platform/upstream/openblas.git
21 months agofix build error 51/287951/1 accepted/tizen_6.5_unified tizen_6.5 accepted/tizen/6.5/unified/20230417.052928 accepted/tizen/6.5/unified/20230419.041406 accepted/tizen/6.5/unified/20230423.211904
Inki Dae [Mon, 25 Apr 2022 05:20:42 +0000 (14:20 +0900)]
fix build error

Fixed below build error,
[   13s] getarch_2nd.c: In function 'main':
[   13s] getarch_2nd.c:12:35: error: 'SGEMM_DEFAULT_UNROLL_M' undeclared (first use in this function); did you mean 'XGEMM_DEFAULT_UNROLL_M'?
[   13s]    12 |     printf("SGEMM_UNROLL_M=%d\n", SGEMM_DEFAULT_UNROLL_M);
[   13s]       |                                   ^~~~~~~~~~~~~~~~~~~~~~
[   13s]       |                                   XGEMM_DEFAULT_UNROLL_M

Reported-by : hyunggi.lee@samsung.com
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Change-Id: Idba959362d6e8e4078ebfca2d3844ca03878e0b1

4 years agopackaging: correct openblas library version accepted/tizen_6.0_unified_hotfix tizen_6.0_hotfix accepted/tizen/6.0/unified/20201030.111239 accepted/tizen/6.0/unified/hotfix/20201102.231835 accepted/tizen/6.5/unified/20211028.224325 accepted/tizen/unified/20200117.072026 submit/tizen/20200117.065412 submit/tizen_6.0/20201029.205503 submit/tizen_6.0_hotfix/20201102.192903 submit/tizen_6.0_hotfix/20201103.115103 submit/tizen_6.5/20211028.163601 tizen_6.0.m2_release tizen_6.5.m2_release
Inki Dae [Fri, 17 Jan 2020 06:25:10 +0000 (15:25 +0900)]
packaging: correct openblas library version

Change-Id: Ibc5b8384048d39f3ea2e475eb974d26501d29ae0
Signed-off-by: Inki Dae <inki.dae@samsung.com>
4 years agoRevert "Fix build error due to toolchain upgrade(gcc6->gcc9)"
Inki Dae [Fri, 17 Jan 2020 06:24:52 +0000 (15:24 +0900)]
Revert "Fix build error due to toolchain upgrade(gcc6->gcc9)"

This reverts commit 15c0f76ff281e84bedd186fcb7f10739e0ff175b.

Change-Id: Ied000e23c9ac19bf508310925bc91c422d9ffdc8
Signed-off-by: Inki Dae <inki.dae@samsung.com>
4 years agoFix build error due to toolchain upgrade(gcc6->gcc9)
Tae-Young Chung [Tue, 14 Jan 2020 03:51:06 +0000 (12:51 +0900)]
Fix build error due to toolchain upgrade(gcc6->gcc9)

Fix SIGSEGV; openblas failed to single, double, complex precision tests with
"Segmentation fault - invalid memory reference" log message while building with x86_64.

Change-Id: I4d3ce9608c34e81236c53057b2ce6efc92f6685a
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
4 years agoApply DEP to libopenblas_openmp
Tae-Young Chung [Fri, 17 May 2019 02:23:39 +0000 (11:23 +0900)]
Apply DEP to libopenblas_openmp

Change-Id: Iff5ce567ab5c7ed3604f6b3bc58a16b82572436a
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
4 years agoUpdate License
Tae-Young Chung [Wed, 12 Sep 2018 04:39:44 +0000 (13:39 +0900)]
Update License

Apache2.0 is detected from utest/ctest.h
MIT is detected from relapack/LICENSE
In case of multiple licenses,
all licenses should be notified and
the representative license must be the first(BSD-3-Clause).

Change-Id: Ic68f9d865d41672950edaa3fa6e0d2876ea8137b
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
4 years agoFix ASAN build error
Tae-Young Chung [Fri, 18 May 2018 07:21:26 +0000 (16:21 +0900)]
Fix ASAN build error

Change-Id: Ib04fff8aec20780b141768b8baa1ed72425f696e
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
4 years agoPackaging: add a spec file for Tizen build
Tae-Young Chung [Tue, 27 Mar 2018 06:09:04 +0000 (15:09 +0900)]
Packaging: add a spec file for Tizen build

Change-Id: I3dfd6a7dfe5f74e85b940accfe8c0ac95769277b
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
5 years agoUpdate with 0.3.7 changes upstream/0.3.7 v0.3.7
Martin Kroeker [Sun, 11 Aug 2019 21:23:27 +0000 (23:23 +0200)]
Update with 0.3.7 changes

5 years agoSet version to 0.3.7
Martin Kroeker [Sun, 11 Aug 2019 21:16:45 +0000 (23:16 +0200)]
Set version to 0.3.7

5 years agoSet version to 0.3.7
Martin Kroeker [Sun, 11 Aug 2019 21:16:11 +0000 (23:16 +0200)]
Set version to 0.3.7

5 years agoMerge pull request #2213 from xianyi/develop
Martin Kroeker [Sun, 11 Aug 2019 21:14:49 +0000 (23:14 +0200)]
Merge pull request #2213 from xianyi/develop

Update from develop in preparation of the 0.3.7 release

5 years agoMerge pull request #2212 from martin-frbg/nofort-nolib
Martin Kroeker [Sun, 11 Aug 2019 18:26:34 +0000 (20:26 +0200)]
Merge pull request #2212 from martin-frbg/nofort-nolib

Avoid spurious dependency on the fortran runtime despite NOFORTRAN=1

5 years agoAvoid adding a spurious dependency on the fortran runtime despite NOFORTRAN=1
Martin Kroeker [Sun, 11 Aug 2019 14:24:39 +0000 (16:24 +0200)]
Avoid adding a spurious dependency on the fortran runtime despite NOFORTRAN=1

for cases where a fortran compiler is present but not wanted (e.g. not fully functional)

5 years agoMerge pull request #2211 from martin-frbg/arm64_gcc_trivial
Martin Kroeker [Sun, 11 Aug 2019 14:08:05 +0000 (16:08 +0200)]
Merge pull request #2211 from martin-frbg/arm64_gcc_trivial

Silence two nuisance warnings from gcc

5 years agoSilence two nuisance warnings from gcc
Martin Kroeker [Sun, 11 Aug 2019 10:46:05 +0000 (12:46 +0200)]
Silence two nuisance warnings from gcc

5 years agoMerge pull request #2208 from martin-frbg/munmap-debug
Martin Kroeker [Fri, 9 Aug 2019 05:55:35 +0000 (07:55 +0200)]
Merge pull request #2208 from martin-frbg/munmap-debug

Provide more information on mmap/munmap failure

5 years agoMerge pull request #2206 from martin-frbg/zen-dtrmm
Martin Kroeker [Fri, 9 Aug 2019 05:55:20 +0000 (07:55 +0200)]
Merge pull request #2206 from martin-frbg/zen-dtrmm

Replace vpermpd with vpermilpd in the Haswell DTRMM kernel

5 years agoMerge pull request #2199 from martin-frbg/zen-dtrsm
Martin Kroeker [Fri, 9 Aug 2019 05:55:02 +0000 (07:55 +0200)]
Merge pull request #2199 from martin-frbg/zen-dtrsm

Replace most vpermpd calls in the Haswell DTRSM_RN kernel

5 years agoAdd files via upload
Martin Kroeker [Thu, 8 Aug 2019 22:08:11 +0000 (00:08 +0200)]
Add files via upload

5 years agoProvide more information on mmap/munmap failure
Martin Kroeker [Thu, 8 Aug 2019 21:15:35 +0000 (23:15 +0200)]
Provide more information on mmap/munmap failure

for #2207

5 years agoReplace most vpermpd calls in the Haswell DTRSM_RN kernel
Martin Kroeker [Sat, 3 Aug 2019 10:40:13 +0000 (12:40 +0200)]
Replace most vpermpd calls in the Haswell DTRSM_RN kernel

5 years agoMerge pull request #2198 from martin-frbg/icelake
Martin Kroeker [Fri, 2 Aug 2019 06:36:14 +0000 (08:36 +0200)]
Merge pull request #2198 from martin-frbg/icelake

Update CPUID recognition for Intel Ice Lake

5 years agoAdd CPUID identification of Intel Ice Lake
Martin Kroeker [Thu, 1 Aug 2019 20:52:35 +0000 (22:52 +0200)]
Add CPUID identification of Intel Ice Lake

5 years agoAutodetect Intel Ice Lake (as SKYLAKEX target)
Martin Kroeker [Thu, 1 Aug 2019 20:51:09 +0000 (22:51 +0200)]
Autodetect Intel Ice Lake (as SKYLAKEX target)

