From: Avi Kivity <avi@redhat.com>
To: Anthony Liguori <anthony@codemonkey.ws>
Cc: Jan Kiszka <jan.kiszka@siemens.com>,
"xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>,
Gerd Hoffmann <kraxel@redhat.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Subject: Re: [Qemu-devel] [PATCH v4 0/6] save/restore on Xen
Date: Tue, 24 Jan 2012 17:56:47 +0200 [thread overview]
Message-ID: <4F1ED4BF.20505@redhat.com> (raw)
In-Reply-To: <4F1EAEC7.5050304@codemonkey.ws>
On 01/24/2012 03:14 PM, Anthony Liguori wrote:
> On 01/24/2012 05:10 AM, Avi Kivity wrote:
>> On 01/23/2012 07:18 PM, Anthony Liguori wrote:
>>> Generally speaking, RAM is an independent device in most useful cases.
>>
>> Can you give examples? Do you mean a subdevice with composition, or a
>> really independent device?
>
> I expect we'll have one Ram device. It's size will be configurable.
> One Ram device will hang off of the machine which would be the main ram.
We'll also have a hotpluggable variant which talks to some GPIOs. A
motherboard may support multiple slots with such devices.
> A video card would have a Ram device via composition.
IMO, overkill.
>
> The important consideration about reset is how it propagates. My
> expectation is that we'll propagate reset through the composition tree
> in a preorder transversal. That means in the VGA device reset
> function, it's child device (Ram) has not been reset yet.
Doesn't depth first make more sense? A bus is considered reset after
all devices in the bus have been reset.
>
>>> We really should view RAM as just another device so I don't like the
>>> idea of propagating a global concept of "when RAM is restored" because
>>> that treats it specially compared to other devices.
>>
>> Agree. In fact the first step has been taken as now creating a RAM
>> region with memory_region_init_ram() doesn't register it for migration.
>> The next step would be a VMSTATE_RAM() to make it part of the containing
>> device.
>
> That's not necessary (or wise).
>
> Let's not confuse a Ram device with a MemoryRegion. They are separate
> things and should be treated as separate things. I thought we
> discussed that MemoryRegions are stateless (or at least, there state
> is all derived) and don't need to be serialized?
Well, the actual bits in memory are state. All other attributes are
indeed derived.
I just think that making any device that has a bit of RAM a composed
device is overkill. What do we gain from it? The cost is not trivial.
--
error compiling committee.c: too many arguments to function
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi@redhat.com>
To: Anthony Liguori <anthony@codemonkey.ws>
Cc: Jan Kiszka <jan.kiszka@siemens.com>,
"xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>,
Gerd Hoffmann <kraxel@redhat.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Subject: Re: [PATCH v4 0/6] save/restore on Xen
Date: Tue, 24 Jan 2012 17:56:47 +0200 [thread overview]
Message-ID: <4F1ED4BF.20505@redhat.com> (raw)
In-Reply-To: <4F1EAEC7.5050304@codemonkey.ws>
On 01/24/2012 03:14 PM, Anthony Liguori wrote:
> On 01/24/2012 05:10 AM, Avi Kivity wrote:
>> On 01/23/2012 07:18 PM, Anthony Liguori wrote:
>>> Generally speaking, RAM is an independent device in most useful cases.
>>
>> Can you give examples? Do you mean a subdevice with composition, or a
>> really independent device?
>
> I expect we'll have one Ram device. It's size will be configurable.
> One Ram device will hang off of the machine which would be the main ram.
We'll also have a hotpluggable variant which talks to some GPIOs. A
motherboard may support multiple slots with such devices.
> A video card would have a Ram device via composition.
IMO, overkill.
>
> The important consideration about reset is how it propagates. My
> expectation is that we'll propagate reset through the composition tree
> in a preorder transversal. That means in the VGA device reset
> function, it's child device (Ram) has not been reset yet.
Doesn't depth first make more sense? A bus is considered reset after
all devices in the bus have been reset.
