Re: [PATCH 18/35] Do not disable interrupts in free_page_mlock()

[Mel Gorman]

On Mon, Mar 16, 2009 at 12:05:53PM -0400, Christoph Lameter wrote:
> On Mon, 16 Mar 2009, Mel Gorman wrote:
> 
> > @@ -570,6 +570,8 @@ static void __free_pages_ok(struct page *page, unsigned int order)
> >  	kernel_map_pages(page, 1 << order, 0);
> >
> >  	local_irq_save(flags);
> > +	if (clearMlocked)
> > +		free_page_mlock(page);
> >  	__count_vm_events(PGFREE, 1 << order);
> >  	free_one_page(page_zone(page), page, order,
> >  					get_pageblock_migratetype(page));
> 
> Add an unlikely(clearMblocked) here?
> 

I wasn't sure at the time of writing how likely the case really is but it
makes sense that mlocked() pages are rarely freed. On reflection though,
it makes sense to mark this unlikely().

> > @@ -1036,6 +1039,9 @@ static void free_hot_cold_page(struct page *page, int cold)
> >  	pcp = &zone_pcp(zone, get_cpu())->pcp;
> >  	local_irq_save(flags);
> >  	__count_vm_event(PGFREE);
> > +	if (clearMlocked)
> > +		free_page_mlock(page);
> > +
> >  	if (cold)
> >  		list_add_tail(&page->lru, &pcp->list);
> >  	else
> >
> 
> Same here also make sure tha the __count_vm_events(PGFREE) comes after the
> free_pages_mlock() to preserve symmetry with __free_pages_ok() and maybe
> allow the compiler to do CSE between two invocations of
> __count_vm_events().
> 

Whatever about the latter reasoning about CSE, the symmetry makes sense.
I've made the change. Thanks

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab
--
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/