wernsaar
7aae4a62e7
enabled use of GEMM3M functions
11 years ago
wernsaar
a64fe9bcc9
added optimized sgemv_n kernel for sandybridge
11 years ago
wernsaar
2021d0f9d6
experimentally removed expensive function calls
11 years ago
Isaac Dunham
f7eb81a846
Fix link error on Linux/musl.
get_nprocs() is a GNU convenience function equivalent to POSIX2008
sysconf(_SC_NPROCESSORS_ONLN); the latter should be available in unistd.h
on any current *nix. (OS X supports this call since 10.5, and FreeBSD
currently supports it. But this commit does not change FreeBSD or OS X
versions.)
11 years ago
wernsaar
793175be3a
added experimental support for big numa machines
11 years ago
wernsaar
1d33547222
optimized zgemm kernel for haswell
11 years ago
wernsaar
3ea4dadd30
optimizations for trsm
11 years ago
wernsaar
1b10ff129a
optimizations for trmm
11 years ago
wernsaar
125610d23b
allow to set custom value for ?GEMM_DEFAULT_UNROLL_MN, optimizations for syrk
11 years ago
Zhang Xianyi
c94762bb56
Refs #401 . Added NO_AVX2 flag for old binutils (e.g. RHEL6)
11 years ago
Zhang Xianyi
552119c484
Fixed #407 . Support outputing the CPU corename on runtime.
The user can use char * openblas_get_config() or char * openblas_get_corename().
11 years ago
wernsaar
50e99a52ea
added definitions for PILEDRIVER and HASWELL
11 years ago
wernsaar
be94db096c
disabled *3M functions for x86_64 platforms
11 years ago
Zhang Xianyi
7a8949e0ce
Merge branch 'develop' of https://github.com/TimothyGu/OpenBLAS into TimothyGu-develop
Conflicts:
driver/others/memory.c
11 years ago
Timothy Gu
6c2ead30f0
Remove all trailing whitespace except lapack-netlib
Signed-off-by: Timothy Gu <timothygu99@gmail.com>
11 years ago
Jameson Nash
f41f03ab83
fix #394 . this cleans up some handles after using them, and doesn't disable ALL process privileges upon success
11 years ago
wernsaar
438002204d
Ref #393 : fix for INTERFACE64=0 and ARCH_X86 in divtable
11 years ago
wernsaar
53bfa51ee0
Ref #385 : fixed warnings in dynamic.c
11 years ago
wernsaar
a86d349a51
Ref #380 : enhancements for dynamic_arch
11 years ago
wernsaar
a35a1a9ae7
changed makefiles for lapack development
11 years ago
Olivier Grisel
2c556f093a
Add cast to function pointer to remove warning
11 years ago
Olivier Grisel
3b027d2528
Do not reference pthread_atfork in non-SMP_SERVER mode
11 years ago
Olivier Grisel
49bd98f410
Do not reference pthread_atfork under windows
11 years ago
Olivier Grisel
138a841390
FIX #294 : make OpenBLAS thread-pool resilient to fork via pthread_atfork
11 years ago
Olivier Grisel
046e4013cb
Revert "Refs #294 . Used pthread_atfork to avoid hang after a Unix fork."
This reverts commit 3617c22a56
.
11 years ago
Zhang Xianyi
3617c22a56
Refs #294 . Used pthread_atfork to avoid hang after a Unix fork.
The problem is the mutex we used in blas_server. Thus, we must clear
the mutex before the fork and re-init them at parent and child process.
If you used OpenMP, GOMP has the same problem by now. Please try other OpenMP
implemantation.
11 years ago
Zhang Xianyi
8c7687b419
Refs #338 . Added OPENBLAS_VERBOSE environment variable on runtime
By default, OpenBLAS doesn't output the warning message. You can set
OPENBLAS_VERBOSE (e.g. export OPENBLAS_VERBOSE=1) to enable the warning
message on runtime.
11 years ago
Zhang Xianyi
ab69443bd4
Refs #332 . Added addtional Intel Ivy Bridge and Haswell CPU-id.
11 years ago
Zhang Xianyi
b263e096af
Refs #307 . Delete debug printf.
11 years ago
wernsaar
8191efc420
changed level3.c
12 years ago
wernsaar
0b6e13b689
Merge remote branch 'origin/develop' into haswell
12 years ago
wernsaar
5c648a8984
Merge remote branch 'origin/develop' into haswell
12 years ago
Zhang Xianyi
5048a80032
Refs #283 . Fixed the incorrect usage of long data type for Windows 64.
12 years ago
Zhang Xianyi
a2942456ef
Refs #307 . Fixed the hang bug when free OpenBLAS dll in Windows.
12 years ago
wernsaar
a77c71eaf5
added highly optimized dgemm_kernel for HASWELL
12 years ago
wernsaar
fe8c5666f9
optimized dgemm_kernel for HASWELL
12 years ago
Sébastien Villemot
eae4cfa3f6
Avoid failure on qemu guests declaring an Athlon CPU without 3dnow!
The present patch verifies that, on machines declaring an Athlon CPU model and
family, the 3dnow and 3dnowext feature flags are indeed present. If they are
not, it fallbacks on the most generic x86 kernel. This prevents crashes due to
illegal instruction on qemu guests with a weird configuration.
Closes #272
12 years ago
Zhang Xianyi
2638370844
Init code base for Intel Haswell.
12 years ago
Zhang Xianyi
673e453b3f
Enable bulldozer kernels.
12 years ago
Zhang Xianyi
143cca4dd5
Merge branch 'develop' into bulldozer
12 years ago
Zhang Xianyi
534c5ec919
Fixed #261 . Use strncmp instead of a comparing trick.
12 years ago
Zhang Xianyi
5b504d6c23
Refs #263 . Rollback bulldozer and piledriver kernels to barcelona kernels.
12 years ago
Zhang Xianyi
72b1edaf1b
Merge branch 'develop' into bulldozer
Conflicts:
kernel/x86_64/KERNEL.BULLDOZER
12 years ago
Zhang Xianyi
4471c77905
Fixed #261 . Use strncmp instead of a comparing trick.
12 years ago
Zhang Xianyi
77b572fa0b
Merge branch 'loongson3a' into develop
Conflicts:
Makefile.system
12 years ago
Zhang Xianyi
2a7503e563
Refs #225 . Fixed a bug in GEMM OpenMP threading.
12 years ago
grisuthedragon
c19a488af2
create openblas_get_parallel to retrieve information which
parallelization model is used by OpenBLAS.
12 years ago
Zhang Xianyi
32d2ca3035
Refs #214 , #221 , #246 . Fixed the getrf overflow bug on Windows.
I used a smaller threshold since the stack size is 1MB on windows.
12 years ago
wernsaar
6f008abcef
replaced defined(DOUBLE) by !defined(XDOUBLE)
12 years ago
Zhang Xianyi
f54f5bac9e
Refs #248 . Fixed the LSB compatiable issue for BLAS only.
For example, make CC=lsbcc NO_LAPACK=1.
12 years ago