Bug 581927 - MULI-GPU FireMV 2400 missing pciid + dual screen issue
Summary: MULI-GPU FireMV 2400 missing pciid + dual screen issue
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Fedora
Classification: Fedora
Component: xorg-x11-drv-ati
Version: 13
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: X/OpenGL Maintenance List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-04-13 15:58 UTC by Carsten Clasohm
Modified: 2011-06-27 15:32 UTC (History)
4 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2011-06-27 15:32:40 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
X log from startx (17.47 KB, text/plain)
2010-04-13 15:59 UTC, Carsten Clasohm
no flags Details
X configuration (667 bytes, text/plain)
2010-04-13 16:00 UTC, Carsten Clasohm
no flags Details
RPM patch file (SOURCES/firemv-2400.patch) (4.21 KB, patch)
2010-04-13 16:01 UTC, Carsten Clasohm
no flags Details | Diff
.spec diff (606 bytes, patch)
2010-04-13 16:02 UTC, Carsten Clasohm
no flags Details | Diff
xorg.conf for two separate displays (1.41 KB, text/plain)
2010-04-14 08:44 UTC, Carsten Clasohm
no flags Details
X log for two separate displays (93.56 KB, text/plain)
2010-04-14 08:44 UTC, Carsten Clasohm
no flags Details
xdpyinfo for two separate displays (33.29 KB, text/plain)
2010-04-14 08:44 UTC, Carsten Clasohm
no flags Details
xorg.conf for Xinerama with two monitors (1.44 KB, text/plain)
2010-04-14 08:45 UTC, Carsten Clasohm
no flags Details
X log for Xinerama with two monitors (69.18 KB, text/plain)
2010-04-14 08:45 UTC, Carsten Clasohm
no flags Details
xdpyinfo for Xinerama with two monitors (17.19 KB, text/plain)
2010-04-14 08:45 UTC, Carsten Clasohm
no flags Details
xorg.conf trying to get a 2x2 stretched display (1.60 KB, text/plain)
2010-04-26 14:11 UTC, Gerben Welter
no flags Details
Xorg config for triple-head on firemv 2400 (2.80 KB, text/plain)
2010-11-26 11:19 UTC, Matthew Hall
no flags Details

Description Carsten Clasohm 2010-04-13 15:58:58 UTC
Description of problem:

The radeon driver fails to start with "No devices detected" with an ATI FireMV 2400 PCIE 256M.


Version-Release number of selected component (if applicable):

xorg-x11-drv-ati-6.13.0-1.fc13.i686


How reproducible:

always


Steps to Reproduce:
1. Run system-config-display and select the radeon driver.
2. Run startx.
  
Actual results:

startx aborts with "(EE) No devices detected."


Expected results:

X should start.


Additional info:

For Fedora 12, I did a simple patch to add chip ID 3155. The .patch file and the diff for the .spec is attached. This seemed to work with a single screen. I did have problems with multiple screens though, which is the main use case for a FireMV card.

# lspci -v -d 1002:
03:00.0 VGA compatible controller: ATI Technologies Inc Device 3155 (rev 80) (prog-if 00 [VGA controller])
	Subsystem: ATI Technologies Inc Device 3155
	Flags: bus master, fast devsel, latency 0, IRQ 10
	Memory at e0000000 (32-bit, prefetchable) [size=256M]
	I/O ports at 2100 [size=256]
	Memory at f0200000 (32-bit, non-prefetchable) [size=64K]
	[virtual] Expansion ROM at f0220000 [disabled] [size=128K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00
	Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit+
	Capabilities: [100] Advanced Error Reporting

03:00.1 Display controller: ATI Technologies Inc Device 3175 (rev 80)
	Subsystem: ATI Technologies Inc Device 3154
	Flags: bus master, fast devsel, latency 0
	Memory at f0210000 (32-bit, non-prefetchable) [disabled] [size=64K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00

04:00.0 Display controller: ATI Technologies Inc Device 3155 (rev 80)
	Subsystem: ATI Technologies Inc Device 3155
	Flags: bus master, fast devsel, latency 0, IRQ 5
	Memory at d0000000 (32-bit, prefetchable) [disabled] [size=256M]
	I/O ports at 1100 [disabled] [size=256]
	Memory at f0100000 (32-bit, non-prefetchable) [disabled] [size=64K]
	[virtual] Expansion ROM at f0120000 [disabled] [size=128K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00
	Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit+
	Capabilities: [100] Advanced Error Reporting

04:00.1 Display controller: ATI Technologies Inc Device 3175 (rev 80)
	Subsystem: ATI Technologies Inc Device 3154
	Flags: bus master, fast devsel, latency 0
	Memory at f0110000 (32-bit, non-prefetchable) [disabled] [size=64K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00

Comment 1 Carsten Clasohm 2010-04-13 15:59:55 UTC
Created attachment 406279 [details]
X log from startx

Comment 2 Carsten Clasohm 2010-04-13 16:00:13 UTC
Created attachment 406280 [details]
X configuration

Comment 3 Carsten Clasohm 2010-04-13 16:01:12 UTC
Created attachment 406282 [details]
RPM patch file (SOURCES/firemv-2400.patch)

Comment 4 Carsten Clasohm 2010-04-13 16:02:29 UTC
Created attachment 406283 [details]
.spec diff

Comment 5 Carsten Clasohm 2010-04-14 08:43:15 UTC
I did another test with a patched xorg-x11-drv-ati-6.13.0-1.fc13.i386, adding the chip ID 3155.

This works with one monitor.

If I attach two monitors to plugs 1 and 2, I get two separate displays with the attached xorg.conf-plugs-1-2, but I cannot move the mouse from monitor 1 to monitor 2. The same happens when I use plugs 1 and 3 (Bus ID 3:0:0 and 4:0:0).

When I configure Xinerama with the attached xorg.conf-xinerama, I get a screen size of 6720x1600, with one 1280x1024 and one 1600x1200 monitor connected. Gnome panel is displayed on the left (1280x1024 Dell) screen, and the Computer/Home/Trash icons are displayed on the right (1600x1200 Eizo) screen. Again, I cannot move the mouse from screen 1 to screen 2. The whole 6720x1600 seems to be displayed on the left screen, and I can scroll around by moving the mouse to the screen edges.

Comment 6 Carsten Clasohm 2010-04-14 08:44:00 UTC
Created attachment 406433 [details]
xorg.conf for two separate displays

Comment 7 Carsten Clasohm 2010-04-14 08:44:25 UTC
Created attachment 406434 [details]
X log for two separate displays

Comment 8 Carsten Clasohm 2010-04-14 08:44:45 UTC
Created attachment 406435 [details]
xdpyinfo for two separate displays

Comment 9 Carsten Clasohm 2010-04-14 08:45:11 UTC
Created attachment 406436 [details]
xorg.conf for Xinerama with two monitors

Comment 10 Carsten Clasohm 2010-04-14 08:45:28 UTC
Created attachment 406437 [details]
X log for Xinerama with two monitors

Comment 11 Carsten Clasohm 2010-04-14 08:45:46 UTC
Created attachment 406438 [details]
xdpyinfo for Xinerama with two monitors

Comment 12 Jérôme Glisse 2010-04-19 07:18:49 UTC
I am not sure we can fix this issue, firemv are multi GPU card and we don't have the infrastructure to support multi-gpu yet. My guess for the mouse issue is that you plug each screen on different GPU and only the cursor of one of it is activated.

Comment 13 Dave Airlie 2010-04-19 09:21:41 UTC
I'm going to add pci ids to the F13 kernel + ati driver to enable kms,

kernel-2.6.33.2-53.fc13
xorg-x11-drv-ati-6.13.0-2

should contain the PCI IDs necessary.

In theory you should get kms and it should boot both cards and a simple xorg.conf should work okay.

Comment 14 Gerben Welter 2010-04-26 14:06:51 UTC
Dave,

I'm working with Carsten to get this working. We've tried your updated kernel, but KMS doesn't work. The screen stays black. So we had to disable KMS.

Another part to this problem is that ATI quietly released this card with different chipsets. I've been emailing with Matt Whitney who succesfully setup a quad head configuration using the proprietary driver on RHEL 5.4:

http://matt.whitney-family.org/2009/10/04/quad-monitor-with-an-ati-firemv-2400-pcie/

He has the following pci ids:

1002:3151
1002:3171

and we have:

1002:3155
1002:3175

Modinfo fglrx shows that our pci ids aren't supported and won't load. Having tried several other fglrx versions, it seems that no version of fglrx supports this card revision. So much for Linux support on the box. