floraachy/OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
2,644 Commits 47 Branches 64 Tags
4054d32def49c8a764ac7bfd1e8cf0ea8e1937e6
Commit Graph

2644 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
martin
4054d32def update cmake files 2017-11-24 08:15:40 +01:00
martin
2d52f0f4c3 update cmakefiles for lapack 3.8.0 2017-11-23 21:22:01 +01:00
martin
3be5c3d343 Update LAPACK to 3.8.0 2017-11-23 18:13:35 +01:00
Martin Kroeker
b18730f9e1 Merge pull request #1365 from xoviat/patch-1 
[appveyor] use cmake from conda forge
 2017-11-22 21:13:41 +01:00
xoviat
be9e63c021 [appveyor] use cmake from conda forge 2017-11-21 18:44:02 -06:00
Martin Kroeker
21844238ca Merge pull request #1364 from martin-frbg/shmem-init 
Handle shmem init failures in cpu affinity setup code
 2017-11-19 12:50:16 +01:00
Martin Kroeker
07e7c36dac Handle shmem init failures in cpu affinity setup code 
Failures to obtain or attach shared memory segments would lead to an exit without explanation of the exact cause.
This change introduces a more verbose error message and tries to make the code continue without setting cpu affinity.
Fixes #1351
 2017-11-18 23:57:44 +01:00
Martin Kroeker
9251a2efde Merge pull request #1359 from brada4/develop 
Eliminate mode variable where not needed in syrk interface
 2017-11-18 23:47:17 +01:00
Martin Kroeker
06adbf2e41 Merge pull request #1347 from martin-frbg/issue1322 
Change CBLAS complex functions to take void pointers
 2017-11-18 23:46:57 +01:00
Martin Kroeker
b46e2b57cc Make return parameter of cblas_Xdotc_sub, cblas_Xdotu_sub a void pointer as well 2017-11-18 20:28:02 +01:00
Martin Kroeker
3ce401f51b Make last parameter of cblas_Xdotc_sub/cblas_Xdotu_sub a void pointer as well 2017-11-18 18:58:40 +01:00
Martin Kroeker
1c9f4422b5 Fix declaration of cblas_Xdotc_sub and cblas_Xdotu_sub 
last parameter of cblas_(c,z)dotc_sub and cblas_(c,z)dotu_sub should be void* as well
 2017-11-18 18:56:30 +01:00
Andrew
27575d200a Eliminate mode variable where not needed 2017-11-15 15:32:38 +01:00
Martin Kroeker
6157d0902a Merge pull request #1358 from martin-frbg/unused_vars 
Clean up spurious unused variables in the kernels
 2017-11-15 11:31:43 +01:00
Martin Kroeker
3fea849bbf Remove unused variables from Haswell dtrmm and Bulldozer dtrsm 2017-11-14 23:35:10 +01:00
Martin Kroeker
8f177621bc Remove unused variables at0...at3 from ?symv_U 2017-11-14 23:32:25 +01:00
Martin Kroeker
5f402b7759 Remove unused (loop?) variable j from the gemv_n_4 implementations 2017-11-14 23:29:42 +01:00
Martin Kroeker
65bf0a343c Remove unused variable btpr 2017-11-14 23:25:50 +01:00
Martin Kroeker
acf3d34bc5 Silence an unused variable warning with a cast 
l2 cache size is not universally needed to assign default unrolling limits, but neither putting its declaration inside an ifdef nor cloning it into all ifdef sections that need it really makes sense here.
 2017-11-14 23:23:44 +01:00
Martin Kroeker
8e75f7dcb4 Merge pull request #1353 from xoviat/patch-1 
[appveyor] use flang from conda-forge
 2017-11-10 22:16:31 +01:00
Martin Kroeker
bd3546704c Merge pull request #1356 from martin-frbg/lapack-issue196 
Break out of potentially infinite rescaling loop after 1000 iterations
 2017-11-10 22:15:27 +01:00
Martin Kroeker
2df1e3372d Break out of potentially infinite rescaling loop after 1000 iterations 
Inf values in the input vector will survive rescaling, causing an infinite loop. The value of 1000 is arbitrarily chosen as a large but finite value with the intention to never interfere with regular calculations.
