On Fri, 14 Jan 2011 07:22:56 -0500 Jeff Layton <jlayton@xxxxxxxxxx> wrote: > Testing with the connectathon tests caused this panic: > > [ 388.266918] ------------[ cut here ]------------ > [ 388.267823] kernel BUG at fs/dcache.c:1358! > > [...] > > [ 388.267823] RIP: 0010:[<ffffffff81129146>] [<ffffffff81129146>] d_set_d_op+0x10/0x58 > > [...] > > [ 388.267823] Call Trace: > [ 388.267823] [<ffffffffa0147a3c>] cifs_create+0x5cd/0x6fe [cifs] > [ 388.267823] [<ffffffff81123c5d>] vfs_create+0x70/0x91 > [ 388.267823] [<ffffffff8112499d>] do_last+0x15a/0x2ff > [ 388.267823] [<ffffffff81124e41>] do_filp_open+0x2ff/0x6db > [ 388.267823] [<ffffffff8110ca5f>] ? kmem_cache_free+0x76/0xb4 > [ 388.267823] [<ffffffff81464c6b>] ? _cond_resched+0xe/0x22 > [ 388.267823] [<ffffffff81229300>] ? might_fault+0x21/0x23 > [ 388.267823] [<ffffffff81229400>] ? __strncpy_from_user+0x1f/0x4e > [ 388.267823] [<ffffffff8112eb70>] ? alloc_fd+0x74/0x11f > [ 388.267823] [<ffffffff81118808>] do_sys_open+0x60/0xf2 > [ 388.267823] [<ffffffff810992e3>] ? audit_syscall_entry+0x11c/0x148 > [ 388.267823] [<ffffffff811188ba>] sys_open+0x20/0x22 > [ 388.267823] [<ffffffff811188e4>] sys_creat+0x15/0x17 > [ 388.267823] [<ffffffff8100ac42>] system_call_fastpath+0x16/0x1b > [ 388.436446] ---[ end trace a012e6143ebad719 ]--- > > The problem is that d_op is already set during the lookup. d_set_d_op > will BUG() if you try to set operations when they have already been set. > > Remove all of the d_set_d_op calls except for during lookups and ensure > that it gets set on every lookup. > > Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx> > --- > fs/cifs/dir.c | 32 +++++++------------------------- > fs/cifs/inode.c | 9 --------- > fs/cifs/link.c | 9 ++------- > 3 files changed, 9 insertions(+), 41 deletions(-) > Looks like Al's most recent changes should fix this already. I got this panic on a kernel based on a pull earlier in the day. Please just ignore this patch. Sorry for the noise! -- Jeff Layton <jlayton@xxxxxxxxxx> -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
