{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":444266161,"defaultBranch":"main","name":"hidet","ownerLogin":"hidet-org","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2022-01-04T02:52:45.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/118418068?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1712340322.0","currentOid":""},"activityList":{"items":[{"before":"b4f132358dc2ce73be08f1389003a9b47e4272e8","after":"1035fcb37db75227b3de9ddddf8236772c75780f","ref":"refs/heads/cublas_benchmarks","pushedAt":"2024-04-05T18:22:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"[CuBLAS] Add CuBLAS benchmarks\n\nSome CuBLAS benchmarking results on RTX2080 TI (all measurements are median latencies):\n\nSECTION 1\nFP32 Matrix Multiply: C (bs x m x n) = A (bs x m x k) @ B(bs x k x n)\n\nGroup 1 results with m = 512, n = 512, k = 512\nbs = 1:\ncublas_batched_gemm            69.0us\ncublas_strided_gemm            41.0us\nhidet.ops.matmul optimized     37.0us\nPyTorch                        44.6us\n\nbs = 2:\ncublas_batched_gemm            111.7us\ncublas_strided_gemm            75.8us\nhidet.ops.matmul optimized     69.2us\nPyTorch                        71.7us\n\nbs = 4:\ncublas_batched_gemm            124.9us\ncublas_strided_gemm            97.2us\nhidet.ops.matmul optimized     100.8us\nPyTorch                        96.3us\n\nbs = 8:\ncublas_batched_gemm            190.5us\ncublas_strided_gemm            191.1us\nhidet.ops.matmul optimized     204.7us\nPyTorch                        187.6us\n\nGroup 2 results with m = 1024, n = 1024, k = 2048\nbs = 1:\ncublas_batched_gemm            405.1us\ncublas_strided_gemm            419.2us\nhidet.ops.matmul optimized     370.7us\nPyTorch                        405.1us\n\nbs = 2:\ncublas_batched_gemm            725.3us\ncublas_strided_gemm            859.9us\nhidet.ops.matmul optimized     800.8us\nPyTorch                        719.2us\n\nbs = 4:\ncublas_batched_gemm            1442us\ncublas_strided_gemm            1592us\nhidet.ops.matmul optimized     1606us\nPyTorch                        1466us\n\nbs = 8:\ncublas_batched_gemm            2658us\ncublas_strided_gemm            2830us\nhidet.ops.matmul optimized     3475us\nPyTorch                        2753us\n\nSECTION 2\nFP16 Matrix Multiply: C (bs x m x n) = A (bs x m x k) @ B(bs x k x n)\n\nGroup 1 results with m = 512, n = 512, k = 512\nbs = 1:\ncublas_batched_gemm            63.5us\ncublas_strided_gemm            34.0us\nhidet.ops.matmul optimized     34.9us\nPyTorch                        41.0us\n\nbs = 2:\ncublas_batched_gemm            66.0us\ncublas_strided_gemm            30.2us\nhidet.ops.matmul optimized     64.8us\nPyTorch                        45.1us\n\nbs = 4:\ncublas_batched_gemm            72.7us\ncublas_strided_gemm            32.4us\nhidet.ops.matmul optimized     24.4us\nPyTorch                        46.3us\n\nbs = 8:\ncublas_batched_gemm            81.2us\ncublas_strided_gemm            36.2us\nhidet.ops.matmul optimized     38.5us\nPyTorch                        47.8us\n\nGroup 2 results with m = 1024, n = 1024, k = 2048\nbs = 1:\ncublas_batched_gemm            71.0us\ncublas_strided_gemm            60.1us\nhidet.ops.matmul optimized     65.5us\nPyTorch                        90.6us\n\nbs = 2:\ncublas_batched_gemm            114.8us\ncublas_strided_gemm            112.3us\nhidet.ops.matmul optimized     123.1us\nPyTorch                        160.5us\n\nbs = 4:\ncublas_batched_gemm            225.1us\ncublas_strided_gemm            223.4us\nhidet.ops.matmul optimized     245.6us\nPyTorch                        319.8us\n\nbs = 8:\ncublas_batched_gemm            442.8us\ncublas_strided_gemm            439.1us\nhidet.ops.matmul optimized     733.2us\nPyTorch                        634.8us","shortMessageHtmlLink":"[CuBLAS] Add CuBLAS benchmarks"}},{"before":null,"after":"b4f132358dc2ce73be08f1389003a9b47e4272e8","ref":"refs/heads/cublas_benchmarks","pushedAt":"2024-04-05T18:05:22.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"[CuBLAS] Add CuBLAS benchmarks\n\nSome CuBLAS benchmarking results on RTX2080 TI (all measurements are median latencies):\n\nSECTION 1\nFP32 Matrix Multiply: C (bs x m x n) = A (bs x m x k) @ B(bs x k x n)\n\nGroup 1 results with m = 512, n = 512, k = 512\nbs = 1:\ncublas_batched_gemm            69.0us\ncublas_strided_gemm            41.0us\nhidet.ops.matmul optimized     37.0us\nPyTorch                        44.6us\n\nbs = 2:\ncublas_batched_gemm            111.7us\ncublas_strided_gemm            75.8us\nhidet.ops.matmul optimized     69.2us\nPyTorch                        71.7us\n\nbs = 4:\ncublas_batched_gemm            124.9us\ncublas_strided_gemm            97.2us\nhidet.ops.matmul optimized     100.8us\nPyTorch                        96.3us\n\nbs = 8:\ncublas_batched_gemm            190.5us\ncublas_strided_gemm            191.1us\nhidet.ops.matmul optimized     204.7us\nPyTorch                        187.6us\n\nGroup 2 results with m = 1024, n = 1024, k = 2048\nbs = 1:\ncublas_batched_gemm            405.1us\ncublas_strided_gemm            419.2us\nhidet.ops.matmul optimized     370.7us\nPyTorch                        405.1us\n\nbs = 2:\ncublas_batched_gemm            725.3us\ncublas_strided_gemm            859.9us\nhidet.ops.matmul optimized     800.8us\nPyTorch                        719.2us\n\nbs = 4:\ncublas_batched_gemm            1442us\ncublas_strided_gemm            1592us\nhidet.ops.matmul optimized     1606us\nPyTorch                        1466us\n\nbs = 8:\ncublas_batched_gemm            2658us\ncublas_strided_gemm            2830us\nhidet.ops.matmul optimized     3475us\nPyTorch                        2753us\n\nSECTION 2\nFP16 Matrix Multiply: C (bs x m x n) = A (bs x m x k) @ B(bs x k x n)\n\nGroup 1 results with m = 512, n = 512, k = 512\nbs = 1:\ncublas_batched_gemm            63.5us\ncublas_strided_gemm            34.0us\nhidet.ops.matmul optimized     34.9us\nPyTorch                        41.0us\n\nbs = 2:\ncublas_batched_gemm            66.0us\ncublas_strided_gemm            30.2us\nhidet.ops.matmul optimized     64.8us\nPyTorch                        45.1us\n\nbs = 4:\ncublas_batched_gemm            72.7us\ncublas_strided_gemm            32.4us\nhidet.ops.matmul optimized     24.4us\nPyTorch                        46.3us\n\nbs = 8:\ncublas_batched_gemm            81.2us\ncublas_strided_gemm            36.2us\nhidet.ops.matmul optimized     38.5us\nPyTorch                        47.8us\n\nGroup 2 results with m = 1024, n = 1024, k = 2048\nbs = 1:\ncublas_batched_gemm            71.0us\ncublas_strided_gemm            60.1us\nhidet.ops.matmul optimized     65.5us\nPyTorch                        90.6us\n\nbs = 2:\ncublas_batched_gemm            114.8us\ncublas_strided_gemm            112.3us\nhidet.ops.matmul optimized     123.1us\nPyTorch                        160.5us\n\nbs = 4:\ncublas_batched_gemm            225.1us\ncublas_strided_gemm            223.4us\nhidet.ops.matmul optimized     245.6us\nPyTorch                        319.8us\n\nbs = 8:\ncublas_batched_gemm            442.8us\ncublas_strided_gemm            439.1us\nhidet.ops.matmul optimized     733.2us\nPyTorch                        634.8us","shortMessageHtmlLink":"[CuBLAS] Add CuBLAS benchmarks"}},{"before":"33d8bddb56e30f336f127590ca13b6995cd44ca1","after":"531b8d35d2c6fde0418991f245ce05d95be8b19c","ref":"refs/heads/main","pushedAt":"2024-04-03T15:44:16.000Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"[Version] Bump version to v0.4.0.dev (#443)","shortMessageHtmlLink":"[Version] Bump version to v0.4.0.dev (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2223279099\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/443\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/443/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/443\">#443</a>)"}},{"before":null,"after":"0986cd965d5d052594cbd7a09c6222811dcd40b9","ref":"refs/heads/bump","pushedAt":"2024-04-03T15:43:34.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"Bump version to v0.4.0.dev","shortMessageHtmlLink":"Bump version to v0.4.0.dev"}},{"before":"c53af567289508ae60088802a602fbda012acad1","after":"33d8bddb56e30f336f127590ca13b6995cd44ca1","ref":"refs/heads/main","pushedAt":"2024-04-03T14:47:15.000Z","pushType":"pr_merge","commitsCount":18,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"[RC] Release candidate for version 0.3.1 (#442)","shortMessageHtmlLink":"[RC] Release candidate for version 0.3.1 (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2222792706\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/442\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/442/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/442\">#442</a>)"}},{"before":null,"after":"df05f8338f450c95a5e7276b5bfee6bbd63c18fe","ref":"refs/heads/rc0.3.1","pushedAt":"2024-04-03T12:24:10.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"bump version to 0.3.1","shortMessageHtmlLink":"bump version to 0.3.1"}},{"before":"c7a22ebdc703088766bbf4571bb5158585ed18fd","after":"c53af567289508ae60088802a602fbda012acad1","ref":"refs/heads/main","pushedAt":"2024-03-20T22:56:33.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"[Fix] Skip a failed test due to huggingface transformers update (#439)\n\nFix the transformers version to 4.37 to pass the CI (the new version of\r\ntransformers introduced some breaking change).\r\n\r\nTodo: update our modeling to make it compatible with the new version of\r\ntransformers.","shortMessageHtmlLink":"[Fix] Skip a failed test due to huggingface transformers update (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2184794184\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/439\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/439/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/439\">#439</a>)"}},{"before":"f7ce7ef087188d789083a49a19472312f5e7a4fe","after":"68192fdfd9a5174d20daf005adc102c21976ed0f","ref":"refs/heads/cudnn_conv","pushedAt":"2024-03-14T23:04:55.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"[CUDNN] Add CuDNN performance benchmarks","shortMessageHtmlLink":"[CUDNN] Add CuDNN performance benchmarks"}},{"before":"ab1c51d841e91201dd81132bd71e1576c6f8e9b0","after":"f7ce7ef087188d789083a49a19472312f5e7a4fe","ref":"refs/heads/cudnn_conv","pushedAt":"2024-03-14T22:54:52.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"[CUDNN] Add CuDNN performance benchmarks","shortMessageHtmlLink":"[CUDNN] Add CuDNN performance benchmarks"}},{"before":"86d9aca17a9d93ef77796387e762d2cf79dddc33","after":"ab1c51d841e91201dd81132bd71e1576c6f8e9b0","ref":"refs/heads/cudnn_conv","pushedAt":"2024-03-14T22:51:09.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"[CUDNN] Add CuDNN performance benchmarks","shortMessageHtmlLink":"[CUDNN] Add CuDNN performance benchmarks"}},{"before":"0dc5e53f97ee58870d467e2ea83686ba4e12ba42","after":"5e0a3d2b55ae9b7a19404756639bed044ed2156f","ref":"refs/heads/yaoyaoding-patch-1","pushedAt":"2024-03-14T15:16:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"Update test_llama.py","shortMessageHtmlLink":"Update test_llama.py"}},{"before":"a694d2b5b0f92ac39d7e65361a7d6389ad00e246","after":"0dc5e53f97ee58870d467e2ea83686ba4e12ba42","ref":"refs/heads/yaoyaoding-patch-1","pushedAt":"2024-03-14T15:16:04.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"Update requirements-dev.txt","shortMessageHtmlLink":"Update requirements-dev.txt"}},{"before":null,"after":"a694d2b5b0f92ac39d7e65361a7d6389ad00e246","ref":"refs/heads/yaoyaoding-patch-1","pushedAt":"2024-03-13T20:02:55.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"[Fix] Skip a failed test due to huggingface transformers update\n\nThis PR skips llama modeling test becase the huggingface transformers have updated the modeling of llama. \r\n\r\nTodo: update our modeling to make it compatible with the new version of transformers.","shortMessageHtmlLink":"[Fix] Skip a failed test due to huggingface transformers update"}},{"before":"b7c9026116caa66657a8074d3ee470c029e3366d","after":"c7a22ebdc703088766bbf4571bb5158585ed18fd","ref":"refs/heads/main","pushedAt":"2024-03-07T21:55:42.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"hjjq","name":"Hanjie","path":"/hjjq","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/50634613?s=80&v=4"},"commit":{"message":"[CI] Add back tests (#436)","shortMessageHtmlLink":"[CI] Add back tests (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2174874794\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/436\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/436/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/436\">#436</a>)"}},{"before":"075da22d73f709921ace2fb87bad70b057ffbcbb","after":"86d9aca17a9d93ef77796387e762d2cf79dddc33","ref":"refs/heads/cudnn_conv","pushedAt":"2024-03-07T02:28:07.