OpenMathLib/OpenBLAS
1
0
Fork 0
mirror of https://github.com/OpenMathLib/OpenBLAS synced 2026-05-31 00:45:48 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
Files
b329e45288c2e7fc0ef15c4e8a7b3c8dfd74a930
OpenBLAS/kernel/x86_64
History
Martin Kroeker 5378046abd roll back DGEMM kernels to 4x8 when compiling for DYNAMIC_ARCH
2021-12-06 19:43:54 +01:00
..
amax_atom.S
…
amax_sse2.S
…
amax_sse.S
…
amax.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
asum_atom.S
…
asum_sse2.S
…
asum_sse.S
…
asum.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
axpy_atom.S
…
axpy_sse2.S
…
axpy_sse.S
…
axpy.S
…
bf16_common_macros.h
x86_64: BFLOAT16: fix build warning
2021-09-28 18:30:06 +08:00
bf16to.c
Add bfloat16 based dot and conversion with single/double
2020-09-04 02:31:25 +08:00
builtin_stinit.S
…
cabs.S
…
casum_microk_skylakex-2.c
Initialize abs_mask1 with itself to silence a gcc warning
2021-09-15 22:10:43 +02:00
casum.c
fix error declare function blas_level1_thread_with_return_value
2020-12-02 09:51:52 +08:00
caxpy_microk_bulldozer-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
caxpy_microk_haswell-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
caxpy_microk_sandy-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
caxpy_microk_steamroller-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
caxpy.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
cdot_microk_bulldozer-2.c
…
cdot_microk_haswell-2.c
…
cdot_microk_sandy-2.c
…
cdot_microk_steamroller-2.c
…
cdot.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
cgemm3m_kernel_8x4_haswell.c
Update cgemm3m_kernel_8x4_haswell.c
2019-12-27 18:23:29 +08:00
cgemm_kernel_4x2_bulldozer.S
…
cgemm_kernel_4x2_piledriver.S
…
cgemm_kernel_4x8_sandy.S
…
cgemm_kernel_8x2_haswell.c
Update cgemm_kernel_8x2_haswell.c
2020-02-27 22:26:15 +08:00
cgemm_kernel_8x2_haswell.S
…
cgemm_kernel_8x2_sandy.S
…
cgemm_kernel_8x2_skylakex.c
AVX512 CGEMM & ZGEMM kernels
2019-11-11 20:04:52 +08:00
cgemv_n_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
cgemv_n_microk_bulldozer-4.c
…
cgemv_n_microk_haswell-4.c
…
cgemv_n.S
…
cgemv_t_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
cgemv_t_microk_bulldozer-4.c
…
cgemv_t_microk_haswell-4.c
…
cgemv_t.S
…
copy_sse2.S
Convert aligned moves to unaligned
2020-04-13 14:58:52 +02:00
copy_sse.S
…
copy.S
…
cscal_microk_bulldozer-2.c
…
cscal_microk_haswell-2.c
…
cscal_microk_steamroller-2.c
…
cscal.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
ctrsm_kernel_LN_bulldozer.c
…
ctrsm_kernel_LT_bulldozer.c
…
ctrsm_kernel_RN_bulldozer.c
…
ctrsm_kernel_RT_bulldozer.c
…
dasum_microk_haswell-2.c
Add casts
2021-09-11 13:38:28 +02:00
dasum_microk_skylakex-2.c
Add casts
2021-09-14 21:41:53 +02:00
dasum.c
Use Haswell optimizations for Zen as well
2021-02-11 09:26:15 +01:00
daxpy_bulldozer.S
…
daxpy_microk_bulldozer-2.c
…
daxpy_microk_haswell-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
daxpy_microk_nehalem-2.c
…
daxpy_microk_piledriver-2.c
…
daxpy_microk_sandy-2.c
…
daxpy_microk_skylakex-2.c
…
daxpy_microk_steamroller-2.c
…
daxpy.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dcopy_bulldozer.S
…
ddot_bulldozer.S
…
ddot_microk_bulldozer-2.c
…
