Date: Wed, 14 Nov 2012 16:56:18 -0500 From: Justin Hibbits <chmeeedalf@gmail.com> To: matt <sendtomatt@gmail.com> Cc: Andreas Tobler <andreast-list@fgznet.ch>, x11@freebsd.org, freebsd-ppc@freebsd.org Subject: Re: PPC DRM graphics testing Message-ID: <20121114165618.18affd74@narn.knownspace> In-Reply-To: <50A1CA6D.40907@gmail.com> References: <50972E9E.3010101@gmail.com> <50974ECD.5010702@fgznet.ch> <50988FE0.9030806@gmail.com> <50989EA0.5020509@fgznet.ch> <5098CA4F.7020306@gmail.com> <509A8B3D.8030703@fgznet.ch> <509A9D11.6010605@gmail.com> <509A9FD4.4050403@fgznet.ch> <509ACFAA.8050802@fgznet.ch> <509CA2FF.7070902@gmail.com> <509D7BC5.1020301@fgznet.ch> <50A008D5.2060003@fgznet.ch> <50A1CA6D.40907@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 12 Nov 2012 20:19:57 -0800 matt <sendtomatt@gmail.com> wrote: > On 11/11/12 12:21, Andreas Tobler wrote: > > On 09.11.12 22:55, Andreas Tobler wrote: > >> On 09.11.12 07:30, matt wrote: > >>> On 11/07/12 13:16, Andreas Tobler wrote: > >>>> So. Newer ati packages did not improve. I then decided to start > >>>> from a clean base, no debug Xorg, no debug radeon driver etc. I > >>>> cleaned, deinstalled and reinstalled first the xorg-server then > >>>> the dri and afterwards the same for the xf86-video-ati. The > >>>> order of dri and ati does not matter I think. Having done this, > >>>> I was able to run the 3D-demos.... yeah! The kernel has WITTNES > >>>> and friends commented, also DRM_DEBUG is off. Here the excerpts > >>>> from the Xorg.log: X.Org X Server 1.7.7 Module radeon: compiled > >>>> for 1.7.7, module version = 6.14.3 Andreas > >>> It did not work here :( > >>> I am using WITH_NEW_XORG, could be that > >> That is one difference, don't know how big/important. I will try > >> to play with WITH_NEW_XORG. > > For the record. I updated to NEW_ORG, portmaster -a after adding > > WITH_NEW_XORG to /etc/make.conf. Additionally I had to rebuild the > > video, keyboard and input drivers. > > > > > > Started X, everthing seems fine. I could run the glx demos but the > > content was crap. There was some animation but with the glxgears > > demo I couldn't identify the gears itself. I only saw some rotating > > rays. Updated to mesa-demos 8.0.1, no improvement. > > > > At least no hardlock. > > > > I'll see how I can dig into. > > > > Andreas > > > "OLD_XORG" causes an immediate hard lock with games/minetest (I wanted > to rule out mesa-demos as it seems to compile its own version of mesa > vs libGL?) I cannot SSH in or do anything except reboot. I assume you > are also running CURRENT. If that is the case, I think we can say > works on hardware (g5 agp) for certain, and it doesn't seem to work > so far on the one mdd g4 tested. I am upgrading to WITH_NEW_XORG > again, because it got further than an immediate hardlock in OLD_XORG > for me. I did not install xfce this time, just minetest and blackbox. > I tried xorg-server with AIGLX off as well as on too. > > So possibilities: > Radeon 9000 has some hardware quirk > G4 MDD has some hardware quirk > My radeon 9000 is broken > My G4 MDD is broken > Remotely possible CPU=7450 or -O2 is breaking things? > > Matt > I have another data point: G4 DPMDD, Radeon 9600Pro, Xorg server 1.7.7 Xorg segfaults on startup: (II) XINPUT: Adding extended input device "Keyboard0" (type: KEYBOARD) Segmentation fault: 11 at address 0x90 Fatal server error: Caught signal 11 (Segmentation fault: 11). Server aborting Unfortunately, gdb(1) is useless for backtracing, because it doesn't support threads on ppc32. - Justin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20121114165618.18affd74>