From: "Joel Fernandes (Google)" <j...@joelfernandes.org> Android needs to mremap large regions of memory during memory management related operations. The mremap system call can be really slow if THP is not enabled. The bottleneck is move_page_tables, which is copying each pte at a time, and can be really slow across a large map. Turning on THP may not be a viable option, and is not for us. This patch speeds up the performance for non-THP system by copying at the PMD level when possible.
The speedup is an order of magnitude on x86 (~20x). On a 1GB mremap, the mremap completion times drops from 3.4-3.6 milliseconds to 144-160 microseconds. Before: Total mremap time for 1GB data: 3521942 nanoseconds. Total mremap time for 1GB data: 3449229 nanoseconds. Total mremap time for 1GB data: 3488230 nanoseconds. After: Total mremap time for 1GB data: 150279 nanoseconds. Total mremap time for 1GB data: 144665 nanoseconds. Total mremap time for 1GB data: 158708 nanoseconds. If THP is enabled the optimization is mostly skipped except in certain situations. [j...@joelfernandes.org: fix 'move_normal_pmd' unused function warning] Link: http://lkml.kernel.org/r/20181108224457.gb209...@google.com Link: http://lkml.kernel.org/r/20181108181201.88826-3-joe...@google.com Signed-off-by: Joel Fernandes (Google) <j...@joelfernandes.org> Acked-by: Kirill A. Shutemov <kir...@shutemov.name> Reviewed-by: William Kucharski <william.kuchar...@oracle.com> Cc: Julia Lawall <julia.law...@lip6.fr> Cc: Michal Hocko <mho...@kernel.org> Cc: Will Deacon <will.dea...@arm.com> Signed-off-by: Andrew Morton <a...@linux-foundation.org> Signed-off-by: Linus Torvalds <torva...@linux-foundation.org> (cherry picked from commit 2c91bd4a4e2e530582d6fd643ea7b86b27907151) Signed-off-by: Konstantin Khorenko <khore...@virtuozzo.com> Signed-off-by: Kirill Tkhai <ktk...@virtuozzo.com> Patchset description: Patch "mm: speed up mremap by 20x on large regions" introduces optimization: when a moved region has source and destination addresses and size equal to multiple of PMD_SIZE; PTEs are not really copyed. Instead of this, new PMD pointer is changed to point to old PTEs, while old PMD is cleared. This may be useful, when CRIU remaps large memory areas on restore (but really, alignment to PMD_SIZE is not very often, though possible). (cherry-picked from ms commit 2c91bd4a4e2e530582d6fd643ea7b86b27907151) Signed-off-by: Andrey Zhadchenko <andrey.zhadche...@virtuozzo.com> diff --git a/arch/Kconfig b/arch/Kconfig index c2142af..4d7f52a 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -669,6 +669,11 @@ config HAVE_IRQ_TIME_ACCOUNTING Archs need to ensure they use a high enough resolution clock to support irq time accounting and then call enable_sched_clock_irqtime(). +config HAVE_MOVE_PMD + bool + help + Archs that select this are able to move page tables at the PMD level. + config HAVE_ARCH_TRANSPARENT_HUGEPAGE bool diff --git a/mm/mremap.c b/mm/mremap.c index 8cc4fc1..e482d8a 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -191,6 +191,52 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd, drop_rmap_locks(vma); } +#ifdef CONFIG_HAVE_MOVE_PMD +static bool move_normal_pmd(struct vm_area_struct *vma, unsigned long old_addr, + unsigned long new_addr, unsigned long old_end, + pmd_t *old_pmd, pmd_t *new_pmd) +{ + spinlock_t *old_ptl, *new_ptl; + struct mm_struct *mm = vma->vm_mm; + pmd_t pmd; + + if ((old_addr & ~PMD_MASK) || (new_addr & ~PMD_MASK) + || old_end - old_addr < PMD_SIZE) + return false; + + /* + * The destination pmd shouldn't be established, free_pgtables() + * should have release it. + */ + if (WARN_ON(!pmd_none(*new_pmd))) + return false; + + /* + * We don't have to worry about the ordering of src and dst + * ptlocks because exclusive mmap_sem prevents deadlock. + */ + old_ptl = pmd_lock(vma->vm_mm, old_pmd); + new_ptl = pmd_lockptr(mm, new_pmd); + if (new_ptl != old_ptl) + spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING); + + /* Clear the pmd */ + pmd = *old_pmd; + pmd_clear(old_pmd); + + VM_BUG_ON(!pmd_none(*new_pmd)); + + /* Set the new pmd */ + set_pmd_at(mm, new_addr, new_pmd, pmd); + flush_tlb_range(vma, old_addr, old_addr + PMD_SIZE); + if (new_ptl != old_ptl) + spin_unlock(new_ptl); + spin_unlock(old_ptl); + + return true; +} +#endif + unsigned long move_page_tables(struct vm_area_struct *vma, unsigned long old_addr, struct vm_area_struct *new_vma, unsigned long new_addr, unsigned long len, @@ -237,6 +283,23 @@ unsigned long move_page_tables(struct vm_area_struct *vma, split_huge_pmd(vma, old_pmd, old_addr); if (pmd_trans_unstable(old_pmd)) continue; + } else if (extent == PMD_SIZE) { +#ifdef CONFIG_HAVE_MOVE_PMD + /* + * If the extent is PMD-sized, try to speed the move by + * moving at the PMD level if possible. + */ + bool moved; + + if (need_rmap_locks) + take_rmap_locks(vma); + moved = move_normal_pmd(vma, old_addr, new_addr, + old_end, old_pmd, new_pmd); + if (need_rmap_locks) + drop_rmap_locks(vma); + if (moved) + continue; +#endif } if (pte_alloc(new_vma->vm_mm, new_pmd, new_addr)) break; -- 1.8.3.1 _______________________________________________ Devel mailing list Devel@openvz.org https://lists.openvz.org/mailman/listinfo/devel
