Mailinglist Archive: radeonhd (408 mails)

[Rafał Miłecki <zajec5@xxxxxxxxx>]

2009/6/4 David C. Rankin <drankinatty@xxxxxxxxxxxxxxxxxx>:
> On Wednesday 03 June 2009 03:27:22 Rafał Miłecki wrote:
> > 2009/5/31 David C. Rankin <drankinatty@xxxxxxxxxxxxxxxxxx>:
> > >        Here is the fatal server error experienced on logout with the
> > > downclock branch:
> > >
> > > (II) RADEONHD(0): Modeline "1440x900"x0.0   96.31  1440 1504 1536 1760
> > >  900 903 906 912 -hsync -vsync (54.7 kHz)
> > >
> > > Backtrace:
> > > 0: /usr/bin/X(xorg_backtrace+0x26) [0x4efd76]
> > > 1: /usr/bin/X(xf86SigHandler+0x39) [0x49be99]
> > > 2: /lib/libc.so.6 [0x7fdc985f8ec0]
> > > 3: /usr/lib/xorg/modules/extensions//libglx.so [0x7fdc97ae0318]
> > > 4:
> > > /usr/lib/xorg/modules/dri/r300_dri.so(__driUtilUpdateDrawableInfo+0xc1)
> > > [0x7fdc8d5af611]
> > > 5: /usr/lib/xorg/modules/dri/r300_dri.so(radeonGetLock+0x57)
> > > [0x7fdc8d5b55e7] 6:
> > > /usr/lib/xorg/modules/dri/r300_dri.so(r300FlushCmdBuf+0x97)
> > > [0x7fdc8d5c0987] 7:
> > > /usr/lib/xorg/modules/dri/r300_dri.so(r300DestroyContext+0x417)
> > > [0x7fdc8d5bad37] 8: /usr/lib/xorg/modules/dri/r300_dri.so
> > > [0x7fdc8d5af3f0]
> > > 9: /usr/lib/xorg/modules/extensions//libglx.so [0x7fdc97ae1bf9]
> > > 10: /usr/lib/xorg/modules/extensions//libglx.so(__glXFreeContext+0x6c)
> > > [0x7fdc97ad830c]
> > > 11: /usr/lib/xorg/modules/extensions//libglx.so [0x7fdc97ad86ab]
> > > 12: /usr/bin/X(CallCallbacks+0x44) [0x451834]
> > > 13: /usr/bin/X(CloseDownClient+0x58) [0x448008]
> > > 14: /usr/bin/X(Dispatch+0x1f8) [0x44d838]
> > > 15: /usr/bin/X(main+0x3bd) [0x43389d]
> > > 16: /lib/libc.so.6(__libc_start_main+0xfd) [0x7fdc985e555d]
> > > 17: /usr/bin/X [0x432d29]
> > >
> > > Fatal server error:
> > > Caught signal 11.  Server aborting
> >
> > What do you actually mean by downclock branch? Is there one?
>
>        (laughing) Well, I mean whatever I pulled and compiled when you showed 
> me how
> to use git to pull in the additional packages to test the downclocking and
> apply the patches and edit the file to enable it ;-)
>
>        Whether that is a git branch, twig or stick -- I just dunno??

Uh, OK, I gone totally wrong while reading this thread :)

My experimental code for downclocking memory doesn't seem to work well
in your case. It's enought to mention this patch of your log:
> (II) RADEONHD(0): Attempting to set Memory Clock to 100000
> (II) RADEONHD(0): Current Memory Clock: 0

It's weird for me that my code cause crash... wouldn't expect that.
Lock up - OK, but crashing sounds weird.

Anyway, if it doesn't happen with standard (engine only) downclocking,
just ignore this. Downclocking memory is done wrong and especially in
your case shouldn't be used.

-- 
Rafał Miłecki
--
To unsubscribe, e-mail: radeonhd+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: radeonhd+help@xxxxxxxxxxxx