>
>>> We really should view RAM as just another device so I don't like the
>>> idea of propagating a global concept of "when RAM is restored" because
>>> that treats it specially compared to other devices.
>>
>> Agree. In fact the first step has been taken as now creating a RAM
>> region with memory_region_init_ram() doesn't register it for migration.
>> The next step would be a VMSTATE_RAM() to make it part of the containing
>> device.
>
> That's not necessary (or wise).
>
> Let's not confuse a Ram device with a MemoryRegion. They are separate
> things and should be treated as separate things. I thought we
> discussed that MemoryRegions are stateless (or at least, there state
> is all derived) and don't need to be serialized?
Well, the actual bits in memory are state. All other attributes are
indeed derived.
I just think that making any device that has a bit of RAM a composed
device is overkill. What do we gain from it? The cost is not trivial.
--
error compiling committee.c: too many arguments to function
next prev parent reply other threads:[~2012-01-24 15:57 UTC|newest]
Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-20 17:20 [Qemu-devel] [PATCH v4 0/6] save/restore on Xen Stefano Stabellini
2012-01-20 17:20 ` Stefano Stabellini
2012-01-20 17:21 ` [Qemu-devel] [PATCH v4 1/6] vl.c: do not save the RAM state when Xen is enabled Stefano Stabellini
2012-01-20 17:21 ` Stefano Stabellini
2012-01-23 15:58 ` [Qemu-devel] " Anthony Liguori
2012-01-23 15:58 ` Anthony Liguori
2012-01-20 17:21 ` [Qemu-devel] [PATCH v4 2/6] xen mapcache: check if memory region has moved Stefano Stabellini
2012-01-20 17:21 ` Stefano Stabellini
2012-01-20 17:21 ` [Qemu-devel] [PATCH v4 3/6] Set runstate to INMIGRATE earlier Stefano Stabellini
2012-01-20 17:21 ` Stefano Stabellini
2012-01-20 17:21 ` [Qemu-devel] [PATCH v4 4/6] cirrus_vga: do not reset videoram on resume Stefano Stabellini
2012-01-20 17:21 ` Stefano Stabellini
2012-01-20 17:21 ` [Qemu-devel] [PATCH v4 5/6] xen: record physmap changes to xenstore Stefano Stabellini
2012-01-20 17:21 ` Stefano Stabellini
2012-01-20 17:21 ` [Qemu-devel] [PATCH v4 6/6] xen: change memory access behavior during migration Stefano Stabellini
2012-01-20 17:21 ` Stefano Stabellini
2012-01-20 17:59 ` [Qemu-devel] [PATCH v4 0/6] save/restore on Xen Jan Kiszka
2012-01-20 17:59 ` Jan Kiszka
2012-01-23 10:47 ` [Qemu-devel] " Stefano Stabellini
2012-01-23 10:47 ` Stefano Stabellini
2012-01-23 11:50 ` [Qemu-devel] " Jan Kiszka
2012-01-23 11:50 ` Jan Kiszka
2012-01-23 11:59 ` [Qemu-devel] " Stefano Stabellini
2012-01-23 11:59 ` Stefano Stabellini
2012-01-23 12:10 ` [Qemu-devel] " Jan Kiszka
2012-01-23 12:10 ` Jan Kiszka
2012-01-23 14:46 ` [Qemu-devel] " Stefano Stabellini
2012-01-23 14:46 ` Stefano Stabellini
2012-01-23 15:46 ` [Qemu-devel] " Jan Kiszka
2012-01-23 15:46 ` Jan Kiszka
2012-01-23 16:16 ` [Qemu-devel] " Stefano Stabellini
2012-01-23 16:16 ` Stefano Stabellini
2012-01-23 16:22 ` [Qemu-devel] " Anthony Liguori
2012-01-23 16:22 ` Anthony Liguori
2012-01-23 17:13 ` [Qemu-devel] " Jan Kiszka
2012-01-23 17:13 ` Jan Kiszka
2012-01-23 17:18 ` [Qemu-devel] " Anthony Liguori
2012-01-23 17:18 ` Anthony Liguori
2012-01-23 17:31 ` [Qemu-devel] " Jan Kiszka
2012-01-23 17:31 ` Jan Kiszka
2012-01-23 17:36 ` [Qemu-devel] " Anthony Liguori
2012-01-23 17:36 ` Anthony Liguori
2012-01-24 10:21 ` [Qemu-devel] " Gerd Hoffmann
2012-01-24 10:21 ` Gerd Hoffmann
2012-01-24 11:13 ` [Qemu-devel] " Avi Kivity
2012-01-24 11:13 ` Avi Kivity
2012-01-24 12:00 ` [Qemu-devel] " Stefano Stabellini
2012-01-24 12:00 ` Stefano Stabellini
2012-01-24 15:39 ` [Qemu-devel] " Avi Kivity
2012-01-24 15:39 ` Avi Kivity
2012-01-24 13:18 ` [Qemu-devel] " Anthony Liguori
2012-01-24 13:18 ` Anthony Liguori
2012-01-24 15:43 ` [Qemu-devel] " Avi Kivity
2012-01-24 15:43 ` Avi Kivity
2012-01-24 13:25 ` [Qemu-devel] " Anthony Liguori
2012-01-24 13:25 ` Anthony Liguori
2012-01-24 15:47 ` [Qemu-devel] " Avi Kivity
2012-01-24 15:47 ` Avi Kivity
2012-01-24 11:10 ` [Qemu-devel] " Avi Kivity
2012-01-24 11:10 ` Avi Kivity
2012-01-24 11:27 ` [Qemu-devel] " Paolo Bonzini
2012-01-24 11:27 ` Paolo Bonzini
2012-01-24 11:32 ` [Qemu-devel] " Avi Kivity
2012-01-24 11:32 ` Avi Kivity
2012-01-24 11:44 ` [Qemu-devel] " Paolo Bonzini
2012-01-24 11:44 ` Paolo Bonzini
2012-01-24 15:48 ` [Qemu-devel] " Avi Kivity
2012-01-24 15:48 ` Avi Kivity
2012-01-24 11:52 ` [Qemu-devel] " Stefano Stabellini
2012-01-24 11:52 ` Stefano Stabellini
2012-01-24 13:15 ` [Qemu-devel] " Anthony Liguori
2012-01-24 13:15 ` Anthony Liguori
2012-01-24 13:14 ` [Qemu-devel] " Anthony Liguori
2012-01-24 13:14 ` Anthony Liguori
2012-01-24 15:56 ` Avi Kivity [this message]
2012-01-24 15:56 ` Avi Kivity
2012-01-24 17:51 ` [Qemu-devel] " Anthony Liguori
2012-01-24 17:51 ` Anthony Liguori
2012-01-23 16:00 ` [Qemu-devel] " Anthony Liguori
2012-01-23 16:00 ` Anthony Liguori
2012-01-23 16:46 ` [Qemu-devel] " Stefano Stabellini
2012-01-23 16:46 ` Stefano Stabellini
2012-01-23 16:54 ` [Qemu-devel] " Anthony Liguori
2012-01-23 16:54 ` Anthony Liguori
2012-01-23 17:05 ` [Qemu-devel] " Stefano Stabellini
2012-01-23 17:05 ` Stefano Stabellini
2012-01-23 17:07 ` [Qemu-devel] " Anthony Liguori
2012-01-23 17:07 ` Anthony Liguori
-- strict thread matches above, loose matches on Subject: below --
2012-01-25 13:04 [Qemu-devel] " Stefano Stabellini
2012-01-31 15:50 ` Stefano Stabellini
2012-02-13 12:20 ` Stefano Stabellini
2012-02-21 10:22 ` Stefano Stabellini
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=4F1ED4BF.20505@redhat.com \
--to=avi@redhat.com \
--cc=anthony@codemonkey.ws \
--cc=jan.kiszka@siemens.com \
--cc=kraxel@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefano.stabellini@eu.citrix.com \
--cc=xen-devel@lists.xensource.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.