On Mon, 2019-06-03 at 10:18 -0400, Steve Dickson wrote:
>
> On 5/31/19 12:02 PM, J. Bruce Fields wrote:
> > On Tue, May 28, 2019 at 04:31:19PM -0400, Trond Myklebust wrote:
> > > @@ -373,21 +390,22 @@ static char *next_mnt(void **v, char *p)
> > > FILE *f;
> > > struct mntent *me;
> > > size_t l = strlen(p);
> > > + char *mnt_dir = NULL;
> > > +
> > > if (*v == NULL) {
> > > f = setmntent("/etc/mtab", "r");
> > > *v = f;
> > > } else
> > > f = *v;
> > > - while ((me = getmntent(f)) != NULL && l > 1 &&
> > > - (strncmp(me->mnt_dir, p, l) != 0 ||
> > > - me->mnt_dir[l] != '/'))
> > > - ;
> > > - if (me == NULL) {
> > > - endmntent(f);
> > > - *v = NULL;
> > > - return NULL;
> > > + while ((me = getmntent(f)) != NULL && l > 1) {
> > > + mnt_dir = nfsd_path_strip_root(me->mnt_dir);
> > > +
> > > + if (strncmp(mnt_dir, p, l) == 0 && mnt_dir[l] != '/')
> > > + return mnt_dir;
> >
> > That should be "mnt_dir[l] == '/'", right?
> Comment says
> /* Iterate through /etc/mtab, finding mountpoints
> * at or below a given path
> */
>
> So I don't think the actual '/' should returned, Trond?
Yep. You're both right. I've got an incremental patch to fix this.
--
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@xxxxxxxxxxxxxxx
