Chao Liu f4acec502e Restructure gridwise and blockwise GEMM, add tensor contraction and FWD-v4r5 (#36) ...

* experimenting magic number division

* overhauling fwd-v4r4 to clearly reflect transformation graph

* added fwd-v4r5

* bug fix for make_dynamic_naive_tensor_descriptor_aligned_v2

* bug fix and added sanity-check in transform_dynamic_tensor_descriptor

* added conv_driver_v2

[ROCm/composable_kernel commit: 30072aec37]

2021-06-09 23:53:08 -05:00

..

cmake-cuda.sh

Code clean up (#20)

2020-06-23 20:31:27 -05:00

cmake-rocm3.1.sh

Code clean up (#20)

2020-06-23 20:31:27 -05:00

cmake-rocm.sh

Use DynamicBuffer instead of raw pointer (#32)

2021-05-12 13:10:42 -05:00

compile-rocm3.1.sh

Code clean up (#20)

2020-06-23 20:31:27 -05:00

count_vgpr.sh

Dynamic tensor descriptor (#24)

2021-03-25 13:51:11 -05:00

docker-rocm3.7.sh

No raw index calculation (#31)

2021-05-11 00:09:25 -05:00

extract_asm-cuda.sh

Added bwd data v3r1 v4r1, tweaking v1 (#10)

2020-01-20 10:20:03 -06:00

hipclang_opt.sh

Dynamic tensor descriptor (#24)

2021-03-25 13:51:11 -05:00

run.sh

Restructure gridwise and blockwise GEMM, add tensor contraction and FWD-v4r5 (#36)

2021-06-09 23:53:08 -05:00