On Wed, Jan 21, 2015 at 12:05:39PM +0100, Sabrina Dubroca wrote: > 2015-01-21, 04:36:38 +0000, Al Viro wrote: > > On Tue, Jan 20, 2015 at 08:01:26PM -0800, Guenter Roeck wrote: > > > With this patch: > > > > > > sys_mkdir .:40775 returned -17 > > > sys_mkdir usr:40775 returned 0 > > > sys_mkdir usr/lib:40775 returned 0 > > > sys_mkdir usr/share:40755 returned 0 > > > sys_mkdir usr/share/udhcpc:40755 returned 0 > > > sys_mkdir usr/bin:40775 returned 0 > > > sys_mkdir usr/sbin:40775 returned 0 > > > sys_mkdir mnt:40775 returned 0 > > > sys_mkdir proc:40775 returned 0 > > > sys_mkdir root:40775 returned 0 > > > sys_mkdir lib:40775 returned 0 > > > sys_mkdir lib/modules:40775 returned 0 > > > ... > > > > > > and the problem is fixed. > > This patch also works for me. > > > > ... except that it simply confirms that something's fishy with getname_kernel() > > of ->name of struct filename returned by getname(). IOW, I still do not > > understand the mechanism of breakage there. > > I'm not so sure about that. I tried to copy name to a new string in > do_path_lookup and that didn't help. > > Now, I've removed the > > putname(filename); > > line from do_path_lookup and I don't get the panic. That would indicate that somehow the refcount got unbalanced. Looking more closely it seems like the various audit_*() function do take a reference, but maybe that's not enough. But debugging this further I see no indication that the memory is ever freed, or otherwise corrupted. I did collect a bit more data, perhaps that's useful. I started seeing this issue as well on devices that boot over NFS. After reading this thread I also realized that another warning that I was seeing might be related: [ 28.261930] Warning: unable to open an initial console. I've added a couple of printks and see that the reason for this is that /dev/console doesn't get created. /dev however does get created. [ 11.786627] sys_mkdir dev:40755 returned 0 ... [ 11.978748] sys_mknod dev/console:20600 returned -2 The chain that fails turns out to be this: sys_mknod() sys_mknodat() user_path_create() kern_path_create() do_path_lookup() filename_lookup() path_lookupat() path_init() link_path_walk() walk_component() walk_components() ends up calling lookup_slow() and the result is that inode == NULL and d_is_negative(path->dentry) returns true, therefore causing -ENOENT to be returned. I tried to figure out why inode would be NULL at that point or why d_is_negative() returned true, but I ended up getting completely lost, so I thought it best to report my findings before I confuse everything. Is there anything else I can investigate to track this down? Thierry
Attachment:
pgpcAiu2lhGD6.pgp
Description: PGP signature
