Re: [Mesa-dev] [PATCH 4/6] i965: dump scheduling cycle estimates

2015-10-05 Thread Iago Toral
On Fri, 2015-10-02 at 17:43 -0400, Connor Abbott wrote: > On Fri, Oct 2, 2015 at 5:37 PM, Connor Abbott wrote: > > The heuristic we're using is rather lame, since it assumes everything is > > non-uniform and loops execute 50 times, but it should be enough for > > measuring

Re: [Mesa-dev] [PATCH 4/6] i965: dump scheduling cycle estimates

2015-10-02 Thread Connor Abbott
On Fri, Oct 2, 2015 at 5:37 PM, Connor Abbott wrote: > The heuristic we're using is rather lame, since it assumes everything is > non-uniform and loops execute 50 times, but it should be enough for > measuring improvements in the scheduler that don't result in a change in >

[Mesa-dev] [PATCH 4/6] i965: dump scheduling cycle estimates

2015-10-02 Thread Connor Abbott
The heuristic we're using is rather lame, since it assumes everything is non-uniform and loops execute 50 times, but it should be enough for measuring improvements in the scheduler that don't result in a change in the number of instructions. Signed-off-by: Connor Abbott ---