[vulkan] Allow u8/u16 when only 8/16-bit storage is supported (no shaderInt16)#8759

Open

yunusberndt wants to merge 26 commits into

taichi-dev:masterfrom

yunusberndt:Vulkan-patch-for-AMD

Commits on Aug 28, 2025

Commits on Aug 29, 2025

Commits on Aug 31, 2025

fix: resolve Vulkan build and compatibility issues for AMD GPUs
yunusberndt
committed
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot]
committed

Commits on Sep 8, 2025

Enhance Vulkan support for AMD GPUs by refining 16-bit storage capabilities
yunusberndt
committed
reverted gitignore to upstream standard
yunusberndt
committed
Merge branch 'Vulkan-patch-for-AMD' of https://github.com/yunusberndt/taichi into Vulkan-patch-for-AMD
yunusberndt
committed
Refine Vulkan device creation and update build configuration
yunusberndt
committed
Fix formatting issues: black, clang-format, trailing whitespace, and end-of-file fixes
yunusberndt
committed
Fix pre-commit formatting issues: Black 25.1.0 and clang-format
yunusberndt
committed
Fix Vulkan 1.1 version check for AMD 8-bit/16-bit support
yunusberndt
committed
Add debug output for AMD GPU capability detection
yunusberndt
committed
Add detailed debug output for feature detection values
yunusberndt
committed

Commits on Sep 9, 2025