Fix #142: AMD GPU detection on Windows - #156
Merged
Merged
Conversation
- Add configure_gpu_environment() method to set GGML_* environment variables before backend initialization - Fix GPU layer assignment by ensuring OpenCL/Vulkan/CUDA backends have proper environment setup - Add comprehensive regression tests for all GPU backend environment configuration - Fix compiler warnings with documented #[allow(dead_code)] attributes for conditionally-used MoeConfig fields Root cause: GPU backends require environment variables set before llama.cpp initialization. AMD GPUs detected by clinfo but layers assigned to CPU due to missing GGML_OPENCL/GGML_VULKAN variables. Tests: All 6 release gates pass with zero warnings. Regression tests added for GPU backend validation.
- Prefix unused variables with underscores to suppress warnings - Variables are conditionally used based on feature flags
Michael-A-Kuykendall
added a commit
that referenced
this pull request
Dec 19, 2025
* Fix #142: AMD GPU detection on Windows - Add configure_gpu_environment() method to set GGML_* environment variables before backend initialization - Fix GPU layer assignment by ensuring OpenCL/Vulkan/CUDA backends have proper environment setup - Add comprehensive regression tests for all GPU backend environment configuration - Fix compiler warnings with documented #[allow(dead_code)] attributes for conditionally-used MoeConfig fields Root cause: GPU backends require environment variables set before llama.cpp initialization. AMD GPUs detected by clinfo but layers assigned to CPU due to missing GGML_OPENCL/GGML_VULKAN variables. Tests: All 6 release gates pass with zero warnings. Regression tests added for GPU backend validation. * Fix compiler warnings in issue 142 regression test - Prefix unused variables with underscores to suppress warnings - Variables are conditionally used based on feature flags
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
1 participant
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.
Fixes AMD GPU detection issue on Windows where GPUs were detected by clinfo but all layers were assigned to CPU instead of GPU.
Root Cause: GPU backend environment variables (GGML_OPENCL, GGML_VULKAN, GGML_CUDA) were not set before llama.cpp backend initialization.
Changes:
Testing: All 6 release gates pass with zero warnings. Regression tests verify environment variable configuration for all GPU backends.
Issue: #142