public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Matt Helsley <matthltc@us.ibm.com>
To: Linux-Kernel <linux-kernel@vger.kernel.org>
Cc: Jes Sorensen <jes@sgi.com>,
	LSE-Tech <lse-tech@lists.sourceforge.net>,
	Chandra S Seetharaman <sekharan@us.ibm.com>,
	Christoph Hellwig <hch@lst.de>, Al Viro <viro@zeniv.linux.org.uk>,
	Steve Grubb <sgrubb@redhat.com>,
	linux-audit@redhat.com, Paul Jackson <pj@sgi.com>,
	Andrew Morton <akpm@osdl.org>
Subject: [PATCH 6/9] Task Watchers v2: Register IRQ flag tracing task watcher
Date: Thu, 02 Nov 2006 20:23:03 -0800	[thread overview]
Message-ID: <20061103042749.944489000@us.ibm.com> (raw)
In-Reply-To: 20061103042257.274316000@us.ibm.com

[-- Attachment #1: task-watchers-register-trace-irqflags --]
[-- Type: text/plain, Size: 4284 bytes --]

Register an irq-flag-tracing task watcher instead of hooking into
copy_process().

Signed-off-by: Matt Helsley <matthltc@us.ibm.com>
---
 kernel/fork.c       |   19 -------------------
 kernel/irq/handle.c |   24 ++++++++++++++++++++++++
 2 files changed, 24 insertions(+), 19 deletions(-)

Benchmark results:
System: 4 1.7GHz ppc64 (Power 4+) processors, 30968600MB RAM, 2.6.19-rc2-mm2 kernel

Clone	Number of Children Cloned
	5000		7500		10000		12500		15000		17500
	---------------------------------------------------------------------------------------
Mean	17826.5 	18077.4 	18160.1 	18263.6 	18343	 	18350.8
Dev	305.841 	306.331 	283.323 	284.761 	292.732 	292.882
Err (%)	1.71565 	1.69455 	1.56014 	1.55917 	1.59588 	1.59602

Fork	Number of Children Forked
	5000		7500		10000		12500		15000		17500
	---------------------------------------------------------------------------------------
Mean	17813.5 	18062.4 	18140.5 	18246.7 	18237.8 	18275.2
Dev	305.816 	294.914 	294.779 	294.727 	323.996 	300.176
Err (%)	1.71677 	1.63275 	1.62498 	1.61523 	1.77651 	1.64253

Kernbench:
Elapsed: 124.4s User: 439.787s System: 46.485s CPU: 390.3%
439.70user 46.43system 2:04.64elapsed 390%CPU (0avgtext+0avgdata 0maxresident)k
439.92user 46.38system 2:04.47elapsed 390%CPU (0avgtext+0avgdata 0maxresident)k
439.79user 46.62system 2:04.44elapsed 390%CPU (0avgtext+0avgdata 0maxresident)k
439.83user 46.46system 2:04.29elapsed 391%CPU (0avgtext+0avgdata 0maxresident)k
439.73user 46.47system 2:04.12elapsed 391%CPU (0avgtext+0avgdata 0maxresident)k
439.83user 46.49system 2:04.10elapsed 391%CPU (0avgtext+0avgdata 0maxresident)k
439.76user 46.42system 2:04.41elapsed 390%CPU (0avgtext+0avgdata 0maxresident)k
439.70user 46.64system 2:04.30elapsed 391%CPU (0avgtext+0avgdata 0maxresident)k
439.79user 46.47system 2:04.76elapsed 389%CPU (0avgtext+0avgdata 0maxresident)k
439.82user 46.47system 2:04.47elapsed 390%CPU (0avgtext+0avgdata 0maxresident)k

Index: linux-2.6.19-rc2-mm2/kernel/fork.c
===================================================================
--- linux-2.6.19-rc2-mm2.orig/kernel/fork.c
+++ linux-2.6.19-rc2-mm2/kernel/fork.c
@@ -1052,29 +1052,10 @@ static struct task_struct *copy_process(
 		p->tgid = current->tgid;
 
