projects / platform / upstream / openblas.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ff74319) | patch
Optimize sdot/ddot for POWER10
authorRajalakshmi Srinivasaraghavan <rajis@linux.ibm.com>
Sat, 7 Nov 2020 21:21:58 +0000 (15:21 -0600)
committerRajalakshmi Srinivasaraghavan <rajis@linux.ibm.com>
Sat, 7 Nov 2020 21:21:58 +0000 (15:21 -0600)
commit6e364981a8af0f72ad9e62a69fe62fdedc18255b
tree6e9657a0691093f63adad1fa85ca7ed4541f9227tree | snapshot
parentff74319ea5c5a3757af970076967389e63a0469bcommit | diff
Optimize sdot/ddot for POWER10

This patch makes use of new POWER10 vector pair instructions for
loads and stores.
kernel/power/KERNEL.POWER10 diff | blob | history
kernel/power/ddot_microk_power10.c [new file with mode: 0644] blob
kernel/power/ddot_power10.c [new file with mode: 0644] blob
kernel/power/sdot_microk_power10.c [new file with mode: 0644] blob
kernel/power/sdot_power10.c [new file with mode: 0644] blob
Domain: Machine Learning / ML Framework;