thelittlefireman 748d4225b9 ggml-cuda: HIP: optimize packed byte subtraction (__vsubss4 -> __vsub4) (#29478) ... * ggml-cuda: HIP: optimize non-saturating packed byte subtraction (`__vsubss4`) * CI: ignore 1 spilled vgpr in fattn_vec --------- Co-authored-by: Carl Philipp Klemm <carl@uvos.xyz>		2026-09-29 19:59:19 +02:00
..
gcn-cdna-vgpr-check.py	ggml-cuda: HIP: optimize packed byte subtraction (__vsubss4 -> __vsub4) (#29478)	2026-09-29 19:59:19 +02:00