{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":505398471,"defaultBranch":"LA.UM.8.11","name":"biofrost-kernel-realme-trinket","ownerLogin":"mcdofrenchfreis","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2022-06-20T10:41:17.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/45391090?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1711702954.0","currentOid":""},"activityList":{"items":[{"before":"8227a820241afe10a638d4c75e34aba6edd8bbd3","after":"8a6d1c22939783ce1cf69a73a17550a8b8a6b030","ref":"refs/heads/labs","pushedAt":"2024-04-30T02:05:38.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"cpufreq: schedutil: Set default up/down rate limits to 500/1000 us\n\nThis is empirically observed to yield good performance with reduced power\nconsumption via having the down rate limit configured to be 2x longer than\nthe up rate limit. This reduces bouncing between CPU frequencies by\nstalling down-clocking, which not only improves performance, but also\ncounter-intuitively improves power consumption.\n\nThe short up/down rate limits also provide improved interactivity and\nreal-time response.\n\nSigned-off-by: Sultan Alsawaf ","shortMessageHtmlLink":"cpufreq: schedutil: Set default up/down rate limits to 500/1000 us"}},{"before":"a0d112dbe69c4bca914e3dbbf8c1dfb78458ad8a","after":"8227a820241afe10a638d4c75e34aba6edd8bbd3","ref":"refs/heads/labs","pushedAt":"2024-04-30T01:47:46.000Z","pushType":"push","commitsCount":7,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"cpufreq: schedutil: Set default up/down rate limits to 500/1000 us\n\nThis is empirically observed to yield good performance with reduced power\nconsumption via having the down rate limit configured to be 2x longer than\nthe up rate limit. This reduces bouncing between CPU frequencies by\nstalling down-clocking, which not only improves performance, but also\ncounter-intuitively improves power consumption.\n\nThe short up/down rate limits also provide improved interactivity and\nreal-time response.\n\nSigned-off-by: Sultan Alsawaf ","shortMessageHtmlLink":"cpufreq: schedutil: Set default up/down rate limits to 500/1000 us"}},{"before":"455aca61a95247601d994c2e44caab0beda69108","after":"a0d112dbe69c4bca914e3dbbf8c1dfb78458ad8a","ref":"refs/heads/labs","pushedAt":"2024-04-30T01:40:28.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"biofrost_defconfig: Reduce SLMK Timeout\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"biofrost_defconfig: Reduce SLMK Timeout"}},{"before":"efcf742c98db786387382393e2936f567b410a5f","after":"455aca61a95247601d994c2e44caab0beda69108","ref":"refs/heads/labs","pushedAt":"2024-04-30T01:33:48.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: make it more swappy\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: make it more swappy"}},{"before":"c91347241d23ba13e31df77791bd40f1acc72d94","after":"efcf742c98db786387382393e2936f567b410a5f","ref":"refs/heads/labs","pushedAt":"2024-04-29T18:19:25.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: make it more swappy\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: make it more swappy"}},{"before":"66f134f861ee402e3c2b0993bca8a7f14f665ee9","after":"c91347241d23ba13e31df77791bd40f1acc72d94","ref":"refs/heads/labs","pushedAt":"2024-04-29T18:16:38.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: make it more swappy\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: make it more swappy"}},{"before":"d1fb619dce7367d80a5c50d0909527c0e425eeb0","after":"66f134f861ee402e3c2b0993bca8a7f14f665ee9","ref":"refs/heads/labs","pushedAt":"2024-04-29T18:11:43.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: make it more swappy\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: make it more swappy"}},{"before":"d4455f3b9167171bb72fabafe4f1b44579bce55a","after":"d1fb619dce7367d80a5c50d0909527c0e425eeb0","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:55:33.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: make it more swappy\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: make it more swappy"}},{"before":"4351374f5b7b4d0e58c9d50b7cdd4530e8cb841f","after":"d4455f3b9167171bb72fabafe4f1b44579bce55a","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:48:39.