ddot_microk_haswell-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
ddot_microk_nehalem-2.c
…
ddot_microk_piledriver-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
ddot_microk_sandy-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
ddot_microk_skylakex-2.c
…
ddot_microk_steamroller-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
ddot.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dgemm_beta_skylakex.c
…
dgemm_kernel_4x4_haswell.S
…
dgemm_kernel_4x8_haswell.S
…
dgemm_kernel_4x8_sandy.S
…
dgemm_kernel_4x8_skylakex_2.c
Update dgemm_kernel_4x8_skylakex_2.c
2019-11-28 19:56:35 +08:00
dgemm_kernel_4x8_skylakex.c
…
dgemm_kernel_6x4_piledriver.S
…
dgemm_kernel_8x2_bulldozer.S
…
dgemm_kernel_8x2_piledriver.S
…
dgemm_kernel_8x8_skylakex.c
Update dgemm_kernel_8x8_skylakex.c
2019-10-18 15:00:17 +08:00
dgemm_kernel_16x2_haswell.S
…
dgemm_kernel_16x2_skylakex.c
GEMM: skylake: improve the performance when m is small
2021-04-28 13:56:06 +00:00
dgemm_kernel_16x2_skylakex.S
…
dgemm_ncopy_2.S
…
dgemm_ncopy_4.S
…
dgemm_ncopy_8_bulldozer.S
…
dgemm_ncopy_8_skylakex.c
…
dgemm_ncopy_8.S
…
dgemm_small_kernel_nn_skylakex.c
Small Matrix: skylakex: fix build error in old compiler
2021-08-05 04:43:47 +00:00
dgemm_small_kernel_nt_skylakex.c
Small Matrix: skylakex: fix build error in old compiler
2021-08-05 04:43:47 +00:00
dgemm_small_kernel_permit_skylakex.c
Small Matrix: skylakex: add DGEMM_SMALL_M_PERMIT and tune for TN kernel
2021-08-02 07:06:54 +00:00
dgemm_small_kernel_tn_skylakex.c
Small Matrix: skylakex: fix build error in old compiler
2021-08-05 04:43:47 +00:00
dgemm_small_kernel_tt_skylakex.c
Small Matrix: skylakex: fix build error in old compiler
2021-08-05 04:43:47 +00:00
dgemm_tcopy_2.S
…
dgemm_tcopy_4.S
…
dgemm_tcopy_8_bulldozer.S
…
dgemm_tcopy_8_skylakex.c
…
dgemm_tcopy_8.S
…
dgemm_tcopy_16_skylakex.c
Fix build with -Werror=return-type
2020-10-21 08:43:39 +02:00
dgemv_n_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dgemv_n_atom.S
…
dgemv_n_bulldozer.S
…
dgemv_n_microk_haswell-4.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
dgemv_n_microk_nehalem-4.c
…
dgemv_n_microk_piledriver-4.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
dgemv_n_microk_skylakex-4.c
…
dgemv_n.S
…
dgemv_t_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dgemv_t_atom.S
…
dgemv_t_bulldozer.S
…
dgemv_t_microk_haswell-4.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
dgemv_t.S
…
dger_microk_sandy-2.c
…
dger.c
…
dot_atom.S
…
dot_sse2.S
…
dot_sse.S
…
dot.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
drot_microk_haswell-2.c
replace spurious avx512 requirement with fma check
2021-04-26 21:55:30 +02:00
drot_microk_skylakex-2.c
Improve the performance of rot by using AVX512 and AVX2 intrinsic
2020-11-05 15:12:36 +08:00
drot.c
Use Haswell optimizations for Zen as well
2021-02-11 09:24:16 +01:00
dscal_microk_bulldozer-2.c
…
dscal_microk_haswell-2.c
…
dscal_microk_sandy-2.c
…
dscal_microk_skylakex-2.c
…
dscal.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dsymv_L_microk_bulldozer-2.c
…
dsymv_L_microk_haswell-2.c
…
dsymv_L_microk_nehalem-2.c
…
dsymv_L_microk_sandy-2.c
…
dsymv_L_microk_skylakex-2.c
Duplicate earlier Clang 9.0.0 workaround for corresponding Apple Clang version
