRY and TPARG_FL_TEVENT are mutually exclusive. + * TPARG_FL_KERNEL and TPARG_FL_USER are also mutually exclusive. + * TPARG_FL_FPROBE and TPARG_FL_TPOINT are optional but it should be with + * TPARG_FL_KERNEL. + */ #define TPARG_FL_RETURN BIT(0) #define TPARG_FL_KERNEL BIT(1) #define TPARG_FL_FENTRY BIT(2) -#define TPARG_FL_TPOINT BIT(3) +#define TPARG_FL_TEVENT BIT(3) #define TPARG_FL_USER BIT(4) -#define TPARG_FL_MASK GENMASK(4, 0) +#define TPARG_FL_FPROBE BIT(5) +#define TPARG_FL_TPOINT BIT(6) +#define TPARG_FL_LOC_MASK GENMASK(4, 0) + +static inline bool tparg_is_function_entry(unsigned int flags) +{ + return (flags & TPARG_FL_LOC_MASK) == (TPARG_FL_KERNEL | TPARG_FL_FENTRY); +} + +static inline bool tparg_is_function_return(unsigned int flags) +{ + return (flags & TPARG_FL_LOC_MASK) == (TPARG_FL_KERNEL | TPARG_FL_RETURN); +} + +struct traceprobe_parse_context { + struct trace_event_call *event; + /* BTF related parameters */ + const char *funcname; /* Function name in BTF */ + const struct btf_type *proto; /* Prototype of the function */ + const struct btf_param *params; /* Parameter of the function */ + s32 nr_params; /* The number of the parameters */ + struct btf *btf; /* The BTF to be used */ + const struct btf_type *last_type; /* Saved type */ + u32 last_bitoffs; /* Saved bitoffs */ + u32 last_bitsize; /* Saved bitsize */ + struct trace_probe *tp; + unsigned int flags; + int offset; +}; extern int traceprobe_parse_probe_arg(struct trace_probe *tp, int i, - const char *argv, unsigned int flags); + const char *argv, + struct traceprobe_parse_context *ctx); +const char **traceprobe_expand_meta_args(int argc, const char *argv[], + int *new_argc, char *buf, int bufsize, + struct traceprobe_parse_context *ctx); +extern int traceprobe_expand_dentry_args(int argc, const char *argv[], char **buf); extern int traceprobe_update_arg(struct probe_arg *arg); extern void traceprobe_free_probe_arg(struct probe_arg *arg); +/* + * If either traceprobe_parse_probe_arg() or traceprobe_expand_meta_args() is called, + * this MUST be called for clean up the context and return a resource. + */ +void traceprobe_finish_parse(struct traceprobe_parse_context *ctx); +static inline void traceprobe_free_parse_ctx(struct traceprobe_parse_context *ctx) +{ + traceprobe_finish_parse(ctx); + kfree(ctx); +} + +DEFINE_FREE(traceprobe_parse_context, struct traceprobe_parse_context *, + if (_T) traceprobe_free_parse_ctx(_T)) + extern int traceprobe_split_symbol_offset(char *symbol, long *offset); int traceprobe_parse_event_name(const char **pevent, const char **pgroup, char *buf, int offset); @@ -401,11 +488,14 @@ extern int traceprobe_define_arg_fields(struct trace_event_call *event_call, C(REFCNT_OPEN_BRACE, "Reference counter brace is not closed"), \ C(BAD_REFCNT_SUFFIX, "Reference counter has wrong suffix"), \ C(BAD_UPROBE_OFFS, "Invalid uprobe offset"), \ - C(MAXACT_NO_KPROBE, "Maxactive is not for kprobe"), \ + C(BAD_MAXACT_TYPE, "Maxactive is only for function exit"), \ C(BAD_MAXACT, "Invalid maxactive number"), \ C(MAXACT_TOO_BIG, "Maxactive is too big"), \ C(BAD_PROBE_ADDR, "Invalid probed address or symbol"), \ + C(NON_UNIQ_SYMBOL, "The symbol is not unique"), \ C(BAD_RETPROBE, "Retprobe address must be an function entry"), \ + C(NO_TRACEPOINT, "Tracepoint is not found"), \ + C(BAD_TP_NAME, "Invalid character in tracepoint name"),\ C(BAD_ADDR_SUFFIX, "Invalid probed address suffix"), \ C(NO_GROUP_NAME, "Group name is not specified"), \ C(GROUP_TOO_LONG, "Group name is too long"), \ @@ -415,6 +505,7 @@ extern int traceprobe_define_arg_fields(struct trace_event_call *event_call, C(BAD_EVENT_NAME, "Event name must follow the same rules as C identifiers"), \ C(EVENT_EXIST, "Given group/event name is already used by another event"), \ C(RETVAL_ON_PROBE, "$retval is not available on probe"), \ + C(NO_RETVAL, "This function returns 'void' type"), \ C(BAD_STACK_NUM, "Invalid stack number"), \ C(BAD_ARG_NUM, "Invalid argument number"), \ C(BAD_VAR, "Invalid $-valiable specified"), \ @@ -453,7 +544,24 @@ extern int traceprobe_define_arg_fields(struct trace_event_call *event_call, C(NO_EVENT_INFO, "This requires both group and event name to attach"),\ C(BAD_ATTACH_EVENT, "Attached event does not exist"),\ C(BAD_ATTACH_ARG, "Attached event does not have this field"),\ - C(NO_EP_FILTER, "No filter rule after 'if'"), + C(NO_EP_FILTER, "No filter rule after 'if'"), \ + C(NOSUP_BTFARG, "BTF is not available or not supported"), \ + C(NO_BTFARG, "This variable is not found at this probe point"),\ + C(NO_BTF_ENTRY, "No BTF entry for this probe point"), \ + C(BAD_VAR_ARGS, "$arg* must be an independent parameter without name etc."),\ + C(NOFENTRY_ARGS, "$arg* can be used only on function entry or exit"), \ + C(DOUBLE_ARGS, "$arg* can be used only once in the parameters"), \ + C(ARGS_2LONG, "$arg* failed because the argument list is too long"), \ + C(ARGIDX_2BIG, "$argN index is too big"), \ + C(NO_PTR_STRCT, "This is not a pointer to union/structure."), \ + C(NOSUP_DAT_ARG, "Non pointer structure/union argument is not supported."),\ + C(BAD_HYPHEN, "Failed to parse single hyphen. Forgot '>'?"), \ + C(NO_BTF_FIELD, "This field is not found."), \ + C(BAD_BTF_TID, "Failed to get BTF type info."),\ + C(BAD_TYPE4STR, "This type does not fit for string."),\ + C(NEED_STRING_TYPE, "$comm and immediate-string only accepts string type"),\ + C(TOO_MANY_ARGS, "Too many arguments are specified"), \ + C(TOO_MANY_EARGS, "Too many entry arguments specified"), #undef C #define C(a, b) TP_ERR_##a @@ -470,10 +578,17 @@ struct trace_probe_log { int index; }; -void trace_probe_log_init(const char *subsystem, int argc, const char **argv); +const char *trace_probe_log_init(const char *subsystem, int argc, const char **argv); void trace_probe_log_set_index(int index); void trace_probe_log_clear(void); void __trace_probe_log_err(int offset, int err); +DEFINE_FREE(trace_probe_log_clear, const char *, if (_T) trace_probe_log_clear()) + #define trace_probe_log_err(offs, err) \ __trace_probe_log_err(offs, TP_ERR_##err) + +struct uprobe_dispatch_data { + struct trace_uprobe *tu; + unsigned long bp_addr; +}; diff --git a/kernel/trace/trace_probe_kernel.h b/kernel/trace/trace_probe_kernel.h index 77dbd9ff9782..bb723eefd7b7 100644 --- a/kernel/trace/trace_probe_kernel.h +++ b/kernel/trace/trace_probe_kernel.h @@ -2,8 +2,6 @@ #ifndef __TRACE_PROBE_KERNEL_H_ #define __TRACE_PROBE_KERNEL_H_ -#define FAULT_STRING "(fault)" - /* * This depends on trace_probe.h, but can not include it due to * the way trace_probe_tmpl.h is used by trace_kprobe.c and trace_eprobe.c. @@ -12,31 +10,23 @@ */ /* Return the length of string -- including null terminal byte */ static nokprobe_inline int -kern_fetch_store_strlen_user(unsigned long addr) +fetch_store_strlen_user(unsigned long addr) { const void __user *uaddr = (__force const void __user *)addr; - int ret; - ret = strnlen_user_nofault(uaddr, MAX_STRING_SIZE); - /* - * strnlen_user_nofault returns zero on fault, insert the - * FAULT_STRING when that occurs. - */ - if (ret <= 0) - return strlen(FAULT_STRING) + 1; - return ret; + return strnlen_user_nofault(uaddr, MAX_STRING_SIZE); } /* Return the length of string -- including null terminal byte */ static nokprobe_inline int -kern_fetch_store_strlen(unsigned long addr) +fetch_store_strlen(unsigned long addr) { int ret, len = 0; u8 c; #ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE if (addr < TASK_SIZE) - return kern_fetch_store_strlen_user(addr); + return fetch_store_strlen_user(addr); #endif do { @@ -44,18 +34,14 @@ kern_fetch_store_strlen(unsigned long addr) len++; } while (c && ret == 0 && len < MAX_STRING_SIZE); - /* For faults, return enough to hold the FAULT_STRING */ - return (ret < 0) ? strlen(FAULT_STRING) + 1 : len; + return (ret < 0) ? ret : len; } -static nokprobe_inline void set_data_loc(int ret, void *dest, void *__dest, void *base, int len) +static nokprobe_inline void set_data_loc(int ret, void *dest, void *__dest, void *base) { - if (ret >= 0) { - *(u32 *)dest = make_data_loc(ret, __dest - base); - } else { - strscpy(__dest, FAULT_STRING, len); - ret = strlen(__dest) + 1; - } + if (ret < 0) + ret = 0; + *(u32 *)dest = make_data_loc(ret, __dest - base); } /* @@ -63,7 +49,7 @@ static nokprobe_inline void set_data_loc(int ret, void *dest, void *__dest, void * with max length and relative data location. */ static nokprobe_inline int -kern_fetch_store_string_user(unsigned long addr, void *dest, void *base) +fetch_store_string_user(unsigned long addr, void *dest, void *base) { const void __user *uaddr = (__force const void __user *)addr; int maxlen = get_loc_len(*(u32 *)dest); @@ -76,7 +62,7 @@ kern_fetch_store_string_user(unsigned long addr, void *dest, void *base) __dest = get_loc_data(dest, base); ret = strncpy_from_user_nofault(__dest, uaddr, maxlen); - set_data_loc(ret, dest, __dest, base, maxlen); + set_data_loc(ret, dest, __dest, base); return ret; } @@ -86,7 +72,7 @@ kern_fetch_store_string_user(unsigned long addr, void *dest, void *base) * length and relative data location. */ static nokprobe_inline int -kern_fetch_store_string(unsigned long addr, void *dest, void *base) +fetch_store_string(unsigned long addr, void *dest, void *base) { int maxlen = get_loc_len(*(u32 *)dest); void *__dest; @@ -94,7 +80,7 @@ kern_fetch_store_string(unsigned long addr, void *dest, void *base) #ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE if ((unsigned long)addr < TASK_SIZE) - return kern_fetch_store_string_user(addr, dest, base); + return fetch_store_string_user(addr, dest, base); #endif if (unlikely(!maxlen)) @@ -107,9 +93,27 @@ kern_fetch_store_string(unsigned long addr, void *dest, void *base) * probing. */ ret = strncpy_from_kernel_nofault(__dest, (void *)addr, maxlen); - set_data_loc(ret, dest, __dest, base, maxlen); + set_data_loc(ret, dest, __dest, base); return ret; } +static nokprobe_inline int +probe_mem_read_user(void *dest, void *src, size_t size) +{ + const void __user *uaddr = (__force const void __user *)src; + + return copy_from_user_nofault(dest, uaddr, size); +} + +static nokprobe_inline int +probe_mem_read(void *dest, void *src, size_t size) +{ +#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE + if ((unsigned long)src < TASK_SIZE) + return probe_mem_read_user(dest, src, size); +#endif + return copy_from_kernel_nofault(dest, src, size); +} + #endif /* __TRACE_PROBE_KERNEL_H_ */ diff --git a/kernel/trace/trace_probe_tmpl.h b/kernel/trace/trace_probe_tmpl.h index 5cea672243f6..f39b37fcdb3b 100644 --- a/kernel/trace/trace_probe_tmpl.h +++ b/kernel/trace/trace_probe_tmpl.h @@ -54,7 +54,7 @@ fetch_apply_bitfield(struct fetch_insn *code, void *buf) * If dest is NULL, don't store result and return required dynamic data size. */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, void *dest, void *base); static nokprobe_inline int fetch_store_strlen(unsigned long addr); static nokprobe_inline int @@ -98,6 +98,26 @@ fetch_store_symstring(unsigned long addr, void *dest, void *base) return sprint_symbol(__dest, addr); } +/* common part of process_fetch_insn*/ +static nokprobe_inline int +process_common_fetch_insn(struct fetch_insn *code, unsigned long *val) +{ + switch (code->op) { + case FETCH_OP_IMM: + *val = code->immediate; + break; + case FETCH_OP_COMM: + *val = (unsigned long)current->comm; + break; + case FETCH_OP_DATA: + *val = (unsigned long)code->data; + break; + default: + return -EILSEQ; + } + return 0; +} + /* From the 2nd stage, routine is same */ static nokprobe_inline int process_fetch_insn_bottom(struct fetch_insn *code, unsigned long val, @@ -136,11 +156,11 @@ stage3: code++; goto array; case FETCH_OP_ST_USTRING: - ret += fetch_store_strlen_user(val + code->offset); + ret = fetch_store_strlen_user(val + code->offset); code++; goto array; case FETCH_OP_ST_SYMSTR: - ret += fetch_store_symstrlen(val + code->offset); + ret = fetch_store_symstrlen(val + code->offset); code++; goto array; default: @@ -184,6 +204,8 @@ stage3: array: /* the last stage: Loop on array */ if (code->op == FETCH_OP_LP_ARRAY) { + if (ret < 0) + ret = 0; total += ret; if (++i < code->param) { code = s3; @@ -210,7 +232,7 @@ array: /* Sum up total data length for dynamic arrays (strings) */ static nokprobe_inline int -__get_data_size(struct trace_probe *tp, struct pt_regs *regs) +__get_data_size(struct trace_probe *tp, void *regs, void *edata) { struct probe_arg *arg; int i, len, ret = 0; @@ -218,7 +240,7 @@ __get_data_size(struct trace_probe *tp, struct pt_regs *regs) for (i = 0; i < tp->nr_args; i++) { arg = tp->args + i; if (unlikely(arg->dynamic)) { - len = process_fetch_insn(arg->code, regs, NULL, NULL); + len = process_fetch_insn(arg->code, regs, edata, NULL, NULL); if (len > 0) ret += len; } @@ -229,7 +251,7 @@ __get_data_size(struct trace_probe *tp, struct pt_regs *regs) /* Store the value of each argument */ static nokprobe_inline void -store_trace_args(void *data, struct trace_probe *tp, void *rec, +store_trace_args(void *data, struct trace_probe *tp, void *rec, void *edata, int header_size, int maxlen) { struct probe_arg *arg; @@ -244,40 +266,10 @@ store_trace_args(void *data, struct trace_probe *tp, void *rec, /* Point the dynamic data area if needed */ if (unlikely(arg->dynamic)) *dl = make_data_loc(maxlen, dyndata - base); - ret = process_fetch_insn(arg->code, rec, dl, base); - if (unlikely(ret < 0 && arg->dynamic)) { - *dl = make_data_loc(0, dyndata - base); - } else { + ret = process_fetch_insn(arg->code, rec, edata, dl, base); + if (arg->dynamic && likely(ret > 0)) { dyndata += ret; maxlen -= ret; } } } - -static inline int -print_probe_args(struct trace_seq *s, struct probe_arg *args, int nr_args, - u8 *data, void *field) -{ - void *p; - int i, j; - - for (i = 0; i < nr_args; i++) { - struct probe_arg *a = args + i; - - trace_seq_printf(s, " %s=", a->name); - if (likely(!a->count)) { - if (!a->type->print(s, data + a->offset, field)) - return -ENOMEM; - continue; - } - trace_seq_putc(s, '{'); - p = data + a->offset; - for (j = 0; j < a->count; j++) { - if (!a->type->print(s, p, field)) - return -ENOMEM; - trace_seq_putc(s, j == a->count - 1 ? '}' : ','); - p += a->type->size; - } - } - return 0; -} diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index c9ffdcfe622e..c46d584ded3b 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -8,6 +8,7 @@ #include <linux/module.h> #include <linux/kallsyms.h> #include <linux/uaccess.h> +#include <linux/kmemleak.h> #include <linux/ftrace.h> #include <trace/events/sched.h> @@ -148,3 +149,518 @@ void tracing_stop_tgid_record(void) { tracing_stop_sched_switch(RECORD_TGID); } + +/* + * The tgid_map array maps from pid to tgid; i.e. the value stored at index i + * is the tgid last observed corresponding to pid=i. + */ +static int *tgid_map; + +/* The maximum valid index into tgid_map. */ +static size_t tgid_map_max; + +#define SAVED_CMDLINES_DEFAULT 128 +#define NO_CMDLINE_MAP UINT_MAX +/* + * Preemption must be disabled before acquiring trace_cmdline_lock. + * The various trace_arrays' max_lock must be acquired in a context + * where interrupt is disabled. + */ +static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; +struct saved_cmdlines_buffer { + unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; + unsigned *map_cmdline_to_pid; + unsigned cmdline_num; + int cmdline_idx; + char saved_cmdlines[]; +}; +static struct saved_cmdlines_buffer *savedcmd; + +/* Holds the size of a cmdline and pid element */ +#define SAVED_CMDLINE_MAP_ELEMENT_SIZE(s) \ + (TASK_COMM_LEN + sizeof((s)->map_cmdline_to_pid[0])) + +static inline char *get_saved_cmdlines(int idx) +{ + return &savedcmd->saved_cmdlines[idx * TASK_COMM_LEN]; +} + +static inline void set_cmdline(int idx, const char *cmdline) +{ + strscpy(get_saved_cmdlines(idx), cmdline, TASK_COMM_LEN); +} + +static void free_saved_cmdlines_buffer(struct saved_cmdlines_buffer *s) +{ + int order = get_order(sizeof(*s) + s->cmdline_num * TASK_COMM_LEN); + + kmemleak_free(s); + free_pages((unsigned long)s, order); +} + +static struct saved_cmdlines_buffer *allocate_cmdlines_buffer(unsigned int val) +{ + struct saved_cmdlines_buffer *s; + struct page *page; + int orig_size, size; + int order; + + /* Figure out how much is needed to hold the given number of cmdlines */ + orig_size = sizeof(*s) + val * SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); + order = get_order(orig_size); + size = 1 << (order + PAGE_SHIFT); + page = alloc_pages(GFP_KERNEL, order); + if (!page) + return NULL; + + s = page_address(page); + kmemleak_alloc(s, size, 1, GFP_KERNEL); + memset(s, 0, sizeof(*s)); + + /* Round up to actual allocation */ + val = (size - sizeof(*s)) / SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); + s->cmdline_num = val; + + /* Place map_cmdline_to_pid array right after saved_cmdlines */ + s->map_cmdline_to_pid = (unsigned *)&s->saved_cmdlines[val * TASK_COMM_LEN]; + + memset(&s->map_pid_to_cmdline, NO_CMDLINE_MAP, + sizeof(s->map_pid_to_cmdline)); + memset(s->map_cmdline_to_pid, NO_CMDLINE_MAP, + val * sizeof(*s->map_cmdline_to_pid)); + + return s; +} + +int trace_create_savedcmd(void) +{ + savedcmd = allocate_cmdlines_buffer(SAVED_CMDLINES_DEFAULT); + + return savedcmd ? 0 : -ENOMEM; +} + +int trace_save_cmdline(struct task_struct *tsk) +{ + unsigned tpid, idx; + + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + BUILD_BUG_ON(!is_power_of_2(PID_MAX_DEFAULT)); + + tpid = tsk->pid & (PID_MAX_DEFAULT - 1); + + /* + * It's not the end of the world if we don't get + * the lock, but we also don't want to spin + * nor do we want to disable interrupts, + * so if we miss here, then better luck next time. + * + * This is called within the scheduler and wake up, so interrupts + * had better been disabled and run queue lock been held. + */ + lockdep_assert_preemption_disabled(); + if (!arch_spin_trylock(&trace_cmdline_lock)) + return 0; + + idx = savedcmd->map_pid_to_cmdline[tpid]; + if (idx == NO_CMDLINE_MAP) { + idx = (savedcmd->cmdline_idx + 1) % savedcmd->cmdline_num; + + savedcmd->map_pid_to_cmdline[tpid] = idx; + savedcmd->cmdline_idx = idx; + } + + savedcmd->map_cmdline_to_pid[idx] = tsk->pid; + set_cmdline(idx, tsk->comm); + + arch_spin_unlock(&trace_cmdline_lock); + + return 1; +} + +static void __trace_find_cmdline(int pid, char comm[]) +{ + unsigned map; + int tpid; + + if (!pid) { + strcpy(comm, "<idle>"); + return; + } + + if (WARN_ON_ONCE(pid < 0)) { + strcpy(comm, "<XXX>"); + return; + } + + tpid = pid & (PID_MAX_DEFAULT - 1); + map = savedcmd->map_pid_to_cmdline[tpid]; + if (map != NO_CMDLINE_MAP) { + tpid = savedcmd->map_cmdline_to_pid[map]; + if (tpid == pid) { + strscpy(comm, get_saved_cmdlines(map), TASK_COMM_LEN); + return; + } + } + strcpy(comm, "<...>"); +} + +void trace_find_cmdline(int pid, char comm[]) +{ + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + + __trace_find_cmdline(pid, comm); + + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); +} + +static int *trace_find_tgid_ptr(int pid) +{ + /* + * Pairs with the smp_store_release in set_tracer_flag() to ensure that + * if we observe a non-NULL tgid_map then we also observe the correct + * tgid_map_max. + */ + int *map = smp_load_acquire(&tgid_map); + + if (unlikely(!map || pid > tgid_map_max)) + return NULL; + + return &map[pid]; +} + +int trace_find_tgid(int pid) +{ + int *ptr = trace_find_tgid_ptr(pid); + + return ptr ? *ptr : 0; +} + +static int trace_save_tgid(struct task_struct *tsk) +{ + int *ptr; + + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + ptr = trace_find_tgid_ptr(tsk->pid); + if (!ptr) + return 0; + + *ptr = tsk->tgid; + return 1; +} + +static bool tracing_record_taskinfo_skip(int flags) +{ + if (unlikely(!(flags & (TRACE_RECORD_CMDLINE | TRACE_RECORD_TGID)))) + return true; + if (!__this_cpu_read(trace_taskinfo_save)) + return true; + return false; +} + +/** + * tracing_record_taskinfo - record the task info of a task + * + * @task: task to record + * @flags: TRACE_RECORD_CMDLINE for recording comm + * TRACE_RECORD_TGID for recording tgid + */ +void tracing_record_taskinfo(struct task_struct *task, int flags) +{ + bool done; + + if (tracing_record_taskinfo_skip(flags)) + return; + + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(task); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(task); + + /* If recording any information failed, retry again soon. */ + if (!done) + return; + + __this_cpu_write(trace_taskinfo_save, false); +} + +/** + * tracing_record_taskinfo_sched_switch - record task info for sched_switch + * + * @prev: previous task during sched_switch + * @next: next task during sched_switch + * @flags: TRACE_RECORD_CMDLINE for recording comm + * TRACE_RECORD_TGID for recording tgid + */ +void tracing_record_taskinfo_sched_switch(struct task_struct *prev, + struct task_struct *next, int flags) +{ + bool done; + + if (tracing_record_taskinfo_skip(flags)) + return; + + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(prev); + done &= !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(next); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(prev); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(next); + + /* If recording any information failed, retry again soon. */ + if (!done) + return; + + __this_cpu_write(trace_taskinfo_save, false); +} + +/* Helpers to record a specific task information */ +void tracing_record_cmdline(struct task_struct *task) +{ + tracing_record_taskinfo(task, TRACE_RECORD_CMDLINE); +} + +void tracing_record_tgid(struct task_struct *task) +{ + tracing_record_taskinfo(task, TRACE_RECORD_TGID); +} + +int trace_alloc_tgid_map(void) +{ + int *map; + + if (tgid_map) + return 0; + + tgid_map_max = init_pid_ns.pid_max; + map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), + GFP_KERNEL); + if (!map) + return -ENOMEM; + + /* + * Pairs with smp_load_acquire() in + * trace_find_tgid_ptr() to ensure that if it observes + * the tgid_map we just allocated then it also observes + * the corresponding tgid_map_max value. + */ + smp_store_release(&tgid_map, map); + return 0; +} + +static void *saved_tgids_next(struct seq_file *m, void *v, loff_t *pos) +{ + int pid = ++(*pos); + + return trace_find_tgid_ptr(pid); +} + +static void *saved_tgids_start(struct seq_file *m, loff_t *pos) +{ + int pid = *pos; + + return trace_find_tgid_ptr(pid); +} + +static void saved_tgids_stop(struct seq_file *m, void *v) +{ +} + +static int saved_tgids_show(struct seq_file *m, void *v) +{ + int *entry = (int *)v; + int pid = entry - tgid_map; + int tgid = *entry; + + if (tgid == 0) + return SEQ_SKIP; + + seq_printf(m, "%d %d\n", pid, tgid); + return 0; +} + +static const struct seq_operations tracing_saved_tgids_seq_ops = { + .start = saved_tgids_start, + .stop = saved_tgids_stop, + .next = saved_tgids_next, + .show = saved_tgids_show, +}; + +static int tracing_saved_tgids_open(struct inode *inode, struct file *filp) +{ + int ret; + + ret = tracing_check_open_get_tr(NULL); + if (ret) + return ret; + + return seq_open(filp, &tracing_saved_tgids_seq_ops); +} + + +const struct file_operations tracing_saved_tgids_fops = { + .open = tracing_saved_tgids_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static void *saved_cmdlines_next(struct seq_file *m, void *v, loff_t *pos) +{ + unsigned int *ptr = v; + + if (*pos || m->count) + ptr++; + + (*pos)++; + + for (; ptr < &savedcmd->map_cmdline_to_pid[savedcmd->cmdline_num]; + ptr++) { + if (*ptr == -1 || *ptr == NO_CMDLINE_MAP) + continue; + + return ptr; + } + + return NULL; +} + +static void *saved_cmdlines_start(struct seq_file *m, loff_t *pos) +{ + void *v; + loff_t l = 0; + + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + + v = &savedcmd->map_cmdline_to_pid[0]; + while (l <= *pos) { + v = saved_cmdlines_next(m, v, &l); + if (!v) + return NULL; + } + + return v; +} + +static void saved_cmdlines_stop(struct seq_file *m, void *v) +{ + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); +} + +static int saved_cmdlines_show(struct seq_file *m, void *v) +{ + char buf[TASK_COMM_LEN]; + unsigned int *pid = v; + + __trace_find_cmdline(*pid, buf); + seq_printf(m, "%d %s\n", *pid, buf); + return 0; +} + +static const struct seq_operations tracing_saved_cmdlines_seq_ops = { + .start = saved_cmdlines_start, + .next = saved_cmdlines_next, + .stop = saved_cmdlines_stop, + .show = saved_cmdlines_show, +}; + +static int tracing_saved_cmdlines_open(struct inode *inode, struct file *filp) +{ + int ret; + + ret = tracing_check_open_get_tr(NULL); + if (ret) + return ret; + + return seq_open(filp, &tracing_saved_cmdlines_seq_ops); +} + +const struct file_operations tracing_saved_cmdlines_fops = { + .open = tracing_saved_cmdlines_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static ssize_t +tracing_saved_cmdlines_size_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char buf[64]; + int r; + + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + r = scnprintf(buf, sizeof(buf), "%u\n", savedcmd->cmdline_num); + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +void trace_free_saved_cmdlines_buffer(void) +{ + free_saved_cmdlines_buffer(savedcmd); +} + +static int tracing_resize_saved_cmdlines(unsigned int val) +{ + struct saved_cmdlines_buffer *s, *savedcmd_temp; + + s = allocate_cmdlines_buffer(val); + if (!s) + return -ENOMEM; + + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + savedcmd_temp = savedcmd; + savedcmd = s; + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); + free_saved_cmdlines_buffer(savedcmd_temp); + + return 0; +} + +static ssize_t +tracing_saved_cmdlines_size_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + unsigned long val; + int ret; + + ret = kstrtoul_from_user(ubuf, cnt, 10, &val); + if (ret) + return ret; + + /* must have at least 1 entry or less than PID_MAX_DEFAULT */ + if (!val || val > PID_MAX_DEFAULT) + return -EINVAL; + + ret = tracing_resize_saved_cmdlines((unsigned int)val); + if (ret < 0) + return ret; + + *ppos += cnt; + + return cnt; +} + +const struct file_operations tracing_saved_cmdlines_size_fops = { + .open = tracing_open_generic, + .read = tracing_saved_cmdlines_size_read, + .write = tracing_saved_cmdlines_size_write, +}; diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 330aee1c1a49..8faa73d3bba1 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -41,7 +41,7 @@ static void stop_func_tracer(struct trace_array *tr, int graph); static int save_flags; #ifdef CONFIG_FUNCTION_GRAPH_TRACER -# define is_graph(tr) ((tr)->trace_flags & TRACE_ITER_DISPLAY_GRAPH) +# define is_graph(tr) ((tr)->trace_flags & TRACE_ITER(DISPLAY_GRAPH)) #else # define is_graph(tr) false #endif @@ -83,14 +83,14 @@ func_prolog_preempt_disable(struct trace_array *tr, goto out_enable; *data = per_cpu_ptr(tr->array_buffer.data, cpu); - disabled = atomic_inc_return(&(*data)->disabled); + disabled = local_inc_return(&(*data)->disabled); if (unlikely(disabled != 1)) goto out; return 1; out: - atomic_dec(&(*data)->disabled); + local_dec(&(*data)->disabled); out_enable: preempt_enable_notrace(); @@ -112,14 +112,17 @@ static int wakeup_display_graph(struct trace_array *tr, int set) return start_func_tracer(tr, set); } -static int wakeup_graph_entry(struct ftrace_graph_ent *trace) +static int wakeup_graph_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; unsigned int trace_ctx; + u64 *calltime; int ret = 0; - if (ftrace_graph_ignore_func(trace)) + if (ftrace_graph_ignore_func(gops, trace)) return 0; /* * Do not trace a function if it's filtered by set_graph_notrace. @@ -134,27 +137,40 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace) if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return 0; - ret = __trace_graph_entry(tr, trace, trace_ctx); - atomic_dec(&data->disabled); + calltime = fgraph_reserve_data(gops->idx, sizeof(*calltime)); + if (calltime) { + *calltime = trace_clock_local(); + ret = __trace_graph_entry(tr, trace, trace_ctx); + } + local_dec(&data->disabled); preempt_enable_notrace(); return ret; } -static void wakeup_graph_return(struct ftrace_graph_ret *trace) +static void wakeup_graph_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; unsigned int trace_ctx; + u64 *calltime; + u64 rettime; + int size; - ftrace_graph_addr_finish(trace); + ftrace_graph_addr_finish(gops, trace); if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return; - __trace_graph_return(tr, trace, trace_ctx); - atomic_dec(&data->disabled); + rettime = trace_clock_local(); + + calltime = fgraph_retrieve_data(gops->idx, &size); + if (calltime) + __trace_graph_return(tr, trace, trace_ctx, *calltime, rettime); + local_dec(&data->disabled); preempt_enable_notrace(); return; } @@ -220,10 +236,10 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip, return; local_irq_save(flags); - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, fregs); local_irq_restore(flags); - atomic_dec(&data->disabled); + local_dec(&data->disabled); preempt_enable_notrace(); } @@ -231,8 +247,8 @@ static int register_wakeup_function(struct trace_array *tr, int graph, int set) { int ret; - /* 'set' is set if TRACE_ITER_FUNCTION is about to be set */ - if (function_enabled || (!set && !