a64fe9bcc9
added optimized sgemv_n kernel for sandybridge
2014-09-06 08:41:53 +02:00
2021d0f9d6
experimentally removed expensive function calls
2014-09-05 15:05:53 +02:00
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.)
2014-08-03 15:06:30 -07:00
793175be3a
added experimental support for big numa machines
2014-08-02 13:40:16 +02:00
1d33547222
optimized zgemm kernel for haswell
2014-07-27 11:51:42 +02:00
3ea4dadd30
optimizations for trsm
2014-07-25 11:59:17 +02:00
1b10ff129a
optimizations for trmm
2014-07-25 10:00:23 +02:00
125610d23b
allow to set custom value for ?GEMM_DEFAULT_UNROLL_MN, optimizations for syrk
2014-07-24 18:43:31 +02:00
c94762bb56
Refs #401 . Added NO_AVX2 flag for old binutils (e.g. RHEL6)
2014-07-16 08:38:25 +08:00
552119c484
Fixed #407 . Support outputing the CPU corename on runtime.
...
The user can use char * openblas_get_config() or char * openblas_get_corename().
2014-07-08 12:48:08 +08:00
50e99a52ea
added definitions for PILEDRIVER and HASWELL
2014-07-06 12:08:27 +02:00
be94db096c
disabled *3M functions for x86_64 platforms
2014-07-01 16:18:05 +02:00
7a8949e0ce
Merge branch 'develop' of https://github.com/TimothyGu/OpenBLAS into TimothyGu-develop
...
Conflicts:
driver/others/memory.c
2014-06-28 20:51:31 +08:00
6c2ead30f0
Remove all trailing whitespace except lapack-netlib
...
Signed-off-by: Timothy Gu <timothygu99@gmail.com>
2014-06-27 12:05:18 -07:00
f41f03ab83
fix #394 . this cleans up some handles after using them, and doesn't disable ALL process privileges upon success
2014-06-27 12:16:57 -04:00
438002204d
Ref #393 : fix for INTERFACE64=0 and ARCH_X86 in divtable
2014-06-21 12:29:23 +02:00
53bfa51ee0
Ref #385 : fixed warnings in dynamic.c
2014-06-12 18:17:08 +02:00
a86d349a51
Ref #380 : enhancements for dynamic_arch
2014-06-12 14:20:03 +02:00
a35a1a9ae7
changed makefiles for lapack development
2014-05-07 11:33:02 +02:00
2c556f093a
Add cast to function pointer to remove warning
2014-02-25 11:08:32 +01:00
3b027d2528
Do not reference pthread_atfork in non-SMP_SERVER mode
2014-02-25 11:08:32 +01:00
49bd98f410
Do not reference pthread_atfork under windows
2014-02-19 19:25:48 +01:00
138a841390
FIX #294 : make OpenBLAS thread-pool resilient to fork via pthread_atfork
2014-02-19 19:01:15 +01:00
046e4013cb
Revert "Refs #294 . Used pthread_atfork to avoid hang after a Unix fork."
...
This reverts commit 3617c22a56
2014-02-19 18:32:54 +01:00
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.
2014-02-18 15:36:04 +08:00
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.
2014-01-24 02:05:59 +08:00
ab69443bd4
Refs #332 . Added addtional Intel Ivy Bridge and Haswell CPU-id.
2014-01-05 23:44:29 +08:00
b263e096af
Refs #307 . Delete debug printf.
2013-12-31 15:53:13 +08:00
8191efc420
changed level3.c
2013-12-01 13:52:08 +01:00
0b6e13b689
Merge remote branch 'origin/develop' into haswell
2013-12-01 13:38:11 +01:00
5c648a8984
Merge remote branch 'origin/develop' into haswell
2013-12-01 11:25:33 +01:00
5048a80032
Refs #283 . Fixed the incorrect usage of long data type for Windows 64.
2013-11-14 13:46:42 +08:00
a2942456ef
Refs #307 . Fixed the hang bug when free OpenBLAS dll in Windows.
2013-11-13 10:00:18 +08:00
a77c71eaf5
added highly optimized dgemm_kernel for HASWELL
2013-10-28 10:23:47 +01:00
fe8c5666f9
optimized dgemm_kernel for HASWELL
2013-10-20 16:52:26 +02:00
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
2013-08-28 14:29:42 +02:00
2638370844
Init code base for Intel Haswell.
2013-08-13 00:54:59 +08:00
673e453b3f
Enable bulldozer kernels.
2013-08-05 16:07:54 +08:00
143cca4dd5
Merge branch 'develop' into bulldozer
2013-08-05 15:51:53 +08:00
534c5ec919
Fixed #261 . Use strncmp instead of a comparing trick.
2013-07-29 16:48:35 +08:00
5b504d6c23
Refs #263 . Rollback bulldozer and piledriver kernels to barcelona kernels.
2013-07-28 17:39:24 +08:00
72b1edaf1b
Merge branch 'develop' into bulldozer
...
Conflicts:
kernel/x86_64/KERNEL.BULLDOZER
2013-07-28 06:38:25 +02:00
4471c77905
Fixed #261 . Use strncmp instead of a comparing trick.
2013-07-26 23:43:54 +08:00
77b572fa0b
Merge branch 'loongson3a' into develop
...
Conflicts:
Makefile.system
2013-07-20 22:33:17 +08:00
2a7503e563
Refs #225 . Fixed a bug in GEMM OpenMP threading.
2013-07-15 09:56:19 +08:00
c19a488af2
create openblas_get_parallel to retrieve information which
...
parallelization model is used by OpenBLAS.
2013-07-11 21:39:19 +08:00
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.
2013-07-11 03:20:02 +08:00
6f008abcef
replaced defined(DOUBLE) by !defined(XDOUBLE)
2013-07-09 18:17:50 +02:00
f54f5bac9e
Refs #248 . Fixed the LSB compatiable issue for BLAS only.
...
For example, make CC=lsbcc NO_LAPACK=1.
2013-07-09 15:38:03 +08:00
5d3312142a
Refs #221 #246 . Fixed the overflowing stack bug in mutlithreading BLAS3.
...
When NUM_THREADS(MAX_CPU_NUNBERS) is very large ,e.g. 256.
typedef struct {
volatile BLASLONG working[MAX_CPU_NUMBER][CACHE_LINE_SIZE * DIVIDE_RATE];
} job_t;
job_t job[MAX_CPU_NUMBER];
The job array is equal 8MB.
Thus, We use malloc instead of stack allocation.
2013-07-08 01:07:05 +08:00
wernsaar
Isaac Dunham
Zhang Xianyi
Timothy Gu
Jameson Nash
Olivier Grisel
Sébastien Villemot
grisuthedragon