Re: VFS deadlock ?

[Dave Jones <davej@redhat.com>]

On Thu, Mar 21, 2013 at 09:26:17PM +0000, Al Viro wrote:
 > On Thu, Mar 21, 2013 at 02:18:15PM -0700, Linus Torvalds wrote:
 > > On Thu, Mar 21, 2013 at 2:02 PM, Dave Jones <davej@redhat.com> wrote:
 > > >
 > > > here we go...
 > > >
 > > > WARNING: at fs/namei.c:2335 lock_rename+0x156/0x160()
 > > > p1=irda p2=irda
 > > 
 > > Ok, good. I ssupect it's /proc or /sys, we do have that entry there.
 > > 
 > > But in fact I suspect we do want the parent name after all, because I
 > > think we have multiple "irda" directories. There's the one in
 > > /proc/net/ (added by net/irda/irproc.c), and there's a sysctl  CTL_DIR
 > > "irda" directory (kernel/sysctl_binary.c). And there might even be a
 > > few other ones in /sys too, thanks to the ldisc called "irda" etc.
 > > 
 > > I don't see where the shared inode comes from, but I suspect that
 > > would be easier to guess if we actually see which particular case it
 > > ends up being..
 > 
 > Well, something like
 > 	static char path[4096];
 > 	d_absolute_path(p1, path, 4096);
 > 	printk(KERN_ERR "%s %s %d %d"
 > 		path, p1->d_sb->s_type->name, d_unlinked(p1), d_unlinked(p2));
 > might be useful - pathname within fs, fs type and which of those suckers are
 > unlinked...

uh..

fs/namei.c:2342:3: warning: passing argument 1 of ‘d_absolute_path’ from incompatible pointer type [enabled by default]
   d_absolute_path(p1, path, 4096);
   ^
In file included from include/linux/fs.h:8:0,
                 from fs/namei.c:21:
include/linux/dcache.h:337:14: note: expected ‘const struct path *’ but argument is of type ‘struct dentry *’
 extern char *d_absolute_path(const struct path *, char *, int);
              ^


How do I go from the dentry in p1 to the path d_absolute_path expects ?

	Dave