[experimental] metal execution #6868

Performance Gate Passed

✅ 1000 untouched benchmarks
🆕 12 new benchmarks
⏩ 1466 skipped benchmarks¹

	Mode	Benchmark	`BASE`	`HEAD`	Efficiency
🆕	Simulation	`for_decompress_i32[(100000, "100K")]`	N/A	742.1 µs	N/A
🆕	Simulation	`for_decompress_i64[(1000000, "1M")]`	N/A	11.4 ms	N/A
🆕	Simulation	`for_decompress_i32[(1000000, "1M")]`	N/A	4.1 ms	N/A
🆕	Simulation	`for_decompress_i32[(10000000, "10M")]`	N/A	58 ms	N/A
🆕	Simulation	`for_decompress_i64[(100000, "100K")]`	N/A	1.5 ms	N/A
🆕	Simulation	`for_decompress_u32[(1000000, "1M")]`	N/A	4.1 ms	N/A
🆕	Simulation	`for_decompress_i64[(10000000, "10M")]`	N/A	116 ms	N/A
🆕	Simulation	`for_decompress_u64[(100000, "100K")]`	N/A	1.5 ms	N/A
🆕	Simulation	`for_decompress_u64[(10000000, "10M")]`	N/A	116 ms	N/A
🆕	Simulation	`for_decompress_u64[(1000000, "1M")]`	N/A	11.4 ms	N/A
🆕	Simulation	`for_decompress_u32[(100000, "100K")]`	N/A	742.1 µs	N/A
🆕	Simulation	`for_decompress_u32[(10000000, "10M")]`	N/A	58 ms	N/A

_{Comparing metal (cf80187) with develop (1bb7e68)}

1466 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