Re: [PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Mandeep Singh Baines
On Wed, Feb 20, 2013 at 4:42 PM, Andrew Morton wrote: > On Wed, 20 Feb 2013 16:28:07 -0800 > Mandeep Singh Baines wrote: > >> > Backtraces aren't *that* bad. We'll easily be able to tell which of >> > the two callsites triggered the trace. >> > >> >> Let's say there was a try_to_freeze() that go

Re: [PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Andrew Morton
On Wed, 20 Feb 2013 16:28:07 -0800 Mandeep Singh Baines wrote: > > Backtraces aren't *that* bad. We'll easily be able to tell which of > > the two callsites triggered the trace. > > > > Let's say there was a try_to_freeze() that got inlined indirectly > (multiple levels of inline) into do_exit.

Re: [PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Mandeep Singh Baines
On Wed, Feb 20, 2013 at 4:20 PM, Andrew Morton wrote: > On Wed, 20 Feb 2013 16:17:39 -0800 > Mandeep Singh Baines wrote: > >> On Wed, Feb 20, 2013 at 3:24 PM, Andrew Morton >> wrote: >> > On Wed, 20 Feb 2013 15:17:16 -0800 >> > Mandeep Singh Baines wrote: >> > >> >> We shouldn't try_to_freeze i

Re: [PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Andrew Morton
On Wed, 20 Feb 2013 16:17:39 -0800 Mandeep Singh Baines wrote: > On Wed, Feb 20, 2013 at 3:24 PM, Andrew Morton > wrote: > > On Wed, 20 Feb 2013 15:17:16 -0800 > > Mandeep Singh Baines wrote: > > > >> We shouldn't try_to_freeze if locks are held. > >> > >> ... > >> > >> @@ -43,6 +44,9 @@ extern

Re: [PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Mandeep Singh Baines
On Wed, Feb 20, 2013 at 3:24 PM, Andrew Morton wrote: > On Wed, 20 Feb 2013 15:17:16 -0800 > Mandeep Singh Baines wrote: > >> We shouldn't try_to_freeze if locks are held. >> >> ... >> >> @@ -43,6 +44,9 @@ extern void thaw_kernel_threads(void); >> >> + if (!(current->flags & PF_NOFREEZE)) >>

Re: [PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Andrew Morton
On Wed, 20 Feb 2013 15:17:16 -0800 Mandeep Singh Baines wrote: > We shouldn't try_to_freeze if locks are held. > > ... > > @@ -43,6 +44,9 @@ extern void thaw_kernel_threads(void); > > + if (!(current->flags & PF_NOFREEZE)) > + debug_check_no_locks_held(current, > + >

[PATCH v4] lockdep: check that no locks held at freeze time

2013-02-20 Thread Mandeep Singh Baines
We shouldn't try_to_freeze if locks are held. Changes since v1: * LKML: <20130215111635.ga26...@gmail.com> Ingo Molnar * Added a msg string that gets passed in. * LKML: <20130215154449.gd30...@redhat.com> Oleg Nesterov * Check PF_NOFREEZE in try_to_freeze(). Changes since v2: * LKML: <20130216