From: Yinghai Lu <yinghai@kernel.org>
To: "Luck, Tony" <tony.luck@intel.com>
Cc: Tejun Heo <tj@kernel.org>, Tang Chen <tangchen@cn.fujitsu.com>,
Tang Chen <imtangchen@gmail.com>,
"H. Peter Anvin" <hpa@zytor.com>,
"Moore, Robert" <robert.moore@intel.com>,
"Zheng, Lv" <lv.zheng@intel.com>,
"Rafael J. Wysocki" <rjw@sisk.pl>, Len Brown <lenb@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>,
Andrew Morton <akpm@linux-foundation.org>,
Thomas Renninger <trenn@suse.de>,
Jiang Liu <jiang.liu@huawei.com>,
Wen Congyang <wency@cn.fujitsu.com>,
Lai Jiangshan <laijs@cn.fujitsu.com>,
Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>,
Taku Izumi <izumi.taku@jp.fujitsu.com>,
Mel Gorman <mgorman@suse.de>, Minchan Kim <minchan@kernel.org>,
"mina86@mina86.com" <mina86@mina86.com>,
"gong.chen@linux.intel.com" <gong.chen@linux.intel.com>,
Vasilis Liaskovitis <vasilis.liaskovitis@profitbricks.com>,
"lwoodman@redhat.com" <lwoodman@red>
Subject: Re: [PATCH part5 0/7] Arrange hotpluggable memory as ZONE_MOVABLE.
Date: Thu, 15 Aug 2013 12:34:31 -0700 [thread overview]
Message-ID: <CAE9FiQVAsHLRCcrtFLqgifAJ6jfbDK7y_OFGUDArSWPKk+2qnQ@mail.gmail.com> (raw)
In-Reply-To: <3908561D78D1C84285E8C5FCA982C28F31CBD145@ORSMSX106.amr.corp.intel.com>
On Thu, Aug 15, 2013 at 12:08 PM, Luck, Tony <tony.luck@intel.com> wrote:
>> That is what my patchset want to do.
>> put page tables on the same node like node data.
>> with that, hotplug and normal case will be the same code path.
>
> Page tables are a big issue if we have 4K mappings (8 byte entry per
> 4K page means 2MB of page tables per GB of memory) ... but only
> used for DEBUG cases, right?
yes.
>
> If we use 2M mappings, then allocations are 512x smaller - so only
> 4K per GB - hard to justify spreading that across nodes.
>
> If we can use 1GB mappings - then another 512x reduction to 8 bytes per GB (or 8KB per TB)
Yes. 4k for 512G.
Just make all cases use same code path even for DEBUG_PAGEALLOC with 4k page
mapping.
>
> Aren't page structures a bigger issue? ~64 bytes per 4K page. Do we
> make sure these get allocated from the NUMA node that they describe?
> This should not hurt the ZONE_MOVEABLE-ness of this - although they are
> kernel structures they can be freed when the node is removed (at that point
> they describe memory that is no longer present). From a scalability perspective
> we don't want to run node0 low on memory by using it for every other node.
> From a NUMA perspective we want the page_t that describes a page to be
> in the same locality as the page itself.
yes, that is vmemmap, and it is already numa aware.
Thanks
Yinghai
--
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>
next prev parent reply other threads:[~2013-08-15 19:34 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-08 10:16 [PATCH part5 0/7] Arrange hotpluggable memory as ZONE_MOVABLE Tang Chen
2013-08-08 10:16 ` [PATCH part5 1/7] x86: get pg_data_t's memory from other node Tang Chen
2013-08-12 14:39 ` Tejun Heo
2013-08-12 15:12 ` Tang Chen
2013-08-08 10:16 ` [PATCH part5 2/7] x86, numa, mem_hotplug: Skip all the regions the kernel resides in Tang Chen
2013-08-08 10:16 ` [PATCH part5 3/7] memblock, numa: Introduce flag into memblock Tang Chen
2013-08-08 10:16 ` [PATCH part5 4/7] memblock, mem_hotplug: Introduce MEMBLOCK_HOTPLUG flag to mark hotpluggable regions Tang Chen
2013-08-08 10:16 ` [PATCH part5 5/7] memblock, mem_hotplug: Make memblock skip hotpluggable regions by default Tang Chen
