qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin O'Connor <kevin@koconnor.net>
To: Gleb Natapov <gleb@redhat.com>
Cc: Gerhard Wiesinger <lists@wiesinger.com>,
	Gerd Hoffmann <kraxel@redhat.com>,
	qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] Boot order problem and disable iPXE/gPXE
Date: Sun, 8 Jan 2012 10:47:47 -0500	[thread overview]
Message-ID: <20120108154747.GA32376@morn.localdomain> (raw)
In-Reply-To: <20120108151745.GE2167@redhat.com>

On Sun, Jan 08, 2012 at 05:17:45PM +0200, Gleb Natapov wrote:
> On Sun, Jan 08, 2012 at 09:58:53AM -0500, Kevin O'Connor wrote:
> > On Sun, Jan 08, 2012 at 10:31:05AM +0200, Gleb Natapov wrote:
> > > Cool! What about non pnp roms that also register bcv? Should we use HD
> > > priority for those too?
> > 
> > A non-pnp rom can't register a BCV, but it can hook int13 (or hook
> > int19 or do a direct boot).  Right now, if a user selects a "legacy
> > option rom" from the boot menu, SeaBIOS will still prioritize HD
> > booting even if the rom doesn't hook int13.  It's an open question
> > whether that is the right thing to do.  In practice, though, there
> > aren't many legacy option roms, so it's probably not a big deal.  The
> > "vapic.bin" should probably be upgraded to use a PNP header just so it
> > doesn't show up in the boot menu.
> > 
> I was talking about this code in optionroms.c:
> 
>         if (! pnp) {
>             // Legacy rom.
>             boot_add_bcv(FLATPTR_TO_SEG(rom), OPTION_ROM_INITVECTOR, 0
>                          , getRomPriority(sources, rom, 0));
>             continue;
>         }
> 
> Isn't it register bcv for each non pnp option rom found?

Yes.

>As far as I
> understand this is done in order to postpone executing such rom until
> it is absolutely clear that user wants to use it for booting.

No - the option rom will always be executed.  The purpose of placing
it in the BCV list is to order its execution with respect to other
BCVs so that if the legacy option rom hooks int13 it will do so after
other drives have had a chance.

Should a legacy rom do direct boot or hook int19 it will take priority
over everything else - irrespective of seabios priorities or what the
user selects in the boot menu.  This lack of control is what spawned
the whole PNP rom stuff.

>If
> boot_add_bcv() will unconditionally give HD boot priority to such bcv it
> will have higher priority than CD or pxe boot option, no?

As above, the BCV priorities are only relevant when compared to the
priority of other BCVs.  So, no - it should have no impact on CD/PXE.

As I mentioned in my previous post, SeaBIOS does currently prioritizes
hard drive booting when the user selects "Legacy option rom" from the
boot menu - it's unclear if this is the right thing to do.

To be honest, this wasn't all that clear to me yesterday - I had to
think on it a bit.  This legacy boot stuff is crazy.

-Kevin

  reply	other threads:[~2012-01-08 15:47 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-04 18:02 [Qemu-devel] Boot order problem and disable iPXE/gPXE Gerhard Wiesinger
2012-01-05  3:03 ` Kevin O'Connor
2012-01-05  7:39   ` Gerhard Wiesinger
2012-01-06  1:42     ` Kevin O'Connor
2012-01-06  8:20       ` Gerhard Wiesinger
2012-01-06  8:59         ` Gerd Hoffmann
2012-01-05  7:50 ` Gleb Natapov
2012-01-05  7:58 ` Gerd Hoffmann
2012-01-05  8:23   ` Gerhard Wiesinger
2012-01-05  8:32     ` Gleb Natapov
2012-01-05  9:09       ` Gerhard Wiesinger
2012-01-05  9:20         ` Gleb Natapov
2012-01-05  9:27           ` Gerhard Wiesinger
2012-01-05  9:29             ` Gleb Natapov
2012-01-05  9:36               ` Gerhard Wiesinger
2012-01-05  9:42                 ` Gleb Natapov
2012-01-05  9:56                   ` Gerhard Wiesinger
2012-01-05 10:00                     ` Gleb Natapov
2012-01-05  9:42             ` Gerd Hoffmann
2012-01-06  2:05           ` Kevin O'Connor
2012-01-06  5:27             ` Gleb Natapov
2012-01-07  1:53               ` Kevin O'Connor
2012-01-07  9:00                 ` Gerhard Wiesinger
2012-01-07 16:28                   ` Kevin O'Connor
2012-01-07 22:47                     ` Gerhard Wiesinger
2012-01-07 22:53                       ` Kevin O'Connor
2012-01-08  8:31                     ` Gleb Natapov
2012-01-08 14:58                       ` Kevin O'Connor
2012-01-08 15:17                         ` Gleb Natapov
2012-01-08 15:47                           ` Kevin O'Connor [this message]
2012-01-25 20:30                             ` Gerhard Wiesinger
2012-01-09  9:10                 ` Gerd Hoffmann
2012-01-07  8:35             ` Gerhard Wiesinger
2012-01-07 16:20               ` Kevin O'Connor
2012-01-07 22:46                 ` Gerhard Wiesinger
2012-01-07 22:59                   ` Gerhard Wiesinger
2012-01-08 11:59                     ` Gerhard Wiesinger
2012-01-05  8:57     ` Gerd Hoffmann
2012-01-05  9:24       ` Gerhard Wiesinger
2012-01-05  9:53         ` Gerd Hoffmann
2012-01-05 10:08           ` Gerhard Wiesinger
2012-01-05 10:13             ` Gerd Hoffmann
2012-01-05 11:13               ` Gerhard Wiesinger
2012-01-05 11:16                 ` Gleb Natapov

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=20120108154747.GA32376@morn.localdomain \
    --to=kevin@koconnor.net \
    --cc=gleb@redhat.com \
    --cc=kraxel@redhat.com \
    --cc=lists@wiesinger.com \
    --cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).