Messages in this thread | | | Date | Mon, 15 Oct 2012 22:20:14 +0900 | Subject | Re: [PATCH v4 3/4] fat (exportfs): rebuild directory-inode if fat_dget() fails | From | Namjae Jeon <> |
| |
2012/10/15 OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>: > Namjae Jeon <linkinjeon@gmail.com> writes: > >>> OK, I think I got where is wrong. If it is the subdir of rootdir, >>> fat_dget() should get the inode of root? >> That is correct. Whenever fat_dget() is called for sub-directories >> which is in root folder, we assign root inode as parent inode. >>> >>> So, I guess if parent_logstart == 0, it is BUG(). >> Would you explain more why you think it is BUG() ? > > Because root dentry is never expired until umount. So, fat_dget() > shouldn't never fail to get inode for subdir of rootdir. Otherwise, > "stale_rw" will not be working in the case even if there is cache. > > I.e. I think we don't need to check parent_logstart == 0. Yes, You're right. I found this condition is really not needed when I check again. I will remove this code. Thanks OGAWA! > > Thanks. > -- > OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
| |