Vulkan Embedding Fix (#7360)

Browse source

* Fix empty Vulkan host buffers

Add fp32 fp16 matmul shader

Fix matmul shader alignment

* Remove deprecated tensor->backend uses

* Fix Vulkan validation errors on embedding models with no offloaded layers

* Fix Vulkan llava segfault when not offloading layers

...

This commit is contained in:

0cc4m 2024-05-19 17:19:53 +02:00 • committed by GitHub

parent e4e6f67be6

commit f030ec1f7a

No known key found for this signature in database

GPG key ID: B5690EEEBB952194

3 changed files with 8000 additions and 4397 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

12204

ggml-vulkan-shaders.hpp

View file

File diff suppressed because it is too large Load diff