On 07/11/2018 04:07 AM, Peter Robinson wrote:
On Tue, Jul 10, 2018 at 9:51 PM, Cole Robinson
<crobinso(a)redhat.com> wrote:
> On 07/10/2018 04:22 PM, Zbigniew Jędrzejewski-Szmek wrote:
>> On Tue, Jul 03, 2018 at 10:14:43PM -0700, Thomas Daede wrote:
>>> On 07/03/2018 05:15 AM, Jan Kurik wrote:
>>>> Move to uEFI as the default boot mechanism for ARMv7 devices.
>>>
>>> Will this work with virt-manager too? Currently, while aarch64 boots via
>>> uEFI there, it seems that armv7 is only supported by manually specifying
>>> a kernel and initrd.
>>
>> Ping.
>>
>
> Currently this won't work with virt-manager.
Yes, I'm aware.
> If this arm32 stuff works similar to aarch64, then what we need is:
> extend the edk2 package to build working arm32 images, some packaging
> changes to libvirt, and some straightforward-but-not-trivial changes to
> virt-manager to match.
My intention here was to look at the aarch64 support and copy that and
submit a patch. I would love help and I intended to reach out but have
been thumped with other escalations and problems so haven't had the
chance yet.
There is already an edk2-arm subpackage [1] in the Fedora builds and
it appears to work with my very basic testing.
> Gerd Hoffman has a nightly edk2 repo which builds arm32 images. If
> someone can figure out the magic qemu incantation to get those working
> with Fedora, similar to how aarch64 works, I can do the rest
Awesome, will reach out to work with you, thank you!
FYI I've filed a bug to track virt-manager work, so if anyone is
interested jump on the CC list
https://bugzilla.redhat.com/show_bug.cgi?id=1613996
Thanks,
Cole