linux-kernel - [PATCH v3] mm/hugetlb: avoid looping to the same hugepage if !pages and !vmas

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-Id: <1567581712-5992-1-git-send-email-totty.lu@gmail.com>
Date:   Wed,  4 Sep 2019 15:21:52 +0800
From:   Zhigang Lu <totty.lu@...il.com>
To:     luzhigang001@...il.com, mike.kravetz@...cle.com,
        willy@...radead.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Cc:     Zhigang Lu <tonnylu@...cent.com>
Subject: [PATCH v3] mm/hugetlb: avoid looping to the same hugepage if !pages and !vmas

From: Zhigang Lu <tonnylu@...cent.com>

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.

Signed-off-by: Zhigang Lu <tonnylu@...cent.com>
Reviewed-by: Haozhong Zhang <hzhongzhang@...cent.com>
Reviewed-by: Zongming Zhang <knightzhang@...cent.com>
Reviewed-by: Mike Kravetz <mike.kravetz@...cle.com>
Acked-by: Matthew Wilcox <willy@...radead.org>
---
 mm/hugetlb.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 6d7296d..a096fb3 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -4391,6 +4391,21 @@ 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);
-- 
1.8.3.1