AMDGPU: Do not generate non-temporal hint when Load_Tr intrinsic did not specify it #79104
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Some upcoming intrinsics use these new types
Some upcoming intrinsics use these new types
It could be possible for getIConstantVRegValWithLookThrough to return std::nullopt.
It could be possible for getIConstantVRegValWithLookThrough to return std::nullopt.
…not specify it Summary: int_amdgcn_global_load_tr did not specify non-temporal load transpose, thus we should not genetrate the non-temporal hint for the load. We need to implement getTgtMemIntrinsic to create the corresponding MemSDNode. And we don't set the non-temporal flag because the intrinsic did not specify it. NOTE: We need to implement getTgtMemIntrinsic for any memory intrinsics.
jayfoad
reviewed
Jan 23, 2024
| @@ -1348,6 +1348,14 @@ bool SITargetLowering::getTgtMemIntrinsic(IntrinsicInfo &Info, | |||
| MachineMemOperand::MOVolatile; | |||
| return true; | |||
| } | |||
| case Intrinsic::amdgcn_global_load_tr: { | |||
There was a problem hiding this comment.
This case should also be handled in getAdrModeArguments below.
piotrAMD
reviewed
Jan 23, 2024
| @@ -13,9 +13,8 @@ define amdgpu_kernel void @global_load_tr_b64(ptr addrspace(1) %addr, ptr addrsp | |||
| ; GFX12-SDAG-W32-NEXT: s_load_b128 s[0:3], s[0:1], 0x24 | |||
| ; GFX12-SDAG-W32-NEXT: v_mov_b32_e32 v2, 0 | |||
| ; GFX12-SDAG-W32-NEXT: s_wait_kmcnt 0x0 | |||
| ; GFX12-SDAG-W32-NEXT: global_load_tr_b64 v[0:1], v2, s[0:1] offset:32 th:TH_LOAD_NT | |||
| ; GFX12-SDAG-W32-NEXT: global_load_tr_b64 v[0:1], v2, s[0:1] offset:32 | |||
…not specify it Summary: int_amdgcn_global_load_tr did not specify non-temporal load transpose, thus we should not genetrate the non-temporal hint for the load. We need to implement getTgtMemIntrinsic to create the corresponding MemSDNode. And we don't set the non-temporal flag because the intrinsic did not specify it. NOTE: We need to implement getTgtMemIntrinsic for any memory intrinsics.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
int_amdgcn_global_load_tr did not specify non-temporal load transpose, thus we should
not genetrate the non-temporal hint for the load. We need to implement getTgtMemIntrinsic
to create the corresponding MemSDNode. And we don't set the non-temporal flag because
the intrinsic did not specify it.
NOTE: We need to implement getTgtMemIntrinsic for any memory intrinsics.