{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":86621987,"defaultBranch":"master","name":"cuda_scheduling_examiner_mirror","ownerLogin":"yalue","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2017-03-29T19:36:02.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/839991?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1657560540.1541278","currentOid":""},"activityList":{"items":[{"before":"ef7d8b21b842818ffd0004014b6eca42c13686e9","after":"1c0bd05950fb5e168a83fbef2242f43fafac44d3","ref":"refs/heads/master","pushedAt":"2024-06-03T19:44:08.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Don't assume CPU core 1 and 0 are available\n\nThe available CPUs may be arbitrary, such as on systems using a\npartitioned scheduler. CPU core 1 may not be available, and so we\nneed to use the same logic to determine the first-available CPU as\nwe already used to determine all the subsequently-available CPUs.","shortMessageHtmlLink":"Don't assume CPU core 1 and 0 are available"}},{"before":"eafa26dac48968657aaa80850887ed1fbb29e4e2","after":"ef7d8b21b842818ffd0004014b6eca42c13686e9","ref":"refs/heads/master","pushedAt":"2024-05-30T23:56:28.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Python 3 support for `view_blocksbysm.py` via `graphics.py` update (#15)\n\n* Python 3 support for `view_blocksbysm.py` via `graphics.py` update\r\n\r\n`graphics.py` was previously Version 3.5---this was not a real\r\nversion, but something likely created by Tanya off of Version 3.4.\r\n\r\nUnfortunately, \"Version 3.5\" changed the API of `graphics.py`,\r\nmaking `view_blocksbysm.py` incompatible with newer versions.\r\n\r\nThis commit takes Version 4.5 of `graphics.py` from the March 14th,\r\n2011 archive of http://mcsp.wartburg.edu/zelle/python/graphics.py\r\non archive.org, and minimally modifies it to persist the API changes\r\ndepended on by `view_blocksbysm.py`. Note that these modifications\r\n(as before) break image support in `graphics.py`.\r\n\r\n* Remove `gray68` foreground color in `view_blocksbysm.py`\r\n\r\nThis has no matching background or arrow color.\r\n\r\nAlso remove some rapidly repeated colors from the foreground color\r\nlist, as it's better to have a long period before color repetition.\r\n\r\n* Update comments to reflect that all view_*.py scripts support Python 3\r\n\r\n* Support unique color-pattern shading for up to 64 stream\r\n\r\nPreviously would repeat color-pattern shades within less than 20\r\nstreams.\r\n\r\nImplemented by manually enumerating all permutations.","shortMessageHtmlLink":"Python 3 support for <code>view_blocksbysm.py</code> via <code>graphics.py</code> update (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2326195524\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/15\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/15/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/15\">#15</a>)"}},{"before":"c7a05b44412e65d5a28d253b43630fe9431cbc22","after":"eafa26dac48968657aaa80850887ed1fbb29e4e2","ref":"refs/heads/master","pushedAt":"2024-05-30T15:28:25.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Support an arbitrary number of streams in `view_blocksbysm.py` (#14)\n\nNecessary for plots such as Fig. 12 in \"Hardware Compute\r\nPartitioning on NVIDIA GPUs\" by Bakita and Anderson at RTAS'23.\r\n\r\nPreviously only up to 25 streams could be plotted.\r\n\r\nAlso two comment fixes.","shortMessageHtmlLink":"Support an arbitrary number of streams in <code>view_blocksbysm.py</code> (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2325941458\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/14\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/14/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/14\">#14</a>)"}},{"before":"58463825cabed3e845b7f7fa01c1e01f61524f1c","after":"c7a05b44412e65d5a28d253b43630fe9431cbc22","ref":"refs/heads/master","pushedAt":"2024-05-30T14:52:18.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Usability and Python 3 support improvements (#13)\n\n* More Python 3 support and bugfixes in the plotting scripts\r\n\r\n- `view_blocksbysm.py`: Support plotting more simultaneous streams\r\n  by including background and arrow colors for already-defined\r\n  foreground colors, as well as cycle through the pattern map, rather\r\n  than attempting an out-of-bounds access.\r\n- `view_stackplots.py`: Support Python 3, include a legend,\r\n  crosshatch filled regions, add x and y labels, change limit for\r\n  the y axis to always be the total number of GPU threads, add an\r\n  option for plotting as a percent of GPU capacity, add an option\r\n  to save the plot, and switch to argparse-based argument parsing\r\n  (as previously done for `view_blocksbysm.py` and others).\r\n- `view_timelines.py`: Support Python 3 and switch to argparse-based\r\n  argument parsing (adding regex support).\r\n- `view_times_cdf.py`: Add support for Matplotlib 3+, fix a Python 3\r\n  compatibility issue (removal of .next()), and add support for regex-\r\n  -based filename specification.\r\n- `view_times_pdf.py`: Add support for Matplotlib 3+ and fix a Python\r\n  3 compatibility issue (removal of .next()).\r\n\r\n* Shrink `view_blocksbysm.py` block labels\r\n\r\nGPUs now have far more SMs than when this script was originally\r\nwritten, making the labels unreadably overlap in many situations.