| Age | Commit message (Collapse) | Author |
|---|
|
Now that kptr_off_tab has been refactored into btf_record, and can hold
more than one specific field type, accomodate bpf_spin_lock and
bpf_timer as well.
While they don't require any more metadata than offset, having all
special fields in one place allows us to share the same code for
allocated user defined types and handle both map values and these
allocated objects in a similar fashion.
As an optimization, we still keep spin_lock_off and timer_off offsets in
the btf_record structure, just to avoid having to find the btf_field
struct each time their offset is needed. This is mostly needed to
manipulate such objects in a map value at runtime. It's ok to hardcode
just one offset as more than one field is disallowed.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20221103191013.1236066-8-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
To prepare the BPF verifier to handle special fields in both map values
and program allocated types coming from program BTF, we need to refactor
the kptr_off_tab handling code into something more generic and reusable
across both cases to avoid code duplication.
Later patches also require passing this data to helpers at runtime, so
that they can work on user defined types, initialize them, destruct
them, etc.
The main observation is that both map values and such allocated types
point to a type in program BTF, hence they can be handled similarly. We
can prepare a field metadata table for both cases and store them in
struct bpf_map or struct btf depending on the use case.
Hence, refactor the code into generic btf_record and btf_field member
structs. The btf_record represents the fields of a specific btf_type in
user BTF. The cnt indicates the number of special fields we successfully
recognized, and field_mask is a bitmask of fields that were found, to
enable quick determination of availability of a certain field.
Subsequently, refactor the rest of the code to work with these generic
types, remove assumptions about kptr and kptr_off_tab, rename variables
to more meaningful names, etc.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20221103191013.1236066-7-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Daniel Borkmann says:
====================
bpf 2022-11-04
We've added 8 non-merge commits during the last 3 day(s) which contain
a total of 10 files changed, 113 insertions(+), 16 deletions(-).
The main changes are:
1) Fix memory leak upon allocation failure in BPF verifier's stack state
tracking, from Kees Cook.
2) Fix address leakage when BPF progs release reference to an object,
from Youlin Li.
3) Fix BPF CI breakage from buggy in.h uapi header dependency,
from Andrii Nakryiko.
4) Fix bpftool pin sub-command's argument parsing, from Pu Lehui.
5) Fix BPF sockmap lockdep warning by cancelling psock work outside
of socket lock, from Cong Wang.
6) Follow-up for BPF sockmap to fix sk_forward_alloc accounting,
from Wang Yufen.
bpf-for-netdev
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
selftests/bpf: Add verifier test for release_reference()
bpf: Fix wrong reg type conversion in release_reference()
bpf, sock_map: Move cancel_work_sync() out of sock lock
tools/headers: Pull in stddef.h to uapi to fix BPF selftests build in CI
net/ipv4: Fix linux/in.h header dependencies
bpftool: Fix NULL pointer dereference when pin {PROG, MAP, LINK} without FILE
bpf, sockmap: Fix the sk->sk_forward_alloc warning of sk_stream_kill_queues
bpf, verifier: Fix memory leak in array reallocation for stack state
====================
Link: https://lore.kernel.org/r/20221104000445.30761-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
It is not scalable to maintain a list of types that can have non-zero
ref_obj_id. It is never set for scalars anyway, so just remove the
conditional on register types and print it whenever it is non-zero.
Acked-by: Dave Marchevsky <davemarchevsky@fb.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20221103191013.1236066-6-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
For the case where allow_ptr_leaks is false, code is checking whether
slot type is STACK_INVALID and STACK_SPILL and rejecting other cases.
This is a consequence of incorrectly checking for register type instead
of the slot type (NOT_INIT and SCALAR_VALUE respectively). Fix the
check.
Fixes: 01f810ace9ed ("bpf: Allow variable-offset stack access")
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20221103191013.1236066-5-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
When support was added for spilled PTR_TO_BTF_ID to be accessed by
helper memory access, the stack slot was not overwritten to STACK_MISC
(and that too is only safe when env->allow_ptr_leaks is true).
This means that helpers who take ARG_PTR_TO_MEM and write to it may
essentially overwrite the value while the verifier continues to track
the slot for spilled register.
This can cause issues when PTR_TO_BTF_ID is spilled to stack, and then
overwritten by helper write access, which can then be passed to BPF
helpers or kfuncs.
Handle this by falling back to the case introduced in a later commit,
which will also handle PTR_TO_BTF_ID along with other pointer types,
i.e. cd17d38f8b28 ("bpf: Permits pointers on stack for helper calls").
Finally, include a comment on why REG_LIVE_WRITTEN is not being set when
clobber is set to true. In short, the reason is that while when clobber
is unset, we know that we won't be writing, when it is true, we *may*
write to any of the stack slots in that range. It may be a partial or
complete write, to just one or many stack slots.
We cannot be sure, hence to be conservative, we leave things as is and
never set REG_LIVE_WRITTEN for any stack slot. However, clobber still
needs to reset them to STACK_MISC assuming writes happened. However read
marks still need to be propagated upwards from liveness point of view,
as parent stack slot's contents may still continue to matter to child
states.
Cc: Yonghong Song <yhs@meta.com>
Fixes: 1d68f22b3d53 ("bpf: Handle spilled PTR_TO_BTF_ID properly when checking stack_boundary")
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20221103191013.1236066-4-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
This is useful in particular to mark the pointer as volatile, so that
compiler treats each load and store to the field as a volatile access.
