diff options
| author | Nicolas Pitre <nicolas.pitre@linaro.org> | 2014-12-03 22:43:06 +0300 |
|---|---|---|
| committer | John Stultz <john.stultz@linaro.org> | 2015-01-24 04:21:31 +0300 |
| commit | 8b618628b2bf83512fc8df5e8672619d65adfdfb (patch) | |
| tree | c295bed4f4e702f5f39fcce622258e55242b1d93 /kernel/time | |
| parent | eaa27f34e91a14cdceed26ed6c6793ec1d186115 (diff) | |
| download | linux-8b618628b2bf83512fc8df5e8672619d65adfdfb.tar.xz | |
ktime: Optimize ktime_divns for constant divisors
At least on ARM, do_div() is optimized to turn constant divisors into
an inline multiplication by the reciprocal value at compile time.
However this optimization is missed entirely whenever ktime_divns() is
used and the slow out-of-line division code is used all the time.
Let ktime_divns() use do_div() inline whenever the divisor is constant
and small enough. This will make things like ktime_to_us() and
ktime_to_ms() much faster.
Cc: Arnd Bergmann <arnd.bergmann@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Nicolas Pitre <nico@linaro.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Diffstat (limited to 'kernel/time')
| -rw-r--r-- | kernel/time/hrtimer.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 37e50aadd471..890535c41c2d 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -266,7 +266,7 @@ lock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags) /* * Divide a ktime value by a nanosecond value */ -u64 ktime_divns(const ktime_t kt, s64 div) +u64 __ktime_divns(const ktime_t kt, s64 div) { u64 dclc; int sft = 0; @@ -282,7 +282,7 @@ u64 ktime_divns(const ktime_t kt, s64 div) return dclc; } -EXPORT_SYMBOL_GPL(ktime_divns); +EXPORT_SYMBOL_GPL(__ktime_divns); #endif /* BITS_PER_LONG >= 64 */ /* |