3 Commits

Include renames

Author	SHA1	Message	Date
Johannes Gäßler	7049736b2d	CUDA: fix numerical issues in tile FA kernel (#16540)	2025-10-13 17:29:45 +03:00
Johannes Gäßler	11f0af5504	CUDA: faster tile FA, add oob checks, more HSs (#16492)	2025-10-11 20:54:32 +02:00
Johannes Gäßler	79bc429262	CUDA: faster tile FA (Pascal/AMD), headsize 256 (#15769)	2025-09-07 00:26:28 +02:00