Vulkan Improvements (llama/5835)

Browse Source

* Improve dequant shaders, add fast q4_0 dequant

* Optimize dmmv non-kquants for GCN

Remove unnecessary SPIR-V shader duplication

* Fix q4_0 dequant dispatch sizes

Fix backend free bug

* Optimize dequant shaders for q4_1, q5_0, q5_1 and q8_0

* Add unary and binary op shader templates

* Fix Vulkan check results

* Enable non-contiguous support for simple ops

* Add argsort

Basic q4_0 mmq shader and unit test

* Speed up q4_0 dequant code, enable mmq for q4_0

* Rework matmul pipeline selection

* Add soft_max alibi support

* Add q4_1, q5_0, q5_1 and q8_0 dequant mat mat mul shaders

* Add environment variable GGML_VK_FORCE_MAX_ALLOCATION_SIZE to limit max buffer size

Rename GGML_VULKAN_DISABLE_F16 to GGML_VK_DISABLE_F16 for consistency

...

This commit is contained in:

0cc4m

2024-03-05 13:33:42 +01:00

committed by Georgi Gerganov

parent 9d9a405cfd

commit 9c399689ec

2 changed files with 1341 additions and 788 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

2128

ggml-vulkan.cpp

File diff suppressed because it is too large Load Diff