Add support for FP16 to openBLAS and shgemm on RISCV #5290
Conversation
…r RISCV64_ZVL256B Added HFLOAT16 support for RISCV64 Added shgemm_kernel_8x8 for RISCV64_ZVL128B and shgemm_kernel_16x8 for RISCV64_ZVL256B based on HFLOAT16 The instruction sets used are ZVFH and ZFH, which need to be supported by RVV1.0 Related to issue OpenMathLib#5279 Co-authored-by Linjin Li <[email protected]>
Added shgemm_kernel_8x8 for RISCV64_ZVL128B and shgemm_kernel_16x8 fo…
- modify the macro conditions in Makefile.system - Delete development test code Related to issue#5279
|
I think we'll need some adjustments in interface/gemm.c to disable the "small matrix" pathway, adjust the function name used in error messages, and also keep hfloat16 out of the optimum thread number computations, now that it is separate from bfloat16. Please see attached (and add this or similar to the PR if you agree) Also it probably makes sense to seperate the two types to such an extent that one can be built without the other - I have further modified your version of gensymbol(.pl) to take another parameter for the BUILD_HFLOAT16, and adjusted exports/Makefile accordingly: |
|
In similar vein, the affected files in the benchmark folder probably need adjusting as well, please check |
|
Lastly, I found that I cannot build the "generic" replacements on a non-RISCV host unless I remove the Unfortunately, I have not had the time to try a CMake build yet. |
-add HFLOAT16 and BUILD_HFLOAT16 macro define to distinguish BFLOAT16 and BUILD_BFLOAT16
-add shgemm for RISCV_ZVL128B and RISCV_ZVL256B
-using fp16 on RISCV requires zfh and zvfh instruction sets
-enable fp16 support in Makefile.rule
Related to issue #5279
Co-authored-by Ao Dong