Bug 2273 - IBM T40P + Radeon 9000 M9 + Dual head
Summary: IBM T40P + Radeon 9000 M9 + Dual head
Status: RESOLVED DUPLICATE of bug 3621
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/Radeon (show other bugs)
Version: git
Hardware: x86 (IA32) Linux (All)
: high normal
Assignee: Xorg Project Team
QA Contact:
URL:
Whiteboard:
Keywords:
: 2367 (view as bug list)
Depends on:
Blocks:
 
Reported: 2005-01-13 08:04 UTC by michael sommers
Modified: 2005-09-22 12:54 UTC (History)
5 users (show)

See Also:
i915 platform:
i915 features:


Attachments
crtc-output separation (35.43 KB, patch)
2005-02-19 15:15 UTC, Alex Deucher
no flags Details | Splinter Review

Description michael sommers 2005-01-13 08:04:55 UTC
I've been trying to operate both the laptop panel + the external DVI for a while
now without any success.  I can work either the laptop panel or the DVI port
separately but not both of them together.   After some games as explained below
X finally detects both heads but is not able to operate both of them together.

I have looked through all the bugs I could find with similar issues and have
tried plenty of various configuration changes without luck. 

Versions tried: XFree 4.3 (debian version) / 4.4 (website version) and now Xorg
6.8.1 /  6.8.1 with Hui's patch (from bug 1559) / cvs head from today.

XFree behaves pretty much like an unpatched 6.8.1, while 6.8.1 and the cvs head
versions produce about the same results.

Basically HEAD and any version with Huj's patch produce much better results but
there seems to be one last problem, the monitor itself does not get detected for
some reason, attached are Xorg.log and my config.

Since I don't know how to attach a file here I will only show excerpts of the
Xorg.log that may be relevant.

(II) RADEON(0): 
(II) RADEON(0): Primary:
 Monitor   -- TMDS
 Connector -- DVI-D
 DAC Type  -- TVDAC/ExtDAC
 TMDS Type -- NONE
 DDC Type  -- NONE
(II) RADEON(0): Secondary:
 Monitor   -- CRT
 Connector -- VGA
 DAC Type  -- Primary
 TMDS Type -- NONE
 DDC Type  -- VGA_DDC
(II) RADEON(0): PLL parameters: rf=2700 rd=12 min=20000 max=35000; xclk=20000
(II) RADEON(0): DFP table revision: 3
(WW) RADEON(0): Failed to detect secondary monitor DDC, default HSync and VRefre
sh used


I have tried both
Option "MonitorLayout"             "TMDS,AUTO"
 
and 

Option "MonitorLayout"             "TMDS,LVDS"

both produce the same problem, though slightly different text in the log excerpt
above.  both cases do not detect the second monitor.

Any input would be greatly appreciated.   I will respond to any query and
provide any details necessary.

Thanks!
Comment 1 Alex Deucher 2005-01-13 10:20:05 UTC
This configuration is currently not supported.  The problem is many LVDS + TMDS
solutions use both the internal LVDS and internal TMDS controllers.  The driver
always uses crtc1 for the internal LVDS and internal TMDS and crtc2 for external
tmds.  Your configuration would need crtc1 driving LVDS and crtc2 drving
internal TMDS.  The radeon driver needs to be restructured so that either crtc
can driver any output.  It's on my todo list, but I don't know when I'll get to it.
Comment 2 Alex Deucher 2005-01-28 14:37:22 UTC
*** Bug 2367 has been marked as a duplicate of this bug. ***
Comment 3 Alex Warg 2005-02-01 01:20:54 UTC
I have a small patch that enables support for DVI and the internal LCD
on my T40p. All you have to do is to specify the option "MonitorLayout" to
"LVDS, TMDS" and use the MergedFB feature.
I can switch among the pseudo-Xinerama-modes and clone-modes on the fly.

The patch here:
http://os.inf.tu-dresden.de/~aw11/xorg/radeon_t40p.diff.gz
Comment 4 Alex Deucher 2005-02-19 15:15:29 UTC
Created attachment 1943 [details] [review]
crtc-output separation

can you try this patch?  you'll need to use:
Option "monitorlayout" "LVDS, TMDS"
Comment 5 Alex Deucher 2005-02-19 15:17:33 UTC
(In reply to comment #4)
> Created an attachment (id=1943) [edit]
> crtc-output separation
> 
> can you try this patch?  you'll need to use:
> Option "monitorlayout" "LVDS, TMDS"
> 

I also have binaries here:
elfloader:
http://www.botchco.com/alex/xorg/radeon_drv.o
dllloader:
http://www.botchco.com/alex/xorg/radeon_drv.so
Comment 6 Alex Deucher 2005-02-19 22:35:48 UTC
latest patch can be found here:
http://www.botchco.com/alex/xorg/crtc_to_output.patch
Comment 7 Alex Deucher 2005-02-20 16:56:44 UTC
updated patch

https://bugs.freedesktop.org/show_bug.cgi?id=2273
Comment 8 Erik Slagter 2005-05-18 02:42:08 UTC
I have the same problem.

Using xorg 6.8.2(-30), Ati Radeon Mobility M300, Dell Inspiron 9300.

I can attach a VGA monitor with my current setup and use it fine (MonitorLayout
"LVDS, CRT").

If I attach a DVI monitor, with most combinations of options I don't get a
picture at the DVI monitor at all, with some I get a picture only at the
external monitor, never at both.

In text mode, it sort of works, I can switch between the internal screen and the
external monitor using FN+F8.

I have tried the binary module mentioned here, the result is that I don't even
get a picture on the internal screen, with all combinations of MonitorLayout. I
don't know if it drives the external DVI, I don't have access to one at the
moment. IMHO it's not that important, as not even the laptop's panel works with
this module. I reverted to the mainstream module.
Comment 9 Erik Slagter 2005-06-02 03:36:12 UTC
Is there still any development going on in this area?

I now have my own DVI monitor, so I can test at will. I also consider putting
the support in myself, but for that I'll need some support I guess.
Comment 10 Alex Deucher 2005-06-02 05:53:16 UTC
(In reply to comment #9)
> Is there still any development going on in this area?
> 
> I now have my own DVI monitor, so I can test at will. I also consider putting
> the support in myself, but for that I'll need some support I guess.

I don't have any time to mess with this at the moment, but if you want to work
on it, I'd be happy to answer any questions.
Comment 11 Erik Slagter 2005-06-05 04:14:23 UTC
Ok.

BTW I found a bug in radeon_macros.h 1.5 -> 1.6 (6_8_99_8 -> 6_8_99_9), the
changed definition of a certain macro (see also #3369 which it's supposed to
fix) breaks the detection of my second screen completely (even with
MonitorLayout enabled). So I am using 1.5 for "development". 
Comment 12 Alex Deucher 2005-06-25 02:12:23 UTC
please try the patch on bug 3621 and report your results there.
Comment 13 T. Hood 2005-09-23 05:54:58 UTC

*** This bug has been marked as a duplicate of 3621 ***


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.