Skip to content

Actions: liuliu/ccv

Actions

analyze

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
analyze.yaml
45 workflow run results
45 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Add ccv_cnnp_debug. analyze #45: Commit ef7bd53 pushed by liuliu
September 28, 2024 05:59 6h 20m 45s unstable
unstable
September 28, 2024 05:59 6h 20m 45s
Fix a typo. analyze #44: Commit d2622a3 pushed by liuliu
September 16, 2024 04:50 1h 34m 29s unstable
unstable
September 16, 2024 04:50 1h 34m 29s
Fix bug where the shader cache is not used properly. analyze #43: Commit 67887cd pushed by liuliu
September 16, 2024 04:46 10m 19s unstable
unstable
September 16, 2024 04:46 10m 19s
Fix a bug on the flag is not inspected properly. analyze #42: Commit d53469e pushed by liuliu
September 16, 2024 00:12 2h 51m 42s unstable
unstable
September 16, 2024 00:12 2h 51m 42s
Make sure we do low precision intermediate the same way as Swift repo. analyze #41: Commit eda29de pushed by liuliu
September 15, 2024 23:47 2h 12m 30s unstable
unstable
September 15, 2024 23:47 2h 12m 30s
Temporarily gate against BF16. analyze #40: Commit ffd6604 pushed by liuliu
September 15, 2024 22:39 1h 34m 43s unstable
unstable
September 15, 2024 22:39 1h 34m 43s
Switch SDPA to default to high precision and only you can optionally … analyze #39: Commit 2134068 pushed by liuliu
September 15, 2024 22:04 1h 20m 10s unstable
unstable
September 15, 2024 22:04 1h 20m 10s
Add code to support switching load offset to be computed immediately … analyze #38: Commit 546e9bc pushed by liuliu
September 15, 2024 21:52 50m 42s unstable
unstable
September 15, 2024 21:52 50m 42s
Pass in lse from the op. analyze #37: Commit a58acd2 pushed by liuliu
September 15, 2024 16:57 10m 23s unstable
unstable
September 15, 2024 16:57 10m 23s
Fix a bug caused low precision intermediates not working. analyze #36: Commit c67441f pushed by liuliu
September 14, 2024 18:17 10m 8s unstable
unstable
September 14, 2024 18:17 10m 8s
Integrated into mfa call flow. analyze #35: Commit 6737683 pushed by liuliu
September 13, 2024 23:59 41m 57s unstable
unstable
September 13, 2024 23:59 41m 57s
Add AttentionDescriptor+Parameters and still passes. analyze #34: Commit 290be87 pushed by liuliu
September 13, 2024 19:46 10m 22s unstable
unstable
September 13, 2024 19:46 10m 22s
Pass square_attention_test. analyze #33: Commit 3cc2117 pushed by liuliu
September 13, 2024 18:23 10m 26s unstable
unstable
September 13, 2024 18:23 10m 26s
Further fix translation errors now forward with this particular confi… analyze #32: Commit d71a4f5 pushed by liuliu
September 13, 2024 05:12 1h 5m 5s unstable
unstable
September 13, 2024 05:12 1h 5m 5s
Fix some obvious issues. analyze #31: Commit 20bba89 pushed by liuliu
September 13, 2024 02:51 31m 36s unstable
unstable
September 13, 2024 02:51 31m 36s
Add translated AttentionKernel. Need to do validate on the files. analyze #30: Commit 6ac4534 pushed by liuliu
September 12, 2024 00:02 51m 44s unstable
unstable
September 12, 2024 00:02 51m 44s
Speculative update to explicitly wait for async copy done when store. analyze #29: Commit ca70311 pushed by liuliu
August 20, 2024 18:23 25m 34s unstable
unstable
August 20, 2024 18:23 25m 34s
Force the kernel selection to be on registerPrecisionC = FP32 only. analyze #28: Commit 34bad96 pushed by liuliu
August 19, 2024 17:52 49m 22s unstable
unstable
August 19, 2024 17:52 49m 22s
Explicitly set register_float. analyze #27: Commit 0181946 pushed by liuliu
August 18, 2024 23:26 1h 35m 52s unstable
unstable
August 18, 2024 23:26 1h 35m 52s
Default to accumulator at FP32. analyze #26: Commit 88ef7bc pushed by liuliu
August 18, 2024 21:53 2m 16s unstable
unstable
August 18, 2024 21:53 2m 16s
Make MPS GEMM more flexible on batch stride. analyze #25: Commit ddd3f97 pushed by liuliu
August 18, 2024 05:38 8m 27s unstable
unstable
August 18, 2024 05:38 8m 27s
For M3 / M4, leading block dimensions always fix to 32. analyze #24: Commit 8d6197e pushed by liuliu
August 16, 2024 23:53 14m 58s unstable
unstable
August 16, 2024 23:53 14m 58s
Test some shapes. analyze #23: Commit cde0b15 pushed by liuliu
August 16, 2024 17:58 18m 30s unstable
unstable
August 16, 2024 17:58 18m 30s
Added new GEMM MFA implementation that is optimized for M3 / M4 devices. analyze #22: Commit 5c5bc18 pushed by liuliu
August 16, 2024 00:08 6h 9m 39s unstable
unstable
August 16, 2024 00:08 6h 9m 39s
Change the flags for gemm and how sdpa upcast should work. analyze #21: Commit 6c30517 pushed by liuliu
August 12, 2024 16:28 19m 2s unstable
unstable
August 12, 2024 16:28 19m 2s