| Message ID | 4A76FAFE.6050906@users.sourceforge.net |
|---|---|
| State | New, archived |
| Headers | show |
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 5edc2bf..23c9475 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -99,7 +99,7 @@ static int jffs2_do_readpage_nolock (struct inode *inode, struct page *pg) kunmap(pg); D2(printk(KERN_DEBUG "readpage finished\n")); - return 0; + return ret; } int jffs2_do_readpage_unlock(struct inode *inode, struct page *pg)
[JFFS2] Fix return value from jffs2_do_readpage_nolock() This fixes "kernel BUG at fs/jffs2/file.c:251!". This pseudocode hopefully illustrates the scenario that triggers it: jffs2_write_begin { jffs2_do_readpage_nolock { jffs2_read_inode_range { jffs2_read_dnode { Data CRC 33c102e9 != calculated CRC 0ef77e7b for node at 005d42e4 return -EIO; } } ClearPageUptodate(pg); return 0; } } jffs2_write_end { BUG_ON(!PageUptodate(pg)); } Signed-off-by: Anders Grafström <grfstrm@users.sourceforge.net> --- This issue has resurfaced on linux-mtd a couple of times now. So here's the patch for it once more. This time with a proper commit message. fs/jffs2/file.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)