{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":335575155,"defaultBranch":"master","name":"opensbi","ownerLogin":"starfive-tech","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2021-02-03T09:41:45.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/66111875?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1715837090.0","currentOid":""},"activityList":{"items":[{"before":null,"after":"cbe45cb6ab205fa51d6d19bd8bbc9cb9810f952d","ref":"refs/heads/starfive-v1.4-dubhe","pushedAt":"2024-05-16T05:24:36.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"rebeccasf","name":"Rebecca Chang Swee Fun","path":"/rebeccasf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9360972?s=80&v=4"},"commit":{"message":"platform: generic: starfive: Add Dubhe CMO SBI extension\n\nAdd L2 cache flush and L2 cache invalidate SBI calls for Dubhe\n\nSigned-off-by: Chee Hong Ang ","shortMessageHtmlLink":"platform: generic: starfive: Add Dubhe CMO SBI extension"}},{"before":"f00a13f0586c15765ce75f3820d1bbc0c07faba0","after":null,"ref":"refs/tags/JH7110_VF2_6.1_v5.11.4","pushedAt":"2024-03-14T09:06:52.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"andyhu-stf","name":"Andy Hu","path":"/andyhu-stf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/89558762?s=80&v=4"}},{"before":"5b9a160efb75851553bf81b9efec4a6c2bee51e7","after":null,"ref":"refs/tags/JH7110_VF2_515_v5.11.4","pushedAt":"2024-03-14T09:05:54.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"andyhu-stf","name":"Andy Hu","path":"/andyhu-stf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/89558762?s=80&v=4"}},{"before":"a10e0837f215fb4ed25f195f2c511a41deb9c324","after":"8ce81d0fbcbb10d4ad9e924c8cd64925b79e0cb6","ref":"refs/heads/starfive-v1.2-dubhe","pushedAt":"2024-02-28T02:59:19.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"rebeccasf","name":"Rebecca Chang Swee Fun","path":"/rebeccasf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9360972?s=80&v=4"},"commit":{"message":"platform: generic: starfive: Changing naming from jh8100 to dubhe\n\nSigned-off-by: Wei Liang Lim ","shortMessageHtmlLink":"platform: generic: starfive: Changing naming from jh8100 to dubhe"}},{"before":null,"after":"cc7f99946dad984b33259a3db8426deda9989ddd","ref":"refs/heads/rtthread_AMP","pushedAt":"2023-12-27T14:15:37.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"andyhu-stf","name":"Andy Hu","path":"/andyhu-stf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/89558762?s=80&v=4"},"commit":{"message":"sbi: add amp ipi call function support\n\nAdd amp IPI call and set amp data address support.\n\nSigned-off-by: Minda Chen ","shortMessageHtmlLink":"sbi: add amp ipi call function support"}},{"before":"01f315f483e1608997d0c459527547d8568faa05","after":null,"ref":"refs/tags/REL_DUBHE_SEPT_2023","pushedAt":"2023-10-25T02:26:01.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"rebeccasf","name":"Rebecca Chang Swee Fun","path":"/rebeccasf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9360972?s=80&v=4"}},{"before":null,"after":"a10e0837f215fb4ed25f195f2c511a41deb9c324","ref":"refs/heads/starfive-v1.2-dubhe","pushedAt":"2023-10-24T02:57:27.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"rebeccasf","name":"Rebecca Chang Swee Fun","path":"/rebeccasf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9360972?s=80&v=4"},"commit":{"message":"lib: sbi: Fix Priv spec version for [m|s]counteren and mcountinhibit CSRs\n\n[ Upstream commit 976895c57e3b0655466e20c6eb0d7c670b9e8452 ]\n\nFix Priv spec version typo in commit d4b563c881d6 (\"lib: sbi: Remove MCOUNTEREN\nand SCOUNTEREN hart features\").\n\nAt least Priv spec v1.11 is required for [m|s]counteren and mcountinhibit CSRs.\n\nFixes: d4b563c881d6 (\"lib: sbi: Remove MCOUNTEREN and SCOUNTEREN hart features\")\nSigned-off-by: Ley Foon Tan \nReviewed-by: Anup Patel ","shortMessageHtmlLink":"lib: sbi: Fix Priv spec version for [m|s]counteren and mcountinhibit …"}},{"before":"ced60104df4f835e1d6bd0697b38394de64e301a","after":"c6a092cd80112529cb2e92e180767ff5341b22a3","ref":"refs/heads/JH7110_VisionFive2_devel","pushedAt":"2023-03-23T14:28:53.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"andyhu-stf","name":"Andy Hu","path":"/andyhu-stf","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/89558762?s=80&v=4"},"commit":{"message":"lib: sbi: Clear IPIs before init_warm_startup in non-boot harts\n\nSince commit 50d4fde1c5a4 (\"lib: Remove redundant sbi_platform_ipi_clear()\ncalls\"), the IPI sent from the boot hart in wake_coldboot_harts() is not\ncleared in the secondary harts until they reach sbi_ipi_init(). However,\nsbi_hsm_init() and sbi_hsm_hart_wait() are called earlier, so a secondary\nhart might enter sbi_hsm_hart_wait() with an already pending IPI.