UPSTREAM PR #21539: vulkan: Support Q1_0

[loci-dev]

Note

Source pull request: ggml-org/llama.cpp#21539

Overview

Add Q1_0 support to ggml-vulkan. Supports get_rows, set_rows, mul_mat(id). Does not support the q8_1 dp4 path (though this is probably worth adding in a followon), since we get the most benefit with smaller quants.

Additional information

None.

Requirements

• I have read and agree with the contributing guidelines
• AI usage disclosure: I used Claude to write the majority of this change. I reviewed all the code and had it rewrite several things for consistency, and I rewrote the coopmat2 dequant function myself.

[loci-review]

No meaningful performance changes were detected across 125311 analyzed functions in the following binaries: build.bin.libllama.so, build.bin.llama-tts, build.bin.llama-cvector-generator, build.bin.llama-bench, build.bin.libmtmd.so, build.bin.libggml.so, build.bin.libggml-cpu.so, build.bin.libggml-base.so, build.bin.llama-tokenize, build.bin.llama-gemma3-cli, build.bin.llama-gguf-split, build.bin.llama-llava-cli, build.bin.llama-minicpmv-cli, build.bin.llama-quantize, build.bin.llama-qwen2vl-cli.

💬 Questions? Tag @loci-dev