2020-05-05 10:44:50 +02:00
dsymv_L.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dsymv_U_microk_bulldozer-2.c
…
dsymv_U_microk_haswell-2.c
…
dsymv_U_microk_nehalem-2.c
…
dsymv_U_microk_sandy-2.c
…
dsymv_U.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
dtobf16_microk_cooperlake.c
Add bfloat16 based dot and conversion with single/double
2020-09-04 02:31:25 +08:00
dtrmm_kernel_4x8_haswell.c
…
dtrsm_kernel_LN_bulldozer.c
…
dtrsm_kernel_LT_8x2_bulldozer.S
…
dtrsm_kernel_RN_8x2_bulldozer.S
…
dtrsm_kernel_RN_haswell.c
…
dtrsm_kernel_RT_bulldozer.c
…
gemm_beta.S
…
gemm_kernel_2x8_nehalem.S
…
gemm_kernel_4x2_atom.S
…
gemm_kernel_4x4_barcelona.S
…
gemm_kernel_4x4_core2.S
…
gemm_kernel_4x4_penryn.S
…
gemm_kernel_4x4_sse2.S
…
gemm_kernel_4x4_sse3.S
…
gemm_kernel_4x8_nano.S
…
gemm_kernel_4x8_nehalem.S
…
gemm_kernel_8x4_barcelona.S
…
gemm_kernel_8x4_core2.S
…
gemm_kernel_8x4_penryn.S
…
gemm_kernel_8x4_sse3.S
…
gemm_kernel_8x4_sse.S
…
gemm_ncopy_2_bulldozer.S
…
gemm_ncopy_2.S
…
gemm_ncopy_4_opteron.S
…
gemm_ncopy_4.S
…
gemm_tcopy_2_bulldozer.S
…
gemm_tcopy_2.S
…
gemm_tcopy_4_opteron.S
…
gemm_tcopy_4.S
…
iamax_sse2.S
…
iamax_sse.S
Silence a redefinition warning
2020-10-15 19:08:12 +02:00
iamax.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
izamax_sse2.S
…
izamax_sse.S
…
izamax.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
KERNEL
Remove premature entry for DOMATCOPY_RT
2021-03-18 21:53:50 +01:00
KERNEL.ATOM
…
KERNEL.BARCELONA
…
KERNEL.BOBCAT
…
KERNEL.BULLDOZER
…
KERNEL.COOPERLAKE
sbgemm: cooperlake: change kernel size to 16x4
2021-09-07 21:30:45 +08:00
KERNEL.CORE2
…
KERNEL.DUNNINGTON
…
KERNEL.EXCAVATOR
…
KERNEL.generic
…
KERNEL.HASWELL
Improve the performance of rot by using AVX512 and AVX2 intrinsic
2020-11-05 15:12:36 +08:00
KERNEL.NANO
…
KERNEL.NEHALEM
…
KERNEL.OPTERON
…
KERNEL.OPTERON_SSE3
…
KERNEL.PENRYN
…
KERNEL.PILEDRIVER
…
KERNEL.PRESCOTT
…
KERNEL.SANDYBRIDGE
…
KERNEL.SAPPHIRERAPIDS
sbgemm: spr: disable small matrix path by default
2021-10-17 19:08:03 -07:00
KERNEL.SKYLAKEX
roll back DGEMM kernels to 4x8 when compiling for DYNAMIC_ARCH
2021-12-06 19:43:54 +01:00
KERNEL.STEAMROLLER
…
KERNEL.ZEN
Enable optimized srot/drot kernels from Haswell
2021-02-11 09:23:05 +01:00
lsame.S
…
Makefile
…
mcount.S
…
nrm2_sse.S
…
nrm2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
omatcopy_rt.c
Update omatcopy_rt.c
2021-02-24 09:34:14 +01:00
qconjg.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qdot.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qgemm_kernel_2x2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qgemv_n.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qgemv_t.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qtrsm_kernel_LN_2x2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qtrsm_kernel_LT_2x2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
qtrsm_kernel_RT_2x2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
rot_sse2.S
…
rot_sse.S
…
rot.S
…
sasum_microk_haswell-2.c
Add casts
2021-09-11 13:38:28 +02:00
sasum_microk_skylakex-2.c
Add casts
2021-09-14 21:41:53 +02:00
sasum.c
Use Haswell optimizations for Zen as well
2021-02-11 09:25:36 +01:00
saxpy_microk_haswell-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
