Commit Graph

  • e4663be46a added symv benchmark wernsaar 2014-07-21 07:50:54 +0200
  • 11637b6926 add benchmark for ger wernsaar 2014-07-21 06:25:42 +0200
  • 80bf3e6a35 Merge pull request #419 from wernsaar/develop Zhang Xianyi 2014-07-20 23:35:17 +0800
  • 6acbafe45b added sgemv_n microkernel for haswell wernsaar 2014-07-20 14:52:25 +0200
  • 5392d11b04 optimized sgemv_n_microk_sandy.c wernsaar 2014-07-20 14:08:04 +0200
  • c0fe95fb72 added sgemv_n microkernel for sandybridge wernsaar 2014-07-20 13:17:47 +0200
  • d9d4077c93 added sgemv_t microkernel for haswell wernsaar 2014-07-20 11:30:32 +0200
  • 02eb72ac42 bugfix in sgemv_t_microk_sandy.c wernsaar 2014-07-20 10:48:41 +0200
  • c06f9986d4 added sgemv_t microkernel for sandybridge wernsaar 2014-07-20 10:21:08 +0200
  • 2cce125c79 added optimized sgemv_t for bulldozer and piledriver wernsaar 2014-07-19 15:48:07 +0200
  • b3938fe371 don't use this sgemv_n on Windows wernsaar 2014-07-19 07:15:34 +0200
  • e6668dd83b Merge pull request #414 from staticfloat/sf/symlinkfix Zhang Xianyi 2014-07-18 23:13:18 +0800
  • c8a4a56177 performance optimizations for sgemv_n wernsaar 2014-07-18 11:25:21 +0200
  • 3c5732615d added blocked sgemv_n and microkernel for bulldozer and piledriver wernsaar 2014-07-17 23:15:07 +0200
  • f20c0f9819 Merge branch 'develop' v0.2.10 Zhang Xianyi 2014-07-17 15:15:57 +0800
  • 134fa320e6 Refs #415. Fixed the x86/i386 compiling bug with DYNAMIC_ARCH=1. Zhang Xianyi 2014-07-17 15:02:01 +0800
  • a79df1ff49 Don't create an absolute symlink when installing on Darwin Elliot Saba 2014-07-16 15:31:27 -0400
  • 7ceb25d7b3 changed string GFORTRAN to lowercase wernsaar 2014-07-16 17:08:43 +0200
  • 21b5347fbe Merge branch 'develop' Zhang Xianyi 2014-07-16 18:04:30 +0800
  • f2eb480738 OpenBLAS 0.2.10 version. Zhang Xianyi 2014-07-16 18:04:18 +0800
  • c94762bb56 Refs #401. Added NO_AVX2 flag for old binutils (e.g. RHEL6) Zhang Xianyi 2014-07-16 08:37:55 +0800
  • 51413925bd adjust number of threads for small size in cgemv and zgemv wernsaar 2014-07-15 16:27:02 +0200
  • b985cea65d adjust number of threads for sgemv and dgemv wernsaar 2014-07-15 16:04:46 +0200
  • d286daa2ba adjusted number of threads for small size wernsaar 2014-07-15 14:41:35 +0200
  • bcb115b55b added benchmark for gemv wernsaar 2014-07-15 13:35:36 +0200
  • 3dd094f17a Merge pull request #413 from wernsaar/develop Zhang Xianyi 2014-07-14 22:39:22 +0800
  • 339ab34c4c added additional test value to dstest.in wernsaar 2014-07-13 18:29:19 +0200
  • 7424e2b609 added additional test value wernsaar 2014-07-13 18:26:38 +0200
  • 73594cff73 segment violation in x86_64 sgemv kernels wernsaar 2014-07-13 10:49:43 +0200
  • 880597b301 segment violation in sgemv kernels wernsaar 2014-07-13 10:46:14 +0200
  • 9c835431d0 modified pathes to atlas, mkl and acml wernsaar 2014-07-12 16:20:29 +0200
  • 1d4ffddf69 added conf option for number of loops wernsaar 2014-07-12 11:54:39 +0200
  • b0e7810a6b added her2k benchmark wernsaar 2014-07-11 16:31:05 +0200
  • 2b92a8c499 added herk benchmark wernsaar 2014-07-11 16:16:48 +0200
  • 274b8dc91a add hemm benchmark wernsaar 2014-07-11 15:26:34 +0200
  • 74b237ca22 added syr2k benchmark wernsaar 2014-07-11 14:48:25 +0200
