[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87a7n070tj.fsf@intel.com>
Date: Fri, 26 Oct 2018 16:19:52 -0700
From: Vinicius Costa Gomes <vinicius.gomes@...el.com>
To: Adrian Hunter <adrian.hunter@...el.com>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Ingo Molnar <mingo@...nel.org>
Cc: Clark Williams <williams@...hat.com>, linux-kernel@...r.kernel.org,
linux-perf-users@...r.kernel.org,
Arnaldo Carvalho de Melo <acme@...hat.com>,
David Ahern <dsahern@...il.com>, Jiri Olsa <jolsa@...nel.org>,
Namhyung Kim <namhyung@...nel.org>,
Wang Nan <wangnan0@...wei.com>
Subject: Re: [PATCH 11/11] perf tools: Stop fallbacking to kallsyms for vdso symbols lookup
Hi,
Adrian Hunter <adrian.hunter@...el.com> writes:
> On 18/10/18 1:55 AM, Arnaldo Carvalho de Melo wrote:
>> From: Arnaldo Carvalho de Melo <acme@...hat.com>
>>
>> David reports that:
>>
>> <quote>
>> Perf has this hack where it uses the kernel symbol map as a backup when
>> a symbol can't be found in the user's symbol table(s).
>
> I don't think this is a complete fix because it exposes new problems.
This commit broke function name resolution for 'perf record -g' for me.
What I mean is, with this commit applied:
$ ./tools/perf/perf record -g -- sleep 1
$ ./tools/perf/perf report
'perf report' doesn't seem to be able to show the function names of the
trace.
If I revert this commit, function names are resolved fine.
> This code caters for branches from kernel space to user space and vice
> versa. That is, since there is only one cpumode so it is certain to be
> wrong for either 'ip' or 'addr' when they are not both in the kernel
> or both in userspace.
>
>>
>>
>> Cc: Adrian Hunter <adrian.hunter@...el.com>
>> Cc: David Ahern <dsahern@...il.com>
>> Cc: Jiri Olsa <jolsa@...nel.org>
>> Cc: Namhyung Kim <namhyung@...nel.org>
>> Cc: Wang Nan <wangnan0@...wei.com>
>> Link: https://lkml.kernel.org/n/tip-cs7skq9pp0kjypiju6o7trse@git.kernel.org
>> Signed-off-by: Arnaldo Carvalho de Melo <acme@...hat.com>
>> ---
>> tools/perf/util/event.c | 21 ++-------------------
>> 1 file changed, 2 insertions(+), 19 deletions(-)
>>
>> diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
>> index 0988eb3b844b..bc646185f8d9 100644
>> --- a/tools/perf/util/event.c
>> +++ b/tools/perf/util/event.c
>> @@ -1561,26 +1561,9 @@ struct map *thread__find_map(struct thread *thread, u8 cpumode, u64 addr,
>>
>> return NULL;
>> }
>> -try_again:
>> +
>> al->map = map_groups__find(mg, al->addr);
>> - if (al->map == NULL) {
>> - /*
>> - * If this is outside of all known maps, and is a negative
>> - * address, try to look it up in the kernel dso, as it might be
>> - * a vsyscall or vdso (which executes in user-mode).
>> - *
>> - * XXX This is nasty, we should have a symbol list in the
>> - * "[vdso]" dso, but for now lets use the old trick of looking
>> - * in the whole kernel symbol list.
>> - */
>> - if (cpumode == PERF_RECORD_MISC_USER && machine &&
>> - mg != &machine->kmaps &&
>> - machine__kernel_ip(machine, al->addr)) {
>> - mg = &machine->kmaps;
>> - load_map = true;
>> - goto try_again;
>> - }
>> - } else {
>> + if (al->map != NULL) {
>> /*
>> * Kernel maps might be changed when loading symbols so loading
>> * must be done prior to using kernel maps.
>>
--
Vinicius
Powered by blists - more mailing lists