Test: CI build optimizations (CUDA compute capabilities and non-curses mode from cl/945255455)#4142

Draft

vkarampudi wants to merge 6 commits into

tensorflow:masterfrom

vkarampudi:test-gpu-docker-fixes

Commits on Jul 9, 2026

Test: CI build optimizations (CUDA compute capabilities and non-curses mode from cl/945255455)
vkarampudi
committed
Fix: Prune sm_60 from .bazelrc and add --noshow_progress to Dockerfiles to eliminate 5hr GPU build and 2MiB log clipping
vkarampudi
committed

Commits on Jul 10, 2026

Fix: Prune sm_70 from .bazelrc to resolve Volta Clang CUDA build errors and speed up compilation
vkarampudi
committed
Fix: Add cuda_clang_fixes.h compatibility header and include flags to resolve NCCL 2.29 typeof and uint4 __stwt errors under Clang
vkarampudi
committed

Commits on Jul 11, 2026

Fix: Append typeof and uint4 __stwt overloads directly to clang_cuda_runtime_wrapper.h via rules_ml_toolchain patch_cmds to avoid un-sandboxed header file not found errors
vkarampudi
committed

Commits on Jul 12, 2026

Fix GPU Docker build failures.
vkarampudi
committed