Commit acbfb087 authored by Zhigang Lu's avatar Zhigang Lu Committed by Linus Torvalds
Browse files

mm/hugetlb: avoid looping to the same hugepage if !pages and !vmas 

When mmapping an existing hugetlbfs file with MAP_POPULATE, we find it
is very time consuming.  For example, mmapping a 128GB file takes about
50 milliseconds.  Sampling with perfevent shows it spends 99% time in
the same_page loop in follow_hugetlb_page().

samples: 205  of event 'cycles', Event count (approx.): 136686374
-  99.04%  test_mmap_huget  [kernel.kallsyms]  [k] follow_hugetlb_page
        follow_hugetlb_page
        __get_user_pages
        __mlock_vma_pages_range
        __mm_populate
        vm_mmap_pgoff
        sys_mmap_pgoff
        sys_mmap
        system_call_fastpath
        __mmap64

follow_hugetlb_page() is called with pages=NULL and vmas=NULL, so for
each hugepage, we run into the same_page loop for pages_per_huge_page()
times, but doing nothing.  With this change, it takes less then 1
millisecond to mmap a 128GB file in hugetlbfs.

Link: http://lkml.kernel.org/r/1567581712-5992-1-git-send-email-totty.lu@gmail.com


Signed-off-by: default avatarZhigang Lu <tonnylu@tencent.com>
Reviewed-by: default avatarHaozhong Zhang <hzhongzhang@tencent.com>
Reviewed-by: default avatarZongming Zhang <knightzhang@tencent.com>
Reviewed-by: default avatarMike Kravetz <mike.kravetz@oracle.com>
Acked-by: default avatarMatthew Wilcox <willy@infradead.org>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 188b04a7

mm/hugetlb.c

+15 −0
Original line number Diff line number Diff line
@@ -4338,6 +4338,21 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, 
				break;

			}

		}



		/*

		 * If subpage information not requested, update counters

		 * and skip the same_page loop below.

		 */

		if (!pages && !vmas && !pfn_offset &&

		    (vaddr + huge_page_size(h) < vma->vm_end) &&

		    (remainder >= pages_per_huge_page(h))) {

			vaddr += huge_page_size(h);

			remainder -= pages_per_huge_page(h);

			i += pages_per_huge_page(h);

			spin_unlock(ptl);

			continue;

		}



same_page:

		if (pages) {

			pages[i] = mem_map_offset(page, pfn_offset);