Cold-Air-Networks/mirror-llvm-project
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 7

[NFC][AMDGPU] Cleanup in SOPInstructions.td (#210367)

Browse source 
- Eliminate code which was accidentally merged during a merge-conflict
(https://github.com/llvm/llvm-project/pull/203155/changes#r3560059941)
- Remove comments as they are specific to usubsat.
This commit is contained in:
Anshil Gandhi 2026-07-17 13:13:52 -04:00 committed by GitHub
parent 53c048b373
commit a67eac3e14
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 0 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
llvm/lib/Target/AMDGPU/SOPInstructions.td
View file

@ -2073,8 +2073,6 @@ let AddedComplexity = 20 in {
>;
}
// Uniform saturating unsigned subtraction: usubsat(a, b) = max(a, b) - b
// If a >= b, result is a - b; otherwise max(a, b) = b, so b - b = 0.
// Higher complexity to prefer scalar over VALU patterns for uniform values.
let AddedComplexity = 20 in {
def : GCNPat<
@ -2087,13 +2085,6 @@ let AddedComplexity = 20 in {
>;
} // End AddedComplexity = 20
// V_ADD_I32_e32/S_ADD_U32 produces carry in VCC/SCC. For the vector
// case, the sgpr-copies pass will fix this to use the vector version.
def : GCNPat <
(i32 (addc i32:$src0, i32:$src1)),
(S_ADD_U32 $src0, $src1)
>;
// FIXME: We need to use COPY_TO_REGCLASS to work-around the fact that
// REG_SEQUENCE patterns don't support instructions with multiple
// outputs.