From mboxrd@z Thu Jan 1 00:00:00 1970 From: akpm@linux-foundation.org Subject: + mmap-avoid-merging-cloned-vmas.patch added to -mm tree Date: Tue, 29 Mar 2011 15:37:49 -0700 Message-ID: <201103292237.p2TMbolo009763@imap1.linux-foundation.org> Reply-To: linux-kernel@vger.kernel.org Return-path: Received: from smtp1.linux-foundation.org ([140.211.169.13]:41350 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751197Ab1C2Wh7 (ORCPT ); Tue, 29 Mar 2011 18:37:59 -0400 Sender: mm-commits-owner@vger.kernel.org List-Id: mm-commits@vger.kernel.org To: mm-commits@vger.kernel.org Cc: shaohua.li@intel.com, andi@firstfloor.org, hughd@google.com, riel@redhat.com The patch titled mmap: avoid merging cloned VMAs has been added to the -mm tree. Its filename is mmap-avoid-merging-cloned-vmas.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: mmap: avoid merging cloned VMAs From: Shaohua Li Avoid merging a VMA with another VMA which is cloned from the parent process. The cloned VMA shares the anon_vma lock with the parent process's VMA. If we do the merge, more vmas (even the new range is only for current process) use the perent process's anon_vma lock. This introduces scalability issues. find_mergeable_anon_vma() already considers this case. Signed-off-by: Shaohua Li Cc: Rik van Riel Cc: Hugh Dickins Cc: Andi Kleen Signed-off-by: Andrew Morton --- mm/mmap.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff -puN mm/mmap.c~mmap-avoid-merging-cloned-vmas mm/mmap.c --- a/mm/mmap.c~mmap-avoid-merging-cloned-vmas +++ a/mm/mmap.c @@ -699,9 +699,17 @@ static inline int is_mergeable_vma(struc } static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1, - struct anon_vma *anon_vma2) + struct anon_vma *anon_vma2, + struct vm_area_struct *vma) { - return !anon_vma1 || !anon_vma2 || (anon_vma1 == anon_vma2); + /* + * The list_is_singular() test is to avoid merging VMA cloned from + * parents. This can improve scalability caused by anon_vma lock. + */ + if ((!anon_vma1 || !anon_vma2) && (!vma || + list_is_singular(&vma->anon_vma_chain))) + return 1; + return anon_vma1 == anon_vma2; } /* @@ -720,7 +728,7 @@ can_vma_merge_before(struct vm_area_stru struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff) { if (is_mergeable_vma(vma, file, vm_flags) && - is_mergeable_anon_vma(anon_vma, vma->anon_vma)) { + is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { if (vma->vm_pgoff == vm_pgoff) return 1; } @@ -739,7 +747,7 @@ can_vma_merge_after(struct vm_area_struc struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff) { if (is_mergeable_vma(vma, file, vm_flags) && - is_mergeable_anon_vma(anon_vma, vma->anon_vma)) { + is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { pgoff_t vm_pglen; vm_pglen = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; if (vma->vm_pgoff + vm_pglen == vm_pgoff) @@ -817,7 +825,7 @@ struct vm_area_struct *vma_merge(struct can_vma_merge_before(next, vm_flags, anon_vma, file, pgoff+pglen) && is_mergeable_anon_vma(prev->anon_vma, - next->anon_vma)) { + next->anon_vma, NULL)) { /* cases 1, 6 */ err = vma_adjust(prev, prev->vm_start, next->vm_end, prev->vm_pgoff, NULL); _ Patches currently in -mm which might be from shaohua.li@intel.com are origin.patch mmap-avoid-unnecessary-anon_vma-lock.patch mmap-avoid-merging-cloned-vmas.patch mm-batch-activate_page-to-reduce-lock-contention.patch