All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: mingo@kernel.org, alexander.shishkin@linux.intel.com, eranian@google.com
Cc: linux-kernel@vger.kernel.org, vince@deater.net,
	dvyukov@google.com, andi@firstfloor.org, jolsa@redhat.com,
	peterz@infradead.org
Subject: [RFC][PATCH 11/12] perf: Specialize perf_event_exit_task()
Date: Mon, 11 Jan 2016 17:25:09 +0100	[thread overview]
Message-ID: <20160111163229.352636767@infradead.org> (raw)
In-Reply-To: 20160111162458.427203780@infradead.org

[-- Attachment #1: peterz-perf-fixes-14.patch --]
[-- Type: text/plain, Size: 1953 bytes --]

The perf_remove_from_context() usage in __perf_event_exit_task() is
different from the other usages in that this site has already
detached and scheduled out the task context.

This will stand in the way of stronger assertions checking the (task)
context scheduling invariants.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 kernel/events/core.c |   18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -8726,7 +8726,13 @@ __perf_event_exit_task(struct perf_event
 	 * Do destroy all inherited groups, we don't care about those
 	 * and being thorough is better.
 	 */
-	perf_remove_from_context(child_event, !!child_event->parent);
+	raw_spin_lock_irq(&child_ctx->lock);
+	WARN_ON_ONCE(child_ctx->is_active);
+
+	if (!!child_event->parent)
+		perf_group_detach(child_event);
+	list_del_event(child_event, child_ctx);
+	raw_spin_unlock_irq(&child_ctx->lock);
 
 	/*
 	 * It can happen that the parent exits first, and has events
@@ -8746,17 +8752,15 @@ static void perf_event_exit_task_context
 {
 	struct perf_event *child_event, *next;
 	struct perf_event_context *child_ctx, *clone_ctx = NULL;
-	unsigned long flags;
 
 	if (likely(!child->perf_event_ctxp[ctxn]))
 		return;
 
-	local_irq_save(flags);
+	local_irq_disable();
+	WARN_ON_ONCE(child != current);
 	/*
 	 * We can't reschedule here because interrupts are disabled,
-	 * and either child is current or it is a task that can't be
-	 * scheduled, so we are now safe from rescheduling changing
-	 * our context.
+	 * and child must be current.
 	 */
 	child_ctx = rcu_dereference_raw(child->perf_event_ctxp[ctxn]);
 
@@ -8776,7 +8780,7 @@ static void perf_event_exit_task_context
 	 */
 	clone_ctx = unclone_ctx(child_ctx);
 	update_context_time(child_ctx);
-	raw_spin_unlock_irqrestore(&child_ctx->lock, flags);
+	raw_spin_unlock_irq(&child_ctx->lock);
 
 	if (clone_ctx)
 		put_ctx(clone_ctx);

  parent reply	other threads:[~2016-01-11 16:34 UTC|newest]

Thread overview: 67+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-11 16:24 [RFC][PATCH 00/12] various perf fixes Peter Zijlstra
2016-01-11 16:24 ` [RFC][PATCH 01/12] perf: Add lockdep assertions Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 02/12] perf: Fix cgroup event scheduling Peter Zijlstra
2016-01-11 19:43   ` Stephane Eranian
2016-01-11 22:03     ` Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 03/12] perf: Fix cgroup scheduling in enable_on_exec Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 04/12] perf: Remove stale comment Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 05/12] perf: Fix enable_on_exec event scheduling Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 06/12] perf: Use task_ctx_sched_out() Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 07/12] perf: Simplify/fix perf_event_enable() event scheduling Peter Zijlstra
2016-03-08 10:04   ` James Morse
2016-03-08 10:26     ` Peter Zijlstra
2016-03-08 10:42       ` James Morse
2016-03-08 11:29         ` Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 08/12] perf: Optimize perf_sched_events usage Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 09/12] perf: Make ctx->is_active and cpuctx->task_ctx consistent Peter Zijlstra
2016-01-11 16:25 ` [RFC][PATCH 10/12] perf: Fix task context scheduling Peter Zijlstra
2016-01-11 16:25 ` Peter Zijlstra [this message]
2016-01-11 16:25 ` [RFC][PATCH 12/12] perf: Collapse and fix event_function_call() users Peter Zijlstra
2016-01-13 10:50   ` Peter Zijlstra
2016-01-13 13:46     ` Alexander Shishkin
2016-01-13 17:30       ` Peter Zijlstra
2016-01-13 15:00   ` Alexander Shishkin
2016-01-13 15:38     ` Alexander Shishkin
2016-01-13 18:10     ` Peter Zijlstra
2016-01-13 20:44       ` Peter Zijlstra
2016-01-14 10:44     ` Peter Zijlstra
2016-01-14 16:30       ` Peter Zijlstra
2016-02-17 22:38   ` Sasha Levin
2016-02-18  7:46     ` Peter Zijlstra
2016-02-18 12:37       ` Peter Zijlstra
2016-01-11 18:44 ` [RFC][PATCH 00/12] various perf fixes Dmitry Vyukov
2016-01-11 19:56   ` Andi Kleen
2016-01-11 22:00   ` Peter Zijlstra
2016-01-12  9:59     ` Ingo Molnar
2016-01-12 10:11 ` Ingo Molnar
2016-01-12 10:57   ` Dmitry Vyukov
2016-01-12 11:00     ` Dmitry Vyukov
2016-01-12 11:01       ` Dmitry Vyukov
2016-01-12 11:26         ` Dmitry Vyukov
2016-01-12 11:35           ` Dmitry Vyukov
2016-01-12 12:01           ` Peter Zijlstra
2016-01-13 15:18           ` Alexander Shishkin
2016-01-13 15:22             ` Dmitry Vyukov
2016-01-13 15:35               ` Alexander Shishkin
2016-01-14  9:35           ` Peter Zijlstra
2016-01-14 10:05             ` Dmitry Vyukov
2016-02-15 15:04               ` Dmitry Vyukov
2016-02-15 15:38                 ` Peter Zijlstra
2016-02-15 15:47                   ` Dmitry Vyukov
2016-02-15 16:01                   ` Vince Weaver
2016-02-15 16:17               ` Peter Zijlstra
2016-02-15 16:29                 ` Dmitry Vyukov
2016-02-15 16:29               ` Peter Zijlstra
2016-02-15 16:35                 ` Dmitry Vyukov
2016-02-15 16:38                   ` Dmitry Vyukov
2016-02-15 16:52                     ` Peter Zijlstra
2016-02-15 17:04                       ` Dmitry Vyukov
2016-02-15 17:07                         ` Peter Zijlstra
2016-02-15 17:45                           ` Dmitry Vyukov
2016-02-15 18:01                             ` Peter Zijlstra
2016-02-15 18:33                               ` Dmitry Vyukov
2016-02-15 16:41                   ` Peter Zijlstra
2016-02-15 16:54                     ` Dmitry Vyukov
2016-02-15 16:59                       ` Peter Zijlstra
2016-01-12 13:13   ` Peter Zijlstra
2016-01-12 13:31     ` 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=20160111163229.352636767@infradead.org \
    --to=peterz@infradead.org \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=andi@firstfloor.org \
    --cc=dvyukov@google.com \
    --cc=eranian@google.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.