UPSTREAM PR #18431: ggml-cuda: (cmake) expand "native" into concrete architectures

[loci-dev]

Mirrored from ggml-org/llama.cpp#18431

Fixes #18430.

#18413 caused error on blackwell local build because native will be treated as 120-real but is not captured by the regex.

Tested with or without docker (with a blackwell GPU). Seems OK.

[loci-review]

Explore the complete analysis inside the Version Insights

Perfect! I've retrieved the summary report for your project. Here's what the analysis shows:

Summary Report for llama.cpp PR #730

Key Findings:

• ✅ No significant performance changes detected - All modified functions show performance changes of less than 2%
• ✅ No regressions in response time or throughput time
• ✅ Performance-neutral changes - The pull request maintains stable performance compared to the base version

Project Information:

• Repository: llama.cpp (owner: auroralabs-loci)
• Pull Request: UPSTREAM PR #18431: ggml-cuda: (cmake) expand "native" into concrete architectures #730
• Comparing version 105f68ff-a65a-4885-a8a3-f3fa9f8b2d34 against base 2e5a4c4b-9082-45a3-bf95-9f818a494db0

This is a positive result, indicating that the changes in PR #730 don't introduce any performance degradation and maintain the existing performance characteristics of the codebase.