(tr->trace_flags & TRACE_ITER_FUNCTION))) + /* 'set' is set if TRACE_ITER(FUNCTION) is about to be set */ + if (function_enabled || (!set && !(tr->trace_flags & TRACE_ITER(FUNCTION)))) return 0; if (graph) @@ -261,7 +277,7 @@ static void unregister_wakeup_function(struct trace_array *tr, int graph) static int wakeup_function_set(struct trace_array *tr, u32 mask, int set) { - if (!(mask & TRACE_ITER_FUNCTION)) + if (!(mask & TRACE_ITER(FUNCTION))) return 0; if (set) @@ -305,10 +321,10 @@ __trace_function(struct trace_array *tr, if (is_graph(tr)) trace_graph_function(tr, ip, parent_ip, trace_ctx); else - trace_function(tr, ip, parent_ip, trace_ctx); + trace_function(tr, ip, parent_ip, trace_ctx, NULL); } -static int wakeup_flag_changed(struct trace_array *tr, u32 mask, int set) +static int wakeup_flag_changed(struct trace_array *tr, u64 mask, int set) { struct tracer *tracer = tr->current_trace; @@ -316,7 +332,7 @@ static int wakeup_flag_changed(struct trace_array *tr, u32 mask, int set) return 0; #ifdef CONFIG_FUNCTION_GRAPH_TRACER - if (mask & TRACE_ITER_DISPLAY_GRAPH) + if (mask & TRACE_ITER(DISPLAY_GRAPH)) return wakeup_display_graph(tr, set); #endif @@ -374,7 +390,6 @@ tracing_sched_switch_trace(struct trace_array *tr, struct task_struct *next, unsigned int trace_ctx) { - struct trace_event_call *call = &event_context_switch; struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct ctx_switch_entry *entry; @@ -392,8 +407,7 @@ tracing_sched_switch_trace(struct trace_array *tr, entry->next_state = task_state_index(next); entry->next_cpu = task_cpu(next); - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } static void @@ -402,7 +416,6 @@ tracing_sched_wakeup_trace(struct trace_array *tr, struct task_struct *curr, unsigned int trace_ctx) { - struct trace_event_call *call = &event_wakeup; struct ring_buffer_event *event; struct ctx_switch_entry *entry; struct trace_buffer *buffer = tr->array_buffer.buffer; @@ -420,8 +433,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr, entry->next_state = task_state_index(wakee); entry->next_cpu = task_cpu(wakee); - if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } static void notrace @@ -455,7 +467,7 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, /* disable local data, not wakeup_cpu data */ cpu = raw_smp_processor_id(); - disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); + disabled = local_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); if (likely(disabled != 1)) goto out; @@ -492,7 +504,7 @@ out_unlock: arch_spin_unlock(&wakeup_lock); local_irq_restore(flags); out: - atomic_dec(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); + local_dec(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); } static void __wakeup_reset(struct trace_array *tr) @@ -543,11 +555,11 @@ probe_wakeup(void *ignore, struct task_struct *p) * - wakeup_dl handles tasks belonging to sched_dl class only. */ if (tracing_dl || (wakeup_dl && !dl_task(p)) || - (wakeup_rt && !dl_task(p) && !rt_task(p)) || + (wakeup_rt && !rt_or_dl_task(p)) || (!dl_task(p) && (p->prio >= wakeup_prio || p->prio >= current->prio))) return; - disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); + disabled = local_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); if (unlikely(disabled != 1)) goto out; @@ -594,7 +606,7 @@ probe_wakeup(void *ignore, struct task_struct *p) out_locked: arch_spin_unlock(&wakeup_lock); out: - atomic_dec(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); + local_dec(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); } static void start_wakeup_tracer(struct trace_array *tr) @@ -669,8 +681,8 @@ static int __wakeup_tracer_init(struct trace_array *tr) save_flags = tr->trace_flags; /* non overwrite screws up the latency tracers */ - set_tracer_flag(tr, TRACE_ITER_OVERWRITE, 1); - set_tracer_flag(tr, TRACE_ITER_LATENCY_FMT, 1); + set_tracer_flag(tr, TRACE_ITER(OVERWRITE), 1); + set_tracer_flag(tr, TRACE_ITER(LATENCY_FMT), 1); tr->max_latency = 0; wakeup_trace = tr; @@ -713,15 +725,15 @@ static int wakeup_dl_tracer_init(struct trace_array *tr) static void wakeup_tracer_reset(struct trace_array *tr) { - int lat_flag = save_flags & TRACE_ITER_LATENCY_FMT; - int overwrite_flag = save_flags & TRACE_ITER_OVERWRITE; + int lat_flag = save_flags & TRACE_ITER(LATENCY_FMT); + int overwrite_flag = save_flags & TRACE_ITER(OVERWRITE); stop_wakeup_tracer(tr); /* make sure we put back any tasks we are tracing */ wakeup_reset(tr); - set_tracer_flag(tr, TRACE_ITER_LATENCY_FMT, lat_flag); - set_tracer_flag(tr, TRACE_ITER_OVERWRITE, overwrite_flag); + set_tracer_flag(tr, TRACE_ITER(LATENCY_FMT), lat_flag); + set_tracer_flag(tr, TRACE_ITER(OVERWRITE), overwrite_flag); ftrace_reset_array_ops(tr); wakeup_busy = false; } diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index ff0536cea968..d88c44f1dfa5 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -17,6 +17,7 @@ static inline int trace_valid_entry(struct trace_entry *entry) case TRACE_PRINT: case TRACE_BRANCH: case TRACE_GRAPH_ENT: + case TRACE_GRAPH_RETADDR_ENT: case TRACE_GRAPH_RET: return 1; } @@ -756,19 +757,284 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) #ifdef CONFIG_FUNCTION_GRAPH_TRACER +#ifdef CONFIG_DYNAMIC_FTRACE + +#define CHAR_NUMBER 123 +#define SHORT_NUMBER 12345 +#define WORD_NUMBER 1234567890 +#define LONG_NUMBER 1234567890123456789LL +#define ERRSTR_BUFLEN 128 + +struct fgraph_fixture { + struct fgraph_ops gops; + int store_size; + const char *store_type_name; + char error_str_buf[ERRSTR_BUFLEN]; + char *error_str; +}; + +static __init int store_entry(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + struct fgraph_fixture *fixture = container_of(gops, struct fgraph_fixture, gops); + const char *type = fixture->store_type_name; + int size = fixture->store_size; + void *p; + + p = fgraph_reserve_data(gops->idx, size); + if (!p) { + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, + "Failed to reserve %s\n", type); + return 0; + } + + switch (size) { + case 1: + *(char *)p = CHAR_NUMBER; + break; + case 2: + *(short *)p = SHORT_NUMBER; + break; + case 4: + *(int *)p = WORD_NUMBER; + break; + case 8: + *(long long *)p = LONG_NUMBER; + break; + } + + return 1; +} + +static __init void store_return(struct ftrace_graph_ret *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) +{ + struct fgraph_fixture *fixture = container_of(gops, struct fgraph_fixture, gops); + const char *type = fixture->store_type_name; + long long expect = 0; + long long found = -1; + int size; + char *p; + + p = fgraph_retrieve_data(gops->idx, &size); + if (!p) { + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, + "Failed to retrieve %s\n", type); + return; + } + if (fixture->store_size > size) { + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, + "Retrieved size %d is smaller than expected %d\n", + size, (int)fixture->store_size); + return; + } + + switch (fixture->store_size) { + case 1: + expect = CHAR_NUMBER; + found = *(char *)p; + break; + case 2: + expect = SHORT_NUMBER; + found = *(short *)p; + break; + case 4: + expect = WORD_NUMBER; + found = *(int *)p; + break; + case 8: + expect = LONG_NUMBER; + found = *(long long *)p; + break; + } + + if (found != expect) { + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, + "%s returned not %lld but %lld\n", type, expect, found); + return; + } + fixture->error_str = NULL; +} + +static int __init init_fgraph_fixture(struct fgraph_fixture *fixture) +{ + char *func_name; + int len; + + snprintf(fixture->error_str_buf, ERRSTR_BUFLEN, + "Failed to execute storage %s\n", fixture->store_type_name); + fixture->error_str = fixture->error_str_buf; + + func_name = "*" __stringify(DYN_FTRACE_TEST_NAME); + len = strlen(func_name); + + return ftrace_set_filter(&fixture->gops.ops, func_name, len, 1); +} + +/* Test fgraph storage for each size */ +static int __init test_graph_storage_single(struct fgraph_fixture *fixture) +{ + int size = fixture->store_size; + int ret; + + pr_cont("PASSED\n"); + pr_info("Testing fgraph storage of %d byte%s: ", size, str_plural(size)); + + ret = init_fgraph_fixture(fixture); + if (ret && ret != -ENODEV) { + pr_cont("*Could not set filter* "); + return -1; + } + + ret = register_ftrace_graph(&fixture->gops); + if (ret) { + pr_warn("Failed to init store_bytes fgraph tracing\n"); + return -1; + } + + DYN_FTRACE_TEST_NAME(); + + unregister_ftrace_graph(&fixture->gops); + + if (fixture->error_str) { + pr_cont("*** %s ***", fixture->error_str); + return -1; + } + + return 0; +} + +static struct fgraph_fixture store_bytes[4] __initdata = { + [0] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 1, + .store_type_name = "byte", + }, + [1] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 2, + .store_type_name = "short", + }, + [2] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 4, + .store_type_name = "word", + }, + [3] = { + .gops = { + .entryfunc = store_entry, + .retfunc = store_return, + }, + .store_size = 8, + .store_type_name = "long long", + }, +}; + +static __init int test_graph_storage_multi(void) +{ + struct fgraph_fixture *fixture; + bool printed = false; + int i, j, ret; + + pr_cont("PASSED\n"); + pr_info("Testing multiple fgraph storage on a function: "); + + for (i = 0; i < ARRAY_SIZE(store_bytes); i++) { + fixture = &store_bytes[i]; + ret = init_fgraph_fixture(fixture); + if (ret && ret != -ENODEV) { + pr_cont("*Could not set filter* "); + printed = true; + goto out2; + } + } + + for (j = 0; j < ARRAY_SIZE(store_bytes); j++) { + fixture = &store_bytes[j]; + ret = register_ftrace_graph(&fixture->gops); + if (ret) { + pr_warn("Failed to init store_bytes fgraph tracing\n"); + printed = true; + goto out1; + } + } + + DYN_FTRACE_TEST_NAME(); +out1: + while (--j >= 0) { + fixture = &store_bytes[j]; + unregister_ftrace_graph(&fixture->gops); + + if (fixture->error_str && !printed) { + pr_cont("*** %s ***", fixture->error_str); + printed = true; + } + } +out2: + while (--i >= 0) { + fixture = &store_bytes[i]; + ftrace_free_filter(&fixture->gops.ops); + + if (fixture->error_str && !printed) { + pr_cont("*** %s ***", fixture->error_str); + printed = true; + } + } + return printed ? -1 : 0; +} + +/* Test the storage passed across function_graph entry and return */ +static __init int test_graph_storage(void) +{ + int ret; + + ret = test_graph_storage_single(&store_bytes[0]); + if (ret) + return ret; + ret = test_graph_storage_single(&store_bytes[1]); + if (ret) + return ret; + ret = test_graph_storage_single(&store_bytes[2]); + if (ret) + return ret; + ret = test_graph_storage_single(&store_bytes[3]); + if (ret) + return ret; + ret = test_graph_storage_multi(); + if (ret) + return ret; + return 0; +} +#else +static inline int test_graph_storage(void) { return 0; } +#endif /* CONFIG_DYNAMIC_FTRACE */ + /* Maximum number of functions to trace before diagnosing a hang */ #define GRAPH_MAX_FUNC_TEST 100000000 static unsigned int graph_hang_thresh; /* Wrap the real function entry probe to avoid possible hanging */ -static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace) +static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace, + struct fgraph_ops *gops, + struct ftrace_regs *fregs) { /* This is harmlessly racy, we want to approximately detect a hang */ if (unlikely(++graph_hang_thresh > GRAPH_MAX_FUNC_TEST)) { ftrace_graph_stop(); printk(KERN_WARNING "BUG: Function graph tracer hang!