Bug 12048

Summary: Xpress 200M (RC410, 5a62) gives black screen when DRI is enabled
Product: xorg Reporter: Brice Goglin <brice.goglin>
Component: Driver/RadeonAssignee: xf86-video-ati maintainers <xorg-driver-ati>
Status: RESOLVED FIXED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium CC: avillaci, jacob, wberrier
Version: git   
Hardware: Other   
OS: All   
URL: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=437321
Whiteboard:
i915 platform: i915 features:
Bug Depends on:    
Bug Blocks: 11749    
Attachments:
Description Flags
Patch: Disable DRI by default for RS482 cards
none
Disable DRI by default for RS482 cards
none
rename to PCI_CHIP_RS485_5975 everywhere
none
Result of running revenge-1.0 on Radeon XPRESS 200M none

Description Brice Goglin 2007-08-18 01:32:31 UTC
Bug reported by Zack Weinberg on the Debian BTS 2 weeks ago, discussed with airlied on IRC a couple times.

Since 6.6.192 and kernel 2.6.22, the radeon driver enables DRI on
  PCI:*(1:5:0) ATI Technologies Inc RC410 [Radeon Xpress 200M] rev 0
It results in a black screen for Zack.

The log with 6.6.193 and DRI is available at: http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=42;filename=xorg.log.6_6_193-1_dri;att=6;bug=437321
With DRI disabled in xorg.conf: http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=42;filename=xorg.log.6_6_193-1_no-dri;att=5;bug=437321

With AIGLX disabled (requested by airlied):
http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=52;filename=xorg.log.6_6_193-1_no-aiglx;att=1;bug=437321

And finally, airlied requested the dmesg output when debug is enabled in the drm kernel module:
http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=62;bug=437321

I am marking as a blocker for 6.7 in case it's easy to fix. But, this "regression" may be easily worked around by disabling DRI, feel free to unblock if there's too much work needed to fix this...
Comment 1 Alex Villacís Lasso 2007-09-20 14:56:26 UTC
I also have the exact same symptoms:

01:05.0 VGA compatible controller: ATI Technologies Inc RC410 [Radeon Xpress 200] (prog-if 00 [VGA])
        Subsystem: Intel Corporation Unknown device d600
        Flags: bus master, 66MHz, medium devsel, latency 64, IRQ 21
        Memory at d8000000 (32-bit, prefetchable) [size=128M]
        I/O ports at ee00 [size=256]
        Memory at fdef0000 (32-bit, non-prefetchable) [size=64K]
        [virtual] Expansion ROM at fde00000 [disabled] [size=128K]
        Capabilities: <access denied>

Testing on Fedora Core 7, all packages updated as of 2007-09-20. All gits mentioned updated as of 2007-09-20.

Compiled libdrm and mesa from git and put symlinks (after backing up original files), and also compiled current git for xf86-video-ati . On startup, screen blanks out (LED on monitor turns from green to amber, like a powersaving mode).  Keyboard does not respond to Ctrl-Alt-Backspace. Machine is still responsive, since I can ssh from another location and kill X. At least X can return to text mode on kill.
Comment 2 Alex Deucher 2007-09-20 15:39:43 UTC
I suspect we need to make adjustments to the memory controller when the DRI is enabled.  Hopefully we'll get some info from AMD soon.
Comment 3 Tormod Volden 2007-10-03 05:07:06 UTC
From https://bugs.launchpad.net/bugs/139241 (and duplicates) it seems like the [1002:5975] card has to have DRI disabled as well.
Comment 4 Tormod Volden 2007-10-03 11:00:25 UTC
Created attachment 11888 [details] [review]
Patch: Disable DRI by default for RS482 cards
Comment 5 Tormod Volden 2007-10-03 11:04:59 UTC
Created attachment 11889 [details] [review]
Disable DRI by default for RS482 cards
Comment 6 Tormod Volden 2007-10-05 16:27:38 UTC
Created attachment 11910 [details] [review]
rename to PCI_CHIP_RS485_5975 everywhere

Ok, I mean RS485 cards. They are just called RS482_5975 in the code.

Since bugzilla refuses me to open a new bug for that, I'll attach a patch for renaming it to RS485_5975 here. The other patch has to corrected accordingly.
Comment 7 Alex Deucher 2007-10-05 17:11:51 UTC
committed:
314e8fdbe5369e4c1786d176d03d0d2a046f25d5
fe66cc4e02a52e1d014ae5de192c65e8abd7f58a
Comment 8 Wade Berrier 2007-10-08 15:49:18 UTC
I'm seeing this on suse 10.3 with xorg 7.2.  With gateway laptop ml3109.

lspci:

01:05.0 VGA compatible controller: ATI Technologies Inc RC410 [Radeon Xpress 200M] (prog-if 00 [VGA])
        Subsystem: Gateway 2000 Unknown device 0318
        Flags: bus master, 66MHz, medium devsel, latency 66, IRQ 17
        Memory at d0000000 (32-bit, prefetchable) [size=256M]
        I/O ports at 9000 [size=256]
        Memory at c0000000 (32-bit, non-prefetchable) [size=64K]
        [virtual] Expansion ROM at c0020000 [disabled] [size=128K]
        Capabilities: [50] Power Management version 2
        Capabilities: [80] Message Signalled Interrupts: Mask- 64bit- Queue=0/0 Enable-

More than a blank screen though, my keyboard doesn't respond (no caps lock light).  But, I can still ssh to the laptop...
Comment 9 Alex Villacís Lasso 2007-11-21 09:33:17 UTC
With fglrx driver 8.42.3 the ATI Technologies Inc RC410 [Radeon Xpress 200] integrated chip in my machine works correctly with DRI enabled (but only at 1280x1024, and the screen is capable of at least 1400x1050). Is there anything I can do on my machine that will provide information on how to enable DRI correctly on the open-source driver? 
Comment 10 Alex Villacís Lasso 2007-11-21 09:54:34 UTC
Created attachment 12671 [details]
Result of running revenge-1.0 on Radeon XPRESS 200M

This is the result of running revenge 1.0 on my machine: ATI Technologies Inc RC410 [Radeon Xpress 200]. Beryl is running but with metacity as window manager.
Comment 11 Alex Deucher 2008-12-03 00:09:32 UTC
The DRI should work on that chip now.

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.