Lewis Crawford 17efa572c3

[InstSimplify] Optimize maximumnum and minimumnum (#139581) ...

Add support for the new maximumnum and minimumnum intrinsics in various
optimizations in InstSimplify.

Also, change the behavior of optimizing maxnum(sNaN, x) to simplify to
qNaN instead of x to better match the LLVM IR spec, and add more tests
for sNaN behavior for all 3 max/min intrinsic types.

2025-10-07 14:23:32 +01:00

..

addrspacecast.ll

[ValueTracking] Return true for AddrSpaceCast in canCreateUndefOrPoison (#144686)

2025-06-24 08:43:47 +08:00

amdgcn-demanded-vector-elts-inseltpoison.ll

…

amdgcn-demanded-vector-elts.ll

…

amdgcn-intrinsics-gfx8.ll

…

amdgcn-intrinsics.ll

[AMDGPU] Add the support for 45-bit buffer resource (#159702)

2025-09-24 11:12:02 -04:00

amdgcn-simplify-image-buffer-stores.ll

…

bitcast-fold-lane-ops.ll

…

fma_legacy.ll

…

fmed3-fpext-fold.ll

…

fmed3.ll

[InstSimplify] Optimize maximumnum and minimumnum (#139581)

2025-10-07 14:23:32 +01:00

fmul_legacy.ll

…

image-d16.ll

[AMDGPU] Support D16 folding for image.sample with multiple extractelement and fptrunc users (#141758)

2025-06-18 09:00:07 +08:00

issue68120.ll

…

lane-index-simplify-demanded-bits.ll

…

lit.local.cfg

…

llvm.amdgcn.readfirstlane.ll

…

llvm.amdgcn.readlane.ll

…

llvm.amdgcn.wavefrontsize.ll

…

mbcnt.ll

…

memcpy-from-constant.ll

[InstCombine] Strip leading zero indices from GEP (#155415)

2025-09-01 09:58:11 +02:00

mfma-scale.ll

…

permlane64.ll

…

phi-with-incoming-from-load.ll

…

ptr-replace-alloca.ll

InstCombine: Check GEP operand is available (#160438)

2025-09-25 17:20:20 +09:00

rcp-contract-rsq.ll

…

select-from-load.ll

…

simplify-amdgcn.cvt.off.f32.i4.ll

…

simplify-demanded-vector-elts-lane-intrinsics.ll

…

tan.ll

…

trivially-uniform.ll

…

wmma-f8f6f4.ll

[AMDGPU] gfx1250 v_wmma_scale[16]_f32_16x16x128_f8f6f4 codegen (#152036)

2025-08-04 19:16:34 -07:00