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: make it more swappy\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: make it more swappy"}},{"before":"0ae05f6fcd8c4c290d4f1b33b3088730fb3228c8","after":"4351374f5b7b4d0e58c9d50b7cdd4530e8cb841f","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:44:09.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"walt: update kp-controlled tunables\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"walt: update kp-controlled tunables"}},{"before":"0ce17ebea2d6293b93a9addbec81873e740bc9b7","after":"0ae05f6fcd8c4c290d4f1b33b3088730fb3228c8","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:43:19.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"walt: update kp-controlled tunables\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"walt: update kp-controlled tunables"}},{"before":"6ad3ea6aada4b5747a9b7382f575e967d6066b82","after":"0ce17ebea2d6293b93a9addbec81873e740bc9b7","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:36:30.000Z","pushType":"push","commitsCount":9,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"sched/cass: fix build","shortMessageHtmlLink":"sched/cass: fix build"}},{"before":"5b8ed6e3176d892d6e1e00875f8d425c63acd441","after":"6ad3ea6aada4b5747a9b7382f575e967d6066b82","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:30:51.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"sched/cass: Backport few changes intended for >=4.20 kernels\n\nOur current CASS is an adapted version for <4.20 kernels. Currently, this misses\nout available_idle_cpu() and sched_idle_cpu() from the original commit [1] which\nare intended for >=4.20 kernels. Backport the original changes since we are not\nupstream and we don't care about diverging from it.\n\nMoreover, without sched_idle_cpu(), jitter takes a big hit.\n\nFor this, from commit 6d53328e37351 (\"UPSTREAM: sched/fair:\nStart tracking SCHED_IDLE tasks count in cfs_rq\") to commit\n7c616803d0472 (\"UPSTREAM: sched/fair: Load balance aggressively\nfor SCHED_IDLE CPUs\") are needed as dependencies.\n\n[1]: https://github.com/kerneltoast/android_kernel_google_gs201/commit/d968506906803\n\nSigned-off-by: Tashfin Shakeer Rhythm ","shortMessageHtmlLink":"sched/cass: Backport few changes intended for >=4.20 kernels"}},{"before":"8ac4145db9ceedc713a4eb516effcc3bfff9cb69","after":"5b8ed6e3176d892d6e1e00875f8d425c63acd441","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:27:59.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"sched/cass: fast-forward","shortMessageHtmlLink":"sched/cass: fast-forward"}},{"before":"fae1970eeef9c5bbff85877fef6d5a885b66856b","after":"8ac4145db9ceedc713a4eb516effcc3bfff9cb69","ref":"refs/heads/labs","pushedAt":"2024-04-29T17:15:27.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"UPSTREAM: sched/fair: Fallback to sched-idle CPU if idle CPU isn't found\n\nWe try to find an idle CPU to run the next task, but in case we don't\nfind an idle CPU it is better to pick a CPU which will run the task the\nsoonest, for performance reason.\n\nA CPU which isn't idle but has only SCHED_IDLE activity queued on it\nshould be a good target based on this criteria as any normal fair task\nwill most likely preempt the currently running SCHED_IDLE task\nimmediately. In fact, choosing a SCHED_IDLE CPU over a fully idle one\nshall give better results as it should be able to run the task sooner\nthan an idle CPU (which requires to be woken up from an idle state).\n\nThis patch updates both fast and slow paths with this optimization.\n\nSigned-off-by: Viresh Kumar \nSigned-off-by: Peter Zijlstra (Intel) \nCc: Daniel Lezcano \nCc: Linus Torvalds \nCc: Peter Zijlstra \nCc: Thomas Gleixner \nCc: Vincent Guittot \nCc: chris.redpath@arm.com\nCc: quentin.perret@linaro.org\nCc: songliubraving@fb.com\nCc: steven.sistare@oracle.com\nCc: subhra.mazumdar@oracle.com\nCc: tkjos@google.com\nLink: https://lkml.kernel.org/r/eeafa25fdeb6f6edd5b2da716bc8f0ba7708cbcf.1561523542.git.viresh.kumar@linaro.org\nSigned-off-by: Ingo Molnar \nSigned-off-by: Tashfin Shakeer Rhythm ","shortMessageHtmlLink":"UPSTREAM: sched/fair: Fallback to sched-idle CPU if idle CPU isn't found"}},{"before":"345b0612ed5fea34f1f4d2369d1a3a4d2e63cea4","after":"4597410aa74dc8c522c0b9df6f9ab5eee041a19e","ref":"refs/heads/inline/unified","pushedAt":"2024-03-29T09:02:46.