The patch titled
sys_time() speedup
has been added to the -mm tree. Its filename is
sys_time-speedup.patch
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: sys_time() speedup
From: Ingo Molnar <mingo@xxxxxxx>
Improve performance of sys_time(). sys_time() returns time in seconds, but
it does so by calling do_gettimeofday() and then returning the tv_sec
portion of the GTOD time. But the data structure "xtime", which is updated
by every timer/scheduler tick, already offers HZ granularity time.
The patch improves the sysbench OLTP macrobenchmark significantly:
2.6.22-rc6:
#threads
1: transactions: 3733 (373.21 per sec.)
2: transactions: 6676 (667.46 per sec.)
3: transactions: 6957 (695.50 per sec.)
4: transactions: 7055 (705.48 per sec.)
5: transactions: 6596 (659.33 per sec.)
2.6.22-rc6 + sys_time.patch:
1: transactions: 4005 (400.47 per sec.)
2: transactions: 7379 (737.77 per sec.)
3: transactions: 7347 (734.49 per sec.)
4: transactions: 7468 (746.65 per sec.)
5: transactions: 7428 (742.47 per sec.)
Mixed API uses of gettimeofday() and time() are guaranteed to be coherent
via the use of a at-most-once-per-second slowpath that updates xtime.
Signed-off-by: Ingo Molnar <mingo@xxxxxxx>
Cc: John Stultz <johnstul@xxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Roman Zippel <zippel@xxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
kernel/time.c | 27 ++++++++++++++++++++++-----
1 files changed, 22 insertions(+), 5 deletions(-)
diff -puN kernel/time.c~sys_time-speedup kernel/time.c
--- a/kernel/time.c~sys_time-speedup
+++ a/kernel/time.c
@@ -57,14 +57,17 @@ EXPORT_SYMBOL(sys_tz);
*/
asmlinkage long sys_time(time_t __user * tloc)
{
- time_t i;
- struct timeval tv;
+ /*
+ * We read xtime.tv_sec atomically - it's updated
+ * atomically by update_wall_time(), so no need to
+ * even read-lock the xtime seqlock:
+ */
+ time_t i = xtime.tv_sec;
- do_gettimeofday(&tv);
- i = tv.tv_sec;
+ smp_rmb(); /* sys_time() results are coherent */
if (tloc) {
- if (put_user(i,tloc))
+ if (put_user(i, tloc))
i = -EFAULT;
}
return i;
@@ -373,6 +376,20 @@ void do_gettimeofday (struct timeval *tv
tv->tv_sec = sec;
tv->tv_usec = usec;
+
+ /*
+ * Make sure xtime.tv_sec [returned by sys_time()] always
+ * follows the gettimeofday() result precisely. This
+ * condition is extremely unlikely, it can hit at most
+ * once per second:
+ */
+ if (unlikely(xtime.tv_sec != tv->tv_sec)) {
+ unsigned long flags;
+
+ write_seqlock_irqsave(&xtime_lock);
+ update_wall_time();
+ write_seqlock_irqrestore(&xtime_lock);
+ }
}
EXPORT_SYMBOL(do_gettimeofday);
_
Patches currently in -mm which might be from mingo@xxxxxxx are
git-acpi-add-exports.patch
git-kvm.patch
git-selinux.patch
x86_64-irq-check-remote-irr-bit-before-migrating-level-triggered-irq-v3.patch
only-allow-nonlinear-vmas-for-ram-backed-filesystems.patch
cpuset-remove-sched-domain-hooks-from-cpusets.patch
introduce-write_trylock_irqsave.patch
use-write_trylock_irqsave-in-ptrace_attach.patch
fix-stop_machine_run-problem-with-naughty-real-time-process.patch
cpu-hotplug-fix-ksoftirqd-termination-on-cpu-hotplug-with-naughty-realtime-process.patch
cpu-hotplug-fix-ksoftirqd-termination-on-cpu-hotplug-with-naughty-realtime-process-fix.patch
pie-randomization.patch
vdso-print-fatal-signals.patch
remove-clockevents_releaserequest_device.patch
add-a-flag-to-indicate-deferrable-timers-in-proc-timer_stats.patch
introduce-o_cloexec-take-2.patch
introduce-o_cloexec-parisc-fix.patch
o_cloexec-for-scm_rights.patch
o_cloexec-for-scm_rights-fix.patch
o_cloexec-for-scm_rights-fix-2.patch
improve-behaviour-of-spurious-irq-detect.patch
improve-behaviour-of-spurious-irq-detect-fix.patch
allow-softlockup-to-be-runtime-disabled.patch
sys_time-speedup.patch
cfs-scheduler.patch
cfs-scheduler-vs-detach-schedh-from-mmh.patch
cfs-scheduler-v14-rc2-mm1.patch
cfs-scheduler-warning-fixes.patch
cfs-scheduler-v15-rc3-mm1.patch
fs-proc-basec-make-a-struct-static.patch
cfs-warning-fixes.patch
schedstats-fix-printk-format.patch
cfs-scheduler-v16.patch
sched-add-above-background-load-function.patch
mm-implement-swap-prefetching.patch
fix-raw_spinlock_t-vs-lockdep.patch
lockdep-sanitise-config_prove_locking.patch
lockdep-reduce-the-ifdeffery.patch
lockstat-core-infrastructure.patch
lockstat-core-infrastructure-fix.patch
lockstat-core-infrastructure-fix-fix.patch
lockstat-core-infrastructure-fix-fix-fix.patch
lockstat-human-readability-tweaks.patch
lockstat-hook-into-spinlock_t-rwlock_t-rwsem-and-mutex.patch
lockdep-various-fixes.patch
lockdep-various-fixes-checkpatch.patch
lockdep-fixup-sk_callback_lock-annotation.patch
lockstat-measure-lock-bouncing.patch
lockstat-measure-lock-bouncing-checkpatch.patch
lockstat-better-class-name-representation.patch
detect-atomic-counter-underflows.patch
make-frame_pointer-default=y.patch
mutex-subsystem-synchro-test-module.patch
lockdep-show-held-locks-when-showing-a-stackdump.patch
kmap_atomic-debugging.patch
random-warning-squishes.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
