On Sun, May 22, 2016 at 01:34:59PM +0800, Wang Shilong wrote: > On Sun, May 22, 2016 at 10:19 AM, Theodore Ts'o <tytso@xxxxxxx> wrote: > > Creating a file system with project quotas can fail if mke2fs is built > > using hardening options. This is because quota_compute_usage() used > > ext2fs_get_next_inode() instead of ext2fs_get_inode_full(), and a > > small inode was passed into quota_data_add, when a large inode needs > > to be used. As a result get_dq() would end up dereferencing undefined > > space in the stack. Without the hardening options, this would be > > zero, so "mke2fs -t ext4 -O project.quota -I 256 test.img" would work > > essentially by accident. > > > > Fix this by using ext2fs_get_inode_full() so that a large inode is > > available to quota_data_inodes(). > > > > Signed-off-by: Theodore Ts'o <tytso@xxxxxxx> > > I thought Li Xi sent updated e2fsprogs including this fixing parts. > maybe because you merged early version patches. There was a separate patch that broke ABI backwards compatibility of e2fsprogs' shared libraries, which I rejected on those grounds, perhaps that's what you thinking of? It wasn't clear that the patch was in fact fixing a problem, as opposed to just being a clean up. So I didn't realize there was a problem that needed fixing. - Ted -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html