lkml.org 
[lkml]   [2015]   [May]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 1/3] mm/page_alloc: don't break highest order freepage if steal
On 05/12/2015 09:51 AM, Vlastimil Babka wrote:
>> {
>> struct page *page;
>> + bool steal_fallback;
>>
>> -retry_reserve:
>> +retry:
>> page = __rmqueue_smallest(zone, order, migratetype);
>>
>> if (unlikely(!page) && migratetype != MIGRATE_RESERVE) {
>> if (migratetype == MIGRATE_MOVABLE)
>> page = __rmqueue_cma_fallback(zone, order);
>>
>> - if (!page)
>> - page = __rmqueue_fallback(zone, order, migratetype);
>> + if (page)
>> + goto out;
>> +
>> + steal_fallback = __rmqueue_fallback(zone, order, migratetype);

Oh and the variable can be probably replaced by calling
__rmqueue_fallback directly in the if() below.

>>
>> /*
>> * Use MIGRATE_RESERVE rather than fail an allocation. goto
>> * is used because __rmqueue_smallest is an inline function
>> * and we want just one call site
>> */
>> - if (!page) {
>> + if (!steal_fallback)
>> migratetype = MIGRATE_RESERVE;
>> - goto retry_reserve;
>> - }
>> +
>> + goto retry;
>> }
>>
>> +out:
>> trace_mm_page_alloc_zone_locked(page, order, migratetype);
>> return page;
>> }
>>
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>



\
 
 \ /
  Last update: 2015-05-12 10:21    [W:0.091 / U:25.436 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site