Skip to content

Actions: liuliu/ccv

Actions

cuda-int-tests

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
cuda-int-tests.yaml
46 workflow run results
46 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Add ccv_cnnp_debug. cuda-int-tests #46: Commit ef7bd53 pushed by liuliu
September 28, 2024 05:59 6h 10m 40s unstable
unstable
September 28, 2024 05:59 6h 10m 40s
Fix a typo. cuda-int-tests #45: Commit d2622a3 pushed by liuliu
September 16, 2024 04:50 1h 50m 20s unstable
unstable
September 16, 2024 04:50 1h 50m 20s
Fix bug where the shader cache is not used properly. cuda-int-tests #44: Commit 67887cd pushed by liuliu
September 16, 2024 04:46 41m 32s unstable
unstable
September 16, 2024 04:46 41m 32s
Fix a bug on the flag is not inspected properly. cuda-int-tests #43: Commit d53469e pushed by liuliu
September 16, 2024 00:12 3h 9m 3s unstable
unstable
September 16, 2024 00:12 3h 9m 3s
Make sure we do low precision intermediate the same way as Swift repo. cuda-int-tests #42: Commit eda29de pushed by liuliu
September 15, 2024 23:47 1h 38m 17s unstable
unstable
September 15, 2024 23:47 1h 38m 17s
Temporarily gate against BF16. cuda-int-tests #41: Commit ffd6604 pushed by liuliu
September 15, 2024 22:39 2h 6m 9s unstable
unstable
September 15, 2024 22:39 2h 6m 9s
Switch SDPA to default to high precision and only you can optionally … cuda-int-tests #40: Commit 2134068 pushed by liuliu
September 15, 2024 22:04 1h 36m 6s unstable
unstable
September 15, 2024 22:04 1h 36m 6s
Add code to support switching load offset to be computed immediately … cuda-int-tests #39: Commit 546e9bc pushed by liuliu
September 15, 2024 21:52 40m 19s unstable
unstable
September 15, 2024 21:52 40m 19s
Pass in lse from the op. cuda-int-tests #38: Commit a58acd2 pushed by liuliu
September 15, 2024 16:57 50m 14s unstable
unstable
September 15, 2024 16:57 50m 14s
Fix a bug caused low precision intermediates not working. cuda-int-tests #37: Commit c67441f pushed by liuliu
September 14, 2024 18:17 41m 36s unstable
unstable
September 14, 2024 18:17 41m 36s
Integrated into mfa call flow. cuda-int-tests #36: Commit 6737683 pushed by liuliu
September 13, 2024 23:59 31m 38s unstable
unstable
September 13, 2024 23:59 31m 38s
Add AttentionDescriptor+Parameters and still passes. cuda-int-tests #35: Commit 290be87 pushed by liuliu
September 13, 2024 19:46 1h 5m 30s unstable
unstable
September 13, 2024 19:46 1h 5m 30s
Pass square_attention_test. cuda-int-tests #34: Commit 3cc2117 pushed by liuliu
September 13, 2024 18:23 1h 4m 56s unstable
unstable
September 13, 2024 18:23 1h 4m 56s
Further fix translation errors now forward with this particular confi… cuda-int-tests #33: Commit d71a4f5 pushed by liuliu
September 13, 2024 05:12 22m 15s unstable
unstable
September 13, 2024 05:12 22m 15s
Fix some obvious issues. cuda-int-tests #32: Commit 20bba89 pushed by liuliu
September 13, 2024 02:51 2h 14m 20s unstable
unstable
September 13, 2024 02:51 2h 14m 20s
Add translated AttentionKernel. Need to do validate on the files. cuda-int-tests #31: Commit 6ac4534 pushed by liuliu
September 12, 2024 00:02 17m 48s unstable
unstable
September 12, 2024 00:02 17m 48s
Speculative update to explicitly wait for async copy done when store. cuda-int-tests #30: Commit ca70311 pushed by liuliu
August 20, 2024 18:23 2h 23m 17s unstable
unstable
August 20, 2024 18:23 2h 23m 17s
Force the kernel selection to be on registerPrecisionC = FP32 only. cuda-int-tests #29: Commit 34bad96 pushed by liuliu
August 19, 2024 17:52 1h 11m 33s unstable
unstable
August 19, 2024 17:52 1h 11m 33s
Explicitly set register_float. cuda-int-tests #28: Commit 0181946 pushed by liuliu
August 18, 2024 23:26 1h 23m 26s unstable
unstable
August 18, 2024 23:26 1h 23m 26s
Default to accumulator at FP32. cuda-int-tests #27: Commit 88ef7bc pushed by liuliu
August 18, 2024 21:53 47m 57s unstable
unstable
August 18, 2024 21:53 47m 57s
Make MPS GEMM more flexible on batch stride. cuda-int-tests #26: Commit ddd3f97 pushed by liuliu
August 18, 2024 05:38 12m 28s unstable
unstable
August 18, 2024 05:38 12m 28s
For M3 / M4, leading block dimensions always fix to 32. cuda-int-tests #25: Commit 8d6197e pushed by liuliu
August 16, 2024 23:53 24m 32s unstable
unstable
August 16, 2024 23:53 24m 32s
Test some shapes. cuda-int-tests #24: Commit cde0b15 pushed by liuliu
August 16, 2024 17:58 9m 57s unstable
unstable
August 16, 2024 17:58 9m 57s
Added new GEMM MFA implementation that is optimized for M3 / M4 devices. cuda-int-tests #23: Commit 5c5bc18 pushed by liuliu
August 16, 2024 00:08 6h 14m 25s unstable
unstable
August 16, 2024 00:08 6h 14m 25s
Change the flags for gemm and how sdpa upcast should work. cuda-int-tests #22: Commit 6c30517 pushed by liuliu
August 12, 2024 16:28 6h 38m 20s unstable
unstable
August 12, 2024 16:28 6h 38m 20s