[PATCH bpf-next 1/3] arm64: perf: Don't report the caller twice in caller-regs callchains

[Andrii Nakryiko <andrii@kernel.org>]

perf_arch_fetch_caller_regs() sets regs->pc to the return address of
the function it is used in (CALLER_ADDR0), but points regs->regs[29]
at that function's own frame record, whose saved LR is that same
return address. perf_callchain_kernel() reports regs->pc and then
unwinds from the frame record at x29, so the caller shows up twice at
the top of every kernel callchain built from perf_fetch_caller_regs():
tracepoint events, software events emitted by __perf_sw_event_sched()
(context-switches, cgroup-switches, cpu-migrations), and
bpf_get_stack() and bpf_get_stackid() in tracepoint, raw tracepoint,
tp_btf and fentry/fexit programs. E.g., for a tracepoint program
attached to sched_process_fork:

  __traceiter_sched_process_fork+0x44
  __traceiter_sched_process_fork+0x44
  kernel_clone+0x250
  ...

Use __builtin_frame_address(1), the caller's frame record, the same
way kunwind_init_from_caller() does, so that pc and x29 describe the
same frame.

x86 had the same duplicate entry, fixed by commit 6d08340d1e35
("Revert "perf/x86: Always store regs->ip in perf_callchain_kernel()"").

Fixes: 5b09a094f2fb ("arm64: perf: Fix callchain parse error with kernel tracepoint events")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
 arch/arm64/include/asm/perf_event.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/perf_event.h b/arch/arm64/include/asm/perf_event.h
index ee45b4e77347..13efb46f0fe8 100644
--- a/arch/arm64/include/asm/perf_event.h
+++ b/arch/arm64/include/asm/perf_event.h
@@ -15,7 +15,7 @@
 
 #define perf_arch_fetch_caller_regs(regs, __ip) { \
 	(regs)->pc = (__ip);    \
-	(regs)->regs[29] = (unsigned long) __builtin_frame_address(0); \
+	(regs)->regs[29] = (unsigned long) __builtin_frame_address(1); \
 	(regs)->sp = current_stack_pointer; \
 	(regs)->pstate = PSR_MODE_EL1h;	\
 }
-- 
2.53.1-Meta