From: David Hildenbrand <david@redhat.com>
To: Michal Hocko <mhocko@suse.com>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
linux-hyperv@vger.kernel.org, xen-devel@lists.xenproject.org,
linux-acpi@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Vlastimil Babka <vbabka@suse.cz>,
Oscar Salvador <osalvador@suse.de>,
Alexander Duyck <alexander.h.duyck@linux.intel.com>,
Mel Gorman <mgorman@techsingularity.net>,
Dave Hansen <dave.hansen@intel.com>,
Wei Yang <richard.weiyang@linux.alibaba.com>,
Mike Rapoport <rppt@kernel.org>,
"K. Y. Srinivasan" <kys@microsoft.com>,
Haiyang Zhang <haiyangz@microsoft.com>,
Stephen Hemminger <sthemmin@microsoft.com>,
Wei Liu <wei.liu@kernel.org>
Subject: Re: [PATCH v1 4/5] mm/page_alloc: place pages to tail in __free_pages_core()
Date: Fri, 2 Oct 2020 17:10:03 +0200 [thread overview]
Message-ID: <7bf5d426-1fdd-4525-25da-a68d92b6c11d@redhat.com> (raw)
In-Reply-To: <20201002134115.GJ4555@dhcp22.suse.cz>
On 02.10.20 15:41, Michal Hocko wrote:
> On Mon 28-09-20 20:21:09, David Hildenbrand wrote:
>> __free_pages_core() is used when exposing fresh memory to the buddy
>> during system boot and when onlining memory in generic_online_page().
>>
>> generic_online_page() is used in two cases:
>>
>> 1. Direct memory onlining in online_pages().
>> 2. Deferred memory onlining in memory-ballooning-like mechanisms (HyperV
>> balloon and virtio-mem), when parts of a section are kept
>> fake-offline to be fake-onlined later on.
>>
>> In 1, we already place pages to the tail of the freelist. Pages will be
>> freed to MIGRATE_ISOLATE lists first and moved to the tail of the freelists
>> via undo_isolate_page_range().
>>
>> In 2, we currently don't implement a proper rule. In case of virtio-mem,
>> where we currently always online MAX_ORDER - 1 pages, the pages will be
>> placed to the HEAD of the freelist - undesireable. While the hyper-v
>> balloon calls generic_online_page() with single pages, usually it will
>> call it on successive single pages in a larger block.
>>
>> The pages are fresh, so place them to the tail of the freelists and avoid
>> the PCP. In __free_pages_core(), remove the now superflouos call to
>> set_page_refcounted() and add a comment regarding page initialization and
>> the refcount.
>>
>> Note: In 2. we currently don't shuffle. If ever relevant (page shuffling
>> is usually of limited use in virtualized environments), we might want to
>> shuffle after a sequence of generic_online_page() calls in the
>> relevant callers.
>
> It took some time to get through all the freeing paths with subtle
> differences but this looks reasonable. You are mentioning that this
> influences a boot time free memory ordering as well but only very
> briefly. I do not expect this to make a huge difference but who knows.
> It makes some sense to add pages in the order they show up in the
> physical address ordering.
I think boot memory is mostly exposed in the physical address ordering.
In that case, higher addresses will now be used less likely immediately
after this patch. I also don't think it's an issue - if we still detect
it's an issue it's fairly easy to change again.
Thanks!
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2020-10-02 15:10 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-28 18:21 [PATCH v1 0/5] mm: place pages to the freelist tail when onling and undoing isolation David Hildenbrand
2020-09-28 18:21 ` [PATCH v1 1/5] mm/page_alloc: convert "report" flag of __free_one_page() to a proper flag David Hildenbrand
2020-09-28 20:11 ` Pankaj Gupta
2020-09-29 8:58 ` Wei Yang
2020-10-02 13:17 ` Michal Hocko
2020-10-02 13:41 ` Matthew Wilcox
2020-10-02 14:48 ` David Hildenbrand
2020-10-02 14:57 ` David Hildenbrand
2020-09-28 18:21 ` [PATCH v1 2/5] mm/page_alloc: place pages to tail in __putback_isolated_page() David Hildenbrand
2020-09-28 20:38 ` Pankaj Gupta
2020-09-29 9:10 ` Wei Yang
2020-10-02 13:19 ` Michal Hocko
2020-09-28 18:21 ` [PATCH v1 3/5] mm/page_alloc: always move pages to the tail of the freelist in unset_migratetype_isolate() David Hildenbrand
2020-09-28 20:55 ` Pankaj Gupta
2020-09-29 9:18 ` Wei Yang
2020-09-29 10:12 ` David Hildenbrand
2020-09-30 7:48 ` Wei Yang
2020-10-02 13:24 ` Michal Hocko
2020-10-02 15:20 ` David Hildenbrand
2020-10-05 6:56 ` Michal Hocko
2020-10-05 8:20 ` Mel Gorman
2020-10-05 9:11 ` David Hildenbrand
2020-09-28 18:21 ` [PATCH v1 4/5] mm/page_alloc: place pages to tail in __free_pages_core() David Hildenbrand
2020-09-28 20:33 ` Pankaj Gupta
2020-09-29 9:36 ` Wei Yang
2020-09-29 10:14 ` David Hildenbrand
2020-10-02 13:41 ` Michal Hocko
2020-10-02 15:10 ` David Hildenbrand [this message]
2020-09-28 18:21 ` [PATCH v1 5/5] mm/memory_hotplug: update comment regarding zone shuffling David Hildenbrand
2020-09-29 9:40 ` Wei Yang
2020-10-02 13:41 ` Michal Hocko
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=7bf5d426-1fdd-4525-25da-a68d92b6c11d@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alexander.h.duyck@linux.intel.com \
--cc=dave.hansen@intel.com \
--cc=haiyangz@microsoft.com \
--cc=kys@microsoft.com \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-hyperv@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@suse.com \
--cc=osalvador@suse.de \
--cc=richard.weiyang@linux.alibaba.com \
--cc=rppt@kernel.org \
--cc=sthemmin@microsoft.com \
--cc=vbabka@suse.cz \
--cc=wei.liu@kernel.org \
--cc=xen-devel@lists.xenproject.org \
/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).