diff options
-rw-r--r-- | CHANGELOG | 5 | ||||
-rw-r--r-- | doc/tuning.md | 6 |
2 files changed, 11 insertions, 0 deletions
@@ -1,3 +1,8 @@ +Development version (next version) +- Fix a correctness issue with DGEMM on SM 7.5 Turing GPUs +- Various minor fixes and enhancements +- Added tuned parameters for various devices (see doc/tuning.md) + Version 1.5.2 - Changed XAMAX/XAMIN to more likely return first rather than last min/max index, updated API docs - Added batched routines to pyclblast diff --git a/doc/tuning.md b/doc/tuning.md index 793bce88..200cb7cc 100644 --- a/doc/tuning.md +++ b/doc/tuning.md @@ -34,6 +34,11 @@ The CLBlast library is already tuned for the most commonly used OpenCL devices a - Tesla K20m - Tesla K40m - Tesla P100 16GB + - Tesla T4 + - Tesla V100 + - Quadro T2000 + - Quadro GV100 + - A100 * AMD GPUs: - Radeon HD 6750M - Radeon HD 6770M @@ -66,6 +71,7 @@ The CLBlast library is already tuned for the most commonly used OpenCL devices a - Core i7-4790K - Core i7-5930K - Core i7-6770HQ + - Core i9-9980HK - Xeon E5-2630 v3 - Xeon E5-2630 v4 * Other devices: |