From: Kevin O'Connor <kevin@koconnor.net>
To: Anthony Liguori <anthony@codemonkey.ws>
Cc: Peter Maydell <peter.maydell@linaro.org>,
Stefan Hajnoczi <stefanha@gmail.com>,
Bug 1180923 <1180923@bugs.launchpad.net>,
qemu-devel <qemu-devel@nongnu.org>, TC1988 <panie@tiscali.it>,
Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [Qemu-devel] [Bug 1180923] [NEW] unused memory filled with 0x00 instead of 0xFF
Date: Mon, 20 May 2013 21:31:49 -0400 [thread overview]
Message-ID: <20130521013149.GA17641@morn.localdomain> (raw)
In-Reply-To: <87ehd5wqaq.fsf@codemonkey.ws>
On Fri, May 17, 2013 at 10:59:25AM -0500, Anthony Liguori wrote:
> Paolo Bonzini <pbonzini@redhat.com> writes:
> > So you can see it in three ways:
> >
> > 1) It's not a QEMU problem, but a firmware problem. You could initialize
> > the UMBs with 0xFF in SeaBIOS for example.
>
> SeaBIOS is not supposed to do any initialization. Once it's RAM,
> there's no guarantee of what the contents will be. The problem is that
> it's supposed to be ROM.
>
> I doubt it's still there, but BIOSes often had an option to disable
> ROM shadowing expressly because it breaks applications that assume that
> this space is ROM after BIOS loads.
>
> Perhaps SeaBIOS could support such an option but we can't support it in
> KVM at the moment.
It would not be easy to make SeaBIOS work with a truly read-only rom.
SeaBIOS is designed to store static settings in the f-segment (where
it can be accessed relative to %cs at runtime), and undoing that would
be quite difficult.
However, it is not too difficult to make SeaBIOS turn the
0xc0000-0x100000 area into read-only memory after initialization. The
biggest stumbling block is that SeaBIOS stores read/write variables in
the e-segment today, but it's possible to store them at the end of the
640k region instead (I posted a patch for this at:
http://www.seabios.org/pipermail/seabios/2013-February/005693.html ).
Once the memory 0xc0000-0x100000 ram can be read-only, setting the pam
registers to mark it as read-only and memset'ing the region to 0xff is
straight forward.
-Kevin
next prev parent reply other threads:[~2013-05-21 1:31 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-05-16 17:34 [Qemu-devel] [Bug 1180923] [NEW] unused memory filled with 0x00 instead of 0xFF TC1988
2013-05-16 17:40 ` [Qemu-devel] [Bug 1180923] " TC1988
2013-05-16 17:45 ` Battler
2013-05-16 19:20 ` [Qemu-devel] [Bug 1180923] [NEW] " Paolo Bonzini
2013-05-16 19:52 ` Stefan Hajnoczi
2013-05-16 20:00 ` Peter Maydell
2013-05-17 7:25 ` Paolo Bonzini
2013-05-17 15:59 ` Anthony Liguori
2013-05-21 1:31 ` Kevin O'Connor [this message]
2013-05-16 19:20 ` [Qemu-devel] [Bug 1180923] " Battler
2013-05-17 14:26 ` Battler
2013-05-17 14:36 ` Battler
2013-05-17 16:43 ` Battler
2020-08-07 9:25 ` Thomas Huth
2020-10-07 4:17 ` Launchpad Bug Tracker
2022-07-31 20:59 ` David Glover
2022-08-01 6:29 ` 罗勇刚(Yonggang Luo)
2022-08-01 6:29 ` Yonggang Luo
2022-08-01 6:30 ` Yonggang Luo
2023-06-12 21:57 ` Evren
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=20130521013149.GA17641@morn.localdomain \
--to=kevin@koconnor.net \
--cc=1180923@bugs.launchpad.net \
--cc=anthony@codemonkey.ws \
--cc=panie@tiscali.it \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@gmail.com \
/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).