linux-kernel - [PATCH 3/3 v2] mm: Batch page_check_references in shrink_page_list sharing the same i_mmap

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]

Date:	Mon, 10 Sep 2012 09:19:32 -0700
From:	Tim Chen <tim.c.chen@...ux.intel.com>
To:	Andrew Morton <akpm@...ux-foundation.org>,
	Mel Gorman <mel@....ul.ie>, Minchan Kim <minchan@...nel.org>,
	Johannes Weiner <hannes@...xchg.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	Andrea Arcangeli <aarcange@...hat.com>,
	David Rientjes <rientjes@...gle.com>,
	Michal Hocko <mhocko@...e.cz>,
	Xiao Guangrong <xiaoguangrong@...ux.vnet.ibm.com>,
	Paul Gortmaker <paul.gortmaker@...driver.com>
Cc:	Andi Kleen <ak@...ux.intel.com>, linux-mm@...ck.org,
	linux-kernel <linux-kernel@...r.kernel.org>,
	Alex Shi <alex.shi@...el.com>,
	Matthew Wilcox <willy@...ux.intel.com>,
	"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
	Fengguang Wu <fengguang.wu@...el.com>
Subject: [PATCH 3/3 v2] mm: Batch page_check_references in shrink_page_list
 sharing the same i_mmap_mutex

In shrink_page_list, call to page_referenced_file and try_to_unmap will cause the
acquisition/release of mapping->i_mmap_mutex for each page in the page
list.  However, it is very likely that successive pages in the list
share the same mapping and we can reduce the frequency of i_mmap_mutex
acquisition by holding the mutex in shrink_page_list before calling
__page_referenced and __try_to_unmap. This improves the
performance when the system has a lot page reclamations for file mapped
pages if workloads are using a lot of memory for page cache.

Tim

---
Signed-off-by: Tim Chen <tim.c.chen@...ux.intel.com>
Signed-off-by: Matthew Wilcox <willy@...ux.intel.com>
--- 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index d4ab646..0428639 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -786,7 +786,7 @@ static enum page_references page_check_references(struct page *page,
 	int referenced_ptes, referenced_page;
 	unsigned long vm_flags;
 
-	referenced_ptes = page_referenced(page, 1, mz->mem_cgroup, &vm_flags);
+	referenced_ptes = __page_referenced(page, 1, mz->mem_cgroup, &vm_flags);
 	referenced_page = TestClearPageReferenced(page);
 
 	/* Lumpy reclaim - ignore references */
@@ -856,6 +856,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 	unsigned long nr_congested = 0;
 	unsigned long nr_reclaimed = 0;
 	unsigned long nr_writeback = 0;
+	struct mutex *i_mmap_mutex = NULL;
 
 	cond_resched();
 
@@ -909,7 +910,15 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 			}
 		}
 
+		if (needs_page_mmap_mutex(page) &&
+			       i_mmap_mutex != &page->mapping->i_mmap_mutex) {
+			if (i_mmap_mutex)
+				mutex_unlock(i_mmap_mutex);
+			i_mmap_mutex = &page->mapping->i_mmap_mutex;
+			mutex_lock(i_mmap_mutex);
+		}
 		references = page_check_references(page, mz, sc);
+
 		switch (references) {
 		case PAGEREF_ACTIVATE:
 			goto activate_locked;
@@ -939,7 +948,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 		 * processes. Try to unmap it here.
 		 */
 		if (page_mapped(page) && mapping) {
-			switch (try_to_unmap(page, TTU_UNMAP)) {
+			switch (__try_to_unmap(page, TTU_UNMAP)) {
 			case SWAP_FAIL:
 				goto activate_locked;
 			case SWAP_AGAIN:
@@ -1090,6 +1099,8 @@ keep_lumpy:
 		VM_BUG_ON(PageLRU(page) || PageUnevictable(page));
 	}
 
+	if (i_mmap_mutex)
+		mutex_unlock(i_mmap_mutex);
 	nr_reclaimed += __remove_mapping_batch(&unmap_pages, &ret_pages,
 					       &free_pages);
 






--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/