a4c96eca6
(refs/pull/543/head)
Fix a buffer overflow with MAX_STACK_ALLOC size in dgemv_t by
2015-04-15 09:41:45 +0200
fb02cb0a4
Merge pull request #540 from wernsaar/develop by
2015-04-14 15:53:09 +0200
baa0363ea
(refs/pull/540/head)
add optimized ddot-kernel for piledriver by
2015-04-14 15:09:13 +0200
34ba66606
add optimized daxpy-kernel for piledriver by
2015-04-14 14:23:29 +0200
f615dc760
added optimized saxpy kernel for steamroller by
2015-04-14 09:09:39 +0200
331c41763
optimized saxpy for piledriver by
2015-04-14 08:34:11 +0200
6c3a0b5d4
Enable MAX_STACK_ALLOC by default. by
2015-04-13 23:23:40 -0500
fd9fd4293
Refs #478, #482. Fixed bug on previous commit. by
2015-04-13 23:22:27 -0500
979848197
Refs #478, #482. Fix segfault bug for gemv_t with MAX_ALLOC_STACK flag. by
2015-04-13 19:45:27 -0500
d7a17ad85
optimized sdot-kernel for pilediver by
2015-04-13 13:19:21 +0200
d35f6c63c
add optimized daxpy-kernel for steamroller by
2015-04-13 12:22:43 +0200
166d76e86
added optimized sdot-kernel for steamroller by
2015-04-11 08:48:18 +0200
f9f127d83
added optimized ddot kernel for steamroller by
2015-04-10 16:18:03 +0200
62231ab33
Merge pull request #538 from wernsaar/develop by
2015-04-10 16:03:37 +0200
3119def9a
(refs/pull/538/head)
updated cdot and zdot by
2015-04-10 11:10:31 +0200
33b332372
add optimized cdot- and zdot-kernel for sandybridge by
2015-04-10 09:37:26 +0200
fd838c75b
add optimized cdot- and zdot-kernel for haswell by
2015-04-09 15:13:52 +0200
b57a60dac
updated cdot and zdot for piledriver by
2015-04-09 10:33:46 +0200
5c5116397
added optimized cdot- and zdot-kernel for steamroller by
2015-04-09 09:45:23 +0200
9299d8cfd
added optimized cdot- and zdot-kernels for bulldozer by
2015-04-08 16:29:55 +0200
0a3d3b945
Refs #535. Fix the wrong vector instruction in sgemm sandy bridge kernel. by
2015-04-08 03:55:49 +0800
4f680a7d6
Merge pull request #534 from wernsaar/develop by
2015-04-07 12:48:11 -0500
ba926e807
(refs/pull/534/head)
added cdot- and zdot benchmark by
2015-04-07 11:56:06 +0200
60c6dec6e
updated some lines for bulldozer by
2015-04-06 18:47:16 +0200
47898cca3
added optimized saxpy- and daxpy-kernel for sandybridge by
2015-04-06 16:05:16 +0200
53bb92428
added optimized saxpy- and daxpy-kernel for haswell by
2015-04-06 12:33:16 +0200
1e80b8b0d
Merge pull request #531 from wernsaar/develop by
2015-04-05 16:42:39 -0500
a901b065d
(refs/pull/531/head)
added optimized ddot-kernel for sandybridge by
2015-04-05 20:19:38 +0200
3937e2a0a
add optimized sdot-kernel for sandybridge by
2015-04-05 19:47:05 +0200
9707d608d
removed double definition line by
2015-04-05 18:35:34 +0200
701b9d755
added optimized sdot- and ddot-kernel for HASWELL by
2015-04-05 17:57:53 +0200
8977b3f23
Refs #529. Support Intel Broadwell by Haswell kernels. by
2015-04-02 11:08:03 -0500
f6426395e
Merge pull request #527 from xantares/patch-1 by
2015-03-30 10:16:11 -0500
0ac787eef
(refs/pull/527/head)
fix mingw install by
2015-03-30 09:30:55 +0200
e5b96e55a
Fix build bug for ARM64. by
2015-03-24 15:27:17 -0500
d0c51c4de
(tag: v0.2.14)
Merge branch 'develop' by
2015-03-24 15:07:07 -0500
a3491e1e8
Update the doc for 0.2.14. by
2015-03-24 15:05:59 -0500
e81a5d61e
Merge branch 'develop' of github.com:xianyi/OpenBLAS into develop by
2015-03-24 12:17:12 -0500
c674fa32b
Add ARM targets. by
2015-03-24 12:17:04 -0500
e34911a73
Fix compiling bug for ARM with setting BINARY. by
2015-03-24 17:15:33 +0000
76dcaf228
Merge pull request #521 from maxlevesque/patch-1 by
2015-03-21 12:26:35 -0500
770fac92e
(refs/pull/521/head)
Correct typo /proc/ instead of /pros/ by
2015-03-20 23:25:11 +0100
e95d64333
Refs #519. Avoid calling strncpy. by
2015-03-19 15:57:22 -0500
75c40bcc4
Refs #520. Fixed ONLY_CBLAS=1 compiling bug on OSX. by
2015-03-19 11:51:36 -0500
b62f9f412
Merge pull request #518 from ton/issue-508 by
2015-03-18 13:00:07 -0500
b6438dede
(refs/pull/518/head)
Fix issue #508 by
2015-03-18 13:22:43 +0100
1d183dcda
Added lapacke sources. by
2015-02-25 16:51:08 -0600
cdefdb21c
Refs #492. Fixed c/zsyr bug with negative incx. by
2015-02-26 06:37:03 +0800
e19bf3a28
Removed MSVC cpuid func when using clang. by
2015-02-25 14:44:49 -0600
3649cfbd7
Fixed EPILOGUE for clang. by
2015-02-25 12:23:26 -0600
5ae899375
Added intrinsics for MSVC. by
2015-02-25 11:52:51 -0600
84d90d6ed
Fixed some compiler errors/warnings for clang. by
2015-02-25 11:52:25 -0600
518e2424a
Fixed bad filename for cpuid.S compile. by
2015-02-25 11:51:29 -0600
ea7f9dacf
Refs #509. Fixed geadd building bug with DYNAMIC_ARCH=1. by
2015-02-26 01:47:11 +0800
bf5dbb7e2
Refs#509. Merge branch 'grisuthedragon-develop' into develop by
2015-02-26 01:44:19 +0800
00e373aea
Added LAPACK sources directly to add_library call instead of OBJECT. by
2015-02-25 10:18:18 -0600
9eaea02f3
Added additional gemm defines for complex types. by
2015-02-25 09:39:11 -0600
ab7043373
Fixed bug generating trmv complex source names. by
2015-02-24 15:18:16 -0600
504cdb10e
Added check for MSVC before enabling fortran. by
2015-02-24 14:31:45 -0600
a8002b0c5
Separated getarch ASM file when using MSVC. by
2015-02-24 14:31:18 -0600
0553476fb
Added TRANS defines for complex sources in lapack. by
2015-02-24 14:30:35 -0600
2416d9dba
Fixed TRANSA defines for complex sources in driver/level2. by
2015-02-24 13:18:07 -0600
0d8e227ea
Changed strategy for setting preprocessor definitions. by
2015-02-24 12:26:33 -0600
12d1fb2e4
Fixed incorrect object name in kernel CMakeLists.txt by
2015-02-24 10:30:16 -0600
1b7f42740
Added conj gemv objects for complex build. by
2015-02-23 10:24:31 -0600
b2284647a
More complex objects. by
2015-02-23 07:51:05 -0600
a6116e585
Added some more complex-only objects. by
2015-02-22 17:49:28 -0600
fb5d5bb97
Added defines for complex trmv. by
2015-02-21 12:39:03 -0600
371071d46
Added CONJ defines for trmm/trsm. by
2015-02-21 10:59:02 -0600
8a143516e
Added alternate_name to a couple of the name mangling schemes. by
2015-02-20 17:03:33 -0600
e5897ecb9
Added zherk_kernel.c objects to driver/level3. by
2015-02-19 16:19:56 -0600
714638c18
Added some TRMM objects for complex types. by
2015-02-19 16:11:51 -0600
e27c372e5
Fixed reuse of float_char from parent loop. by
2015-02-19 13:53:29 -0600
f3f2b3d76
Added complex and single netlib-lapack fortran sources to lapack.cmake. by
2015-02-19 12:26:11 -0600
949229804
Added other float types to Makefile.L3. by
2015-02-18 13:01:05 -0600
43725b82c
ParseMakefileVars now replaces Makefile vars with CMake vars. by
2015-02-18 12:23:17 -0600
14fd3d35d
Added checks for missing defines in kernel. by
2015-02-18 10:25:01 -0600
cebc07ceb
ParseMakefileVars now recursively parses included makefiles. by
2015-02-17 22:09:41 -0600
33c5e8db7
Added a helper function for setting the L1 kernel defaults. by
2015-02-17 21:36:23 -0600
67e39bd8f
Added mangled complex filenames to interface and lapack CMakeLists.txt. by
2015-02-17 13:12:30 -0600
9eb149909
Added another param to GenerateNamedObjects to mangle complex source names. by
2015-02-17 10:30:28 -0600
39cc6b21d
(refs/pull/509/head)
Add ATLAS-style ?geadd function by
2015-02-16 13:46:20 +0100
4662a0b13
Changed generate functions to iterate through a list of float types. by
2015-02-15 17:44:37 -0600
e74462a3f
Moved declarations to start of functions to satisfy MSVC C89 implementation. by
2015-02-11 11:16:57 -0600
056ba2675
Changed a number of inline calls to use __inline. by
2015-02-11 11:13:17 -0600
a0d9a7fd8
Changed _Complex types in common_level1.h to use the typedef. by
2015-02-11 11:11:47 -0600
5d3fc092e
Added MSVC defines to common.h. by
2015-02-11 11:10:45 -0600
c94fe7127
Removed incoming-stack-boundary for MSVC. by
2015-02-11 10:54:14 -0600
d60b49e5c
Turned off uninizialized variable warning when compiling lapack-netlib. by
2015-02-10 14:36:43 -0600
64b5a0ef8
Added AUX files from lapack-netlib. by
2015-02-10 14:29:05 -0600
162791e30
Added common objects from kernel Makefile. by
2015-02-10 12:42:05 -0600
8743093bd
Added aux files from lapack-netlib. by
2015-02-10 11:47:46 -0600
96cf6779c
Added DLA sources from lapack-netlib. by
2015-02-10 11:01:01 -0600
3b20b6242
Fixed trti2 name. by
2015-02-09 15:29:28 -0600
6ddbfea70
Added generic laswp object. by
2015-02-09 15:15:58 -0600
c0624a26b
Fixed some dgemm_copy function names. by
2015-02-09 14:34:29 -0600
4bfaf1ce6
Removed some list appends I missed. by
2015-02-09 12:56:55 -0600
e8c39138c
Removed return value from GenerateNamedObjects. by
2015-02-09 12:28:09 -0600
f99279922
Added the rest of Makefile.L3. by
2015-02-09 10:47:35 -0600
4c65afcce
Changed kernel filenames to vars. These will need to be read from KERNEL. by
2015-02-09 09:52:14 -0600