{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":127913487,"defaultBranch":"master","name":"code_saturne","ownerLogin":"code-saturne","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2018-04-03T13:35:03.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/38041830?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1717358289.0","currentOid":""},"activityList":{"items":[{"before":"fb95dbf22110b777bd15e12f4b2dbab426efc6a9","after":"93de3e1018ce6789c342315bf29dada9cbb43345","ref":"refs/heads/master","pushedAt":"2024-06-09T23:25:28.000Z","pushType":"push","commitsCount":7,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Multigrid: allow tuning max level for device (i.e. GPU).\n\nFor the time being, this is a global setting, defined using\n`cs_multigrid_set_max_grid_level_for_device.`","shortMessageHtmlLink":"Multigrid: allow tuning max level for device (i.e. GPU)."}},{"before":"7b5cd965a63a16c88941de2589ef31fed9fbbab0","after":"fb95dbf22110b777bd15e12f4b2dbab426efc6a9","ref":"refs/heads/master","pushedAt":"2024-06-09T19:36:40.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix crash in debug mode in zero-layer mesh extrusion.","shortMessageHtmlLink":"Fix crash in debug mode in zero-layer mesh extrusion."}},{"before":"de11c98e0d70ab6726808b524e74cd0527345e86","after":"7b5cd965a63a16c88941de2589ef31fed9fbbab0","ref":"refs/heads/master","pushedAt":"2024-06-08T13:17:59.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix compilation with gcc 14.1, with a few additional warning and style fixes.","shortMessageHtmlLink":"Fix compilation with gcc 14.1, with a few additional warning and styl…"}},{"before":"9bff4ff5009876a697dcb0e36a989c5b000a666e","after":"de11c98e0d70ab6726808b524e74cd0527345e86","ref":"refs/heads/master","pushedAt":"2024-06-08T11:47:24.000Z","pushType":"push","commitsCount":14,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"SpMV tuning: use fixed number of iterations.\n\nAlso initialize each SpMV with untimed dry run to avoid possible perturbation\nof measures by external library initialization time.","shortMessageHtmlLink":"SpMV tuning: use fixed number of iterations."}},{"before":"a9e01e6ab0bc69c57c67ae46599f96b09f4f14d9","after":"9bff4ff5009876a697dcb0e36a989c5b000a666e","ref":"refs/heads/master","pushedAt":"2024-06-04T16:37:28.000Z","pushType":"push","commitsCount":18,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Multigrid parirwise aggregation: fix hang in some zero local cell cases.","shortMessageHtmlLink":"Multigrid parirwise aggregation: fix hang in some zero local cell cases."}},{"before":"66e34584da74e517d6b799936276789d4dfae998","after":"14d65be831e858eb4e48f545811b4349c8e9f7a5","ref":"refs/heads/v8.1","pushedAt":"2024-06-02T19:58:09.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix extra preprocessing for staged restarts.\n\nCherry picked from commits 5643c02e, 90451b87, and 35ce78f9.\n\nWhen a restart was to use an unmodified mesh (\"restart/mesh_input.csm\"),\nthe case domain's self.mesh_input entry was set in the \"stage\" step,\nbut not restored in the \"preprocess\" step in case of a 2-step submit,\nleading to extra preprocessing which could lead to inconsistent\nrestart errors.\n\nWe now check this case so as to avoid unwanted preprocessing.\n\nWhen restarting with no mesh change, add previous mesh_input to checkpoint.\n\nThis fixes GitLab issue #579.","shortMessageHtmlLink":"Fix extra preprocessing for staged restarts."}},{"before":"e4136e7411b64abf75b8b7d0b8f568373a89fc53","after":"3337bbf280af2e38619973bed13016bf7690e355","ref":"refs/heads/v8.0","pushedAt":"2024-06-02T19:58:09.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix extra preprocessing for staged restarts. starts.\n    \nWhen a restart was to use an unmodified mesh (\"restart/mesh_input.csm\"),\nthe case domain's self.mesh_input entry was set in the \"stage\" step,\nbut not restored in the \"preprocess\" step in case of a 2-step submit,\nleading to extra preprocessing which could lead to inconsistent\nrestart errors.\n    \nWe now check this case so as to avoid unwanted preprocessing.\n    \nThis fixes GitLab issue #579.","shortMessageHtmlLink":"Fix extra preprocessing for staged restarts. starts."}},{"before":"28d4cdec46b27e99b3e17c0a503389ed65993614","after":"a9e01e6ab0bc69c57c67ae46599f96b09f4f14d9","ref":"refs/heads/master","pushedAt":"2024-06-02T19:58:09.000Z","pushType":"push","commitsCount":6,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Improve setting of ascent levels for multigrid V cycle preconditioner.\n\nWhen running a single cycle, as in preconditioning, allow ascent smoothing\nat the finest level instead of starting a new cycle. This allows correctly\nhandling different number of descent and ascent cycles at that level,\nand avoids an extra SpMV and restriction.","shortMessageHtmlLink":"Improve setting of ascent levels for multigrid V cycle preconditioner."}},{"before":"33ac33dc8328ba3a62b6be016d73660405c4e6b0","after":"28d4cdec46b27e99b3e17c0a503389ed65993614","ref":"refs/heads/master","pushedAt":"2024-05-29T23:47:01.000Z","pushType":"push","commitsCount":7,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Add missing test (probably important only for readability/comprehension).","shortMessageHtmlLink":"Add missing test (probably important only for readability/comprehensi…"}},{"before":"0bcda372d445e6367a9fbc503cddd653698aa326","after":"33ac33dc8328ba3a62b6be016d73660405c4e6b0","ref":"refs/heads/master","pushedAt":"2024-05-28T19:50:18.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"[GPU dispatch] dispatch cs_upwind_gradient","shortMessageHtmlLink":"[GPU dispatch] dispatch cs_upwind_gradient"}},{"before":"8697c09136fb8ee67a5c80ce85a99f8e7c45d579","after":"0bcda372d445e6367a9fbc503cddd653698aa326","ref":"refs/heads/master","pushedAt":"2024-05-28T00:06:51.000Z","pushType":"push","commitsCount":8,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix base multigrid behavior on GPU.\n\n- Use cs_dispatch for some multigrid operations.\n- Enable fine levels to tun on GPU and coarse levels on CPU.\n- Automatically replace Gauss-Seidel with Jacobi on GPU.\n\nFurther tuning will be needed, but the algorithm at least seems\nfunctional.","shortMessageHtmlLink":"Fix base multigrid behavior on GPU."}},{"before":"a2fb49541557b81bfd79086b9ea75b0f5c5cffc8","after":"8697c09136fb8ee67a5c80ce85a99f8e7c45d579","ref":"refs/heads/master","pushedAt":"2024-05-26T19:45:59.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Avoid macros leading to redundant computations in wall functions.","shortMessageHtmlLink":"Avoid macros leading to redundant computations in wall functions."}},{"before":"9b28f11805b3c7fb3db48ac4ec7aefe00c5a6c56","after":"a2fb49541557b81bfd79086b9ea75b0f5c5cffc8","ref":"refs/heads/master","pushedAt":"2024-05-26T17:37:05.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fixes for SYCL.","shortMessageHtmlLink":"Fixes for SYCL."}},{"before":"0e51528e2d3cb14778541108b98ff1be0c2f765d","after":"9b28f11805b3c7fb3db48ac4ec7aefe00c5a6c56","ref":"refs/heads/master","pushedAt":"2024-05-24T17:44:41.000Z","pushType":"push","commitsCount":8,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Enable GPU dispatch for unsteady tensor convection+diffusion.\n\nThe code is not tested on GPU yet, only on CPU, so it is probable\nthe memory mapping of some variables in the tensor call paths\nwill need to be adapted to host/device shared allocations.","shortMessageHtmlLink":"Enable GPU dispatch for unsteady tensor convection+diffusion."