saxpy_microk_nehalem-2.c
…
saxpy_microk_piledriver-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
saxpy_microk_sandy-2.c
…
saxpy_microk_skylakex-2.c
…
saxpy.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sbdot_microk_cooperlake.c
x86_64: BFLOAT16: fix build warning
2021-09-28 18:30:06 +08:00
sbdot.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sbgemm_block_microk_cooperlake.c
x86_64: BFLOAT16: fix build warning
2021-09-28 18:30:06 +08:00
sbgemm_kernel_16x4_cooperlake.c
sbgemm: cooperlake: prefetch A & B
2021-09-07 21:30:46 +08:00
sbgemm_kernel_16x16_spr_tmpl.c
sbgemm: spr: optimization for tmp_c buffer
2021-10-17 19:08:03 -07:00
sbgemm_kernel_16x16_spr.c
sbgemm: spr: kernel handle alpha != 1.0
2021-10-17 19:08:03 -07:00
sbgemm_microk_cooperlake_template.c
sbgemm: cooperlake: enable SBGEMM by small matrix path
2021-08-30 17:40:30 +08:00
sbgemm_ncopy_4_cooperlake.c
sbgemm: cooperlake: kernel works for NN
2021-09-07 21:30:45 +08:00
sbgemm_ncopy_16_cooperlake.c
sbgemm: cooperlake: implement ncopy_16
2021-09-07 21:30:46 +08:00
sbgemm_oncopy_16_spr.c
sbgemm: spr: oncopy: use tile load/store instead
2021-10-17 19:08:03 -07:00
sbgemm_otcopy_16_spr.c
sbgemm: spr: implement otcopy_16
2021-10-17 19:08:03 -07:00
sbgemm_small_kernel_nn_cooperlake.c
sbgemm: cooperlake: enable SBGEMM by small matrix path
2021-08-30 17:40:30 +08:00
sbgemm_small_kernel_nt_cooperlake.c
sbgemm: cooperlake: enable SBGEMM by small matrix path
2021-08-30 17:40:30 +08:00
sbgemm_small_kernel_permit_cooperlake.c
sbgemm: cooperlake: tuning for small matrix
2021-09-07 21:30:46 +08:00
sbgemm_small_kernel_permit_spr.c
sbgemm: spr: disable small matrix path by default
2021-10-17 19:08:03 -07:00
sbgemm_small_kernel_template_cooperlake.c
sbgemm: cooperlake: make sure hot buffer aligned to 64
2021-08-30 17:40:30 +08:00
sbgemm_small_kernel_tn_cooperlake.c
sbgemm: cooperlake: enable SBGEMM by small matrix path
2021-08-30 17:40:30 +08:00
sbgemm_small_kernel_tt_cooperlake.c
sbgemm: cooperlake: enable SBGEMM by small matrix path
2021-08-30 17:40:30 +08:00
sbgemm_tcopy_4_cooperlake.c
sbgemm: cooperlake: add n24 kernel for tcopy_4
2021-09-07 21:30:46 +08:00
sbgemm_tcopy_16_cooperlake.c
sbgemm: cooperlake: implement tcopy_4
2021-09-07 21:30:46 +08:00
sbgemv_n_microk_cooperlake_template.c
x86_64: BFLOAT16: fix build warning
2021-09-28 18:30:06 +08:00
sbgemv_n_microk_cooperlake.c
Implementation of BF16 based gemv
2020-10-29 02:08:23 +08:00
sbgemv_n.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sbgemv_t_microk_cooperlake_template.c
x86_64: BFLOAT16: fix build warning
2021-09-28 18:30:06 +08:00
sbgemv_t_microk_cooperlake.c
Implementation of BF16 based gemv
2020-10-29 02:08:23 +08:00
sbgemv_t.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
scal_atom.S
…
scal_sse2.S
…
scal_sse.S
…
scal.S
…
sdot_microk_bulldozer-2.c
…
sdot_microk_haswell-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
sdot_microk_nehalem-2.c
…
sdot_microk_sandy-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
sdot_microk_skylakex-2.c
…
sdot_microk_steamroller-2.c
…
sdot.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sgemm_beta_skylakex.c
sbgemm: cooperlake: add dummy source files
2021-09-07 21:30:45 +08:00
