ftrace: arm64: move from REGS to ARGS

This commit replaces arm64's support for FTRACE_WITH_REGS with support
for FTRACE_WITH_ARGS. This removes some overhead and complexity, and
removes some latent issues with inconsistent presentation of struct
pt_regs (which can only be reliably saved/restored at exception
boundaries).

FTRACE_WITH_REGS has been supported on arm64 since commit:

  3b23e4991fb66f6d ("arm64: implement ftrace with regs")

As noted in the commit message, the major reasons for implementing
FTRACE_WITH_REGS were:

(1) To make it possible to use the ftrace graph tracer with pointer
    authentication, where it's necessary to snapshot/manipulate the LR
    before it is signed by the instrumented function.

(2) To make it possible to implement LIVEPATCH in future, where we need
    to hook function entry before an instrumented function manipulates
    the stack or argument registers. Practically speaking, we need to
    preserve the argument/return registers, PC, LR, and SP.

Neither of these need a struct pt_regs, and only require the set of
registers which are live at function call/return boundaries. Our calling
convention is defined by "Procedure Call Standard for the Arm® 64-bit
Architecture (AArch64)" (AKA "AAPCS64"), which can currently be found
at:

  https://github.com/ARM-software/abi-aa/blob/main/aapcs64/aapcs64.rst

Per AAPCS64, all function call argument and return values are held in
the following GPRs:

* X0 - X7 : parameter / result registers
* X8      : indirect result location register
* SP      : stack pointer (AKA SP)

Additionally, ad function call boundaries, the following GPRs hold
context/return information:

* X29 : frame pointer (AKA FP)
* X30 : link register (AKA LR)

... and for ftrace we need to capture the instrumented address:

 * PC  : program counter

No other GPRs are relevant, as none of the other arguments hold
parameters or return values:

* X9  - X17 : temporaries, may be clobbered
* X18       : shadow call stack pointer (or temorary)
* X19 - X28 : callee saved

This patch implements FTRACE_WITH_ARGS for arm64, only saving/restoring
the minimal set of registers necessary. This is always sufficient to
manipulate control flow (e.g. for live-patching) or to manipulate
function arguments and return values.

This reduces the necessary stack usage from 336 bytes for pt_regs down
to 112 bytes for ftrace_regs + 32 bytes for two frame records, freeing
up 188 bytes. This could be reduced further with changes to the
unwinder.

As there is no longer a need to save different sets of registers for
different features, we no longer need distinct `ftrace_caller` and
`ftrace_regs_caller` trampolines. This allows the trampoline assembly to
be simpler, and simplifies code which previously had to handle the two
trampolines.

I've tested this with the ftrace selftests, where there are no
unexpected failures.

Co-developed-by: Florent Revest <revest@chromium.org>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Florent Revest <revest@chromium.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Will Deacon <will@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20221103170520.931305-5-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>

1 files changed, 48 insertions, 34 deletions

diff --git a/arch/arm64/kernel/ftrace.c b/arch/arm64/kernel/ftrace.c
index 8745175f4a75..5cf990d052ba 100644
--- a/arch/arm64/kernel/ftrace.c
+++ b/arch/arm64/kernel/ftrace.c
@@ -17,6 +17,49 @@
 #include <asm/insn.h>
 #include <asm/patching.h>
 
