Beware of nvidia-graphics 1.0-7664

Axel Thimm Axel.Thimm at ATrpms.net
Tue Jun 7 12:13:23 UTC 2005


On Tue, Jun 07, 2005 at 01:56:17PM +0200, Nils Philippsen wrote:
> On Mon, 2005-06-06 at 16:14 +0200, Axel Thimm wrote:
> > On Mon, Jun 06, 2005 at 03:41:58PM +0200, Nils Philippsen wrote:
> > > On Mon, 2005-06-06 at 11:22 +0200, Axel Thimm wrote:
> > > > On Mon, Jun 06, 2005 at 10:24:08AM +0200, Nils Philippsen wrote:
> > > > > Note that this driver version broke suspend/resume for me (oh the joys
> > > > > of proprietary software with a proprietary development model), the
> > > > > display gets garbled. After some dabbling with it I found a workaround,
> > > > > though: Setting the display to DPMS suspend, then on again brought the
> > > > > card back into a semi-sane state (X works, text console are still
> > > > > garbled).
> > > > 
> > > > Yes, I had to find that out myself when the screensaver kicked in
> > > > killing two hours of work ... :/
> > > 
> > > Hmm, screensavers work just fine here, I only had problems with
> > > suspend/resume until I implemented the workaround.
> > 
> > My symptoms are display goes black, but never returns. The system works
> > fine though, but a headless laptop is no joy ... :/
> 
> Have you tried the workaround yet?

No, I was bitten before reading this and haven't had the chance to
test it. Also, my xset says I have DPMS suspend on, so I'm probably
already using the workaround. My symptoms were not a garbled display,
but a permanently turned off one, perhaps different bugs altogether.

> > > > Also XvMC libs are broken :/
> > > 
> > > Care to elaborate? I don't see these in the livna RPMs I use, but maybe
> > > they're just missing ;-).
> > 
> > libXvMCNVIDIA_dynamic.so.1 is refencing _nv0019XvMCdynamic which does
> > not exist. You'll only hit it if you use XvMC, e.g. mplayer/mythtv/xine etc.
> 
> Well, mplayer doesn't use libXvMC here:
> 
> nils at gibraltar:~> rpm -q mplayer --requires | grep -i xvmc
> nils at gibraltar:~>

It probably depends on your mplayer rpm. mplayer -vo help lists the
output devices:

$ mplayer -vo help
MPlayer 1.0pre7-3.4.3 (C) 2000-2005 MPlayer Team
CPU: Intel Pentium M Banias (Family: 6, Stepping: 5)
Detected cache-line size is 64 bytes
CPUflags:  MMX: 1 MMX2: 1 3DNow: 0 3DNow2: 0 SSE: 1 SSE2: 1


Available video output drivers:
        ivtvosd IVTV OSD
        xvmc    XVideo Motion Compensation
        xv      X11/Xv
        x11     X11 ( XImage/Shm )
        xover   General X11 driver for overlay capable video output
	drivers
        gl      X11 (OpenGL)
        gl2     X11 (OpenGL) - multiple textures version
        dga     DGA ( Direct Graphic Access V2.0 )
        sdl     SDL YUV/RGB/BGR renderer (SDL v1.1.7+ only!)
        aa      AAlib
        vesa    VESA VBE 2.0 video output
        xvidix  X11 (VIDIX)
        cvidix  console VIDIX
        null    Null video output
        mpegpes Mpeg-PES file
        ivtv    IVTV-Mpeg file
        yuv4mpeg        yuv4mpeg output for mjpegtools
        png     PNG file
        jpeg    JPEG file
        tga     Targa output
        pnm     PPM/PGM/PGMYUV file
        md5sum  md5sum of each frame


> Perhaps you shouldn't build mplayer against nvidia's libraries?

You don't have a real choice on a fanless low-CPU-power system
relaying on the GPU's acceleration. That's what XvMC is about. :(

There is a wrapper lib that can be used to divert the XvMC
implementation at runtime. That would protect against broken (unused)
libs.
-- 
Axel.Thimm at ATrpms.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/fedora-devel-list/attachments/20050607/b983b79d/attachment.sig>


More information about the fedora-devel-list mailing list