ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-03 05:01:25 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
d6a4605e1c77d73f5387bb4e5735411dafd6fb83
composable_kernel/include/ck/tensor_operation/gpu
History
darren-amd d6a4605e1c Change flag to CK_GFX90A_DENORM_WORKAROUND (#1817) 
* Change flag from CK_WORKAROUND_DENORM_FIX to CK_GFX90A_DENORM_WORKAROUND for more clarity. Also changed the definition macros to be more clear.
2025-01-28 09:58:39 -05:00
..
block
Implement the fp16xint4 scale weight only kernel for Ali (#1786)
2025-01-03 18:35:21 +08:00
device
Fix and optimize dynamic unary elementwise (#1818)
2025-01-16 13:48:39 -08:00
element
Fix and optimize dynamic unary elementwise (#1818)
2025-01-16 13:48:39 -08:00
grid
Change flag to CK_GFX90A_DENORM_WORKAROUND (#1817)
2025-01-28 09:58:39 -05:00
thread
Grouped convolution backward weight special vector size loads (#1772)
2025-01-10 22:02:30 +08:00
warp
terminology clean-up (#1792)
2025-01-03 16:38:22 -08:00