mmu_notifier: pass through vma to invalidate_range and invalidate_page v2
New user of the mmu_notifier interface need to lookup vma in order to
perform the invalidation operation. Instead of redoing a vma lookup
inside the callback just pass through the vma from the call site where
it is already available.
This needs small refactoring in memory.c to call invalidate_range on
vma boundary while previously it was call once for larger range. The
overhead might be offseted by the fact that mmu_notifier listener now
work on smaller and exact range.
Changed since v1 :
- Only passthrough the vma.
- Commit comment.
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
(cherry picked from commit
eeb9f7693d2a86a7caa368e2c51d3ac09da04a8e)
Conflicts:
drivers/gpu/drm/i915/i915_gem_userptr.c
kernel/events/uprobes.c
mm/huge_memory.c
mm/hugetlb.c
mm/migrate.c
mm/rmap.c
mm/mprotect.c
Bug
200074285
Change-Id: I076973244386b359b1db6c6c2fe0c68d289b19ff
Signed-off-by: Sri Krishna chowdary <schowdary@nvidia.com>
Reviewed-on: http://git-master/r/678276
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>