From: Andrea Arcangeli <andrea@suse.de>
To: Daniel Phillips <phillips@bonn-fries.net>
Cc: Russell King <rmk@arm.linux.org.uk>, linux-kernel@vger.kernel.org
Subject: Re: Bug: Discontigmem virt_to_page() [Alpha,ARM,Mips64?]
Date: Thu, 2 May 2002 15:34:02 +0200 [thread overview]
Message-ID: <20020502153402.A11414@dualathlon.random> (raw)
In-Reply-To: <20020426192711.D18350@flint.arm.linux.org.uk> <E172isy-0001rL-00@starship> <20020502034318.T11414@dualathlon.random> <E172k3S-0001sH-00@starship>
On Wed, May 01, 2002 at 04:41:17AM +0200, Daniel Phillips wrote:
> On Thursday 02 May 2002 03:43, Andrea Arcangeli wrote:
> > On Wed, May 01, 2002 at 03:26:22AM +0200, Daniel Phillips wrote:
> > > For your information, the mem_map lives in *virtual* memory, it does not
> > > need to change location, only the kernel page tables need to be updated,
> > > to allow a section of kernel memory to be moved to a different physical
> > > location. For user memory, this was always possible, now it is possible
> > > for kernel memory as well. Naturally, it's not all you have to do to get
> > > hotplug memory, but it's a big step in that direction.
> >
> > what kernel pagetables?
>
> The normal page tables that are used to map kernel memory.
>
> > pagetables for space that you left free for what?
>
> These page tables have not been left free for anything. The nice thing about
> page tables is that you can change the page table entries to point wherever
> you want. (I know you know this.) This is what config_nonlinear supports,
> and that is why it's called config_nonlinear. When we want to remap part of
> the kernel memory to a different piece of physical memory, we just need to
> fill in some pte's. The tricky part is knowing how to fill in the ptes, and
> config_nonlinear takes care of that.
>
> > You waste virtual space for that at the very least on x86 that is
> > just very tigh, at this point kernel virtual space is more costly than
> > physical space these days. And nevertheless most archs doesn't have
> > pagetables at all to read and write the page structures. yes it's
> > virtual memory but it's a direct mapping.
>
> Most architectures? That's quite possibly an exaggeration. Some
> architectures - MIPS32 for example - make this difficult or impossible,
> and so what? Those can't do software hotplug memory, sorry.
alpha is the same as mips I think. sparc would be the same too if
there's any discontigmem sparc. Dunno of arm. We're talking about
architectures needing discontigmem, 99% percent of users doesn't need
discontigmem in the first place, you never need discontigmem in x86 and
even in new-numa you don't need discontigmem, you want to pass through
discontigmem only to get the numa topology description that the current
discontigmem provides via the pgdat.
Andrea
next prev parent reply other threads:[~2002-05-02 13:34 UTC|newest]
Thread overview: 152+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-04-26 18:27 Bug: Discontigmem virt_to_page() [Alpha,ARM,Mips64?] Russell King
2002-04-26 22:46 ` Andrea Arcangeli
2002-04-29 17:50 ` Martin J. Bligh
2002-04-29 22:00 ` Roman Zippel
2002-04-30 0:43 ` Andrea Arcangeli
2002-04-27 22:10 ` Daniel Phillips
2002-04-29 13:35 ` Andrea Arcangeli
2002-04-29 23:02 ` Daniel Phillips
2002-05-01 2:23 ` Andrea Arcangeli
2002-04-30 23:12 ` Daniel Phillips
2002-05-01 1:05 ` Daniel Phillips
2002-05-02 0:47 ` Andrea Arcangeli
2002-05-01 1:26 ` Daniel Phillips
2002-05-02 1:43 ` Andrea Arcangeli
2002-05-01 2:41 ` Daniel Phillips
2002-05-02 13:34 ` Andrea Arcangeli [this message]
2002-05-02 15:18 ` Martin J. Bligh
2002-05-02 15:35 ` Andrea Arcangeli
2002-05-01 15:42 ` Daniel Phillips
2002-05-02 16:06 ` Andrea Arcangeli
2002-05-02 16:10 ` Martin J. Bligh
2002-05-02 16:40 ` Andrea Arcangeli
2002-05-02 17:16 ` William Lee Irwin III
2002-05-02 18:41 ` Andrea Arcangeli
2002-05-02 19:19 ` William Lee Irwin III
2002-05-02 19:27 ` Daniel Phillips
2002-05-02 19:38 ` William Lee Irwin III
2002-05-02 19:58 ` Daniel Phillips
2002-05-03 6:28 ` Andrea Arcangeli
2002-05-03 6:10 ` Andrea Arcangeli
2002-05-02 22:20 ` Martin J. Bligh
2002-05-02 21:28 ` William Lee Irwin III
2002-05-02 21:52 ` Kurt Ferreira
2002-05-02 21:55 ` William Lee Irwin III
2002-05-03 6:38 ` Andrea Arcangeli
2002-05-03 6:58 ` Martin J. Bligh
2002-05-03 6:04 ` Andrea Arcangeli
2002-05-03 6:33 ` Martin J. Bligh
2002-05-03 8:38 ` Andrea Arcangeli
2002-05-03 9:26 ` William Lee Irwin III
2002-05-03 15:38 ` Martin J. Bligh
2002-05-03 15:17 ` Virtual address space exhaustion (was Discontigmem virt_to_page() ) Martin J. Bligh
2002-05-03 15:58 ` Andrea Arcangeli
2002-05-03 16:10 ` Martin J. Bligh
2002-05-03 16:25 ` Andrea Arcangeli
2002-05-03 16:02 ` Daniel Phillips
2002-05-03 16:20 ` Andrea Arcangeli
2002-05-03 16:41 ` Daniel Phillips
2002-05-03 16:58 ` Andrea Arcangeli
2002-05-03 18:08 ` Daniel Phillips
2002-05-03 9:24 ` Bug: Discontigmem virt_to_page() [Alpha,ARM,Mips64?] William Lee Irwin III
2002-05-03 10:30 ` Andrea Arcangeli
2002-05-03 11:09 ` William Lee Irwin III
2002-05-03 11:27 ` Andrea Arcangeli
2002-05-03 15:42 ` Martin J. Bligh
2002-05-03 15:32 ` Martin J. Bligh
2002-05-02 19:22 ` Daniel Phillips
2002-05-03 6:06 ` Andrea Arcangeli
2002-05-02 18:25 ` Daniel Phillips
2002-05-02 18:44 ` Andrea Arcangeli
2002-05-02 19:31 ` Martin J. Bligh
2002-05-02 18:57 ` Andrea Arcangeli
2002-05-02 19:08 ` Daniel Phillips
2002-05-03 5:15 ` Andrea Arcangeli
2002-05-05 23:54 ` Daniel Phillips
2002-05-06 0:28 ` Andrea Arcangeli
2002-05-06 0:34 ` Daniel Phillips
2002-05-06 1:01 ` Andrea Arcangeli
2002-05-06 0:55 ` Russell King
2002-05-06 1:07 ` Daniel Phillips
2002-05-06 1:20 ` Andrea Arcangeli
2002-05-06 1:24 ` Daniel Phillips
2002-05-06 1:42 ` Andrea Arcangeli
2002-05-06 1:48 ` Daniel Phillips
2002-05-06 2:06 ` Andrea Arcangeli
2002-05-06 17:40 ` Daniel Phillips
2002-05-06 19:09 ` Martin J. Bligh
2002-05-06 1:09 ` Andrea Arcangeli