5 years agoReplace vpermpd with vpermilpd in the Haswell DTRMM kernel
Martin Kroeker [Sun, 28 Jul 2019 21:17:28 +0000 (23:17 +0200)]
Replace vpermpd with vpermilpd in the Haswell DTRMM kernel

to improve performance on AMD Zen (#2180) applying wjc404's improvement of the DGEMM kernel from #2186

5 years agoMerge pull request #2196 from wjc404/develop
Martin Kroeker [Sun, 28 Jul 2019 21:11:40 +0000 (23:11 +0200)]
Merge pull request #2196 from wjc404/develop

Add vbroadcastsd kernel to dgemm_kernel_4x8_haswell.S

5 years agoAdd files via upload
wjc404 [Sat, 27 Jul 2019 23:39:09 +0000 (07:39 +0800)]
Add files via upload

5 years agoMerge pull request #2112 from ffontaine/develop
Martin Kroeker [Sat, 27 Jul 2019 11:00:13 +0000 (13:00 +0200)]
Merge pull request #2112 from ffontaine/develop

Makefile.arm: remove -march flags

5 years agoMerge pull request #2193 from martin-frbg/makeutest
Martin Kroeker [Wed, 24 Jul 2019 18:19:21 +0000 (20:19 +0200)]
Merge pull request #2193 from martin-frbg/makeutest

Override special make variables

5 years agoUnset special make variables in ctest Makefile as well
Martin Kroeker [Wed, 24 Jul 2019 13:26:09 +0000 (15:26 +0200)]
Unset special make variables in ctest Makefile as well

5 years agoOverride special make variables
Martin Kroeker [Tue, 23 Jul 2019 14:56:40 +0000 (16:56 +0200)]
Override special make variables

as seen in https://github.com/xianyi/OpenBLAS/issues/1912#issuecomment-514183900 , any external setting of TARGET_ARCH (which could result from building OpenBLAS as part of a larger project that actually uses this variable) would cause the utest build to fail.
(Other subtargets appear to be unaffected as they do not use implicit make rules)

5 years agoMerge pull request #2191 from tylerjereddy/conditional_updates
Martin Kroeker [Tue, 23 Jul 2019 14:20:39 +0000 (16:20 +0200)]
Merge pull request #2191 from tylerjereddy/conditional_updates

MAINT: remove legacy CMake endif()

5 years agoMerge pull request #2190 from martin-frbg/zdot-zen
Martin Kroeker [Tue, 23 Jul 2019 14:15:08 +0000 (16:15 +0200)]
Merge pull request #2190 from martin-frbg/zdot-zen

Replace vpermpd with vpermilpd in the Haswell/Zen zdot microkernel

5 years agoMerge pull request #2189 from wjc404/develop
Martin Kroeker [Tue, 23 Jul 2019 06:32:56 +0000 (08:32 +0200)]
Merge pull request #2189 from wjc404/develop

Update dgemm_kernel_4x8_haswell.S for reducing cache misses

5 years agoMAINT: remove legacy CMake endif()
Tyler Reddy [Tue, 23 Jul 2019 03:24:57 +0000 (21:24 -0600)]
MAINT: remove legacy CMake endif()

* clean up a case where CMake endif()
contained the conditional used in the
if(), which is no longer needed /
discouraged since our minimum required
CMake version supports the modern syntax

5 years agoReplace vpermpd with vpermilpd
Martin Kroeker [Mon, 22 Jul 2019 06:28:16 +0000 (08:28 +0200)]
Replace vpermpd with vpermilpd

to improve performance on Zen/Zen2 (as demonstrated by wjc404 in #2180)

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Sat, 20 Jul 2019 17:10:32 +0000 (01:10 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Sat, 20 Jul 2019 16:47:45 +0000 (00:47 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoAdd files via upload
wjc404 [Sat, 20 Jul 2019 14:08:22 +0000 (22:08 +0800)]
Add files via upload

5 years agoAdd files via upload
wjc404 [Sat, 20 Jul 2019 14:04:41 +0000 (22:04 +0800)]
Add files via upload

5 years agoAdd files via upload
wjc404 [Sat, 20 Jul 2019 06:33:37 +0000 (14:33 +0800)]
Add files via upload

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Fri, 19 Jul 2019 15:58:24 +0000 (23:58 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoAdd files via upload
wjc404 [Fri, 19 Jul 2019 15:47:58 +0000 (23:47 +0800)]
Add files via upload

5 years agoMerge pull request #2186 from wjc404/develop
Martin Kroeker [Thu, 18 Jul 2019 14:04:44 +0000 (16:04 +0200)]
Merge pull request #2186 from wjc404/develop