The alternative is having to define and use READ_ONCE and WRITE_ONCE in
the BPF program.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20221103191013.1236066-3-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
test_gen_kprobe_cmd() only free buf in fail path, hence buf will leak
when there is no failure. Move kfree(buf) from fail path to common path
to prevent the memleak. The same reason and solution in
test_gen_kretprobe_cmd().
unreferenced object 0xffff888143b14000 (size 2048):
comm "insmod", pid 52490, jiffies 4301890980 (age 40.553s)
hex dump (first 32 bytes):
70 3a 6b 70 72 6f 62 65 73 2f 67 65 6e 5f 6b 70 p:kprobes/gen_kp
72 6f 62 65 5f 74 65 73 74 20 64 6f 5f 73 79 73 robe_test do_sys
backtrace:
[<000000006d7b836b>] kmalloc_trace+0x27/0xa0
[<0000000009528b5b>] 0xffffffffa059006f
[<000000008408b580>] do_one_initcall+0x87/0x2a0
[<00000000c4980a7e>] do_init_module+0xdf/0x320
[<00000000d775aad0>] load_module+0x3006/0x3390
[<00000000e9a74b80>] __do_sys_finit_module+0x113/0x1b0
[<000000003726480d>] do_syscall_64+0x35/0x80
[<000000003441e93b>] entry_SYSCALL_64_after_hwframe+0x46/0xb0
Link: https://lore.kernel.org/all/20221102072954.26555-1-shangxiaojing@huawei.com/
Fixes: 64836248dda2 ("tracing: Add kprobe event command generation test module")
Cc: stable@vger.kernel.org
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
Since commit ab51e15d535e ("fprobe: Introduce FPROBE_FL_KPROBE_SHARED flag
for fprobe") introduced fprobe_kprobe_handler() for fprobe::ops::func,
unregister_fprobe() fails to unregister the registered if user specifies
FPROBE_FL_KPROBE_SHARED flag.
Moreover, __register_ftrace_function() is possible to change the
ftrace_ops::func, thus we have to check fprobe::ops::saved_func instead.
To check it correctly, it should confirm the fprobe::ops::saved_func is
either fprobe_handler() or fprobe_kprobe_handler().
Link: https://lore.kernel.org/all/166677683946.1459107.15997653945538644683.stgit@devnote3/
Fixes: cad9931f64dc ("fprobe: Add ftrace based probe APIs")
Cc: stable@vger.kernel.org
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
Check if fp->rethook succeeded to be allocated. Otherwise, if
rethook_alloc() fails, then we end up dereferencing a NULL pointer in
rethook_add_node().
Link: https://lore.kernel.org/all/20221025031209.954836-1-rafaelmendsr@gmail.com/
Fixes: 5b0ab78998e3 ("fprobe: Add exit_handler support")
Cc: stable@vger.kernel.org
Signed-off-by: Rafael Mendonca <rafaelmendsr@gmail.com>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
In aggregate kprobe case, when arm_kprobe failed,
we need set the kp->flags with KPROBE_FLAG_DISABLED again.
If not, the 'kp' kprobe will been considered as enabled
but it actually not enabled.
Link: https://lore.kernel.org/all/20220902155820.34755-1-liq3ea@163.com/
Fixes: 12310e343755 ("kprobes: Propagate error from arm_kprobe_ftrace()")
Cc: stable@vger.kernel.org
Signed-off-by: Li Qiang <liq3ea@163.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
|
|
Some helper functions will allocate memory. To avoid memory leaks, the
verifier requires the eBPF program to release these memories by calling
the corresponding helper functions.
When a resource is released, all pointer registers corresponding to the
resource should be invalidated. The verifier use release_references() to
do this job, by apply __mark_reg_unknown() to each relevant register.
It will give these registers the type of SCALAR_VALUE. A register that
will contain a pointer value at runtime, but of type SCALAR_VALUE, which
may allow the unprivileged user to get a kernel pointer by storing this
register into a map.
Using __mark_reg_not_init() while NOT allow_ptr_leaks can mitigate this
problem.
Fixes: fd978bf7fd31 ("bpf: Add reference tracking to verifier")
Signed-off-by: Youlin Li <liulin063@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20221103093440.3161-1-liulin063@gmail.com
|
|
No conflicts.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The actual maximum image size formula in hibernate_preallocate_memory()
is as follows:
max_size = (count - (size + PAGES_FOR_IO)) / 2
- 2 * DIV_ROUND_UP(reserved_size, PAGE_SIZE);
but the one in the kerneldoc comment of the function is different and
incorrect.
Fixes: ddeb64870810 ("PM / Hibernate: Add sysfs knob to control size of memory for drivers")
Signed-off-by: xiongxin <xiongxin@kylinos.cn>
[ rjw: Subject and changelog rewrite ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
|
KASAN reported a use-after-free with ftrace ops [1]. It was found from
vmcore that perf had registered two ops with the same content
successively, both dynamic. After unregistering the second ops, a
use-after-free occurred.
In ftrace_shutdown(), when the second ops is unregistered, the
FTRACE_UPDATE_CALLS command is not set because there is another enabled
ops with the same content. Also, both ops are dynamic and the ftrace
callback function is ftrace_ops_list_func, so the
FTRACE_UPDATE_TRACE_FUNC command will not be set. Eventually the value
of 'command' will be 0 and ftrace_shutdown() will skip the rcu
synchronization.
However, ftrace may be activated. When the ops is released, another CPU
may be accessing the ops. Add the missing synchronization to fix this
problem.
