projects / platform / upstream / openblas.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: caf3394) | patch
ARM64: Use THUNDERX2T99 Neon Kernels for ARMV8
authorAshwin Sekhar T K <ashwin.sekhar@cavium.com>
Wed, 17 Oct 2018 15:11:27 +0000 (08:11 -0700)
committerAshwin Sekhar T K <ashwin.sekhar@cavium.com>
Wed, 17 Oct 2018 17:44:37 +0000 (10:44 -0700)
commit21f46a1cf2cefbdedf89878e3a6324578d0fe8ca
tree328d70d33b375e5832ccbd47551d5596b9233279tree | snapshot
parentcaf339412f9e828ffd3e43ec4b58ecd992eeff7acommit | diff
ARM64: Use THUNDERX2T99 Neon Kernels for ARMV8

Currently the generic ARMV8 target uses C implementations
for many routines. Replace these with the neon implementations
written for THUNDERX2T99 target which are upto 6x faster for
certain routines.
driver/others/parameter.c diff | blob | history
interface/swap.c diff | blob | history
kernel/arm64/KERNEL.ARMV8 diff | blob | history
param.h diff | blob | history
Domain: Machine Learning / ML Framework;