]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
ggml : optimize cuda ssm_scan using warp-level reduction (llama/18505)
authorAadeshveer Singh <redacted>
Tue, 6 Jan 2026 18:24:34 +0000 (23:54 +0530)
committerGeorgi Gerganov <redacted>
Wed, 14 Jan 2026 07:11:59 +0000 (09:11 +0200)
commit436f30d05f5bfb7791dff01d9d80aa7e6c3f94d6
treeca91abdbf56cb604a8cb0bfddef255a84cab1f59
parentdbec71f6cf0b10eabc625b91cf0dcfaa368bf7f2
ggml : optimize cuda ssm_scan using warp-level reduction (llama/18505)

* ggml : optimize cuda ssm_scan using warp-level reduction

* ggml : apply code review suggestions (style, const, constexpr)

* ggml : add TODO regarding stride consistency
ggml/src/ggml-cuda/ssm-scan.cu