提交 ba56e91c 编写于 作者: S Seth, Rohit 提交者: Linus Torvalds

[PATCH] mm: page_alloc: increase size of per-cpu-pages

Increase the page allocator's per-cpu magazines from 1/4MB to 1/2MB.

Over 100+ runs for a workload, the difference in mean is about 2%.  The best
results for both are almost same.  Though the max variation in results with
1/2MB is only 2.2%, whereas with 1/4MB it is 12%.
Signed-off-by: NRohit Seth <rohit.seth@intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
上级 fcdae29a
...@@ -1721,29 +1721,29 @@ static int __devinit zone_batchsize(struct zone *zone) ...@@ -1721,29 +1721,29 @@ static int __devinit zone_batchsize(struct zone *zone)
/* /*
* The per-cpu-pages pools are set to around 1000th of the * The per-cpu-pages pools are set to around 1000th of the
* size of the zone. But no more than 1/4 of a meg - there's * size of the zone. But no more than 1/2 of a meg.
* no point in going beyond the size of L2 cache.
* *
* OK, so we don't know how big the cache is. So guess. * OK, so we don't know how big the cache is. So guess.
*/ */
batch = zone->present_pages / 1024; batch = zone->present_pages / 1024;
if (batch * PAGE_SIZE > 256 * 1024) if (batch * PAGE_SIZE > 512 * 1024)
batch = (256 * 1024) / PAGE_SIZE; batch = (512 * 1024) / PAGE_SIZE;
batch /= 4; /* We effectively *= 4 below */ batch /= 4; /* We effectively *= 4 below */
if (batch < 1) if (batch < 1)
batch = 1; batch = 1;
/* /*
* Clamp the batch to a 2^n - 1 value. Having a power * We will be trying to allcoate bigger chunks of contiguous
* of 2 value was found to be more likely to have * memory of the order of fls(batch). This should result in
* suboptimal cache aliasing properties in some cases. * better cache coloring.
* *
* For example if 2 tasks are alternately allocating * A sanity check also to ensure that batch is still in limits.
* batches of pages, one task can end up with a lot
* of pages of one half of the possible page colors
* and the other with pages of the other colors.
*/ */
batch = (1 << fls(batch + batch/2)) - 1; batch = (1 << fls(batch + batch/2));
if (fls(batch) >= (PAGE_SHIFT + MAX_ORDER - 2))
batch = PAGE_SHIFT + ((MAX_ORDER - 1 - PAGE_SHIFT)/2);
return batch; return batch;
} }
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册