From: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
To: mingo@elte.hu
Cc: linux-kernel@vger.kernel.org,
Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Paul Mackerras <paulus@samba.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Thomas Gleixner <tglx@linutronix.de>,
Greg Kroah-Hartman <gregkh@suse.de>
Subject: [PATCH 0/5] lockdep: Add information of file and line to lockdep_map
Date: Thu, 7 Jan 2010 19:39:50 +0900 [thread overview]
Message-ID: <1262860795-5745-1-git-send-email-mitake@dcl.info.waseda.ac.jp> (raw)
In-Reply-To: <4B45B9C1.2040900@dcl.info.waseda.ac.jp>
There are a lot of lock instances with same names (e.g. port_lock).
This patch series add __FILE__ and __LINE__ to lockdep_map,
and these will be used for trace lock events.
Example use from perf lock map:
| 0xffffea0004c992b8: __pte_lockptr(page) (src: include/linux/mm.h, line: 952)
| 0xffffea0004b112b8: __pte_lockptr(page) (src: include/linux/mm.h, line: 952)
| 0xffffea0004a3f2b8: __pte_lockptr(page) (src: include/linux/mm.h, line: 952)
| 0xffffea0004cd5228: __pte_lockptr(page) (src: include/linux/mm.h, line: 952)
| 0xffff8800b91e2b28: &sb->s_type->i_lock_key (src: fs/inode.c, line: 166)
| 0xffff8800bb9d7ae0: key (src: kernel/wait.c, line: 16)
| 0xffff8800aa07dae0: &dentry->d_lock (src: fs/dcache.c, line: 944)
| 0xffff8800b07fbae0: &dentry->d_lock (src: fs/dcache.c, line: 944)
| 0xffff8800b07f3ae0: &dentry->d_lock (src: fs/dcache.c, line: 944)
| 0xffff8800bf15fae0: &sighand->siglock (src: kernel/fork.c, line: 1490)
| 0xffff8800b90f7ae0: &dentry->d_lock (src: fs/dcache.c, line: 944)
| ...
(This output of perf lock map is produced by my local version,
I'll send this later.)
And sadly, as Peter Zijlstra predicted, this produces certain overhead.
Before appling this series:
| % sudo ./perf lock rec perf bench sched messaging
| # Running sched/messaging benchmark...
| # 20 sender and receiver processes per group
| # 10 groups == 400 processes run
|
| Total time: 3.834 [sec]
After:
sudo ./perf lock rec perf bench sched messaging
| # Running sched/messaging benchmark...
| # 20 sender and receiver processes per group
| # 10 groups == 400 processes run
|
| Total time: 5.415 [sec]
| [ perf record: Woken up 0 times to write data ]
| [ perf record: Captured and wrote 53.512 MB perf.data (~2337993 samples) ]
But raw exec of perf bench sched messaging is this:
| % perf bench sched messaging
| # Running sched/messaging benchmark...
| # 20 sender and receiver processes per group
| # 10 groups == 400 processes run
|
| Total time: 0.498 [sec]
Tracing lock events already produces amount of overhead.
I think the overhead produced by this series is not a fatal problem,
radically optimization is required...
Could you merge this into perf/lock branch, Ingo?
Signed-off-by: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Hitoshi Mitake (5):
lockdep: Add file and line to initialize sequence of spin and rw lock
lockdep: Add file and line to initialize sequence of rwsem
lockdep: Add file and line to initialize sequence of rwsem
lockdep: Add file and line to initialize sequence of mutex
lockdep: Fix the way to initialize class_mutex for information of
file and line
arch/x86/include/asm/rwsem.h | 9 +++++++--
drivers/base/class.c | 3 ++-
include/linux/mutex-debug.h | 2 +-
include/linux/mutex.h | 12 +++++++++---
include/linux/rwsem-spinlock.h | 11 ++++++++---
include/linux/spinlock.h | 12 ++++++++----
include/linux/spinlock_types.h | 12 ++++++++++--
kernel/mutex-debug.c | 5 +++--
kernel/mutex-debug.h | 3 ++-
kernel/mutex.c | 5 +++--
kernel/mutex.h | 2 +-
lib/rwsem-spinlock.c | 5 +++--
lib/rwsem.c | 5 +++--
lib/spinlock_debug.c | 10 ++++++----
14 files changed, 66 insertions(+), 30 deletions(-)
next prev parent reply other threads:[~2010-01-07 10:40 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-16 3:28 [PATCH 1/2] perf: Add util/include/linuxhash.h to include hash.h of kernel Hitoshi Mitake
2009-12-16 3:28 ` [PATCH 2/2] perf lock: Fix output of tracing lock events Hitoshi Mitake
2009-12-16 8:19 ` [tip:perf/lock] " tip-bot for Hitoshi Mitake
2009-12-17 8:50 ` [PATCH 2/2] " Ingo Molnar
2009-12-17 9:24 ` Peter Zijlstra
2009-12-17 10:09 ` Ingo Molnar
2009-12-17 10:26 ` Peter Zijlstra
2009-12-17 10:51 ` Ingo Molnar
2009-12-26 13:43 ` Hitoshi Mitake
2009-12-28 10:01 ` Peter Zijlstra
2009-12-31 13:24 ` Hitoshi Mitake
2010-01-07 10:38 ` Hitoshi Mitake
2010-01-07 10:39 ` Hitoshi Mitake [this message]
2010-01-13 9:52 ` [PATCH 0/5] lockdep: Add information of file and line to lockdep_map Peter Zijlstra
2010-01-13 10:09 ` Ingo Molnar
2010-01-16 13:01 ` Hitoshi Mitake
2010-01-18 7:20 ` Peter Zijlstra
2010-01-26 5:56 ` [PATCH] Add type of locks to lock trace events Hitoshi Mitake
2010-01-28 10:21 ` Peter Zijlstra
2010-01-29 17:29 ` Frederic Weisbecker
2010-01-29 17:41 ` Peter Zijlstra
2010-01-29 22:12 ` Frederic Weisbecker
2010-02-24 9:02 ` [PATCH] Separate lock events with types Hitoshi Mitake
2010-03-26 23:33 ` Frederic Weisbecker
2010-04-05 10:37 ` Hitoshi Mitake
2010-04-06 8:26 ` Peter Zijlstra
2010-04-06 9:44 ` Frederic Weisbecker
2010-01-07 10:39 ` [PATCH 1/5] lockdep: Add file and line to initialize sequence of spin and rw lock Hitoshi Mitake
2010-01-07 10:39 ` [PATCH 2/5] lockdep: Add file and line to initialize sequence of rwsem Hitoshi Mitake
2010-01-07 10:39 ` [PATCH 3/5] " Hitoshi Mitake
2010-01-07 10:39 ` [PATCH 4/5] lockdep: Add file and line to initialize sequence of mutex Hitoshi Mitake
2010-01-07 10:39 ` [PATCH 5/5] lockdep: Fix the way to initialize class_mutex for information of file and line Hitoshi Mitake
2010-01-13 10:00 ` Ingo Molnar
2010-01-13 23:17 ` Greg KH
2010-01-13 23:19 ` Greg KH
2010-01-16 12:55 ` Hitoshi Mitake
2009-12-16 8:18 ` [tip:perf/lock] perf: Add util/include/linuxhash.h to include hash.h of kernel tip-bot for Hitoshi Mitake
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=1262860795-5745-1-git-send-email-mitake@dcl.info.waseda.ac.jp \
--to=mitake@dcl.info.waseda.ac.jp \
--cc=a.p.zijlstra@chello.nl \
--cc=fweisbec@gmail.com \
--cc=gregkh@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
--cc=tglx@linutronix.de \
/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).