sgemm_direct_performant.c
[WIP] Refactor the driver code for direct SGEMM (#2782)
2020-08-19 14:51:09 +02:00
sgemm_direct_skylakex.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sgemm_kernel_8x4_bulldozer.S
…
sgemm_kernel_8x4_haswell_2.c
Strip UTF8 byte order marker from source
2020-06-26 09:00:43 +02:00
sgemm_kernel_8x4_haswell.c
Update sgemm_kernel_8x4_haswell.c
2020-02-06 01:47:46 +00:00
sgemm_kernel_8x8_sandy.S
…
sgemm_kernel_16x2_bulldozer.S
…
sgemm_kernel_16x2_piledriver.S
…
sgemm_kernel_16x4_haswell.S
…
sgemm_kernel_16x4_sandy.S
…
sgemm_kernel_16x4_skylakex_2.c
AVX512 STRMM kernel
2020-02-16 22:58:00 +08:00
sgemm_kernel_16x4_skylakex_3.c
Use "old" compute(24) function with clang due to register limitations
2021-04-06 19:58:32 +02:00
sgemm_kernel_16x4_skylakex.c
make skylakex sgemm code more friendly for readers
2020-01-13 16:28:41 +08:00
sgemm_kernel_16x4_skylakex.S
…
sgemm_ncopy_4_skylakex.c
…
sgemm_small_kernel_nn_skylakex.c
Small Matrix: skylakex: sgemm nn: fix n6 conflicts with n4
2021-08-02 07:06:54 +00:00
sgemm_small_kernel_nt_skylakex.c
Small Matrix: skylakex: fix build error in old compiler
2021-08-05 04:43:47 +00:00
sgemm_small_kernel_permit_skylakex.c
Small Matrix: skylakex: add sgemm tt kernel
2021-08-02 07:06:54 +00:00
sgemm_small_kernel_tn_skylakex.c
Small Matrix: skylakex: add sgemm tn kernel
2021-08-02 07:06:54 +00:00
sgemm_small_kernel_tt_skylakex.c
Small Matrix: skylakex: fix build error in old compiler
2021-08-05 04:43:47 +00:00
sgemm_tcopy_16_skylakex.c
…
sgemv_n_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sgemv_n_microk_bulldozer-4.c
…
sgemv_n_microk_haswell-4.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
sgemv_n_microk_nehalem-4.c
…
sgemv_n_microk_sandy-4.c
…
sgemv_n_microk_skylakex-8.c
optimize on sgemv_n for small n
2021-04-30 12:14:58 -04:00
sgemv_n.c
…
sgemv_n.S
…
sgemv_t_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
sgemv_t_microk_bulldozer-4.c
…
sgemv_t_microk_haswell-4.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
sgemv_t_microk_nehalem-4.c
…
sgemv_t_microk_sandy-4.c
…
sgemv_t_microk_skylakex_template.c
sgemv: skylakex: fix build warning
2021-08-25 07:13:00 +00:00
sgemv_t_microk_skylakex.c
Optimized sgemv_t for small N based on AVX512
2021-06-08 15:08:28 -04:00
sgemv_t.c
…
sgemv_t.S
…
sger_microk_sandy-2.c
…
sger.c
…
srot_microk_haswell-2.c
Remove spurious AVX512 requirement and add AVX2/FMA3 guard
2021-03-06 14:35:49 +01:00
srot_microk_skylakex-2.c
Improve the performance of rot by using AVX512 and AVX2 intrinsic
2020-11-05 15:12:36 +08:00
srot.c
Use Haswell optimizations for Zen as well
2021-02-11 09:24:51 +01:00
ssymv_L_microk_bulldozer-2.c
…
ssymv_L_microk_haswell-2.c
…
ssymv_L_microk_nehalem-2.c
…
ssymv_L_microk_sandy-2.c
…
ssymv_L.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
ssymv_U_microk_bulldozer-2.c
…
ssymv_U_microk_haswell-2.c
…
ssymv_U_microk_nehalem-2.c
…
ssymv_U_microk_sandy-2.c
…
ssymv_U.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
staticbuffer.S
…
stobf16_microk_cooperlake.c
Add bfloat16 based dot and conversion with single/double
2020-09-04 02:31:25 +08:00
strsm_kernel_8x4_haswell_L_common.h
Strip UTF8 byte order marker from source
2020-06-26 09:00:43 +02:00
