]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 14f8352)
Update Vulkan RoPE implementation (#7818)
author0cc4m <redacted>
Tue, 11 Jun 2024 19:20:29 +0000 (21:20 +0200)
committerGitHub <redacted>
Tue, 11 Jun 2024 19:20:29 +0000 (21:20 +0200)
commitef52d1d16afc695d798396cdd13594ea5e45a9dd
treed6ef7a296ce1df22eb803650da442a1c5ca55340tree
parent14f83526cd27f638c856ea6eff08110b9860eb2acommit | diff
Update Vulkan RoPE implementation (#7818)

* Update Vulkan RoPE implementation

* Return nullptr on alloc_buffer when allocation fails, instead of throwing an exception

Minor fixes

* Fix segfault when running out of VRAM

Co-authored-by: slaren <redacted>
---------

Co-authored-by: slaren <redacted>
ggml-alloc.c diff | blob | history
ggml-vulkan-shaders.hpp diff | blob | history
ggml-vulkan.cpp diff | blob | history
ggml_vk_generate_shaders.py diff | blob | history
Packaging of ggml-org/llama.cpp