[PATCH] D126389: [AMDGPU] Improve codegen of extractelement/insertelement in some cases

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added a comment.

In D126389#3540098 <https://reviews.llvm.org/D126389#3540098>, @jpages wrote:

> In D126389#3539881 <https://reviews.llvm.org/D126389#3539881>, @foad wrote:
>
>> In D126389#3538346 <https://reviews.llvm.org/D126389#3538346>, @rampitec wrote:
>>
>>> Any performance numbers? The 8 element case was driven by a specific customer program and the performance of the cmp/select was better than movrel.
>>
>> I don't know why that would be. Maybe the performance characteristics are different on GFX10+ compared to GFX9.
>>
>> Also on GFX10+ sgpr usage does not affect occupancy, so perhaps the heuristic could be tweaked to make it more likely to use s_movrel (not v_movrel) on GFX10+.
>
> I will try to get some performance numbers on specific games. Do you know if this performance problem was specific to an architecture? 
> Like Jay said, I could tweak the heuristic for this and only generate it for GFX10+.

All the tests were done on gfx9 flavors. It is also worth noting gfx9 does not have movrel, but uses s_set_gpr_idx_on instead, which may be the difference.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D126389/new/

https://reviews.llvm.org/D126389