718efcec6
(refs/pull/2013/head)
Fix out-of-bounds memory access in gemm_beta by
2019-02-13 22:08:37 +0100
f9d67bb5e
Fix out-of-bounds memory access in gemm_beta by
2019-02-13 22:06:41 +0100
76bb74fcd
Merge pull request #2012 from maamountki/z14 by
2019-02-13 20:15:56 +0100
0a54c98b9
(refs/pull/2012/head)
[ZARCH] Modify constraints by
2019-02-13 21:06:25 +0200
bec54ae36
[ZARCH] Fix caxpy by
2019-02-13 12:54:35 +0200
63d7bad8a
Merge pull request #2010 from martin-frbg/issue2009 by
2019-02-12 23:24:02 +0100
ab1630f9f
(refs/pull/2010/head)
Fix declaration of arguments in inline assembly by
2019-02-12 16:14:02 +0100
b824fa70e
Fix declaration of assembly arguments in SSYMV and DSYMV microkernels by
2019-02-12 16:00:18 +0100
91481a3e4
Fix declaration of input arguments in inline assembly by
2019-02-12 15:51:43 +0100
dc6ac9eab
Fix declaration of input arguments in the x86_64 s/dGEMV_T and s/dGEMV_N kernels by
2019-02-12 15:33:48 +0100
f58367410
[ZARCH] Fix cgemv_t_4 by
2019-02-12 13:12:28 +0200
77fe70019
[ZARCH] Fix constraints and source code formatting by
2019-02-11 16:01:13 +0200
03a2bf260
Fix potential memory leak in cpu enumeration on Linux (#2008) by
2019-02-10 23:24:45 +0100
ad86d7ceb
(refs/pull/2008/head)
Disentangle pathways for cpu counts above and below 1024 by
2019-02-09 18:49:16 +0100
8b37c8a14
Fixup ifdefs for non-glibc linux by
2019-02-08 11:54:49 +0100
cd6425add
Fix declarations of cpuset(p) by
2019-02-08 09:17:23 +0100
aab2aaa36
Fix potential memory leak in cpu enumeration with glibc by
2019-02-07 23:12:43 +0100
8c58865aa
(refs/pull/2007/head)
Fix potential memory leak in cpu enumeration by
2019-02-07 22:39:03 +0100
69edc5bbe
Restore dropped patches in the non-TLS branch of memory.c (#2004) by
2019-02-07 20:06:13 +0100
1ba2b2e63
(refs/pull/2004/head)
Fix misedit that caused hang with OPENMP by
2019-02-07 14:18:19 +0100
c330ea1bc
Restore dropped patches in the non-TLS branch of memory.c by
2019-02-06 22:40:03 +0100
703977016
[ZARCH] Undo the last commit by
2019-02-06 20:11:44 +0200
641767f84
Merge pull request #2001 from martin-frbg/cmake-dynlist by
2019-02-06 08:39:24 +0100
af6e2253a
Merge pull request #2000 from martin-frbg/issue1989 by
2019-02-06 00:29:30 +0100
5952e586c
(refs/pull/2001/head)
Support DYNAMIC_LIST option in cmake by
2019-02-05 23:51:40 +0100
f10408aae
Merge pull request #1999 from martin-frbg/issue1996-2 by
2019-02-05 22:02:11 +0100
d70ae3ab4
(refs/pull/2000/head)
Make c_check robust against old or incomplete perl installations by
2019-02-05 20:06:34 +0100
1391fc46d
(refs/pull/1999/head)
fix second instance of complex.h for c++ as well by
2019-02-05 19:29:33 +0100
11a43e811
[ZARCH] Set alignment hint for vl/vst by
2019-02-05 19:17:08 +0200
817fe9865
Merge pull request #1998 from martin-frbg/issue1992 by
2019-02-05 17:39:59 +0100
f4b82d7bc
(refs/pull/1998/head)
Include complex rather than complex.h in C++ contexts by
