Wangyang Guo
45fdf951b6
sbgemm: cooperlake: reorder ptr increase for performance
4 years ago
Wangyang Guo
cece3541ab
sbgemm: cooperlake: fix bug in m64n12
4 years ago
Wangyang Guo
9df0953cde
sbgemm: cooperlake: kernel works for NN
4 years ago
Wangyang Guo
2ec9f3a8aa
sbgemm: cooperlake: change kernel size to 16x4
4 years ago
Wangyang Guo
ef8f5fecc8
sbgemm: cooperlake: implement sbgemm_tcopy_32
4 years ago
Wangyang Guo
4c294336e6
sbgemm: cooperlake: add dummy source files
4 years ago
Wangyang Guo
619588fbab
sbgemm: remove unnecessary b0 files
4 years ago
Wangyang Guo
f39301935c
sbgemm: cooperlake: make sure hot buffer aligned to 64
4 years ago
Wangyang Guo
7d27b182fc
sbgemm: cooperlake: enable SBGEMM by small matrix path
4 years ago
Martin Kroeker
bec9d9f63d
Merge pull request #3335 from guowangy/small-matrix-latest
Add GEMM optimization for small matrix and single/double kernel for skylakex
4 years ago
Wangyang Guo
dbbb39199f
sgemv: skylakex: fix build warning
4 years ago
Wangyang Guo
e9acb46431
sgemv: skylakex: bug fix for sgemv_t kernel in corner case
4 years ago
Wangyang Guo
f9dba63c28
Small Matrix: skylakex: remove unnecessary b0 source files
4 years ago
Wangyang Guo
44d0032f3b
Small Matrix: skylakex: fix build error in old compiler
4 years ago
Chen, Guobing
5d86becdae
Add all SBGEMM kernels for IA AVX512-BF16 based platforms
Added all SBGEMM kernels including NN/NT/TN/TT for both ColMajor and
RowMajor, based on AVX512-BF16 ISA set on IA.
Signed-off-by: Chen, Guobing <guobing.chen@intel.com>
4 years ago
Wangyang Guo
fa777f5517
Small Matrix: skylakex: add DGEMM_SMALL_M_PERMIT and tune for TN kernel
4 years ago
Wangyang Guo
8592c21af4
Small Matrix: skylakex: dgemm nn: fix typo in idx load
4 years ago
Wangyang Guo
3e79f6d89a
Small Matrix: skylakex: add dgemm tn kernel
4 years ago
Wangyang Guo
323d7da4f7
Small Matrix: skylakex: add dgemm tt kernel
4 years ago
Wangyang Guo
f57fc932ac
Small Matrix: skylakex: add dgemm nt kernel
4 years ago
Wangyang Guo
91ec21202b
Small Matrix: skylakex: add dgemm nn kernel
4 years ago
Wangyang Guo
72e070539c
Small Matrix: skylakex: add sgemm tt kernel
4 years ago
Wangyang Guo
02c6e764f2
Small Matrix: skylakex: add SGEMM_SMALL_M_PERMIT and tune for TN kernel
4 years ago
Wangyang Guo
642c393879
Small Matrix: skylakex: add sgemm tn kernel
4 years ago
Wangyang Guo
ae3f5c737c
Small Matrix: skylakex: sgemm nt: optimize for M < 12
4 years ago
Wangyang Guo
0d72d75bf9
Small Matrix: skylakex: add sgemm nt kernel
4 years ago
Wangyang Guo
ca7682e3a3
Small Matrix: skylakex: sgemm nn: fix n6 conflicts with n4
4 years ago
Wangyang Guo
9967e61abb
Small Matrix: skylakex: sgemm nn: fix error when beta not zero
4 years ago
Wangyang Guo
a87736346f
Small Matrix: skylakex: sgemm nn: add n6 to improve performance
4 years ago
Wangyang Guo
4c9d9940fd
Small Matrix: skylakex: sgemm nn: reduce store 4 N at a time
4 years ago
Wangyang Guo
13b32f69b7
Small Matrix: skylakex: sgemm nn: reduce store 4 M at a time
4 years ago
Wangyang Guo
3d8c6d9607
Small Matrix: skylakex: sgemm nn: clean up unused code
4 years ago
Wangyang Guo
49b61a3f30
Small Matrix: skylakex: sgemm_nn: optimize for M <= 8
4 years ago
Wangyang Guo
f88470323b
Optimize M < 16 using AVX512 mask
4 years ago
Wangyang Guo
9186456a12
small matrix: SkylakeX: add SGEMM NN kernel
4 years ago
Martin Kroeker
5b4b385ecf
Temporarily disable the SkylakeX sgemv_t microkernel due to LAPACK testsuite failures
4 years ago
Ma, Yu
706a08d4a0
Optimized sgemv_t for small N based on AVX512
4 years ago
Martin Kroeker
5f677e782e
Merge pull request #3196 from guowangy/skylakex-gemm-batch-k
GEMM: skylake: improve the performance when m is small
4 years ago
Martin Kroeker
02087a62e7
Merge pull request #3205 from intelmy/sgemv_n_opt
optimize on sgemv_n for small n
4 years ago
Martin Kroeker
8b90e5f202
Drop redundant inclusion of complex.h
4 years ago
Martin Kroeker
c0ca63ea46
Fix missing conditionals for non-SKX kernels
4 years ago
pnp
3d4ccd2a13
fix for build error
4 years ago
pnp
c59652f0ce
optimize on sgemv_n for small n
4 years ago
Wangyang Guo
aa7b3dc3db
GEMM: skylake: improve the performance when m is small
4 years ago
Martin Kroeker
3d511f0e66
replace spurious avx512 requirement with fma check
4 years ago
Martin Kroeker
2dfb24730d
Use "old" compute(24) function with clang due to register limitations
4 years ago
Martin Kroeker
7b8f580941
Merge pull request #3156 from martin-frbg/omatcopy_d
Move x86_64 DOMATCOPY_RT back to the C implementation
4 years ago
Martin Kroeker
0f5e86a0d9
Remove premature entry for DOMATCOPY_RT
4 years ago
Martin Kroeker
7b294a99fd
Move common.h back to the top of the file so that SKYLAKEX (from config.h) is defined in time
4 years ago
Martin Kroeker
0934568d9c
Move includes under the ifdef for compilers w/o intrinsics support
4 years ago