From: Peter Zijlstra <peterz@infradead.org>
To: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>,
linux-kernel@vger.kernel.org, vince@deater.net,
eranian@google.com, johannes@sipsolutions.net,
Arnaldo Carvalho de Melo <acme@infradead.org>
Subject: Re: [PATCH 4/7] perf: Free aux pages in unmap path
Date: Fri, 4 Dec 2015 23:17:23 +0100 [thread overview]
Message-ID: <20151204221723.GR11639@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <20151204170206.GI17308@twins.programming.kicks-ass.net>
On Fri, Dec 04, 2015 at 06:02:06PM +0100, Peter Zijlstra wrote:
> The only solution I've come up with is:
>
> struct rb_aux *aux = rb->aux;
>
> if (aux && vma->vm_pgoff == aux->pgoff) {
> ctx = perf_event_ctx_lock(event);
Can't do this at all, see the comment in put_event(). perf_read_group()
accesses user memory (and hence causes faults, which in turn take
mmap_sem) while holding ctx::mutex.
So neither this, not what you proposed can work.
Will need moar thinking.
> if (!atomic_dec_and_mutex_lock(&aux->mmap_count, &event->mmap_mutex) {
> /* we now hold both ctx::mutex and event::mmap_mutex */
> rb->aux = NULL;
> ring_buffer_put(rb); /* aux had a reference */
> _perf_event_stop(event);
> ring_buffer_put_aux(aux); /* should be last */
> mutex_unlock(&event->mmap_mutex);
> }
> mutex_unlock(&ctx->mutex);
> }
>
>
next prev parent reply other threads:[~2015-12-04 22:17 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-03 10:32 [PATCH 0/7] perf: Untangle aux refcounting Alexander Shishkin
2015-12-03 10:32 ` [PATCH 1/7] perf: Refuse to begin aux transaction after aux_mmap_count drops Alexander Shishkin
2015-12-03 10:32 ` [PATCH 2/7] perf: Generalize task_function_call()ers Alexander Shishkin
2015-12-03 17:34 ` Peter Zijlstra
2015-12-08 16:42 ` Alexander Shishkin
2015-12-08 16:57 ` Peter Zijlstra
2015-12-17 13:40 ` Peter Zijlstra
2015-12-17 14:25 ` Alexander Shishkin
2015-12-17 15:07 ` Peter Zijlstra
2015-12-18 9:01 ` Peter Zijlstra
2015-12-18 15:07 ` Alexander Shishkin
2015-12-18 16:47 ` Peter Zijlstra
2015-12-18 17:41 ` Alexander Shishkin
2015-12-21 14:39 ` Alexander Shishkin
2016-01-11 10:44 ` Alexander Shishkin
2015-12-03 10:32 ` [PATCH 3/7] perf: Add a helper to stop running events Alexander Shishkin
2015-12-03 10:32 ` [PATCH 4/7] perf: Free aux pages in unmap path Alexander Shishkin
2015-12-04 17:02 ` Peter Zijlstra
2015-12-04 22:17 ` Peter Zijlstra [this message]
2015-12-07 16:16 ` Peter Zijlstra
2015-12-09 9:57 ` Alexander Shishkin
2015-12-09 10:56 ` Peter Zijlstra
2015-12-10 11:20 ` Alexander Shishkin
2015-12-10 12:58 ` Alexander Shishkin
2015-12-03 10:32 ` [PATCH 5/7] perf: Document aux api usage Alexander Shishkin
2015-12-03 20:36 ` Mathieu Poirier
2015-12-03 10:32 ` [PATCH 6/7] perf/x86/intel/pt: Move transaction start/stop to pmu start/stop callbacks Alexander Shishkin
2015-12-03 10:32 ` [PATCH 7/7] perf/x86/intel/bts: Move transaction start/stop to " Alexander Shishkin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20151204221723.GR11639@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=acme@infradead.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=eranian@google.com \
--cc=johannes@sipsolutions.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=vince@deater.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome