Skip to content

Releases: ROCm/hipBLASLt

hipBLASLt 0.12.1 for ROCm 6.4.1

20 May 13:15
4d62e13
Compare
Choose a tag to compare

Resolved issues

  • Fixed an accuracy issue that occurred for some solutions using an FP32 or TF32 data type with a TT transpose.

hipBLASLt 0.12.0 for ROCm 6.4.0

11 Apr 13:34
Compare
Choose a tag to compare

Added

  • Support roctx if HIPBLASLT_ENABLE_MARKER=1 is set
  • Output the profile logging if HIPBLASLT_LOG_MASK=64 is set
  • Support FP16 compute type
  • Add memory bandwidth information in hipblaslt-bench output
  • Support user offline tuning mechanism
  • Add more samples

Changed

  • Output the bench command along with solution index if HIPBLASLT_LOG_MASK=32 is set

Optimized

  • Improve the overall performance of XF32/FP16/BF16/FP8/BF8 data type
  • Reduce library size

Resolved issues

  • Fix multi-threads bug
  • Fix multi-streams bug

hipBLASLt 0.10.0 for ROCm 6.3.3

19 Feb 17:47
b4e5042
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.3.3 did not change. The library was rebuilt for the updated ROCm 6.3.3 stack.

hipBLASLt 0.10.0 for ROCm 6.3.2

28 Jan 15:43
b4e5042
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.3.2 did not change. The library was rebuilt for the updated ROCm 6.3.2 stack.

hipBLASLt 0.10.0 for ROCm 6.3.1

20 Dec 16:12
b4e5042
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.3.1 did not change. The library was rebuilt for the updated ROCm 6.3.1 stack.

hipBLASLt 0.10.0 for ROCm 6.3.0

03 Dec 19:49
9f30df5
Compare
Choose a tag to compare

Added

  • Support the V2 CPP extension API for backward compatibility
  • Support for data type Int8 in with Int8 out
  • Support for data type FP32/FP64 for gfx110x
  • Add the Extension API hipblaslt_ext::matmulIsTuned
  • Output atol and rtol for hipblaslt-bench validation
  • Output the bench command for hipblaslt CPP ext API path if HIPBLASLT_LOG_MASK=32 is set
  • Support odd sizes for FP8/BF8 GEMM

Changed

  • Reorganize and add more sample code
  • Add a dependency with the hipblas-common package and remove the dependency with the hipblas package

Optimized

  • Support fused kernel for HIPBLASLT_MATMUL_DESC_AMAX_D_POINTER for FP8/BF8 data type
  • Improve the library loading time
  • Improve the overall performance of first returned solution

Upcoming changes

  • The V1 CPP extension API will be deprecated in a future release of hipBLASLt

hipBLASLt 0.8.0 for ROCm 6.2.4

06 Nov 19:55
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.2.4 did not change. The library was rebuilt for the updated ROCm 6.2.4 stack.

hipBLASLt 0.8.0 for ROCm 6.2.2

27 Sep 16:01
44efd61
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.2.2 did not change. The library was rebuilt for the updated ROCm 6.2.2 stack.

hipBLASLt 0.8.0 for ROCm 6.2.1

20 Sep 19:57
44efd61
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.2.1 did not change. The library was rebuilt for the updated ROCm 6.2.1 stack.

hipBLASLt 0.8.0 for ROCm 6.2.0

02 Aug 16:15
Compare
Choose a tag to compare

hipBLASLt code for ROCm 6.2.0 did not change. The library was rebuilt for the updated ROCm 6.2.0 stack.