Hi Andreas, On Tue, May 23, 2017 at 10:19:35AM -0600, Andreas Dilger wrote: > On May 21, 2017, at 12:23 AM, Eric Biggers <ebiggers3@xxxxxxxxx> wrote: > > > > From: Eric Biggers <ebiggers@xxxxxxxxxx> > > > > ext2fs_inode_data_blocks2() calculates an inode's data block count by > > subtracting the external xattr block, if any, from the total blocks. > > But on bigalloc filesystems, the xattr "block" is actually a whole > > cluster, so ext2fs_inode_data_blocks2() would return a too-large value. > > > > It seems this could have caused several different problems, but the one > > I encountered was that xfstest generic/399 failed in the "bigalloc" > > config because e2fsck incorrectly considered a symlink on the filesystem > > to be corrupted at the end of the test. This happened because e2fsck > > incorrectly calculated a nonzero data block count for a "fast" symlink > > with an external xattr block and therefore treated it as a "slow" > > symlink, which failed validation. > > I thought we changed this to detect "fast" inodes by i_size < 60 rather > than using the blocks count, because the blocks count was (and apparently > continues to be) unreliable for determining fast vs. slow symlinks. > > However, ext4_inode_is_fast_symlink() still checks blocks count. In > "[PATCH] ext4: fix reading new encrypted symlinks on no-journal filesystems" > we discussed whether this was safe, and it appears to be OK from my > analysis. > > We just continue to hit problems when extrapolating various blocks counts > to detect fast symlinks rather than just using the same mechanism we use > at creation time, which is "len > EXT4_N_BLOCKS * 4". > > Cheers, Andreas > Yes, I still think we probably should do that. This bug needed to be fixed anyway though, since ext2fs_inode_data_blocks2() is used for a bit more than just distinguishing between fast and slow symlinks. Eric