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"syscall: Increase bpf fake uname to 5.4\n\nChange-Id: I50bfa0d35d81f1c8cc21530ea0524a6752d0d34c\nSigned-off-by: shygosh ","shortMessageHtmlLink":"syscall: Increase bpf fake uname to 5.4"}},{"before":null,"after":"4597410aa74dc8c522c0b9df6f9ab5eee041a19e","ref":"refs/heads/bpf","pushedAt":"2024-03-29T09:02:34.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"syscall: Increase bpf fake uname to 5.4\n\nChange-Id: I50bfa0d35d81f1c8cc21530ea0524a6752d0d34c\nSigned-off-by: shygosh ","shortMessageHtmlLink":"syscall: Increase bpf fake uname to 5.4"}},{"before":"cab656db4572cda6c2d9433c84372d2fa6f9e086","after":"fae1970eeef9c5bbff85877fef6d5a885b66856b","ref":"refs/heads/labs","pushedAt":"2024-03-08T15:09:26.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"alarmtimer: Increase wakeup safety margin\n\nUsers reported delayed notifications.\n\nSigned-off-by: Andrzej Perczak \nSigned-off-by: Tashfin Shakeer Rhythm ","shortMessageHtmlLink":"alarmtimer: Increase wakeup safety margin"}},{"before":"8050dd565048047935bc95f0dac6968a336e0cb6","after":"cab656db4572cda6c2d9433c84372d2fa6f9e086","ref":"refs/heads/labs","pushedAt":"2024-03-08T15:04:59.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"kernel/sched: Mitigate non-boosted tasks preempting boosted tasks\n\n* by default uclamp_boosted doesn't do rcu locking like schedtune_task_boost, basically its like schedtune_task_boost_rcu_locked. With this we might get the same benefits from [1]\n\n[1]:https://github.com/kdrag0n/proton_zf6/commit/567ea3c45e9c42eaa86e56bda9137a8431f66880#\n\nSigned-off-by: dlwlrma123 \nSigned-off-by: Dakkshesh \nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"kernel/sched: Mitigate non-boosted tasks preempting boosted tasks"}},{"before":"8fd3d8e22b538b40e7227e8724072495ad6c6177","after":"d9c8f48759f4e74dfa12deeb9c535562febb1073","ref":"refs/heads/mglru-new","pushedAt":"2024-02-26T08:00:17.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"biofrost_defconfig: Lesser aggressive SLMK\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"biofrost_defconfig: Lesser aggressive SLMK"}},{"before":"e9c8bd6653f0faf8b1a7b72c832ca7b41c30b442","after":"8fd3d8e22b538b40e7227e8724072495ad6c6177","ref":"refs/heads/mglru-new","pushedAt":"2024-02-26T07:59:44.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"mm: compaction: avoid 100% CPU usage during compaction when a task is killed\n\n\"howaboutsynergy\" reported via kernel buzilla number 204165 that\ncompact_zone_order was consuming 100% CPU during a stress test for\nprolonged periods of time. Specifically the following command, which\nshould exit in 10 seconds, was taking an excessive time to finish while\nthe CPU was pegged at 100%.\n\n stress -m 220 --vm-bytes 1000000000 --timeout 10\n\nTracing indicated a pattern as follows\n\n stress-3923 [007] 519.106208: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106212: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106216: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106219: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106223: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106227: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106231: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106235: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106238: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106242: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n\nNote that compaction is entered in rapid succession while scanning and\nisolating nothing. The problem is that when a task that is compacting\nreceives a fatal signal, it retries indefinitely instead of exiting\nwhile making no progress as a fatal signal is pending.\n\nIt's not easy to trigger this condition although enabling zswap helps on\nthe basis that the timing is altered. A very small window has to be hit\nfor the problem to occur (signal delivered while compacting and\nisolating a PFN for migration that is not aligned to SWAP_CLUSTER_MAX).