linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Lameter <clameter@sgi.com>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: apw@shadowen.org, linux-mm@kvack.org, linux-arch@vger.kernel.org,
	npiggin@suse.de, mel@csn.ul.ie
Subject: Re: [PATCH 3/7] Generic Virtual Memmap support for SPARSEMEM
Date: Fri, 13 Jul 2007 16:17:32 -0700 (PDT)	[thread overview]
Message-ID: <Pine.LNX.4.64.0707131612530.26795@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <20070714081210.1440db40.kamezawa.hiroyu@jp.fujitsu.com>

On Sat, 14 Jul 2007, KAMEZAWA Hiroyuki wrote:

> Just because this patch takes care of boot path. Maybe small problem.

Ahh. I just looked at it. Yes we did not modify the hotplug path. It needs
to call the new vmemmap alloc functions.

> Basically, I welcome this patch. I like this.

Yes you proposed the initial version of this last year. Thanks.

> If we can remove DISCONTIG+VMEMMAP after this is merged, we can say good-bye
> to terrible CONFIG_HOLES_IN_ZONE :)

Right. Horrible stuff. Lots of useless cachelines that have to be 
references in critical paths.

> Note
> >From memory hotplug development/enhancement view, I have following thinking now.
>  
>  1. memmap's section is *not* aligned to "big page size". We have to take care
>     of this at adding support for memory_hotplug/unplug.

You can call the functions for virtual memmap allocation directly. They 
are already generic and will call the page allocator instead of the 
bootmem allocator if the system is already. They will give you the 
properly aligned memory. Perhaps you can just change a few lines 
in sparse_add_one_section to call the vmemmap functions instead?

>  2. With an appropriate patch, we can allocate new section's memmap from
>     itself. This will reduce possibility of memory hotplug failure becasue of
>     large size kmalloc/vmalloc. And it guarantees locality of memmap.
>     But maybe need some amount of work for implementing this in clean way.
>     This will depend on vmemmap.

That is a good idea. Maybe do the simple approach first and then the other 
one?

> 
>  3. removin memmap code for memory unplug will be necessary. But there is no code
>     for removing memmap in usual SPARSEMEM. So this is not real problem of vmemmap
>     now. 

Right. It would have to be added later anyways.

--
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>

  reply	other threads:[~2007-07-13 23:17 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-07-13 13:34 [PATCH 0/7] Sparsemem Virtual Memmap V5 Andy Whitcroft
2007-07-13 13:35 ` [PATCH 1/7] sparsemem: clean up spelling error in comments Andy Whitcroft
2007-07-13 13:35 ` [PATCH 2/7] sparsemem: record when a section has a valid mem_map Andy Whitcroft
2007-07-13 14:24   ` [PATCH] Bah, hoisted by my own petard. Below is an updated version Andy Whitcroft
2007-07-13 13:36 ` [PATCH 3/7] Generic Virtual Memmap support for SPARSEMEM Andy Whitcroft
2007-07-13 14:51   ` KAMEZAWA Hiroyuki
2007-07-13 22:42     ` Christoph Lameter
2007-07-13 23:12       ` KAMEZAWA Hiroyuki
2007-07-13 23:17         ` Christoph Lameter [this message]
2007-07-13 23:25           ` KAMEZAWA Hiroyuki
2007-07-14 15:20   ` Christoph Hellwig
2007-07-14 16:06     ` Christoph Lameter
2007-07-14 16:33       ` Christoph Hellwig
2007-07-23 19:36         ` Christoph Lameter
2007-07-30 14:39     ` Andy Whitcroft
2007-07-30 18:35       ` Christoph Lameter
2007-07-13 13:36 ` [PATCH 4/7] x86_64: SPARSEMEM_VMEMMAP 2M page size support Andy Whitcroft
2007-07-19 23:25   ` Andrew Morton
2007-07-13 13:37 ` [PATCH 5/7] IA64: SPARSEMEM_VMEMMAP 16K " Andy Whitcroft
2007-07-13 13:37 ` [PATCH 6/7] SPARC64: SPARSEMEM_VMEMMAP support Andy Whitcroft
2007-07-13 17:00   ` Christoph Lameter
2007-07-13 13:38 ` [PATCH 7/7] ppc64: " Andy Whitcroft
2007-07-13 17:04 ` [PATCH 0/7] Sparsemem Virtual Memmap V5 Christoph Lameter
2007-07-13 17:40   ` Andrew Morton
2007-07-13 18:23     ` Christoph Lameter
2007-07-14  8:57       ` Russell King
2007-07-14 15:10         ` Christoph Lameter
2007-07-14 17:16           ` Russell King
2007-07-13 20:08     ` Roman Zippel
2007-07-13 22:02     ` Luck, Tony
2007-07-13 22:21       ` Christoph Lameter
2007-07-13 22:37         ` Luck, Tony
2007-07-13 22:54           ` Christoph Lameter
2007-07-13 23:27             ` KAMEZAWA Hiroyuki
2007-07-13 23:28               ` Christoph Lameter
2007-07-14  8:49         ` Nick Piggin
2007-07-14 15:07           ` Christoph Lameter
2007-07-13 22:43     ` David Miller, Andrew Morton
2007-07-26  8:05 ` Paul Mundt

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=Pine.LNX.4.64.0707131612530.26795@schroedinger.engr.sgi.com \
    --to=clameter@sgi.com \
    --cc=apw@shadowen.org \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=npiggin@suse.de \
    /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