[1]
BUG: KASAN: use-after-free in __ftrace_ops_list_func kernel/trace/ftrace.c:7020 [inline]
BUG: KASAN: use-after-free in ftrace_ops_list_func+0x2b0/0x31c kernel/trace/ftrace.c:7049
Read of size 8 at addr ffff56551965bbc8 by task syz-executor.2/14468
CPU: 1 PID: 14468 Comm: syz-executor.2 Not tainted 5.10.0 #7
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x0/0x40c arch/arm64/kernel/stacktrace.c:132
show_stack+0x30/0x40 arch/arm64/kernel/stacktrace.c:196
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x1b4/0x248 lib/dump_stack.c:118
print_address_description.constprop.0+0x28/0x48c mm/kasan/report.c:387
__kasan_report mm/kasan/report.c:547 [inline]
kasan_report+0x118/0x210 mm/kasan/report.c:564
check_memory_region_inline mm/kasan/generic.c:187 [inline]
__asan_load8+0x98/0xc0 mm/kasan/generic.c:253
__ftrace_ops_list_func kernel/trace/ftrace.c:7020 [inline]
ftrace_ops_list_func+0x2b0/0x31c kernel/trace/ftrace.c:7049
ftrace_graph_call+0x0/0x4
__might_sleep+0x8/0x100 include/linux/perf_event.h:1170
__might_fault mm/memory.c:5183 [inline]
__might_fault+0x58/0x70 mm/memory.c:5171
do_strncpy_from_user lib/strncpy_from_user.c:41 [inline]
strncpy_from_user+0x1f4/0x4b0 lib/strncpy_from_user.c:139
getname_flags+0xb0/0x31c fs/namei.c:149
getname+0x2c/0x40 fs/namei.c:209
[...]
Allocated by task 14445:
kasan_save_stack+0x24/0x50 mm/kasan/common.c:48
kasan_set_track mm/kasan/common.c:56 [inline]
__kasan_kmalloc mm/kasan/common.c:479 [inline]
__kasan_kmalloc.constprop.0+0x110/0x13c mm/kasan/common.c:449
kasan_kmalloc+0xc/0x14 mm/kasan/common.c:493
kmem_cache_alloc_trace+0x440/0x924 mm/slub.c:2950
kmalloc include/linux/slab.h:563 [inline]
kzalloc include/linux/slab.h:675 [inline]
perf_event_alloc.part.0+0xb4/0x1350 kernel/events/core.c:11230
perf_event_alloc kernel/events/core.c:11733 [inline]
__do_sys_perf_event_open kernel/events/core.c:11831 [inline]
__se_sys_perf_event_open+0x550/0x15f4 kernel/events/core.c:11723
__arm64_sys_perf_event_open+0x6c/0x80 kernel/events/core.c:11723
[...]
Freed by task 14445:
kasan_save_stack+0x24/0x50 mm/kasan/common.c:48
kasan_set_track+0x24/0x34 mm/kasan/common.c:56
kasan_set_free_info+0x20/0x40 mm/kasan/generic.c:358
__kasan_slab_free.part.0+0x11c/0x1b0 mm/kasan/common.c:437
__kasan_slab_free mm/kasan/common.c:445 [inline]
kasan_slab_free+0x2c/0x40 mm/kasan/common.c:446
slab_free_hook mm/slub.c:1569 [inline]
slab_free_freelist_hook mm/slub.c:1608 [inline]
slab_free mm/slub.c:3179 [inline]
kfree+0x12c/0xc10 mm/slub.c:4176
perf_event_alloc.part.0+0xa0c/0x1350 kernel/events/core.c:11434
perf_event_alloc kernel/events/core.c:11733 [inline]
__do_sys_perf_event_open kernel/events/core.c:11831 [inline]
__se_sys_perf_event_open+0x550/0x15f4 kernel/events/core.c:11723
[...]
Link: https://lore.kernel.org/linux-trace-kernel/20221103031010.166498-1-lihuafei1@huawei.com
Fixes: edb096e00724f ("ftrace: Fix memleak when unregistering dynamic ops when tracing disabled")
Cc: stable@vger.kernel.org
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Li Huafei <lihuafei1@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
On some machines the number of listed CPUs may be bigger than the actual
CPUs that exist. The tracing subsystem allocates a per_cpu directory with
access to the per CPU ring buffer via a cpuX file. But to save space, the
ring buffer will only allocate buffers for online CPUs, even though the
CPU array will be as big as the nr_cpu_ids.
With the addition of waking waiters on the ring buffer when closing the
file, the ring_buffer_wake_waiters() now needs to make sure that the
buffer is allocated (with the irq_work allocated with it) before trying to
wake waiters, as it will cause a NULL pointer dereference.
While debugging this, I added a NULL check for the buffer itself (which is
OK to do), and also NULL pointer checks against buffer->buffers (which is
not fine, and will WARN) as well as making sure the CPU number passed in
is within the nr_cpu_ids (which is also not fine if it isn't).
Link: https://lore.kernel.org/all/87h6zklb6n.wl-tiwai@suse.de/
Link: https://lore.kernel.org/all/CAM6Wdxc0KRJMXVAA0Y=u6Jh2V=uWB-_Fn6M4xRuNppfXzL1mUg@mail.gmail.com/
Link: https://lkml.kernel.org/linux-trace-kernel/20221101191009.1e7378c8@rorschach.local.home
Cc: stable@vger.kernel.org
Cc: Steven Noonan <steven.noonan@gmail.com>
Bugzilla: https://bugzilla.opensuse.org/show_bug.cgi?id=1204705
Reported-by: Takashi Iwai <tiwai@suse.de>
Reported-by: Roland Ruckerbauer <roland.rucky@gmail.com>
Fixes: f3ddb74ad079 ("tracing: Wake up ring buffer waiters on closing of the file")
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Daniel Borkmann says:
====================
bpf-next 2022-11-02
We've added 70 non-merge commits during the last 14 day(s) which contain
a total of 96 files changed, 3203 insertions(+), 640 deletions(-).
The main changes are:
1) Make cgroup local storage available to non-cgroup attached BPF programs
such as tc BPF ones, from Yonghong Song.
2) Avoid unnecessary deadlock detection and failures wrt BPF task storage
helpers, from Martin KaFai Lau.
