[mmotm 0521][PATCH 0/12] various OOM fixes for 2.6.35
in [Kernel]
|
Prev: [PATCH 3/6] sctp multistream scheduling: provide the default FCFS scheduling
Next: [PATCH 03/12] oom: the points calculation of child processes must use find_lock_task_mm() too
From: KOSAKI Motohiro on 3 Jun 2010 01:50 Hi This patch series is collection of various OOM bugfixes. I think all of patches can send to 2.6.35. Recently, David Rientjes and Luis Claudio R. Goncalves posted other various imporovement. I'll collect such 2.6.36 items and I plan to push -mm at next week. patch lists ------------------------------------- oom: select_bad_process: check PF_KTHREAD instead of !mm to skip kthreads oom: introduce find_lock_task_mm() to fix !mm false positives oom: the points calculation of child processes must use find_lock_task_mm() too oom: __oom_kill_task() must use find_lock_task_mm() too oom: make oom_unkillable() helper function oom: remove warning for in mm-less task __oom_kill_process() oom: Fix child process iteration properly oom: dump_tasks() use find_lock_task_mm() too oom: remove PF_EXITING check completely oom: sacrifice child with highest badness score for parent oom: remove special handling for pagefault ooms oom: give current access to memory reserves if it has been killed diffstat ------------ mm/oom_kill.c | 303 ++++++++++++++++++++++++++++++-------------------------- 1 files changed, 162 insertions(+), 141 deletions(-) Changes since last post ------------------------- - Drop Luis's "give the dying task a higher priority" patch - Add "remove PF_EXITING check completely" patch - Drop Oleg's "oom: select_bad_process: PF_EXITING check should take ->mm into account" because conflict against "remove PF_EXITING check completely" - Add "oom: sacrifice child with highest badness score for parent" - Add "oom: remove special handling for pagefault ooms" - Add "oom: give current access to memory reserves if it has been killed" -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |