vulkan: add pipeline barriers for memcpy read operations#23770
Merged
Conversation
github-actions
Bot
added
Vulkan
jeffbolznv
reviewed
May 27, 2026
0cc4m
force-pushed
the
0cc4m/vulkan-host-pipeline-barriers
branch
from
May 27, 2026 15:15
f0e0f66 to
0a1ebc5
Compare
0cc4m
force-pushed
the
0cc4m/vulkan-host-pipeline-barriers
branch
from
June 12, 2026 07:11
0a1ebc5 to
d998fbf
Compare
Contributor
Author
|
I forgot to push this further, let's get it done now. @jeffbolznv This may not be all barriers that are needed, but for now it should unblock the two PRs depending on it. Any concerns? |
Contributor
|
Have somebody done perf testing to make sure this doesn't add a bunch of overhead? I'm less worried about it since it's limited to uma devices. It would be nice to check perf on dgx spark but I don't have access to one. But I'll +1 it to unblock things. |
jeffbolznv
approved these changes
Jun 12, 2026
Contributor
Author
|
Good idea, here's Qwen on DGX Spark with this PR, #24326 and #22930:
And the same on AMD Strix Halo:
|
Contributor
Author
|
@ggml-org/maintainers Another approval needed. |
ngxson
approved these changes
Jun 12, 2026
0cc4m
changed the title
anaisbetts
pushed a commit
to anaisbetts/llama.cpp
that referenced
this pull request
Jun 16, 2026
) * vulkan: add pipeline barriers for memcpy read/write operations * remove unnecessary host write pipeline barriers (cherry picked from commit 3e7bd4f)
fukuro-kun
pushed a commit
to fukuro-kun/fukuro-llama-cpp-turboquant
that referenced
this pull request
Jun 23, 2026
fukuro-kun
pushed a commit
to fukuro-kun/fukuro-llama-cpp-turboquant
that referenced
this pull request
Jun 23, 2026
…l-org#23770)" This reverts commit d34db42.
fukuro-kun
pushed a commit
to fukuro-kun/fukuro-llama-cpp-turboquant
that referenced
this pull request
Jul 5, 2026
fukuro-kun
pushed a commit
to fukuro-kun/fukuro-llama-cpp-turboquant
that referenced
this pull request
Jul 5, 2026
…l-org#23770)" This reverts commit d0bb87b.
adrianhoehne
pushed a commit
to adrianhoehne/llama.cpp
that referenced
this pull request
Jul 5, 2026
fukuro-kun
pushed a commit
to fukuro-kun/fukuro-llama-cpp-turboquant
that referenced
this pull request
Jul 12, 2026
…i#10 UMA nicht umsetzbar AtomicBot-ai#6 MUL_MAT_ID Subgroup (PR ggml-org#15524): bereits vollständig im Fork integriert - MatMulIdType::SUBGROUP, mul_mm_id_funcs.glsl, matmul_id_subgroup_* Shader - Pipelines aktiv (subgroup_ballot && subgroup_require_full_support) - Vorheriger Revert war fehlerhafter Merge, aktueller Stand sauber AtomicBot-ai#10 UMA Zero-Copy (PR ggml-org#22462): nicht implementieren - Verwandte UMA-PRs (ggml-org#22455, ggml-org#22930, ggml-org#23770) bereits revertiert - RCA-Masterplan: System-RAM langsamer als GTT für GPU-Compute auf Mars - PR ggml-org#22462 ist open/unstable
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Overview
Insert Vulkan pipeline barriers before/after memcpy operations for host-visible memory. This resolves the Intel read issue in #22930, and I also added the opposite direction barriers for writes. I am not sure if this is the best way. @jeffbolznv @rillomas
Requirements