2002-05-06 1:13 ` Daniel Phillips
2002-05-06 2:03 ` Daniel Phillips
2002-05-06 2:31 ` Andrea Arcangeli
2002-05-06 8:57 ` Russell King
2002-05-06 8:54 ` Roman Zippel
2002-05-06 15:26 ` Daniel Phillips
2002-05-06 19:07 ` Roman Zippel
2002-05-08 15:57 ` Daniel Phillips
2002-05-08 23:11 ` Roman Zippel
2002-05-09 16:08 ` Daniel Phillips
2002-05-09 22:06 ` Roman Zippel
2002-05-09 22:22 ` Daniel Phillips
2002-05-09 23:00 ` Roman Zippel
2002-05-09 23:22 ` Daniel Phillips
2002-05-10 0:13 ` Roman Zippel
2002-05-02 22:39 ` Martin J. Bligh
2002-05-03 7:04 ` Andrea Arcangeli
2002-05-02 23:42 ` Daniel Phillips
2002-05-03 7:45 ` Andrea Arcangeli
2002-05-02 16:07 ` Martin J. Bligh
2002-05-02 16:58 ` Gerrit Huizenga
2002-05-02 18:10 ` Andrea Arcangeli
2002-05-02 19:28 ` Gerrit Huizenga
2002-05-02 22:23 ` Martin J. Bligh
2002-05-03 6:20 ` Andrea Arcangeli
2002-05-03 6:39 ` Martin J. Bligh
2002-05-02 16:00 ` William Lee Irwin III
2002-05-02 2:37 ` William Lee Irwin III
2002-05-02 15:59 ` Andrea Arcangeli
2002-05-02 16:06 ` William Lee Irwin III
2002-05-01 18:05 ` Jesse Barnes
2002-05-01 23:17 ` Andrea Arcangeli
2002-05-01 23:23 ` discontiguous memory platforms Jesse Barnes
2002-05-02 0:51 ` Ralf Baechle
2002-05-02 1:27 ` Andrea Arcangeli
2002-05-02 1:32 ` Ralf Baechle
2002-05-02 8:50 ` Roman Zippel
2002-05-01 13:21 ` Daniel Phillips
2002-05-02 14:00 ` Roman Zippel
2002-05-01 14:08 ` Daniel Phillips
2002-05-02 17:56 ` Roman Zippel
2002-05-01 17:59 ` Daniel Phillips
2002-05-02 18:26 ` Roman Zippel
2002-05-02 18:32 ` Daniel Phillips
2002-05-02 19:40 ` Roman Zippel
2002-05-02 20:14 ` Daniel Phillips
2002-05-03 6:34 ` Andrea Arcangeli
2002-05-03 9:33 ` Roman Zippel
2002-05-03 6:30 ` Andrea Arcangeli
2002-05-02 18:35 ` Geert Uytterhoeven
2002-05-02 18:39 ` Daniel Phillips
2002-05-02 0:20 ` Bug: Discontigmem virt_to_page() [Alpha,ARM,Mips64?] Anton Blanchard
2002-05-01 1:35 ` Daniel Phillips
2002-05-02 1:45 ` William Lee Irwin III
2002-05-01 2:02 ` Daniel Phillips
2002-05-02 2:33 ` William Lee Irwin III
2002-05-01 2:44 ` Daniel Phillips
2002-05-02 1:46 ` Andrea Arcangeli
2002-05-01 1:56 ` Daniel Phillips
2002-05-02 1:01 ` Andrea Arcangeli
2002-05-02 15:28 ` Anton Blanchard
2002-05-01 16:10 ` Daniel Phillips
2002-05-02 15:59 ` Dave Engebretsen
2002-05-01 17:24 ` Daniel Phillips
2002-05-02 16:44 ` Dave Engebretsen
2002-05-02 16:31 ` William Lee Irwin III
2002-05-02 16:21 ` Dave Engebretsen
2002-05-02 17:28 ` William Lee Irwin III
2002-05-02 23:05 ` Daniel Phillips
2002-05-03 0:05 ` William Lee Irwin III
2002-05-03 1:19 ` Daniel Phillips
2002-05-03 19:47 ` Dave Engebretsen
2002-05-03 22:06 ` Daniel Phillips
2002-05-03 23:52 ` David Mosberger
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=20020502153402.A11414@dualathlon.random \
--to=andrea@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=phillips@bonn-fries.net \
--cc=rmk@arm.linux.org.uk \
/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