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, Arnaldo Carvalho de Melo <acme@infradead.org>,
Jiri Olsa <jolsa@kernel.org>
Subject: Re: [PATCH] perf: Synchronously cleanup child events
Date: Fri, 15 Jan 2016 18:57:59 +0100 [thread overview]
Message-ID: <20160115175759.GL3421@worktop> (raw)
In-Reply-To: <1452866861-17680-1-git-send-email-alexander.shishkin@linux.intel.com>
On Fri, Jan 15, 2016 at 04:07:41PM +0200, Alexander Shishkin wrote:
> int perf_event_release_kernel(struct perf_event *event)
> {
> + struct perf_event *child, *tmp;
> + LIST_HEAD(child_list);
>
> + if (!is_kernel_event(event))
> + perf_remove_from_owner(event);
>
> + event->owner = NULL;
>
> + /*
> + * event::child_mutex nests inside ctx::lock, so move children
> + * to a safe place first and avoid inversion
> + */
> + mutex_lock(&event->child_mutex);
> + list_splice_init(&event->child_list, &child_list);
> + mutex_unlock(&event->child_mutex);
I suspect this races against inherit_event(), like:
inherit_event() perf_event_release_kernel()
if (is_orphaned_event(parent_event) /* false */
event->owner = NULL
mutex_lock(child_mutex);
list_splice
mutex_unlock(child_mutex);
mutex_lock(child_mutex);
list_add_tail
mutex_unlock(child_mutex);
Something like this would fix that I think, not sure its the best way
though...
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -979,8 +979,8 @@ static void put_ctx(struct perf_event_co
* Lock order:
* task_struct::perf_event_mutex
* perf_event_context::mutex
- * perf_event_context::lock
* perf_event::child_mutex;
+ * perf_event_context::lock
* perf_event::mmap_mutex
* mmap_sem
*/
@@ -8956,6 +8958,16 @@ inherit_event(struct perf_event *parent_
if (parent_event->parent)
parent_event = parent_event->parent;
+ WARN_ON_ONCE(parent_event->ctx->parent_ctx);
+ /*
+ * Serialize against perf_event_kernel_release()'s orphanage..
+ */
+ mutex_lock(&parent_event->child_mutex);
+ if (is_orphaned_event(parent_event)) {
+ mutex_unlock(&parent_event->child_mutex);
+ return NULL;
+ }
+
child_event = perf_event_alloc(&parent_event->attr,
parent_event->cpu,
child,
@@ -8964,8 +8976,8 @@ inherit_event(struct perf_event *parent_
if (IS_ERR(child_event))
return child_event;
- if (is_orphaned_event(parent_event) ||
- !atomic_long_inc_not_zero(&parent_event->refcount)) {
+ if (!atomic_long_inc_not_zero(&parent_event->refcount)) {
+ mutex_unlock(&parent_event->child_mutex);
free_event(child_event);
return NULL;
}
@@ -9013,8 +9025,6 @@ inherit_event(struct perf_event *parent_
/*
* Link this into the parent event's child list
*/
- WARN_ON_ONCE(parent_event->ctx->parent_ctx);
- mutex_lock(&parent_event->child_mutex);
list_add_tail(&child_event->child_list, &parent_event->child_list);
mutex_unlock(&parent_event->child_mutex);
next prev parent reply other threads:[~2016-01-15 17:58 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-15 11:22 [PATCH] perf: Cleanup user's child events Alexander Shishkin
2016-01-15 12:54 ` Peter Zijlstra
2016-01-15 13:05 ` Alexander Shishkin
2016-01-15 13:09 ` Peter Zijlstra
2016-01-15 14:07 ` [PATCH] perf: Synchronously cleanup " Alexander Shishkin
2016-01-15 17:57 ` Peter Zijlstra [this message]
2016-01-18 12:07 ` Alexander Shishkin
2016-01-18 12:37 ` Alexander Shishkin
2016-01-18 14:44 ` Peter Zijlstra
2016-01-19 15:12 ` [PATCH v2] " Alexander Shishkin
2016-01-19 20:05 ` Peter Zijlstra
2016-01-19 21:58 ` Alexei Starovoitov
2016-01-20 8:32 ` Peter Zijlstra
2016-01-21 4:55 ` Alexei Starovoitov
2016-01-20 7:04 ` Alexander Shishkin
2016-01-20 8:03 ` Peter Zijlstra
2016-01-22 11:35 ` Alexander Shishkin
2016-01-22 12:12 ` Peter Zijlstra
2016-01-22 12:38 ` Peter Zijlstra
2016-01-22 19:44 ` Alexei Starovoitov
2016-01-25 11:48 ` Peter Zijlstra
2016-01-25 14:54 ` Peter Zijlstra
2016-01-25 21:04 ` Peter Zijlstra
2016-01-26 4:59 ` Alexei Starovoitov
2016-01-26 16:16 ` Peter Zijlstra
2016-01-26 17:24 ` Peter Zijlstra
2016-01-26 23:31 ` Alexei Starovoitov
2016-01-27 9:58 ` Peter Zijlstra
2016-01-27 17:52 ` Alexei Starovoitov
2016-01-29 11:28 ` [tip:perf/urgent] perf/bpf: Convert perf_event_array to use struct file tip-bot for Alexei Starovoitov
2016-01-29 20:01 ` Alexei Starovoitov
2016-01-19 20:07 ` [PATCH v2] perf: Synchronously cleanup child events Peter Zijlstra
2016-01-19 7:45 ` [PATCH] " Ingo Molnar
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=20160115175759.GL3421@worktop \
--to=peterz@infradead.org \
--cc=acme@infradead.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=eranian@google.com \
--cc=jolsa@kernel.org \
--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;
as well as URLs for NNTP newsgroup(s).