On Wed, 21 Nov 2012 22:36:56 +0800
Jiang Liu wrote:
> > void mod_zone_managed_pages(struct zone *zone, signed long delta)
> > {
> > WARN_ON(system_state != SYSTEM_BOOTING &&
> > !is_locked_memory_hotplug());
> > zone->managed_pages += delta;
> > }
> This seems a little overhea
On 11/21/2012 03:31 AM, Andrew Morton wrote:
@@ -106,6 +106,7 @@ static void get_page_bootmem(unsigned long info,
struct page *page,
void __ref put_page_bootmem(struct page *page)
{
unsigned long type;
+ static DEFINE_MUTEX(ppb_lock);
type = (un
On Tue, 20 Nov 2012 22:56:11 +0800
Jiang Liu wrote:
> On 11/20/2012 07:38 AM, Andrew Morton wrote:
> > On Mon, 19 Nov 2012 00:07:26 +0800
> > Jiang Liu wrote:
>
> ...
>
> > Also, the existing comment tells us that spanned_pages and
> > present_pages are protected by span_seqlock but has not been
On 11/20/2012 07:38 AM, Andrew Morton wrote:
> On Mon, 19 Nov 2012 00:07:26 +0800
> Jiang Liu wrote:
>
>> Currently a zone's present_pages is calcuated as below, which is
>> inaccurate and may cause trouble to memory hotplug.
>> spanned_pages - absent_pages - memmap_pages - dma_reserve.
>>
>
On Mon, 19 Nov 2012 00:07:26 +0800
Jiang Liu wrote:
> Currently a zone's present_pages is calcuated as below, which is
> inaccurate and may cause trouble to memory hotplug.
> spanned_pages - absent_pages - memmap_pages - dma_reserve.
>
> During fixing bugs caused by inaccurate zone->presen
Currently a zone's present_pages is calcuated as below, which is
inaccurate and may cause trouble to memory hotplug.
spanned_pages - absent_pages - memmap_pages - dma_reserve.
During fixing bugs caused by inaccurate zone->present_pages, we found
zone->present_pages has been abused. The fie
6 matches
Mail list logo
