3 people 6881c754e1 use HIP_VERSION variables to guard against duplicate atomicAdd definitions (#41802) ...

Signed-off-by: Philip Maybank <pmaybank@amd.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>

2026-05-08 18:44:37 -04:00

..

awq

[Kernel] Fix awq error when n is not divisable by 128 (#13227)

2025-02-13 20:07:05 -08:00

fused_kernels

[ROCm] Enable fused_silu_mul_block_quant on ROCm (#38817)

2026-04-08 11:23:32 -05:00

gguf

[Bugfix][ROCm] Fix for warp_size uses on host (#21205)

2025-07-24 00:37:19 -07:00

gptq

use HIP_VERSION variables to guard against duplicate atomicAdd definitions (#41802)

2026-05-08 18:44:37 -04:00

gptq_allspark

[Refactor] Rename gptq_marlin to marlin to match MoE (#32952)

2026-01-23 16:48:12 -05:00

hadamard/hadacore

Use narrow over indexing in hadacore_transform to prep for ABI stable (#28756)

2025-11-15 01:10:15 -08:00

machete

[NVIDIA] Bugfix NVFP4 DGX Spark and RTX50 (#38423)

2026-03-30 09:36:18 -07:00

marlin

[Kernel] Add MXFP8 to Marlin GEMM/MoE and refactor Mxfp8LinearOp (#34664)

2026-04-01 09:41:42 -07:00

w8a8

[Bugfix] Fix broken explicit unquantized kv cache dtype support (#38922)

2026-04-09 22:27:53 -07:00

activation_kernels.cu

[Bugfix]fix output Nan/Inf in marlin if dtype=float16 (#33972)

2026-03-27 16:36:08 -07:00

utils.cuh

[Feature][ROCm]Enable fusion pass for torch.compile on ROCm (#15050)

2025-03-31 04:42:18 -07:00