vulkan: deduplicate Microsoft Direct3D12 devices #16689
Merged
+9
−1
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… from the `vulkan-dozen` driver
0cc4m
reviewed
Oct 25, 2025
ggml/src/ggml-vulkan/ggml-vulkan.cpp
Outdated
| return std::equal(std::begin(old_id.deviceUUID), std::end(old_id.deviceUUID), std::begin(new_id.deviceUUID)); | ||
|
|
||
| bool equals = std::equal(std::begin(old_id.deviceUUID), std::end(old_id.deviceUUID), std::begin(new_id.deviceUUID)); | ||
| equals |= old_id.deviceLUIDValid && new_id.deviceLUIDValid && |
There was a problem hiding this comment.
|= is a bitwise or, please use a logical or.
0cc4m
approved these changes
Oct 26, 2025
pwilkin
pushed a commit
to pwilkin/llama.cpp
that referenced
this pull request
Oct 27, 2025
* fix: deduplicate and deprioritize Microsoft Direct3D12 vulkan devices from the `vulkan-dozen` driver * style: indent * fix: decrease priority * fix: switch to `||`
theo77186
pushed a commit
to theo77186/llama.cpp
that referenced
this pull request
Oct 28, 2025
* fix: deduplicate and deprioritize Microsoft Direct3D12 vulkan devices from the `vulkan-dozen` driver * style: indent * fix: decrease priority * fix: switch to `||`
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
A Microsoft Direct3D12 device appears on some Windows machines, which comes from the Microsoft Dozen (Dzn) Vulkan driver that is built on top of Direct3D 12.
This PR fixes an issue where sometimes both the native device entry and the Direct3D12 device entry are picked up by llama.cpp, and since both point to the same underlying device, after using all the memory of the first device, buffer allocations on the second will always fail.
When a native Vulkan driver is already available for a given device, we now prioritize it over the Dozen driver for reduced overhead.
I found a note on VulkanHub about the Direct3D12 device that was helpful in figuring this out.
Before
After