3) Add LLVM disassembler as default library for dumping JITed code
in bpftool, from Quentin Monnet.
4) Various kprobe_multi_link fixes related to kernel modules,
from Jiri Olsa.
5) Optimize x86-64 JIT with emitting BMI2-based shift instructions,
from Jie Meng.
6) Improve BPF verifier's memory type compatibility for map key/value
arguments, from Dave Marchevsky.
7) Only create mmap-able data section maps in libbpf when data is exposed
via skeletons, from Andrii Nakryiko.
8) Add an autoattach option for bpftool to load all object assets,
from Wang Yufen.
9) Various memory handling fixes for libbpf and BPF selftests,
from Xu Kuohai.
10) Initial support for BPF selftest's vmtest.sh on arm64,
from Manu Bretelle.
11) Improve libbpf's BTF handling to dedup identical structs,
from Alan Maguire.
12) Add BPF CI and denylist documentation for BPF selftests,
from Daniel Müller.
13) Check BPF cpumap max_entries before doing allocation work,
from Florian Lehner.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (70 commits)
samples/bpf: Fix typo in README
bpf: Remove the obsolte u64_stats_fetch_*_irq() users.
bpf: check max_entries before allocating memory
bpf: Fix a typo in comment for DFS algorithm
bpftool: Fix spelling mistake "disasembler" -> "disassembler"
selftests/bpf: Fix bpftool synctypes checking failure
selftests/bpf: Panic on hard/soft lockup
docs/bpf: Add documentation for new cgroup local storage
selftests/bpf: Add test cgrp_local_storage to DENYLIST.s390x
selftests/bpf: Add selftests for new cgroup local storage
selftests/bpf: Fix test test_libbpf_str/bpf_map_type_str
bpftool: Support new cgroup local storage
libbpf: Support new cgroup local storage
bpf: Implement cgroup storage available to non-cgroup-attached bpf progs
bpf: Refactor some inode/task/sk storage functions for reuse
bpf: Make struct cgroup btf id global
selftests/bpf: Tracing prog can still do lookup under busy lock
selftests/bpf: Ensure no task storage failure for bpf_lsm.s prog due to deadlock detection
bpf: Add new bpf_task_storage_delete proto with no deadlock detection
bpf: bpf_task_storage_delete_recur does lookup first before the deadlock check
...
====================
Link: https://lore.kernel.org/r/20221102062120.5724-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Running the test currently fails on non-SMP systems, despite being
enabled by default. This means that running the test with:
./tools/testing/kunit/kunit.py run --arch x86_64 hw_breakpoint
results in every hw_breakpoint test failing with:
# test_one_cpu: failed to initialize: -22
not ok 1 - test_one_cpu
Instead, use kunit_skip(), which will mark the test as skipped, and give
a more comprehensible message:
ok 1 - test_one_cpu # SKIP not enough cpus
This makes it more obvious that the test is not suited to the test
environment, and so wasn't run, rather than having run and failed.
Signed-off-by: David Gow <davidgow@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Latypov <dlatypov@google.com>
Acked-by: Marco Elver <elver@google.com>
Link: https://lore.kernel.org/r/20221026141040.1609203-1-davidgow@google.com
|
|
A common exploit pattern for ROP attacks is to abuse prepare_kernel_cred()
in order to construct escalated privileges[1]. Instead of providing a
short-hand argument (NULL) to the "daemon" argument to indicate using
init_cred as the base cred, require that "daemon" is always set to
an actual task. Replace all existing callers that were passing NULL
with &init_task.
Future attacks will need to have sufficiently powerful read/write
primitives to have found an appropriately privileged task and written it
to the ROP stack as an argument to succeed, which is similarly difficult
to the prior effort needed to escalate privileges before struct cred
existed: locate the current cred and overwrite the uid member.
This has the added benefit of meaning that prepare_kernel_cred() can no
longer exceed the privileges of the init task, which may have changed from
the original init_cred (e.g. dropping capabilities from the bounding set).
[1] https://google.com/search?q=commit_creds(prepare_kernel_cred(0))
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: David Howells <dhowells@redhat.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Steve French <sfrench@samba.org>
Cc: Ronnie Sahlberg <lsahlber@redhat.com>
Cc: Shyam Prasad N <sprasad@microsoft.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: Anna Schumaker <anna@kernel.org>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: Jeff Layton <jlayton@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: "Michal Koutný" <mkoutny@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Cc: linux-nfs@vger.kernel.org
Cc: netdev@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Acked-by: Russ Weight <russell.h.weight@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
Link: https://lore.kernel.org/r/20221026232943.never.775-kees@kernel.org
|
|
If an error (NULL) is returned by krealloc(), callers of realloc_array()
were setting their allocation pointers to NULL, but on error krealloc()
does not touch the original allocation. This would result in a memory
resource leak. Instead, free the old allocation on the error handling
path.
The memory leak information is as follows as also reported by Zhengchao:
unreferenced object 0xffff888019801800 (size 256):
comm "bpf_repo", pid 6490, jiffies 4294959200 (age 17.170s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<00000000b211474b>] __kmalloc_node_track_caller+0x45/0xc0
[<0000000086712a0b>] krealloc+0x83/0xd0
[<00000000139aab02>] realloc_array+0x82/0xe2
[<00000000b1ca41d1>] grow_stack_state+0xfb/0x186
[<00000000cd6f36d2>] check_mem_access.cold+0x141/0x1341
[<0000000081780455>] do_check_common+0x5358/0xb350
[<0000000015f6b091>] bpf_check.cold+0xc3/0x29d
[<000000002973c690>] bpf_prog_load+0x13db/0x2240
[<00000000028d1644>] __sys_bpf+0x1605/0x4ce0
[<00000000053f29bd>] __x64_sys_bpf+0x75/0xb0
[<0000000056fedaf5>] do_syscall_64+0x35/0x80
[<000000002bd58261>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
Fixes: c69431aab67a ("bpf: verifier: Improve function state reallocation")
Reported-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reported-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Bill Wendling <morbo@google.com>
Cc: Lorenz Bauer <oss@lmb.io>
Link: https://lore.kernel.org/bpf/20221029025433.2533810-1-keescook@chromium.org
|
|
This is a full revert of commit:
f1389181622a ("kallsyms: Take callthunks into account")
The commit assumes a number of things that are not quite right.
Notably it assumes every symbol has PADDING_BYTES in front of it that
are not claimed by another symbol.
This is not true; even when compiled with:
-fpatchable-function-entry=${PADDING_BYTES},${PADDING_BYTES}
Notably things like .cold subfunctions do not need to adhere to this
change in ABI. It it also not true when build with CFI_CLANG, which
claims these PADDING_BYTES in the __cfi_##name symbol.
Once the prefix bytes are not consistent and or otherwise claimed the
approach this patch takes goes out the window and kallsym resolution
will report invalid symbol names.
Therefore revert this to make room for another approach.
Reported-by: Reported-by: kernel test robot <yujie.liu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Yujie Liu <yujie.liu@intel.com>
Link: https://lore.kernel.org/r/202210241614.2ae4c1f5-yujie.liu@intel.com
Link: https://lkml.kernel.org/r/20221028194453.330970755@infradead.org
|
|
At the moment the AMD encrypted platform reserves 6% of RAM for SWIOTLB
or 1GB, whichever is less. However it is possible that there is no block
big enough in the low memory which make SWIOTLB allocation fail and
the kernel continues without DMA. In such case a VM hangs on DMA.
This moves alloc+remap to a helper and calls it from a loop where
the size is halved on each iteration.
This updates default_nslabs on successful allocation which looks like
an oversight as not doing so should have broken callers of
swiotlb_size_or_default().
Signed-off-by: Alexey Kardashevskiy <aik@amd.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@amd.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Now that the 32bit UP oddity is gone and 32bit uses always a sequence
count, there is no need for the fetch_irq() variants anymore.
Convert to the regular interface.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/bpf/20221026123110.331690-1-bigeasy@linutronix.de
|
|
6ab428604f72 ("cgroup: Implement DEBUG_CGROUP_REF") added a config option
which forces cgroup refcnt functions to be not inlined so that they can be
kprobed for debugging. However, it forgot export them when the config is
enabled breaking modules which make use of css reference counting.
Fix it by adding CGROUP_REF_EXPORT() macro to cgroup_refcnt.h which is
defined to EXPORT_SYMBOL_GPL when CONFIG_DEBUG_CGROUP_REF is set.
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: 6ab428604f72 ("cgroup: Implement DEBUG_CGROUP_REF")
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Borislav Petkov:
- Rename a perf memory level event define to denote it is of CXL type
- Add Alder and Raptor Lakes support to RAPL
- Make sure raw sample data is output with tracepoints
* tag 'perf_urgent_for_v6.1_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/mem: Rename PERF_MEM_LVLNUM_EXTN_MEM to PERF_MEM_LVLNUM_CXL
perf/x86/rapl: Add support for Intel Raptor Lake
perf/x86/rapl: Add support for Intel AlderLake-N
perf: Fix missing raw data on tracepoint events
|
|
The commit d583d360a620 ("psi: Fix psi state corruption when schedule()
races with cgroup move") fixed a race problem by making cgroup_move_task()
use task->psi_flags instead of looking at the scheduler state.
We can extend task->psi_flags usage to CPU migration, which should be
a minor optimization for performance and code simplicity.
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Link: https://lore.kernel.org/r/20220926081931.45420-1-zhouchengming@bytedance.com
|
|
Psi polling mechanism is trying to minimize the number of wakeups to
run psi_poll_work and is currently relying on timer_pending() to detect
when this work is already scheduled. This provides a window of opportunity
for psi_group_change to schedule an immediate psi_poll_work after
poll_timer_fn got called but before psi_poll_work could reschedule itself.
Below is the depiction of this entire window:
poll_timer_fn
wake_up_interruptible(&group->poll_wait);
psi_poll_worker
wait_event_interruptible(group->poll_wait, ...)
psi_poll_work
psi_schedule_poll_work
if (timer_pending(&group->poll_timer)) return;
...
mod_timer(&group->poll_timer, jiffies + delay);
Prior to 461daba06bdc we used to rely on poll_scheduled atomic which was
reset and set back inside psi_poll_work and therefore this race window
was much smaller.
The larger window causes increased number of wakeups and our partners
report visible power regression of ~10mA after applying 461daba06bdc.
Bring back the poll_scheduled atomic and make this race window even
narrower by resetting poll_scheduled only when we reach polling expiration
time. This does not completely eliminate the possibility of extra wakeups
caused by a race with psi_group_change however it will limit it to the
worst case scenario of one extra wakeup per every tracking window (0.5s
in the worst case).
This patch also ensures correct ordering between clearing poll_scheduled
flag and obtaining changed_states using memory barrier. Correct ordering
between updating changed_states and setting poll_scheduled is ensured by
atomic_xchg operation.
By tracing the number of immediate rescheduling attempts performed by
psi_group_change and the number of these attempts being blocked due to
psi monitor being already active, we can assess the effects of this change:
Before the patch:
Run#1 Run#2 Run#3
Immediate reschedules attempted: 684365 1385156 1261240
Immediate reschedules blocked: 682846 1381654 1258682
Immediate reschedules (delta): 1519 3502 2558
Immediate reschedules (% of attempted): 0.22% 0.25% 0.20%
After the patch:
Run#1 Run#2 Run#3
Immediate reschedules attempted: 882244 770298 426218
Immediate reschedules blocked: 881996 769796 426074
Immediate reschedules (delta): 248 502 144
Immediate reschedules (% of attempted): 0.03% 0.07% 0.03%
The number of non-blocked immediate reschedules dropped from 0.22-0.25%
to 0.03-0.07%. The drop is attributed to the decrease in the race window
size and the fact that we allow this race only when psi monitors reach
polling window expiration time.
