Skip to content

Conversation

netrunnereve
Copy link
Collaborator

Basically Vulkan headers 1.3.301 and newer contain vk::detail::DispatchLoaderDynamic and the older ones have vk::DispatchLoaderDynamic instead.

@netrunnereve netrunnereve requested a review from 0cc4m as a code owner September 30, 2025 03:28
@github-actions github-actions bot added Vulkan Issues specific to the Vulkan backend ggml changes relating to the ggml tensor library for machine learning labels Sep 30, 2025
@jeffbolznv
Copy link
Collaborator

I looked back in the vulkan-hpp commit history, I think this change was made between header versions 281 and 282.

@geerlingguy
Copy link

I've tested on my Raspberry Pi 5 running Pi OS 12 (based on Debian 12), applying this PR/patch, and recompiling llama.cpp, I am able to build it successfully again.

Before I was getting the errors shown in #16311.

@netrunnereve
Copy link
Collaborator Author

I looked back in the vulkan-hpp commit history, I think this change was made between header versions 281 and 282.

Nope it happened between 300 and 301. See this diff: KhronosGroup/Vulkan-Headers@cbcad3c#diff-a83f956c7aeb2b6dee7ffa2f249c0d98a2cfad6e157fe8438e372a5267afaeff

@0cc4m 0cc4m merged commit 132d673 into ggml-org:master Oct 1, 2025
61 of 68 checks passed
@netrunnereve netrunnereve deleted the dispatchloaderdynamic branch October 1, 2025 16:56
yael-works pushed a commit to yael-works/llama.cpp that referenced this pull request Oct 15, 2025
…gml-org#16345)

* make ggml_vk_default_dispatcher support older vulkan headers

* simpilfy with using
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants