Martin Kroeker
880bc1d1db
Acknowledge past CZI EOSS 1/EOSS 3 funding
3 years ago
Martin Kroeker
d0ba257de0
Merge pull request #3704 from XiWeiGu/loongarch64_dynamic_arch
LoongArch64: Add DYNAMIC_ARCH support
3 years ago
Martin Kroeker
78da6a750a
Merge pull request #3705 from RajalakshmiSR/bf16ppc
POWER: Enable bfloat16 kernels by default
3 years ago
Rajalakshmi Srinivasaraghavan
1d97405c02
POWER: Enable bfloat16 kernels by default
This patch enables bfloat16 kernels by default for POWER processors.
Tested on Linux POWER8, POWER9, POWER10 and AIX POWER10 systems.
3 years ago
gxw
fbfe1daf6e
LoongArch64: Add DYNAMIC_ARCH support
3 years ago
Martin Kroeker
047a279f09
Merge pull request #3702 from martin-frbg/issue3687
Add openblas_getaffinity() extension (Linux-only)
3 years ago
Martin Kroeker
30473b6a9d
add openblas_getaffinity()
3 years ago
Martin Kroeker
8668571040
add openblas_getaffinity()
3 years ago
Martin Kroeker
daca01622b
fix detection of Neoverse V1 and user-enforced selection of N2 in ARM64 DYNAMIC_ARCH ( #3700 )
* fix detection of Neoverse V1 and user-enforced selection of N2
3 years ago
Martin Kroeker
c322aab685
Merge pull request #3684 from imzhuhl/neoversen2_dynamic_arch
Neoverse N2: DYNAMIC_ARCH
3 years ago
Martin Kroeker
cf796aee8c
Merge pull request #3699 from martin-frbg/issue3692
Add c_check recognition of Fujitsu fcc for Fugaku A64FX
3 years ago
Martin Kroeker
28d40ba60b
Merge pull request #3696 from XiWeiGu/loongson2k1000
LoongArch64: Add core LOONGSON2K1000 and LOONGSONGENERIC
3 years ago
Martin Kroeker
692848d20c
typo fix
3 years ago
Martin Kroeker
76ea7739dd
Merge pull request #3698 from martin-frbg/issue3697
utest needs to be linked against libm on QNX as well
3 years ago
Martin Kroeker
f8c5bdfbab
Treat Fujitsu fcc on Fugaku like clang
3 years ago
Martin Kroeker
70001e1e9e
Add Fujitsu compiler
3 years ago
Martin Kroeker
cf37182260
Add Fujitsu compiler (fcc)
3 years ago
Martin Kroeker
68d86ea150
Add Fujitsu compiler
3 years ago
Martin Kroeker
7aaa0ce0e8
utest needs to be linked against libm on QNX as well
3 years ago
Martin Kroeker
cd8e57040c
Merge pull request #3691 from martin-frbg/issue3679-sparc
SPARC: fix DNRM2 returning INF instead of zero due to intermediate overflow
3 years ago
gxw
3573306a69
LoongArch64: Add core LOONGSON2K1000 and LOONGSONGENERIC
3 years ago
Martin Kroeker
a4303ae378
Merge pull request #3695 from martin-frbg/ppc6nrm2
PPC6: Fix DNRM2 returning INF instead of zero due to intermediate overflow
3 years ago
Martin Kroeker
31377d04f0
Merge pull request #3694 from martin-frbg/traviswait
Add back travis_wait to keep ppc jobs from getting cancelled
3 years ago
Martin Kroeker
6c118b7977
Fix DNRM2 returning INF instead of zero due to intermediate overflow
3 years ago
Martin Kroeker
b60415a347
Add back travis_wait to keep ppc jobs from getting cancelled
3 years ago
Martin Kroeker
c43ec53bdd
Merge pull request #3690 from RajalakshmiSR/cdotp10
POWER: Fix complex dot function failures
3 years ago
Martin Kroeker
b7c65d08cb
Merge pull request #3689 from RajalakshmiSR/dgemvgcc10
POWER10: dgemv builtin rename
3 years ago
Martin Kroeker
fcbbd8c25c
Merge pull request #3682 from XiWeiGu/develop
Fix dnrm2_tiny testcase failure
3 years ago
Martin Kroeker
06ef015234
fix DNRM2 returning INF instead of zero due to intermediate overflow
3 years ago
Rajalakshmi Srinivasaraghavan
a612e78a97
POWER: Fix complex dot function failures
There are some test failures in complex dot functions when compiling with gcc12.
The machine constraints used now do not update all the four elements in the
expected result array. Fixing this with a reduced level of optimization.
This is not changing any performance numbers but will be converted to C code in future.
3 years ago
Rajalakshmi Srinivasaraghavan
432fd99445
POWER10: dgemv builtin rename
Add check to use correct builtin name for older versions
of gcc10 compilers.
3 years ago
gxw
4dd05e526b
LoongArch64: Fix dnrm2_tiny testcase failure
3 years ago
Martin Kroeker
7da799dc66
Merge pull request #3686 from martin-frbg/issue3685
Fix Fortran-less CTEST build option
3 years ago
Martin Kroeker
6e018b84c4
Fix function prototypes and INTERFACE64 support
3 years ago
Martin Kroeker
ccd87cc472
Fix switching between Fortran and C build
3 years ago
Honglin Zhu
d5ca477f42
Neoverse N2: DYNAMIC_ARCH
3 years ago
gxw
cce4b1d956
MIPS64: Fix dnrm2_tiny testcase failure
3 years ago
Martin Kroeker
7918ba11c2
Merge pull request #3680 from martin-frbg/issue3636-2
Guard against sysconf(__SC_NPROCESSORS_CONF) returning zero at runtime
3 years ago
Martin Kroeker
69148ae795
Guard against sysconf returning zero processors
3 years ago
Martin Kroeker
e9260f5451
Guard against system call returning zero processors
3 years ago
Martin Kroeker
4cfd6f110a
Merge pull request #3678 from martin-frbg/issue3677
Eliminate uses of CREAL on left-hand side of assignments
3 years ago
Martin Kroeker
e12d474780
Eliminate uses of CREAL on left-hand side of assignments
3 years ago
Martin Kroeker
686e6d7c10
Merge pull request #3676 from martin-frbg/dnrm2-utest
Add DNRM2 regression test for issues 2998 and 3654
3 years ago
Martin Kroeker
c5041ae270
properly embed test_dnrm2
3 years ago
Martin Kroeker
8e6f719ad3
use huge_val not huge_valf for portability
3 years ago
Martin Kroeker
af88494f87
old systems may not have inf in math.h
3 years ago
Martin Kroeker
ee41b6eb24
Add DNRM2 regression test for issues 2998 and 3654
3 years ago
Martin Kroeker
bf8998a9f4
Merge pull request #3675 from martin-frbg/issue3654
workaround ThunderX2 DNRM2 fault with ssq=inf,scale=0
3 years ago
Martin Kroeker
9e29598575
workaround fault with ssq=inf,scale=0
3 years ago
Martin Kroeker
3df3d622eb
Merge pull request #3672 from imzhuhl/neoversen2_bf16
sbgemm support for ARM Neoverse N2
3 years ago