[PC-BSD Testing] Blue men Fix
pcbsd-testing-l at jelal.kn-bremen.de
Sat Nov 17 07:22:21 PST 2012
[And again I forgot to use the subscribed address...]
Ok I got it working...
On Wed, Nov 14, 2012 at 11:28:04PM +0100, Juergen Lock wrote:
> I got a reply by Nvidia's Andy Ritger, he is not subscribed so cannot
> post himself but I quote his reply here:
> On Wed, Nov 14, 2012 at 10:59:51AM -0800, Andy Ritger wrote:
> > Hi Juergen,
> > I assume the issue reported here is the Flash U/V plane swap bug worked
> > around by this commit:
> > http://cgit.freedesktop.org/~aplattner/libvdpau/commit/?id=ca9e637c61e80145f0625a590c91429db67d0a40
> > The latest Linux libvdpau library built/shipped by NVIDIA should contain
> > that fix, and you should also be able to build libvdpau yourself from
> > the source here:
> > http://cgit.freedesktop.org/~aplattner/libvdpau/
> > I don't know how the PC-BSD nvidia-driver ports gets its libvdpau,
> > hopefully the above gives enough information to help sort this out.
> > Let me know if there is anything I can do to help.
> > Thanks,
> > - Andy Ritger
> [end quote]
> I think PCBSD gets its Linux libvdpau from the FreeBSD nvidia driver
> like FreeBSD itself, like in the version I tested that's currently in
> FreeBSD ports,
> If this is fixed in the latest version all is fine, then just the
> FreeBSD port,
> and possibly PCBSD needs to be updated, I've Cc'd the port maintainer.
I meanwhile tested the linux libvdpau from NVIDIA-FreeBSD-x86_64-310.19.tar.gz
with the rest of the 304.60 port and can confirm the issue is fixed
(or should I say worked around) there, the running-in-flash detection
of the workaround need linprocfs mounted tho.
So all that's missing now is an update to the nvidia-driver port
More information about the Testing