Bug#585777: Radeon gamma bug, display too bright when using VGA output

[Alex Deucher]

On Tue, Apr 26, 2011 at 3:56 AM, Andrew Rule <a.rule at freeuk.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi,
>
> I also have this problem.  It appeared for me when I upgraded from Xorg
> 1.7.x to Xorg 1.9.5
>
> I am using
> "ATI Radeon Mobility 9200 (M9+) 5C63 (AGP)" (ChipID = 0x5c63)
> as X reports it, and
> Linux amos 2.6.38.3 #2 SMP PREEMPT Sat Apr 16 22:03:19 BST 2011 i686
> GNU/Linux
> from kernel.org
>
> I tried setting Gamma in /etc/X11/xorg.conf, but it didn't pick it up.
> If I run xgamma after logging in to X, it *does* change the gamma - but
> still not enough.  So I can set Gamma to 0.3 in the file, but if I set
> it using xgamma, it actually darkens the screen somewhat - but it is
> still not right.  Output from xgamma:
> - -> Red  0.300, Green  0.300, Blue  0.300
> <- Red  0.300, Green  0.300, Blue  0.300
>

It's not the gamma, but the dac adj values.

>
> I tried
> radeontool regset DAC_MACRO_CNTL 0x00000808
> but it didn't work for me.  (Perhaps that is not a surprise, I am using
> a tower PC, not a laptop.)

You are probably using the secondary dac.  The dac adjust values for
the TV dac are in
TV_DAC_CNTL.

>
> Perhaps I should not mention this here, but X is also unstable for me
> under KDE now, so I'm using sawfish for the time being.  Specifically,
> it crashes when notifications popup in KDE.
>
> Another problem I have noticed is that when scrolling a window under
> another window, the contents of the lower window smudges.  O, and popup
> hints from firefox come up in the wrong place for the search box.
>
> And, as you can see from the attached files, I am not using KMS -
> because I found that unstable since upgrading the kernel from 2.6.36.1
> to 2.6.38.x (although previously the virtual terminals had not worked at
> all, and now they do occasionally; but it was stable under X, now it is
> not.)

What is unstable with KMS?  KMS is more likely to be fixed that UMS at
this point.  If you want to fix UMS, you'll probably have to bisect
xf86-video-ati to find out what change caused the breakage.  Your time
would be better off spend bisecting to see what caused the instability
with KMS.

Alex

>
> Please ask further questions if you need.  I'm gonna try downgradeing X,
> cos I'd like to have a screen I can look at pictures on :-)
>
> Sincerely,
>
> Andrew Rule
>
> PS please tell me if I should open a new bug for some of this, I wasn't
> sure.  But the gamma stuff seemed so similar I thought it might be the
> same bug.
>
> PPS: package versions:
> firmware-linux/testing uptodate 0.29
> libc6/testing uptodate 2.11.2-11
> libdrm-radeon1/testing upgradeable from 2.4.23-3 to 2.4.24-2
> libdrm2/testing upgradeable from 2.4.23-3 to 2.4.24-2
> libpciaccess0/testing uptodate 0.12.1-1
> libpixman-1-0/testing uptodate 0.21.4-2
> libudev0/testing upgradeable from 166-1 to 167-3
> xserver-xorg-core/testing uptodate 2:1.9.5-1
> xserver-xorg-video-radeon/testing uptodate 1:6.14.1-1
>
> I upgraded to:
> firmware-linux/testing uptodate 0.29
> libc6/testing uptodate 2.11.2-11
> libdrm-radeon1/testing uptodate 2.4.24-2
> libdrm2/testing uptodate 2.4.24-2
> libpciaccess0/testing uptodate 0.12.1-1
> libpixman-1-0/testing uptodate 0.21.4-2
> libudev0/testing upgradeable from 166-1 to 167-3
> xserver-xorg-core/testing uptodate 2:1.9.5-1
> xserver-xorg-video-radeon/testing uptodate 1:6.14.1-1
>
> with no obvious change, certainly not to the gamma issue.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.11 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>
> iEYEARECAAYFAk22epYACgkQAXpg5UuOD0UtzwCgmLNnarwPog1J+v6TmvAwOA3Q
> PC8AnjW6/ejWvwW7ewVIIMq7MZWs+J72
> =NH8l
> -----END PGP SIGNATURE-----
>
> _______________________________________________
> xorg-driver-ati mailing list
> xorg-driver-ati at lists.x.org
> http://lists.x.org/mailman/listinfo/xorg-driver-ati
>
>