| Message ID | 20230825214551.136149-1-tytso@mit.edu |
|---|---|
| State | Accepted |
| Headers | show |
| Series | libext2fs: don't truncate the orphan file inode if it is newly allocated | expand |
* Theodore Ts'o <tytso@mit.edu>: > In ext2fs_create_orphan_file(), don't try truncating inode for the > orphan file if ext2fs_create_orphan_file() allocated the inode. This > avoids problems where the newly allocated inode in the inode table > might contain garbage; if the metadata checksum feature is enabled, > this will generally result in the function failing with a checksum > invalid error, but this can cause mke2fs (which calls > ext2fs_create_orphan_file) to fail. > > Signed-off-by: Theodore Ts'o <tytso@mit.edu> > --- > lib/ext2fs/orphan.c | 21 ++++++++++----------- > 1 file changed, 10 insertions(+), 11 deletions(-) > > diff --git a/lib/ext2fs/orphan.c b/lib/ext2fs/orphan.c > index e25f20ca2..c2f83567f 100644 > --- a/lib/ext2fs/orphan.c > +++ b/lib/ext2fs/orphan.c > @@ -127,22 +127,21 @@ errcode_t ext2fs_create_orphan_file(ext2_filsys fs, blk_t num_blocks) > struct mkorphan_info oi; > struct ext4_orphan_block_tail *ob_tail; > > - if (!ino) { > + if (ino) { > + err = ext2fs_read_inode(fs, ino, &inode); > + if (err) > + return err; > + if (EXT2_I_SIZE(&inode)) { > + err = ext2fs_truncate_orphan_file(fs); > + if (err) > + return err; > + } > + } else { > err = ext2fs_new_inode(fs, EXT2_ROOT_INO, LINUX_S_IFREG | 0600, > 0, &ino); > if (err) > return err; > ext2fs_inode_alloc_stats2(fs, ino, +1, 0); > - ext2fs_mark_ib_dirty(fs); > - } > - > - err = ext2fs_read_inode(fs, ino, &inode); > - if (err) > - return err; > - if (EXT2_I_SIZE(&inode)) { > - err = ext2fs_truncate_orphan_file(fs); > - if (err) > - return err; > } > > memset(&inode, 0, sizeof(struct ext2_inode)); > -- > 2.31.0 > With this patch to mke2fs, I ran ext4/049 on a 6.5 kernel using kvm-xfstests' 4k test scenario. It passed 110/110 trials. Without the patch, it fails 100% of the time. So, it addresses the failures I've been seeing. Thanks! Tested-by: Eric Whitney <enwlinux@gmail.com>
diff --git a/lib/ext2fs/orphan.c b/lib/ext2fs/orphan.c index e25f20ca2..c2f83567f 100644 --- a/lib/ext2fs/orphan.c +++ b/lib/ext2fs/orphan.c @@ -127,22 +127,21 @@ errcode_t ext2fs_create_orphan_file(ext2_filsys fs, blk_t num_blocks) struct mkorphan_info oi; struct ext4_orphan_block_tail *ob_tail; - if (!ino) { + if (ino) { + err = ext2fs_read_inode(fs, ino, &inode); + if (err) + return err; + if (EXT2_I_SIZE(&inode)) { + err = ext2fs_truncate_orphan_file(fs); + if (err) + return err; + } + } else { err = ext2fs_new_inode(fs, EXT2_ROOT_INO, LINUX_S_IFREG | 0600, 0, &ino); if (err) return err; ext2fs_inode_alloc_stats2(fs, ino, +1, 0); - ext2fs_mark_ib_dirty(fs); - } - - err = ext2fs_read_inode(fs, ino, &inode); - if (err) - return err; - if (EXT2_I_SIZE(&inode)) { - err = ext2fs_truncate_orphan_file(fs); - if (err) - return err; } memset(&inode, 0, sizeof(struct ext2_inode));
In ext2fs_create_orphan_file(), don't try truncating inode for the orphan file if ext2fs_create_orphan_file() allocated the inode. This avoids problems where the newly allocated inode in the inode table might contain garbage; if the metadata checksum feature is enabled, this will generally result in the function failing with a checksum invalid error, but this can cause mke2fs (which calls ext2fs_create_orphan_file) to fail. Signed-off-by: Theodore Ts'o <tytso@mit.edu> --- lib/ext2fs/orphan.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-)