On 5/11/2026 8:36 PM, Philippe Mathieu-Daudé wrote:
> On 11/5/26 23:07, Pierrick Bouvier wrote:
>> On 5/11/2026 1:46 PM, Cédric Le Goater wrote:
>>> On 5/11/26 19:59, Pierrick Bouvier wrote:
>>>> On 5/11/2026 10:48 AM, Philippe Mathieu-Daudé wrote:
>>>>> On 11/5/26 17:48, Cédric Le Goater wrote:
>>>>>> Hello,
>>>>>>
>>>>>> On 5/6/26 15:55, Philippe Mathieu-Daudé wrote:
>>>>>>> Introduce a source set common to system / user. Start it
>>>>>>> with the files built in both sets: 'cpu_models_user.c'
>>>>>>> and 'gdbstub.c' No logical change intended.
>>>>>>>
>>>>>>> Signed-off-by: Philippe Mathieu-Daudé <[email protected]>
>>>>>>> Reviewed-by: Ilya Leoshkevich <[email protected]>
>>>>>>> Message-Id: <[email protected]>
>>>>>>
>>>>>> This change introduced a regression with PCI passthrough which
>>>>>> stopped working. Guest kernel reports :
>>>>>>
>>>>>>      [    0.156501] zpci: PCI is not supported because CPU
>>>>>> facilities 69
>>>>>> or 71 are not available
>>>>>>
>>>>>> and other devices (ap, ccw) are impacted too I think.
>>>>>
>>>>> Is it something we could test with the mainstream CI, or does this
>>>>> requires specific hardware?
>>>>>
>>>>
>>>> It could be tested with a tcg arm vm + nested VM with PCI passthrough.
>>>
>>> Yes, adding VFIO tests to the QEMU upstream CI would be great.
>>>
>>>> However, the problem is that you need to cross-compile QEMU in CI to
>>>> test current version to launch the nested VM.
> 
> I guess remember one of your scripts close to that setup, mounting
> the same directory in nested guest with:
> 
>   -virtfs \
>    local,path=$(pwd),mount_tag=host,security_model=mapped,readonly=off
> 
> Same could be used with s390x / x86 I suppose.

Yes, it's another possibility.
I just like the idea of generating the image on the fly, and not have to
host it/update it anywhere. Considering this, it's "free" to include a
binary at creation vs a mount network.

For reference, this is the script to create an ext4 image from container
[1], and the custom init booting [2] and running a custom command.

[1] https://github.com/p-b-o/qemu-linux-stack/blob/master/build_rootfs.sh
[2] https://github.com/p-b-o/qemu-linux-stack/blob/master/rootfs/common/init

The scripts in themselves are not important, but it shows how small they
are, and thus, easy to develop and maintain.

Reply via email to