Files
composable_kernel/include/ck
Kiefer van Teutem 04d4dd1ada Replace grouped conv bwd wei wmmaV3 bilin/scale bf16f32bf16 support with bf16bf16bf16 (#3470)
* Replace grouped convolution bwd weight wmma v3 bilinear and scale bf16f32bf16 support with bf16bf16bf16 support. Update tests.

* Tentative fix for bwd weight bilinear bf16bf16bf16, seems like the bilinear elementwise overload for this case (bf16, f32 accu, bf16) was wrong.

[ROCm/composable_kernel commit: 88ae445580]
2025-12-29 12:58:29 +01:00
..
2025-11-27 08:48:00 +01:00