Re: [PATCH] mm: poison struct page for ptlock

[Hugh Dickins]

On Sun, 6 Nov 2005, Andrew Morton wrote:
> Hugh Dickins <hugh@xxxxxxxxxxx> wrote:
> >
> > Suppress split ptlock on arches which may use one page for multiple page
> >  tables.  Reconsider what better to do (particularly on ppc64) later on.
> 
> But why is that a problem per-se?  A few "pagetable pages" will share the
> same lock.  Deadlocky when we take two pagetable locks?

Good point, it may not be as disastrous as I was imagining.  Certainly
deadlocky in copy_page_range and mremap move at present, but that can
be easily surmounted.  However, there is the slab lru issue too: if it's
convenient for an arch to use slab cache, I'd prefer not to make life
difficult for them.  I'll come back to this but not tonight.

Hugh
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/