]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 19bba67)
cuda: reset cuda context after reading memory size (#23935)
authorRuben Ortlam <redacted>
Mon, 8 Jun 2026 08:22:44 +0000 (10:22 +0200)
committerGitHub <redacted>
Mon, 8 Jun 2026 08:22:44 +0000 (10:22 +0200)
commit0f7fada56bea32c9e0db3874b801d887ff6c4529
tree17a90a991b5ddb980abaaf75a44a31a08fbfe50atree
parent19bba67c1f4db723c60a0d421aa0788bf4ddc699commit | diff
cuda: reset cuda context after reading memory size (#23935)

* cuda: reset device in get_memory function if no backend is active

* also count device and host buffers

* exclude hip and musa from counting and device reset

* use device mutex instead of atomic

* undo backend_free function move
ggml/src/ggml-cuda/ggml-cuda.cu diff | blob | history
Packaging of ggml-org/llama.cpp