diff options
| author | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2007-02-16 10:28:22 +0100 |
|---|---|---|
| committer | Linus Torvalds <torvalds@woody.linux-foundation.org> | 2007-02-16 17:14:00 +0100 |
| commit | 1f2ea0837dbc263ce2a2512c4e73c83df68a6a55 (patch) | |
| tree | 8e033c747852671f3d4d2cb0bc677dab9ecfc441 /kernel | |
| parent | [PATCH] time: x86_64: re-enable vsyscall support for x86_64 (diff) | |
| download | linux-1f2ea0837dbc263ce2a2512c4e73c83df68a6a55.tar.xz linux-1f2ea0837dbc263ce2a2512c4e73c83df68a6a55.zip | |
[PATCH] posix timers: RCU optimization for clock_gettime()
Use RCU to avoid the need to acquire tasklist_lock in the single-threaded
case of clock_gettime(). It still acquires tasklist_lock when for a
(potentially multithreaded) process. This change allows realtime
applications to frequently monitor CPU consumption of individual tasks, as
requested (and now deployed) by some off-list users.
This has been in Ingo Molnar's -rt patchset since late 2005 with no
problems reported, and tests successfully on 2.6.20-rc6, so I believe that
it is long-since ready for mainline adoption.
[paulmck@linux.vnet.ibm.com: fix exit()/posix_cpu_clock_get() race spotted by Oleg]
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'kernel')
| -rw-r--r-- | kernel/posix-cpu-timers.c | 15 |
1 files changed, 10 insertions, 5 deletions
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 7c3e1e6dfb5b..657f77697415 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -304,7 +304,7 @@ int posix_cpu_clock_get(const clockid_t which_clock, struct timespec *tp) * should be able to see it. */ struct task_struct *p; - read_lock(&tasklist_lock); + rcu_read_lock(); p = find_task_by_pid(pid); if (p) { if (CPUCLOCK_PERTHREAD(which_clock)) { @@ -312,12 +312,17 @@ int posix_cpu_clock_get(const clockid_t which_clock, struct timespec *tp) error = cpu_clock_sample(which_clock, p, &rtn); } - } else if (p->tgid == pid && p->signal) { - error = cpu_clock_sample_group(which_clock, - p, &rtn); + } else { + read_lock(&tasklist_lock); + if (p->tgid == pid && p->signal) { + error = + cpu_clock_sample_group(which_clock, + p, &rtn); + } + read_unlock(&tasklist_lock); } } - read_unlock(&tasklist_lock); + rcu_read_unlock(); } if (error) |