 2017-11-10 20:02:21 +01:00
Martin Kroeker
4271b2b158 Merge pull request #1354 from martin-frbg/shmem 
Try to handle shmget or shmat failing
 2017-11-10 09:11:03 +01:00
Martin Kroeker
148493df89 Merge branch 'develop' into shmem 2017-11-09 23:25:15 +01:00
Martin Kroeker
415555a9c1 Merge branch 'develop' into shmem 2017-11-09 23:20:54 +01:00
Martin Kroeker
2a6fef9a55 Try to handle shmget or shmat failing 
also replaces one verbatim sched_yield with the YIELDING macro for consistency as suggested in #1351
 2017-11-09 23:16:13 +01:00
xoviat
307305aeb5 [appeyor] use flang from conda-forge 
This flang will be updated in the future. We leave cmake because it's
not yet released with fortran support
 2017-11-09 15:10:02 -06:00
Martin Kroeker
cc26cdce0c Merge pull request #1352 from martin-frbg/issue1351 
Output an error message when shmat() fails
 2017-11-09 21:08:16 +01:00
Martin Kroeker
d8576826c4 Output an error message when shmat() fails 
Observed in #1351 with SELinux as the likely culprit. Without the message, the user saw a segfault with no apparent reason
 2017-11-09 17:31:44 +01:00
Martin Kroeker
c6968edec4 Merge pull request #1350 from insertinterestingnamehere/flang 
WIP: Support for Flang on Windows
 2017-11-08 11:40:13 +01:00
Isuru Fernando
9268314290 Fix gensymbol script 2017-11-06 21:12:38 -06:00
Ian Henriksen
3ace0fda3f Merge pull request #1 from xoviat/patch-1 
[appveyor] fixes
 2017-11-06 15:17:24 -06:00
xoviat
3cfc64404a [appveyor] fixes 2017-11-06 15:05:20 -06:00
Ian Henriksen
72956e8950 Build MATGEN LAPACK routines by default when building with CMake. 2017-11-06 14:47:27 -06:00
Ian Henriksen
505dc08635 Update lapacke.cmake with routines added in LAPACK 3.7.0. 2017-11-06 14:43:33 -06:00
Ian Henriksen
61587b0670 Update lapack.cmake with additional routines from LAPACK version 3.7.0. 2017-11-06 14:41:02 -06:00
Ian Henriksen
632fc75d77 Allow using compilers other than gfortran in conjunction with 
MSVC or clang-cl.
 2017-11-06 14:39:12 -06:00
Martin Kroeker
2c222f1faa Modify complex CBLAS functions to take void pointers 
Modify complex CBLAS functions to take void pointers instead of float or double arguments (to bring the prototypes in line with netlib and other implementations' cblas.h)
 2017-11-05 15:53:14 +01:00
Martin Kroeker
66ac898f64 Change prototypes of all complex functions to use void* 
Change prototypes of complex functions to use void pointers like the other implementations of CBLAS
 2017-11-05 15:42:33 +01:00
Martin Kroeker
ab87ee6b48 Merge pull request #1329 from martin-frbg/dsdot 
(Trivial) optimized dsdot implementation for HASWELL
 2017-10-25 19:13:38 +02:00
Martin Kroeker
a07807caac Eliminate loop code when called as/from dsdot 2017-10-25 16:45:41 +02:00
Martin Kroeker
b71f4fe681 Merge pull request #1334 from ashwinyes/develop_aarch64_20171024_addlocallabels 
ARM64: Convert all labels to local labels
 2017-10-24 19:50:03 +02:00
Ashwin Sekhar T K
a0128aa489 ARM64: Convert all labels to local labels 
While debugging/profiling applications using perf or other tools, the
kernels appear scattered in the profile reports. This is because the labels
within the kernels are not local and each label is shown as a separate
function.

To avoid this, all the labels within the kernels are changed to local
labels.
 2017-10-24 11:40:05 +00:00
Martin Kroeker
627133f9ad Merge pull request #1333 from martin-frbg/haswell32 
Fix 32bit HASWELL builds
 2017-10-24 11:25:03 +02:00
Martin Kroeker
0e2cf102e1 Fix 32bit HASWELL 2017-10-24 10:07:44 +02:00
Martin Kroeker
5e3e91d0fc Split the microkernel workload into chunks of 32 floats for dsdot mode to limit loss of precision 2017-10-22 18:18:51 +02:00
Martin Kroeker
28c3fa8950 Add dsdot 2017-10-16 23:29:03 +02:00
Martin Kroeker
8ac87c1cb6 Implement DSDOT with unchanged sdot microkernels 2017-10-16 23:27:51 +02:00
Martin Kroeker
b7cee00455 Merge pull request #1327 from martin-frbg/cmake-relapack 
Make ReLAPACK available in cmake builds
 2017-10-12 20:26:35 +02:00
Martin Kroeker
962b20a9bb Optionally add ReLAPACK to LIB_COMPONENTS 2017-10-12 17:02:01 +02:00