Commit Graph

  • 6a13a94e71 added gesv benchmark Werner Saar 2015-06-02 13:35:49 +0200
  • eff43d3289 Merge pull request #585 from wernsaar/develop wernsaar 2015-05-31 15:01:54 +0200
  • 9c4817d07b bugfix for Makefile on mac Werner Saar 2015-05-31 14:16:51 +0200
  • 319f3a0451 Merge pull request #584 from wernsaar/develop wernsaar 2015-05-29 13:27:20 +0200
  • 02c7766f68 bugfixes, to build benchmarks with mingw on Windows OS Werner Saar 2015-05-29 12:56:22 +0200
  • f38cb67ca8 Merge pull request #581 from wernsaar/develop wernsaar 2015-05-23 12:58:15 +0200
  • eea2e30b74 bugfix for arm locking Werner Saar 2015-05-23 11:40:40 +0200
  • 19b8fd2aed smp lock bugfix Werner Saar 2015-05-23 10:58:38 +0200
  • 0cc5212741 Merge pull request #580 from wernsaar/develop wernsaar 2015-05-23 09:46:39 +0200
  • c47c8e8cf5 added blas level1 swap benchmark Werner Saar 2015-05-21 08:51:42 +0200
  • a11555c715 Support Android NDK armeabi-v7a-hard ABI. (-mfloat-abi=hard) Zhang Xianyi 2015-05-20 21:57:27 -0500
  • 897d03518e Merge pull request #578 from wernsaar/develop wernsaar 2015-05-20 11:56:02 +0200
  • 23fbc5728e added blas level1 copy benchmark Werner Saar 2015-05-20 11:05:00 +0200
  • 6d40fa587f Fix f_check bug. Zhang Xianyi 2015-05-19 12:04:45 -0500
  • 22dcd79959 Merge pull request #577 from wernsaar/develop wernsaar 2015-05-19 10:59:24 +0200
  • ea4df0aad3 Ref #574: Bugfix for armv6 memory barrier Werner Saar 2015-05-19 10:43:12 +0200
  • e127fb8fd8 1) Refs #575. Remove g77 from compiler list. 2) If OpenBLAS cannot find Fortran compiler, it will only build BLAS (without LAPACK). Zhang Xianyi 2015-05-19 00:01:04 -0500
  • 8c05970767 Merge 4a20fec92d into 7fb718a7d8 Ivan Ukhov 2015-05-18 20:41:27 +0000
  • 4a20fec92d Make gfortran the first to look for Ivan Ukhov 2015-05-18 16:39:15 -0400
  • 7fb718a7d8 Merge pull request #572 from wernsaar/develop wernsaar 2015-05-18 13:47:38 +0200
  • 24f58c8bb1 added optimized cscal and zscal kernels for steamroller Werner Saar 2015-05-18 12:40:07 +0200
  • 95b1faf667 added optimized cscal and zscal kernels for steamroller and piledriver Werner Saar 2015-05-18 10:50:57 +0200
  • 2d9e406050 added optimized cscal kernel for sandybridge Werner Saar 2015-05-18 08:46:06 +0200
  • 59083e3ce1 added optimized cscal kernel for bulldozer Werner Saar 2015-05-18 07:33:52 +0200
  • 685be40339 Merge pull request #571 from wernsaar/develop wernsaar 2015-05-17 14:09:14 +0200
  • 31c9e399e9 added optimized cscal kernel for haswell Werner Saar 2015-05-17 13:44:09 +0200
  • 7de6bb9889 added optimized zscal kernel for bulldozer Werner Saar 2015-05-17 11:45:19 +0200
  • d63034303b added optimized zscal kernel for haswell Werner Saar 2015-05-16 16:41:45 +0200
  • 51ff17d46e Add AMD Excavator target. Zhang Xianyi 2015-05-13 16:16:30 -0500
  • 905534942a Merge pull request #568 from wernsaar/develop wernsaar 2015-05-13 13:48:08 +0200
  • 18e90ee2e3 bugfix: added static to functions Werner Saar 2015-05-13 13:31:26 +0200
  • e00cccc41e added optimized dscal kernel for piledriver Werner Saar 2015-05-13 13:05:35 +0200
  • 73f09bf64f optimized dscal kernel for increment != 1 Werner Saar 2015-05-13 12:14:39 +0200
