[ROCm][Windows] Enable AOTriton runtime compile on Windows (#165538)#2726

Open

tvukovic-amd wants to merge 31 commits intorelease/2.9from

always_build_aotriton_runtime_on_windows

Commits on Oct 10, 2025

[release/2.8] Enable wheels
jithunnair-amd
committed
[release/2.8] Upgrade numpy versions; Use different package versions for py3.9;

jithunnair-amd
and
ethanwee1
committed
Use ROCm/triton and update triton.txt
jithunnair-amd
committed
Add related_commits file (#2396)

pragupta
authored and
jithunnair-amd
committed
Add QA automation scripts for running PyTorch unit tests
jithunnair-amd
committed
[AUTOGENERATED] [release/2.5] [ROCm][layer_norm] Use __builtin_amdgcn_rcpf(x) instead of 1.f/x (#1800)

authored and
jithunnair-amd
committed
[release/2.7] Update test_binary_ufuncs.py after numpy upgrade (#2289)

ethanwee1
authored and
jithunnair-amd
committed
Clean up CUDA state between tests (#2335)

rraminen
authored and
jithunnair-amd
committed
[AUTOGENERATED] [release/2.7] [release/2.6] Fix dtype before comparing torch and numpy tensors (#2362)

authored and
jithunnair-amd
committed
[release/2.8] enable py3.13 (#2366)

ethanwee1
authored and
jithunnair-amd
committed
[SWDEV-539076] Initial naive foreach autotune support (#2377)

jataylo
authored and
jithunnair-amd
committed
[release/2.7] [SWDEV-543214] Reland #2416 Fix warps runtime (#2421)

jataylo
and
jithunnair-amd
committed
[AUTOGENERATED] [release/2.8] [release/2.7] [SWDEV-543214] Reland #2416 Fix warps runtime part 2 (#2455)

authored and
jithunnair-amd
committed
[AUTOGENERATED] [release/2.8] [SWDEV-539215] - Autotune support for persistent reduction and no_x_dim removal (#2454)

authored and
jithunnair-amd
committed
[SWDEV-539119] [release/2.8] Add fast_tanh support (#2484)

jataylo
authored and
jithunnair-amd
committed
[AUTOGENERATED] [release/2.8] Change triton package name depending on rocm version (#2529)

authored and
jithunnair-amd
committed
[release/2.8] Define uint32 t when ROCM_VERSION >= 70000 (#2513)

rraminen
authored and
jithunnair-amd
committed
[AUTOGENERATED] [release/2.8] [ROCm] OffsetCalc Unroll Optimization (#2598)

authored and
jithunnair-amd
committed
Bug fix and optimisation for persistent reduction kernel tuning (#2596)

jataylo
authored and
jithunnair-amd
committed
[ROCm] Fix indexing_backward_kernel perf (#2667)

jerrymannil
authored and
jithunnair-amd
committed
[ROCm] Improve perf for elementwise broadcast with mixed dtype (#2672)

jerrymannil
authored and
jithunnair-amd
committed
[ROCm] Implement float32 copy kernel (#2683)

jerrymannil
authored and
jithunnair-amd
committed
Bump triton to 3.5.x

pragupta
authored and
jithunnair-amd
committed
Update fbgemm submodule to avoid ck errors

pragupta
authored and
jithunnair-amd
committed

Commits on Oct 14, 2025

Commits on Oct 15, 2025

Update version to 2.9.0
jithunnair-amd
authored

Commits on Oct 16, 2025

[ROCm] Fix non-stride-one backwards indexing performance

rocm-repo-management-api[bot]
and
doru1004
authored

Commits on Oct 17, 2025