public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Suparna Bhattacharya <suparna@in.ibm.com>
To: Corey Minyard <cminyard@mvista.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>,
	Kenneth Sumrall <ken@mvista.com>,
	linux-kernel@vger.kernel.org, lkcd-devel@lists.sourceforge.net
Subject: Re: Kexec, DMA, and SMP
Date: Thu, 13 Feb 2003 16:43:27 +0530	[thread overview]
Message-ID: <20030213164327.A14429@in.ibm.com> (raw)
In-Reply-To: <3E4A70EA.4020504@mvista.com>; from cminyard@mvista.com on Wed, Feb 12, 2003 at 10:06:02AM -0600

On Wed, Feb 12, 2003 at 10:06:02AM -0600, Corey Minyard wrote:
> |
> |It is trivial if you don't let alloc_pages give the memory to anyone for
> |any purpose.
> 
> Ok, agreed, if you reserve a section of physical memory just for kexec 
> to copy it's kernel into, it will
> prevent DMA from clobbering something from the time kexec copies the 
> kernel there to the time
> decompressing starts.
> 
> Another thought.  If you add a delay with all other processors and 
> interrupts off, the disk devices
> will run out of things to do.
> 
> Once you add all the necessary quiesce functions, these can go away.
> 
> I do doubt these will make a big difference, though.  The problem we 
> were seeing was with the
> shared control structures in memory.  The new kernel laid memory out a 
> little differently and
> things like buffer pointers were overwritten with new data.  This, in 
> turn, cause the device to
> do random things.  I would guess this is the most likely scenario, since 

The trick is that the new kernel's allocations are also from the
reserved area. (Using the same technique that MCL relies on to
avoid allocating from and stomping over the pages containing 
crash dump).

So the memory used by the old and new kernel are mutually 
exclusive.

Regards
Suparna

-- 
Suparna Bhattacharya (suparna@in.ibm.com)
Linux Technology Center
IBM Software Labs, India


  reply	other threads:[~2003-02-13 10:58 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <3E448745.9040707@mvista.com>
     [not found] ` <m1isvuzjj2.fsf@frodo.biederman.org>
2003-02-08 20:18   ` Kexec, DMA, and SMP Corey Minyard
2003-02-09 18:39     ` Eric W. Biederman
2003-02-10 11:14       ` Kexec on 2.5.59 problems ? Suparna Bhattacharya
2003-02-10 17:09         ` [Fastboot] " Andy Pfiffer
2003-02-10 18:07           ` Eric W. Biederman
2003-02-11  7:21             ` Suparna Bhattacharya
2003-02-11 17:04               ` Andy Pfiffer
2003-02-11 23:46                 ` Andy Pfiffer
2003-02-12  4:29                   ` Eric W. Biederman
2003-02-12 22:31                     ` Andy Pfiffer
2003-02-13  9:50                       ` Suparna Bhattacharya
2003-02-13 15:10                         ` Eric W. Biederman
2003-02-18 10:59                           ` Suparna Bhattacharya
2003-02-18 15:06                             ` Eric W. Biederman
2003-02-10 12:12       ` Kexec, DMA, and SMP Suparna Bhattacharya
2003-02-10 13:56         ` Corey Minyard
2003-02-10 15:07           ` Suparna Bhattacharya
2003-02-10 15:22             ` Corey Minyard
2003-02-10 17:56         ` Eric W. Biederman
2003-02-11  1:35           ` Kenneth Sumrall
2003-02-11  5:08             ` Eric W. Biederman
2003-02-11 17:09               ` Stephen Hemminger
2003-02-11 12:55           ` Suparna Bhattacharya
2003-02-11 13:40             ` Suparna Bhattacharya
2003-02-11 14:06               ` Corey Minyard
2003-02-11 14:40                 ` Suparna Bhattacharya
2003-02-11 15:20                   ` Corey Minyard
2003-02-12  4:28                     ` Eric W. Biederman
2003-02-12 14:17                       ` Corey Minyard
2003-02-12 14:51                         ` Eric W. Biederman
2003-02-12 16:06                           ` Corey Minyard
2003-02-13 11:13                             ` Suparna Bhattacharya [this message]
2003-02-14  3:13                             ` Werner Almesberger
2003-02-14 14:20                               ` Corey Minyard
2003-02-14 18:10                                 ` Werner Almesberger
2003-02-14 18:23                                   ` Corey Minyard
2003-02-14 19:26                                     ` Zwane Mwaikambo
2003-02-14 19:44                                       ` Werner Almesberger
2003-02-14 20:00                                         ` Corey Minyard
2003-02-15  6:03                                           ` Eric W. Biederman
2003-02-16 16:22                                             ` Corey Minyard
2003-02-16 21:48                                               ` Eric W. Biederman
2003-02-17  4:26                                                 ` Corey Minyard
2003-02-17  7:18                                                   ` Eric W. Biederman
2003-02-17 17:32                                                     ` Corey Minyard
2003-02-12  4:47                     ` Suparna Bhattacharya

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=20030213164327.A14429@in.ibm.com \
    --to=suparna@in.ibm.com \
    --cc=cminyard@mvista.com \
    --cc=ebiederm@xmission.com \
    --cc=ken@mvista.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkcd-devel@lists.sourceforge.net \
    /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