2013-08-14 21:54 ` Naoya Horiguchi
2013-08-15 5:15 ` Tang Chen
2013-08-08 10:16 ` [PATCH part5 6/7] mem-hotplug: Introduce movablenode boot option to {en|dis}able using SRAT Tang Chen
2013-08-08 10:16 ` [PATCH part5 7/7] x86, numa, acpi, memory-hotplug: Make movablenode have higher priority Tang Chen
2013-08-09 16:32 ` [PATCH part5 0/7] Arrange hotpluggable memory as ZONE_MOVABLE Tejun Heo
2013-08-12 6:33 ` Tang Chen
2013-08-12 8:54 ` Tang Chen
2013-08-12 14:50 ` Tejun Heo
2013-08-12 15:14 ` H. Peter Anvin
2013-08-12 15:23 ` Tejun Heo
2013-08-12 16:29 ` Tang Chen
2013-08-12 16:46 ` Tejun Heo
2013-08-12 18:23 ` Tang Chen
2013-08-12 20:20 ` Tejun Heo
2013-08-12 20:49 ` Luck, Tony
2013-08-12 20:54 ` Tejun Heo
2013-08-12 20:57 ` H. Peter Anvin
2013-08-12 21:06 ` Yinghai Lu
2013-08-12 21:08 ` Tejun Heo
2013-08-12 21:12 ` H. Peter Anvin
2013-08-12 21:14 ` Tejun Heo
2013-08-12 21:11 ` H. Peter Anvin
2013-08-12 21:11 ` Luck, Tony
2013-08-12 21:25 ` Yinghai Lu
2013-08-12 21:28 ` H. Peter Anvin
2013-08-13 5:14 ` H. Peter Anvin
2013-08-13 6:14 ` Tang Chen
2013-08-13 9:56 ` Tang Chen
2013-08-13 14:38 ` Tejun Heo
2013-08-13 22:33 ` Yinghai Lu
2013-08-14 1:22 ` Tang Chen
2013-08-15 19:06 ` Toshi Kani
2013-08-15 20:28 ` Yinghai Lu
2013-08-16 2:08 ` Tang Chen
2013-08-16 4:21 ` Yinghai Lu
2013-08-19 3:07 ` Tang Chen
2013-08-19 3:28 ` Yinghai Lu
2013-08-15 8:42 ` Tang Chen
2013-08-15 12:19 ` Tejun Heo
2013-08-15 12:44 ` Tang Chen
2013-08-15 12:49 ` Tejun Heo
2013-08-15 12:52 ` Tang Chen
2013-08-15 14:37 ` Yinghai Lu
2013-08-15 14:45 ` Tejun Heo
2013-08-15 15:05 ` Yinghai Lu
2013-08-15 15:10 ` Tejun Heo
2013-08-15 19:49 ` Toshi Kani
2013-08-15 19:08 ` Luck, Tony
2013-08-15 19:34 ` Yinghai Lu [this message]
2013-08-15 14:35 ` Yinghai Lu
2013-08-16 1:16 ` Tang Chen
2013-08-12 15:41 ` Tang Chen
2013-08-12 15:46 ` Tejun Heo
2013-08-12 16:19 ` Tang Chen
2013-08-12 16:22 ` Tejun Heo
2013-08-12 17:01 ` Tang Chen
2013-08-12 17:23 ` H. Peter Anvin
2013-08-14 18:22 ` KOSAKI Motohiro
2013-08-12 18:07 ` Tejun Heo
2013-08-14 18:15 ` KOSAKI Motohiro
2013-08-14 18:23 ` Tejun Heo
2013-08-14 19:40 ` KOSAKI Motohiro
2013-08-14 19:55 ` Tejun Heo
2013-08-14 20:29 ` KOSAKI Motohiro
2013-08-14 20:30 ` H. Peter Anvin
2013-08-14 20:35 ` Tejun Heo
2013-08-14 21:17 ` KOSAKI Motohiro
2013-08-14 21:36 ` Tejun Heo
2013-08-15 1:08 ` KOSAKI Motohiro
2013-08-15 1:21 ` Tejun Heo
2013-08-15 1:33 ` Tejun Heo
2013-08-15 1:44 ` KOSAKI Motohiro
2013-08-15 2:22 ` Tejun Heo
2013-08-15 1:38 ` KOSAKI Motohiro
2013-08-15 1:51 ` Tejun Heo
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=CAE9FiQVAsHLRCcrtFLqgifAJ6jfbDK7y_OFGUDArSWPKk+2qnQ@mail.gmail.com \
--to=yinghai@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=gong.chen@linux.intel.com \
--cc=hpa@zytor.com \
--cc=imtangchen@gmail.com \
--cc=isimatu.yasuaki@jp.fujitsu.com \
--cc=izumi.taku@jp.fujitsu.com \
--cc=jiang.liu@huawei.com \
--cc=laijs@cn.fujitsu.com \
--cc=lenb@kernel.org \
--cc=lv.zheng@intel.com \
--cc=lwoodman@red \
--cc=mgorman@suse.de \
--cc=mina86@mina86.com \
--cc=minchan@kernel.org \
--cc=mingo@elte.hu \
--cc=rjw@sisk.pl \
--cc=robert.moore@intel.com \
--cc=tangchen@cn.fujitsu.com \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
--cc=tony.luck@intel.com \
--cc=trenn@suse.de \
--cc=vasilis.liaskovitis@profitbricks.com \
--cc=wency@cn.fujitsu.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).