From: Aaron Lu <aaron.lu@intel.com>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Huang Ying <ying.huang@intel.com>,
Dave Hansen <dave.hansen@linux.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>,
Matthew Wilcox <willy@infradead.org>,
Daniel Jordan <daniel.m.jordan@oracle.com>,
Tariq Toukan <tariqt@mellanox.com>,
Jesper Dangaard Brouer <brouer@redhat.com>
Subject: Re: [RFC v4 PATCH 3/5] mm/rmqueue_bulk: alloc without touching individual page structure
Date: Wed, 17 Oct 2018 22:23:27 +0800 [thread overview]
Message-ID: <20181017142327.GB9167@intel.com> (raw)
In-Reply-To: <20181017112042.GK5819@techsingularity.net>
On Wed, Oct 17, 2018 at 12:20:42PM +0100, Mel Gorman wrote:
> On Wed, Oct 17, 2018 at 02:33:28PM +0800, Aaron Lu wrote:
> > Profile on Intel Skylake server shows the most time consuming part
> > under zone->lock on allocation path is accessing those to-be-returned
> > page's "struct page" on the free_list inside zone->lock. One explanation
> > is, different CPUs are releasing pages to the head of free_list and
> > those page's 'struct page' may very well be cache cold for the allocating
> > CPU when it grabs these pages from free_list' head. The purpose here
> > is to avoid touching these pages one by one inside zone->lock.
> >
>
> I didn't read this one in depth because it's somewhat ortogonal to the
> lazy buddy merging which I think would benefit from being finalised and
> ensuring that there are no reductions in high-order allocation success
> rates. Pages being allocated on one CPU and freed on another is not that
> unusual -- ping-pong workloads or things like netperf used to exhibit
> this sort of pattern.
>
> However, this part stuck out
>
> > +static inline void zone_wait_cluster_alloc(struct zone *zone)
> > +{
> > + while (atomic_read(&zone->cluster.in_progress))
> > + cpu_relax();
> > +}
> > +
>
> RT has had problems with cpu_relax in the past but more importantly, as
> this delay for parallel compactions and allocations of contig ranges,
> we could be stuck here for very long periods of time with interrupts
The longest possible time is one CPU accessing pcp->batch number cold
cachelines. Reason:
When zone_wait_cluster_alloc() is called, we already held zone lock so
no more allocations are possible. Waiting in_progress to become zero
means waiting any CPU that increased in_progress to finish processing
their allocated pages. Since they will at most allocate pcp->batch pages
and worse case are all these page structres are cache cold, so the
longest wait time is one CPU accessing pcp->batch number cold cache lines.
I have no idea if this time is too long though.
> disabled. It gets even worse if it's from an interrupt context such as
> jumbo frame allocation or a high-order slab allocation that is atomic.
My understanding is atomic allocation won't trigger compaction, no?
> These potentially large periods of time with interrupts disabled is very
> hazardous.
I see and agree, thanks for pointing this out.
Hopefully, the above mentioned worst case time won't be regarded as
unbound or too long.
> It may be necessary to consider instead minimising the number
> of struct page update when merging to PCP and then either increasing the
> size of the PCP or allowing it to exceed pcp->high for short periods of
> time to batch the struct page updates.
I don't quite follow this part. It doesn't seem possible we can exceed
pcp->high in allocation path, or are you talking about free path?
And thanks a lot for the review!
next prev parent reply other threads:[~2018-10-17 14:23 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-17 6:33 [RFC v4 PATCH 0/5] Eliminate zone->lock contention for will-it-scale/page_fault1 and parallel free Aaron Lu
2018-10-17 6:33 ` [RFC v4 PATCH 1/5] mm/page_alloc: use helper functions to add/remove a page to/from buddy Aaron Lu
2018-10-17 9:51 ` Mel Gorman
2018-10-17 6:33 ` [RFC v4 PATCH 2/5] mm/__free_one_page: skip merge for order-0 page unless compaction failed Aaron Lu
2018-10-17 10:44 ` Mel Gorman
2018-10-17 13:10 ` Aaron Lu
2018-10-17 13:58 ` Mel Gorman
2018-10-17 14:59 ` Aaron Lu
2018-10-18 11:16 ` Mel Gorman
2018-10-19 5:57 ` Aaron Lu
2018-10-19 8:54 ` Mel Gorman
2018-10-19 15:00 ` Daniel Jordan
2018-10-20 9:00 ` Aaron Lu
2018-10-17 17:03 ` Vlastimil Babka
2018-10-18 6:48 ` Aaron Lu
2018-10-18 8:23 ` Vlastimil Babka
2018-10-18 11:07 ` Aaron Lu
2018-10-17 6:33 ` [RFC v4 PATCH 3/5] mm/rmqueue_bulk: alloc without touching individual page structure Aaron Lu
2018-10-17 11:20 ` Mel Gorman
2018-10-17 14:23 ` Aaron Lu [this message]
2018-10-18 11:20 ` Mel Gorman
2018-10-18 13:21 ` Aaron Lu
2018-10-22 9:37 ` Vlastimil Babka
2018-10-23 2:19 ` Aaron Lu
2018-10-17 6:33 ` [RFC v4 PATCH 4/5] mm/free_pcppages_bulk: reduce overhead of cluster operation on free path Aaron Lu
2018-10-17 6:33 ` [RFC v4 PATCH 5/5] mm/can_skip_merge(): make it more aggressive to attempt cluster alloc/free 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=20181017142327.GB9167@intel.com \
--to=aaron.lu@intel.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=brouer@redhat.com \
--cc=daniel.m.jordan@oracle.com \
--cc=dave.hansen@linux.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=tariqt@mellanox.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).