llama.cpp
Vulkan Optimizations and Fixes
#8959
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
9
Changes
View On
GitHub
Vulkan Optimizations and Fixes
#8959
0cc4m
merged 9 commits into
master
from
0cc4m/vulkan-optimization
Optimize Vulkan REPEAT performance
0645ed5c
Use Vulkan GLSL fused multiply-add instruction where possible
f78487b8
Add GGML_VULKAN_PERF option to output performance data per operator
4c6a7bb0
Rework and fix Vulkan descriptor set and descriptor pool handling
efe6aca5
Fix float32 concat f16 shader validation error
9e0ac989
github-actions
added
Vulkan
github-actions
added
ggml
mofosyne
added
Review Complexity : Medium
mofosyne
added
performance
mofosyne
added
bugfix
Add Vulkan GROUP_NORM eps parameter
61d83887
Merge upstream changes, fix conflicts
4f197e33
Fix validation error with transfer queue memory barrier flags
5ae33eb9
0cc4m
marked this pull request as ready for review
1 year ago
ggerganov
commented on 2024-08-11
0cc4m
requested a review
from
ggerganov
1 year ago
slaren
approved these changes on 2024-08-14
ggerganov
approved these changes on 2024-08-14
Remove trailing whitespaces
12d214f4
0cc4m
merged
5fd89a70
into master
1 year ago
0cc4m
deleted the 0cc4m/vulkan-optimization branch
1 year ago
Login to write a write a comment.
Login via GitHub
Reviewers
ggerganov
slaren
Assignees
No one assigned
Labels
performance
Vulkan
bugfix
Review Complexity : Medium
ggml
Milestone
No milestone
Login to write a write a comment.
Login via GitHub