diff options
| author |   Martin Schwidefsky <schwidefsky@de.ibm.com> | 2017-03-01 09:47:57 +0100 |
|---|---|---|
| committer | Martin Schwidefsky <schwidefsky@de.ibm.com> | 2017-03-01 09:59:28 +0100 |
| commit | d03bd0454b101adb94d0b5a9cc11396182943cb4 (patch) | |
| tree | 1ff875a7337cbbd3ba85f8990e3d4103139c41cf /arch/s390 | |
| parent | s390/cputime: provide archicture specific cputime_to_nsecs (diff) | |
| download | linux-dev-d03bd0454b101adb94d0b5a9cc11396182943cb4.tar.xz linux-dev-d03bd0454b101adb94d0b5a9cc11396182943cb4.zip | |
s390/timex: micro optimization for tod_to_ns
The conversion of a TOD value to nano-seconds currently uses a 32/32 bit
split with the calculation for "nsecs = (TOD * 125) >> 9". Using a
55/9 bit split saves an instruction.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Diffstat (limited to 'arch/s390')
| -rw-r--r-- | arch/s390/include/asm/timex.h | 12 |
1 files changed, 4 insertions, 8 deletions
diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h index 354344dcc198..118535123f34 100644 --- a/arch/s390/include/asm/timex.h +++ b/arch/s390/include/asm/timex.h @@ -206,20 +206,16 @@ static inline unsigned long long get_tod_clock_monotonic(void) * ns = (todval * 125) >> 9; * * In order to avoid an overflow with the multiplication we can rewrite this. - * With a split todval == 2^32 * th + tl (th upper 32 bits, tl lower 32 bits) + * With a split todval == 2^9 * th + tl (th upper 55 bits, tl lower 9 bits) * we end up with * - * ns = ((2^32 * th + tl) * 125 ) >> 9; - * -> ns = (2^23 * th * 125) + ((tl * 125) >> 9); + * ns = ((2^9 * th + tl) * 125 ) >> 9; + * -> ns = (th * 125) + ((tl * 125) >> 9); * */ static inline unsigned long long tod_to_ns(unsigned long long todval) { - unsigned long long ns; - - ns = ((todval >> 32) << 23) * 125; - ns += ((todval & 0xffffffff) * 125) >> 9; - return ns; + return ((todval >> 9) * 125) + (((todval & 0x1ff) * 125) >> 9); } #endif |