public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: paulus <paulus@samba.org>,
	stephane eranian <eranian@googlemail.com>,
	Robert Richter <robert.richter@amd.com>,
	Will Deacon <will.deacon@arm.com>,
	Paul Mundt <lethal@linux-sh.org>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Cyrill Gorcunov <gorcunov@gmail.com>,
	Lin Ming <ming.m.lin@intel.com>,
	Yanmin <yanmin_zhang@linux.intel.com>,
	Deng-Cheng Zhu <dengcheng.zhu@gmail.com>,
	David Miller <davem@davemloft.net>, Ingo Molnar <mingo@elte.hu>
Cc: linux-kernel@vger.kernel.org, Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: [RFC][PATCH 18/19] perf: Provide a separate task context for swevents
Date: Tue, 07 Sep 2010 18:46:51 +0200	[thread overview]
Message-ID: <20100907165012.227790649@chello.nl> (raw)
In-Reply-To: 20100907164633.879932949@chello.nl

[-- Attachment #1: perf-swevent-context.patch --]
[-- Type: text/plain, Size: 2332 bytes --]

Since software events are always schedulable, mixing them up with
hardware events (who are not) can lead to funny scheduling oddities.

Giving them their own context solves this.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/sched.h  |    1 +
 kernel/hw_breakpoint.c |    2 ++
 kernel/perf_event.c    |    8 ++++++++
 3 files changed, 11 insertions(+)

Index: linux-2.6/include/linux/sched.h
===================================================================
--- linux-2.6.orig/include/linux/sched.h
+++ linux-2.6/include/linux/sched.h
@@ -1164,6 +1164,7 @@ struct rcu_node;
 enum perf_event_task_context {
 	perf_invalid_context = -1,
 	perf_hw_context = 0,
+	perf_sw_context,
 	perf_nr_task_contexts,
 };
 
Index: linux-2.6/kernel/perf_event.c
===================================================================
--- linux-2.6.orig/kernel/perf_event.c
+++ linux-2.6/kernel/perf_event.c
@@ -4691,6 +4691,8 @@ static int perf_swevent_init(struct perf
 }
 
 static struct pmu perf_swevent = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= perf_swevent_init,
 	.add		= perf_swevent_add,
 	.del		= perf_swevent_del,
@@ -4782,6 +4784,8 @@ static int perf_tp_event_init(struct per
 }
 
 static struct pmu perf_tracepoint = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= perf_tp_event_init,
 	.add		= perf_trace_add,
 	.del		= perf_trace_del,
@@ -4970,6 +4974,8 @@ static int cpu_clock_event_init(struct p
 }
 
 static struct pmu perf_cpu_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= cpu_clock_event_init,
 	.add		= cpu_clock_event_add,
 	.del		= cpu_clock_event_del,
@@ -5045,6 +5051,8 @@ static int task_clock_event_init(struct 
 }
 
 static struct pmu perf_task_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= task_clock_event_init,
 	.add		= task_clock_event_add,
 	.del		= task_clock_event_del,
Index: linux-2.6/kernel/hw_breakpoint.c
===================================================================
--- linux-2.6.orig/kernel/hw_breakpoint.c
+++ linux-2.6/kernel/hw_breakpoint.c
@@ -610,6 +610,8 @@ static void hw_breakpoint_stop(struct pe
 }
 
 static struct pmu perf_breakpoint = {
+	.task_ctx_nr	= perf_sw_context, /* could eventually get its own */
+
 	.event_init	= hw_breakpoint_event_init,
 	.add		= hw_breakpoint_add,
 	.del		= hw_breakpoint_del,



  parent reply	other threads:[~2010-09-07 16:56 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-07 16:46 [RFC][PATCH 00/19] perf pmu interface changes -v4 Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 01/19] perf: Fix CPU hotplug Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 02/19] perf: deconstify struct pmu Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 03/19] perf: register pmu implementations Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 04/19] perf: Unindent labels Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 05/19] perf: Reduce perf_disable() usage Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 06/19] perf: Per PMU disable Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 07/19] perf: Default PMU ops Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 08/19] perf: Shrink hw_perf_event Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 09/19] perf: Rework the PMU methods Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 10/19] perf: Remove the sysfs bits Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 11/19] perf: Separate find_get_context() from event initialization Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 12/19] perf: Remove the swevent hash-table from the cpu context Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 13/19] perf: Per cpu-context rotation timer Peter Zijlstra
2010-09-07 17:07   ` Peter Zijlstra
2010-09-07 17:33     ` Thomas Gleixner
2010-09-07 17:36       ` Peter Zijlstra
2010-09-08 13:50         ` Thomas Gleixner
2010-09-08 13:56           ` Peter Zijlstra
2010-09-07 17:08   ` Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 14/19] perf: Per-pmu-per-cpu contexts Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 15/19] perf: Move some code around Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 16/19] perf: Clean up perf_event_context allocation Peter Zijlstra
2010-09-07 16:46 ` [RFC][PATCH 17/19] perf: Multiple task contexts Peter Zijlstra
2010-09-07 16:46 ` Peter Zijlstra [this message]
2010-09-07 16:46 ` [RFC][PATCH 19/19] perf: Optimize context ops Peter Zijlstra
2010-09-10  4:39 ` [RFC][PATCH 00/19] perf pmu interface changes -v4 Paul Mackerras
2010-09-10  7:43   ` Peter Zijlstra
2010-09-14 16:56   ` Peter Zijlstra

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=20100907165012.227790649@chello.nl \
    --to=a.p.zijlstra@chello.nl \
    --cc=davem@davemloft.net \
    --cc=dengcheng.zhu@gmail.com \
    --cc=eranian@googlemail.com \
    --cc=fweisbec@gmail.com \
    --cc=gorcunov@gmail.com \
    --cc=lethal@linux-sh.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ming.m.lin@intel.com \
    --cc=mingo@elte.hu \
    --cc=paulus@samba.org \
    --cc=robert.richter@amd.com \
    --cc=will.deacon@arm.com \
    --cc=yanmin_zhang@linux.intel.com \
    /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