Matthias GehreandGitHub 77c1d9fe9b [ROCm][Perf] Tune wvSplitK on gfx1151 (#40784) ...

Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>

2026-06-25 14:17:46 +08:00

..

attention.cu

[ROCm] Fix OOB During Model Warmup With ROCM_ATTN and MRV2 (#46548)

2026-06-24 10:53:21 -05:00

moe_q_gemm_rdna3.cu

[ROCm][Perf] Fused MoE W4A16 HIP kernel for AMD RDNA3 (gfx1100) (#44075)

2026-06-06 15:30:39 -05:00

ops.h

[ROCm][Perf] Fused MoE W4A16 HIP kernel for AMD RDNA3 (gfx1100) (#44075)

2026-06-06 15:30:39 -05:00

q_gemm_rdna3_wmma.cu

[Kernel][ROCm] Native W4A16 kernel for AMD RDNA3 (gfx1100) — fp16 + bf16 (#41394)

2026-05-29 11:04:40 +00:00

q_gemm_rdna3.cu

[Kernel][ROCm] Native W4A16 kernel for AMD RDNA3 (gfx1100) — fp16 + bf16 (#41394)

2026-05-29 11:04:40 +00:00

qdq_4_rdna3.cuh

[Kernel][ROCm] Native W4A16 kernel for AMD RDNA3 (gfx1100) — fp16 + bf16 (#41394)

2026-05-29 11:04:40 +00:00

skinny_gemms.cu

[ROCm][Perf] Tune wvSplitK on gfx1151 (#40784)

2026-06-25 14:17:46 +08:00

torch_bindings.cpp

[ROCm][Perf] Fused MoE W4A16 HIP kernel for AMD RDNA3 (gfx1100) (#44075)

2026-06-06 15:30:39 -05:00