757 Commits (44a5ce33ba2eda1f9d0d2a458791f2f0304a1876)

Author SHA1 Message Date
  Martin Kroeker 44a5ce33ba
move pragma 3 years ago
  Martin Kroeker ea74832602
move pragma 3 years ago
  Martin Kroeker 90b2d9fe3e
move pragma 3 years ago
  Martin Kroeker c7585781de
move pragma 3 years ago
  Martin Kroeker 7f0b14eaff
move pragma 3 years ago
  Martin Kroeker 49589a6304
move pragma 3 years ago
  Martin Kroeker 8c6261c1f7
move pragma 3 years ago
  Martin Kroeker 002df7cdfd
Add guards around the pragma 3 years ago
  Martin Kroeker 7c59443ea2
Add guards around the pragma 3 years ago
  Martin Kroeker 07ef20ea76
Add guards around the pragma 3 years ago
  Martin Kroeker 7a93103b47
Add guards around the pragma 3 years ago
  Martin Kroeker c4b0767785
Add guards around the pragma 3 years ago
  Martin Kroeker 746f6c0230
Add guards around the pragma 3 years ago
  Martin Kroeker b2b31a8024
Add guards around the pragma 3 years ago
  Martin Kroeker 83568f37e4
disable gcc tree vectorizer 3 years ago
  Martin Kroeker ed030a0b0e
disable gcc tree vectorizer 3 years ago
  Martin Kroeker ce4cdd52a3
disable gcc tree vectorizer 3 years ago
  Martin Kroeker 801fe19f21
disable gcc tree vectorizer 3 years ago
  Martin Kroeker 45b1dc7113
disable gcc tree vectorizer 3 years ago
  Martin Kroeker e8a94437d9
try pragma to disable tree vectorizer 3 years ago
  Martin Kroeker e5d0749420
try pragma to disable tree vectorizer 3 years ago
  Martin Kroeker dc49edd4e6
Revert "roll back DGEMM kernel ... for DYNAMIC_ARCH" 4 years ago
  Caroline Newcombe 5cc1111383 fix unsafe read of Y in assembly kernel 4 years ago
  Wangyang Guo 225683218c Small Matrix: use proper inline asm input constraint for AVX512 mask 4 years ago
  Martin Kroeker 9c626e466e
really fix definition of SHUFFLE_MAGIC_NO 4 years ago
  Martin Kroeker 9d7429406f
Declare SHUFFLE_MAGIC_NO as const to placate clang 4 years ago
  Martin Kroeker 522f809825
Merge pull request #3542 from martin-frbg/issue3540 4 years ago
  Mosè Giordano abbc947edb Fix compilation of Skylake AVX512 kernels with GCC 6 4 years ago
  Martin Kroeker c62f8e2c01
Prevent compiler attempts to use k0 as mask register 4 years ago
  Martin Kroeker 80eb581c83
Fix non-portable u_int64_t 4 years ago
  Martin Kroeker 73ffabe6ba
Guard uses of _mm512_reduce_add_p? 4 years ago
  Martin Kroeker 7b146e590c
fix function typecast 4 years ago
  Martin Kroeker e9a0e52201
fix function typecast 4 years ago
  Martin Kroeker d1ee6ff73f
fix function typecasts 4 years ago
  Martin Kroeker 5378046abd
roll back DGEMM kernels to 4x8 when compiling for DYNAMIC_ARCH 4 years ago
  Caroline Newcombe feeb8283a5 Fix unsafe read during final iteration of zsymv_L_sse2.S 4 years ago
  Wangyang Guo 63a103ba6e sbgemm: spr: disable small matrix path by default 4 years ago
  Wangyang Guo 82194ea9d2 sbgemm: spr: implement otcopy_16 4 years ago
  Wangyang Guo 8632380a96 sbgemm: spr: reuse ncopy_16 from cooperlake as incopy 4 years ago
  Wangyang Guo 6bc8204ce5 sbgemm: spr: optimization for tmp_c buffer 4 years ago
  Wangyang Guo f018aa342a sbgemm: spr: kernel handle alpha != 1.0 4 years ago
  Wangyang Guo a52456b168 sbgemm: spr: oncopy: use tile load/store instead 4 years ago
  Wangyang Guo f2485352a6 sbgemm: spr: only load A once in tail_k handling 4 years ago
  Wangyang Guo 9ab33228bb sbgemm: spr: process k2 and odd k at the same time 4 years ago
  Wangyang Guo 10d52646e2 sbgemm: spr: oncopy: avoid handling too much pointer at a time 4 years ago
  Wangyang Guo 88154ed02d sbgemm: spr: reduce tile conf loading by seperate tail k handling 4 years ago
  Wangyang Guo a70bfb52d5 sbgemm: spr: kernel works for NN case when alpha is 1.0 4 years ago
  Wangyang Guo 6051c86741 sbgemm: spr: kernel works for m32 in NN case 4 years ago
  Wangyang Guo d0b253ac6e sbgemm: spr: implement oncopy_16 4 years ago
  Wangyang Guo 1d48b7cb16 sbgemm: spr: add dummy source files 4 years ago