\r\n\r\nThis removes the space between the kernel name and the block number,\r\nand shrinks the label font size to 10pt. 10pt is the same as the\r\nIEEEtran footnote size, making it acceptably readable when inserted\r\nin a paper (barring any intermediate scaling).","shortMessageHtmlLink":"Usability and Python 3 support improvements (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2325859394\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/13\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/13/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/13\">#13</a>)"}},{"before":"74baca910b4e95a787209c9601f051afca0a36be","after":"58463825cabed3e845b7f7fa01c1e01f61524f1c","ref":"refs/heads/master","pushedAt":"2023-10-24T17:00:14.000Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Merge pull request #12 from JoshuaJB/master\n\nImprove release marker stacking logic in view_blocksbysm.py","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1959650784\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/12\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/12/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/12\">#12</a> from JoshuaJB/master"}},{"before":"f815580a2756508f1529bee567964bc144caed35","after":"74baca910b4e95a787209c9601f051afca0a36be","ref":"refs/heads/master","pushedAt":"2023-07-03T14:25:47.000Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Merge pull request #11 from JoshuaJB/partitioning\n\nGPU partitioning from \"Hardware Compute Partitioning on NVIDIA GPUs\" (RTAS'23)","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1786150175\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/11\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/11/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/11\">#11</a> from JoshuaJB/partitioning"}},{"before":"5d6ca16b7ec38c43f811277025ef12d1e0b51654","after":"f815580a2756508f1529bee567964bc144caed35","ref":"refs/heads/master","pushedAt":"2023-07-03T14:23:39.000Z","pushType":"pr_merge","commitsCount":3,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Merge pull request #10 from JoshuaJB/master\n\nFix linking and more broadly apply the GlobalTimer64() improvements","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1786133174\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/10\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/10/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/10\">#10</a> from JoshuaJB/master"}},{"before":"2c9139a3b12157baba502b1bbc9ed14413cd7124","after":"5d6ca16b7ec38c43f811277025ef12d1e0b51654","ref":"refs/heads/master","pushedAt":"2023-06-29T00:52:24.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Update README.md\n\nNow that there's been some more recent work on this project, it doesn't make as much sense to refer to the hip_plugin_framework project as being more current.","shortMessageHtmlLink":"Update README.md"}},{"before":"4908795a447b94beeb05c0317426860585230f7a","after":"2c9139a3b12157baba502b1bbc9ed14413cd7124","ref":"refs/heads/master","pushedAt":"2023-06-29T00:45:25.000Z","pushType":"pr_merge","commitsCount":3,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Merge pull request #9 from JoshuaJB/master\n\nIncrease flexibility of some commonly used scripts + cleanup","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1779174802\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/9\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/9/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/9\">#9</a> from JoshuaJB/master"}},{"before":"5f94c3e77cfb11d345af8b002a872c8a46fc9dbc","after":"4908795a447b94beeb05c0317426860585230f7a","ref":"refs/heads/master","pushedAt":"2023-06-26T14:41:51.000Z","pushType":"pr_merge","commitsCount":4,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Merge pull request #8 from JoshuaJB/master\n\nImprovements to clocking and timing infrastructure","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1774779433\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/8\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/8/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/8\">#8</a> from JoshuaJB/master"}},{"before":"5b9251bbb6324ebc7ef86efc0414df72b44bc0c2","after":"5f94c3e77cfb11d345af8b002a872c8a46fc9dbc","ref":"refs/heads/master","pushedAt":"2023-06-22T00:32:10.007Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"yalue","name":"Nathan O","path":"/yalue","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/839991?s=80&v=4"},"commit":{"message":"Merge pull request #7 from JoshuaJB/master\n\nRemove a critical dimensioning error in matrix_multiply.cu","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"1768513224\" data-permission-text=\"Title is private\" data-url=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/issues/7\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/yalue/cuda_scheduling_examiner_mirror/pull/7/hovercard\" href=\"https://github.com/yalue/cuda_scheduling_examiner_mirror/pull/7\">#7</a> from JoshuaJB/master"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEW1DeCAA","startCursor":null,"endCursor":null}},"title":"Activity · yalue/cuda_scheduling_examiner_mirror"}