Re: vfs: lockdep splat with prepare_bprm_creds

[Oleg Nesterov <oleg@redhat.com>]

On 03/15, Al Viro wrote:
>
> On Fri, Mar 15, 2013 at 12:07:14AM -0400, Sasha Levin wrote:
> > Hi all,
> >
> > While fuzzing with trinity inside a KVM tools guest running latest -next kernel
> > I've stumbled on the following.
> >
> > Dave Jones reported something similar, but that seemed to involve cgroup's mutex
> > and didn't seem like it was the same issue as this one.
>
> Lovely...  It's an execve() attempt on a "binary" that is, in fact, a procfs
> file (/proc/<pid>/stack),

probably... other lock_trace() callers can't generate this lockdep output
afaics.

> with its ->read() trying to grab ->cred_guard_mutex.
> The fact that it's seq_file-based is irrelevant here - all that matters is
> that we have ->read() for some file trying to grab ->cred_guard_mutex.

Yes, perhaps the patch below makes sense anyway as a cleanup, but obviously
it can't help.

Cough... I am shy to disclose my ignorance, but could you explain how
open_exec()->do_filp_open(MAY_EXEC) can succeed in this case? At least
acl_permission_check() looks as if open_exec() should fail...

Just curious, thanks in advance.

Oleg.

--- x/fs/proc/base.c
+++ x/fs/proc/base.c
@@ -317,12 +317,12 @@ static int proc_pid_stack(struct seq_fil
 	err = lock_trace(task);
 	if (!err) {
 		save_stack_trace_tsk(task, &trace);
+		unlock_trace(task);
 
 		for (i = 0; i < trace.nr_entries; i++) {
 			seq_printf(m, "[<%pK>] %pS\n",
 				   (void *)entries[i], (void *)entries[i]);
 		}
-		unlock_trace(task);
 	}
 	kfree(entries);