[XLA:GPU] Use DeviceDescription instead of hard-coding warp size as 32#2938

Open

amd-jianli12 wants to merge 5 commits intor2.18-rocm-enhancedfrom

r2.18-rocm-enhanced-warpsize

Commits on Apr 23, 2025

Use DeviceDescription instead of hard-coding warp size as 32

majnemer
authored and
amd-jianli12
committed

Commits on May 8, 2025

[ROCM] Adapt legacy reduction algorithm to warp size 64
amd-jianli12
committed

Commits on May 12, 2025

[ROCM] Fix GpuKernelTilingTest.UnnestedTransposeC128TypeRun
amd-jianli12
committed

Commits on May 13, 2025

[ROCM] Fix ParallelReductionTest.CouldBeThreeReductionGroups
amd-jianli12
committed

Commits on May 15, 2025

[ROCM] Clean up XLA unit tests for warp size 64
amd-jianli12
committed