+#ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
+struct fregs_offset {
+	const char *name;
+	int offset;
+};
+
+#define FREGS_OFFSET(n, field)				\
+{							\
+	.name = n,					\
+	.offset = offsetof(struct ftrace_regs, field),	\
+}
+
+static const struct fregs_offset fregs_offsets[] = {
+	FREGS_OFFSET("x0", regs[0]),
+	FREGS_OFFSET("x1", regs[1]),
+	FREGS_OFFSET("x2", regs[2]),
+	FREGS_OFFSET("x3", regs[3]),
+	FREGS_OFFSET("x4", regs[4]),
+	FREGS_OFFSET("x5", regs[5]),
+	FREGS_OFFSET("x6", regs[6]),
+	FREGS_OFFSET("x7", regs[7]),
+	FREGS_OFFSET("x8", regs[8]),
+
+	FREGS_OFFSET("x29", fp),
+	FREGS_OFFSET("x30", lr),
+	FREGS_OFFSET("lr", lr),
+
+	FREGS_OFFSET("sp", sp),
+	FREGS_OFFSET("pc", pc),
+};
+
+int ftrace_regs_query_register_offset(const char *name)
+{
+	for (int i = 0; i < ARRAY_SIZE(fregs_offsets); i++) {
+		const struct fregs_offset *roff = &fregs_offsets[i];
+		if (!strcmp(roff->name, name))
+			return roff->offset;
+	}
+
+	return -EINVAL;
+}
+#endif
+
 #ifdef CONFIG_DYNAMIC_FTRACE
 /*
  * Replace a single instruction, which may be a branch or NOP.
@@ -70,9 +113,6 @@ static struct plt_entry *get_ftrace_plt(struct module *mod, unsigned long addr)
 
 	if (addr == FTRACE_ADDR)
 		return &plt[FTRACE_PLT_IDX];
-	if (addr == FTRACE_REGS_ADDR &&
-	    IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_REGS))
-		return &plt[FTRACE_REGS_PLT_IDX];
 #endif
 	return NULL;
 }
@@ -154,25 +194,7 @@ int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
 	return ftrace_modify_code(pc, old, new, true);
 }
 
-#ifdef CONFIG_DYNAMIC_FTRACE_WITH_REGS
-int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
-			unsigned long addr)
-{
-	unsigned long pc = rec->ip;
-	u32 old, new;
-
-	if (!ftrace_find_callable_addr(rec, NULL, &old_addr))
-		return -EINVAL;
-	if (!ftrace_find_callable_addr(rec, NULL, &addr))
-		return -EINVAL;
-
-	old = aarch64_insn_gen_branch_imm(pc, old_addr,
-					  AARCH64_INSN_BRANCH_LINK);
-	new = aarch64_insn_gen_branch_imm(pc, addr, AARCH64_INSN_BRANCH_LINK);
-
-	return ftrace_modify_code(pc, old, new, true);
-}
-
+#ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
 /*
  * The compiler has inserted two NOPs before the regular function prologue.
  * All instrumented functions follow the AAPCS, so x0-x8 and x19-x30 are live,
@@ -228,7 +250,7 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
 	 *
 	 * Note: 'mod' is only set at module load time.
 	 */
-	if (!IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_REGS) &&
+	if (!IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_ARGS) &&
 	    IS_ENABLED(CONFIG_ARM64_MODULE_PLTS) && mod) {
 		return aarch64_insn_patch_text_nosync((void *)pc, new);
 	}
@@ -279,19 +301,11 @@ void prepare_ftrace_return(unsigned long self_addr, unsigned long *parent,
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 
-#ifdef CONFIG_DYNAMIC_FTRACE_WITH_REGS
+#ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	/*
-	 * When DYNAMIC_FTRACE_WITH_REGS is selected, `fregs` can never be NULL
-	 * and arch_ftrace_get_regs(fregs) will always give a non-NULL pt_regs
-	 * in which we can safely modify the LR.
-	 */
-	struct pt_regs *regs = arch_ftrace_get_regs(fregs);
-	unsigned long *parent = (unsigned long *)&procedure_link_pointer(regs);
-
-	prepare_ftrace_return(ip, parent, frame_pointer(regs));
+	prepare_ftrace_return(ip, &fregs->lr, fregs->fp);
 }
 #else
 /*
@@ -323,6 +337,6 @@ int ftrace_disable_ftrace_graph_caller(void)
 {
 	return ftrace_modify_graph_caller(false);
 }
-#endif /* CONFIG_DYNAMIC_FTRACE_WITH_REGS */
+#endif /* CONFIG_DYNAMIC_FTRACE_WITH_ARGS */
 #endif /* CONFIG_DYNAMIC_FTRACE */
 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */