From: Dave Hansen <dave@linux.vnet.ibm.com>
To: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Christoph Lameter <cl@linux.com>,
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
David Rientjes <rientjes@google.com>,
Pekka Enberg <penberg@kernel.org>, Michal Hocko <mhocko@suse.cz>,
Andrew Morton <akpm@linux-foundation.org>,
Hugh Dickins <hughd@google.com>,
linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>,
linux-parisc@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
x86 maintainers <x86@kernel.org>, Tejun Heo <tj@kernel.org>,
Mel Gorman <mel@csn.ul.ie>, Andy Whitcroft <apw@shadowen.org>
Subject: Re: [PATCH] convert parisc to sparsemem (was Re: [PATCH v3] mm: make expand_downwards symmetrical to expand_upwards)
Date: Wed, 27 Apr 2011 09:36:29 -0700 [thread overview]
Message-ID: <1303922189.9516.33.camel@nimitz> (raw)
In-Reply-To: <1303583657.4116.11.camel@mulgrave.site>
On Sat, 2011-04-23 at 13:34 -0500, James Bottomley wrote:
> This is the preliminary conversion. It's very nasty on parisc because
> the memory allocation isn't symmetric anymore: under DISCONTIGMEM, we
> push all memory into bootmem and then let free_all_bootmem() do the
> magic for us;
Urg, that's unfortunate. I bet we could fairly easily teach the bootmem
allocator to allow a couple of bootmem_data's to hang off of an
individual pgdat. Put each pmem_ranges in one of those instead of a
pgdat. That would at least help with the bitmap size explosion and
extra loops.
> now we have to do separate initialisations for ranges
> because SPARSEMEM can't do multi-range boot memory. It's also got the
> horrible hack that I only use the first found range for bootmem. I'm
> not sure if this is correct (it won't be if the first found range can be
> under about 50MB because we'll run out of bootmem during boot) ... we
> might have to sort the ranges and use the larges, but that will involve
> us in even more hackery around the bootmem reservations code.
>
> The boot sequence got a few seconds slower because now all of the loops
> over our pfn ranges actually have to skip through the holes (which takes
> time for 64GB).
Which iterations were these, btw? All of the ones I saw the patch touch
seemed to be running over just a single pmem_range.
> All in all, I've not been very impressed with SPARSEMEM over
> DISCONTIGMEM. It seems to have a lot of rough edges (necessitating
> exception code) which DISCONTIGMEM just copes with.
We definitely need to look at extending it to cover bootmem-time a bit.
Is that even worth it these days with the no-bootmem bits around?
-- Dave
--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2011-04-27 16:37 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-15 13:51 [PATCH] mm: make expand_downwards symmetrical to expand_upwards Michal Hocko
2011-04-18 3:00 ` Hugh Dickins
2011-04-18 10:01 ` [PATCH v2] " Michal Hocko
2011-04-18 20:56 ` Andrew Morton
[not found] ` <20110419091022.GA21689@tiehlicka.suse.cz>
2011-04-19 11:09 ` [PATCH followup] mm: get rid of CONFIG_STACK_GROWSUP || CONFIG_IA64 Michal Hocko
2011-04-20 0:33 ` KOSAKI Motohiro
2011-04-20 6:59 ` Michal Hocko
2011-04-20 7:08 ` KOSAKI Motohiro
2011-04-26 7:59 ` Michal Hocko
2011-04-19 11:10 ` [PATCH v3] mm: make expand_downwards symmetrical to expand_upwards Michal Hocko
2011-04-19 15:46 ` James Bottomley
2011-04-19 16:06 ` [PATCH v3] mm: make expand_downwards symmetrical to John David Anglin
2011-04-19 16:59 ` James Bottomley
2011-04-19 16:07 ` [PATCH v3] mm: make expand_downwards symmetrical to expand_upwards James Bottomley
2011-04-19 17:05 ` Pekka Enberg
2011-04-19 17:11 ` James Bottomley
2011-04-19 17:15 ` Christoph Lameter
2011-04-19 17:48 ` James Bottomley
2011-04-19 18:10 ` Christoph Lameter
2011-04-19 18:20 ` James Bottomley
2011-04-19 18:35 ` Christoph Lameter
2011-04-19 19:49 ` James Bottomley
2011-04-19 20:56 ` Christoph Lameter
2011-04-19 21:21 ` James Bottomley
2011-04-19 21:39 ` Christoph Lameter
2011-04-19 21:48 ` James Bottomley
2011-04-19 21:58 ` Christoph Lameter
2011-04-20 1:23 ` KOSAKI Motohiro
2011-04-20 2:48 ` James Bottomley
2011-04-20 2:57 ` KOSAKI Motohiro
2011-04-20 13:50 ` Christoph Lameter
2011-04-21 13:32 ` Tejun Heo
2011-04-20 5:53 ` Pekka Enberg
2011-04-20 7:15 ` KOSAKI Motohiro
2011-04-20 7:34 ` Pekka Enberg
2011-04-20 8:40 ` KOSAKI Motohiro
2011-04-20 16:32 ` James Bottomley
2011-04-20 16:50 ` Christoph Lameter
2011-04-20 18:09 ` James Bottomley
2011-04-20 21:18 ` David Rientjes
2011-04-20 22:15 ` James Bottomley
2011-04-20 23:12 ` David Rientjes
2011-04-21 13:16 ` KOSAKI Motohiro
2011-04-21 16:37 ` James Bottomley
2011-04-21 18:33 ` Christoph Lameter
2011-04-21 18:45 ` Dave Hansen
2011-04-22 18:19 ` James Bottomley
2011-04-22 20:24 ` Dave Hansen
2011-04-22 20:35 ` James Bottomley
2011-04-22 21:33 ` James Bottomley
2011-04-23 18:34 ` [PATCH] convert parisc to sparsemem (was Re: [PATCH v3] mm: make expand_downwards symmetrical to expand_upwards) James Bottomley
2011-04-24 16:27 ` John David Anglin
2011-04-26 0:32 ` KOSAKI Motohiro
2011-04-27 16:36 ` Dave Hansen [this message]
2011-04-21 20:05 ` [PATCH v3] mm: make expand_downwards symmetrical to expand_upwards James Bottomley
2011-04-21 21:07 ` Christoph Lameter
2011-04-21 21:22 ` James Bottomley
2011-04-21 19:33 ` David Rientjes
2011-04-22 0:34 ` KOSAKI Motohiro
2011-04-21 13:03 ` KOSAKI Motohiro
2011-04-21 19:38 ` David Rientjes
2011-04-21 20:02 ` Christoph Lameter
2011-04-21 21:19 ` David Rientjes
2011-04-21 21:24 ` James Bottomley
2011-04-21 21:34 ` David Rientjes
2011-04-21 21:49 ` James Bottomley
2011-04-21 22:12 ` David Rientjes
2011-04-22 8:02 ` Pekka Enberg
2011-04-22 13:49 ` James Bottomley
2011-04-22 17:00 ` Pekka Enberg
2011-04-22 17:03 ` James Bottomley
2011-04-21 21:41 ` [patch] mm: always set nodes with regular memory in N_NORMAL_MEMORY David Rientjes
2011-04-22 0:36 ` KOSAKI Motohiro
2011-04-20 21:05 ` [PATCH v3] mm: make expand_downwards symmetrical to expand_upwards David Rientjes
2011-04-20 11:20 ` Matthew Wilcox
2011-04-20 11:28 ` Pekka Enberg
2011-04-20 14:15 ` James Bottomley
2011-04-20 14:50 ` Christoph Lameter
2011-04-20 15:02 ` James Bottomley
2011-04-20 15:22 ` Christoph Lameter
2011-04-20 19:25 ` Matthew Wilcox
2011-04-20 21:42 ` David Rientjes
2011-04-21 16:06 ` James Bottomley
2011-04-21 22:19 ` David Rientjes
2011-04-21 22:31 ` James Bottomley
2011-04-20 13:58 ` Christoph Lameter
2011-04-20 21:34 ` David Rientjes
2011-04-20 14:07 ` Christoph Lameter
2011-04-20 2:33 ` James Bottomley
2011-04-19 17:12 ` Christoph Lameter
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=1303922189.9516.33.camel@nimitz \
--to=dave@linux.vnet.ibm.com \
--cc=James.Bottomley@HansenPartnership.com \
--cc=akpm@linux-foundation.org \
--cc=apw@shadowen.org \
--cc=cl@linux.com \
--cc=hughd@google.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-parisc@vger.kernel.org \
--cc=mel@csn.ul.ie \
--cc=mhocko@suse.cz \
--cc=mingo@elte.hu \
--cc=penberg@kernel.org \
--cc=rientjes@google.com \
--cc=tj@kernel.org \
--cc=x86@kernel.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).