Commit Graph

4456 Commits

Author SHA1 Message Date
Martin Kroeker 63994e1cdb
Merge pull request #2405 from susilehtola/patch-1
Fix typo in dynamic_zarch.c
2020-02-11 13:03:35 +01:00
Martin Kroeker 496e3019bc
Merge pull request #2404 from martin-frbg/issue2395
Fix spurious application of USE_TRMM in cmake builds
2020-02-11 13:00:36 +01:00
Martin Kroeker 169be3f097
Merge pull request #2403 from martin-frbg/issue2400
Fix coretype identification of Intel Cannon Lake, Ice Lake and Goldmont
2020-02-11 13:00:16 +01:00
Martin Kroeker 6ccbb089c2
Merge pull request #2402 from gxw-loongson/develop
Avoid printing the following information on mips and mips64 when check msa
2020-02-11 12:59:53 +01:00
Martin Kroeker 59ebe3636a
Merge pull request #2399 from martin-frbg/buffersize
Make BUFFER_SIZE configurable at build time
2020-02-11 12:56:56 +01:00
Susi Lehtola 5a6bba3061
Patch out instances of Z15 in dynamic_zarch.c
There does not appear to be a Z15 kernel yet, causing link errors from the code. This patch fixes the issue.
2020-02-11 15:07:33 +13:00
Susi Lehtola dff173e50e
Fix typo in dynamic_zarch.c 2020-02-11 14:46:30 +13:00
Martin Kroeker 7e5cbb6f35
Fix bad conditional syntax that caused spurious application of USE_TRMM 2020-02-10 21:17:39 +01:00
Martin Kroeker 303bdb673b
Fix coretype detection for Intel extended models 6 and 7
affecting Goldmont, Cannon Lake, Ice Lake autodetection
2020-02-10 19:17:32 +01:00
gxw 754433f420 Avoid printing the following information on mips and mips64 when check msa:
"unrecognized command line option ‘-mmsa’"
2020-02-10 19:11:45 +08:00
Martin Kroeker 7f0d523b42 Make BUFFER_SIZE configurable 2020-02-09 23:32:57 +01:00
Martin Kroeker c353d8b106
Make BUFFER_SIZE configurable 2020-02-09 23:30:22 +01:00
Martin Kroeker 579be3aa9d Add configuration option for BUFFER_SIZE 2020-02-09 23:28:04 +01:00
Martin Kroeker 449e8ea443
Merge pull request #26 from xianyi/develop
rebase
2020-02-09 23:23:55 +01:00
Martin Kroeker 3bec250cf9
Increment version to 0.3.9.dev 2020-02-09 23:18:44 +01:00
Martin Kroeker f03dd23e90 Increment version to 0.3.9.dev 2020-02-09 23:18:07 +01:00
Martin Kroeker fa93d63365 Merge branch 'release-0.3.0' into develop 2020-02-09 23:16:06 +01:00
Martin Kroeker 90e6c66a57
Merge pull request #2397 from martin-frbg/038changes
Update Changelog with changes from 0.3.8
2020-02-09 23:01:52 +01:00
Martin Kroeker 32d97330b3 Update with changes from 0.3.8 2020-02-09 23:00:36 +01:00
Martin Kroeker 29eaf4b6d7
Merge pull request #25 from xianyi/develop
rebase
2020-02-09 22:48:15 +01:00
Martin Kroeker 47c1bf7f4d typo fixes 2020-02-09 01:06:40 +01:00
Martin Kroeker 2b55f0ad30
Merge pull request #2393 from martin-frbg/issue2388
Provide more documentation in README.md
2020-02-09 01:00:33 +01:00
Martin Kroeker a5b32ab06c
Merge pull request #2390 from martin-frbg/pgi
Small corrections for compilation with PGI compilers
2020-02-09 00:13:40 +01:00
Martin Kroeker 50545b19d0
Update CPU and OS support and document DYNAMIC_ARCH option in README.md
prompted by #2388
2020-02-09 00:06:07 +01:00
Martin Kroeker b3cbd60d7a
Remove PGI from list again as it is actually still not capable 2020-02-08 10:20:13 +01:00
Martin Kroeker 70199d1905
Merge pull request #2389 from Zeyiii/develop
Fix bugs in benchmark of gemv
2020-02-07 16:05:46 +01:00
Martin Kroeker cfe63d8cc2
Remove OpenMP libraries from link list 2020-02-07 16:03:51 +01:00
Martin Kroeker d55b10830f
Remove OpenMP libraries from link list 2020-02-07 16:02:17 +01:00
Martin Kroeker c1c10cbb21
Merge pull request #2384 from wjc404/develop
Optimize AVX512 DGEMM (& DTRMM)
2020-02-07 13:47:12 +01:00
Martin Kroeker 5989841524
Add PGI to avx512-supporting compilers 2020-02-07 13:01:31 +01:00
Martin Kroeker 68a43db358
Fix utest compilation with PGI 2020-02-07 10:15:18 +01:00
Martin Kroeker 9694037b23
Set SUFFIX in tempfile commands, fix bad architecture option for PGI compiler in avx512 test 2020-02-07 10:09:25 +01:00
Martin Kroeker 71faa1c1a7
Merge pull request #24 from xianyi/develop
rebase
2020-02-07 10:03:02 +01:00
wjc404 3447d04eaf
Update dgemm_kernel_16x2_skylakex.c 2020-02-06 02:14:10 +00:00
wjc404 8b5cdcc64c
Update sgemm_kernel_8x4_haswell.c 2020-02-06 01:47:46 +00:00
wjc404 4e00d96a78
Update dgemm_kernel_16x2_skylakex.c 2020-02-06 01:46:36 +00:00
w00421467 ce9ea8f826 Fix another branch 2020-02-05 15:07:18 +08:00
w00421467 0b909203cb Fix bugs in benchmark of gemv 2020-02-05 14:53:37 +08:00
wjc404 096da2f51a
Update dgemm_kernel_16x2_skylakex.c 2020-02-05 13:36:57 +08:00
wjc404 2f96a2c55b
Update trmm_R.c 2020-02-05 10:15:02 +08:00
wjc404 833bd0f8ff
Update trmm_L.c 2020-02-05 10:09:41 +08:00
wjc404 77b8f49556
Update level3_thread.c 2020-02-04 20:33:08 +08:00
wjc404 1c3e20ce48
Update level3.c 2020-02-04 20:30:23 +08:00
wjc404 83b6be7976
Update param.h 2020-02-04 19:55:26 +08:00
wjc404 081b188529
Update KERNEL.SKYLAKEX 2020-02-03 21:38:08 +08:00
wjc404 f3f969f681
Update param.h 2020-02-03 21:34:12 +08:00
wjc404 8019e70211
AVX512 16x2 DGEMM kernel 2020-02-03 21:32:56 +08:00
Martin Kroeker 8d2a796f49
Merge pull request #2378 from martin-frbg/issue2377
Add -march option for AVX512 in cmake as well
2020-01-30 17:07:19 +01:00
Martin Kroeker 8dc9fd4dfe
Add -march option for AVX512 2020-01-30 12:41:18 +01:00
Martin Kroeker abc67bdd74
Merge pull request #2375 from ewanglong/master
fix a few performance drop in some matrix size per data type
2020-01-30 10:27:29 +01:00