From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
To: akpm@linux-foundation.org, Ingo Molnar <mingo@elte.hu>,
linux-kernel@vger.kernel.org
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
Andi Kleen <ak@suse.de>,
linux-mm@kvack.org, Dave Hansen <haveblue@us.ibm.com>,
Masami Hiramatsu <mhiramat@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
"Frank Ch. Eigler" <fche@redhat.com>,
Hideo AOKI <haoki@redhat.com>,
Takashi Nishiie <t-nishiie@np.css.fujitsu.com>,
Steven Rostedt <rostedt@goodmis.org>,
Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Subject: [patch 10/15] LTTng instrumentation - memory page faults
Date: Wed, 09 Jul 2008 10:59:39 -0400 [thread overview]
Message-ID: <20080709150046.737002691@polymtl.ca> (raw)
In-Reply-To: 20080709145929.352201601@polymtl.ca
[-- Attachment #1: lttng-instrumentation-memory.patch --]
[-- Type: text/plain, Size: 3639 bytes --]
Instrument the page fault entry and exit. Useful to detect delays caused by page
faults and bad memory usage patterns.
Those tracepoints are used by LTTng.
About the performance impact of tracepoints (which is comparable to markers),
even without immediate values optimizations, tests done by Hideo Aoki on ia64
show no regression. His test case was using hackbench on a kernel where
scheduler instrumentation (about 5 events in code scheduler code) was added.
See the "Tracepoints" patch header for performance result detail.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: Andi Kleen <ak@suse.de>
CC: linux-mm@kvack.org
CC: Dave Hansen <haveblue@us.ibm.com>
CC: Masami Hiramatsu <mhiramat@redhat.com>
CC: 'Peter Zijlstra' <peterz@infradead.org>
CC: "Frank Ch. Eigler" <fche@redhat.com>
CC: 'Ingo Molnar' <mingo@elte.hu>
CC: 'Hideo AOKI' <haoki@redhat.com>
CC: Takashi Nishiie <t-nishiie@np.css.fujitsu.com>
CC: 'Steven Rostedt' <rostedt@goodmis.org>
CC: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
---
mm/memory-trace.h | 14 ++++++++++++++
mm/memory.c | 33 ++++++++++++++++++++++++---------
2 files changed, 38 insertions(+), 9 deletions(-)
Index: linux-2.6-lttng/mm/memory.c
===================================================================
--- linux-2.6-lttng.orig/mm/memory.c 2008-07-09 10:58:31.000000000 -0400
+++ linux-2.6-lttng/mm/memory.c 2008-07-09 10:58:34.000000000 -0400
@@ -61,6 +61,7 @@
#include <linux/swapops.h>
#include <linux/elf.h>
+#include "memory-trace.h"
#ifndef CONFIG_NEED_MULTIPLE_NODES
/* use the per-pgdat data instead for discontigmem - mbligh */
@@ -2664,30 +2665,44 @@ unlock:
int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
unsigned long address, int write_access)
{
+ int res;
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
+ trace_memory_handle_fault_entry(mm, vma, address, write_access);
+
__set_current_state(TASK_RUNNING);
count_vm_event(PGFAULT);
- if (unlikely(is_vm_hugetlb_page(vma)))
- return hugetlb_fault(mm, vma, address, write_access);
+ if (unlikely(is_vm_hugetlb_page(vma))) {
+ res = hugetlb_fault(mm, vma, address, write_access);
+ goto end;
+ }
pgd = pgd_offset(mm, address);
pud = pud_alloc(mm, pgd, address);
- if (!pud)
- return VM_FAULT_OOM;
+ if (!pud) {
+ res = VM_FAULT_OOM;
+ goto end;
+ }
pmd = pmd_alloc(mm, pud, address);
- if (!pmd)
- return VM_FAULT_OOM;
+ if (!pmd) {
+ res = VM_FAULT_OOM;
+ goto end;
+ }
pte = pte_alloc_map(mm, pmd, address);
- if (!pte)
- return VM_FAULT_OOM;
+ if (!pte) {
+ res = VM_FAULT_OOM;
+ goto end;
+ }
- return handle_pte_fault(mm, vma, address, pte, pmd, write_access);
+ res = handle_pte_fault(mm, vma, address, pte, pmd, write_access);
+end:
+ trace_memory_handle_fault_exit(res);
+ return res;
}
#ifndef __PAGETABLE_PUD_FOLDED
Index: linux-2.6-lttng/mm/memory-trace.h
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ linux-2.6-lttng/mm/memory-trace.h 2008-07-09 10:58:34.000000000 -0400
@@ -0,0 +1,14 @@
+#ifndef _MEMORY_TRACE_H
+#define _MEMORY_TRACE_H
+
+#include <linux/tracepoint.h>
+
+DEFINE_TRACE(memory_handle_fault_entry,
+ TPPROTO(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long address, int write_access),
+ TPARGS(mm, vma, address, write_access));
