From mboxrd@z Thu Jan 1 00:00:00 1970 From: akpm@linux-foundation.org Subject: + mm-mremap-calculate-extent-in-one-place.patch added to -mm tree Date: Sat, 18 Jan 2020 16:07:13 -0800 Message-ID: <20200119000713.SM_le%akpm@linux-foundation.org> Reply-To: linux-kernel@vger.kernel.org Return-path: Received: from mail.kernel.org ([198.145.29.99]:54946 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727012AbgASAHO (ORCPT ); Sat, 18 Jan 2020 19:07:14 -0500 Sender: mm-commits-owner@vger.kernel.org List-Id: mm-commits@vger.kernel.org To: mm-commits@vger.kernel.org, yang.shi@linux.alibaba.com, thellstrom@vmware.com, kirill@shutemov.name, dan.j.williams@intel.com, aneesh.kumar@linux.ibm.com, richardw.yang@linux.intel.com The patch titled Subject: mm/mremap: calculate extent in one place has been added to the -mm tree. Its filename is mm-mremap-calculate-extent-in-one-place.patch This patch should soon appear at http://ozlabs.org/~akpm/mmots/broken-out/mm-mremap-calculate-extent-in-one-place.patch and later at http://ozlabs.org/~akpm/mmotm/broken-out/mm-mremap-calculate-extent-in-one-place.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/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next and is updated there every 3-4 working days ------------------------------------------------------ From: Wei Yang Subject: mm/mremap: calculate extent in one place Page tables is moved on the base of PMD. This requires both source and destination range should meet the requirement. Current code works well since move_huge_pmd() and move_normal_pmd() would check old_addr and new_addr again. And then return to move_ptes() if the either of them is not aligned. In stead of calculating the extent separately, it is better to calculate in one place, so we know it is not necessary to try move pmd. By doing so, the logic seems a little clear. Link: http://lkml.kernel.org/r/20200117232254.2792-5-richardw.yang@linux.intel.com Signed-off-by: Wei Yang Cc: Aneesh Kumar K.V Cc: Dan Williams Cc: Kirill A. Shutemov Cc: Thomas Hellstrom Cc: Yang Shi Signed-off-by: Andrew Morton --- mm/mremap.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) --- a/mm/mremap.c~mm-mremap-calculate-extent-in-one-place +++ a/mm/mremap.c @@ -240,7 +240,7 @@ unsigned long move_page_tables(struct vm unsigned long new_addr, unsigned long len, bool need_rmap_locks) { - unsigned long extent, next, old_end; + unsigned long extent, old_next, new_next, old_end; struct mmu_notifier_range range; pmd_t *old_pmd, *new_pmd; @@ -253,8 +253,9 @@ unsigned long move_page_tables(struct vm for (; old_addr < old_end; old_addr += extent, new_addr += extent) { cond_resched(); - next = pmd_addr_end(old_addr, old_end); - extent = next - old_addr; + old_next = pmd_addr_end(old_addr, old_end); + new_next = pmd_addr_end(new_addr, new_addr + len); + extent = min((old_next - old_addr), (new_next - new_addr)); old_pmd = get_old_pmd(vma->vm_mm, old_addr); if (!old_pmd) continue; @@ -298,9 +299,6 @@ unsigned long move_page_tables(struct vm if (pte_alloc(new_vma->vm_mm, new_pmd)) break; - next = pmd_addr_end(new_addr, new_addr + len); - if (extent > next - new_addr) - extent = next - new_addr; move_ptes(vma, old_pmd, old_addr, old_addr + extent, new_vma, new_pmd, new_addr, need_rmap_locks); } _ Patches currently in -mm which might be from richardw.yang@linux.intel.com are mm-thp-remove-the-defer-list-related-code-since-this-will-not-happen.patch mm-gupc-use-is_vm_hugetlb_page-to-check-whether-to-follow-huge.patch mm-mremap-format-the-check-in-move_normal_pmd-same-as-move_huge_pmd.patch mm-mremap-it-is-sure-to-have-enough-space-when-extent-meets-requirement.patch mm-mremap-use-pmd_addr_end-to-calculate-next-in-move_page_tables.patch mm-mremap-calculate-extent-in-one-place.patch mm-mremap-start-addresses-are-properly-aligned.patch mm-huge_memoryc-use-head-to-check-huge-zero-page.patch mm-huge_memoryc-use-head-to-emphasize-the-purpose-of-page.patch mm-huge_memoryc-reduce-critical-section-protected-by-split_queue_lock.patch mm-remove-dead-code-totalram_pages_set.patch