  • c353abd38c added syrk benchmark wernsaar 2014-07-11 14:21:25 +0200
  • 0acce17979 added trsm benchmark wernsaar 2014-07-11 13:51:08 +0200
  • 2016a685e6 added trmm benchmark wernsaar 2014-07-11 13:20:42 +0200
  • 1b9a6aac30 added benchmark for symm wernsaar 2014-07-11 12:47:48 +0200
  • e27433ab6a added gemm benchmark and modified Makefile for benchmark wernsaar 2014-07-11 11:09:47 +0200
  • 7961404a40 Merge pull request #411 from wernsaar/develop Zhang Xianyi 2014-07-10 22:38:15 +0800
  • cedc1f4b14 Ref #410: disabled optimized potri functions ( single threading bug) wernsaar 2014-07-10 13:42:32 +0200
  • 0884b73c69 Lapack-test Windows 32bit now error free wernsaar 2014-07-10 11:01:47 +0200
  • 9bd9472ae9 Lapack-test: cleanup of x86 32bit KERNEL file wernsaar 2014-07-09 16:08:19 +0200
  • 2e2473f390 Merge pull request #409 from wernsaar/develop Zhang Xianyi 2014-07-09 21:11:00 +0800
  • c4a423a642 bugfixes for lapack on ARM Platform wernsaar 2014-07-09 12:21:39 +0200
  • f9991fd5f6 Merge branch 'develop' v0.2.10.rc2 Zhang Xianyi 2014-07-09 08:48:00 +0800
  • 47688e24e9 OpenBLAS 0.2.10 rc2 version. Zhang Xianyi 2014-07-09 08:47:36 +0800
  • 61ef0c3419 added cross compiler examples for 32bit and 64bit ARM wernsaar 2014-07-08 12:55:18 +0200
  • 698e77dba4 Refs #406. Fixed utest building bug. Zhang Xianyi 2014-07-08 17:26:49 +0800
  • 2081f6e8ff Lapack bug114: replaced cgesvd.f and zgesvd.f wernsaar 2014-07-08 10:21:10 +0200
  • dc6b809f15 Lapack bug117: replaced zstemr.f wernsaar 2014-07-08 10:08:34 +0200
  • 0f08684649 Lapack bug118: replaced clanhf.f and zlanhf.f wernsaar 2014-07-08 09:57:40 +0200
  • 552119c484 Fixed #407. Support outputing the CPU corename on runtime. The user can use char * openblas_get_config() or char * openblas_get_corename(). Zhang Xianyi 2014-07-08 12:48:08 +0800
  • a8e51bcc44 Merge 9bca2bed63 into 94d3cfaa10 Tony Kelman 2014-07-07 22:20:42 +0000
  • 9bca2bed63 add gotoblas_corename to dll exports list Tony Kelman 2014-07-07 15:20:17 -0700
  • 94d3cfaa10 Merge pull request #404 from wernsaar/develop Zhang Xianyi 2014-07-07 00:39:33 +0800
  • 13348b2137 removed reference to daxpy_bulldozer kernel (Windows bug in lapack-test) wernsaar 2014-07-06 16:39:32 +0200
  • 783a7d2202 bugfix for fortran compiler wernsaar 2014-07-06 13:33:42 +0200
  • 50e99a52ea added definitions for PILEDRIVER and HASWELL wernsaar 2014-07-06 12:08:27 +0200
  • 9964ed2f79 bugfix for CORE2 wernsaar 2014-07-06 11:47:28 +0200
  • d5b976f92d fallback to zgemm_kernel_4x2_sse.S wernsaar 2014-07-06 11:05:28 +0200
  • f7267d9b0e added missing definition for DUNNINGTON wernsaar 2014-07-06 10:17:07 +0200
  • e0c080a28c removed reference to zgemm_kernel_4x2_sse3.S (bug in lapack-test) wernsaar 2014-07-05 16:13:17 +0200
  • e80b144932 enabled compiling of *3M functions wernsaar 2014-07-02 14:11:53 +0200
  • 02a504c0b8 fixed my bug in ger.c wernsaar 2014-07-02 10:39:33 +0200
  • be94db096c disabled *3M functions for x86_64 platforms wernsaar 2014-07-01 16:18:05 +0200
  • b079df9ef4 added optimized sdot- and dsdot-kernel, written in C wernsaar 2014-06-30 14:46:38 +0200
  • aee61456a4 disabled SMP for sbmv and zsbmv again wernsaar 2014-06-29 21:18:38 +0200
  • 01a119abfc enabled SMP for sbmv and zsbmv, but only for 64bit binaries wernsaar 2014-06-29 20:35:56 +0200
  • 1fad2b759f enabled smp for ger.c and zger.c, but only for 64bit binaries wernsaar 2014-06-29 16:43:04 +0200
  • e1e83a1b71 modification, to run blas-test on Windows wernsaar 2014-06-29 10:15:29 +0200
  • da3d70420a Merge branch 'develop' v0.2.10.rc1 Zhang Xianyi 2014-06-29 10:46:22 +0800
  • 1127f5a2d7 OpenBLAS 0.2.10 rc1 version. Zhang Xianyi 2014-06-29 10:45:50 +0800
  • 0ae4cc2803 Merge branch 'wernsaar-develop' into develop Zhang Xianyi 2014-06-29 10:40:54 +0800
  • 99efbbbad5 Fixed #395. Enable optimized cgemm for Sandybridge. Added optimized sdot kernel. Fixed c/zgemm, zgemv computational error of haswell, piledriver, bullldozer, and barcelona on Windows. Zhang Xianyi 2014-06-29 10:34:51 +0800
  • 22e5aee2dd fixed zgemv bug for older AMD Processors wernsaar 2014-06-28 19:04:49 +0200
  • 249917700d Merge branch 'TimothyGu-develop' into develop Fixed #398. Remove all trailing whitespace except lapack-netlib. Zhang Xianyi 2014-06-28 20:52:07 +0800
  • 7a8949e0ce Merge branch 'develop' of https://github.com/TimothyGu/OpenBLAS into TimothyGu-develop Zhang Xianyi 2014-06-28 20:51:31 +0800
  • b82108f899 Merge pull request #399 from TimothyGu/upstr Zhang Xianyi 2014-06-28 20:40:23 +0800
  • 8373ad4ec2 Merge pull request #397 from vtjnash/develop Zhang Xianyi 2014-06-28 20:38:48 +0800
  • 35d37e124f bugfix for barcelona zgemv-kernel wernsaar 2014-06-28 12:36:11 +0200
  • d8ba46efdb bugfix for bulldozer cgemm-, zgemm- and zgemv-kernel wernsaar 2014-06-28 12:16:20 +0200
  • a15f22a1f6 bugfix for piledriver cgemm-, zgemm- and zgemv-kernel wernsaar 2014-06-28 11:46:58 +0200
  • b94ea89f52 bugfix for haswell cgemm- and zgemm-kernel wernsaar 2014-06-28 10:22:40 +0200
  • 35f668bb14 bugfix for cgemm_kernel_8x2_sandy.S wernsaar 2014-06-28 10:01:56 +0200
  • 4ebbf758f5 .gitignore: add some more entries concerned with kernel Timothy Gu 2014-06-27 13:58:42 -0700
  • 8615d6ec87 Build import libs as .dll.a instead of .lib Timothy Gu 2014-06-27 11:58:14 -0700
  • 6c2ead30f0 Remove all trailing whitespace except lapack-netlib Timothy Gu 2014-06-27 12:05:18 -0700
  • f41f03ab83 fix #394. this cleans up some handles after using them, and doesn't disable ALL process privileges upon success Jameson Nash 2014-06-27 12:10:04 -0400
  • 365e8de346 added optimized cgemm-kernel for SANDYBRIDGE wernsaar 2014-06-27 13:40:29 +0200
  • 578d1b6219 added DSDOT definition and enabled optimized sdot kernel wernsaar 2014-06-27 11:30:29 +0200
  • a6ae079b17 added blas-test from lapack wernsaar 2014-06-27 10:12:19 +0200
  • d10db52edb Merge pull request #390 from wernsaar/develop Zhang Xianyi 2014-06-27 14:57:06 +0800
  • dabab2b5f4 added new optimized sgemm kernel for SANDYBRIGE wernsaar 2014-06-26 21:42:08 +0200
  • aa2709c4e0 enabled optimized dgemm kernel for NEHALEM wernsaar 2014-06-26 12:22:29 +0200
  • 9d6f2b594e Fortran flag -frecursive is disabled by default wernsaar 2014-06-25 13:55:19 +0200
  • a13bcc1716 enabled optimized sgemv kernel for barcelona and piledriver wernsaar 2014-06-25 13:50:57 +0200
  • d2c82d7543 enabled optimized sgemv kernel for HASWELL wernsaar 2014-06-25 12:56:45 +0200