Repository navigation
ci : run test-backend-ops as a dedicated ci/run.sh test - #28740
Merged
Merged
Conversation
github-actions
Bot
added
testing
ggerganov
changed the title
Run test-backend-ops as a separate gg test in ci/run.sh so it is executed outside ctest. With GG_BUILD_HIGH_PERF it keeps the existing CPU-only invocation (-b CPU); otherwise it runs all available backends without a backend filter. Remove the dedicated backend-ops workflow and keep test-backend-ops as a built target that is not registered with ctest to avoid duplicate runs. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp
Move the test-backend-ops gg test before test-llama-archs. Enable GG_BUILD_HIGH_PERF and LLAMA_ARG_THREADS on the Graviton4 KleidiAI job and use the standard self-hosted results/mnt paths. Add TODO markers for decoupling tests from libllama. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp
Pass -j $(nproc) to test-backend-ops in both high-perf and all-backend modes. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp
ggerganov
force-pushed
the
gg/ci-workflow-backend-ops
branch
from
September 11, 2026 14:37
55a7cdc to
6c60722
Compare
Member
Author
|
@ggml-org/amd The |
Member
Author
|
The fyi @ggml-org/ggml-vulkan |
Contributor
|
Likely a moltenvk bug. We could try to workaround this by adding locks in a few places, or should we just try to run that CI job without -j? |
Member
Author
|
For now I've disabled |
Contributor
The stream-capture discrepancy is something I'll look into further but #28782 is the minimal fix on the llama.cpp end to unblock |
Member
Author
|
@jeffbolznv The parallel |
Contributor
|
I think I have a good lead on this, but it's very intermittent and I need time to make sure my fixes really work. |
Contributor
|
#28830 should fix the vulkan failure. |
pl752
pushed a commit
to pl752/llama.cpp
that referenced
this pull request
Sep 15, 2026
* ci : run test-backend-ops as a dedicated gg test Run test-backend-ops as a separate gg test in ci/run.sh so it is executed outside ctest. With GG_BUILD_HIGH_PERF it keeps the existing CPU-only invocation (-b CPU); otherwise it runs all available backends without a backend filter. Remove the dedicated backend-ops workflow and keep test-backend-ops as a built target that is not registered with ctest to avoid duplicate runs. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops earlier and enable high-perf on kleidiai Move the test-backend-ops gg test before test-llama-archs. Enable GG_BUILD_HIGH_PERF and LLAMA_ARG_THREADS on the Graviton4 KleidiAI job and use the standard self-hosted results/mnt paths. Add TODO markers for decoupling tests from libllama. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops in parallel Pass -j $(nproc) to test-backend-ops in both high-perf and all-backend modes. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : disable parallel tests for ROCm * cont : disable parallel tests with MoltenVK
quimmedes
pushed a commit
to quimmedes/cafe-llama.cpp
that referenced
this pull request
Sep 16, 2026
* ci : run test-backend-ops as a dedicated gg test Run test-backend-ops as a separate gg test in ci/run.sh so it is executed outside ctest. With GG_BUILD_HIGH_PERF it keeps the existing CPU-only invocation (-b CPU); otherwise it runs all available backends without a backend filter. Remove the dedicated backend-ops workflow and keep test-backend-ops as a built target that is not registered with ctest to avoid duplicate runs. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops earlier and enable high-perf on kleidiai Move the test-backend-ops gg test before test-llama-archs. Enable GG_BUILD_HIGH_PERF and LLAMA_ARG_THREADS on the Graviton4 KleidiAI job and use the standard self-hosted results/mnt paths. Add TODO markers for decoupling tests from libllama. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops in parallel Pass -j $(nproc) to test-backend-ops in both high-perf and all-backend modes. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : disable parallel tests for ROCm * cont : disable parallel tests with MoltenVK
zsogitbe
pushed a commit
to zsogitbe/llama.cpp
that referenced
this pull request
Sep 17, 2026
* ci : run test-backend-ops as a dedicated gg test Run test-backend-ops as a separate gg test in ci/run.sh so it is executed outside ctest. With GG_BUILD_HIGH_PERF it keeps the existing CPU-only invocation (-b CPU); otherwise it runs all available backends without a backend filter. Remove the dedicated backend-ops workflow and keep test-backend-ops as a built target that is not registered with ctest to avoid duplicate runs. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops earlier and enable high-perf on kleidiai Move the test-backend-ops gg test before test-llama-archs. Enable GG_BUILD_HIGH_PERF and LLAMA_ARG_THREADS on the Graviton4 KleidiAI job and use the standard self-hosted results/mnt paths. Add TODO markers for decoupling tests from libllama. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops in parallel Pass -j $(nproc) to test-backend-ops in both high-perf and all-backend modes. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : disable parallel tests for ROCm * cont : disable parallel tests with MoltenVK
frostyautumnleaf
pushed a commit
to frostyautumnleaf/llama.cpp
that referenced
this pull request
Oct 5, 2026
* ci : run test-backend-ops as a dedicated gg test Run test-backend-ops as a separate gg test in ci/run.sh so it is executed outside ctest. With GG_BUILD_HIGH_PERF it keeps the existing CPU-only invocation (-b CPU); otherwise it runs all available backends without a backend filter. Remove the dedicated backend-ops workflow and keep test-backend-ops as a built target that is not registered with ctest to avoid duplicate runs. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops earlier and enable high-perf on kleidiai Move the test-backend-ops gg test before test-llama-archs. Enable GG_BUILD_HIGH_PERF and LLAMA_ARG_THREADS on the Graviton4 KleidiAI job and use the standard self-hosted results/mnt paths. Add TODO markers for decoupling tests from libllama. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : run test-backend-ops in parallel Pass -j $(nproc) to test-backend-ops in both high-perf and all-backend modes. Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp * ci : disable parallel tests for ROCm * cont : disable parallel tests with MoltenVK
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
Run
test-backend-opsas a dedicated test inci/run.shoutside ctest. WhenGG_BUILD_HIGH_PERFis set it keeps the existing CPU-only invocation (-b CPU); otherwise it runs all available backends without a backend filter. The test is parallelized with-j $(nproc).Keep
test-backend-opsas a built target but not registered with ctest to avoid duplicate execution, and remove the staletest-backend-opsexclusions from the OpenVINO, Vulkan and WebGPU workflows.Enable
GG_BUILD_HIGH_PERFandLLAMA_ARG_THREADSon the Graviton4 KleidiAI job and switch it to the standard self-hosted results/mnt paths.Additional information
Adds TODO markers for decoupling tests from
libllama.Requirements