Rajalakshmi Srinivasaraghavan 0826d68f93 POWER10: Change the packing format for bfloat16 ...

As the new MMA instructions need the inputs in 4x2 order for bfloat16,
changing the format in copy/packing code.  This avoids permute instructions
in the gemm kernel inner loop.

2020-10-13 16:05:10 -05:00

..

alpha

Add implementations of ssum/dsum and csum/zsum

2019-03-30 22:05:11 +01:00

arm

Optimize the performance of sum by using universal intrinsics

2020-10-12 19:48:53 +08:00

arm64

Merge pull request #2867 from Qiyu8/usimd-floatdot

2020-10-10 12:10:25 +02:00

generic

Rename "HALF" and "sh" to "BFLOAT16" and "sb"

2020-10-11 23:56:17 +02:00

ia64

Add ia64 implementation of ?sum

2019-03-30 22:18:03 +01:00

mips

Delete KERNEL.1004K

2020-04-19 15:44:30 +02:00

mips64

Fix compilation problem on loongson platform

2020-04-09 19:28:15 +08:00

power

POWER10: Change the packing format for bfloat16

2020-10-13 16:05:10 -05:00

simd

Optimize the performance of dot by using universal intrinsics in X86/ARM

2020-09-28 20:36:53 +08:00

sparc

Add SPARC implementation of ?sum

2019-03-30 22:25:06 +01:00

x86

Enable COOPERLAKE build target

2020-08-13 06:18:00 +08:00

x86_64

Merge pull request #2886 from martin-frbg/issue_2767

2020-10-13 00:04:35 +02:00

zarch

s390x: fix cscal and zscal implementations

2020-09-21 13:10:05 +02:00

CMakeLists.txt

Change "HALF" and "sh" to "BFLOAT16" and "sb"

2020-10-12 00:00:55 +02:00

Makefile

Add whitelist of DYNAMIC_ARCH kernels for which -msse3 needs to be enabled

2020-10-11 01:06:46 +02:00

Makefile.L1

Fix build issues with bfloat16

2020-10-13 11:00:22 -05:00

Makefile.L2

Allow compiling only a subset of kernels for specific variable types

2020-10-11 14:52:09 +02:00

Makefile.L3

Fix build issues with bfloat16

2020-10-13 11:00:22 -05:00

Makefile.LA

Support NO_LAPACK=1 to build the lib without LAPACK functions.

2011-03-04 11:51:32 +08:00

setparam-ref.c

Change "HALF" and "sh" to "BFLOAT16" and "sb"

2020-10-12 00:00:55 +02:00