\n\nThis was reproduced locally -- 16G single socket system, 8G swap, 30%\nzswap configured, vm-bytes 22000000000 using Colin Kings stress-ng\nimplementation from github running in a loop until the problem hits).\nTracing recorded the problem occurring almost 200K times in a short\nwindow. With this patch, the problem hit 4 times but the task existed\nnormally instead of consuming CPU.\n\nThis problem has existed for some time but it was made worse by commit\ncf66f0700c8f (\"mm, compaction: do not consider a need to reschedule as\ncontention\"). Before that commit, if the same condition was hit then\nlocks would be quickly contended and compaction would exit that way.\n\nChange-Id: I67b546921390d17b393c1f3f2f195db9de499255\nBugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=204165\nLink: http://lkml.kernel.org/r/20190718085708.GE24383@techsingularity.net\nFixes: cf66f0700c8f (\"mm, compaction: do not consider a need to reschedule as contention\")\nSigned-off-by: Mel Gorman \nReviewed-by: Vlastimil Babka \nCc: \t[5.1+]\nSigned-off-by: Andrew Morton \nSigned-off-by: Linus Torvalds \nGit-Commit: 670105a25608affe01cb0ccdc2a1f4bd2327172b\nGit-Repo: git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git\n[vinmenon@codeaurora.org: trivial conflict fixes]\nSigned-off-by: Vinayak Menon \nSigned-off-by: Adam W. Willis \nSigned-off-by: atndko ","shortMessageHtmlLink":"mm: compaction: avoid 100% CPU usage during compaction when a task is…"}},{"before":"73bb195a8c186df8797cd622b17d7414a5dae5a5","after":"e9c8bd6653f0faf8b1a7b72c832ca7b41c30b442","ref":"refs/heads/mglru-new","pushedAt":"2024-02-09T06:25:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"mm: compaction: avoid 100% CPU usage during compaction when a task is killed\n\n\"howaboutsynergy\" reported via kernel buzilla number 204165 that\ncompact_zone_order was consuming 100% CPU during a stress test for\nprolonged periods of time. Specifically the following command, which\nshould exit in 10 seconds, was taking an excessive time to finish while\nthe CPU was pegged at 100%.\n\n stress -m 220 --vm-bytes 1000000000 --timeout 10\n\nTracing indicated a pattern as follows\n\n stress-3923 [007] 519.106208: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106212: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106216: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106219: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106223: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106227: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106231: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106235: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106238: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n stress-3923 [007] 519.106242: mm_compaction_isolate_migratepages: range=(0x70bb80 ~ 0x70bb80) nr_scanned=0 nr_taken=0\n\nNote that compaction is entered in rapid succession while scanning and\nisolating nothing. The problem is that when a task that is compacting\nreceives a fatal signal, it retries indefinitely instead of exiting\nwhile making no progress as a fatal signal is pending.\n\nIt's not easy to trigger this condition although enabling zswap helps on\nthe basis that the timing is altered. A very small window has to be hit\nfor the problem to occur (signal delivered while compacting and\nisolating a PFN for migration that is not aligned to SWAP_CLUSTER_MAX).\n\nThis was reproduced locally -- 16G single socket system, 8G swap, 30%\nzswap configured, vm-bytes 22000000000 using Colin Kings stress-ng\nimplementation from github running in a loop until the problem hits).\nTracing recorded the problem occurring almost 200K times in a short\nwindow. With this patch, the problem hit 4 times but the task existed\nnormally instead of consuming CPU.\n\nThis problem has existed for some time but it was made worse by commit\ncf66f0700c8f (\"mm, compaction: do not consider a need to reschedule as\ncontention\"). Before that commit, if the same condition was hit then\nlocks would be quickly contended and compaction would exit that way.\n\nChange-Id: I67b546921390d17b393c1f3f2f195db9de499255\nBugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=204165\nLink: http://lkml.kernel.org/r/20190718085708.GE24383@techsingularity.net\nFixes: cf66f0700c8f (\"mm, compaction: do not consider a need to reschedule as contention\")\nSigned-off-by: Mel Gorman \nReviewed-by: Vlastimil Babka \nCc: \t[5.1+]\nSigned-off-by: Andrew Morton \nSigned-off-by: Linus Torvalds \nGit-Commit: 670105a25608affe01cb0ccdc2a1f4bd2327172b\nGit-Repo: git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git\n[vinmenon@codeaurora.org: trivial conflict fixes]\nSigned-off-by: Vinayak Menon \nSigned-off-by: Adam W. Willis \nSigned-off-by: atndko ","shortMessageHtmlLink":"mm: compaction: avoid 100% CPU usage during compaction when a task is…"}},{"before":"20f784dd84c58c7459e5f345a4bf1620935221fc","after":"73bb195a8c186df8797cd622b17d7414a5dae5a5","ref":"refs/heads/mglru-new","pushedAt":"2024-02-09T05:54:36.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: drop direct reclaim swappiness to 20\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"vmscan: drop direct reclaim swappiness to 20"}},{"before":"bd9cd9277de391ba7f53c611809b7181fae368e1","after":"20f784dd84c58c7459e5f345a4bf1620935221fc","ref":"refs/heads/mglru-new","pushedAt":"2024-02-08T06:27:42.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"mm: multi-gen LRU: Turn minimum cache TTL to 1000 ms\n\nSigned-off-by: 0wnerDied ","shortMessageHtmlLink":"mm: multi-gen LRU: Turn minimum cache TTL to 1000 ms"}},{"before":"01d7c331598ca2bf67a24a32bc5b824ff448aa37","after":"7c1533c3839f6d47ef5c0c5fa6042af280ac8f73","ref":"refs/heads/labs-zram","pushedAt":"2024-02-05T13:15:53.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"simple_lmk: Wait 28ms before starting killing again\n\nThis is a temporary fix for extreme memory pressure such as running yuzu emulator, sultan will come up with a better fix.\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"simple_lmk: Wait 28ms before starting killing again"}},{"before":"b518065aec203c954f4893b9fbc483ed428fd655","after":"bd9cd9277de391ba7f53c611809b7181fae368e1","ref":"refs/heads/mglru-new","pushedAt":"2024-02-05T04:31:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"biofrost_defconfig: Readjust SLMK\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"biofrost_defconfig: Readjust SLMK"}},{"before":"73dc585d63f94ff791d12ee8e025013d02ae6776","after":"b518065aec203c954f4893b9fbc483ed428fd655","ref":"refs/heads/mglru-new","pushedAt":"2024-02-05T04:28:26.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"biofrost_defconfig: Readjust SLMK\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"biofrost_defconfig: Readjust SLMK"}},{"before":"df6de07b5398f818452b3c978a8c10aac32c735f","after":"12deaa992644b57f96fffc2008fe7fcea83fe45c","ref":"refs/heads/backup/zram-slmk","pushedAt":"2024-02-05T04:01:53.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"sched/walt: update per-kprofile tuning\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"sched/walt: update per-kprofile tuning"}},{"before":"f33a3dd2a35c076e3198f7de3ba52934943e87e4","after":"df6de07b5398f818452b3c978a8c10aac32c735f","ref":"refs/heads/backup/zram-slmk","pushedAt":"2024-02-05T03:38:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"biofrost_defconfig: Enable and Tune SLMK\n\nSigned-off-by: mcdofrenchfreis ","shortMessageHtmlLink":"biofrost_defconfig: Enable and Tune SLMK"}},{"before":"9c2db5c4bdc2ad871c46a277554f61d35c2f7f2e","after":"e4c78e47a524c40ca00361824301799ef2c75be8","ref":"refs/heads/mglru-zram","pushedAt":"2024-02-04T06:29:22.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"mcdofrenchfreis","name":"Evan","path":"/mcdofrenchfreis","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/45391090?s=80&v=4"},"commit":{"message":"vmscan: Reduce swapping aggressiveness to 10\n\nSince Android 12, memory management on Androids has been affected\nsignificantly, especially on devices with <=4GB RAMs.\n\nIn such situation, reducing swappiness parameter seems to have helped alot\nin multitasking and it seems fluid now.\n\nSigned-off-by: Panchajanya1999 ","shortMessageHtmlLink":"vmscan: Reduce swapping aggressiveness to 10"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEPYso0gA","startCursor":null,"endCursor":null}},"title":"Activity · mcdofrenchfreis/biofrost-kernel-realme-trinket"}