Fixes: 461daba06bdc ("psi: eliminate kthread_worker from psi trigger scheduling mechanism")
Reported-by: Kathleen Chang <yt.chang@mediatek.com>
Reported-by: Wenju Xu <wenju.xu@mediatek.com>
Reported-by: Jonathan Chen <jonathan.jmchen@mediatek.com>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chengming Zhou <zhouchengming@bytedance.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Tested-by: SH Chen <show-hong.chen@mediatek.com>
Link: https://lore.kernel.org/r/20221028194541.813985-1-surenb@google.com
|
|
Pavan reported a problem that PSI avgs_work idle shutoff is not
working at all. Because PSI_NONIDLE condition would be observed in
psi_avgs_work()->collect_percpu_times()->get_recent_times() even if
only the kworker running avgs_work on the CPU.
Although commit 1b69ac6b40eb ("psi: fix aggregation idle shut-off")
avoided the ping-pong wake problem when the worker sleep, psi_avgs_work()
still will always re-arm the avgs_work, so shutoff is not working.
This patch changes to use PSI_STATE_RESCHEDULE to flag whether to
re-arm avgs_work in get_recent_times(). For the current CPU, we re-arm
avgs_work only when (NR_RUNNING > 1 || NR_IOWAIT > 0 || NR_MEMSTALL > 0),
for other CPUs we can just check PSI_NONIDLE delta. The new flag
is only used in psi_avgs_work(), so we check in get_recent_times()
that current_work() is avgs_work.
One potential problem is that the brief period of non-idle time
incurred between the aggregation run and the kworker's dequeue will
be stranded in the per-cpu buckets until avgs_work run next time.
The buckets can hold 4s worth of time, and future activity will wake
the avgs_work with a 2s delay, giving us 2s worth of data we can leave
behind when shut off the avgs_work. If the kworker run other works after
avgs_work shut off and doesn't have any scheduler activities for 2s,
this maybe a problem.
Reported-by: Pavan Kondeti <quic_pkondeti@quicinc.com>
Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Suren Baghdasaryan <surenb@google.com>
Tested-by: Chengming Zhou <zhouchengming@bytedance.com>
Link: https://lore.kernel.org/r/20221014110551.22695-1-zhouchengming@bytedance.com
|
|
When a pending event exists and growth is less than the threshold, the
current logic is to skip this trigger without generating event. However,
from e6df4ead85d9 ("psi: fix possible trigger missing in the window"),
our purpose is to generate event as long as pending event exists and the
rate meets the limit, no matter what growth is.
This patch handles this case properly.
Fixes: e6df4ead85d9 ("psi: fix possible trigger missing in the window")
Signed-off-by: Hao Lee <haolee.swjtu@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Suren Baghdasaryan <surenb@google.com>
Link: https://lore.kernel.org/r/20220919072356.GA29069@haolee.io
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael Wysocki:
"These make the intel_pstate driver work as expected on all hybrid
platforms to date (regardless of possible platform firmware issues),
fix hybrid sleep on systems using suspend-to-idle by default, make the
generic power domains code handle disabled idle states properly and
update pm-graph.
Specifics:
- Make intel_pstate use what is known about the hardware instead of
relying on information from the platform firmware (ACPI CPPC in
particular) to establish the relationship between the HWP CPU
performance levels and frequencies on all hybrid platforms
available to date (Rafael Wysocki)
- Allow hybrid sleep to use suspend-to-idle as a system suspend
method if it is the current suspend method of choice (Mario
Limonciello)
- Fix handling of unavailable/disabled idle states in the generic
power domains code (Sudeep Holla)
- Update the pm-graph suite of utilities to version 5.10 which is
fixes-mostly and does not add any new features (Todd Brandt)"
* tag 'pm-6.1-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
PM: domains: Fix handling of unavailable/disabled idle states
pm-graph v5.10
cpufreq: intel_pstate: hybrid: Use known scaling factor for P-cores
cpufreq: intel_pstate: Read all MSRs on the target CPU
PM: hibernate: Allow hybrid sleep to work with s2idle
|
|
For maps of type BPF_MAP_TYPE_CPUMAP memory is allocated first before
checking the max_entries argument. If then max_entries is greater than
NR_CPUS additional work needs to be done to free allocated memory before
an error is returned.
This changes moves the check on max_entries before the allocation
happens.
Signed-off-by: Florian Lehner <dev@der-flo.net>
Link: https://lore.kernel.org/r/20221028183405.59554-1-dev@der-flo.net
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
|
|
It's really difficult to debug when cgroup or css refs leak. Let's add a
debug option to force the refcnt function to not be inlined so that they can
be kprobed for debugging.
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
There is a typo in comment for DFS algorithm in bpf/verifier.c. The top
element should not be popped until all its neighbors have been checked.
Fix it.
Fixes: 475fb78fbf48 ("bpf: verifier (add branch/goto checks)")
Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20221027034458.2925218-1-xukuohai@huaweicloud.com
|
|
Use the event group trees to iterate only perf_tracepoint events.
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
|
|
There have been various issues and limitations with the way perf uses
(task) contexts to track events. Most notable is the single hardware
PMU task context, which has resulted in a number of yucky things (both
proposed and merged).
Notably:
- HW breakpoint PMU
- ARM big.little PMU / Intel ADL PMU
- Intel Branch Monitoring PMU
- AMD IBS PMU
- S390 cpum_cf PMU
- PowerPC trace_imc PMU
*Current design:*
Currently we have a per task and per cpu perf_event_contexts:
task_struct::perf_events_ctxp[] <-> perf_event_context <-> perf_cpu_context
^ | ^ | ^
`---------------------------------' | `--> pmu ---'
v ^
perf_event ------'
Each task has an array of pointers to a perf_event_context. Each
perf_event_context has a direct relation to a PMU and a group of
events for that PMU. The task related perf_event_context's have a
pointer back to that task.
Each PMU has a per-cpu pointer to a per-cpu perf_cpu_context, which
includes a perf_event_context, which again has a direct relation to
that PMU, and a group of events for that PMU.
The perf_cpu_context also tracks which task context is currently
associated with that CPU and includes a few other things like the
hrtimer for rotation etc.
Each perf_event is then associated with its PMU and one
perf_event_context.
*Proposed design:*
New design proposed by this patch reduce to a single task context and
a single CPU context but adds some intermediate data-structures:
task_struct::perf_event_ctxp -> perf_event_context <- perf_cpu_context
^ | ^ ^
`---------------------------' | |
| | perf_cpu_pmu_context <--.
| `----. ^ |
| | | |
| v v |
| ,--> perf_event_pmu_context |
| | |
| | |
v v |
perf_event ---> pmu ----------------'
With the new design, perf_event_context will hold all events for all
pmus in the (respective pinned/flexible) rbtrees. This can be achieved
by adding pmu to rbtree key:
{cpu, pmu, cgroup, group_index}
Each perf_event_context carries a list of perf_event_pmu_context which
is used to hold per-pmu-per-context state. For example, it keeps track
of currently active events for that pmu, a pmu specific task_ctx_data,
a flag to tell whether rotation is required or not etc.
Additionally, perf_cpu_pmu_context is used to hold per-pmu-per-cpu
state like hrtimer details to drive the event rotation, a pointer to
perf_event_pmu_context of currently running task and some other
ancillary information.
Each perf_event is associated to it's pmu, perf_event_context and
perf_event_pmu_context.
Further optimizations to current implementation are possible. For
example, ctx_resched() can be optimized to reschedule only single pmu
events.
Much thanks to Ravi for picking this up and pushing it towards
completion.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Co-developed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20221008062424.313-1-ravi.bangoria@amd.com
|
|
The do_set_cpus_allowed() function is used by either kthread_bind() or
select_fallback_rq(). In both cases the user affinity (if any) should be
destroyed too.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220922180041.1768141-6-longman@redhat.com
|
|
It was found that the user requested affinity via sched_setaffinity()
can be easily overwritten by other kernel subsystems without an easy way
to reset it back to what the user requested. For example, any change
to the current cpuset hierarchy may reset the cpumask of the tasks in
the affected cpusets to the default cpuset value even if those tasks
have pre-existing user requested affinity. That is especially easy to
trigger under a cgroup v2 environment where writing "+cpuset" to the
root cgroup's cgroup.subtree_control file will reset the cpus affinity
of all the processes in the system.
That is problematic in a nohz_full environment where the tasks running
in the nohz_full CPUs usually have their cpus affinity explicitly set
and will behave incorrectly if cpus affinity changes.
Fix this problem by looking at user_cpus_ptr in __set_cpus_allowed_ptr()
and use it to restrcit the given cpumask unless there is no overlap. In
that case, it will fallback to the given one. The SCA_USER flag is
reused to indicate intent to set user_cpus_ptr and so user_cpus_ptr
masking should be skipped. In addition, masking should also be skipped
if any of the SCA_MIGRATE_* flag is set.
All callers of set_cpus_allowed_ptr() will be affected by this change.
A scratch cpumask is added to percpu runqueues structure for doing
additional masking when user_cpus_ptr is set.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220922180041.1768141-4-longman@redhat.com
|
|
Unconditionally preserve the user requested cpumask on
sched_setaffinity() calls. This allows using it outside of the fairly
narrow restrict_cpus_allowed_ptr() use-case and fix some cpuset issues
that currently suffer destruction of cpumasks.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220922180041.1768141-3-longman@redhat.com
|
|
In order to prepare for passing through additional data through the
affinity call-chains, convert the mask and flags argument into a
structure.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220922180041.1768141-5-longman@redhat.com
|
|
affine_move_task() assumes task_rq_lock() has been called and it does
an implicit task_rq_unlock() before returning. Add the appropriate
__releases annotations to make this clear.
A typo error in comment is also fixed.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220922180041.1768141-2-longman@redhat.com
|
|
When evaluating the CPU candidates in the perf domain (pd) containing
the previously used CPU (prev_cpu), find_energy_efficient_cpu()
evaluates the energy of the pd:
- without the task (base_energy)
- with the task placed on prev_cpu (if the task fits)
- with the task placed on the CPU with the highest spare capacity,
prev_cpu being excluded from this set
If prev_cpu is already the CPU with the highest spare capacity,
max_spare_cap_cpu will be the CPU with the second highest spare
capacity.
On an Arm64 Juno-r2, with a workload of 10 tasks at a 10% duty cycle,
when prev_cpu and max_spare_cap_cpu are both valid candidates,
prev_spare_cap > max_spare_cap at ~82%.
Thus the energy of the pd when placing the task on max_spare_cap_cpu
is computed with no possible positive outcome 82% most of the time.
Do not consider max_spare_cap_cpu as a valid candidate if
prev_spare_cap > max_spare_cap.
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20221006081052.3862167-2-pierre.gondois@arm.com
|
|
We do consider thermal pressure in util_fits_cpu() for uclamp_min only.
With the exception of the biggest cores which by definition are the max
performance point of the system and all tasks by definition should fit.