\n\nsbi_hsm_hart_wait() makes sure the hart leaves the loop only when it is\nactually ready, so a pending unrelated IPI should not cause safety issues.\nHowever, it might be inefficient on certain hardware, because it prevents\n\"wfi\" from stalling the hart even if the hardware supports this, making the\nhart needlessly spin in a \"busy-wait\" loop.\n\nThis behaviour can be observed, for example, in a QEMU VM (QEMU 7.2.0) with\n\"-machine virt\" running a Linux guest. Inserting delays in\nsbi_hsm_hart_start() allows reproducing the issue more reliably.\n\nThe comment in wait_for_coldboot() suggests that the initial IPI is needed\nin the warm resume path, so let us clear it before init_warm_startup()\nonly.\n\nTo do this, sbi_ipi_raw_clear() was created similar to sbi_ipi_raw_send().\n\nSigned-off-by: Evgenii Shatokhin \nReviewed-by: Anup Patel ","shortMessageHtmlLink":"lib: sbi: Clear IPIs before init_warm_startup in non-boot harts"}},{"before":"6f1fe98c2f565a394f258178b0daa3843672395c","after":"c6a092cd80112529cb2e92e180767ff5341b22a3","ref":"refs/heads/master","pushedAt":"2023-03-14T02:55:40.000Z","pushType":"push","commitsCount":225,"pusher":{"login":"MichaelZhuxx","name":null,"path":"/MichaelZhuxx","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78013124?s=80&v=4"},"commit":{"message":"lib: sbi: Clear IPIs before init_warm_startup in non-boot harts\n\nSince commit 50d4fde1c5a4 (\"lib: Remove redundant sbi_platform_ipi_clear()\ncalls\"), the IPI sent from the boot hart in wake_coldboot_harts() is not\ncleared in the secondary harts until they reach sbi_ipi_init(). However,\nsbi_hsm_init() and sbi_hsm_hart_wait() are called earlier, so a secondary\nhart might enter sbi_hsm_hart_wait() with an already pending IPI.\n\nsbi_hsm_hart_wait() makes sure the hart leaves the loop only when it is\nactually ready, so a pending unrelated IPI should not cause safety issues.\nHowever, it might be inefficient on certain hardware, because it prevents\n\"wfi\" from stalling the hart even if the hardware supports this, making the\nhart needlessly spin in a \"busy-wait\" loop.\n\nThis behaviour can be observed, for example, in a QEMU VM (QEMU 7.2.0) with\n\"-machine virt\" running a Linux guest. Inserting delays in\nsbi_hsm_hart_start() allows reproducing the issue more reliably.\n\nThe comment in wait_for_coldboot() suggests that the initial IPI is needed\nin the warm resume path, so let us clear it before init_warm_startup()\nonly.\n\nTo do this, sbi_ipi_raw_clear() was created similar to sbi_ipi_raw_send().\n\nSigned-off-by: Evgenii Shatokhin \nReviewed-by: Anup Patel ","shortMessageHtmlLink":"lib: sbi: Clear IPIs before init_warm_startup in non-boot harts"}},{"before":"c6a092cd80112529cb2e92e180767ff5341b22a3","after":null,"ref":"refs/heads/githubmaster","pushedAt":"2023-03-14T02:46:27.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"MichaelZhuxx","name":null,"path":"/MichaelZhuxx","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78013124?s=80&v=4"}},{"before":null,"after":"c6a092cd80112529cb2e92e180767ff5341b22a3","ref":"refs/heads/githubmaster","pushedAt":"2023-03-14T02:44:42.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"MichaelZhuxx","name":null,"path":"/MichaelZhuxx","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/78013124?s=80&v=4"},"commit":{"message":"lib: sbi: Clear IPIs before init_warm_startup in non-boot harts\n\nSince commit 50d4fde1c5a4 (\"lib: Remove redundant sbi_platform_ipi_clear()\ncalls\"), the IPI sent from the boot hart in wake_coldboot_harts() is not\ncleared in the secondary harts until they reach sbi_ipi_init(). However,\nsbi_hsm_init() and sbi_hsm_hart_wait() are called earlier, so a secondary\nhart might enter sbi_hsm_hart_wait() with an already pending IPI.\n\nsbi_hsm_hart_wait() makes sure the hart leaves the loop only when it is\nactually ready, so a pending unrelated IPI should not cause safety issues.\nHowever, it might be inefficient on certain hardware, because it prevents\n\"wfi\" from stalling the hart even if the hardware supports this, making the\nhart needlessly spin in a \"busy-wait\" loop.\n\nThis behaviour can be observed, for example, in a QEMU VM (QEMU 7.2.0) with\n\"-machine virt\" running a Linux guest. Inserting delays in\nsbi_hsm_hart_start() allows reproducing the issue more reliably.\n\nThe comment in wait_for_coldboot() suggests that the initial IPI is needed\nin the warm resume path, so let us clear it before init_warm_startup()\nonly.\n\nTo do this, sbi_ipi_raw_clear() was created similar to sbi_ipi_raw_send().\n\nSigned-off-by: Evgenii Shatokhin \nReviewed-by: Anup Patel ","shortMessageHtmlLink":"lib: sbi: Clear IPIs before init_warm_startup in non-boot harts"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAESzptJwA","startCursor":null,"endCursor":null}},"title":"Activity · starfive-tech/opensbi"}