]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 8ae32dc)
tests : add -INF blocks to the KQ mask in the FA tests (#16380)
authorGeorgi Gerganov <redacted>
Tue, 7 Oct 2025 05:22:35 +0000 (08:22 +0300)
committerGitHub <redacted>
Tue, 7 Oct 2025 05:22:35 +0000 (08:22 +0300)
commit1d6092fc72f4d10f4486ac95edfd414bc08b62b8
treeb3b7539f2453ba229d3061692f7dffb1e3a00e5etree
parent8ae32dc9ecd7aeebf3a5b43557e0552a0a04cd4fcommit | diff
tests : add -INF blocks to the KQ mask in the FA tests (#16380)

* tests : add -INF blocks to the KQ mask in the FA tests

* cont : bump -INF block size to 64

Co-authored-by: Jeff Bolz <redacted>
* ggml : prevent division by zero in FA CPU op

---------

Co-authored-by: Jeff Bolz <redacted>
ggml/src/ggml-cpu/ops.cpp diff | blob | history
tests/test-backend-ops.cpp diff | blob | history
Packaging of ggml-org/llama.cpp