Bug 15291 - I need option "HDP" "off"
Summary: I need option "HDP" "off"
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/radeonhd (show other bugs)
Version: 7.3 (2007.09)
Hardware: x86-64 (AMD64) Linux (All)
: medium minor
Assignee: Luc Verhaegen
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-03-31 11:58 UTC by Roman Hodek
Modified: 2008-04-07 14:26 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
my xorg.conf (3.22 KB, text/plain)
2008-04-05 08:11 UTC, Roman Hodek
no flags Details
logfile w/o HDP off (28.91 KB, text/x-log)
2008-04-05 08:12 UTC, Roman Hodek
no flags Details
Hardcoded connector table for this device. (1.34 KB, patch)
2008-04-05 14:34 UTC, Egbert Eich
no flags Details | Splinter Review

Description Roman Hodek 2008-03-31 11:58:31 UTC
In section 9.2 of the radeonhd FAQ you request reports if the option "HDP" is required to get a monitor working. Here is one :)

After radeonhd 1.1 had been installed during a Debian update, my X server didn't start anymore. The last (relevant) message was that no valid modes were found, and a few lines above that my output DVI-I_1/analog has been detected as disconnected. A quick look at the wiki revealed what's the problem and how to cure it, no problem so far.

My card is a MSI HD2600 AGP (the fanless one), the BIOS string is "113-MSITV108MS.110 RV630PRO HYNIX 32MX16 512MB". The monitor is a 21'' Eizo, attached via BNC cable (no DDC data!)

Roman
Comment 1 Egbert Eich 2008-03-31 12:33:03 UTC
Roman, could you please provide a log file with -logverbose 7?
Please assign back to the default maintainer when done, thanks.
Comment 2 Egbert Eich 2008-03-31 12:34:09 UTC
Please also run rhd_conntest with the monitor in question attached.
Comment 3 Roman Hodek 2008-04-01 00:25:45 UTC
> Roman, could you please provide a log file with -logverbose 7?

Sure, but I won't have access to that machine until Wednesday evening...

Roman
Comment 4 Roman Hodek 2008-04-05 08:11:54 UTC
Created attachment 15701 [details]
my xorg.conf
Comment 5 Roman Hodek 2008-04-05 08:12:20 UTC
Created attachment 15702 [details]
logfile w/o HDP off
Comment 6 Roman Hodek 2008-04-05 08:22:21 UTC
Sorry, I forgot to produce the requested logfile on Wednesday... but now I've remembered :) The logfile plus my xorg.conf (with HDP off commented out) are attached.

And here's also the rhd_conntest output:

rhd_conntest: v1.1.0, dist of git branch fedora/snapshot, commit 8c6d1b46
Checking connectors on 0x9587, 0x1462, 0x0028  (@01:00:00):
  Load Detection: RHD_OUTPUT_DACB
  HotPlug: RHD_HPD_2
getDDCSpeed: Clock: 600000 Prescale: 0x5dc0
I2CStatus: 1104
getDDCSpeed: Clock: 600000 Prescale: 0x5dc0
I2CStatus: 1104
getDDCSpeed: Clock: 600000 Prescale: 0x5dc0
I2CStatus: 1104
getDDCSpeed: Clock: 600000 Prescale: 0x5dc0
I2CStatus: 1104
  DDC: RHD_DDC_NONE 
Comment 7 Egbert Eich 2008-04-05 14:10:49 UTC
The version of the driver you are using is rather old. Could you update to the current git head and try again, please?
The output of rhd_conntest contradicts the output of the Xserver. The Xserver detects DAC A while rhd_conntest detects DAC B as connected.
I will verify on Monday if rhd_conntest is working correctly with an R6xx card.
Comment 8 Egbert Eich 2008-04-05 14:12:16 UTC
Woops, I mixed up two logs. Sorry.
My bad.
Comment 9 Egbert Eich 2008-04-05 14:34:43 UTC
Created attachment 15708 [details] [review]
Hardcoded connector table for this device.

Roman, could you please get the current git sources apply the attached patch and test again?
Informations on how to build and test should be in http://wiki.x.org/wiki/radeonhd. If it works, please also connect the monitor to the other connector on the card and try if this also works. Chances are that the HPD pin is set wrong here, too. If the latter setup doesn't work, please also provide an rhd_conntest output for it, too.
Comment 10 Egbert Eich 2008-04-06 01:16:50 UTC
Assigning to reporter for further testing. Roman, please assign back to default maintainer when done. Thanks!
Comment 11 Roman Hodek 2008-04-07 11:19:42 UTC
Fine, with the patch it works without the "HDP" "off" again, and also for the second output. So a clear "go" from my side! :)

BTW, I've seen the name "MSI HD2600 AGP" in the table, which is probably what I've written in my starting post. The official full name of the card is "MSI RX2600PRO-T2D512Z/D2", but probably "MSI RX2600PRO AGP" will do (the rest is about video mem, AFAIK).
Comment 12 Egbert Eich 2008-04-07 14:26:36 UTC
Fix pushed.

commit 60dc7e0453167b956a1fb0cc9db89aa22196b1cf
Author: Egbert Eich <eich@freedesktop.org>
Date:   Mon Apr 7 23:23:22 2008 +0200

    Add a hard coded connector table for an MSI HD2600PRO AGP.
    
    Full ID of the card: MSI RX2600PRO-T2D512Z/D2.
    The connector information for this chip lists HPD0 instead of HPD2.

Closing bug.
Roman, thanks for reporting and testing!


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.