All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Aaron Lu <aaron.lu@intel.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	Huang Ying <ying.huang@intel.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Kemi Wang <kemi.wang@intel.com>,
	Tim Chen <tim.c.chen@linux.intel.com>,
	Andi Kleen <ak@linux.intel.com>, Michal Hocko <mhocko@suse.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	Mel Gorman <mgorman@techsingularity.net>,
	Matthew Wilcox <willy@infradead.org>,
	David Rientjes <rientjes@google.com>
Subject: Re: [PATCH v4 3/3 update] mm/free_pcppages_bulk: prefetch buddy while not holding lock
Date: Fri, 9 Mar 2018 13:58:32 -0800	[thread overview]
Message-ID: <20180309135832.988ab6d3d986658d531a79ef@linux-foundation.org> (raw)
In-Reply-To: <20180309082431.GB30868@intel.com>

>
> When a page is freed back to the global pool, its buddy will be checked
> to see if it's possible to do a merge. This requires accessing buddy's
> page structure and that access could take a long time if it's cache cold.
> 
> This patch adds a prefetch to the to-be-freed page's buddy outside of
> zone->lock in hope of accessing buddy's page structure later under
> zone->lock will be faster. Since we *always* do buddy merging and check
> an order-0 page's buddy to try to merge it when it goes into the main
> allocator, the cacheline will always come in, i.e. the prefetched data
> will never be unused.
> 
> Normally, the number of to-be-freed pages(i.e. count) equals to
> pcp->batch (default=31 and has an upper limit of (PAGE_SHIFT * 8)=96 on
> x86_64) but in the case of pcp's pages getting all drained, it will be
> pcp->count which has an upper limit of pcp->high. pcp->high, although
> has a default value of 186 (pcp->batch=31 * 6), can be changed by user
> through /proc/sys/vm/percpu_pagelist_fraction and there is no software
> upper limit so could be large, like several thousand. For this reason,
> only the last pcp->batch number of page's buddy structure is prefetched
> to avoid excessive prefetching. pcp-batch is used because:
> 1 most often, count == pcp->batch;
> 2 it has an upper limit itself so we won't prefetch excessively.
> 
> Considering the possible large value of pcp->high, it also makes
> sense to free the last added page first for cache hot's reason.
> That's where the change of list_add_tail() to list_add() comes in
> as we will free them from head to tail one by one.
> 
> In the meantime, there are two concerns:
> 1 the prefetch could potentially evict existing cachelines, especially
>   for L1D cache since it is not huge;
> 2 there is some additional instruction overhead, namely calculating
>   buddy pfn twice.
> 
> For 1, it's hard to say, this microbenchmark though shows good result but
> the actual benefit of this patch will be workload/CPU dependant;
> For 2, since the calculation is a XOR on two local variables, it's expected
> in many cases that cycles spent will be offset by reduced memory latency
> later. This is especially true for NUMA machines where multiple CPUs are
> contending on zone->lock and the most time consuming part under zone->lock
> is the wait of 'struct page' cacheline of the to-be-freed pages and their
> buddies.
> 
> Test with will-it-scale/page_fault1 full load:
> 
> kernel      Broadwell(2S)  Skylake(2S)   Broadwell(4S)  Skylake(4S)
> v4.16-rc2+  9034215        7971818       13667135       15677465
> patch2/3    9536374 +5.6%  8314710 +4.3% 14070408 +3.0% 16675866 +6.4%
> this patch 10180856 +6.8%  8506369 +2.3% 14756865 +4.9% 17325324 +3.9%
> Note: this patch's performance improvement percent is against patch2/3.
> 
> (Changelog stolen from Dave Hansen and Mel Gorman's comments at
> http://lkml.kernel.org/r/148a42d8-8306-2f2f-7f7c-86bc118f8ccd@intel.com)
> 
> Link: http://lkml.kernel.org/r/20180301062845.26038-4-aaron.lu@intel.com
>
> ...
>
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1141,6 +1141,9 @@ static void free_pcppages_bulk(struct zone *zone, int count,
>  			batch_free = count;
>  
>  		do {
> +			unsigned long pfn, buddy_pfn;
> +			struct page *buddy;
> +
>  			page = list_last_entry(list, struct page, lru);
>  			/* must delete to avoid corrupting pcp list */
>  			list_del(&page->lru);
> @@ -1149,7 +1152,23 @@ static void free_pcppages_bulk(struct zone *zone, int count,
>  			if (bulkfree_pcp_prepare(page))
>  				continue;
>  
> -			list_add_tail(&page->lru, &head);
> +			list_add(&page->lru, &head);

The result here will be that free_pcppages_bulk() frees the pages in
the reverse order?

I don't immediately see a downside to that.  In the (distant) past we
had issues when successive alloc_page() calls would return pages in
descending address order - that totally screwed up scatter-gather page
merging.  But this is the page-freeing path.  Still, something to be
thought about and monitored.