}},{"before":"0c8d8694778d85b9081616ae4e575791b98c3799","after":"0e51528e2d3cb14778541108b98ff1be0c2f765d","ref":"refs/heads/master","pushedAt":"2024-05-24T00:05:28.000Z","pushType":"push","commitsCount":14,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"cs_convection_diffusion_vector: dispatch fixes, improvements, and refactoring.\n\n- More code is shared between the (deprecated) steady and unsteady algorithms,\n  and the transpose gradient is moved back to the vector-only function rather\n  than in the generic strided version.\n- The transpose gradient term is now computed using the dispatch context,\n  with an initial part using a second dispatch for better concurrency.\n- Some missing context wait() calls ware added.","shortMessageHtmlLink":"cs_convection_diffusion_vector: dispatch fixes, improvements, and ref…"}},{"before":"15bc7f4d6ea5a280725d2fa0b5709aea7e8e5087","after":"0c8d8694778d85b9081616ae4e575791b98c3799","ref":"refs/heads/master","pushedAt":"2024-05-22T22:11:26.000Z","pushType":"push","commitsCount":13,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix crash due to typos in commit c0460077 (2024-05-17).","shortMessageHtmlLink":"Fix crash due to typos in commit <a class=\"commit-link\" data-hovercard-type=\"commit\" data-hovercard-url=\"https://github.com/code-saturne/code_saturne/commit/c046007720debe07f578f4799278649263f578f6/hovercard\" href=\"https://github.com/code-saturne/code_saturne/commit/c046007720debe07f578f4799278649263f578f6\"><tt>c046007</tt></a> (2024-05-17)."}},{"before":"c046007720debe07f578f4799278649263f578f6","after":"15bc7f4d6ea5a280725d2fa0b5709aea7e8e5087","ref":"refs/heads/master","pushedAt":"2024-05-18T00:06:11.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Add timers and remove now unused code.","shortMessageHtmlLink":"Add timers and remove now unused code."}},{"before":"a88600a9cdd759029616a33e91c61690f532be5d","after":"c046007720debe07f578f4799278649263f578f6","ref":"refs/heads/master","pushedAt":"2024-05-17T18:53:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"First step of GPU dispatch for vector quantities.\n\nThe code is probably not complete, and only tested on the BUNDLE\ntest case configuration for now, but can provide an example\nfor further work on dispatch to GPU.","shortMessageHtmlLink":"First step of GPU dispatch for vector quantities."}},{"before":"ccf1500de8992edf149704e8cc92d3c3ecf5f096","after":"a88600a9cdd759029616a33e91c61690f532be5d","ref":"refs/heads/master","pushedAt":"2024-05-17T10:51:26.000Z","pushType":"push","commitsCount":14,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix compilation issue in some cases, from inconsistent inline in commit 451b924c.","shortMessageHtmlLink":"Fix compilation issue in some cases, from inconsistent inline in commit "}},{"before":"ba0cfb42f255744c7261c97b7279321875353fd3","after":"ccf1500de8992edf149704e8cc92d3c3ecf5f096","ref":"refs/heads/master","pushedAt":"2024-05-14T16:38:15.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Add missing dispatch wait calls.\n\nAlso, use temporary arry for upwind face counts in\ncs_convection_diffusion_vector, so as to prepare for general dispatch.\nThis will be a bit more costly when logging is required, but les costly in\nthe more common case where verbosity is not increased, as we do not then\nneed a reduction.","shortMessageHtmlLink":"Add missing dispatch wait calls."}},{"before":"5c0d1ffcf648a7df5ba4a3a3554f5fea2384799b","after":"ba0cfb42f255744c7261c97b7279321875353fd3","ref":"refs/heads/master","pushedAt":"2024-05-13T23:01:32.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix misplaced #endif for non-CUDA accelerated code.","shortMessageHtmlLink":"Fix misplaced #endif for non-CUDA accelerated code."}},{"before":"5894a97c041bf6965e193e54170a5d14ab7e6ada","after":"5c0d1ffcf648a7df5ba4a3a3554f5fea2384799b","ref":"refs/heads/master","pushedAt":"2024-05-13T16:44:48.000Z","pushType":"push","commitsCount":17,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"CDO: CIP scaling coefficient is now associated to an equation.