Ilia Mirkin
2017-Aug-21 12:34 UTC
[Nouveau] VDPAU issues in 17.2 and master, and a test request
Hi guys, I just wanted to record some of my observations with nouveau; I haven't had time to debug what happened. All testing done with mplayer -vo vdpau. First off, it seems like master + DRI3 works, up to a point, but then crashes in nouveau pushbuf logic with a missing kref (which means some unknown buffer is being used ... somewhere). Now, in 17.2-rcN (sorry, I forget which one, but recent), neither DRI2 nor DRI3 work. I believe they don't work differently, but both fail. And on master, DRI2 also does not work (which I did by commenting out the dri3 vl screen creation, although one could also achieve this by not having a DRI3-capable X server, which is the default for nouveau). The symptom of "not working" is mostly rejected pushbuf submissions with "unknown handle 0", although sometimes (perhaps consistently in one of the above configuratoins) it's the output that's completely off (some purple-ish thing). As I don't have the hardware myself, I was hoping someone could confirm that this is a nouveau issue and not a more general one. Ideally this would be done by testing VDPAU on both radeonsi and r600, as well as both the DRI3 and the DRI2 paths. Please use mplayer for this, not mpv or mplayer2 or anything else -- all those have various differences in how the rendering is done from mplayer, which could account for a different final result. Thanks, Ilia Mirkin imirkin at alum.mit.edu
Andy Furniss
2017-Aug-21 21:54 UTC
[Nouveau] [Mesa-dev] VDPAU issues in 17.2 and master, and a test request
Ilia Mirkin wrote:> As I don't have the hardware myself, I was hoping someone could > confirm that this is a nouveau issue and not a more general one. > Ideally this would be done by testing VDPAU on both radeonsi and r600, > as well as both the DRI3 and the DRI2 paths. Please use mplayer for > this, not mpv or mplayer2 or anything else -- all those have various > differences in how the rendering is done from mplayer, which could > account for a different final result.Current svn mplayer + git ffmpeg, vdpau vo is OK for me on radeonsi. Option "DRI3" "off" in amdgpu.conf for ddx is also OK.
Ilia Mirkin
2017-Aug-21 21:56 UTC
[Nouveau] [Mesa-dev] VDPAU issues in 17.2 and master, and a test request
On Mon, Aug 21, 2017 at 5:54 PM, Andy Furniss <adf.lists at gmail.com> wrote:> Ilia Mirkin wrote: > >> As I don't have the hardware myself, I was hoping someone could >> confirm that this is a nouveau issue and not a more general one. >> Ideally this would be done by testing VDPAU on both radeonsi and r600, >> as well as both the DRI3 and the DRI2 paths. Please use mplayer for >> this, not mpv or mplayer2 or anything else -- all those have various >> differences in how the rendering is done from mplayer, which could >> account for a different final result. > > > Current svn mplayer + git ffmpeg, vdpau vo is OK for me on radeonsi. > Option "DRI3" "off" in amdgpu.conf for ddx is also OK.Thanks, Andy. What version of mesa was this with?
Possibly Parallel Threads
- [Mesa-dev] VDPAU issues in 17.2 and master, and a test request
- [Mesa-dev] VDPAU issues in 17.2 and master, and a test request
- [Bug 86006] New: [NV84] Nvidia GeForce 8600 GT VDPAU h264 hardware acceleration
- nouveau driver locks up with 4.11 kernel
- NOUVEAU(0): DRI3 on EXA enabled