projects / platform / upstream / openblas.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1a00ef3) | patch
Add an AVX512 enabled DGEMV (n) function
authorArjan van de Ven <arjan@linux.intel.com>
Sat, 11 Aug 2018 17:38:12 +0000 (17:38 +0000)
committerArjan van de Ven <arjan@linux.intel.com>
Sat, 11 Aug 2018 17:38:12 +0000 (17:38 +0000)
commit87bebdbd8aacf30741222b722d5f7bde1e51c739
tree5c361fa1321bc295abd51f4ed86e97f65649becftree | snapshot
parent1a00ef3d271d0335f2f96f955f16acff05f6f6cdcommit | diff
Add an AVX512 enabled DGEMV (n)  function

written in C intrinsics for best readability.
(the same C code works for Haswell as well)

For logistical reasons the code falls back to the existing
haswell AVX2 implementation if the GCC or LLVM compiler is not new enough
kernel/x86_64/dgemv_n_4.c diff | blob | history
kernel/x86_64/dgemv_n_microk_skylakex-4.c [new file with mode: 0644] blob
Domain: Machine Learning / ML Framework;