rocm-7.0.2

ROCm release v7.0.2

rocm-7.0.0

ROCm release v7.0.0

Tensile 4.43.0 for ROCm 6.4.4

Tensile code for ROCm 6.4.4 did not change. The library was rebuilt for the updated ROCm 6.4.4 stack.

Tensile 4.43.0 for ROCm 6.4.3

Tensile code for ROCm 6.4.3 did not change. The library was rebuilt for the updated ROCm 6.4.3 stack.

Tensile 4.43.0 for ROCm 6.4.2

Tensile code for ROCm 6.4.2 did not change. The library was rebuilt for the updated ROCm 6.4.2 stack.

Tensile 4.43.0 for ROCm 6.4.1

Tensile code for ROCm 6.4.1 did not change. The library was rebuilt for the updated ROCm 6.4.1 stack.

Tensile 4.43.0 for ROCm 6.4.0

Added

• Nightly builds with performance statistics
• Cache asm capabilities for reuse
• venv for Tensile create on Linux
• Flag to keep build_tmp when running Tensile
• Generalized profiling scripts
• GFX1151 support
• Single-threaded support in TensileCreateLibrary
• Logic to remove temporary build artifacts

Changed

• Updated Tensile documents (API reference, README.md, and comments)
• Disabled asm-cache for tests
• Used hipcc.bat as a compiler on Windows instead of the Perl script
• Improved clarity of CHANGELOG.md
• Enabled external CI
• Improved Tensile documentation
• Refactored kernel source and header creation
• Refactored writeKernels in TensileCreateLibrary
• Suppressed developer warnings (simplifying the Tensile output)
• Used an explicit cast when invoking min is called
• Used cache abbreviations to compute kernel names

Removed

• OCL backend
• Unsupported tests
• Deep copy in TensileCreateLibrary

Optimized

• Linearized asm register search to reduce build time

Resolved issues

• Fixed Stream-K dynamic grid model
• Fixed logic related to caching asm capabilities
• Fixed accvgpr overflow
• Fixed test failures in SLES containers when running TensileTests
• Fixed a regression that prevents TensileCreateLibrary from completing when fallback logic is not available

Tensile 4.41.0 for ROCm 6.2.4

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

Tensile 4.42.0 for ROCm 6.3.3

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

Tensile 4.42.0 for ROCm 6.3.2

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