Re: NFS causing oops when freeing namespace

2007-01-17 Thread Eric W. Biederman
Oleg Nesterov <[EMAIL PROTECTED]> writes: > On 01/17, Eric W. Biederman wrote: >> >> Cedric Le Goater <[EMAIL PROTECTED]> writes: >> > >> > your first analysis was correct : exit_task_namespaces() should be moved >> > above exit_notify(tsk). It will require some extra fixes for nsproxy >> > thou

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Oleg Nesterov
On 01/17, Eric W. Biederman wrote: > > Cedric Le Goater <[EMAIL PROTECTED]> writes: > > > > your first analysis was correct : exit_task_namespaces() should be moved > > above exit_notify(tsk). It will require some extra fixes for nsproxy > > though. > > I think the only issue is the child_reaper

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Trond Myklebust
On Wed, 2007-01-17 at 15:30 -0700, Eric W. Biederman wrote: > Cedric Le Goater <[EMAIL PROTECTED]> writes: > > > > your first analysis was correct : exit_task_namespaces() should be moved > > above exit_notify(tsk). It will require some extra fixes for nsproxy > > though. > > I think the only is

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Eric W. Biederman
Cedric Le Goater <[EMAIL PROTECTED]> writes: > > your first analysis was correct : exit_task_namespaces() should be moved > above exit_notify(tsk). It will require some extra fixes for nsproxy > though. I think the only issue is the child_reaper and currently we only have one of those. When we

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Cedric Le Goater
Oleg Nesterov wrote: > On 01/17, Cedric Le Goater wrote: >> Oleg Nesterov wrote: >>> On 01/17, Daniel Hokka Zakrisson wrote: It was the only semi-plausible explanation I could come up with. I added a printk in do_exit right before exit_task_namespaces, where sighand was still set, an

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Oleg Nesterov
On 01/17, Cedric Le Goater wrote: > > Oleg Nesterov wrote: > > On 01/17, Daniel Hokka Zakrisson wrote: > >> It was the only semi-plausible explanation I could come up with. I added a > >> printk in do_exit right before exit_task_namespaces, where sighand was > >> still set, and one right before the

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Cedric Le Goater
Oleg Nesterov wrote: > On 01/17, Daniel Hokka Zakrisson wrote: Call Trace: [] _spin_lock_irqsave+0x20/0x90 [] lockd_down+0x125/0x190 [] nfs_free_server+0x6d/0xd0 [] nfs_kill_super+0xc/0x20 [] deactivate_super+0x7d/0xa0 [] release_mounts+0x6e/0x80 [] __

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Oleg Nesterov
On 01/17, Daniel Hokka Zakrisson wrote: > > >> Call Trace: > >> [] _spin_lock_irqsave+0x20/0x90 > >> [] lockd_down+0x125/0x190 > >> [] nfs_free_server+0x6d/0xd0 > >> [] nfs_kill_super+0xc/0x20 > >> [] deactivate_super+0x7d/0xa0 > >> [] release_mounts+0x6e/0x80 > >> [] __put_mnt_ns+0x66/0x80

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Daniel Hokka Zakrisson
Eric W. Biederman wrote: > "Daniel Hokka Zakrisson" <[EMAIL PROTECTED]> writes: > >> The test-case at the bottom causes the following recursive Oopsing on >> 2.6.20-rc5: > > A few more people added to the CC who might have a clue. > >> >> BUG: unable to handle kernel NULL pointer dereference at vir

Re: NFS causing oops when freeing namespace

2007-01-17 Thread Eric W. Biederman
"Daniel Hokka Zakrisson" <[EMAIL PROTECTED]> writes: > The test-case at the bottom causes the following recursive Oopsing on > 2.6.20-rc5: A few more people added to the CC who might have a clue. > > BUG: unable to handle kernel NULL pointer dereference at virtual address > 0504 > printing