Vulkan Shader Refactor, Memory Debugging Option (llama/7947)

Browse Source

* Refactor shaders, extract GLSL code from ggml_vk_generate_shaders.py into vulkan-shaders directory

* Improve debug log code

* Add memory debug output option

* Fix flake8

* Fix unnecessary high llama-3 VRAM use

...

This commit is contained in:

0cc4m

2024-06-18 09:39:40 +03:00

committed by Georgi Gerganov

parent 63a767a134

commit 922971041b

1 changed files with 259 additions and 353 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

ggml-vulkan.cpp

+259 -353

File diff suppressed because it is too large Load Diff