strsm_kernel_8x4_haswell_LN.c
Strip UTF8 byte order marker from source
2020-06-26 09:00:43 +02:00
strsm_kernel_8x4_haswell_LT.c
AVX2 STRSM kernel
2020-03-17 00:34:08 +08:00
strsm_kernel_8x4_haswell_R_common.h
AVX2 STRSM kernel
2020-03-17 00:34:08 +08:00
strsm_kernel_8x4_haswell_RN.c
AVX2 STRSM kernel
2020-03-17 00:34:08 +08:00
strsm_kernel_8x4_haswell_RT.c
AVX2 STRSM kernel
2020-03-17 00:34:08 +08:00
strsm_kernel_LN_bulldozer.c
…
strsm_kernel_LT_bulldozer.c
…
strsm_kernel_RN_bulldozer.c
…
strsm_kernel_RT_bulldozer.c
…
sum.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
swap_sse2.S
…
swap_sse.S
…
swap.S
…
symv_L_sse2.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
symv_L_sse.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
symv_U_sse2.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
symv_U_sse.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
tobf16.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
trsm_kernel_LN_2x8_nehalem.S
…
trsm_kernel_LN_4x2_atom.S
…
trsm_kernel_LN_4x4_barcelona.S
…
trsm_kernel_LN_4x4_core2.S
…
trsm_kernel_LN_4x4_penryn.S
…
trsm_kernel_LN_4x4_sse2.S
…
trsm_kernel_LN_4x4_sse3.S
…
trsm_kernel_LN_4x8_nehalem.S
…
trsm_kernel_LN_8x4_sse.S
…
trsm_kernel_LT_2x8_nehalem.S
…
trsm_kernel_LT_4x2_atom.S
…
trsm_kernel_LT_4x4_barcelona.S
…
trsm_kernel_LT_4x4_core2.S
…
trsm_kernel_LT_4x4_penryn.S
…
trsm_kernel_LT_4x4_sse2.S
…
trsm_kernel_LT_4x4_sse3.S
…
trsm_kernel_LT_4x8_nehalem.S
…
trsm_kernel_LT_8x4_sse.S
…
trsm_kernel_RT_2x8_nehalem.S
…
trsm_kernel_RT_4x2_atom.S
…
trsm_kernel_RT_4x4_barcelona.S
…
trsm_kernel_RT_4x4_core2.S
…
trsm_kernel_RT_4x4_penryn.S
…
trsm_kernel_RT_4x4_sse2.S
…
trsm_kernel_RT_4x4_sse3.S
…
trsm_kernel_RT_4x8_nehalem.S
…
trsm_kernel_RT_8x4_sse.S
…
xdot.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
xgemm3m_kernel_2x2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
xgemm_kernel_1x1.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
xgemv_n.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
xgemv_t.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
xtrsm_kernel_LT_1x1.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zamax_atom.S
…
zamax_sse2.S
…
zamax_sse.S
…
zamax.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zasum_atom.S
…
zasum_microk_skylakex-2.c
Initialize abs_mask1 with itself to silence a gcc warning
2021-09-15 22:11:35 +02:00
zasum_sse2.S
…
zasum_sse.S
…
zasum.c
fix error declare function blas_level1_thread_with_return_value
2020-12-02 09:51:52 +08:00
zasum.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zaxpy_atom.S
…
zaxpy_microk_bulldozer-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
zaxpy_microk_haswell-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
zaxpy_microk_sandy-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
zaxpy_microk_steamroller-2.c
x86_64: clobber all xmm registers after vzeroupper
2020-10-20 02:16:47 +00:00
zaxpy_sse2.S
…
zaxpy_sse.S
…
zaxpy.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zaxpy.S
…
zcopy_sse2.S
…
zcopy_sse.S
…
zcopy.S
…
zdot_atom.S
…
zdot_microk_bulldozer-2.c
…
zdot_microk_haswell-2.c
…
zdot_microk_sandy-2.c
…
zdot_microk_steamroller-2.c
…
zdot_sse2.S
…
zdot_sse.S
…
