OpenBLAS

Commit Graph

Author	SHA1	Message	Date
Zhang Xianyi	32dbeb636d	Refs #221 . Set stack limit to 16MB to prevent a SEGFAULT bug on Mac OS X with DYNAMIC_ARCH=1 & NUM_THREADS=256.	12 years ago
Dan Luu	88ef307cef	Refs #241 . Add Haswell support (using sandybridge optimizations)	12 years ago
Zhang Xianyi	cd1d473ba0	Merge pull request #230 from wernsaar/develop Refs #230. New dgemm and sgemm Kernel for BULLDOZER	12 years ago
wernsaar	25491e42f9	New dgemm kernel for BULLDOZER: dgemm_kernel_8x2_bulldozer.S	12 years ago
Zhang Xianyi	65ffead0cf	Refs #124 . Check XSAVE flag on x86 CPU.	12 years ago
Zhang Xianyi	f1ce74ffdd	Improved the print when OS don't support AVX.	13 years ago
Zhang Xianyi	d744c9590a	In OpenMP threading, preallocate the thread buffer instead of allocating the buffer every time. This patch improved the performance slightly.	13 years ago
Zhang Xianyi	3cc6ae793e	Refs #174 . Return sb pointer when OpenMP or Windows.	13 years ago
Zhang Xianyi	5155e3f509	Refs #174 . Fixed the overflowing buffer bug of multithreading hbmv and sbmv. Instead of using thread 0 buffer, each thread uses its own sb buffer. Thus, it can avoid overflowing thread 0 buffer.	13 years ago
Zhang Xianyi	5c8bf6ae0e	Merge branch 'bulldozer' into develop	13 years ago
Zhang Xianyi	8cdb795438	Refs #187 . Use binary code for xgetbv, which is compatible with old compiler.	13 years ago
Zhang Xianyi	a4ee6f3915	Fixed #172 . Support Intel Xeon E7540.	13 years ago
Zhang Xianyi	fba6b590f2	Merge branch 'master' into develop	13 years ago
Julian Taylor	1138817dd2	add a sanity check on the detected cpu type if we have 64 bit pointers we can't have a 32 bit cpu, so fall back to the 64bit cpu fallback (prescott) E.g. the cpu detection fails in amd qemu64 emulation (family 6 model 2) causing it to use the uninitialized gotoblas_ATHLON	13 years ago
Zhang Xianyi	bdf8d9411e	Refs #163 . Obtain the build configure on runtime. openblas_get_config function returns the configure string. So far, it supports USE64BITINT, NO_CBLAS, NO_LAPACK, NO_LAPACKE, DYNAMIC_ARCH, NO_AFFINITY. Example: #include <stdio.h> extern char * openblas_get_config(); void main() { printf("%s\n",openblas_get_config()); return; }	13 years ago
Zhang Xianyi	bfaaa975e6	Added BULLDOZER target. So far it uses barcelona kernels.	13 years ago
Zhang Xianyi	b7c0fa6bd2	Init AMD Bulldozer codebase.	13 years ago
Zhang Xianyi	6751f7b9a7	Fixed #157 . Only detect the number of physical CPU cores on Mac OSX.	13 years ago
Zhang Xianyi	538c764d2b	Refs #153 . Restore the original CPU affinity when calling openblas_set_num_threads(1). Please read the issue on github.com for the detail.	13 years ago
Zhang Xianyi	6c5899dff5	Don't use xgetbv instruction when NO_AVX=1	13 years ago
Zhang Xianyi	735ca38b8f	Refs #139 . Check OS supporting AVX on runtime.	13 years ago
Zhang Xianyi	f76a384841	Refs #139 . Added NO_AVX flag to use old Nehalem kernels on Sandy Bridge. For example, make NO_AVX=1 or make DYNAMIC_ARCH=1 NO_AVX=1	13 years ago
Jameson Nash	d0e731e8b8	provide support for passing CFLAGS, FFLAGS, PFLAGS, FPFLAGS to make on the command line	13 years ago
Zhang Xianyi	fe4ab95cd5	Refs #136 . Fixed a bug about controlling the number of threads on Windows.	13 years ago
Xianyi Zhang	801383effe	Fixed a hang bug when shutdown blas threads server on Windows. Added the feature about dynamic changing the number of threads on Windows.	13 years ago
Zhang Xianyi	54cd65e47f	Use sandy bridge kernel when DYNAMIC_ARCH=1.	13 years ago
Zhang Xianyi	a55821a2ec	Refs #132 . Kill the threads when unload the library.	13 years ago
Zhang Xianyi	d007cca61d	Refs #134 . Fixed the building bug on IBM Power.	13 years ago
Xianyi Zhang	25f1a573fd	Fixed the build bug when DYNAMIC_ARCH=0.	13 years ago
Sylvestre Ledru	3692b4d631	Improve the detection of sparc	13 years ago
Xianyi Zhang	a507b56ab1	Refs #119 #118 . Fixed disabling hyper threading bug.	13 years ago
Xianyi Zhang	853d16ed7e	Added openblas_set_num_threads dummy function on Windows. We plan to implement this feature in next version.	13 years ago
Zhang Xianyi	422359d09a	Export openblas_set_num_threads in shared library.	13 years ago
Zhang Xianyi	d3b67d0bd8	Refs #113 . Fixed the typo BOBCATE -> BOBCAT	13 years ago
Zhang Xianyi	d6cab3f37e	Refs #113 . Support AMD Bobcate using Barcelona kernel codes. Replace 3DNow! with MMX.	13 years ago
Zhang Xianyi	90d6ad569d	Merge branch 'sandybridge' into develop Just copy the kernel codes from Nehalem. The optimization is ongoing.	13 years ago
Xianyi Zhang	a6adbb299d	Refs #112 . Improved setting thread affinity in Linux. Remove the limit (64) about the number of CPU cores.	13 years ago
Xianyi Zhang	a53c6e2440	Merge branch 'develop' into sandybridge	13 years ago
Zaheer Chothia	a431042475	Fix inconsistent case for OS_* macros (Refs pull request #111 )	13 years ago
Mike Nolta	4e29b6ffc0	FreeBSD: fix OS_FreeBSD -> OS_FREEBSD typos	13 years ago
Xianyi Zhang	19a48b82cf	Init Sandybridge codes based on Nehalem.	14 years ago
Xianyi Zhang	0b89a7a92d	Ref #82 . Disable outputing debug information in alloc_mmap.	14 years ago
Wang Qian	8163ab7e55	Change the block size on Loongson 3B.	14 years ago
Xianyi Zhang	ef6f7f32ae	Fixed mbind bug on Loongson 3B. Check the return value of my_mbind function.	14 years ago
Xianyi Zhang	b95ad4cfaf	Support detecting ICT Loongson-3B CPU.	14 years ago
traz	9fe3049de6	Adding conditional compilation(#if defined(LOONGSON3A)) to avoid affecting the performance of other platforms.	14 years ago
traz	831858b883	Modify aligned address of sa and sb to improve the performance of multi-threads.	14 years ago
Xianyi Zhang	16fc083322	Refs #47 . Fixed the seting parameter bug on Loongson 3A single thread version.	14 years ago
Xianyi Zhang	3c856c0c1a	Check the return value of pthread_create. Update the docs with known issue on Loongson 3A.	14 years ago
Xianyi Zhang	4727fe8abf	Refs #47 . On Loongson 3A, set DGEMM_R parameter depending on different number of threads. It would improve double precision BLAS3 on multi-threads.	14 years ago

1 2

63 Commits (cc522aa21d95968dbba3631723cd478dd73daeb4)