[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20200302093427.49796-1-yuchao0@huawei.com>
Date: Mon, 2 Mar 2020 17:34:27 +0800
From: Chao Yu <yuchao0@...wei.com>
To: <jaegeuk@...nel.org>
CC: <linux-f2fs-devel@...ts.sourceforge.net>,
<linux-kernel@...r.kernel.org>, <chao@...nel.org>,
Chao Yu <yuchao0@...wei.com>
Subject: [PATCH] f2fs: fix to avoid using uninitialized variable
In f2fs_vm_page_mkwrite(), if inode is compress one, and current mmapped
page locates in compressed cluster, we have to call f2fs_get_dnode_of_data()
to get its physical block address before f2fs_wait_on_block_writeback().
Signed-off-by: Chao Yu <yuchao0@...wei.com>
---
fs/f2fs/file.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 09164c6a1d39..caa7ca42791f 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -106,10 +106,18 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
err = f2fs_get_block(&dn, page->index);
f2fs_put_dnode(&dn);
__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
- if (err) {
- unlock_page(page);
- goto out_sem;
- }
+ }
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+ if (!need_alloc) {
+ set_new_dnode(&dn, inode, NULL, NULL, 0);
+ err = f2fs_get_dnode_of_data(&dn, page->index, LOOKUP_NODE);
+ f2fs_put_dnode(&dn);
+ }
+#endif
+ if (err) {
+ unlock_page(page);
+ goto out_sem;
}
f2fs_wait_on_page_writeback(page, DATA, false, true);
--
2.18.0.rc1
Powered by blists - more mailing lists