linux-kernel - [PATCH] mm/huge_memory.c: reduce critical section protected by split_queue

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>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20200110025516.23996-1-richardw.yang@linux.intel.com>
Date:   Fri, 10 Jan 2020 10:55:16 +0800
From:   Wei Yang <richardw.yang@...ux.intel.com>
To:     akpm@...ux-foundation.org
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        kirill.shutemov@...ux.intel.com,
        Wei Yang <richardw.yang@...ux.intel.com>
Subject: [PATCH] mm/huge_memory.c: reduce critical section protected by split_queue_lock

split_queue_lock protects data in struct deferred_split. We can release
the lock after delete the page from deferred_split_queue.

This patch moves the THP accounting out of the lock protection, which is
introduced in commit 65c453778aea ("mm, rmap: account shmem thp pages").

Signed-off-by: Wei Yang <richardw.yang@...ux.intel.com>
---
 mm/huge_memory.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 3c745d298a27..88c541bf44c7 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2810,6 +2810,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 			ds_queue->split_queue_len--;
 			list_del(page_deferred_list(head));
 		}
+		spin_unlock(&ds_queue->split_queue_lock);
 		if (mapping) {
 			if (PageSwapBacked(head))
 				__dec_node_page_state(head, NR_SHMEM_THPS);
@@ -2817,7 +2818,6 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 				__dec_node_page_state(head, NR_FILE_THPS);
 		}
 
-		spin_unlock(&ds_queue->split_queue_lock);
 		__split_huge_page(page, list, end, flags);
 		if (PageSwapCache(head)) {
 			swp_entry_t entry = { .val = page_private(head) };
-- 
2.17.1