584 Commits (45fdf951b64aa9145996727ecda901f00a2eda3c)

Author SHA1 Message Date
  Wangyang Guo 8356a604f0 sbgemm: cooperlake: tuning for block params 4 years ago
  Martin Kroeker cd10d1c03b
Fix typo 4 years ago
  Martin Kroeker 2db1a99aca
Clean up debug messages 4 years ago
  Martin Kroeker 89fc5b8f4f
Fix unmap logic 4 years ago
  Martin Kroeker 7fd12a5e69
Add likely() hints for gcc 4 years ago
  Martin Kroeker 2ba9a567aa
Fix typo 4 years ago
  Martin Kroeker b4b952eece
Add auxiliary tracking space for thread buffer frees too 4 years ago
  Martin Kroeker 7d1becc575
Allocate an auxiliary struct when running out of preconfigured threads 4 years ago
  Martin Kroeker 898212efcd
Actually add the message to the TLS section 4 years ago
  Martin Kroeker 210a1584c5
Rebase source and edit TLS version of the message as well 4 years ago
  Martin Kroeker f2a7a67f5a
Improve the "tried to allocate too many buffers" error message 4 years ago
  Craig Watson 4d7dfe4845 Include Haiku in processor count checks 4 years ago
  JonasZhou 0fca36c8c3 Add cpu detection support for Zhaoxin processors 4 years ago
  River Dillon 2f6326a630 Remove <linux/unistd.h> 4 years ago
  Martin Kroeker 8f22ac552b
Add vendor string Shanghai as successor to Centaur 4 years ago
  Martin Kroeker eb2fdd3af0
Recognize newer Zhaoxin/Centaur processors as Nehalem 4 years ago
  User User-User 750719528a bugz 4 years ago
  User User-User 6423b282a1 dynamic_arch 4 years ago
  Martin Kroeker 307c4c0786
Fix typo 4 years ago
  Martin Kroeker e83df93975
Work around another recent macro name collision with winnt.h 4 years ago
  Martin Kroeker cbfd3c87e1
Recognize Intel Ice Lake SP as Cooper Lake 4 years ago
  Martin Kroeker 623d580b4c
Restore __volatile__ keyword 4 years ago
  Martin Kroeker 186368ddc3
Fix compilation with CLANG 4 years ago
  Martin Kroeker 1a3ad4b670
Fix signatures of the TLS-mode dll_callback and p_process_term functions for Win64 5 years ago
  Peter Hawkins dbbf92c1d1 Fix race in blas_thread_shutdown. 5 years ago
  Martin Kroeker cb429d6b12
Merge pull request #3110 from martin-frbg/issue3108 5 years ago
  Martin Kroeker b0bded3f2f
Fix get_num_procs() in the USE_TLS branch for non-glibc systems 5 years ago
  Martin Kroeker e4e5042e38
Recognize Intel Tiger Lake as SkylakeX 5 years ago
  Martin Kroeker 0cc36770f1
Merge pull request #3073 from xoviat/embedded 5 years ago
  Martin Kroeker eea0c0f2ed
Merge pull request #3085 from alexhenrie/memory_alloc 5 years ago
  Martin Kroeker 0cb9e9fc8d
Remove the VORTEX support bits again for now 5 years ago
  Alex Henrie 113840da12 Fix null pointer check in blas_memory_alloc 5 years ago
  Martin Kroeker deb2e66bcc
Add DYNAMIC_LIST support for ARM64 5 years ago
  xoviat 2e8d6e8690 add functions for embedded 5 years ago
  Martin Kroeker b94dab5250
patch to support power10 in builtin_cpu_is was backported to gcc 10.2, so allow that as wel 5 years ago
  Martin Kroeker 63fa3c3f8f
Require gcc 11 for builtin_cpu_is(power10) 5 years ago
  xoviat b60de4447a add cortex-m platform 5 years ago
  Martin Kroeker 2c445be8ba
Merge pull request #3051 from martin-frbg/rocketlake 5 years ago
  Martin Kroeker 6fe0f1fab9
Label get_cpu_ftr as volatile to keep gcc from rearranging the code 5 years ago
  Martin Kroeker 17c16f2a71
Implement builtin_cpu_is and limit cpu choices to P8 and P9 for NVIDIA compilers 5 years ago
  Martin Kroeker 865676682d
Add Intel Rocket Lake 5 years ago
  Martin Kroeker 6232237dba
Make fallback from P10 to P9 conditional on suitable compiler 5 years ago
  Martin Kroeker 18d8a67485
Merge pull request #2994 from antonblanchard/power10-fixes 5 years ago
  Martin Kroeker 83de62c20d
Merge pull request #3026 from martin-frbg/revert747 5 years ago
  gxw 4b548857d6 Add msa support for loongson 5 years ago
  Martin Kroeker a554712439
remove extra/intermediate size step for min_jj introduced in PR747 5 years ago
  Martin Kroeker 5d26223f4a
remove extra/intermediate size step of min_jj from PR747 5 years ago
  Martin Kroeker bc5b1ddf0d
Merge pull request #3004 from martin-frbg/bsd_getauxval 5 years ago
  Martin Kroeker e7bf8ced6c
Build fix for systems that do not support getauxval 5 years ago
  Martin Kroeker 5fa305172a
Use ifeq instead of ifdef for user-definable options 5 years ago