[Bug 49030] New: Possible recursive locking detected in r600g

* [Bug 49030] New: Possible recursive locking detected in r600g
@ 2012-04-21  4:23 bugzilla-daemon
  2012-04-23  9:24 ` [Bug 49030] " bugzilla-daemon
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: bugzilla-daemon @ 2012-04-21  4:23 UTC (permalink / raw)
  To: dri-devel

https://bugs.freedesktop.org/show_bug.cgi?id=49030

             Bug #: 49030
           Summary: Possible recursive locking detected in r600g
    Classification: Unclassified
           Product: Mesa
           Version: git
          Platform: Other
        OS/Version: All
            Status: NEW
          Severity: minor
          Priority: medium
         Component: Drivers/Gallium/r600
        AssignedTo: dri-devel@lists.freedesktop.org
        ReportedBy: alexandre.f.demers@gmail.com

Here is what I see under dmesg:

=============================================
[ INFO: possible recursive locking detected ]
3.4.0-rc3+ #120 Tainted: G         C  
---------------------------------------------
gnome-shell/1120 is trying to acquire lock:
 (&vm->mutex){+.+...}, at: [<ffffffffa034be9a>] radeon_vm_unbind+0x24/0x40
[radeon]

but task is already holding lock:
 (&vm->mutex){+.+...}, at: [<ffffffffa035b84c>] radeon_cs_ioctl+0x434/0x598
[radeon]

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&vm->mutex);
  lock(&vm->mutex);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

2 locks held by gnome-shell/1120:
 #0:  (&mutex->mutex){+.+.+.}, at: [<ffffffffa035b44e>]
radeon_cs_ioctl+0x36/0x598 [radeon]
 #1:  (&vm->mutex){+.+...}, at: [<ffffffffa035b84c>]
radeon_cs_ioctl+0x434/0x598 [radeon]

stack backtrace:
Pid: 1120, comm: gnome-shell Tainted: G         C   3.4.0-rc3+ #120
Call Trace:
 [<ffffffff8108acd9>] __lock_acquire+0xbf6/0xd11
 [<ffffffff8108a5a3>] ? __lock_acquire+0x4c0/0xd11
 [<ffffffff8108b211>] lock_acquire+0x92/0x104
 [<ffffffffa034be9a>] ? radeon_vm_unbind+0x24/0x40 [radeon]
 [<ffffffff8108a5a3>] ? __lock_acquire+0x4c0/0xd11
 [<ffffffff814d4460>] __mutex_lock_common+0x48/0x34e
 [<ffffffffa034be9a>] ? radeon_vm_unbind+0x24/0x40 [radeon]
 [<ffffffffa034be9a>] ? radeon_vm_unbind+0x24/0x40 [radeon]
 [<ffffffff8108b69e>] ? trace_hardirqs_on_caller+0x123/0x17f
 [<ffffffff814d4839>] mutex_lock_nested+0x2f/0x36
 [<ffffffffa035b84c>] ? radeon_cs_ioctl+0x434/0x598 [radeon]
 [<ffffffffa034be9a>] radeon_vm_unbind+0x24/0x40 [radeon]
 [<ffffffffa034c2f9>] radeon_vm_bind+0xb9/0x1c4 [radeon]
 [<ffffffffa035b857>] radeon_cs_ioctl+0x43f/0x598 [radeon]
 [<ffffffffa01fa9f9>] drm_ioctl+0x2d6/0x3c0 [drm]
 [<ffffffff8128dfed>] ? rcu_read_unlock+0x1c/0x1e
 [<ffffffffa035b418>] ? radeon_cs_finish_pages+0x91/0x91 [radeon]
 [<ffffffff8128e44a>] ? avc_has_perm_flags+0x6b/0x7f
 [<ffffffff8113c20c>] vfs_ioctl+0x24/0x2f
 [<ffffffff8113caf7>] do_vfs_ioctl+0x412/0x455
 [<ffffffff8113cb90>] sys_ioctl+0x56/0x7a
 [<ffffffff814dd3ad>] system_call_fastpath+0x1a/0x1f

It seems to be only a warning. However, if possible, it should be silenced if
there is no reason to have it. I know I reported something similar a couple of
months ago and it was fixed by moving some code around.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

^ permalink raw reply	[flat|nested] 4+ messages in thread