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"Add cudnn conv2d","shortMessageHtmlLink":"Add cudnn conv2d"}},{"before":null,"after":"075da22d73f709921ace2fb87bad70b057ffbcbb","ref":"refs/heads/cudnn_conv","pushedAt":"2024-03-07T02:24:36.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"yudi0201","name":null,"path":"/yudi0201","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/82490690?s=80&v=4"},"commit":{"message":"Add cudnn conv2d","shortMessageHtmlLink":"Add cudnn conv2d"}},{"before":"bcb57b2631dae5c727da6caa7e50ce9fa789c4ea","after":"8d719e838abc967eec19aac0495776686ef7c042","ref":"refs/heads/compile_opt","pushedAt":"2024-03-04T21:47:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"},"commit":{"message":"[Compilation] Check uniqueness of function names after regrouping","shortMessageHtmlLink":"[Compilation] Check uniqueness of function names after regrouping"}},{"before":"51fb21a1e0f149bf60731772f55ec4d99c5b1d7d","after":"bcb57b2631dae5c727da6caa7e50ce9fa789c4ea","ref":"refs/heads/compile_opt","pushedAt":"2024-03-04T21:36:24.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"},"commit":{"message":"[Compilation] Check uniqueness of function names after regrouping","shortMessageHtmlLink":"[Compilation] Check uniqueness of function names after regrouping"}},{"before":"06cf139c74584858a60bb81094972c0b3b389ec7","after":"b7c9026116caa66657a8074d3ee470c029e3366d","ref":"refs/heads/main","pushedAt":"2024-03-04T17:39:11.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"[Fixbug] Fix graph metadata hash (#428)\n\nSame graphs should have the same hash. Graph node traversal used in\r\ngenerating hash is non-deterministic, leading to different hashes for\r\nthe same graph.\r\n\r\nThis can prevent graph runs from using the fast path, since dispatch\r\ntables cannot be found for the current (different) hash!","shortMessageHtmlLink":"[Fixbug] Fix graph metadata hash (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2145150691\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/428\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/428/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/428\">#428</a>)"}},{"before":"140830056c4be5bbe469b46286987d4102de90cb","after":null,"ref":"refs/heads/transpose","pushedAt":"2024-03-01T22:50:15.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"zhiwei-fang","name":null,"path":"/zhiwei-fang","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/156464430?s=80&v=4"}},{"before":"8befb628b737874b489fbea2dee94237f8f0415d","after":"140830056c4be5bbe469b46286987d4102de90cb","ref":"refs/heads/transpose","pushedAt":"2024-03-01T04:51:28.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"zhiwei-fang","name":null,"path":"/zhiwei-fang","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/156464430?s=80&v=4"},"commit":{"message":"transpose 2d v1","shortMessageHtmlLink":"transpose 2d v1"}},{"before":"da5894bcc95c65a1d0d498bd358f3b0e78675477","after":"06cf139c74584858a60bb81094972c0b3b389ec7","ref":"refs/heads/main","pushedAt":"2024-02-27T20:23:29.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"hjjq","name":"Hanjie","path":"/hjjq","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/50634613?s=80&v=4"},"commit":{"message":"[CI] CI migration (#433)","shortMessageHtmlLink":"[CI] CI migration (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2157534511\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/433\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/433/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/433\">#433</a>)"}},{"before":"5f48983419127c50b5ae6614fb8fbbade6e1ec01","after":"da5894bcc95c65a1d0d498bd358f3b0e78675477","ref":"refs/heads/main","pushedAt":"2024-02-22T17:48:50.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"hjjq","name":"Hanjie","path":"/hjjq","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/50634613?s=80&v=4"},"commit":{"message":"[CI] Use slurm for runners (#430)","shortMessageHtmlLink":"[CI] Use slurm for runners (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2149595816\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/430\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/430/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/430\">#430</a>)"}},{"before":"f3ccb8763d18d23fe2c18e2e0310234c346af0eb","after":"5f48983419127c50b5ae6614fb8fbbade6e1ec01","ref":"refs/heads/main","pushedAt":"2024-02-21T18:53:34.