forked from tinygrad/tinygrad
George HotzandGitHub
a73d8717f3
* 50 TFLOP AMD gemm * add lds tiling * register tiling * flip locals * work * comment * remove those
4 lines
24 B
Plaintext
4 lines
24 B
Plaintext
*.s
|
|
*.ll
|
|
fp32_sgemm_amd
|