Update "dgemm_kernel_4x8_haswell.S" for improving performance on zen2 chips

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Wed, 17 Jul 2019 15:50:03 +0000 (23:50 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Wed, 17 Jul 2019 15:47:30 +0000 (23:47 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Wed, 17 Jul 2019 14:39:15 +0000 (22:39 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Wed, 17 Jul 2019 13:27:41 +0000 (21:27 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Wed, 17 Jul 2019 09:02:35 +0000 (17:02 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S
wjc404 [Tue, 16 Jul 2019 16:55:06 +0000 (00:55 +0800)]
Update dgemm_kernel_4x8_haswell.S

5 years agoUpdate dgemm_kernel_4x8_haswell.S for zen2
wjc404 [Tue, 16 Jul 2019 16:46:51 +0000 (00:46 +0800)]
Update dgemm_kernel_4x8_haswell.S for zen2

replaced a bunch of vpermpd instructions with vpermilpd and vperm2f128

5 years agoMerge pull request #2181 from isuruf/install_name
Martin Kroeker [Tue, 9 Jul 2019 18:08:52 +0000 (20:08 +0200)]
Merge pull request #2181 from isuruf/install_name

Change install_name on osx to match linux

5 years agoChange install_name on osx to match linux
Isuru Fernando [Mon, 8 Jul 2019 22:13:21 +0000 (17:13 -0500)]
Change install_name on osx to match linux

5 years agoMerge pull request #2177 from martin-frbg/noaff
Martin Kroeker [Sun, 7 Jul 2019 16:28:21 +0000 (18:28 +0200)]
Merge pull request #2177 from martin-frbg/noaff

Fix surprising behaviour of NO_AFFINITY=0

5 years agoFix surprising behaviour of NO_AFFINITY=0
Martin Kroeker [Sun, 7 Jul 2019 14:04:45 +0000 (16:04 +0200)]
Fix surprising behaviour of NO_AFFINITY=0

5 years agoMerge pull request #2175 from martin-frbg/cmake-mingw-fixes
Martin Kroeker [Sat, 6 Jul 2019 16:07:19 +0000 (18:07 +0200)]
Merge pull request #2175 from martin-frbg/cmake-mingw-fixes

Fix CMAKE compilation with MinGW32 and add it to Appveyor

5 years agoMingw32 needs leading underscore on object names
Martin Kroeker [Sat, 6 Jul 2019 13:07:15 +0000 (15:07 +0200)]
Mingw32 needs leading underscore on object names

(also copy BUNDERSCORE settings for FORTRAN from the corresponding Makefile)

5 years agoMake disabling DYNAMIC_ARCH on unsupported systems work
Martin Kroeker [Sat, 6 Jul 2019 13:05:04 +0000 (15:05 +0200)]
Make disabling DYNAMIC_ARCH on unsupported systems work

needs to be unset in the cache for the change to have any effect

5 years agoAdd getarch flags to disable AVX on x86
Martin Kroeker [Sat, 6 Jul 2019 13:02:39 +0000 (15:02 +0200)]
Add getarch flags to disable AVX on x86

(and other small fixes to match Makefile behaviour)

5 years agoAdd mingw builds to Appveyor config
Martin Kroeker [Sat, 6 Jul 2019 12:30:33 +0000 (14:30 +0200)]
Add mingw builds to Appveyor config

5 years agoUtest needs CBLAS but not necessarily FORTRAN
Martin Kroeker [Sat, 6 Jul 2019 12:29:47 +0000 (14:29 +0200)]
Utest needs CBLAS but not necessarily FORTRAN

5 years agoMerge pull request #2162 from martin-frbg/pgi
Martin Kroeker [Wed, 3 Jul 2019 17:16:30 +0000 (19:16 +0200)]
Merge pull request #2162 from martin-frbg/pgi

Fixes for PGI compiler

5 years agoMerge pull request #2172 from quickwritereader/develop
Martin Kroeker [Mon, 1 Jul 2019 19:06:02 +0000 (21:06 +0200)]
Merge pull request #2172 from quickwritereader/develop

power9 cgemm/ctrmm. new sgemm 8x16

5 years agocgemm/ctrmm power9
AbdelRauf [Tue, 18 Jun 2019 15:55:56 +0000 (15:55 +0000)]
cgemm/ctrmm power9

5 years agoMerge pull request #2170 from pkubaj/patch-1
Martin Kroeker [Sun, 30 Jun 2019 21:29:02 +0000 (23:29 +0200)]
Merge pull request #2170 from pkubaj/patch-1

Fix build on PPC970 for FreeBSD

5 years agoFix build for PPC970 on FreeBSD pt.2
pkubaj [Fri, 28 Jun 2019 10:31:45 +0000 (10:31 +0000)]
Fix build for PPC970 on FreeBSD pt.2

FreeBSD needs those macros too.