> +
> +			/*
> +			 * We are going to put the page back to the global
> +			 * pool, prefetch its buddy to speed up later access
> +			 * under zone->lock. It is believed the overhead of
> +			 * an additional test and calculating buddy_pfn here
> +			 * can be offset by reduced memory latency later. To
> +			 * avoid excessive prefetching due to large count, only
> +			 * prefetch buddy for the last pcp->batch nr of pages.
> +			 */
> +			if (count > pcp->batch)
> +				continue;
> +			pfn = page_to_pfn(page);
> +			buddy_pfn = __find_buddy_pfn(pfn, 0);
> +			buddy = page + (buddy_pfn - pfn);
> +			prefetch(buddy);
>  		} while (--count && --batch_free && !list_empty(list));

This loop hurts my brain, mainly the handling of `count':

	while (count) {
		do {
			batch_free++;
		} while (list_empty(list));

		/* This is the only non-empty list. Free them all. */
		if (batch_free == MIGRATE_PCPTYPES)
			batch_free = count;

		do {
		} while (--count && --batch_free && !list_empty(list));
	}

I guess it kinda makes sense - both loops terminate on count==0.  But
still.  Can it be clarified?

  reply	other threads:[~2018-03-09 21:58 UTC|newest]

Thread overview: 67+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-01  6:28 [PATCH v4 0/3] mm: improve zone->lock scalability Aaron Lu
2018-03-01  6:28 ` Aaron Lu
2018-03-01  6:28 ` [PATCH v4 1/3] mm/free_pcppages_bulk: update pcp->count inside Aaron Lu
2018-03-01  6:28   ` Aaron Lu
2018-03-01 12:11   ` David Rientjes
2018-03-01 12:11     ` David Rientjes
2018-03-01 13:45   ` Michal Hocko
2018-03-01 13:45     ` Michal Hocko
2018-03-12 13:22   ` Vlastimil Babka
2018-03-13  2:11     ` Aaron Lu
2018-03-01  6:28 ` [PATCH v4 2/3] mm/free_pcppages_bulk: do not hold lock when picking pages to free Aaron Lu
2018-03-01  6:28   ` Aaron Lu
2018-03-01 13:55   ` Michal Hocko
2018-03-01 13:55     ` Michal Hocko
2018-03-02  7:15     ` Aaron Lu
2018-03-02  7:15       ` Aaron Lu
2018-03-02 15:34       ` Dave Hansen
2018-03-02 15:34         ` Dave Hansen
2018-03-02  7:31     ` Huang, Ying
2018-03-02  7:31       ` Huang, Ying
2018-03-02  0:01   ` Andrew Morton
2018-03-02  0:01     ` Andrew Morton
2018-03-02  8:01     ` Aaron Lu
2018-03-02  8:01       ` Aaron Lu
2018-03-02 21:23       ` Andrew Morton
2018-03-02 21:23         ` Andrew Morton
2018-03-02 21:25         ` Dave Hansen
2018-03-02 21:25           ` Dave Hansen
2018-03-12 14:22   ` Vlastimil Babka
2018-03-13  3:34     ` Aaron Lu
2018-03-22 15:17       ` Matthew Wilcox
2018-03-26  3:03         ` Aaron Lu
2018-03-01  6:28 ` [PATCH v4 3/3] mm/free_pcppages_bulk: prefetch buddy while not holding lock Aaron Lu
2018-03-01  6:28   ` Aaron Lu
2018-03-01 14:00   ` Michal Hocko
2018-03-01 14:00     ` Michal Hocko
2018-03-02  8:31     ` Aaron Lu
2018-03-02  8:31       ` Aaron Lu
2018-03-02 17:55     ` Vlastimil Babka
2018-03-02 17:55       ` Vlastimil Babka
2018-03-02 18:00       ` Dave Hansen
2018-03-02 18:00         ` Dave Hansen
2018-03-02 18:08         ` Vlastimil Babka
2018-03-02 18:08           ` Vlastimil Babka
2018-03-05 11:41       ` Aaron Lu
2018-03-05 11:41         ` Aaron Lu
2018-03-05 11:48         ` Aaron Lu
2018-03-05 11:48           ` Aaron Lu
2018-03-06  7:55         ` Vlastimil Babka
2018-03-06  7:55           ` Vlastimil Babka
2018-03-06 12:27           ` Aaron Lu
2018-03-06 12:27             ` Aaron Lu
2018-03-06 12:53             ` Matthew Wilcox
2018-03-06 12:53               ` Matthew Wilcox
2018-03-02  0:09   ` Andrew Morton
2018-03-02  0:09     ` Andrew Morton
2018-03-02  8:27     ` Aaron Lu
2018-03-02  8:27       ` Aaron Lu
2018-03-09  8:24       ` [PATCH v4 3/3 update] " Aaron Lu
2018-03-09 21:58         ` Andrew Morton [this message]
2018-03-10 14:46           ` Aaron Lu
2018-03-12 15:05           ` Vlastimil Babka
2018-03-12 17:32         ` Dave Hansen
2018-03-13  3:35           ` Aaron Lu
2018-03-13  7:04             ` Aaron Lu
2018-03-20  9:50               ` Vlastimil Babka
2018-03-20 11:31                 ` [PATCH v4 3/3 update2] " Aaron Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180309135832.988ab6d3d986658d531a79ef@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=aaron.lu@intel.com \
    --cc=ak@linux.intel.com \
    --cc=dave.hansen@intel.com \
    --cc=kemi.wang@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@techsingularity.net \
    --cc=mhocko@suse.com \
    --cc=rientjes@google.com \
    --cc=tim.c.chen@linux.intel.com \
    --cc=vbabka@suse.cz \
    --cc=willy@infradead.org \
    --cc=ying.huang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.