Qianfeng 03fb9b9435 Replace the using of __expf by __ocml_exp_f32 to work-around the test_softmax_rank4 failure (#1394) ...

[ROCm/composable_kernel commit: ee768148f0]

2024-07-17 09:15:05 -07:00

..

block

Merging the gfx12 code into public repo. (#1362)

2024-06-27 00:33:34 -07:00

device

Support access per groups and filter3x3 in grouped conv fwd (#1382)

2024-07-12 11:08:42 -07:00

element

Replace the using of __expf by __ocml_exp_f32 to work-around the test_softmax_rank4 failure (#1394)

2024-07-17 09:15:05 -07:00

grid

Universal streamk with atomics (#1360)

2024-07-05 21:40:30 -07:00

thread

Merging the gfx12 code into public repo. (#1362)

2024-06-27 00:33:34 -07:00

warp

Add structural sparsity xdlops (#1363)

2024-07-04 12:00:14 +02:00