zdot.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zdot.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zgemm3m_kernel_2x8_nehalem.S
…
zgemm3m_kernel_4x2_atom.S
…
zgemm3m_kernel_4x4_barcelona.S
…
zgemm3m_kernel_4x4_core2.S
…
zgemm3m_kernel_4x4_haswell.c
Update zgemm3m_kernel_4x4_haswell.c
2019-12-30 17:33:42 +08:00
zgemm3m_kernel_4x4_penryn.S
…
zgemm3m_kernel_4x4_sse2.S
…
zgemm3m_kernel_4x4_sse3.S
…
zgemm3m_kernel_4x8_nehalem.S
…
zgemm3m_kernel_8x4_barcelona.S
…
zgemm3m_kernel_8x4_core2.S
…
zgemm3m_kernel_8x4_penryn.S
…
zgemm3m_kernel_8x4_sse3.S
…
zgemm3m_kernel_8x4_sse.S
…
zgemm_beta.S
…
zgemm_kernel_1x4_nehalem.S
…
zgemm_kernel_2x1_atom.S
…
zgemm_kernel_2x2_barcelona.S
…
zgemm_kernel_2x2_bulldozer.S
…
zgemm_kernel_2x2_core2.S
…
zgemm_kernel_2x2_penryn.S
…
zgemm_kernel_2x2_piledriver.S
…
zgemm_kernel_2x2_sse2.S
…
zgemm_kernel_2x2_sse3.S
…
zgemm_kernel_2x4_nehalem.S
…
zgemm_kernel_4x2_barcelona.S
…
zgemm_kernel_4x2_core2.S
…
zgemm_kernel_4x2_haswell.c
Update zgemm_kernel_4x2_haswell.c
2020-02-27 22:25:19 +08:00
zgemm_kernel_4x2_haswell.S
…
zgemm_kernel_4x2_penryn.S
…
zgemm_kernel_4x2_skylakex.c
AVX512 CGEMM & ZGEMM kernels
2019-11-11 20:04:52 +08:00
zgemm_kernel_4x2_sse3.S
…
zgemm_kernel_4x2_sse.S
…
zgemm_kernel_4x4_sandy.S
…
zgemm_ncopy_1.S
…
zgemm_ncopy_2.S
…
zgemm_tcopy_1.S
…
zgemm_tcopy_2.S
…
zgemv_n_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zgemv_n_atom.S
…
zgemv_n_dup.S
…
zgemv_n_microk_bulldozer-4.c
…
zgemv_n_microk_haswell-4.c
…
zgemv_n_microk_sandy-4.c
…
zgemv_n.S
…
zgemv_t_4.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zgemv_t_atom.S
…
zgemv_t_dup.S
…
zgemv_t_microk_bulldozer-4.c
…
zgemv_t_microk_haswell-4.c
…
zgemv_t.S
…
znrm2_sse.S
…
znrm2.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zrot_sse2.S
…
zrot_sse.S
…
zrot.S
…
zscal_atom.S
…
zscal_microk_bulldozer-2.c
…
zscal_microk_haswell-2.c
…
zscal_microk_steamroller-2.c
…
zscal_sse2.S
…
zscal_sse.S
…
zscal.c
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zscal.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zsum.S
use emms instead, add WIN guards
2020-10-12 18:15:01 +03:00
zswap_sse2.S
…
zswap_sse.S
…
zswap.S
…
zsymv_L_sse2.S
Fix unsafe read during final iteration of zsymv_L_sse2.S
2021-11-19 14:29:32 -06:00
zsymv_L_sse.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zsymv_U_sse2.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
zsymv_U_sse.S
initial support for Sapphire Rapids platform
2021-10-12 01:30:40 -07:00
ztrsm_kernel_LN_2x1_atom.S
…
ztrsm_kernel_LN_2x2_core2.S
…
ztrsm_kernel_LN_2x2_penryn.S
…
ztrsm_kernel_LN_2x2_sse2.S
…
ztrsm_kernel_LN_2x2_sse3.S
…
ztrsm_kernel_LN_2x4_nehalem.S
…
ztrsm_kernel_LN_4x2_sse.S
…
ztrsm_kernel_LN_bulldozer.c
…
ztrsm_kernel_LT_1x4_nehalem.S
…
ztrsm_kernel_LT_2x1_atom.S
…
ztrsm_kernel_LT_2x2_core2.S
…
ztrsm_kernel_LT_2x2_penryn.S
…
ztrsm_kernel_LT_2x2_sse2.S
…
ztrsm_kernel_LT_2x2_sse3.S
…
ztrsm_kernel_LT_2x4_nehalem.S
…
ztrsm_kernel_LT_4x2_sse.S
…
ztrsm_kernel_LT_bulldozer.c
…
ztrsm_kernel_RN_bulldozer.c
…
ztrsm_kernel_RT_1x4_nehalem.S
…
ztrsm_kernel_RT_2x2_core2.S
…
ztrsm_kernel_RT_2x2_penryn.S
…
ztrsm_kernel_RT_2x2_sse2.S
…
ztrsm_kernel_RT_2x2_sse3.S
…
ztrsm_kernel_RT_2x4_nehalem.S
…
ztrsm_kernel_RT_4x2_sse.S
…
ztrsm_kernel_RT_bulldozer.c
…