| Issue |
178795
|
| Summary |
[X86] Missed Fold: `vgf2p8affineqb(x, m, imm8) ^ C` => `vgf2p8affineqb(x, m, imm8 ^ C)`
|
| Labels |
new issue
|
| Assignees |
|
| Reporter |
WalterKruger
|
The immediate in the `vgf2p8affineqb` instruction applies a XOR after the main affine transformation. This can be folded with standalone XORs by any splatted 8-bit constants.
```asm
affineNot_clang:
gf2p8affineqb xmm0, xmm1, 0
pcmpeqd xmm1, xmm1
pxor xmm0, xmm1
ret
```
```asm
affineNot_tgt:
gf2p8affineqb xmm0, xmm1, 255
ret
```
https://godbolt.org/z/4qhG5fGq3
![]()
_______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