  • 02e772c7e4 added optimized dscal kernel for haswell Werner Saar 2015-05-12 17:19:58 +0200
  • 7aee913991 added optimized dscal kernel for sandybridge Werner Saar 2015-05-12 16:27:43 +0200
  • e50a933037 added optimized dscal kernel for bulldozer Werner Saar 2015-05-12 12:28:44 +0200
  • 5f9011d6ef Merge pull request #566 from powderluv/develop Zhang Xianyi 2015-05-11 20:59:12 -0500
  • ebb9eba987 Fix build with ALLOC_SHM=0 (Android NDK) powderluv 2015-05-10 00:10:26 -0700
  • 8e5a1083bb Refs #532. Improve gemv paralel with small m and large n case. Zhang Xianyi 2015-05-08 05:33:17 +0800
  • 6743beb748 Refs #565. Fix the bug of generate FEXTRALIB. Zhang Xianyi 2015-05-07 13:06:53 +0800
  • bcabf72c08 Refs #565. Merge branch 'andreasnoack-anj/bench' into develop Zhang Xianyi 2015-05-07 12:52:14 +0800
  • cda29f183b Add vecLib benchmarks Andreas Noack 2015-05-06 21:52:34 -0400
  • e52d36450a Merge pull request #564 from wernsaar/develop wernsaar 2015-05-06 11:10:31 +0200
  • f8f2e261fe use only 1 thread if m or n < 2*GEMM_MULTITHREAD_THRESHOLD Werner Saar 2015-05-06 10:41:53 +0200
  • be3c843700 added loops to trsm.c Werner Saar 2015-05-06 09:21:19 +0200
  • e6f57db846 Merge pull request #563 from wernsaar/develop wernsaar 2015-05-05 12:13:35 +0200
  • 9bfd267d51 bugfix for gemm3m tests Werner Saar 2015-05-05 11:58:59 +0200
  • 924bc5372e removed gemm3m functions from normal checks Werner Saar 2015-05-05 11:39:43 +0200
  • 2b83a69650 Merge pull request #561 from wernsaar/develop wernsaar 2015-05-04 11:11:13 +0200
  • 133c11a156 updated dgemv_n kernel for nehalem Werner Saar 2015-04-30 14:38:06 +0200
  • 30f52d53df optimized dgemv_n kernel for haswell Werner Saar 2015-04-30 12:11:39 +0200
  • a124637329 Merge pull request #560 from sebastien-villemot/develop Zhang Xianyi 2015-04-29 11:36:47 -0500
  • 642aaba2e0 Fix detection of ARM architectures in c_check. Sébastien Villemot 2015-04-29 18:14:21 +0200
  • 4c616173e4 Merge pull request #558 from wernsaar/develop wernsaar 2015-04-28 17:30:16 +0200
  • 5e83d80725 optimized dger kernel for sandybridge Werner Saar 2015-04-28 16:58:11 +0200
  • b2e1797dc6 added optimized sger kernel for sandybridge Werner Saar 2015-04-28 15:33:38 +0200
  • e216f686cb optimized saxpy and daxpy for sandybridge Werner Saar 2015-04-28 10:18:32 +0200
  • e42652f772 Merge pull request #554 from wernsaar/develop Zhang Xianyi 2015-04-25 08:11:36 -0500
  • e77db2af31 add benchmarks for zgeru and cgeru Werner Saar 2015-04-25 14:53:07 +0200
  • 37b00841ac Merge pull request #552 from jeromerobert/develop Zhang Xianyi 2015-04-24 14:12:12 -0500
  • fc0e0391f3 bugfixes: replaced int with BLASLONG Werner Saar 2015-04-24 14:30:44 +0200
  • da0f27b9ac Merge pull request #553 from wernsaar/develop wernsaar 2015-04-24 13:57:48 +0200
  • c22068c406 optimized sdot.c for increments != 1 Werner Saar 2015-04-24 13:13:20 +0200
  • dee100d0e4 optimized saxpy.c for increments != 1 Werner Saar 2015-04-24 11:52:59 +0200
  • 0273966abb optimized daxpy kernel for increments != 1 Werner Saar 2015-04-24 11:39:17 +0200
  • 3a67daa954 optimized ddot.c for increments != 1 Werner Saar 2015-04-24 10:56:55 +0200
  • ab567d8443 gemv: Ensure stack buffer is large enough to handle memory alignment Jerome Robert 2015-04-21 10:12:01 +0200
