UPSTREAM PR #18749: Vulkan: Optimize Matmul parameters for AMD GPUs with Coopmat support

[loci-dev]

Mirrored from ggml-org/llama.cpp#18749

I tuned this on AMD Radeon 8060S, but a brief test also showed improvements on AMD RX 9060 XT.

ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat

model	size	params	ngl	test	t/s (ROCm)	t/s (before)	t/s (after)	diff
llama 8B Q4_0	4.33 GiB	8.03 B	99	pp512	815.22 ± 16.88	644.84 ± 4.49	875.24 ± 17.99	+35.7%
llama 8B Q4_0	4.33 GiB	8.03 B	99	pp512 @ d8192	321.66 ± 3.02	384.27 ± 0.29	463.72 ± 1.32	+20.7%
llama 8B Q4_K - Small	4.36 GiB	8.03 B	99	pp512	995.81 ± 32.32	529.50 ± 13.07	793.51 ± 9.67	+49.9%
llama 8B Q4_K - Small	4.36 GiB	8.03 B	99	pp512 @ d8192	352.08 ± 3.17	341.86 ± 1.96	435.55 ± 2.63	+27.4%
llama 8B Q8_0	7.95 GiB	8.03 B	99	pp512	755.86 ± 21.11	422.37 ± 21.61	742.46 ± 15.49	+75.8%
llama 8B Q8_0	7.95 GiB	8.03 B	99	pp512 @ d8192	317.44 ± 2.31	306.97 ± 0.36	419.24 ± 4.59	+36.6%

ggml_vulkan: 0 = AMD Radeon RX 9060 XT (RADV GFX1200) (radv) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat

model	size	params	ngl	test	t/s (ROCm)	t/s (before)	t/s (after)	diff
llama 8B Q4_K - Small	4.36 GiB	8.03 B	99	pp512	648.41 ± 14.37	1437.13 ± 1.77	1902.23 ± 1.65	+32.4%
llama 8B Q4_K - Small	4.36 GiB	8.03 B	99	pp512 @ d8192	410.01 ± 6.22	757.59 ± 2.71	841.76 ± 3.02	+11.1%

[loci-review]

Explore the complete analysis inside the Version Insights

Perfect! I've generated a summary report for your project. Here are the key findings:

Summary Report for llama.cpp PR #884

Performance Impact: ✅ MINIMAL

The analysis comparing the base version (1694ff01-eeb5-11f0-a055-c529586b3e1a) to the target version (b8594261-eec9-11f0-a055-c529586b3e1a) shows:

• No significant response time changes (>2% threshold)
• No significant throughput changes (>2% threshold)

Conclusion

This pull request is performance-neutral, meaning:

• ✅ No performance regressions detected
• ✅ Code changes maintain performance stability
• ✅ Safe to proceed without performance concerns

The PR can move forward without requiring any performance optimization work.