AMDGPU: Fix true16 d16 entry table for DS pseudos #156419
Merged
Conversation
This file contains hidden or 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
arsenm
requested review from
jayfoad,
kosarev,
mariusz-sikora-at-amd,
mbrkusanin,
rampitec,
shiltian and
Sisyph
This stack of pull requests is managed by Graphite. Learn more about stacking. |
rampitec
reviewed
Sep 2, 2025
| : DS_1A1D_NORET_mc<opName, rc> { | ||
| let has_m0_read = 0 in { | ||
| let True16Predicate = UseRealTrue16Insts in { | ||
| def "_t16" : DS_1A1D_NORET<opName#"_t16", VGPR_16>, True16D16Table<NAME#"_D16_HI", NAME>; | ||
| def "_t16" : DS_1A1D_NORET<opName#"_t16", VGPR_16>, | ||
| True16D16Table<NAME#"_D16_HI", NAME#"_gfx9">; |
There was a problem hiding this comment.
Why gfx9, it does not have t16?
There was a problem hiding this comment.
Oh, I see, we do not define gfx11/gfx12 pseudos.
rampitec
approved these changes
Sep 2, 2025
| : DS_1A1D_NORET_mc<opName, rc> { | ||
| let has_m0_read = 0 in { | ||
| let True16Predicate = UseRealTrue16Insts in { | ||
| def "_t16" : DS_1A1D_NORET<opName#"_t16", VGPR_16>, True16D16Table<NAME#"_D16_HI", NAME>; | ||
| def "_t16" : DS_1A1D_NORET<opName#"_t16", VGPR_16>, | ||
| True16D16Table<NAME#"_D16_HI", NAME#"_gfx9">; |
There was a problem hiding this comment.
Oh, I see, we do not define gfx11/gfx12 pseudos.
arsenm
changed the base branch from
users/arsenm/amdgpu/true16-handle-ds-write-b8-b16
to
main
This should be trying to use the _gfx9 variants of DS pseudos, not the base form with m0 uses.
arsenm
force-pushed
the
users/arsenm/amdgpu/fix-true16-d16-table-pseudo-entry
branch
from
849b10b to
21fa4b7
Compare
arsenm
deleted the
users/arsenm/amdgpu/fix-true16-d16-table-pseudo-entry
branch
This was referenced Sep 3, 2025
This was referenced Sep 3, 2025
|
@llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) ChangesThis should be trying to use the _gfx9 variants of DS pseudos, Full diff: https://github.com/llvm/llvm-project/pull/156419.diff 1 Files Affected:
diff --git a/llvm/lib/Target/AMDGPU/DSInstructions.td b/llvm/lib/Target/AMDGPU/DSInstructions.td
index 2ff9dfe8a7014..e6a07ebe1cafb 100644
--- a/llvm/lib/Target/AMDGPU/DSInstructions.td
+++ b/llvm/lib/Target/AMDGPU/DSInstructions.td
@@ -127,11 +127,12 @@ multiclass DS_1A1D_NORET_mc<string opName, RegisterClass rc = VGPR_32> {
}
}
-multiclass DS_1A1D_NORET_t16<string opName, RegisterClass rc = VGPR_32>
+multiclass DS_1A1D_NORET_t16<string opName, RegisterClass rc = VGPR_32>
: DS_1A1D_NORET_mc<opName, rc> {
let has_m0_read = 0 in {
let True16Predicate = UseRealTrue16Insts in {
- def "_t16" : DS_1A1D_NORET<opName#"_t16", VGPR_16>, True16D16Table<NAME#"_D16_HI", NAME>;
+ def "_t16" : DS_1A1D_NORET<opName#"_t16", VGPR_16>,
+ True16D16Table<NAME#"_D16_HI", NAME#"_gfx9">;
}
}
}
|
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.
This should be trying to use the _gfx9 variants of DS pseudos,
not the base form with m0 uses.