[Libosinfo] [PATCH v2] installer: API to query device driver signing requirement

[Zeeshan Ali (Khattak)]

On Fri, Mar 15, 2013 at 4:06 PM, Zeeshan Ali (Khattak)
<zeeshanak at gnome.org> wrote:
> On Fri, Mar 15, 2013 at 11:05 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
>> On Thu, Mar 14, 2013 at 06:27:59PM +0200, Zeeshan Ali (Khattak) wrote:
>>> From: "Zeeshan Ali (Khattak)" <zeeshanak at gnome.org>
>>>
>>> Some OS vendors recommend or require device drivers to be signed by them
>>> before these device drivers could be installed on their OS. This
>>> recommendation/requirement then also applies to the installer scripts of
>>> that particular OS.
>>>
>>> This API enables apps to query such requirement/recommendation, if any.
>>
>> ACK.
>> For what it's worth, the InstallScript level really is the wrong level for
>> this kind of things as multiple install scripts are typically needed to
>> install a single OS, and this API acts on a single install script. As a
>> library user, I don't really care about that level of detail, and having
>> something global to all install scripts needed to install an OS would be
>> plenty enough.
>
> I see your point but this is actually going to happen on InstallConfig
> level and I imagine apps (just like Boxes) will use the same config
> for all scripts of the same profile of the same OS. What does puts me
> off a bit is the fact that same profile scripts of the same OS are not
> really glued together, so it can easily trick app devs in believing
> that they are independent.
>
> I guess this is where the libvirt-builder will be useful, putting all
> these together for apps so they don't need to use these (rather
> low-level) API. I'm hoping to work on that as GNOME 3.10/Fedora 20
> feature.

Meh, I thought this was "install-config,winxp,win7: API to
enable/disable driver signing". However, some of the things I said
above still apply.

-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124