 	retval = notify_task_watchers(WATCH_TASK_INIT, clone_flags, p);
 	if (retval < 0)
 		goto bad_fork_cleanup_delays_binfmt;
-#ifdef CONFIG_TRACE_IRQFLAGS
-	p->irq_events = 0;
-#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
-	p->hardirqs_enabled = 1;
-#else
-	p->hardirqs_enabled = 0;
-#endif
-	p->hardirq_enable_ip = 0;
-	p->hardirq_enable_event = 0;
-	p->hardirq_disable_ip = _THIS_IP_;
-	p->hardirq_disable_event = 0;
-	p->softirqs_enabled = 1;
-	p->softirq_enable_ip = _THIS_IP_;
-	p->softirq_enable_event = 0;
-	p->softirq_disable_ip = 0;
-	p->softirq_disable_event = 0;
-	p->hardirq_context = 0;
-	p->softirq_context = 0;
-#endif
 #ifdef CONFIG_LOCKDEP
 	p->lockdep_depth = 0; /* no locks held yet */
 	p->curr_chain_key = 0;
 	p->lockdep_recursion = 0;
 #endif
Index: linux-2.6.19-rc2-mm2/kernel/irq/handle.c
===================================================================
--- linux-2.6.19-rc2-mm2.orig/kernel/irq/handle.c
+++ linux-2.6.19-rc2-mm2/kernel/irq/handle.c
@@ -13,10 +13,11 @@
 #include <linux/irq.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
+#include <linux/task_watchers.h>
 
 #include "internals.h"
 
 /**
  * handle_bad_irq - handle spurious and unhandled irqs
@@ -266,6 +267,29 @@ void early_init_irq_lock_class(void)
 
 	for (i = 0; i < NR_IRQS; i++)
 		lockdep_set_class(&irq_desc[i].lock, &irq_desc_lock_class);
 }
 
+static int init_task_trace_irqflags(unsigned long clone_flags,
+				    struct task_struct *p)
+{
+	p->irq_events = 0;
+#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW
+	p->hardirqs_enabled = 1;
+#else
+	p->hardirqs_enabled = 0;
+#endif
+	p->hardirq_enable_ip = 0;
+	p->hardirq_enable_event = 0;
+	p->hardirq_disable_ip = _THIS_IP_;
+	p->hardirq_disable_event = 0;
+	p->softirqs_enabled = 1;
+	p->softirq_enable_ip = _THIS_IP_;
+	p->softirq_enable_event = 0;
+	p->softirq_disable_ip = 0;
+	p->softirq_disable_event = 0;
+	p->hardirq_context = 0;
+	p->softirq_context = 0;
+	return 0;
+}
+task_watcher_func(init, init_task_trace_irqflags);
 #endif

--

  parent reply	other threads:[~2006-11-03  4:28 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-11-03  4:22 [PATCH 0/9] Task Watchers v2: Introduction Matt Helsley
2006-11-03  4:22 ` [PATCH 1/9] Task Watchers v2: Task watchers v2 Matt Helsley
2006-11-03 13:22   ` Daniel Walker
2006-11-04  0:43     ` Matt Helsley
2006-11-04  1:13       ` Daniel Walker
2006-11-05  0:12         ` Matt Helsley
2006-11-03  4:22 ` [PATCH 2/9] Task Watchers v2: Register audit task watcher Matt Helsley
2006-11-03  4:23 ` [PATCH 3/9] Task Watchers v2: Register semundo " Matt Helsley
2006-11-03  4:23 ` [PATCH 4/9] Task Watchers v2: Register cpuset " Matt Helsley
2006-11-03  4:23 ` [PATCH 5/9] Task Watchers v2: Register NUMA mempolicy " Matt Helsley
2006-11-03  4:23 ` Matt Helsley [this message]
2006-11-03  4:23 ` [PATCH 7/9] Task Watchers v2: Register lockdep " Matt Helsley
2006-11-03  4:23 ` [PATCH 8/9] Task Watchers v2: Register process keyrings " Matt Helsley
2006-11-03  4:23 ` [PATCH 9/9] Task Watchers v2: Register process events connector Matt Helsley
2006-11-03  8:57 ` [PATCH 0/9] Task Watchers v2: Introduction Paul Jackson
2006-11-03 22:55   ` Matt Helsley

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=20061103042749.944489000@us.ibm.com \
    --to=matthltc@us.ibm.com \
    --cc=akpm@osdl.org \
    --cc=hch@lst.de \
    --cc=jes@sgi.com \
    --cc=linux-audit@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lse-tech@lists.sourceforge.net \
    --cc=pj@sgi.com \
    --cc=sekharan@us.ibm.com \
    --cc=sgrubb@redhat.com \
    --cc=viro@zeniv.linux.org.uk \
    /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