llama.cpp
ef52d1d1 - Update Vulkan RoPE implementation (#7818)

Commit

1 year ago

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 <slarengh@gmail.com> --------- Co-authored-by: slaren <slarengh@gmail.com>

References

#7818 - Update Vulkan RoPE implementation

Author

0cc4m

Parents

14f83526

Files4

ggml-alloc.c
ggml-vulkan-shaders.hpp
ggml-vulkan.cpp
ggml_vk_generate_shaders.py

llama.cpp ef52d1d1 - Update Vulkan RoPE implementation (#7818)

llama.cpp
ef52d1d1 - Update Vulkan RoPE implementation (#7818)