From: Trolle Selander <trolle.selander@eu.citrix.com>
To: Avi Kivity <avi@redhat.com>
Cc: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH] [RFC] Variable video ram size option
Date: Mon, 12 Jan 2009 12:43:17 -0500 [thread overview]
Message-ID: <496B8135.4000304@eu.citrix.com> (raw)
In-Reply-To: <4969A57D.2020708@redhat.com>
Avi Kivity wrote:
> Trolle Selander wrote:
>> These patches are "extracted" from a patch-set I made for Xen. Any
>> xen-specific parts have been removed. The patches do away with the
>> statically defined VGA_RAM_ SIZE and provides a new -videoram option
>> to enable setting the amount of video ram available to the VM.
>> Shrinking default video ram used to 4 Megs (which is all that the
>> default Cirrus card can use, anyway - previously, there's been a
>> "waste" of 4 megs with the statically defined 8 Megs), but allowing
>> users to specify up to 16 Megs of vram, to support the very high
>> resolutions such as 2560x1600x32 that are available on today's
>> high-end displays.
>> The vgabios patch is needed for this to work, both for the variable
>> size video ram, and to add the new resolutions. The defines for the
>> high resolution and wide-screen modes were taken from what is already
>> in qemu-kvm, so getting this into qemu mainline should get qemu/kvm &
>> xen "in sync" in this regard.
>>
>> Special thanks to Stefano Stabellini for comments & constructive
>> criticism, and some save/restore fixes on the Xen version of this
>> patch-set.
>>
>> I'll brace myself for reports of how this patch breaks all the
>> non-x86 architectures that qemu emulates... :)
>> @@ -178,7 +178,7 @@
>> to store the VM snapshots */
>> DriveInfo drives_table[MAX_DRIVES+1];
>> int nb_drives;
>> -static int vga_ram_size;
>> +static int vga_ram_size = 4 * 1024 * 1024;
>> enum vga_retrace_method vga_retrace_method = VGA_RETRACE_DUMB;
>> DisplayState display_state;
>> int nographic;
>
> You're changing the default, which will break existing setups.
>
My bad. That was actually a small xen "leftover", where the previous
default for stdvga didn't matter, since VBE had never been enabled and
the 8 megs allocated for it by default had never actually been used, so
it was "safe" to break the default there. The probably shouldn't be a
"global" default now anyway, but a separate default per emulated card -
current 4 for Cirrus, and 8 for stdvga & vmware. I suppose it would be
possible to also add the ability to set cirrus down to 2 megs for cirrus
too - pretty much everything is already in the code to emulate the 2 meg
model anyway.
All things considered, I consider it rather encouraging that no one has
found anything other than memory defaults & limits to remark on so far. :)
-- Trolle
next prev parent reply other threads:[~2009-01-12 17:43 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-01-09 19:45 [Qemu-devel] [PATCH] [RFC] Variable video ram size option Trolle Selander
2009-01-10 0:02 ` Paul Brook
2009-01-10 1:26 ` Trolle Selander
2009-01-10 6:45 ` Blue Swirl
[not found] ` <4967E7D7.4040605@eu.citrix.com>
[not found] ` <200901091716.00827.paul@codesourcery.com>
2009-01-10 1:27 ` Trolle Selander
[not found] ` <4967ECA7.6040809@eu.citrix.com>
[not found] ` <200901091747.06462.paul@codesourcery.com>
2009-01-10 1:28 ` Trolle Selander
2009-01-10 1:47 ` Jamie Lokier
2009-01-11 7:53 ` Avi Kivity
2009-01-12 17:43 ` Trolle Selander [this message]
2009-01-21 18:07 ` [Qemu-devel] [PATCH] [RFC] Variable video ram size option - revised Trolle Selander
2009-01-21 18:23 ` Daniel P. Berrange
2009-01-21 21:51 ` Anthony Liguori
2009-01-21 21:21 ` [Qemu-devel] " Trolle Selander
-- strict thread matches above, loose matches on Subject: below --
2009-01-12 19:13 [Qemu-devel] [PATCH] [RFC] Variable video ram size option Trolle Selander
2009-01-14 20:07 ` Paul Brook
2009-01-14 20:41 ` Trolle Selander
2009-01-14 21:30 ` Paul Brook
2009-01-14 22:47 ` Trolle Selander
2009-01-15 13:08 ` Jamie Lokier
2009-01-15 15:37 ` Rene Rebe
2009-01-15 16:23 ` Jamie Lokier
2009-01-15 17:36 ` Gerd Hoffmann
2009-01-15 20:57 ` Jamie Lokier
2009-01-15 21:41 ` Gerd Hoffmann
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=496B8135.4000304@eu.citrix.com \
--to=trolle.selander@eu.citrix.com \
--cc=avi@redhat.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).