From: Fabio Fantoni <fabio.fantoni@m2r.biz>
To: Sander Eikelenboom <linux@eikelenboom.it>
Cc: xen-devel@lists.xensource.com,
Ian Campbell <Ian.Campbell@citrix.com>,
Stefano.Stabellini@eu.citrix.com, George.Dunlap@eu.citrix.com,
Ian.Jackson@eu.citrix.com,
Anthony PERARD <anthony.perard@citrix.com>
Subject: Re: [PATCH] libxl: new parameters for upstream qemu's vga interfaces
Date: Mon, 28 Oct 2013 13:57:47 +0100 [thread overview]
Message-ID: <526E5F4B.6060106@m2r.biz> (raw)
In-Reply-To: <525E8091.4010001@m2r.biz>
Il 16/10/2013 14:03, Fabio Fantoni ha scritto:
> Il 10/10/2013 16:39, Fabio Fantoni ha scritto:
>> Il 10/10/2013 14:36, Sander Eikelenboom ha scritto:
>>> Thursday, October 10, 2013, 2:29:46 PM, you wrote:
>>>
>>>> Il 10/10/2013 12:51, Sander Eikelenboom ha scritto:
>>>>> Thursday, October 10, 2013, 12:09:44 PM, you wrote:
>>>>>
>>>>>> Il 10/10/2013 11:32, Ian Campbell ha scritto:
>>>>>>> On Thu, 2013-10-10 at 11:26 +0200, Fabio Fantoni wrote:
>>>>>>>> Il 30/09/2013 12:12, Fabio Fantoni ha scritto:
>>>>>>>>> Change the qemu parameters for upstream qemu vgas to the
>>>>>>>>> new ones (-device), introduced some years ago.
>>>>>>>> Ping
>>>>>>> Please CC the qemu maintainers (Stefano & Anthony), although these
>>>>>>> patches touch the toolstack they are logically qemu patches and
>>>>>>> need
>>>>>>> their ack.
>>>>>> Thanks for your reply, I added Anthony on CC and Stefano was
>>>>>> already on it.
>>>>> Should "None" be added as option as well ?
>>>>>
>>>> There is already a nographic xl parameter that controls the
>>>> corresponding qemu parameter, it should be the same thing.
>>>> And FWIK the none option applies only to the old -vga parameter and
>>>> according to qemu docs/qdev-device-use.txt, the new way to do it is
>>>> with
>>>> -nodefaults (I already made a patch to add it). -nographic is probably
>>>> also deprecated.
>>> In this conversation
>>> http://comments.gmane.org/gmane.comp.emulators.qemu/172385
>>> the -nographic is said to be deprecated by peter maydell.
>>> Didn't know -vga none was also out the door already ...
>
>>
>> I did some test and I found out that now is impossible disable the
>> emulated vga on hvm domUs.
>> I got cirrus vga (the default) even if vga xl parameter is not setted.
>> Also with nographic enabled there is the emulated vga.
>> I think that good solution is:
>> - add this patch and nodefault patch
> update: nodefault patch is already on git
>> - add "none" option to vga xl parameter (that will exclude any
>> emulated vga qemu parameters)
>> - remove -nographic parameter in upstream qemu, making nographic xl
>> parameter deprecated and equivalent to new vga="none" xl parameter.
>>
>> With these changes all should be working and without qemu deprecated
>> parameters (-nographic and -vga).
>> I'm waiting Stefano and/or Anthony replies before write the patches.
>>
>> Another question is about xenfb vga that seems missed on new -device
>> parameter.
>> I used it to have basic Spice support for pv working:
>> http://lists.xen.org/archives/html/xen-devel/2013-09/msg03207.html
>> Anyone can update about it on newer qemu versions please?
>>
>> Thanks for any reply
>
> Ping...
Another ping...
>
>>
>>>>>>>>> Signed-off-by: Fabio Fantoni <fabio.fantoni@m2r.biz>
>>>>>>>>> ---
>>>>>>>>> tools/libxl/libxl_dm.c | 12 +++++-------
>>>>>>>>> 1 file changed, 5 insertions(+), 7 deletions(-)
>>>>>>>>>
>>>>>>>>> diff --git a/tools/libxl/libxl_dm.c b/tools/libxl/libxl_dm.c
>>>>>>>>> index 43c3bec..2c6f5d9 100644
>>>>>>>>> --- a/tools/libxl/libxl_dm.c
>>>>>>>>> +++ b/tools/libxl/libxl_dm.c
>>>>>>>>> @@ -486,15 +486,13 @@ static char **
>>>>>>>>> libxl__build_device_model_args_new(libxl__gc *gc,
>>>>>>>>> switch (b_info->u.hvm.vga.kind) {
>>>>>>>>> case LIBXL_VGA_INTERFACE_TYPE_STD:
>>>>>>>>> - flexarray_vappend(dm_args, "-vga", "std", NULL);
>>>>>>>>> + flexarray_append_pair(dm_args, "-device", "VGA");
>>>>>>>>> break;
>>>>>>>>> case LIBXL_VGA_INTERFACE_TYPE_CIRRUS:
>>>>>>>>> - flexarray_vappend(dm_args, "-vga", "cirrus", NULL);
>>>>>>>>> - if (b_info->video_memkb) {
>>>>>>>>> - flexarray_vappend(dm_args, "-global",
>>>>>>>>> - GCSPRINTF("vga.vram_size_mb=%d",
>>>>>>>>> - libxl__sizekb_to_mb(b_info->video_memkb)), NULL);
>>>>>>>>> - }
>>>>>>>>> + flexarray_append_pair(dm_args, "-device",
>>>>>>>>> "cirrus-vga");
>>>>>>>>> + flexarray_append_pair(dm_args, "-global",
>>>>>>>>> + GCSPRINTF("vga.vram_size_mb=%d",
>>>>>>>>> + libxl__sizekb_to_mb(b_info->video_memkb)));
>>>>>>>>> break;
>>>>>>>>> }
>>>>>
>>>>>
>>>
>>>
>>
>
next prev parent reply other threads:[~2013-10-28 12:57 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-30 10:12 [PATCH] libxl: new parameters for upstream qemu's vga interfaces Fabio Fantoni
2013-10-10 9:26 ` Fabio Fantoni
2013-10-10 9:32 ` Ian Campbell
2013-10-10 10:09 ` Fabio Fantoni
2013-10-10 10:43 ` Stefano Stabellini
2013-10-10 11:58 ` Fabio Fantoni
2013-10-10 10:51 ` Sander Eikelenboom
2013-10-10 12:29 ` Fabio Fantoni
2013-10-10 12:36 ` Sander Eikelenboom
2013-10-10 14:39 ` Fabio Fantoni
2013-10-16 12:03 ` Fabio Fantoni
2013-10-28 12:57 ` Fabio Fantoni [this message]
2013-10-28 16:45 ` Stefano Stabellini
2013-10-29 8:51 ` Fabio Fantoni
2013-11-04 17:52 ` Ian Campbell
2013-11-06 17:09 ` 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=526E5F4B.6060106@m2r.biz \
--to=fabio.fantoni@m2r.biz \
--cc=George.Dunlap@eu.citrix.com \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=Stefano.Stabellini@eu.citrix.com \
--cc=anthony.perard@citrix.com \
--cc=linux@eikelenboom.it \
--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.