ydb - Mirror of YDB github repos

	Commit message (Collapse)	Author	Age	Files	Lines
*	[build] cuda: Fix CUB with CUDA_SRCS()	deshevoy	2026-03-26	1	-2/+12
\| \| \| \| \|	Specify `__CUDA_ARCH_LIST__` explicitly so CUB namespace stay the same across all nvcc invokations commit_hash:2100ccb2307100378bcead498fd34cd11e44c566
*	[build] cuda: Disable some warnings when compiling host code	deshevoy	2026-02-09	1	-1/+8
\| \| \| \| \| \| \|	nvcc disables them implicitly ISSUE: commit_hash:0b68decce1f030902bd770b8b98fc8102c97e738
*	[build] cuda: Add .module_id sanity check to CUDA_SRCS	deshevoy	2026-02-02	1	-0/+1
\| \| \| \| \| \| \|	IDs generated by different cicc invocations should match ISSUE: commit_hash:7cd593cee44b31875e7166709d7614dcfa3f1f14
*	[build] cuda: Fix CUDA_SRCS to support architecture- and family-specific ↵	deshevoy	2026-02-02	1	-11/+3
\| \| \| \| \| \| \| \| \|	features E.g. sm_90a or sm_100f ISSUE: commit_hash:250df064a8abcac925db676565582b5ef05401bb
*	[build] cuda: Fix	deshevoy	2026-02-01	1	-1/+1
\| \| \| \|	commit_hash:f73df3ec27f0695b21e7047ee465b15b201ea06b
*	[build] cuda: Introduce CUDA_SRCS macro utilizing parallelized device code ↵	deshevoy	2026-02-01	1	-0/+45
	compilation Instead of a single graph node launching NVCC to compile .cu for both host and all device architectures CUDA_SRCS generates multiple nodes: - node per each device architecture producing PTX and CUBIN - node merging all PTX and CUBIN files into a single FATBIN blob - node producing .cpp with host code - node compiling host .cpp with embedded FATBIN blob CUDA_ARCHITECTURES variable is used to determine the list of architectures to compile device code for. ISSUE: commit_hash:0a4c2797dd238ae062482af30694df6978301278