From: Peter Chubb <peterc@gelato.unsw.edu.au>
To: David Mosberger-Tang <dmosberger@gmail.com>
Cc: Peter Chubb <peterc@gelato.unsw.edu.au>,
David Miller <davem@davemloft.net>,
clameter@sgi.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org,
torvalds@linux-foundation.org, ianw@gelato.unsw.edu.au
Subject: Re: larger default page sizes...
Date: Wed, 26 Mar 2008 11:57:23 +1100 [thread overview]
Message-ID: <87lk465mks.wl%peter@chubb.wattle.id.au> (raw)
In-Reply-To: <ed5aea430803251734u70f199w10951bc4f0db6262@mail.gmail.com>
>>>>> "David" == David Mosberger-Tang <dmosberger@gmail.com> writes:
David> On Tue, Mar 25, 2008 at 5:41 PM, Peter Chubb
David> <peterc@gelato.unsw.edu.au> wrote:
>> The main issue is that, at least on Itanium, you have to turn off
>> the hardware page table walker for hugepages if you want to mix
>> superpages and standard pages in the same region. (The long format
>> VHPT isn't the panacea we'd like it to be because the hash function
>> it uses depends on the page size).
David> Why not just repeat the PTEs for super-pages? That won't work
David> for huge pages, but for superpages that are a reasonable
David> multiple (e.g., 16-times) the base-page size, it should work
David> nicely.
You end up having to repeat PTEs to fit into Linux's page table
structure *anyway* (unless we can change Linux's page table). But
there's no place in the short format hardware-walked page table (that
reuses the leaf entries in Linux's table) for a page size. And if you
use some of the holes in the format, the hardware walker doesn't
understand it --- so you have to turn off the hardware walker for
*any* regions where there might be a superpage.
If you use the long format VHPT, you have a choice: load the
hash table with just the translation that caused the miss, load all
possible hash entries that could have caused the miss for the page, or
preload the hash table when the page is instantiated, with all
possible entries that could hash to the huge page. I don't remember
the details, but I seem to remember all these being bad choices for
one reason or other ... Ian, can you elaborate?
--
Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au
http://www.ertos.nicta.com.au ERTOS within National ICT Australia
--
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:[~2008-03-26 0:57 UTC|newest]
Thread overview: 87+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-21 6:17 [00/14] Virtual Compound Page Support V3 Christoph Lameter
2008-03-21 6:17 ` [01/14] vcompound: Return page array on vunmap Christoph Lameter
2008-03-21 6:17 ` [02/14] vcompound: pageflags: Add PageVcompound() Christoph Lameter
2008-03-21 6:17 ` [03/14] vmallocinfo: Support display of vcompound for a virtual compound page Christoph Lameter
2008-03-21 7:55 ` Eric Dumazet
2008-03-21 17:32 ` Christoph Lameter
2008-03-21 6:17 ` [04/14] vcompound: Core piece Christoph Lameter
2008-03-22 12:10 ` KOSAKI Motohiro
2008-03-24 18:28 ` Christoph Lameter
2008-03-21 6:17 ` [05/14] vcompound: Debugging aid Christoph Lameter
2008-03-21 6:17 ` [06/14] vcompound: Virtual fallback for sparsemem Christoph Lameter
2008-03-21 6:17 ` [07/14] vcompound: bit waitqueue support Christoph Lameter
2008-03-21 6:17 ` [08/14] vcompound: Fallback for zone wait table Christoph Lameter
2008-03-21 6:17 ` [09/14] vcompound: crypto: Fallback for temporary order 2 allocation Christoph Lameter
2008-03-21 6:17 ` [10/14] vcompound: slub: Use for buffer to correlate allocation addresses Christoph Lameter
2008-03-21 6:17 ` [11/14] vcompound: Fallbacks for order 1 stack allocations on IA64 and x86 Christoph Lameter
2008-03-21 7:25 ` David Miller, Christoph Lameter
2008-03-21 8:39 ` Ingo Molnar
2008-03-21 17:33 ` Christoph Lameter
2008-03-21 19:02 ` Ingo Molnar
2008-03-21 19:04 ` Christoph Lameter
2008-03-21 17:40 ` Christoph Lameter
2008-03-21 21:57 ` David Miller, Christoph Lameter
2008-03-24 18:27 ` Christoph Lameter
2008-03-24 20:37 ` larger default page sizes David Miller, Christoph Lameter
2008-03-24 21:05 ` Christoph Lameter
2008-03-24 21:43 ` David Miller, Christoph Lameter
2008-03-25 17:48 ` Christoph Lameter
2008-03-25 23:22 ` David Miller, Christoph Lameter
2008-03-25 23:41 ` Peter Chubb
2008-03-25 23:49 ` David Miller, Peter Chubb
2008-03-26 0:25 ` Peter Chubb
2008-03-26 0:31 ` David Miller, Peter Chubb
2008-03-26 0:34 ` David Mosberger-Tang
2008-03-26 0:39 ` David Miller, David Mosberger-Tang
2008-03-26 0:57 ` Peter Chubb [this message]
2008-03-26 4:16 ` John Marvin
2008-03-26 4:36 ` David Miller, John Marvin
2008-03-24 21:25 ` Luck, Tony
2008-03-24 21:46 ` David Miller, Luck, Tony
2008-03-25 3:29 ` Paul Mackerras
2008-03-25 4:15 ` David Miller, Paul Mackerras
2008-03-25 11:50 ` Paul Mackerras
2008-03-25 23:32 ` David Miller, Paul Mackerras
2008-03-25 23:49 ` Luck, Tony
2008-03-26 0:16 ` David Miller, Luck, Tony
2008-03-26 15:54 ` Nish Aravamudan
2008-03-26 17:05 ` Luck, Tony
2008-03-26 18:54 ` Mel Gorman
2008-03-25 12:05 ` Andi Kleen
2008-03-25 21:27 ` Paul Mackerras
2008-03-26 5:24 ` Paul Mackerras
2008-03-26 15:59 ` Linus Torvalds
2008-03-27 1:08 ` Paul Mackerras
2008-03-26 17:56 ` Christoph Lameter
2008-03-26 23:21 ` David Miller, Christoph Lameter
2008-03-27 3:00 ` Paul Mackerras
2008-03-25 18:27 ` Dave Hansen
2008-03-24 21:13 ` [11/14] vcompound: Fallbacks for order 1 stack allocations on IA64 and x86 Luck, Tony
2008-03-25 17:42 ` Christoph Lameter
2008-03-25 19:09 ` Luck, Tony
2008-03-25 19:25 ` Christoph Lameter
2008-03-21 22:30 ` Andi Kleen
2008-03-24 19:53 ` Christoph Lameter
2008-03-25 7:51 ` Andi Kleen
2008-03-25 17:55 ` Christoph Lameter
2008-03-25 18:07 ` Andi Kleen
2008-03-21 6:17 ` [12/14] vcompound: Avoid vmalloc in e1000 driver Christoph Lameter
2008-03-21 17:27 ` Kok, Auke
2008-03-21 6:17 ` [13/14] vcompound: Use vcompound for swap_map Christoph Lameter
2008-03-21 21:25 ` Andi Kleen
2008-03-21 21:33 ` Christoph Lameter
2008-03-24 19:54 ` Christoph Lameter
2008-03-25 7:52 ` Andi Kleen
2008-03-25 17:45 ` Christoph Lameter
2008-03-25 17:55 ` Andi Kleen
2008-03-25 17:51 ` Christoph Lameter
2008-03-21 6:17 ` [14/14] vcompound: Avoid vmalloc for ehash_locks Christoph Lameter
2008-03-21 7:02 ` Eric Dumazet
2008-03-21 7:03 ` Christoph Lameter
2008-03-21 7:31 ` David Miller, Christoph Lameter
2008-03-21 7:42 ` Eric Dumazet
2008-03-21 7:31 ` David Miller, Eric Dumazet
2008-03-21 17:31 ` Christoph Lameter
2008-03-22 18:40 ` [00/14] Virtual Compound Page Support V3 Arjan van de Ven
2008-03-24 18:31 ` Christoph Lameter
2008-03-24 19:29 ` 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=87lk465mks.wl%peter@chubb.wattle.id.au \
--to=peterc@gelato.unsw.edu.au \
--cc=clameter@sgi.com \
--cc=davem@davemloft.net \
--cc=dmosberger@gmail.com \
--cc=ianw@gelato.unsw.edu.au \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=torvalds@linux-foundation.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