Re: [PATCH] perf/x86: always include regs->ip in callchain

[Song Liu]

> On May 22, 2019, at 6:58 AM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> 
> On Tue, May 21, 2019 at 01:48:13PM -0700, Song Liu wrote:
>> Commit d15d356887e7 removes regs->ip for !perf_hw_regs(regs) case. This
>> breaks tests like test_stacktrace_map from selftests/bpf/tests_prog.
> 
> That test is broken by something else; just the one entry is wrong too.
> 
> That said, yes the patch below is actually correct, but the above
> description is misleading at best.

How about we change it to: 

Commit d15d356887e7 removes regs->ip for !perf_hw_regs(regs) case. This
patch adds regs->ip back.


Shall I send v2 with the change? 

Thanks,
Song

> 
>> This patch adds regs->ip back.
>> 
>> Fixes: d15d356887e7 ("perf/x86: Make perf callchains work without CONFIG_FRAME_POINTER")
>> Cc: Kairui Song <kasong@xxxxxxxxxx>
>> Cc: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
>> Signed-off-by: Song Liu <songliubraving@xxxxxx>
>> ---
>> arch/x86/events/core.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>> 
>> diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
>> index f315425d8468..7b8a9eb4d5fd 100644
>> --- a/arch/x86/events/core.c
>> +++ b/arch/x86/events/core.c
>> @@ -2402,9 +2402,9 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re
>> 		return;
>> 	}
>> 
>> +	if (perf_callchain_store(entry, regs->ip))
>> +		return;
>> 	if (perf_hw_regs(regs)) {
>> -		if (perf_callchain_store(entry, regs->ip))
>> -			return;
>> 		unwind_start(&state, current, regs, NULL);
>> 	} else {
>> 		unwind_start(&state, current, NULL, (void *)regs->sp);
>> -- 
>> 2.17.1
>>