[ofa-general] Re: [PATCH 08 of 11] anon-vma-rwsem

[Peter Zijlstra]

On Thu, 2008-05-08 at 09:11 -0700, Linus Torvalds wrote:
> 
> On Thu, 8 May 2008, Linus Torvalds wrote:
> > 
> > Also, we'd need to make it 
> > 
> > 	unsigned short flag:1;
> > 
> > _and_ change spinlock_types.h to make the spinlock size actually match the 
> > required size (right now we make it an "unsigned int slock" even when we 
> > actually only use 16 bits).
> 
> Btw, this is an issue only on 32-bit x86, because on 64-bit one we already 
> have the padding due to the alignment of the 64-bit pointers in the 
> list_head (so there's already empty space there).
> 
> On 32-bit, the alignment of list-head is obviously just 32 bits, so right 
> now the structure is "perfectly packed" and doesn't have any empty space. 
> But that's just because the spinlock is unnecessarily big.
> 
> (Of course, if anybody really uses NR_CPUS >= 256 on 32-bit x86, then the 
> structure really will grow. That's a very odd configuration, though, and 
> not one I feel we really need to care about).

Another possibility, would something like this work?

 
 /*
  * null out the begin function, no new begin calls can be made
  */
 rcu_assing_pointer(my_notifier.invalidate_start_begin, NULL); 

 /*
  * lock/unlock all rmap locks in any order - this ensures that any
  * pending start() will have its end() function called.
  */
 mm_barrier(mm);

 /*
  * now that no new start() call can be made and all start()/end() pairs
  * are complete we can remove the notifier.
  */
 mmu_notifier_remove(mm, my_notifier);


This requires a mmu_notifier instance per attached mm and that
__mmu_notifier_invalidate_range_start() uses rcu_dereference() to obtain
the function.

But I think its enough to ensure that:

  for each start an end will be called

It can however happen that end is called without start - but we could
handle that I think.