2017-11-01 10:33:59

by Wang Nan

[permalink] [raw]
Subject: Re: [PATCH 1/2] perf mmap: Fix perf backward recording



On 2017/11/1 17:49, Namhyung Kim wrote:
> Hi,
>
> On Wed, Nov 01, 2017 at 05:53:26AM +0000, Wang Nan wrote:
>> perf record backward recording doesn't work as we expected: it never
>> overwrite when ring buffer full.
>>
>> Test:
>>
>> (Run a busy printing python task background like this:
>>
>> while True:
>> print 123
>>
>> send SIGUSR2 to perf to capture snapshot.)

[SNIP]

>>
>> Signed-off-by: Wang Nan <[email protected]>
>> ---
>> tools/perf/util/evlist.c | 8 +++++++-
>> 1 file changed, 7 insertions(+), 1 deletion(-)
>>
>> diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
>> index c6c891e..4c5daba 100644
>> --- a/tools/perf/util/evlist.c
>> +++ b/tools/perf/util/evlist.c
>> @@ -799,22 +799,28 @@ perf_evlist__should_poll(struct perf_evlist *evlist __maybe_unused,
>> }
>>
>> static int perf_evlist__mmap_per_evsel(struct perf_evlist *evlist, int idx,
>> - struct mmap_params *mp, int cpu_idx,
>> + struct mmap_params *_mp, int cpu_idx,
>> int thread, int *_output, int *_output_backward)
>> {
>> struct perf_evsel *evsel;
>> int revent;
>> int evlist_cpu = cpu_map__cpu(evlist->cpus, cpu_idx);
>> + struct mmap_params *mp;
>>
>> evlist__for_each_entry(evlist, evsel) {
>> struct perf_mmap *maps = evlist->mmap;
>> + struct mmap_params rdonly_mp;
>> int *output = _output;
>> int fd;
>> int cpu;
>>
>> + mp = _mp;
>> if (evsel->attr.write_backward) {
>> output = _output_backward;
>> maps = evlist->backward_mmap;
>> + rdonly_mp = *_mp;
>> + rdonly_mp.prot &= ~PROT_WRITE;
>> + mp = &rdonly_mp;
>>
>> if (!maps) {
>> maps = perf_evlist__alloc_mmap(evlist);
>> --
> What about this instead (not tested)?
>
>
> diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
> index c6c891e154a6..27ebe355e794 100644
> --- a/tools/perf/util/evlist.c
> +++ b/tools/perf/util/evlist.c
> @@ -838,6 +838,11 @@ static int perf_evlist__mmap_per_evsel(struct perf_evlist *evlist, int idx,
> if (*output == -1) {
> *output = fd;
>
> + if (evsel->attr.write_backward)
> + mp->prot = PROT_READ;
> + else
> + mp->prot = PROT_READ | PROT_WRITE;
> +

If evlist->overwrite is true, PROT_WRITE should be unset even if
write_backward is
not set. If you want to delay the setting of mp->prot, you need to consider
both evlist->overwrite and evsel->attr.write_backward.

Then why not removing mp->prot totally, and add a prot argument to
perf_mmap__mmap,
since prot setting is always decided independently?

> if (perf_mmap__mmap(&maps[idx], mp, *output) < 0)
> return -1;
> } else {
> @@ -1058,9 +1063,7 @@ int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,
> struct perf_evsel *evsel;
> const struct cpu_map *cpus = evlist->cpus;
> const struct thread_map *threads = evlist->threads;
> - struct mmap_params mp = {
> - .prot = PROT_READ | (overwrite ? 0 : PROT_WRITE),
> - };
> + struct mmap_params mp;
>
> if (!evlist->mmap)
> evlist->mmap = perf_evlist__alloc_mmap(evlist);

The 'overwrite' argument in perf_evlist__mmap_ex() controls
evlist->overwrite.

Thank you.


From 1582856934106290327@xxx Wed Nov 01 09:53:24 +0000 2017
X-GM-THRID: 1582841972981207976
X-Gmail-Labels: Inbox,Category Forums