Even under thermal pressure, the capacity of the biggest CPU is the
highest in the system and should still fit every task. Except when it
reaches capacity inversion point, then this is no longer true.
We can handle this by using the inverted capacity as capacity_orig in
util_fits_cpu(). Which not only addresses the problem above, but also
ensure uclamp_max now considers the inverted capacity. Force fitting
a task when a CPU is in this adverse state will contribute to making the
thermal throttling last longer.
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-10-qais.yousef@arm.com
|
|
Check each performance domain to see if thermal pressure is causing its
capacity to be lower than another performance domain.
We assume that each performance domain has CPUs with the same
capacities, which is similar to an assumption made in energy_model.c
We also assume that thermal pressure impacts all CPUs in a performance
domain equally.
If there're multiple performance domains with the same capacity_orig, we
will trigger a capacity inversion if the domain is under thermal
pressure.
The new cpu_in_capacity_inversion() should help users to know when
information about capacity_orig are not reliable and can opt in to use
the inverted capacity as the 'actual' capacity_orig.
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-9-qais.yousef@arm.com
|
|
condition
If the utilization of the woken up task is 0, we skip the energy
calculation because it has no impact.
But if the task is boosted (uclamp_min != 0) will have an impact on task
placement and frequency selection. Only skip if the util is truly
0 after applying uclamp values.
Change uclamp_task_cpu() signature to avoid unnecessary additional calls
to uclamp_eff_get(). feec() is the only user now.
Fixes: 732cd75b8c920 ("sched/fair: Select an energy-efficient CPU on task wake-up")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-8-qais.yousef@arm.com
|
|
So that it is now uclamp aware.
This fixes a major problem of busy tasks capped with UCLAMP_MAX keeping
the system in overutilized state which disables EAS and leads to wasting
energy in the long run.
Without this patch running a busy background activity like JIT
compilation on Pixel 6 causes the system to be in overutilized state
74.5% of the time.
With this patch this goes down to 9.79%.
It also fixes another problem when long running tasks that have their
UCLAMP_MIN changed while running such that they need to upmigrate to
honour the new UCLAMP_MIN value. The upmigration doesn't get triggered
because overutilized state never gets set in this state, hence misfit
migration never happens at tick in this case until the task wakes up
again.
Fixes: af24bde8df202 ("sched/uclamp: Add uclamp support to energy_compute()")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-7-qais.yousef@arm.com
|
|
Use the new util_fits_cpu() to ensure migration margin and capacity
pressure are taken into account correctly when uclamp is being used
otherwise we will fail to consider CPUs as fitting in scenarios where
they should.
s/asym_fits_capacity/asym_fits_cpu/ to better reflect what it does now.
Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-6-qais.yousef@arm.com
|
|
Use the new util_fits_cpu() to ensure migration margin and capacity
pressure are taken into account correctly when uclamp is being used
otherwise we will fail to consider CPUs as fitting in scenarios where
they should.
Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-5-qais.yousef@arm.com
|
|
As reported by Yun Hsiang [1], if a task has its uclamp_min >= 0.8 * 1024,
it'll always pick the previous CPU because fits_capacity() will always
return false in this case.
The new util_fits_cpu() logic should handle this correctly for us beside
more corner cases where similar failures could occur, like when using
UCLAMP_MAX.
We open code uclamp_rq_util_with() except for the clamp() part,
util_fits_cpu() needs the 'raw' values to be passed to it.
Also introduce uclamp_rq_{set, get}() shorthand accessors to get uclamp
value for the rq. Makes the code more readable and ensures the right
rules (use READ_ONCE/WRITE_ONCE) are respected transparently.
[1] https://lists.linaro.org/pipermail/eas-dev/2020-July/001488.html
Fixes: 1d42509e475c ("sched/fair: Make EAS wakeup placement consider uclamp restrictions")
Reported-by: Yun Hsiang <hsiang023167@gmail.com>
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-4-qais.yousef@arm.com
|
|
So that the new uclamp rules in regard to migration margin and capacity
pressure are taken into account correctly.
Fixes: a7008c07a568 ("sched/fair: Make task_fits_capacity() consider uclamp restrictions")
Co-developed-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-3-qais.yousef@arm.com
|
|
fits_capacity() verifies that a util is within 20% margin of the
capacity of a CPU, which is an attempt to speed up upmigration.
But when uclamp is used, this 20% margin is problematic because for
example if a task is boosted to 1024, then it will not fit on any CPU
according to fits_capacity() logic.
Or if a task is boosted to capacity_orig_of(medium_cpu). The task will
end up on big instead on the desired medium CPU.
Similar corner cases exist for uclamp and usage of capacity_of().
Slightest irq pressure on biggest CPU for example will make a 1024
boosted task look like it can't fit.
What we really want is for uclamp comparisons to ignore the migration
margin and capacity pressure, yet retain them for when checking the
_actual_ util signal.
For example, task p:
p->util_avg = 300
p->uclamp[UCLAMP_MIN] = 1024
Will fit a big CPU. But
p->util_avg = 900
p->uclamp[UCLAMP_MIN] = 1024
will not, this should trigger overutilized state because the big CPU is
now *actually* being saturated.
Similar reasoning applies to capping tasks with UCLAMP_MAX. For example:
p->util_avg = 1024
p->uclamp[UCLAMP_MAX] = capacity_orig_of(medium_cpu)
Should fit the task on medium cpus without triggering overutilized
state.
Inlined comments expand more on desired behavior in more scenarios.
Introduce new util_fits_cpu() function which encapsulates the new logic.
The new function is not used anywhere yet, but will be used to update
various users of fits_capacity() in later patches.
Fixes: af24bde8df202 ("sched/uclamp: Add uclamp support to energy_compute()")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220804143609.515789-2-qais.yousef@arm.com
|
