On Mon, Oct 01, 2018 at 04:33:36PM -0700, Darrick J. Wong wrote: > On Tue, Oct 02, 2018 at 09:21:28AM +1000, Dave Chinner wrote: > > On Mon, Oct 01, 2018 at 03:45:00PM -0700, Darrick J. Wong wrote: > > > From: Darrick J. Wong <darrick.wong@xxxxxxxxxx> > > > > > > Teach callers of inode->i_op->get_link in the vfs code to check for a > > > NULL return value and return an error status instead of blindly > > > dereferencing the returned NULL pointer. > > > > > > Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx> > > > --- > > > fs/namei.c | 4 ++++ > > > 1 file changed, 4 insertions(+) > > > > > > diff --git a/fs/namei.c b/fs/namei.c > > > index 0cab6494978c..0744ab981fa0 100644 > > > --- a/fs/namei.c > > > +++ b/fs/namei.c > > > @@ -4737,6 +4737,8 @@ int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen) > > > if (IS_ERR(link)) > > > return PTR_ERR(link); > > > } > > > + if (!link) > > > + return -EUCLEAN; > > > > If we are going to start returning this as a filesystem corruption > > error from the VFS, can we please start with adding > > > > #define EFSCORRUPTED EUCLEAN > > > > into one of the global error definition headers? The code makes much > > more sense when it uses EFSCORRUPTED.... > > Ok, which header file? include/uapi/asm-generic/errno.h ? Seems reasonable, because that's where EUCLEAN is defined. > We'll finally be able to get rid of the redundant definitions in > ext[24] too. And filesystems like btrfs can be converted from EUCLEAN to EFSCORRUPTED, too. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx
