qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Anthony Liguori <anthony@codemonkey.ws>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: Jan Kiszka <jan.kiszka@siemens.com>,
	qemu-devel@nongnu.org, quintela@redhat.com
Subject: Re: [Qemu-devel] Re: [PATCH 11/11] test-vmstate: add test case to verify we don't change VMState
Date: Wed, 23 Mar 2011 11:27:27 -0500	[thread overview]
Message-ID: <4D8A1F6F.8060003@codemonkey.ws> (raw)
In-Reply-To: <AANLkTimCR0Ck67pXiNTFw1CVS08hy7Z_FbjDp48qKSif@mail.gmail.com>

On 03/23/2011 10:00 AM, Peter Maydell wrote:
> On 23 March 2011 14:52, Anthony Liguori<anthony@codemonkey.ws>  wrote:
>> I think we ought to merge VMStateDescription into DeviceInfo.  For
>> compatibility, we probably need a vmstate_alias name since the device names
>> don't always map 1-1 with the qdev names.  But this should eliminate the
>> problem of reusing VMStateDescriptions for multiple devices.
> That's a feature, not a bug. Consider eg hw/pl110.c -- there
> are two different DeviceInfo devices but since the underlying
> implementation is the same you definitely don't want to have
> two separate VMStateDescription structures to get out of sync.

No, it's a bug.

Migration uses the VMStateDescription name as a section identifier.  The 
section identifiers MUST be unique for a given device.  Otherwise, if 
both devices are present, migration fails miserably.  It also means that 
if the wrong devices are created on the destination, instead of 
predictable failure, you get unpredictable guest corruption.

The Right Way to support what you're describing above is to have a 
single VMStateField array and two VMStateDescriptions.  IOW:

VMStateField pl110_fields[] = {
         VMSTATE_INT32(versatile, pl110_state),
         VMSTATE_UINT32_ARRAY(timing, pl110_state, 4),
         VMSTATE_UINT32(cr, pl110_state),
         VMSTATE_UINT32(upbase, pl110_state),
         VMSTATE_UINT32(lpbase, pl110_state),
         VMSTATE_UINT32(int_status, pl110_state),
         ....
};

VMStateDescription vmstate_pl110 = {
      .name = "pl110",
      ...
      .fields = pl110_fields,
};

VMStateDescription vmstate_pl110_versatile = {
      .name = "pl110-versatile",
      ...
      .fields = pl110_fields,
};

Merging VMStateDescription with DeviceInfo will force this on people 
making it difficult to make this mistake again.

Regards,

Anthony Liguori



> -- PMM
>

  parent reply	other threads:[~2011-03-23 16:27 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-23  0:16 [Qemu-devel] [PATCH 00/11] Add live migration unit tests Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 01/11] Add hard build dependency on glib Anthony Liguori
2011-03-23  8:13   ` Stefan Hajnoczi
2011-03-23  8:19     ` Roy Tam
2011-03-23  8:41       ` Stefan Hajnoczi
2011-03-23  8:58         ` Roy Tam
2011-03-23  9:19           ` [Qemu-devel] " Paolo Bonzini
2011-03-23  9:22           ` [Qemu-devel] " Stefan Hajnoczi
2011-03-23  0:16 ` [Qemu-devel] [PATCH 02/11] vmstate: register all VMStateDescriptions Anthony Liguori
2011-03-23  9:00   ` Alon Levy
2011-03-23 12:30   ` Peter Maydell
2011-03-23 12:37     ` Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 03/11] vmstate: for vmstate types that reuse the same field, make sure name is unique Anthony Liguori
2011-03-23  2:29   ` Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 04/11] sb16: fix migration quirk Anthony Liguori
2011-03-23  9:51   ` [Qemu-devel] " Juan Quintela
2011-03-23 12:16     ` Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 05/11] vga-isa: fix migration by breaking it Anthony Liguori
2011-03-23  9:54   ` [Qemu-devel] " Juan Quintela
2011-03-23  0:16 ` [Qemu-devel] [PATCH 06/11] fdc: fix migration of non-ISA fdc devices Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 07/11] eeprom93xx: Use the new hack macro to avoid duplicate field names Anthony Liguori
2011-03-23  9:58   ` [Qemu-devel] " Juan Quintela
2011-03-23 12:34     ` Anthony Liguori
2011-03-23 14:14       ` Juan Quintela
2011-03-23 14:33         ` Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 08/11] fw_cfg: make sure all VMState fields are unique Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 09/11] pckbd: make non-ISA pckbd use a unique name Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 10/11] vl: add a new -vmstate-dump option to write a VMState JSON schema Anthony Liguori
2011-03-23  0:16 ` [Qemu-devel] [PATCH 11/11] test-vmstate: add test case to verify we don't change VMState Anthony Liguori
2011-03-23  2:31   ` Anthony Liguori
2011-03-23  8:37   ` Stefan Hajnoczi
2011-03-23 10:22   ` Peter Maydell
2011-03-23 12:42     ` Anthony Liguori
2011-03-23 14:17       ` [Qemu-devel] " Juan Quintela
2011-03-23 14:52         ` Anthony Liguori
2011-03-23 15:00           ` Peter Maydell
2011-03-23 15:06             ` Jan Kiszka
2011-03-23 16:27             ` Anthony Liguori [this message]
2011-03-23 16:36               ` Jan Kiszka
2011-03-23 18:42                 ` Anthony Liguori
2011-03-23 16:45               ` Peter Maydell
2011-03-23 17:01                 ` Anthony Liguori
2011-03-23 15:26           ` Juan Quintela
2011-03-23 16:14             ` Anthony Liguori
2011-03-23 16:44               ` Juan Quintela
2011-03-23 14:19     ` Juan Quintela
2011-03-23 14:36       ` Peter Maydell
2011-03-23 15:13         ` Juan Quintela
2011-03-23 15:24           ` Peter Maydell

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=4D8A1F6F.8060003@codemonkey.ws \
    --to=anthony@codemonkey.ws \
    --cc=jan.kiszka@siemens.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.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).