2019-02-05 13:30:13 +0100
61526480f
[ZARCH] Fix copy constraint by
2019-02-05 07:51:19 +0200
81daf6bc3
[ZARCH] Format source code, Fix constraints by
2019-02-05 07:30:38 +0200
a38aa56e7
Merge pull request #1 from xianyi/develop by
2019-02-05 07:25:38 +0200
729e92517
Merge pull request #1996 from quickwritereader/develop by
2019-02-04 16:52:04 +0100
498ac9858
(refs/pull/1996/head)
Note for unused kernels by
2019-02-04 15:41:56 +0000
cd9ea4546
NBMAX=4096 for gemvn, added sgemvn 8x8 for future by
2019-02-04 06:57:11 +0000
d559e65c7
(refs/pull/1995/head)
FIXME by
2019-02-03 22:55:51 +0200
ebf443fe7
clean up trmv interface from scattered workaround by
2019-02-03 13:10:46 +0200
465133e7c
init by
2019-02-03 13:08:03 +0200
f9c5023e0
Merge pull request #1994 from quickwritereader/develop by
2019-02-01 21:04:47 +0100
4abc375a9
(refs/pull/1994/head)
sgemv cgemv pairs by
2019-02-01 13:45:00 +0000
874df6549
Fix incorrect sgemv results for IBM z14 by
2019-02-01 12:58:59 +0100
1f4b61f57
Delete misplaced file sgemv_t_4.c by
2019-02-01 12:57:01 +0100
282230c30
Merge pull request #1993 from martin-frbg/aarnes-zarch by
2019-01-31 21:27:00 +0100
cce574c3e
(refs/pull/1993/head)
Improve the z14 SGEMVT kernel by
2019-01-31 21:24:55 +0100
877023e1e
Fix precision of zarch DSDOT by
2019-01-31 21:22:26 +0100
265142edd
Fix typo in the zarch min/max kernels by
2019-01-31 21:21:40 +0100
885a3c435
USE_TRMM on Z14 by
2019-01-31 21:18:09 +0100
4b512f84d
Add cache sizes for Z14 by
2019-01-31 21:16:44 +0100
72d3e7c9b
Add FORCE Z14 by
2019-01-31 21:15:50 +0100
bdc73a49e
Add parameters for Z14 by
2019-01-31 21:14:37 +0100
1249ee1fd
Add Z14 target by
2019-01-31 21:13:46 +0100
42df9efa0
Merge pull request #1991 from maamountki/z14 by
2019-01-31 19:10:03 +0100
82124729a
(refs/pull/1991/head)
Merge branch 'develop' into z14 by
2019-01-31 19:36:41 +0200
29416cb5a
[ZARCH] Add Z13 version for max/min functions by
2019-01-31 19:11:11 +0200
48b9b94f7
[ZARCH] Improve loading performance for camax/icamax by
2019-01-31 18:52:11 +0200
86a824c97
Fix wrong comparison that made IMIN identical to IMAX by
2019-01-31 15:27:21 +0100
808410c2c
Fix wrong comparison that made IMIN identical to IMAX by
2019-01-31 15:25:15 +0100
eaf20f0e7
Remove ztest by
2019-01-31 09:26:50 +0200
fcd814a8d
[ZARCH] Fix bug in max/min functions by
2019-01-29 17:59:38 +0200
dc4d3bccd
[ZARCH] Fix icamax/icamin by
2019-01-29 03:47:49 +0200
c7143c101
[ZARCH] Fix iamax/imax single precision by
2019-01-28 17:52:23 +0200
04873bb17
[ZARCH] Undo the last commit by
2019-01-28 17:32:24 +0200
c8ef9fb22
[ZARCH] Fix bug in iamax/iamin/imax/imin by
2019-01-28 17:16:18 +0200
5be61f4b4
Merge pull request #1985 from martin-frbg/issue1984 by
2019-01-28 15:44:57 +0100
3d155cff8
Merge pull request #1981 from edisongustavo/develop by
2019-01-28 15:44:42 +0100
7d47f0a82
Merge pull request #1978 from danielgindi/feature/msvc_cmake by
2019-01-28 15:43:35 +0100
a529c71a7
Merge pull request #1962 from brada4/r by
2019-01-28 15:42:57 +0100
ea1716ce2
Update Makefile.rule by
2019-01-27 17:22:26 +0100
0f24b39eb
Reword/expand comments in Makefile.rule by
2019-01-27 15:33:00 +0100
89b60dab8
Merge pull request #1987 from martin-frbg/issue1961 by
2019-01-26 22:25:29 +0100
58dd7e450
(refs/pull/1987/head)
Change ARMV8 target to ARMV7 for BINARY=32 by
2019-01-26 17:52:33 +0100
36b844af8
Change ARMV8 target to ARMV7 when BINARY32 is set by
2019-01-26 17:47:22 +0100
e882b239a
(refs/pull/1985/head)
Correct naming of getrf_parallel object by
2019-01-26 00:45:45 +0100
3f7bb87a2
Merge pull request #1971 from martin-frbg/trsm-threshold by
2019-01-24 09:17:48 +0100
e908ac2a5
(refs/pull/1981/head)
Fix include directory of exported targets by
2019-01-23 15:09:13 +0100
8533aca96
(refs/pull/1971/head)
Avoid penalizing tall skinny matrices by
2019-01-23 10:03:00 +0100
16494cb7c
Merge pull request #1980 from martin-frbg/issue1979 by
2019-01-22 21:10:38 +0100
b56b34a75
(refs/pull/1978/head)
Syntax fix by
2019-01-22 18:55:43 +0100
21eda8b57
(refs/pull/1980/head)
Report SkylakeX as Haswell if compiler does not support AVX512 by
2019-01-22 18:47:12 +0100
24288803b
Adjust test script for correct deployment by
2019-01-22 14:38:01 +0200
f0d834b82
Use VERSION_LESS for comparisons involving software version numbers by
2019-01-22 12:32:24 +0100
63bbd7b0d
Better support for MSVC/Windows in CMake by
2019-01-21 08:35:23 +0200
b11182922
[ZARCH] Update max/min functions by
2019-01-21 15:56:04 +0200
e45e7c341
(refs/pull/1977/head)
Better support for MSVC/Windows in CMake by
2019-01-21 08:35:23 +0200
010d59bfe
Merge pull request #1973 from martin-frbg/issue1464 by
2019-01-20 20:30:11 +0100
7329cf107
upload some buildtests by
2019-01-20 14:58:23 +0100
83b5c6b92
Fix compilation with NO_AVX=1 set by
2019-01-20 12:18:53 +0100
bbfdd6c0f
(refs/pull/1973/head)
Increase Zen SWITCH_RATIO to 16 by
2019-01-19 23:01:31 +0100
cda81cfae
Shift transition to multithreading towards larger matrix sizes by
2019-01-19 00:10:01 +0100
32b0f1168
Fix declaration of input arguments in the Sandybridge GER microkernels (#1967) by
2019-01-18 08:11:39 +0100
b495e5431
Fix declaration of input arguments in the x86_64 SCAL microkernels (#1966) by
2019-01-18 08:11:07 +0100
d5e694025
Fix declaration of input arguments in the x86_64 microkernels for DOT and AXPY (#1965) by
2019-01-17 23:20:32 +0100
24e697ead
Merge pull request #1970 from quickwritereader/develop by
2019-01-17 16:42:11 +0100
3e9fd6359
Bump xcode version to 10.1 to make sure it handles AVX512 by
2019-01-17 16:19:03 +0100
43a457203
(refs/pull/1970/head)
crot fix by
2019-01-17 14:45:31 +0000
ccb2b2175
(refs/pull/1966/head)
Remove stray comma by
2019-01-17 09:48:40 +0100
c17d2f61c
Remove stray comma by
2019-01-17 09:47:12 +0100
3b0b5ce0f
Remove stray comma by
2019-01-17 09:46:05 +0100