  • 3c09cea4b2 Merge pull request #550 from wernsaar/develop wernsaar 2015-04-23 13:27:38 +0200
  • b4f2153dcd added optimized ssymv kernels for sandybridge Werner Saar 2015-04-23 12:19:24 +0200
  • 1c4b0eeae3 added optimized ssymv kernels for haswell Werner Saar 2015-04-23 10:23:13 +0200
  • 406d9d64e9 Merge pull request #549 from wernsaar/develop wernsaar 2015-04-22 12:36:13 +0200
  • 1bec9abb9a added optimized dsymv kernels for sandybridge Werner Saar 2015-04-22 12:09:43 +0200
  • 3814bf60d3 added optimized dsymv kernels for haswell Werner Saar 2015-04-22 10:42:50 +0200
  • 847e19c04e Refs #478,#482, Enable stack alloc for s/dgemv_t.(revert 9798491) Zhang Xianyi 2015-04-20 23:22:40 -0500
  • 46c7b4d5c8 added asum benchmark Werner Saar 2015-04-19 11:24:07 +0200
  • 8e05d291b5 added scal benchmark Werner Saar 2015-04-18 08:41:41 +0200
  • 9da555e5f7 Merge pull request #546 from wernsaar/develop wernsaar 2015-04-16 11:36:51 +0200
  • 6d0db0151f added optimized zaxpy-kernels Werner Saar 2015-04-16 11:19:37 +0200
  • 37b9033c90 Merge pull request #543 from jeromerobert/develop Zhang Xianyi 2015-04-15 11:18:14 -0500
  • 59e7a518c6 Merge pull request #544 from wernsaar/develop wernsaar 2015-04-15 17:04:02 +0200
  • 13889515b3 added optimized caxpy-kernel for sandybridge Werner Saar 2015-04-15 16:29:25 +0200
  • 248c9340c3 added optimized caxpy-kernel for haswell Werner Saar 2015-04-15 15:16:31 +0200
  • e9f33b4ca7 added optimized caxpy-kernel for steamroller Werner Saar 2015-04-15 13:49:23 +0200
  • f5d847122a updated caxpy_microk_bulldozer-2.c and caxpy.c Werner Saar 2015-04-15 11:59:38 +0200
  • a4c96eca67 Fix a buffer overflow with MAX_STACK_ALLOC size in dgemv_t Jerome Robert 2015-04-15 09:41:45 +0200
  • fb02cb0a41 Merge pull request #540 from wernsaar/develop wernsaar 2015-04-14 15:53:09 +0200
  • baa0363ea2 add optimized ddot-kernel for piledriver Werner Saar 2015-04-14 15:09:13 +0200
  • 34ba66606a add optimized daxpy-kernel for piledriver Werner Saar 2015-04-14 14:23:29 +0200
  • f615dc7603 added optimized saxpy kernel for steamroller Werner Saar 2015-04-14 09:09:39 +0200
  • 331c417637 optimized saxpy for piledriver Werner Saar 2015-04-14 08:34:11 +0200
  • 6c3a0b5d46 Enable MAX_STACK_ALLOC by default. Zhang Xianyi 2015-04-13 23:23:40 -0500
  • fd9fd42936 Refs #478, #482. Fixed bug on previous commit. Zhang Xianyi 2015-04-13 23:22:27 -0500
  • 9798481979 Refs #478, #482. Fix segfault bug for gemv_t with MAX_ALLOC_STACK flag. Zhang Xianyi 2015-04-13 19:45:27 -0500
  • d7a17ad85d optimized sdot-kernel for pilediver Werner Saar 2015-04-13 13:19:21 +0200
  • d35f6c63c2 add optimized daxpy-kernel for steamroller Werner Saar 2015-04-13 12:22:43 +0200
  • 166d76e864 added optimized sdot-kernel for steamroller Werner Saar 2015-04-11 08:48:18 +0200
  • f9f127d838 added optimized ddot kernel for steamroller Werner Saar 2015-04-10 16:18:03 +0200
  • 62231ab337 Merge pull request #538 from wernsaar/develop wernsaar 2015-04-10 16:03:37 +0200
  • 3119def9a7 updated cdot and zdot Werner Saar 2015-04-10 11:10:31 +0200
  • 33b332372a add optimized cdot- and zdot-kernel for sandybridge Werner Saar 2015-04-10 09:37:26 +0200