[Mesa-dev] [PATCH 2/2] gallium/os: fix os_time_get_nano() to roll over less

[Jose Fonseca]

On 16/08/17 23:37, Jose Fonseca wrote:
> On 16/08/17 14:22, Brian Paul wrote:
>> From: Frank Richter <frank.richter at dynardo.de>
>>
>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102241
>> Cc: mesa-stable at lists.freedesktop.org
>> Reviewed-by: Brian Paul <brianp at vmware.com>
>> ---
>>   src/gallium/auxiliary/os/os_time.c | 9 ++++++++-
>>   1 file changed, 8 insertions(+), 1 deletion(-)
>>
>> diff --git a/src/gallium/auxiliary/os/os_time.c 
>> b/src/gallium/auxiliary/os/os_time.c
>> index e169139..e4a1cae 100644
>> --- a/src/gallium/auxiliary/os/os_time.c
>> +++ b/src/gallium/auxiliary/os/os_time.c
>> @@ -69,10 +69,17 @@ os_time_get_nano(void)
>>      static LARGE_INTEGER frequency;
>>      LARGE_INTEGER counter;
>> +   int64_t secs, nanosecs;
>>      if(!frequency.QuadPart)
>>         QueryPerformanceFrequency(&frequency);
>>      QueryPerformanceCounter(&counter);
>> -   return counter.QuadPart*INT64_C(1000000000)/frequency.QuadPart;
>> +   /* Compute seconds and nanoseconds parts separately to
>> +    * reduce severity of precision loss.
>> +    */
>> +   secs = counter.QuadPart / frequency.QuadPart;
>> +   nanosecs = (counter.QuadPart % frequency.QuadPart) * 
>> INT64_C(1000000000)
>> +      / frequency.QuadPart;
>> +   return secs*INT64_C(1000000000) + nanosecs;
>>   #else
>>
> 
> Series looks great.
> 
> Thanks!
> 
> Reviewed-by: Jose Fonseca <jfonseca at vmware.com>

BTW, I wonder if we would win by using lldiv().  Because this is often 
use for performance measurements, so these extra division might add some 
impact.

Jose