Re: Bug#818035: libva not working
On 2 July 2016 at 18:51, Nicholas D Steeves <nsteeves@gmail.com> wrote:
> On 2 July 2016 at 08:26, Sebastian Ramacher <sramacher@debian.org> wrote:
>> On 2016-05-23 10:33:46, Jürgen Bausa wrote:
>>> Am Sonntag, 22. Mai 2016, 17:46:28 schrieb Sebastian Ramacher:
>>>
>>> On clean stretch it works. So it must be some other packages as you already
>>> suspected.
>>>
>>> Wouldnt this be a reason to pull i965-va-driver and its dependencies from
>>> stretch for the next jessie release, as the current package does not work?
>>
>> I'm not sure it's suitable for jessie-updates,- it'd be a new feature and not a
>> bug fix. So maybe it'd be more suitable for a backport. Since there is already a
>> backport of libdrm, backports of libva and intel-vaapi-driver should be enough.
>> CCing d-backports@l.d.o to check whether somebody is interested in maintaining
>> backports for libva and intel-vaapi-driver.
>
> I might be interested as I'm already running locally backported
> libva-1.7.1 (needed for smooth fast-forwarding using mpv on Haswell).
> I'll give a intel-vaapi-driver bpo a try later today, and see how it
> goes. That said I am not interested in testing it against the latest
> linux-image bpo, because I track upstream linux-4.4.x. Does this
> disqualify me from maintaining a backport of libva and
> intel-vaapi-driver?
Done. I've been testing my libva-1.7.1 bpo since June 22. I
backported intel-gpu-tools and intel-vaapi-driver yesterday, and so
far everything is 100% ok on my Haswell system. I'm ready to push
libva and intel-vaapi-driver to to Alioth, and I need a sponsor for
intel-gpu-tools. Sebastian, could you please Re: that .gitignore
question in the thread in deb multimedia packagers?
Cheers,
Nicholas
Reply to: