All of lore.kernel.org
 help / color / mirror / Atom feed
From: Michael Tokarev <mjt@tls.msk.ru>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: Stefan Hajnoczi <stefanha@gmail.com>,
	Hitoshi Mitake <h.mitake@gmail.com>,
	qemu-devel@nongnu.org, Anthony Liguori <anthony@codemonkey.ws>
Subject: Re: [Qemu-devel] [PATCH v2] stop using stdio for monitor/serial/etc with -daemonize
Date: Wed, 26 Sep 2012 12:17:39 +0400	[thread overview]
Message-ID: <5062BA23.705@msgid.tls.msk.ru> (raw)
In-Reply-To: <CAFEAcA8F2otWMrJC99xRv3mHtovO9gBxjk-ysqjVASe6QShtxg@mail.gmail.com>

On 26.09.2012 12:00, Peter Maydell wrote:

>> I know lots of people use -nographic -daemonize to run headless
>> guests in background (like, for example, a router).  I guess it
>> come way before -vga option has been introduced, but at least I
>> know about -vga (but not about -vga none).  For one, I never saw
>> -display before.  And it looks like -nographic is a synonym for
>> -display none, and -curses is a synonym for -display curses.

I mean, -nographic is about the same as -vga none -display none.

> -nographic does about three different things at once (and I think
> some of its effects aren't documented). It's a legacy option retained
> for backward compatibility with old command lines.

Sure.  Just like, for example, -stdvga was at the time being.

> If you want something that is non-confusing and makes sense, then
> use -display none to disable graphics, -serial stdio to send serial
> to stdio, and so on. These newer options do one clear thing each
> and can be combined straightforwardly.
> 
>> It looks like we have way too many confusing options doing the
>> same thing.  And I think they should be consistent, at least
>> when they SMELL like they do the same thing, instead of forbidding
>> one or another in some situations.
> 
> I'd love to drop -nographic but we'd break huge numbers of
> existing setups...

So let's make it actually work as expected till we're able to finally
drop it.

What is equivalent of -nographic in terms of -vga/-display/-...?
>From the code it is something like

 -vga none -display none -serial mon:stdio -parallel null

(this is the code I tried to patch).

Note: this, compbined with -daemonize, also has the same issue,
namely, the tty is left in a bad state after qemu process backgrounded,
and for the very same reason: -serial stdio switches the try into
raw mode.  So this should be fixed too -- somehow, either by forbidding
this combination completely or by silently substituting stdio for
-serial with null.  But it will be done in a subsequent patch.

Note also: by forbidding -nographic -daemonize, we'll break lots of
existing setups too, and I still don't see why this combination is
bad, I already demonstrated that it can be made to work in a more
or less reasonable/expected way.

Thanks,

/mjt

  reply	other threads:[~2012-09-26  8:17 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-25 14:48 [Qemu-devel] [PATCH v2] stop using stdio for monitor/serial/etc with -daemonize Michael Tokarev
2012-09-25 21:19 ` Anthony Liguori
2012-09-26  7:09   ` Michael Tokarev
2012-09-26  8:00     ` Peter Maydell
2012-09-26  8:17       ` Michael Tokarev [this message]
2012-09-26  8:43         ` Peter Maydell
2012-09-26 13:46           ` Anthony Liguori
2012-09-26 14:56             ` Michael Tokarev
2012-10-27 11:23               ` Michael Tokarev
2012-10-27 11:33                 ` Peter Maydell
2012-10-27 12:15                   ` Michael Tokarev
2012-10-27 12:48                     ` Blue Swirl
2012-10-27 12:55                       ` Michael Tokarev
2012-10-27 13:11                         ` Michael Tokarev

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=5062BA23.705@msgid.tls.msk.ru \
    --to=mjt@tls.msk.ru \
    --cc=anthony@codemonkey.ws \
    --cc=h.mitake@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.