5 years agoFix build for PPC970 on FreeBSD pt. 1
pkubaj [Fri, 28 Jun 2019 10:29:44 +0000 (10:29 +0000)]
Fix build for PPC970 on FreeBSD pt. 1

FreeBSD needs DCBT_ARG=0 as well.

5 years agoMerge pull request #2169 from pkubaj/develop
Martin Kroeker [Tue, 25 Jun 2019 10:56:33 +0000 (12:56 +0200)]
Merge pull request #2169 from pkubaj/develop

Fix build on FreeBSD/powerpc64.

5 years agoFix build on FreeBSD/powerpc64.
Piotr Kubaj [Tue, 25 Jun 2019 08:58:56 +0000 (10:58 +0200)]
Fix build on FreeBSD/powerpc64.

Signed-off-by: Piotr Kubaj <pkubaj@anongoth.pl>
5 years agoPGI compiler does not like -march=native
Martin Kroeker [Thu, 20 Jun 2019 17:56:01 +0000 (19:56 +0200)]
PGI compiler does not like -march=native

5 years agoMerge pull request #2167 from kavanabhat/dtrmm_power8_segfault
Martin Kroeker [Wed, 19 Jun 2019 12:38:01 +0000 (14:38 +0200)]
Merge pull request #2167 from kavanabhat/dtrmm_power8_segfault

Fix DTRMMKERNEL register save for power8 64-bit mode (Fix for #2166)

5 years agoUpdate dtrmm_kernel_16x4_power8.S
kavanabhat [Wed, 19 Jun 2019 09:57:14 +0000 (15:27 +0530)]
Update dtrmm_kernel_16x4_power8.S

5 years agonew sgemm 8x16
AbdelRauf [Mon, 17 Jun 2019 15:33:38 +0000 (15:33 +0000)]
new sgemm 8x16

5 years agoFix mov syntax
Martin Kroeker [Sun, 16 Jun 2019 16:35:43 +0000 (18:35 +0200)]
Fix mov syntax

5 years agoZero ecx with a mov instruction
Martin Kroeker [Sun, 16 Jun 2019 13:04:10 +0000 (15:04 +0200)]
Zero ecx with a mov instruction

PGI assembler does not like the initialization in the constraints.

5 years agoUpdate Makefile.x86_64
Martin Kroeker [Fri, 14 Jun 2019 06:08:11 +0000 (08:08 +0200)]
Update Makefile.x86_64

5 years agoDo not force gcc options on non-gcc compilers
Martin Kroeker [Thu, 13 Jun 2019 21:01:35 +0000 (23:01 +0200)]
Do not force gcc options on non-gcc compilers

fixes compile failure with pgi 18.10 as reported on OpenBLAS-users

5 years agoMerge pull request #2159 from martin-frbg/issue2149
Martin Kroeker [Mon, 10 Jun 2019 17:12:45 +0000 (19:12 +0200)]
Merge pull request #2159 from martin-frbg/issue2149

Avoid unintentional activation of TLS codepath via USE_TLS=0

5 years agoAvoid unintentional activation of TLS code via USE_TLS=0
Martin Kroeker [Mon, 10 Jun 2019 15:24:15 +0000 (17:24 +0200)]
Avoid unintentional activation of TLS code via USE_TLS=0

fixes #2149

5 years agoMerge pull request #2158 from martin-frbg/issue2143
Martin Kroeker [Mon, 10 Jun 2019 12:08:11 +0000 (14:08 +0200)]
Merge pull request #2158 from martin-frbg/issue2143

Remove any inadvertent use of -march=native from DYNAMIC_ARCH builds

5 years agoRemove any inadvertent use of -march=native from DYNAMIC_ARCH builds
Martin Kroeker [Mon, 10 Jun 2019 07:50:13 +0000 (09:50 +0200)]
Remove any inadvertent use of -march=native from DYNAMIC_ARCH builds

from #2143, -march=native precludes use of more specific options like -march=skylake-avx512 in individual kernels, and defeats the purpose of dynamic arch anyway.

5 years agoMerge pull request #2157 from martin-frbg/2154-2
Martin Kroeker [Sun, 9 Jun 2019 10:19:08 +0000 (12:19 +0200)]
Merge pull request #2157 from martin-frbg/2154-2

Add gfortran workaround for potential ABI violation

5 years agoUpdate fc.cmake
Martin Kroeker [Sun, 9 Jun 2019 07:31:13 +0000 (09:31 +0200)]
Update fc.cmake

5 years agoAdd gfortran workaround for potential ABI violation
Martin Kroeker [Sat, 8 Jun 2019 21:17:03 +0000 (23:17 +0200)]
Add gfortran workaround for potential ABI violation

for #2154

5 years agoMerge pull request #2148 from TiborGY/cpp_thread_test_2
Martin Kroeker [Fri, 7 Jun 2019 11:23:07 +0000 (13:23 +0200)]
Merge pull request #2148 from TiborGY/cpp_thread_test_2

Thread safety tester using C++11 threading (cleaned history)

5 years agoMerge pull request #2156 from martin-frbg/issue2154
Martin Kroeker [Thu, 6 Jun 2019 11:43:12 +0000 (13:43 +0200)]
Merge pull request #2156 from martin-frbg/issue2154

Add gfortran workaround for C->FORTRAN ABI violation

5 years agoAdd gfortran workaround for ABI violations
Martin Kroeker [Thu, 6 Jun 2019 08:24:16 +0000 (10:24 +0200)]
Add gfortran workaround for ABI violations

for #2154 (see gcc bug 90329)

5 years agoAdd gfortran workaround for ABI violations in LAPACKE
Martin Kroeker [Thu, 6 Jun 2019 08:18:40 +0000 (10:18 +0200)]
Add gfortran workaround for ABI violations in LAPACKE

for #2154 (see gcc bug 90329)

5 years agoMerge pull request #2153 from quickwritereader/develop
Martin Kroeker [Thu, 6 Jun 2019 05:42:56 +0000 (07:42 +0200)]
Merge pull request #2153 from quickwritereader/develop

improved power9 zgemm,sgemm

5 years agoconflict resolve
AbdelRauf [Wed, 5 Jun 2019 20:50:50 +0000 (20:50 +0000)]
conflict resolve

5 years agopower9 zgemm ztrmm optimized
AbdelRauf [Wed, 5 Jun 2019 10:30:57 +0000 (10:30 +0000)]
power9 zgemm ztrmm optimized

5 years agoMerge pull request #2145 from martin-frbg/1912-3
Martin Kroeker [Wed, 5 Jun 2019 18:27:45 +0000 (20:27 +0200)]
Merge pull request #2145 from martin-frbg/1912-3

Separate implementations of AMAX and IAMAX on arm

5 years agoMerge pull request #2110 from pc2/cpu-detection
Martin Kroeker [Wed, 5 Jun 2019 18:27:05 +0000 (20:27 +0200)]
Merge pull request #2110 from pc2/cpu-detection

Fix detection of Skylake processors when using GCC

5 years agoc_check: Unlink correct file
Michael Lass [Fri, 3 May 2019 19:22:27 +0000 (21:22 +0200)]
c_check: Unlink correct file

5 years agoFix detection of AVX512 capable compilers in getarch
Michael Lass [Fri, 3 May 2019 19:07:14 +0000 (21:07 +0200)]
Fix detection of AVX512 capable compilers in getarch

21eda8b5 introduced a check in getarch.c to test if the compiler is capable of
AVX512. This check currently fails, since the used __AVX2__ macro is only
defined if getarch itself was compiled with AVX2/AVX512 support. Make sure this
is the case by building getarch with -march=native on x86_64. It is only
supposed to run on the build host anyway.

5 years agosgemm pipeline improved, zgemm rewritten without inner packs, ABI lxvx v20 fixed...
AbdelRauf [Fri, 31 May 2019 22:48:16 +0000 (22:48 +0000)]
sgemm pipeline improved, zgemm rewritten without inner packs, ABI lxvx v20 fixed with vs52

5 years agoDocument NO_AVX512
Martin Kroeker [Mon, 3 Jun 2019 09:01:33 +0000 (11:01 +0200)]
Document NO_AVX512

for #2151

5 years ago add c++ thread test option to Makefile.rule
TiborGY [Sat, 1 Jun 2019 19:36:41 +0000 (21:36 +0200)]
 add c++ thread test option to Makefile.rule

5 years agohook up c++ thread safety test (main Makefile)
TiborGY [Sat, 1 Jun 2019 19:32:52 +0000 (21:32 +0200)]
hook up c++ thread safety test (main Makefile)