+DEFINE_TRACE(memory_handle_fault_exit,
+ TPPROTO(int res),
+ TPARGS(res));
+
+#endif
--
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
next prev parent reply other threads:[~2008-07-09 15:06 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-07-09 14:59 [patch 00/15] Tracepoints v3 for linux-next Mathieu Desnoyers
2008-07-09 14:59 ` [patch 01/15] Kernel Tracepoints Mathieu Desnoyers
2008-07-15 7:50 ` Peter Zijlstra
2008-07-15 13:25 ` Mathieu Desnoyers
2008-07-15 13:59 ` Peter Zijlstra
2008-07-15 14:27 ` Mathieu Desnoyers
2008-07-15 14:42 ` Peter Zijlstra
2008-07-15 15:22 ` Mathieu Desnoyers
2008-07-15 15:31 ` Peter Zijlstra
2008-07-15 15:50 ` Mathieu Desnoyers
2008-08-01 21:10 ` Paul E. McKenney
2008-07-15 16:08 ` Mathieu Desnoyers
2008-07-15 16:25 ` Peter Zijlstra
2008-07-15 16:51 ` Mathieu Desnoyers
2008-08-01 21:10 ` Paul E. McKenney
2008-08-02 0:03 ` Peter Zijlstra
2008-08-02 0:17 ` Paul E. McKenney
2008-08-01 21:10 ` Paul E. McKenney
2008-07-15 16:26 ` Mathieu Desnoyers
2008-08-01 21:10 ` Paul E. McKenney
2008-07-15 17:50 ` Mathieu Desnoyers
2008-07-15 14:03 ` Peter Zijlstra
2008-07-15 14:46 ` Mathieu Desnoyers
2008-07-15 15:13 ` Peter Zijlstra
2008-07-15 18:22 ` Mathieu Desnoyers
2008-07-15 18:33 ` Steven Rostedt
2008-07-15 18:52 ` Masami Hiramatsu
2008-07-15 19:08 ` Mathieu Desnoyers
2008-07-15 19:02 ` Mathieu Desnoyers
2008-07-15 19:52 ` Peter Zijlstra
2008-07-09 14:59 ` [patch 02/15] Tracepoints Documentation Mathieu Desnoyers
2008-07-09 14:59 ` [patch 03/15] Tracepoints Samples Mathieu Desnoyers
2008-07-09 14:59 ` [patch 04/15] LTTng instrumentation - irq Mathieu Desnoyers
2008-07-09 16:39 ` Masami Hiramatsu
2008-07-09 17:05 ` [patch 04/15] LTTng instrumentation - irq (update) Mathieu Desnoyers
2008-07-09 14:59 ` [patch 05/15] LTTng instrumentation - scheduler Mathieu Desnoyers
2008-07-09 15:34 ` [patch 05/15] LTTng instrumentation - scheduler (repost) Mathieu Desnoyers
2008-07-09 15:39 ` Ingo Molnar
2008-07-09 16:00 ` Mathieu Desnoyers
2008-07-09 16:21 ` [patch 05/15] LTTng instrumentation - scheduler (merge ftrace markers) Mathieu Desnoyers
2008-07-09 19:09 ` [PATCH] ftrace port to tracepoints (linux-next) Mathieu Desnoyers
2008-07-10 3:14 ` Takashi Nishiie
2008-07-10 3:57 ` [PATCH] ftrace port to tracepoints (linux-next) (nitpick update) Mathieu Desnoyers
[not found] ` <20080711143709.GB11500@Krystal>
[not found] ` <Pine.LNX.4.58.0807141112540.30484@gandalf.stny.rr.com>
[not found] ` <20080714153334.GA651@Krystal>
[not found] ` <Pine.LNX.4.58.0807141153250.29493@gandalf.stny.rr.com>
2008-07-14 16:25 ` [PATCH] ftrace memory barriers Mathieu Desnoyers
2008-07-14 16:35 ` Steven Rostedt
2008-07-09 14:59 ` [patch 06/15] LTTng instrumentation - timer Mathieu Desnoyers
2008-07-09 14:59 ` [patch 07/15] LTTng instrumentation - kernel Mathieu Desnoyers
2008-07-09 14:59 ` [patch 08/15] LTTng instrumentation - filemap Mathieu Desnoyers
2008-07-09 14:59 ` [patch 09/15] LTTng instrumentation - swap Mathieu Desnoyers
2008-07-09 14:59 ` Mathieu Desnoyers [this message]
2008-07-09 14:59 ` [patch 11/15] LTTng instrumentation - page Mathieu Desnoyers
2008-07-09 14:59 ` [patch 12/15] LTTng instrumentation - hugetlb Mathieu Desnoyers
2008-07-11 14:30 ` [patch 12/15] LTTng instrumentation - hugetlb (update) Mathieu Desnoyers
2008-07-09 14:59 ` [patch 13/15] LTTng instrumentation - net Mathieu Desnoyers
2008-07-09 14:59 ` [patch 14/15] LTTng instrumentation - ipv4 Mathieu Desnoyers
2008-07-09 14:59 ` Mathieu Desnoyers
2008-07-09 17:01 ` [patch 00/15] Tracepoints v3 for linux-next Masami Hiramatsu
2008-07-09 17:11 ` [patch 15/15] LTTng instrumentation - ipv6 Mathieu Desnoyers
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=20080709150046.737002691@polymtl.ca \
--to=mathieu.desnoyers@polymtl.ca \
--cc=ak@suse.de \
--cc=akpm@linux-foundation.org \
--cc=eduard.munteanu@linux360.ro \
--cc=fche@redhat.com \
--cc=haoki@redhat.com \
--cc=haveblue@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhiramat@redhat.com \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=t-nishiie@np.css.fujitsu.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