\n"); - if (ftrace_dump_on_oops) { + if (ftrace_dump_on_oops_enabled()) { ftrace_dump(DUMP_ALL); /* ftrace_dump() disables tracing */ tracing_on(); @@ -776,7 +1042,7 @@ static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace) return 0; } - return trace_graph_entry(trace); + return trace_graph_entry(trace, gops, fregs); } static struct fgraph_ops fgraph_ops __initdata = { @@ -785,14 +1051,7 @@ static struct fgraph_ops fgraph_ops __initdata = { }; #ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS -#ifndef CALL_DEPTH_ACCOUNT -#define CALL_DEPTH_ACCOUNT "" -#endif - -noinline __noclone static void trace_direct_tramp(void) -{ - asm(CALL_DEPTH_ACCOUNT); -} +static struct ftrace_ops direct; #endif /* @@ -819,7 +1078,7 @@ trace_selftest_startup_function_graph(struct tracer *trace, * to detect and recover from possible hangs */ tracing_reset_online_cpus(&tr->array_buffer); - set_graph_array(tr); + fgraph_ops.private = tr; ret = register_ftrace_graph(&fgraph_ops); if (ret) { warn_failed_init_tracer(trace, ret); @@ -855,8 +1114,14 @@ trace_selftest_startup_function_graph(struct tracer *trace, } #ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + /* + * These tests can take some time to run. Make sure on non PREEMPT + * kernels, we do not trigger the softlockup detector. + */ + cond_resched(); + tracing_reset_online_cpus(&tr->array_buffer); - set_graph_array(tr); + fgraph_ops.private = tr; /* * Some archs *cough*PowerPC*cough* add characters to the @@ -870,11 +1135,14 @@ trace_selftest_startup_function_graph(struct tracer *trace, * Register direct function together with graph tracer * and make sure we get graph trace. */ - ret = register_ftrace_direct((unsigned long) DYN_FTRACE_TEST_NAME, - (unsigned long) trace_direct_tramp); + ftrace_set_filter_ip(&direct, (unsigned long)DYN_FTRACE_TEST_NAME, 0, 0); + ret = register_ftrace_direct(&direct, + (unsigned long)ftrace_stub_direct_tramp); if (ret) goto out; + cond_resched(); + ret = register_ftrace_graph(&fgraph_ops); if (ret) { warn_failed_init_tracer(trace, ret); @@ -891,11 +1159,14 @@ trace_selftest_startup_function_graph(struct tracer *trace, unregister_ftrace_graph(&fgraph_ops); - ret = unregister_ftrace_direct((unsigned long) DYN_FTRACE_TEST_NAME, - (unsigned long) trace_direct_tramp); + ret = unregister_ftrace_direct(&direct, + (unsigned long)ftrace_stub_direct_tramp, + true); if (ret) goto out; + cond_resched(); + tracing_start(); if (!ret && !count) { @@ -907,6 +1178,8 @@ trace_selftest_startup_function_graph(struct tracer *trace, ftrace_set_global_filter(NULL, 0, 1); #endif + ret = test_graph_storage(); + /* Don't test dynamic tracing, the function tracer already did */ out: /* Stop it if we failed */ @@ -1216,7 +1489,7 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) /* reset the max latency */ tr->max_latency = 0; - while (p->on_rq) { + while (task_is_runnable(p)) { /* * Sleep to make sure the -deadline thread is asleep too. * On virtual machines we can't rely on timings, diff --git a/kernel/trace/trace_seq.c b/kernel/trace/trace_seq.c index 9c90b3a7dce2..32684ef4fb9d 100644 --- a/kernel/trace/trace_seq.c +++ b/kernel/trace/trace_seq.c @@ -13,12 +13,9 @@ * trace_seq_init() more than once to reset the trace_seq to start * from scratch. * - * The buffer size is currently PAGE_SIZE, although it may become dynamic - * in the future. - * * A write to the buffer will either succeed or fail. That is, unlike * sprintf() there will not be a partial write (well it may write into - * the buffer but it wont update the pointers). This allows users to + * the buffer but it won't update the pointers). This allows users to * try to write something into the trace_seq buffer and if it fails * they can flush it and try again. * @@ -131,6 +128,7 @@ EXPORT_SYMBOL_GPL(trace_seq_bitmask); * trace_seq_vprintf - sequence printing of trace information * @s: trace sequence descriptor * @fmt: printf format string + * @args: Arguments for the format string * * The tracer may use either sequence operations or its own * copy to user routines. To simplify formatting of a trace @@ -369,8 +367,12 @@ EXPORT_SYMBOL_GPL(trace_seq_path); */ int trace_seq_to_user(struct trace_seq *s, char __user *ubuf, int cnt) { + int ret; __trace_seq_init(s); - return seq_buf_to_user(&s->seq, ubuf, cnt); + ret = seq_buf_to_user(&s->seq, ubuf, s->readpos, cnt); + if (ret > 0) + s->readpos += ret; + return ret; } EXPORT_SYMBOL_GPL(trace_seq_to_user); @@ -403,3 +405,26 @@ int trace_seq_hex_dump(struct trace_seq *s, const char *prefix_str, return 1; } EXPORT_SYMBOL(trace_seq_hex_dump); + +/* + * trace_seq_acquire - acquire seq buffer with size len + * @s: trace sequence descriptor + * @len: size of buffer to be acquired + * + * acquire buffer with size of @len from trace_seq for output usage, + * user can fill string into that buffer. + * + * Returns start address of acquired buffer. + * + * it allow multiple usage in one trace output function call. + */ +char *trace_seq_acquire(struct trace_seq *s, unsigned int len) +{ + char *ret = trace_seq_buffer_ptr(s); + + if (!WARN_ON_ONCE(seq_buf_buffer_left(&s->seq) < len)) + seq_buf_commit(&s->seq, len); + + return ret; +} +EXPORT_SYMBOL(trace_seq_acquire); diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 5a48dba912ea..0aa2514a6593 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -32,7 +32,7 @@ static arch_spinlock_t stack_trace_max_lock = DEFINE_PER_CPU(int, disable_stack_tracer); static DEFINE_MUTEX(stack_sysctl_mutex); -int stack_tracer_enabled; +static int stack_tracer_enabled; static void print_max_stack(void) { @@ -514,26 +514,24 @@ static const struct file_operations stack_trace_filter_fops = { #endif /* CONFIG_DYNAMIC_FTRACE */ int -stack_trace_sysctl(struct ctl_table *table, int write, void *buffer, +stack_trace_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int was_enabled; int ret; - mutex_lock(&stack_sysctl_mutex); + guard(mutex)(&stack_sysctl_mutex); was_enabled = !!stack_tracer_enabled; ret = proc_dointvec(table, write, buffer, lenp, ppos); if (ret || !write || (was_enabled == !!stack_tracer_enabled)) - goto out; + return ret; if (stack_tracer_enabled) register_ftrace_function(&trace_ops); else unregister_ftrace_function(&trace_ops); - out: - mutex_unlock(&stack_sysctl_mutex); return ret; } @@ -544,7 +542,7 @@ static __init int enable_stacktrace(char *str) int len; if ((len = str_has_prefix(str, "_filter="))) - strncpy(stack_trace_filter_buf, str + len, COMMAND_LINE_SIZE); + strscpy(stack_trace_filter_buf, str + len); stack_tracer_enabled = 1; return 1; @@ -580,3 +578,23 @@ static __init int stack_trace_init(void) } device_initcall(stack_trace_init); + + +static const struct ctl_table trace_stack_sysctl_table[] = { + { + .procname = "stack_tracer_enabled", + .data = &stack_tracer_enabled, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = stack_trace_sysctl, + }, +}; + +static int __init init_trace_stack_sysctls(void) +{ + register_sysctl_init("kernel", trace_stack_sysctl_table); + return 0; +} +subsys_initcall(init_trace_stack_sysctls); + + diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c index bb247beec447..b3b5586f104d 100644 --- a/kernel/trace/trace_stat.c +++ b/kernel/trace/trace_stat.c @@ -128,7 +128,7 @@ static int stat_seq_init(struct stat_session *session) int ret = 0; int i; - mutex_lock(&session->stat_mutex); + guard(mutex)(&session->stat_mutex); __reset_stat_session(session); if (!ts->stat_cmp) @@ -136,11 +136,11 @@ static int stat_seq_init(struct stat_session *session) stat = ts->stat_start(ts); if (!stat) - goto exit; + return 0; ret = insert_stat(root, stat, ts->stat_cmp); if (ret) - goto exit; + return ret; /* * Iterate over the tracer stat entries and store them in an rbtree. @@ -157,13 +157,10 @@ static int stat_seq_init(struct stat_session *session) goto exit_free_rbtree; } -exit: - mutex_unlock(&session->stat_mutex); return ret; exit_free_rbtree: __reset_stat_session(session); - mutex_unlock(&session->stat_mutex); return ret; } @@ -308,7 +305,7 @@ static int init_stat_file(struct stat_session *session) int register_stat_tracer(struct tracer_stat *trace) { struct stat_session *session, *node; - int ret = -EINVAL; + int ret; if (!trace) return -EINVAL; @@ -316,18 +313,18 @@ int register_stat_tracer(struct tracer_stat *trace) if (!trace->stat_start || !trace->stat_next || !trace->stat_show) return -EINVAL; + guard(mutex)(&all_stat_sessions_mutex); + /* Already registered? */ - mutex_lock(&all_stat_sessions_mutex); list_for_each_entry(node, &all_stat_sessions, session_list) { if (node->ts == trace) - goto out; + return -EINVAL; } - ret = -ENOMEM; /* Init the session */ session = kzalloc(sizeof(*session), GFP_KERNEL); if (!session) - goto out; + return -ENOMEM; session->ts = trace; INIT_LIST_HEAD(&session->session_list); @@ -336,16 +333,13 @@ int register_stat_tracer(struct tracer_stat *trace) ret = init_stat_file(session); if (ret) { destroy_session(session); - goto out; + return ret; } - ret = 0; /* Register */ list_add_tail(&session->session_list, &all_stat_sessions); - out: - mutex_unlock(&all_stat_sessions_mutex); - return ret; + return 0; } void unregister_stat_tracer(struct tracer_stat *trace) diff --git a/kernel/trace/trace_synth.h b/kernel/trace/trace_synth.h index b29595fe3ac5..43f6fb6078db 100644 --- a/kernel/trace/trace_synth.h +++ b/kernel/trace/trace_synth.h @@ -18,6 +18,7 @@ struct synth_field { bool is_signed; bool is_string; bool is_dynamic; + bool is_stack; }; struct synth_event { diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 942ddbdace4a..e96d0063cbcf 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include <trace/syscall.h> #include <trace/events/syscalls.h> +#include <linux/kernel_stat.h> #include <linux/syscalls.h> #include <linux/slab.h> #include <linux/kernel.h> @@ -123,6 +124,119 @@ const char *get_syscall_name(int syscall) return entry->name; } +/* Added to user strings or arrays when max limit is reached */ +#define EXTRA "..." + +static void get_dynamic_len_ptr(struct syscall_trace_enter *trace, + struct syscall_metadata *entry, + int *offset_p, int *len_p, unsigned char **ptr_p) +{ + unsigned char *ptr; + int offset = *offset_p; + int val; + + /* This arg points to a user space string */ + ptr = (void *)trace->args + sizeof(long) * entry->nb_args + offset; + val = *(int *)ptr; + + /* The value is a dynamic string (len << 16 | offset) */ + ptr = (void *)trace + (val & 0xffff); + *len_p = val >> 16; + offset += 4; + + *ptr_p = ptr; + *offset_p = offset; +} + +static enum print_line_t +sys_enter_openat_print(struct syscall_trace_enter *trace, struct syscall_metadata *entry, + struct trace_seq *s, struct trace_event *event) +{ + unsigned char *ptr; + int offset = 0; + int bits, len; + bool done = false; + static const struct trace_print_flags __flags[] = + { + { O_TMPFILE, "O_TMPFILE" }, + { O_WRONLY, "O_WRONLY" }, + { O_RDWR, "O_RDWR" }, + { O_CREAT, "O_CREAT" }, + { O_EXCL, "O_EXCL" }, + { O_NOCTTY, "O_NOCTTY" }, + { O_TRUNC, "O_TRUNC" }, + { O_APPEND, "O_APPEND" }, + { O_NONBLOCK, "O_NONBLOCK" }, + { O_DSYNC, "O_DSYNC" }, + { O_DIRECT, "O_DIRECT" }, + { O_LARGEFILE, "O_LARGEFILE" }, + { O_DIRECTORY, "O_DIRECTORY" }, + { O_NOFOLLOW, "O_NOFOLLOW" }, + { O_NOATIME, "O_NOATIME" }, + { O_CLOEXEC, "O_CLOEXEC" }, + { -1, NULL } + }; + + trace_seq_printf(s, "%s(", entry->name); + + for (int i = 0; !done && i < entry->nb_args; i++) { + + if (trace_seq_has_overflowed(s)) + goto end; + + if (i) + trace_seq_puts(s, ", "); + + switch (i) { + case 2: + bits = trace->args[2]; + + trace_seq_puts(s, "flags: "); + + /* No need to show mode when not creating the file */ + if (!(bits & (O_CREAT|O_TMPFILE))) + done = true; + + if (!(bits & O_ACCMODE)) { + if (!bits) { + trace_seq_puts(s, "O_RDONLY"); + continue; + } + trace_seq_puts(s, "O_RDONLY|"); + } + + trace_print_flags_seq(s, "|", bits, __flags); + /* + * trace_print_flags_seq() adds a '\0' to the + * buffer, but this needs to append more to the seq. + */ + if (!trace_seq_has_overflowed(s)) + trace_seq_pop(s); + + continue; + case 3: + trace_seq_printf(s, "%s: 0%03o", entry->args[i], + (unsigned int)trace->args[i]); + continue; + } + + trace_seq_printf(s, "%s: %lu", entry->args[i], + trace->args[i]); + + if (!(BIT(i) & entry->user_mask)) + continue; + + get_dynamic_len_ptr(trace, entry, &offset, &len, &ptr); + trace_seq_printf(s, " \"%.*s\"", len, ptr); + } + + trace_seq_putc(s, ')'); +end: + trace_seq_putc(s, '\n'); + + return trace_handle_return(s); +} + static enum print_line_t print_syscall_enter(struct trace_iterator *iter, int flags, struct trace_event *event) @@ -132,7 +246,9 @@ print_syscall_enter(struct trace_iterator *iter, int flags, struct trace_entry *ent = iter->ent; struct syscall_trace_enter *trace; struct syscall_metadata *entry; - int i, syscall; + int i, syscall, val, len; + unsigned char *ptr; + int offset = 0; trace = (typeof(trace))ent; syscall = trace->nr; @@ -146,21 +262,80 @@ print_syscall_enter(struct trace_iterator *iter, int flags, goto end; } + switch (entry->syscall_nr) { + case __NR_openat: + if (!tr || !(tr->trace_flags & TRACE_ITER(VERBOSE))) + return sys_enter_openat_print(trace, entry, s, event); + break; + default: + break; + } + trace_seq_printf(s, "%s(", entry->name); for (i = 0; i < entry->nb_args; i++) { + bool printable = false; + char *str; if (trace_seq_has_overflowed(s)) goto end; + if (i) + trace_seq_puts(s, ", "); + /* parameter types */ - if (tr && tr->trace_flags & TRACE_ITER_VERBOSE) + if (tr && tr->trace_flags & TRACE_ITER(VERBOSE)) trace_seq_printf(s, "%s ", entry->types[i]); /* parameter values */ - trace_seq_printf(s, "%s: %lx%s", entry->args[i], - trace->args[i], - i == entry->nb_args - 1 ? "" : ", "); + if (trace->args[i] < 10) + trace_seq_printf(s, "%s: %lu", entry->args[i], + trace->args[i]); + else + trace_seq_printf(s, "%s: 0x%lx", entry->args[i], + trace->args[i]); + + if (!(BIT(i) & entry->user_mask)) + continue; + + get_dynamic_len_ptr(trace, entry, &offset, &len, &ptr); + + if (entry->user_arg_size < 0 || entry->user_arg_is_str) { + trace_seq_printf(s, " \"%.*s\"", len, ptr); + continue; + } + + val = trace->args[entry->user_arg_size]; + + str = ptr; + trace_seq_puts(s, " ("); + for (int x = 0; x < len; x++, ptr++) { + if (isascii(*ptr) && isprint(*ptr)) + printable = true; + if (x) + trace_seq_putc(s, ':'); + trace_seq_printf(s, "%02x", *ptr); + } + if (len < val) + trace_seq_printf(s, ", %s", EXTRA); + + trace_seq_putc(s, ')'); + + /* If nothing is printable, don't bother printing anything */ + if (!printable) + continue; + + trace_seq_puts(s, " \""); + for (int x = 0; x < len; x++) { + if (isascii(str[x]) && isprint(str[x])) + trace_seq_putc(s, str[x]); + else + trace_seq_putc(s, '.'); + } + if (len < val) + trace_seq_printf(s, "\"%s", EXTRA); + else + trace_seq_putc(s, '"'); } trace_seq_putc(s, ')'); @@ -206,26 +381,107 @@ print_syscall_exit(struct trace_iterator *iter, int flags, .size = sizeof(_type), .align = __alignof__(_type), \ .is_signed = is_signed_type(_type), .filter_type = FILTER_OTHER } +/* When len=0, we just calculate the needed length */ +#define LEN_OR_ZERO (len ? len - pos : 0) + +static int __init +sys_enter_openat_print_fmt(struct syscall_metadata *entry, char *buf, int len) +{ + int pos = 0; + + pos += snprintf(buf + pos, LEN_OR_ZERO, + "\"dfd: 0x%%08lx, filename: 0x%%08lx \\\"%%s\\\", flags: %%s%%s, mode: 0%%03o\","); + pos += snprintf(buf + pos, LEN_OR_ZERO, + " ((unsigned long)(REC->dfd)),"); + pos += snprintf(buf + pos, LEN_OR_ZERO, + " ((unsigned long)(REC->filename)),"); + pos += snprintf(buf + pos, LEN_OR_ZERO, + " __get_str(__filename_val),"); + pos += snprintf(buf + pos, LEN_OR_ZERO, + " (REC->flags & ~3) && !(REC->flags & 3) ? \"O_RDONLY|\" : \"\", "); + pos += snprintf(buf + pos, LEN_OR_ZERO, + " REC->flags ? __print_flags(REC->flags, \"|\", "); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_WRONLY\" }, ", O_WRONLY); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_RDWR\" }, ", O_RDWR); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_CREAT\" }, ", O_CREAT); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_EXCL\" }, ", O_EXCL); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_NOCTTY\" }, ", O_NOCTTY); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_TRUNC\" }, ", O_TRUNC); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_APPEND\" }, ", O_APPEND); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_NONBLOCK\" }, ", O_NONBLOCK); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_DSYNC\" }, ", O_DSYNC); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_DIRECT\" }, ", O_DIRECT); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_LARGEFILE\" }, ", O_LARGEFILE); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_DIRECTORY\" }, ", O_DIRECTORY); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_NOFOLLOW\" }, ", O_NOFOLLOW); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_NOATIME\" }, ", O_NOATIME); + pos += snprintf(buf + pos, LEN_OR_ZERO, + "{ 0x%x, \"O_CLOEXEC\" }) : \"O_RDONLY\", ", O_CLOEXEC); + + pos += snprintf(buf + pos, LEN_OR_ZERO, + " ((unsigned long)(REC->mode))"); + return pos; +} + static int __init __set_enter_print_fmt(struct syscall_metadata *entry, char *buf, int len) { + bool is_string = entry->user_arg_is_str; int i; int pos = 0; - /* When len=0, we just calculate the needed length */ -#define LEN_OR_ZERO (len ? len - pos : 0) + switch (entry->syscall_nr) { + case __NR_openat: + return sys_enter_openat_print_fmt(entry, buf, len); + default: + break; + } pos += snprintf(buf + pos, LEN_OR_ZERO, "\""); for (i = 0; i < entry->nb_args; i++) { - pos += snprintf(buf + pos, LEN_OR_ZERO, "%s: 0x%%0%zulx%s", - entry->args[i], sizeof(unsigned long), - i == entry->nb_args - 1 ? "" : ", "); + if (i) + pos += snprintf(buf + pos, LEN_OR_ZERO, ", "); + pos += snprintf(buf + pos, LEN_OR_ZERO, "%s: 0x%%0%zulx", + entry->args[i], sizeof(unsigned long)); + + if (!(BIT(i) & entry->user_mask)) + continue; + + /* Add the format for the user space string or array */ + if (entry->user_arg_size < 0 || is_string) + pos += snprintf(buf + pos, LEN_OR_ZERO, " \\\"%%s\\\""); + else + pos += snprintf(buf + pos, LEN_OR_ZERO, " (%%s)"); } pos += snprintf(buf + pos, LEN_OR_ZERO, "\""); for (i = 0; i < entry->nb_args; i++) { pos += snprintf(buf + pos, LEN_OR_ZERO, ", ((unsigned long)(REC->%s))", entry->args[i]); + if (!(BIT(i) & entry->user_mask)) + continue; + /* The user space data for arg has name __<arg>_val */ + if (entry->user_arg_size < 0 || is_string) { + pos += snprintf(buf + pos, LEN_OR_ZERO, ", __get_str(__%s_val)", + entry->args[i]); + } else { + pos += snprintf(buf + pos, LEN_OR_ZERO, ", __print_dynamic_array(__%s_val, 1)", + entry->args[i]); + } } #undef LEN_OR_ZERO @@ -271,8 +527,11 @@ static int __init syscall_enter_define_fields(struct trace_event_call *call) { struct syscall_trace_enter trace; struct syscall_metadata *meta = call->data; + unsigned long mask; + char *arg; int offset = offsetof(typeof(trace), args); int ret = 0; + int len; int i; for (i = 0; i < meta->nb_args; i++) { @@ -285,9 +544,320 @@ static int __init syscall_enter_define_fields(struct trace_event_call *call) offset += sizeof(unsigned long); } + if (ret || !meta->user_mask) + return ret; + + mask = meta->user_mask; + + while (mask) { + int idx = ffs(mask) - 1; + mask &= ~BIT(idx); + + /* + * User space data is faulted into a temporary buffer and then + * added as a dynamic string or array to the end of the event. + * The user space data name for the arg pointer is + * "__<arg>_val". + */ + len = strlen(meta->args[idx]) + sizeof("___val"); + arg = kmalloc(len, GFP_KERNEL); + if (WARN_ON_ONCE(!arg)) { + meta->user_mask = 0; + return -ENOMEM; + } + + snprintf(arg, len, "__%s_val", meta->args[idx]); + + ret = trace_define_field(call, "__data_loc char[]", + arg, offset, sizeof(int), 0, + FILTER_OTHER); + if (ret) { + kfree(arg); + break; + } + offset += 4; + } return ret; } +/* + * Create a per CPU temporary buffer to copy user space pointers into. + * + * SYSCALL_FAULT_USER_MAX is the amount to copy from user space. + * (defined in kernel/trace/trace.h) + + * SYSCALL_FAULT_ARG_SZ is the amount to copy from user space plus the + * nul terminating byte and possibly appended EXTRA (4 bytes). + * + * SYSCALL_FAULT_BUF_SZ holds the size of the per CPU buffer to use + * to copy memory from user space addresses into that will hold + * 3 args as only 3 args are allowed to be copied from system calls. + */ +#define SYSCALL_FAULT_ARG_SZ (SYSCALL_FAULT_USER_MAX + 1 + 4) +#define SYSCALL_FAULT_MAX_CNT 3 +#define SYSCALL_FAULT_BUF_SZ (SYSCALL_FAULT_ARG_SZ * SYSCALL_FAULT_MAX_CNT) + +/* Use the tracing per CPU buffer infrastructure to copy from user space */ +struct syscall_user_buffer { + struct trace_user_buf_info buf; + struct rcu_head rcu; +}; + +static struct syscall_user_buffer *syscall_buffer; + +static int syscall_fault_buffer_enable(void) +{ + struct syscall_user_buffer *sbuf; + int ret; + + lockdep_assert_held(&syscall_trace_lock); + + if (syscall_buffer) { + trace_user_fault_get(&syscall_buffer->buf); + return 0; + } + + sbuf = kmalloc(sizeof(*sbuf), GFP_KERNEL); + if (!sbuf) + return -ENOMEM; + + ret = trace_user_fault_init(&sbuf->buf, SYSCALL_FAULT_BUF_SZ); + if (ret < 0) { + kfree(sbuf); + return ret; + } + + WRITE_ONCE(syscall_buffer, sbuf); + + return 0; +} + +static void rcu_free_syscall_buffer(struct rcu_head *rcu) +{ + struct syscall_user_buffer *sbuf = + container_of(rcu, struct syscall_user_buffer, rcu); + + trace_user_fault_destroy(&sbuf->buf); + kfree(sbuf); +} + + +static void syscall_fault_buffer_disable(void) +{ + struct syscall_user_buffer *sbuf = syscall_buffer; + + lockdep_assert_held(&syscall_trace_lock); + + if (trace_user_fault_put(&sbuf->buf)) + return; + + WRITE_ONCE(syscall_buffer, NULL); + call_rcu_tasks_trace(&sbuf->rcu, rcu_free_syscall_buffer); +} + +struct syscall_args { + char *ptr_array[SYSCALL_FAULT_MAX_CNT]; + int read[SYSCALL_FAULT_MAX_CNT]; + int uargs; +}; + +static int syscall_copy_user(char *buf, const char __user *ptr, + size_t size, void *data) +{ + struct syscall_args *args = data; + int ret; + + for (int i = 0; i < args->uargs; i++, buf += SYSCALL_FAULT_ARG_SZ) { + ptr = (char __user *)args->ptr_array[i]; + ret = strncpy_from_user(buf, ptr, size); + args->read[i] = ret; + } + return 0; +} + +static int syscall_copy_user_array(char *buf, const char __user *ptr, + size_t size, void *data) +{ + struct syscall_args *args = data; + int ret; + + for (int i = 0; i < args->uargs; i++, buf += SYSCALL_FAULT_ARG_SZ) { + ptr = (char __user *)args->ptr_array[i]; + ret = __copy_from_user(buf, ptr, size); + args->read[i] = ret ? -1 : size; + } + return 0; +} + +static char *sys_fault_user(unsigned int buf_size, + struct syscall_metadata *sys_data, + struct syscall_user_buffer *sbuf, + unsigned long *args, + unsigned int data_size[SYSCALL_FAULT_MAX_CNT]) +{ + trace_user_buf_copy syscall_copy = syscall_copy_user; + unsigned long mask = sys_data->user_mask; + unsigned long size = SYSCALL_FAULT_ARG_SZ - 1; + struct syscall_args sargs; + bool array = false; + char *buffer; + char *buf; + int ret; + int i = 0; + + /* The extra is appended to the user data in the buffer */ + BUILD_BUG_ON(SYSCALL_FAULT_USER_MAX + sizeof(EXTRA) >= + SYSCALL_FAULT_ARG_SZ); + + /* + * If this system call event has a size argument, use + * it to define how much of user space memory to read, + * and read it as an array and not a string. + */ + if (sys_data->user_arg_size >= 0) { + array = true; + size = args[sys_data->user_arg_size]; + if (size > SYSCALL_FAULT_ARG_SZ - 1) + size = SYSCALL_FAULT_ARG_SZ - 1; + syscall_copy = syscall_copy_user_array; + } + + while (mask) { + int idx = ffs(mask) - 1; + mask &= ~BIT(idx); + + if (WARN_ON_ONCE(i == SYSCALL_FAULT_MAX_CNT)) + break; + + /* Get the pointer to user space memory to read */ + sargs.ptr_array[i++] = (char *)args[idx]; + } + + sargs.uargs = i; + + /* Clear the values that are not used */ + for (; i < SYSCALL_FAULT_MAX_CNT; i++) { + data_size[i] = -1; /* Denotes no pointer */ + } + + /* A zero size means do not even try */ + if (!buf_size) + return NULL; + + buffer = trace_user_fault_read(&sbuf->buf, NULL, size, + syscall_copy, &sargs); + if (!buffer) + return NULL; + + buf = buffer; + for (i = 0; i < sargs.uargs; i++, buf += SYSCALL_FAULT_ARG_SZ) { + + ret = sargs.read[i]; + if (ret < 0) + continue; + buf[ret] = '\0'; + + /* For strings, replace any non-printable characters with '.' */ + if (!array) { + for (int x = 0; x < ret; x++) { + if (!isprint(buf[x])) + buf[x] = '.'; + } + + size = min(buf_size, SYSCALL_FAULT_USER_MAX); + + /* + * If the text was truncated due to our max limit, + * add "..." to the string. + */ + if (ret > size) { + strscpy(buf + size, EXTRA, sizeof(EXTRA)); + ret = size + sizeof(EXTRA); + } else { + buf[ret++] = '\0'; + } + } else { + ret = min((unsigned int)ret, buf_size); + } + data_size[i] = ret; + } + + return buffer; +} + +static int +syscall_get_data(struct syscall_metadata *sys_data, unsigned long *args, + char **buffer, int *size, int *user_sizes, int *uargs, + int buf_size) +{ + struct syscall_user_buffer *sbuf; + int i; + + /* If the syscall_buffer is NULL, tracing is being shutdown */ + sbuf = READ_ONCE(syscall_buffer); + if (!sbuf) + return -1; + + *buffer = sys_fault_user(buf_size, sys_data, sbuf, args, user_sizes); + /* + * user_size is the amount of data to append. + * Need to add 4 for the meta field that points to + * the user memory at the end of the event and also + * stores its size. + */ + for (i = 0; i < SYSCALL_FAULT_MAX_CNT; i++) { + if (user_sizes[i] < 0) + break; + *size += user_sizes[i] + 4; + } + /* Save the number of user read arguments of this syscall */ + *uargs = i; + return 0; +} + +static void syscall_put_data(struct syscall_metadata *sys_data, + struct syscall_trace_enter *entry, + char *buffer, int size, int *user_sizes, int uargs) +{ + char *buf = buffer; + void *ptr; + int val; + + /* + * Set the pointer to point to the meta data of the event + * that has information about the stored user space memory. + */ + ptr = (void *)entry->args + sizeof(unsigned long) * sys_data->nb_args; + + /* + * The meta data will store the offset of the user data from + * the beginning of the event. That is after the static arguments + * and the meta data fields. + */ + val = (ptr - (void *)entry) + 4 * uargs; + + for (int i = 0; i < uargs; i++) { + + if (i) + val += user_sizes[i - 1]; + + /* Store the offset and the size into the meta data */ + *(int *)ptr = val | (user_sizes[i] << 16); + + /* Skip the meta data */ + ptr += 4; + } + + for (int i = 0; i < uargs; i++, buf += SYSCALL_FAULT_ARG_SZ) { + /* Nothing to do if the user space was empty or faulted */ + if (!user_sizes[i]) + continue; + + memcpy(ptr, buf, user_sizes[i]); + ptr += user_sizes[i]; + } +} + static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) { struct trace_array *tr = data; @@ -296,15 +866,24 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) struct syscall_metadata *sys_data; struct trace_event_buffer fbuffer; unsigned long args[6]; + char *user_ptr; + int user_sizes[SYSCALL_FAULT_MAX_CNT] = {}; int syscall_nr; - int size; + int size = 0; + int uargs = 0; + bool mayfault; + + /* + * Syscall probe called with preemption enabled, but the ring + * buffer and per-cpu data require preemption to be disabled. + */ + might_fault(); syscall_nr = trace_get_syscall_nr(current, regs); if (syscall_nr < 0 || syscall_nr >= NR_syscalls) return; - /* Here we're inside tp handler's rcu_read_lock_sched (__DO_TRACE) */ - trace_file = rcu_dereference_sched(tr->enter_syscall_files[syscall_nr]); + trace_file = READ_ONCE(tr->enter_syscall_files[syscall_nr]); if (!trace_file) return; @@ -315,7 +894,20 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) if (!sys_data) return; - size = sizeof(*entry) + sizeof(unsigned long) * sys_data->nb_args; + /* Check if this syscall event faults in user space memory */ + mayfault = sys_data->user_mask != 0; + + guard(preempt_notrace)(); + + syscall_get_arguments(current, regs, args); + + if (mayfault) { + if (syscall_get_data(sys_data, args, &user_ptr, + &size, user_sizes, &uargs, tr->syscall_buf_sz) < 0) + return; + } + + size += sizeof(*entry) + sizeof(unsigned long) * sys_data->nb_args; entry = trace_event_buffer_reserve(&fbuffer, trace_file, size); if (!entry) @@ -323,9 +915,12 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) entry = ring_buffer_event_data(fbuffer.event); entry->nr = syscall_nr; - syscall_get_arguments(current, regs, args); + memcpy(entry->args, args, sizeof(unsigned long) * sys_data->nb_args); + if (mayfault) + syscall_put_data(sys_data, entry, user_ptr, size, user_sizes, uargs); + trace_event_buffer_commit(&fbuffer); } @@ -338,12 +933,18 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) struct trace_event_buffer fbuffer; int syscall_nr; + /* + * Syscall