ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 23:09:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

RE-enable DL and DPP instances by default. (#1954)

Browse Source 
* enable DL and DPP instances by default

* fix cmake logic
This commit is contained in:
Illia Silin
2025-03-06 21:45:31 -08:00
committed by GitHub
parent 7a4a5d6c08
commit 43c90b5234
4 changed files with 18 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -158,12 +158,12 @@ Additional cmake flags can be used to significantly speed-up the build:
instances of select data types only. The main default data types are fp32 and fp16; you can safely skip
other data types.
* `DL_KERNELS` (default is OFF) must be set to ON in order to build instances, such as `gemm_dl` or
* `DISABLE_DL_KERNELS` (default is OFF) must be set to ON in order not to build instances, such as `gemm_dl` or
`batched_gemm_multi_d_dl`. These instances are useful on architectures like the NAVI2x, as most
other platforms have faster instances, such as `xdl` or `wmma`, available.
* `DPP_KERNELS` (default is OFF) must be set to ON in order to build instances, such as `gemm_dpp`.
These instances are useful on architectures like the NAVI2x, as most other platforms have faster instances, such as `xdl` or `wmma`, available.
* `DISABLE_DPP_KERNELS` (default is OFF) must be set to ON in order not to build instances, such as `gemm_dpp`.
These instances offer a slightly better performance of fp16 gemms on NAVI2x. But on other architectures faster alternatives are available.
* `CK_USE_FP8_ON_UNSUPPORTED_ARCH` (default is OFF) must be set to ON in order to build instances,
such as `gemm_universal`, `gemm_universal_streamk` and `gemm_multiply_multiply` for fp8 data type for GPU targets which do not have native support for fp8 data type, such as gfx908 or gfx90a. These instances are useful on