\n\nThis can be user-defined now and modified equation by equation when\nthe CIP advection scheme is used in a CDO VCb space scheme.","shortMessageHtmlLink":"CDO: CIP scaling coefficient is now associated to an equation."}},{"before":"0a2dfca48fb7388dff472d35b5eeebe992266e5c","after":"5894a97c041bf6965e193e54170a5d14ab7e6ada","ref":"refs/heads/master","pushedAt":"2024-05-12T22:58:26.000Z","pushType":"push","commitsCount":6,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"CUDA: fix memory access errors for Green-Gauss gradient reconstruction.","shortMessageHtmlLink":"CUDA: fix memory access errors for Green-Gauss gradient reconstruction."}},{"before":"8cc40dd7beba43f23a72b1a5b2a45b660b4689a0","after":"0a2dfca48fb7388dff472d35b5eeebe992266e5c","ref":"refs/heads/master","pushedAt":"2024-05-07T21:01:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Convert cs_coal_fp2st from Fortran to C.","shortMessageHtmlLink":"Convert cs_coal_fp2st from Fortran to C."}},{"before":"8bbade551bbcd3099a92b1bfa627b6a81f7a42cd","after":"8cc40dd7beba43f23a72b1a5b2a45b660b4689a0","ref":"refs/heads/master","pushedAt":"2024-05-07T17:40:06.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Fix compiler warning.","shortMessageHtmlLink":"Fix compiler warning."}},{"before":"340c87a6df865af612c42c5f9bafa9fa7a43ba12","after":"8bbade551bbcd3099a92b1bfa627b6a81f7a42cd","ref":"refs/heads/master","pushedAt":"2024-05-06T23:50:15.000Z","pushType":"push","commitsCount":5,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Share other imbrication options.","shortMessageHtmlLink":"Share other imbrication options."}},{"before":"67505f94c4a924f9470bd666c1b6bcdaf11af03c","after":"340c87a6df865af612c42c5f9bafa9fa7a43ba12","ref":"refs/heads/master","pushedAt":"2024-05-06T20:16:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Missing part of commit 67505f94 (just now).","shortMessageHtmlLink":"Missing part of commit <a class=\"commit-link\" data-hovercard-type=\"commit\" data-hovercard-url=\"https://github.com/code-saturne/code_saturne/commit/67505f94c4a924f9470bd666c1b6bcdaf11af03c/hovercard\" href=\"https://github.com/code-saturne/code_saturne/commit/67505f94c4a924f9470bd666c1b6bcdaf11af03c\"><tt>67505f9</tt></a> (just now)."}},{"before":"ea7f4b6529d7113218276998c85df2377b085c74","after":"67505f94c4a924f9470bd666c1b6bcdaf11af03c","ref":"refs/heads/master","pushedAt":"2024-05-06T19:59:47.000Z","pushType":"push","commitsCount":5,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"CUDA: avoid some host/device synchronizations for gradient reconstruction.\n\nThis is still a work in progress, and extraneous h2d/d2h copies\nmay still remain.","shortMessageHtmlLink":"CUDA: avoid some host/device synchronizations for gradient reconstruc…"}},{"before":"2f5bbf08fd3c42ffa922f02c3d90d4adbd1cd337","after":"ea7f4b6529d7113218276998c85df2377b085c74","ref":"refs/heads/master","pushedAt":"2024-05-03T19:27:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"Remove spurious print statement.","shortMessageHtmlLink":"Remove spurious print statement."}},{"before":"667d79e3a7decfb2c8d427f472685cea138674af","after":"2f5bbf08fd3c42ffa922f02c3d90d4adbd1cd337","ref":"refs/heads/master","pushedAt":"2024-05-03T18:53:51.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"YvanFournier","name":"Yvan Fournier","path":"/YvanFournier","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/5770167?s=80&v=4"},"commit":{"message":"GPU: try to reduce host/device communication for halos.","shortMessageHtmlLink":"GPU: try to reduce host/device communication for halos."}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEYJXpfAA","startCursor":null,"endCursor":null}},"title":"Activity · code-saturne/code_saturne"}