mm, oom_reaper: skip mm structs with mmu notifiers
Subsystems (3)
PATCH 1/1 mm, oom_reaper: skip mm structs with mmu notifiers
From: Michal Hocko <[email protected]>
Andrea has noticed that the oom_reaper doesn't invalidate the range via mmu notifiers (mmu_notifier_invalidate_range_start, mmu_notifier_invalidate_range_end) and that can corrupt the memory of the kvm guest for example.
tlb_flush_mmu_tlbonly already invokes mmu notifiers but that is not sufficient as per Andrea:
: mmu_notifier_invalidate_range cannot be used in replacement of : mmu_notifier_invalidate_range_start/end. For KVM
: mmu_notifier_invalidate_range is a noop and rightfully so. A MMU : notifier implementation has to implement either
: ->invalidate_range method or the invalidate_range_start/end : methods, not both. And if you implement invalidate_range_start/end : like KVM is forced to do, calling mmu_notifier_invalidate_range in : common code is a noop for KVM.
:
: For those MMU notifiers that can get away only implementing : ->invalidate_range, the ->invalidate_range is implicitly called by : mmu_notifier_invalidate_range_end(). And only those secondary MMUs : that share the same pagetable with the primary MMU (like AMD : iommuv2) can get away only implementing ->invalidate_range.
As the callback is allowed to sleep and the implementation is out of hand of the MM it is safer to simply bail out if there is an mmu notifier registered. In order to not fail too early make the mm_has_notifiers check under the oom_lock and have a little nap before failing to give the current oom victim some more time to exit.
Changes since v1
- move mm_has_notifiers check after we hold mmap_sem to prevent from any potential races as per Andrea
---
Hi,
I have posted this as an RFC previously [1]. I have updated the changelog to be more clear about the issue and moved the mm_has_notifiers after the lock has been take based on Andrea's suggestion.
Can we merge this?
[1] http://lkml.kernel.org/r/[email protected]
include/linux/mmu_notifier.h | 5 +++++ mm/oom_kill.c | 16 ++++++++++++++++ 2 files changed, 21 insertions(+)
include/linux/mmu_notifier.h +5 -0
@@ -400,6 +400,11 @@ extern void mmu_notifier_synchronize(void);#else /* CONFIG_MMU_NOTIFIER */ +static inline int mm_has_notifiers (struct mm_struct *mm) +{ +return 0 ; +} +static inline void mmu_notifier_release (struct mm_struct *mm) { }
mm/oom_kill.c +16 -0
@@ -40,6 +40,7 @@#include <linux/ratelimit.h> #include <linux/kthread.h> #include <linux/init.h> +#include <linux/mmu_notifier.h> #include <asm/tlb.h> #include "internal.h"
@@ -494,6 +495,21 @@ static bool __oom_reap_task_mm(struct task_struct *tsk, struct mm_struct *mm) goto unlock_oom; } +/* +* If the mm has notifiers then we would need to invalidate them around +* unmap_page_range and that is risky because notifiers can sleep and +* what they do is basically undeterministic. So let's have a short sleep +* to give the oom victim some more time. +* TODO: we really want to get rid of this ugly hack and make sure that +* notifiers cannot block for unbounded amount of time and add +* mmu_notifier_invalidate_range_{start,end} around unmap_page_range +*/ +if (mm_has_notifiers (mm)) { +up_read (&mm->mmap_sem); +schedule_timeout_idle (HZ ); + goto unlock_oom; + } +/* * MMF_OOM_SKIP is set by exit_mmap when the OOM reaper can't * work on the mm anymore. The check for MMF_OOM_SKIP must run
Signature · 7 lines
2.14.1 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 45 lines · Michal Hocko
From: Michal Hocko <[email protected]>
Andrea has noticed that the
oom_reaperdoesn't invalidate the range via mmu notifiers (mmu_notifier_invalidate_range_start,mmu_notifier_invalidate_range_end) and that can corrupt the memory of the kvm guest for example.
tlb_flush_mmu_tlbonlyalready invokes mmu notifiers but that is not sufficient as per Andrea:
:mmu_notifier_invalidate_rangecannot be used in replacement of : mmu_notifier_invalidate_range_start/end. For KVM
:mmu_notifier_invalidate_rangeis a noop and rightfully so. A MMU : notifier implementation has to implement either
: ->invalidate_range method or the invalidate_range_start/end : methods, not both. And if you implement invalidate_range_start/end : like KVM is forced to do, callingmmu_notifier_invalidate_rangein : common code is a noop for KVM.
:
: For those MMU notifiers that can get away only implementing : ->invalidate_range, the ->invalidate_range is implicitly called by :mmu_notifier_invalidate_range_end(). And only those secondary MMUs : that share the same pagetable with the primary MMU (like AMD : iommuv2) can get away only implementing ->invalidate_range.As the callback is allowed to sleep and the implementation is out of hand of the MM it is safer to simply bail out if there is an mmu notifier registered. In order to not fail too early make the
mm_has_notifierscheck under theoom_lockand have a little nap before failing to give the current oom victim some more time to exit.Changes since v1
- movemm_has_notifierscheck after we holdmmap_semto prevent from any potential races as per AndreaFixes aac453635549 ("mm, oom: introduce oom reaper")Noticed-by Andrea Arcangeli <[email protected]>Cc stableSigned-off-by Michal Hocko <[email protected]>---
Hi,
I have posted this as an RFC previously [1]. I have updated the changelog to be more clear about the issue and moved themm_has_notifiersafter the lock has been take based on Andrea's suggestion.
Quoted 7 lines
Can we merge this?
[1] http://lkml.kernel.org/r/[email protected]
include/linux/mmu_notifier.h | 5 +++++ mm/oom_kill.c | 16 ++++++++++++++++ 2 files changed, 21 insertions(+)