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"hjjq","name":"Hanjie","path":"/hjjq","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/50634613?s=80&v=4"},"commit":{"message":"[Compile Server] Fetch repo before checking out (#429)\n\nMinor bug fix.","shortMessageHtmlLink":"[Compile Server] Fetch repo before checking out (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2147471740\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/429\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/429/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/429\">#429</a>)"}},{"before":"5f76caf035bbd29e336db32c36f8ddc1d0727bc2","after":"f3ccb8763d18d23fe2c18e2e0310234c346af0eb","ref":"refs/heads/main","pushedAt":"2024-02-21T18:39:14.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yaoyaoding","name":"Yaoyao Ding","path":"/yaoyaoding","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/23381083?s=80&v=4"},"commit":{"message":"[Fixbug] Fix dynamic memcpy bug (#427)\n\nMinimal failure case:\r\n\r\n```\r\nresize_inputs: Tensor = symbol([1, 3, \"h\", \"w\"], dtype=\"int32\", device=\"cpu\")\r\nresize_outputs = self.resize(resize_inputs.to(self.dtype, self.device))  # (float32, cuda)\r\nresize_graph: FlowGraph = trace_from(resize_outputs, resize_inputs)\r\n\r\nresize_graph.build()\r\n```\r\ncompiles this launch where symbols `h` and `w` are undefined.\r\n\r\n```\r\nDLL void hidet_launch_0(float * __restrict__ x, float * __restrict__ y) {\r\n  cudaMemcpyAsync(y, x, (4 * ((3 * h) * w)), cudaMemcpyHostToDevice, (cudaStream_t)get_cuda_stream());\r\n}\r\n```\r\n\r\nFix is to add exprs to BlackBoxStmt so that symbols defined in exprs can\r\nbe visited during codegen.","shortMessageHtmlLink":"[Fixbug] Fix dynamic memcpy bug (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2145129640\" data-permission-text=\"Title is private\" data-url=\"https://github.com/hidet-org/hidet/issues/427\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/hidet-org/hidet/pull/427/hovercard\" href=\"https://github.com/hidet-org/hidet/pull/427\">#427</a>)"}},{"before":"403a108c72724f779e39c1acc0e8a6fc8bc74dd4","after":"51fb21a1e0f149bf60731772f55ec4d99c5b1d7d","ref":"refs/heads/compile_opt","pushedAt":"2024-02-15T17:53:26.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"},"commit":{"message":"[Compilation] Optimization for compilation process\n\noptimize the compilation behavior for each core. Instead of generating a cuda source file for each core, congragate them into a single source file to reduce compilation overhead","shortMessageHtmlLink":"[Compilation] Optimization for compilation process"}},{"before":"c5fc4dbb41e02052055e013ae21ff7532688468a","after":"403a108c72724f779e39c1acc0e8a6fc8bc74dd4","ref":"refs/heads/compile_opt","pushedAt":"2024-02-15T17:39:44.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"},"commit":{"message":"[Compilation] Optimization for compilation process\n\noptimize the compilation behavior for each core. Instead of generating a cuda source file for each core, congragate them into a single source file to reduce compilation overhead","shortMessageHtmlLink":"[Compilation] Optimization for compilation process"}},{"before":"529030cce2d221c3d3c65a2a7fd88b7added92b7","after":"c5fc4dbb41e02052055e013ae21ff7532688468a","ref":"refs/heads/compile_opt","pushedAt":"2024-02-15T17:18:44.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"},"commit":{"message":"[Compilation] Optimization for compilation process","shortMessageHtmlLink":"[Compilation] Optimization for compilation process"}},{"before":null,"after":"529030cce2d221c3d3c65a2a7fd88b7added92b7","ref":"refs/heads/compile_opt","pushedAt":"2024-02-13T22:28:26.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"},"commit":{"message":"[Compilation] Optimization for compilation process","shortMessageHtmlLink":"[Compilation] Optimization for compilation process"}},{"before":"94d26dcab57674da7107aa46294c401d97166b62","after":null,"ref":"refs/heads/compile_opt","pushedAt":"2024-02-13T19:39:06.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"hunbssfy","name":"Max Hu","path":"/hunbssfy","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6930027?s=80&v=4"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEKSeZUQA","startCursor":null,"endCursor":null}},"title":"Activity · hidet-org/hidet"}