probe called with preemption enabled, but the ring + * buffer and per-cpu data require preemption to be disabled. + */ + might_fault(); + guard(preempt_notrace)(); + syscall_nr = trace_get_syscall_nr(current, regs); if (syscall_nr < 0 || syscall_nr >= NR_syscalls) return; - /* Here we're inside tp handler's rcu_read_lock_sched (__DO_TRACE()) */ - trace_file = rcu_dereference_sched(tr->exit_syscall_files[syscall_nr]); + trace_file = READ_ONCE(tr->exit_syscall_files[syscall_nr]); if (!trace_file) return; @@ -368,39 +969,50 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) static int reg_event_syscall_enter(struct trace_event_file *file, struct trace_event_call *call) { + struct syscall_metadata *sys_data = call->data; struct trace_array *tr = file->tr; int ret = 0; int num; - num = ((struct syscall_metadata *)call->data)->syscall_nr; + num = sys_data->syscall_nr; if (WARN_ON_ONCE(num < 0 || num >= NR_syscalls)) return -ENOSYS; - mutex_lock(&syscall_trace_lock); - if (!tr->sys_refcount_enter) + guard(mutex)(&syscall_trace_lock); + if (sys_data->user_mask) { + ret = syscall_fault_buffer_enable(); + if (ret < 0) + return ret; + } + if (!tr->sys_refcount_enter) { ret = register_trace_sys_enter(ftrace_syscall_enter, tr); - if (!ret) { - rcu_assign_pointer(tr->enter_syscall_files[num], file); - tr->sys_refcount_enter++; + if (ret < 0) { + if (sys_data->user_mask) + syscall_fault_buffer_disable(); + return ret; + } } - mutex_unlock(&syscall_trace_lock); - return ret; + WRITE_ONCE(tr->enter_syscall_files[num], file); + tr->sys_refcount_enter++; + return 0; } static void unreg_event_syscall_enter(struct trace_event_file *file, struct trace_event_call *call) { + struct syscall_metadata *sys_data = call->data; struct trace_array *tr = file->tr; int num; - num = ((struct syscall_metadata *)call->data)->syscall_nr; + num = sys_data->syscall_nr; if (WARN_ON_ONCE(num < 0 || num >= NR_syscalls)) return; - mutex_lock(&syscall_trace_lock); + guard(mutex)(&syscall_trace_lock); tr->sys_refcount_enter--; - RCU_INIT_POINTER(tr->enter_syscall_files[num], NULL); + WRITE_ONCE(tr->enter_syscall_files[num], NULL); if (!tr->sys_refcount_enter) unregister_trace_sys_enter(ftrace_syscall_enter, tr); - mutex_unlock(&syscall_trace_lock); + if (sys_data->user_mask) + syscall_fault_buffer_disable(); } static int reg_event_syscall_exit(struct trace_event_file *file, @@ -417,7 +1029,7 @@ static int reg_event_syscall_exit(struct trace_event_file *file, if (!tr->sys_refcount_exit) ret = register_trace_sys_exit(ftrace_syscall_exit, tr); if (!ret) { - rcu_assign_pointer(tr->exit_syscall_files[num], file); + WRITE_ONCE(tr->exit_syscall_files[num], file); tr->sys_refcount_exit++; } mutex_unlock(&syscall_trace_lock); @@ -435,12 +1047,221 @@ static void unreg_event_syscall_exit(struct trace_event_file *file, return; mutex_lock(&syscall_trace_lock); tr->sys_refcount_exit--; - RCU_INIT_POINTER(tr->exit_syscall_files[num], NULL); + WRITE_ONCE(tr->exit_syscall_files[num], NULL); if (!tr->sys_refcount_exit) unregister_trace_sys_exit(ftrace_syscall_exit, tr); mutex_unlock(&syscall_trace_lock); } +/* + * For system calls that reference user space memory that can + * be recorded into the event, set the system call meta data's user_mask + * to the "args" index that points to the user space memory to retrieve. + */ +static void check_faultable_syscall(struct trace_event_call *call, int nr) +{ + struct syscall_metadata *sys_data = call->data; + unsigned long mask; + + /* Only work on entry */ + if (sys_data->enter_event != call) + return; + + sys_data->user_arg_size = -1; + + switch (nr) { + /* user arg 1 with size arg at 2 */ + case __NR_write: +#ifdef __NR_mq_timedsend + case __NR_mq_timedsend: +#endif + case __NR_pwrite64: + sys_data->user_mask = BIT(1); + sys_data->user_arg_size = 2; + break; + /* user arg 0 with size arg at 1 as string */ + case __NR_setdomainname: + case __NR_sethostname: + sys_data->user_mask = BIT(0); + sys_data->user_arg_size = 1; + sys_data->user_arg_is_str = 1; + break; +#ifdef __NR_kexec_file_load + /* user arg 4 with size arg at 3 as string */ + case __NR_kexec_file_load: + sys_data->user_mask = BIT(4); + sys_data->user_arg_size = 3; + sys_data->user_arg_is_str = 1; + break; +#endif + /* user arg at position 0 */ +#ifdef __NR_access + case __NR_access: +#endif + case __NR_acct: + case __NR_chdir: +#ifdef __NR_chown + case __NR_chown: +#endif +#ifdef __NR_chmod + case __NR_chmod: +#endif + case __NR_chroot: +#ifdef __NR_creat + case __NR_creat: +#endif + case __NR_delete_module: + case __NR_execve: + case __NR_fsopen: +#ifdef __NR_lchown + case __NR_lchown: +#endif +#ifdef __NR_open + case __NR_open: +#endif + case __NR_memfd_create: +#ifdef __NR_mkdir + case __NR_mkdir: +#endif +#ifdef __NR_mknod + case __NR_mknod: +#endif + case __NR_mq_open: + case __NR_mq_unlink: +#ifdef __NR_readlink + case __NR_readlink: +#endif +#ifdef __NR_rmdir + case __NR_rmdir: +#endif + case __NR_shmdt: +#ifdef __NR_statfs + case __NR_statfs: +#endif + case __NR_swapon: + case __NR_swapoff: +#ifdef __NR_truncate + case __NR_truncate: +#endif +#ifdef __NR_unlink + case __NR_unlink: +#endif + case __NR_umount2: +#ifdef __NR_utime + case __NR_utime: +#endif +#ifdef __NR_utimes + case __NR_utimes: +#endif + sys_data->user_mask = BIT(0); + break; + /* user arg at position 1 */ + case __NR_execveat: + case __NR_faccessat: + case __NR_faccessat2: + case __NR_finit_module: + case __NR_fchmodat: + case __NR_fchmodat2: + case __NR_fchownat: + case __NR_fgetxattr: + case __NR_flistxattr: + case __NR_fsetxattr: + case __NR_fspick: + case __NR_fremovexattr: +#ifdef __NR_futimesat + case __NR_futimesat: +#endif + case __NR_inotify_add_watch: + case __NR_mkdirat: + case __NR_mknodat: + case __NR_mount_setattr: + case __NR_name_to_handle_at: +#ifdef __NR_newfstatat + case __NR_newfstatat: +#endif + case __NR_openat: + case __NR_openat2: + case __NR_open_tree: + case __NR_open_tree_attr: + case __NR_readlinkat: + case __NR_quotactl: + case __NR_syslog: + case __NR_statx: + case __NR_unlinkat: +#ifdef __NR_utimensat + case __NR_utimensat: +#endif + sys_data->user_mask = BIT(1); + break; + /* user arg at position 2 */ + case __NR_init_module: + case __NR_fsconfig: + sys_data->user_mask = BIT(2); + break; + /* user arg at position 4 */ + case __NR_fanotify_mark: + sys_data->user_mask = BIT(4); + break; + /* 2 user args, 0 and 1 */ + case __NR_add_key: + case __NR_getxattr: + case __NR_lgetxattr: + case __NR_lremovexattr: +#ifdef __NR_link + case __NR_link: +#endif + case __NR_listxattr: + case __NR_llistxattr: + case __NR_lsetxattr: + case __NR_pivot_root: + case __NR_removexattr: +#ifdef __NR_rename + case __NR_rename: +#endif + case __NR_request_key: + case __NR_setxattr: +#ifdef __NR_symlink + case __NR_symlink: +#endif + sys_data->user_mask = BIT(0) | BIT(1); + break; + /* 2 user args, 0 and 2 */ + case __NR_symlinkat: + sys_data->user_mask = BIT(0) | BIT(2); + break; + /* 2 user args, 1 and 3 */ + case __NR_getxattrat: + case __NR_linkat: + case __NR_listxattrat: + case __NR_move_mount: +#ifdef __NR_renameat + case __NR_renameat: +#endif + case __NR_renameat2: + case __NR_removexattrat: + case __NR_setxattrat: + sys_data->user_mask = BIT(1) | BIT(3); + break; + case __NR_mount: /* Just dev_name and dir_name, TODO add type */ + sys_data->user_mask = BIT(0) | BIT(1) | BIT(2); + break; + default: + sys_data->user_mask = 0; + return; + } + + if (sys_data->user_arg_size < 0) + return; + + /* + * The user_arg_size can only be used when the system call + * is reading only a single address from user space. + */ + mask = sys_data->user_mask; + if (WARN_ON(mask & (mask - 1))) + sys_data->user_arg_size = -1; +} + static int __init init_syscall_trace(struct trace_event_call *call) { int id; @@ -453,6 +1274,8 @@ static int __init init_syscall_trace(struct trace_event_call *call) return -ENOSYS; } + check_faultable_syscall(call, num); + if (set_syscall_print_fmt(call) < 0) return -ENOMEM; @@ -555,12 +1378,16 @@ static int perf_call_bpf_enter(struct trace_event_call *call, struct pt_regs *re struct syscall_trace_enter *rec) { struct syscall_tp_t { - unsigned long long regs; - unsigned long syscall_nr; + struct trace_entry ent; + int syscall_nr; unsigned long args[SYSCALL_DEFINE_MAXARGS]; - } param; + } __aligned(8) param; int i; + BUILD_BUG_ON(sizeof(param.ent) < sizeof(void *)); + + /* bpf prog requires 'regs' to be the first member in the ctx (a.k.a. ¶m) */ + perf_fetch_caller_regs(regs); *(struct pt_regs **)¶m = regs; param.syscall_nr = rec->nr; for (i = 0; i < sys_data->nb_args; i++) @@ -572,12 +1399,25 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) { struct syscall_metadata *sys_data; struct syscall_trace_enter *rec; + struct pt_regs *fake_regs; struct hlist_head *head; unsigned long args[6]; bool valid_prog_array; + bool mayfault; + char *user_ptr; + int user_sizes[SYSCALL_FAULT_MAX_CNT] = {}; + int buf_size = CONFIG_TRACE_SYSCALL_BUF_SIZE_DEFAULT; int syscall_nr; int rctx; - int size; + int size = 0; + int uargs = 0; + + /* + * Syscall probe called with preemption enabled, but the ring + * buffer and per-cpu data require preemption to be disabled. + */ + might_fault(); + guard(preempt_notrace)(); syscall_nr = trace_get_syscall_nr(current, regs); if (syscall_nr < 0 || syscall_nr >= NR_syscalls) @@ -589,26 +1429,39 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) if (!sys_data) return; + syscall_get_arguments(current, regs, args); + + /* Check if this syscall event faults in user space memory */ + mayfault = sys_data->user_mask != 0; + + if (mayfault) { + if (syscall_get_data(sys_data, args, &user_ptr, + &size, user_sizes, &uargs, buf_size) < 0) + return; + } + head = this_cpu_ptr(sys_data->enter_event->perf_events); valid_prog_array = bpf_prog_array_valid(sys_data->enter_event); if (!valid_prog_array && hlist_empty(head)) return; /* get the size after alignment with the u32 buffer size field */ - size = sizeof(unsigned long) * sys_data->nb_args + sizeof(*rec); + size += sizeof(unsigned long) * sys_data->nb_args + sizeof(*rec); size = ALIGN(size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - rec = perf_trace_buf_alloc(size, NULL, &rctx); + rec = perf_trace_buf_alloc(size, &fake_regs, &rctx); if (!rec) return; rec->nr = syscall_nr; - syscall_get_arguments(current, regs, args); memcpy(&rec->args, args, sizeof(unsigned long) * sys_data->nb_args); + if (mayfault) + syscall_put_data(sys_data, rec, user_ptr, size, user_sizes, uargs); + if ((valid_prog_array && - !perf_call_bpf_enter(sys_data->enter_event, regs, sys_data, rec)) || + !perf_call_bpf_enter(sys_data->enter_event, fake_regs, sys_data, rec)) || hlist_empty(head)) { perf_swevent_put_recursion_context(rctx); return; @@ -621,47 +1474,59 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) static int perf_sysenter_enable(struct trace_event_call *call) { - int ret = 0; + struct syscall_metadata *sys_data = call->data; int num; + int ret; - num = ((struct syscall_metadata *)call->data)->syscall_nr; + num = sys_data->syscall_nr; - mutex_lock(&syscall_trace_lock); - if (!sys_perf_refcount_enter) + guard(mutex)(&syscall_trace_lock); + if (sys_data->user_mask) { + ret = syscall_fault_buffer_enable(); + if (ret < 0) + return ret; + } + if (!sys_perf_refcount_enter) { ret = register_trace_sys_enter(perf_syscall_enter, NULL); - if (ret) { - pr_info("event trace: Could not activate syscall entry trace point"); - } else { - set_bit(num, enabled_perf_enter_syscalls); - sys_perf_refcount_enter++; + if (ret) { + pr_info("event trace: Could not activate syscall entry trace point"); + if (sys_data->user_mask) + syscall_fault_buffer_disable(); + return ret; + } } - mutex_unlock(&syscall_trace_lock); - return ret; + set_bit(num, enabled_perf_enter_syscalls); + sys_perf_refcount_enter++; + return 0; } static void perf_sysenter_disable(struct trace_event_call *call) { + struct syscall_metadata *sys_data = call->data; int num; - num = ((struct syscall_metadata *)call->data)->syscall_nr; + num = sys_data->syscall_nr; - mutex_lock(&syscall_trace_lock); + guard(mutex)(&syscall_trace_lock); sys_perf_refcount_enter--; clear_bit(num, enabled_perf_enter_syscalls); if (!sys_perf_refcount_enter) unregister_trace_sys_enter(perf_syscall_enter, NULL); - mutex_unlock(&syscall_trace_lock); + if (sys_data->user_mask) + syscall_fault_buffer_disable(); } static int perf_call_bpf_exit(struct trace_event_call *call, struct pt_regs *regs, struct syscall_trace_exit *rec) { struct syscall_tp_t { - unsigned long long regs; - unsigned long syscall_nr; + struct trace_entry ent; + int syscall_nr; unsigned long ret; - } param; + } __aligned(8) param; + /* bpf prog requires 'regs' to be the first member in the ctx (a.k.a. ¶m) */ + perf_fetch_caller_regs(regs); *(struct pt_regs **)¶m = regs; param.syscall_nr = rec->nr; param.ret = rec->ret; @@ -672,12 +1537,20 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) { struct syscall_metadata *sys_data; struct syscall_trace_exit *rec; + struct pt_regs *fake_regs; struct hlist_head *head; bool valid_prog_array; int syscall_nr; int rctx; int size; + /* + * Syscall probe called with preemption enabled, but the ring + * buffer and per-cpu data require preemption to be disabled. + */ + might_fault(); + guard(preempt_notrace)(); + syscall_nr = trace_get_syscall_nr(current, regs); if (syscall_nr < 0 || syscall_nr >= NR_syscalls) return; @@ -697,7 +1570,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) size = ALIGN(sizeof(*rec) + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - rec = perf_trace_buf_alloc(size, NULL, &rctx); + rec = perf_trace_buf_alloc(size, &fake_regs, &rctx); if (!rec) return; @@ -705,7 +1578,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->ret = syscall_get_return_value(current, regs); if ((valid_prog_array && - !perf_call_bpf_exit(sys_data->exit_event, regs, rec)) || + !perf_call_bpf_exit(sys_data->exit_event, fake_regs, rec)) || hlist_empty(head)) { perf_swevent_put_recursion_context(rctx); return; @@ -717,22 +1590,21 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) static int perf_sysexit_enable(struct trace_event_call *call) { - int ret = 0; int num; num = ((struct syscall_metadata *)call->data)->syscall_nr; - mutex_lock(&syscall_trace_lock); - if (!sys_perf_refcount_exit) - ret = register_trace_sys_exit(perf_syscall_exit, NULL); - if (ret) { - pr_info("event trace: Could not activate syscall exit trace point"); - } else { - set_bit(num, enabled_perf_exit_syscalls); - sys_perf_refcount_exit++; + guard(mutex)(&syscall_trace_lock); + if (!sys_perf_refcount_exit) { + int ret = register_trace_sys_exit(perf_syscall_exit, NULL); + if (ret) { + pr_info("event trace: Could not activate syscall exit trace point"); + return ret; + } } - mutex_unlock(&syscall_trace_lock); - return ret; + set_bit(num, enabled_perf_exit_syscalls); + sys_perf_refcount_exit++; + return 0; } static void perf_sysexit_disable(struct trace_event_call *call) @@ -741,12 +1613,11 @@ static void perf_sysexit_disable(struct trace_event_call *call) num = ((struct syscall_metadata *)call->data)->syscall_nr; - mutex_lock(&syscall_trace_lock); + guard(mutex)(&syscall_trace_lock); sys_perf_refcount_exit--; clear_bit(num, enabled_perf_exit_syscalls); if (!sys_perf_refcount_exit) unregister_trace_sys_exit(perf_syscall_exit, NULL); - mutex_unlock(&syscall_trace_lock); } #endif /* CONFIG_PERF_EVENTS */ diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 8d64b6553aed..1b4f32e2b9bd 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -8,16 +8,19 @@ #define pr_fmt(fmt) "trace_uprobe: " fmt #include <linux/bpf-cgroup.h> -#include <linux/security.h> +#include <linux/cleanup.h> #include <linux/ctype.h> +#include <linux/filter.h> #include <linux/module.h> -#include <linux/uaccess.h> -#include <linux/uprobes.h> #include <linux/namei.h> -#include <linux/string.h> +#include <linux/percpu.h> #include <linux/rculist.h> -#include <linux/filter.h> +#include <linux/security.h> +#include <linux/string.h> +#include <linux/uaccess.h> +#include <linux/uprobes.h> +#include "trace.h" #include "trace_dynevent.h" #include "trace_probe.h" #include "trace_probe_tmpl.h" @@ -58,11 +61,11 @@ struct trace_uprobe { struct dyn_event devent; struct uprobe_consumer consumer; struct path path; - struct inode *inode; char *filename; + struct uprobe *uprobe; unsigned long offset; unsigned long ref_ctr_offset; - unsigned long nhit; + unsigned long __percpu *nhits; struct trace_probe tp; }; @@ -88,14 +91,11 @@ static struct trace_uprobe *to_trace_uprobe(struct dyn_event *ev) static int register_uprobe_event(struct trace_uprobe *tu); static int unregister_uprobe_event(struct trace_uprobe *tu); -struct uprobe_dispatch_data { - struct trace_uprobe *tu; - unsigned long bp_addr; -}; - -static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs); +static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs, + __u64 *data); static int uretprobe_dispatcher(struct uprobe_consumer *con, - unsigned long func, struct pt_regs *regs); + unsigned long func, struct pt_regs *regs, + __u64 *data); #ifdef CONFIG_STACK_GROWSUP static unsigned long adjust_stack_addr(unsigned long addr, unsigned int n) @@ -156,7 +156,7 @@ fetch_store_string(unsigned long addr, void *dest, void *base) return -ENOMEM; if (addr == FETCH_TOKEN_COMM) - ret = strlcpy(dst, current->comm, maxlen); + ret = strscpy(dst, current->comm, maxlen); else ret = strncpy_from_user(dst, src, maxlen); if (ret >= 0) { @@ -170,7 +170,8 @@ fetch_store_string(unsigned long addr, void *dest, void *base) */ ret++; *(u32 *)dest = make_data_loc(ret, (void *)dst - base); - } + } else + *(u32 *)dest = make_data_loc(0, (void *)dst - base); return ret; } @@ -215,11 +216,12 @@ static unsigned long translate_user_vaddr(unsigned long file_offset) /* Note that we don't verify it, since the code does not come from user space */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { struct pt_regs *regs = rec; unsigned long val; + int ret; /* 1st stage: get value from context */ switch (code->op) { @@ -235,20 +237,16 @@ process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, case FETCH_OP_RETVAL: val = regs_return_value(regs); break; - case FETCH_OP_IMM: - val = code->immediate; - break; case FETCH_OP_COMM: val = FETCH_TOKEN_COMM; break; - case FETCH_OP_DATA: - val = (unsigned long)code->data; - break; case FETCH_OP_FOFFS: val = translate_user_vaddr(code->immediate); break; default: - return -EILSEQ; + ret = process_common_fetch_insn(code, &val); + if (ret < 0) + return ret; } code++; @@ -344,7 +342,13 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs, bool is_ret) if (!tu) return ERR_PTR(-ENOMEM); - ret = trace_probe_init(&tu->tp, event, group, true); + tu->nhits = alloc_percpu(unsigned long); + if (!tu->nhits) { + ret = -ENOMEM; + goto error; + } + + ret = trace_probe_init(&tu->tp, event, group, true, nargs); if (ret < 0) goto error; @@ -356,6 +360,7 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs, bool is_ret) return tu; error: + free_percpu(tu->nhits); kfree(tu); return ERR_PTR(ret); @@ -369,6 +374,7 @@ static void free_trace_uprobe(struct trace_uprobe *tu) path_put(&tu->path); trace_probe_cleanup(&tu->tp); kfree(tu->filename); + free_percpu(tu->nhits); kfree(tu); } @@ -494,11 +500,11 @@ static int register_trace_uprobe(struct trace_uprobe *tu) struct trace_uprobe *old_tu; int ret; - mutex_lock(&event_mutex); + guard(mutex)(&event_mutex); ret = validate_ref_ctr_offset(tu); if (ret) - goto end; + return ret; /* register as an event */ old_tu = find_probe_event(trace_probe_name(&tu->tp), @@ -507,11 +513,9 @@ static int register_trace_uprobe(struct trace_uprobe *tu) if (is_ret_probe(tu) != is_ret_probe(old_tu)) { trace_probe_log_set_index(0); trace_probe_log_err(0, DIFF_PROBE_TYPE); - ret = -EEXIST; - } else { - ret = append_trace_uprobe(tu, old_tu); + return -EEXIST; } - goto end; + return append_trace_uprobe(tu, old_tu); } ret = register_uprobe_event(tu); @@ -521,31 +525,33 @@ static int register_trace_uprobe(struct trace_uprobe *tu) trace_probe_log_err(0, EVENT_EXIST); } else pr_warn("Failed to register probe event(%d)\n", ret); - goto end; + return ret; } dyn_event_add(&tu->devent, trace_probe_event_call(&tu->tp)); -end: - mutex_unlock(&event_mutex); - return ret; } +DEFINE_FREE(free_trace_uprobe, struct trace_uprobe *, if (_T) free_trace_uprobe(_T)) + /* * Argument syntax: * - Add uprobe: p|r[:[GRP/][EVENT]] PATH:OFFSET[%return][(REF)] [FETCHARGS] */ static int __trace_uprobe_create(int argc, const char **argv) { - struct trace_uprobe *tu; + struct traceprobe_parse_context *ctx __free(traceprobe_parse_context) = NULL; + struct trace_uprobe *tu __free(free_trace_uprobe) = NULL; + const char *trlog __free(trace_probe_log_clear) = NULL; const char *event = NULL, *group = UPROBE_EVENT_SYSTEM; - char *arg, *filename, *rctr, *rctr_end, *tmp; - char buf[MAX_EVENT_NAME_LEN]; - char gbuf[MAX_EVENT_NAME_LEN]; - enum probe_print_type ptype; - struct path path; + struct path path __free(path_put) = {}; unsigned long offset, ref_ctr_offset; + char *filename __free(kfree) = NULL; + char *arg, *rctr, *rctr_end, *tmp; + char *gbuf __free(kfree) = NULL; + char *buf __free(kfree) = NULL; + enum probe_print_type ptype; bool is_return = false; int i, ret; @@ -564,6 +570,14 @@ static int __trace_uprobe_create(int argc, const char **argv) if (argc < 2) return -ECANCELED; + trlog = trace_probe_log_init("trace_uprobe", argc, argv); + + if (argc - 2 > MAX_TRACE_ARGS) { + trace_probe_log_set_index(2); + trace_probe_log_err(0, TOO_MANY_ARGS); + return -E2BIG; + } + if (argv[0][1] == ':') event = &argv[0][2]; @@ -576,26 +590,20 @@ static int __trace_uprobe_create(int argc, const char **argv) /* Find the last occurrence, in case the path contains ':' too. */ arg = strrchr(filename, ':'); - if (!arg || !isdigit(arg[1])) { - kfree(filename); + if (!arg || !isdigit(arg[1])) return -ECANCELED; - } - trace_probe_log_init("trace_uprobe", argc, argv); trace_probe_log_set_index(1); /* filename is the 2nd argument */ *arg++ = '\0'; ret = kern_path(filename, LOOKUP_FOLLOW, &path); if (ret) { trace_probe_log_err(0, FILE_NOT_FOUND); - kfree(filename); - trace_probe_log_clear(); return ret; } if (!d_is_reg(path.dentry)) { trace_probe_log_err(0, NO_REGULAR_FILE); - ret = -EINVAL; - goto fail_address_parse; + return -EINVAL; } /* Parse reference counter offset if specified. */ @@ -603,16 +611,14 @@ static int __trace_uprobe_create(int argc, const char **argv) if (rctr) { rctr_end = strchr(rctr, ')'); if (!rctr_end) { - ret = -EINVAL; rctr_end = rctr + strlen(rctr); trace_probe_log_err(rctr_end - filename, REFCNT_OPEN_BRACE); - goto fail_address_parse; + return -EINVAL; } else if (rctr_end[1] != '\0') { - ret = -EINVAL; trace_probe_log_err(rctr_end + 1 - filename, BAD_REFCNT_SUFFIX); - goto fail_address_parse; + return -EINVAL; } *rctr++ = '\0'; @@ -620,7 +626,7 @@ static int __trace_uprobe_create(int argc, const char **argv) ret = kstrtoul(rctr, 0, &ref_ctr_offset); if (ret) { trace_probe_log_err(rctr - filename, BAD_REFCNT); - goto fail_address_parse; + return ret; } } @@ -632,8 +638,7 @@ static int __trace_uprobe_create(int argc, const char **argv) is_return = true; } else { trace_probe_log_err(tmp - filename, BAD_ADDR_SUFFIX); - ret = -EINVAL; - goto fail_address_parse; + return -EINVAL; } } @@ -641,16 +646,20 @@ static int __trace_uprobe_create(int argc, const char **argv) ret = kstrtoul(arg, 0, &offset); if (ret) { trace_probe_log_err(arg - filename, BAD_UPROBE_OFFS); - goto fail_address_parse; + return ret; } /* setup a probe */ trace_probe_log_set_index(0); if (event) { + gbuf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!gbuf) + return -ENOMEM; + ret = traceprobe_parse_event_name(&event, &group, gbuf, event - argv[0]); if (ret) - goto fail_address_parse; + return ret; } if (!event) { @@ -658,15 +667,16 @@ static int __trace_uprobe_create(int argc, const char **argv) char *ptr; tail = kstrdup(kbasename(filename), GFP_KERNEL); - if (!tail) { - ret = -ENOMEM; - goto fail_address_parse; - } + if (!tail) + return -ENOMEM; ptr = strpbrk(tail, ".-_"); if (ptr) *ptr = '\0'; + buf = kmalloc(MAX_EVENT_NAME_LEN, GFP_KERNEL); + if (!buf) + return -ENOMEM; snprintf(buf, MAX_EVENT_NAME_LEN, "%c_%s_0x%lx", 'p', tail, offset); event = buf; kfree(tail); @@ -680,42 +690,36 @@ static int __trace_uprobe_create(int argc, const char **argv) ret = PTR_ERR(tu); /* This must return -ENOMEM otherwise there is a bug */ WARN_ON_ONCE(ret != -ENOMEM); - goto fail_address_parse; + return ret; } tu->offset = offset; tu->ref_ctr_offset = ref_ctr_offset; tu->path = path; - tu->filename = filename; + /* Clear @path so that it will not freed by path_put() */ + memset(&path, 0, sizeof(path)); + tu->filename = no_free_ptr(filename); + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + ctx->flags = (is_return ? TPARG_FL_RETURN : 0) | TPARG_FL_USER; /* parse arguments */ - for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + for (i = 0; i < argc; i++) { trace_probe_log_set_index(i + 2); - ret = traceprobe_parse_probe_arg(&tu->tp, i, argv[i], - (is_return ? TPARG_FL_RETURN : 0) | - TPARG_FL_USER); + ret = traceprobe_parse_probe_arg(&tu->tp, i, argv[i], ctx); if (ret) - goto error; + return ret; } ptype = is_ret_probe(tu) ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL; ret = traceprobe_set_print_fmt(&tu->tp, ptype); if (ret < 0) - goto error; + return ret; ret = register_trace_uprobe(tu); if (!ret) - goto out; - -error: - free_trace_uprobe(tu); -out: - trace_probe_log_clear(); - return ret; - -fail_address_parse: - trace_probe_log_clear(); - path_put(&path); - kfree(filename); + tu = NULL; return ret; } @@ -732,7 +736,7 @@ static int create_or_delete_trace_uprobe(const char *raw_command) if (raw_command[0] == '-') return dyn_event_release(raw_command, &trace_uprobe_ops); - ret = trace_uprobe_create(raw_command); + ret = dyn_event_create(raw_command, &trace_uprobe_ops); return ret == -ECANCELED ? -EINVAL : ret; } @@ -819,13 +823,21 @@ static int probes_profile_seq_show(struct seq_file *m, void *v) { struct dyn_event *ev = v; struct trace_uprobe *tu; + unsigned long nhits; + int cpu; if (!is_trace_uprobe(ev)) return 0; tu = to_trace_uprobe(ev); + + nhits = 0; + for_each_possible_cpu(cpu) { + nhits += per_cpu(*tu->nhits, cpu); + } + seq_printf(m, " %s %-44s %15lu\n", tu->filename, - trace_probe_name(&tu->tp), tu->nhit); + trace_probe_name(&tu->tp), nhits); return 0; } @@ -858,9 +870,11 @@ static const struct file_operations uprobe_profile_ops = { struct uprobe_cpu_buffer { struct mutex mutex; void *buf; + int dsize; }; static struct uprobe_cpu_buffer __percpu *uprobe_cpu_buffer; static int uprobe_buffer_refcnt; +#define MAX_UCB_BUFFER_SIZE PAGE_SIZE static int uprobe_buffer_init(void) { @@ -944,12 +958,41 @@ static struct uprobe_cpu_buffer *uprobe_buffer_get(void) static void uprobe_buffer_put(struct uprobe_cpu_buffer *ucb) { + if (!ucb) + return; mutex_unlock(&ucb->mutex); } +static struct uprobe_cpu_buffer *prepare_uprobe_buffer(struct trace_uprobe *tu, + struct pt_regs *regs, + struct uprobe_cpu_buffer **ucbp) +{ + struct uprobe_cpu_buffer *ucb; + int dsize, esize; + + if (*ucbp) + return *ucbp; + + esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); + dsize = __get_data_size(&tu->tp, regs, NULL); + + ucb = uprobe_buffer_get(); + ucb->dsize = tu->tp.size + dsize; + + if (WARN_ON_ONCE(ucb->dsize > MAX_UCB_BUFFER_SIZE)) { + ucb->dsize = MAX_UCB_BUFFER_SIZE; + dsize = MAX_UCB_BUFFER_SIZE - tu->tp.size; + } + + store_trace_args(ucb->buf, &tu->tp, regs, NULL, esize, dsize); + + *ucbp = ucb; + return ucb; +} + static void __uprobe_trace_func(struct trace_uprobe *tu, unsigned long func, struct pt_regs *regs, - struct uprobe_cpu_buffer *ucb, int dsize, + struct uprobe_cpu_buffer *ucb, struct trace_event_file *trace_file) { struct uprobe_trace_entry_head *entry; @@ -960,14 +1003,11 @@ static void __uprobe_trace_func(struct trace_uprobe *tu, WARN_ON(call != trace_file->event_call); - if (WARN_ON_ONCE(tu->tp.size + dsize > PAGE_SIZE)) - return; - if (trace_trigger_soft_disabled(trace_file)) return; esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); - size = esize + tu->tp.size + dsize; + size = esize + ucb->dsize; entry = trace_event_buffer_reserve(&fbuffer, trace_file, size); if (!entry) return; @@ -981,23 +1021,26 @@ static void __uprobe_trace_func(struct trace_uprobe *tu, data = DATAOF_TRACE_ENTRY(entry, false); } - memcpy(data, ucb->buf, tu->tp.size + dsize); + memcpy(data, ucb->buf, ucb->dsize); trace_event_buffer_commit(&fbuffer); } /* uprobe handler */ static int uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs, - struct uprobe_cpu_buffer *ucb, int dsize) + struct uprobe_cpu_buffer **ucbp) { struct event_file_link *link; + struct uprobe_cpu_buffer *ucb; if (is_ret_probe(tu)) return 0; + ucb = prepare_uprobe_buffer(tu, regs, ucbp); + rcu_read_lock(); trace_probe_for_each_link_rcu(link, &tu->tp) - __uprobe_trace_func(tu, 0, regs, ucb, dsize, link->file); + __uprobe_trace_func(tu, 0, regs, ucb, link->file); rcu_read_unlock(); return 0; @@ -1005,13 +1048,16 @@ static int uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs, static void uretprobe_trace_func(struct trace_uprobe *tu, unsigned long func, struct pt_regs *regs, - struct uprobe_cpu_buffer *ucb, int dsize) + struct uprobe_cpu_buffer **ucbp) { struct event_file_link *link; + struct uprobe_cpu_buffer *ucb; + + ucb = prepare_uprobe_buffer(tu, regs, ucbp); rcu_read_lock(); trace_probe_for_each_link_rcu(link, &tu->tp) - __uprobe_trace_func(tu, func, regs, ucb, dsize, link->file); + __uprobe_trace_func(tu, func, regs, ucb, link->file); rcu_read_unlock(); } @@ -1042,7 +1088,7 @@ print_uprobe_event(struct trace_iterator *iter, int flags, struct trace_event *e data = DATAOF_TRACE_ENTRY(entry, false); } - if (print_probe_args(s, tu->tp.args, tu->tp.nr_args, data, entry) < 0) + if (trace_probe_print_args(s, tu->tp.args, tu->tp.nr_args, data, entry) < 0) goto out; trace_seq_putc(s, '\n'); @@ -1051,43 +1097,40 @@ print_uprobe_event(struct trace_iterator *iter, int flags, struct trace_event *e return trace_handle_return(s); } -typedef bool (*filter_func_t)(struct uprobe_consumer *self, - enum uprobe_filter_ctx ctx, - struct mm_struct *mm); +typedef bool (*filter_func_t)(struct uprobe_consumer *self, struct mm_struct *mm); static int trace_uprobe_enable(struct trace_uprobe *tu, filter_func_t filter) { - int ret; + struct inode *inode = d_real_inode(tu->path.dentry); + struct uprobe *uprobe; tu->consumer.filter = filter; - tu->inode = d_real_inode(tu->path.dentry); - - if (tu->ref_ctr_offset) - ret = uprobe_register_refctr(tu->inode, tu->offset, - tu->ref_ctr_offset, &tu->consumer); - else - ret = uprobe_register(tu->inode, tu->offset, &tu->consumer); - - if (ret) - tu->inode = NULL; + uprobe = uprobe_register(inode, tu->offset, tu->ref_ctr_offset, &tu->consumer); + if (IS_ERR(uprobe)) + return PTR_ERR(uprobe); - return ret; + tu->uprobe = uprobe; + return 0; } static void __probe_event_disable(struct trace_probe *tp) { struct trace_uprobe *tu; + bool sync = false; tu = container_of(tp, struct trace_uprobe, tp); WARN_ON(!uprobe_filter_is_empty(tu->tp.event->filter)); list_for_each_entry(tu, trace_probe_probe_list(tp), tp.list) { - if (!tu->inode) + if (!tu->uprobe) continue; - uprobe_unregister(tu->inode, tu->offset, &tu->consumer); - tu->inode = NULL; + uprobe_unregister_nosync(tu->uprobe, &tu->consumer); + sync = true; + tu->uprobe = NULL; } + if (sync) + uprobe_unregister_sync(); } static int probe_event_enable(struct trace_event_call *call, @@ -1203,9 +1246,6 @@ __uprobe_perf_filter(struct trace_uprobe_filter *filter, struct mm_struct *mm) { struct perf_event *event; - if (filter->nr_systemwide) - return true; - list_for_each_entry(event, &filter->perf_events, hw.tp_list) { if (event->hw.target->mm == mm) return true; @@ -1286,7 +1326,7 @@ static int uprobe_perf_close(struct trace_event_call *call, return 0; list_for_each_entry(tu, trace_probe_probe_list(tp), tp.list) { - ret = uprobe_apply(tu->inode, tu->offset, &tu->consumer, false); + ret = uprobe_apply(tu->uprobe, &tu->consumer, false); if (ret) break; } @@ -1310,7 +1350,7 @@ static int uprobe_perf_open(struct trace_event_call *call, return 0; list_for_each_entry(tu, trace_probe_probe_list(tp), tp.list) { - err = uprobe_apply(tu->inode, tu->offset, &tu->consumer, true); + err = uprobe_apply(tu->uprobe, &tu->consumer, true); if (err) { uprobe_perf_close(call, event); break; @@ -1320,8 +1360,7 @@ static int uprobe_perf_open(struct trace_event_call *call, return err; } -static bool uprobe_perf_filter(struct uprobe_consumer *uc, - enum uprobe_filter_ctx ctx, struct mm_struct *mm) +static bool uprobe_perf_filter(struct uprobe_consumer *uc, struct mm_struct *mm) { struct trace_uprobe_filter *filter; struct trace_uprobe *tu; @@ -1330,6 +1369,13 @@ static bool uprobe_perf_filter(struct uprobe_consumer *uc, tu = container_of(uc, struct trace_uprobe, consumer); filter = tu->tp.event->filter; + /* + * speculative short-circuiting check to avoid unnecessarily taking + * filter->rwlock below, if the uprobe has system-wide consumer + */ + if (READ_ONCE(filter->nr_systemwide)) + return true; + read_lock(&filter->rwlock); ret = __uprobe_perf_filter(filter, mm); read_unlock(&filter->rwlock); @@ -1339,10 +1385,11 @@ static bool uprobe_perf_filter(struct uprobe_consumer *uc, static void __uprobe_perf_func(struct trace_uprobe *tu, unsigned long func, struct pt_regs *regs, - struct uprobe_cpu_buffer *ucb, int dsize) + struct uprobe_cpu_buffer **ucbp) { struct trace_event_call *call = trace_probe_event_call(&tu->tp); struct uprobe_trace_entry_head *entry; + struct uprobe_cpu_buffer *ucb; struct hlist_head *head; void *data; int size, esize; @@ -1350,9 +1397,13 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, #ifdef CONFIG_BPF_EVENTS if (bpf_prog_array_valid(call)) { + const struct bpf_prog_array *array; u32 ret; - ret = bpf_prog_run_array_sleepable(call->prog_array, regs, bpf_prog_run); + rcu_read_lock_trace(); + array = rcu_dereference_check(call->prog_array, rcu_read_lock_trace_held()); + ret = bpf_prog_run_array_uprobe(array, regs, bpf_prog_run); + rcu_read_unlock_trace(); if (!ret) return; } @@ -1360,7 +1411,8 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); - size = esize + tu->tp.size + dsize; + ucb = prepare_uprobe_buffer(tu, regs, ucbp); + size = esize + ucb->dsize; size = ALIGN(size + sizeof(u32), sizeof(u64)) - sizeof(u32); if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough")) return; @@ -1383,13 +1435,10 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, data = DATAOF_TRACE_ENTRY(entry, false); } - memcpy(data, ucb->buf, tu->tp.size + dsize); + memcpy(data, ucb->buf, ucb->dsize); - if (size - esize > tu->tp.size + dsize) { - int len = tu->tp.size + dsize; - - memset(data + len, 0, size - esize - len); - } + if (size - esize > ucb->dsize) + memset(data + ucb->dsize, 0, size - esize - ucb->dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); @@ -1399,26 +1448,26 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, /* uprobe profile handler */ static int uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs, - struct uprobe_cpu_buffer *ucb, int dsize) + struct uprobe_cpu_buffer **ucbp) { - if (!uprobe_perf_filter(&tu->consumer, 0, current->mm)) + if (!uprobe_perf_filter(&tu->consumer, current->mm)) return UPROBE_HANDLER_REMOVE; if (!is_ret_probe(tu)) - __uprobe_perf_func(tu, 0, regs, ucb, dsize); + __uprobe_perf_func(tu, 0, regs, ucbp); return 0; } static void uretprobe_perf_func(struct trace_uprobe *tu, unsigned long func, struct pt_regs *regs, - struct uprobe_cpu_buffer *ucb, int dsize) + struct uprobe_cpu_buffer **ucbp) { - __uprobe_perf_func(tu, func, regs, ucb, dsize); + __uprobe_perf_func(tu, func, regs, ucbp); } int bpf_get_uprobe_info(const struct perf_event *event, u32 *fd_type, const char **filename, u64 *probe_offset, - bool perf_type_tracepoint) + u64 *probe_addr, bool perf_type_tracepoint) { const char *pevent = trace_event_name(event->tp_event); const char *group = event->tp_event->class->system; @@ -1435,6 +1484,7 @@ int bpf_get_uprobe_info(const struct perf_event *event, u32 *fd_type, : BPF_FD_TYPE_UPROBE; *filename = tu->filename; *probe_offset = tu->offset; + *probe_addr = tu->ref_ctr_offset; return 0; } #endif /* CONFIG_PERF_EVENTS */ @@ -1473,17 +1523,18 @@ trace_uprobe_register(struct trace_event_call *event, enum trace_reg type, } } -static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs) +static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs, + __u64 *data) { struct trace_uprobe *tu; struct uprobe_dispatch_data udd; - struct uprobe_cpu_buffer *ucb; - int dsize, esize; + struct uprobe_cpu_buffer *ucb = NULL; + unsigned int flags; int ret = 0; - tu = container_of(con, struct trace_uprobe, consumer); - tu->nhit++; + + this_cpu_inc(*tu->nhits); udd.tu = tu; udd.bp_addr = instruction_pointer(regs); @@ -1493,30 +1544,26 @@ static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs) if (WARN_ON_ONCE(!uprobe_cpu_buffer)) return 0; - dsize = __get_data_size(&tu->tp, regs); - esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); - - ucb = uprobe_buffer_get(); - store_trace_args(ucb->buf, &tu->tp, regs, esize, dsize); - - if (trace_probe_test_flag(&tu->tp, TP_FLAG_TRACE)) - ret |= uprobe_trace_func(tu, regs, ucb, dsize); + flags = trace_probe_load_flag(&tu->tp); + if (flags & TP_FLAG_TRACE) + ret |= uprobe_trace_func(tu, regs, &ucb); #ifdef CONFIG_PERF_EVENTS - if (trace_probe_test_flag(&tu->tp, TP_FLAG_PROFILE)) - ret |= uprobe_perf_func(tu, regs, ucb, dsize); + if (flags & TP_FLAG_PROFILE) + ret |= uprobe_perf_func(tu, regs, &ucb); #endif uprobe_buffer_put(ucb); return ret; } static int uretprobe_dispatcher(struct uprobe_consumer *con, - unsigned long func, struct pt_regs *regs) + unsigned long func, struct pt_regs *regs, + __u64 *data) { struct trace_uprobe *tu; struct uprobe_dispatch_data udd; - struct uprobe_cpu_buffer *ucb; - int dsize, esize; + struct uprobe_cpu_buffer *ucb = NULL; + unsigned int flags; tu = container_of(con, struct trace_uprobe, consumer); @@ -1528,18 +1575,13 @@ static int uretprobe_dispatcher(struct uprobe_consumer *con, if (WARN_ON_ONCE(!uprobe_cpu_buffer)) return 0; - dsize = __get_data_size(&tu->tp, regs); - esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); - - ucb = uprobe_buffer_get(); - store_trace_args(ucb->buf, &tu->tp, regs, esize, dsize); - - if (trace_probe_test_flag(&tu->tp, TP_FLAG_TRACE)) - uretprobe_trace_func(tu, func, regs, ucb, dsize); + flags = trace_probe_load_flag(&tu->tp); + if (flags & TP_FLAG_TRACE) + uretprobe_trace_func(tu, func, regs, &ucb); #ifdef CONFIG_PERF_EVENTS - if (trace_probe_test_flag(&tu->tp, TP_FLAG_PROFILE)) - uretprobe_perf_func(tu, func, regs, ucb, dsize); + if (flags & TP_FLAG_PROFILE) + uretprobe_perf_func(tu, func, regs, &ucb); #endif uprobe_buffer_put(ucb); return 0; diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c index c774e560f2f9..7f8da4dab69d 100644 --- a/kernel/trace/tracing_map.c +++ b/kernel/trace/tracing_map.c @@ -454,7 +454,7 @@ static struct tracing_map_elt *get_free_elt(struct tracing_map *map) struct tracing_map_elt *elt = NULL; int idx; - idx = atomic_inc_return(&map->next_elt); + idx = atomic_fetch_add_unless(&map->next_elt, 1, map->max_elts); if (idx < map->max_elts) { elt = *(TRACING_MAP_ELT(map->elts, idx)); if (map->ops && map->ops->elt_init) @@ -574,7 +574,12 @@ __tracing_map_insert(struct tracing_map *map, void *key, bool lookup_only) } memcpy(elt->key, key, map->key_size); - entry->val = elt; + /* + * Ensure the initialization is visible and + * publish the elt. + */ + smp_wmb(); + WRITE_ONCE(entry->val, elt); atomic64_inc(&map->hits); return entry->val; @@ -694,7 +699,7 @@ void tracing_map_clear(struct tracing_map *map) { unsigned int i; - atomic_set(&map->next_elt, -1); + atomic_set(&map->next_elt, 0); atomic64_set(&map->hits, 0); atomic64_set(&map->drops, 0); @@ -778,7 +783,7 @@ struct tracing_map *tracing_map_create(unsigned int map_bits, map->map_bits = map_bits; map->max_elts = (1 << map_bits); - atomic_set(&map->next_elt, -1); + atomic_set(&map->next_elt, 0); map->map_size = (1 << (map_bits + 1)); map->ops = ops; @@ -840,15 +845,11 @@ int tracing_map_init(struct tracing_map *map) static int cmp_entries_dup(const void *A, const void *B) { const struct tracing_map_sort_entry *a, *b; - int ret = 0; a = *(const struct tracing_map_sort_entry **)A; b = *(const struct tracing_map_sort_entry **)B; - if (memcmp(a->key, b->key, a->elt->map->key_size)) - ret = 1; - - return ret; + return memcmp(a->key, b->key, a->elt->map->key_size); } static int cmp_entries_sum(const void *A, const void *B) @@ -1075,7 +1076,7 @@ int tracing_map_sort_entries(struct tracing_map *map, struct tracing_map_sort_entry *sort_entry, **entries; int i, n_entries, ret; - entries = vmalloc(array_size(sizeof(sort_entry), map->max_elts)); + entries = vmalloc_array(map->max_elts, sizeof(sort_entry)); if (!entries) return -ENOMEM; diff --git a/kernel/trace/tracing_map.h b/kernel/trace/tracing_map.h index 2c765ee2a4d4..99c37eeebc16 100644 --- a/kernel/trace/tracing_map.h +++ b/kernel/trace/tracing_map.h @@ -272,10 +272,6 @@ extern u64 tracing_map_read_sum(struct tracing_map_elt *elt, unsigned int i); extern u64 tracing_map_read_var(struct tracing_map_elt *elt, unsigned int i); extern u64 tracing_map_read_var_once(struct tracing_map_elt *elt, unsigned int i); -extern void tracing_map_set_field_descr(struct tracing_map *map, - unsigned int i, - unsigned int key_offset, - tracing_map_cmp_fn_t cmp_fn); extern int tracing_map_sort_entries(struct tracing_map *map, struct tracing_map_sort_key *sort_keys, |