Re: [COUNTERPATCH] mm: avoid overflowing preempt_count() in mmu_take_all_locks()

[Peter Zijlstra <peterz@infradead.org>]

On Thu, 2010-04-01 at 14:17 +0300, Avi Kivity wrote:
> On 04/01/2010 02:13 PM, Avi Kivity wrote:
> >
> >> Anyway, I don't see a reason why we can't convert those locks to
> >> mutexes and get rid of the whole preempt disabled region.
> >
> > If someone is willing to audit all code paths to make sure these locks 
> > are always taken in schedulable context I agree that's a better fix.
> >
> 
>  From mm/rmap.c:
> 
> > /*
> >  * Lock ordering in mm:
> >  *
> >  * inode->i_mutex    (while writing or truncating, not reading or 
> > faulting)
> >  *   inode->i_alloc_sem (vmtruncate_range)
> >  *   mm->mmap_sem
> >  *     page->flags PG_locked (lock_page)
> >  *       mapping->i_mmap_lock
> >  *         anon_vma->lock
> ...
> > *
> >  * (code doesn't rely on that order so it could be switched around)
> >  * ->tasklist_lock
> >  *   anon_vma->lock      (memory_failure, collect_procs_anon)
> >  *     pte map lock
> >  */
> 
> i_mmap_lock is a spinlock, and tasklist_lock is a rwlock, so some 
> changes will be needed.

i_mmap_lock will need to change just as well, mm_take_all_locks() uses
both anon_vma->lock and mapping->i_mmap_lock.

I've almost got a patch done that converts those two, still need to look
where that tasklist_lock muck happens.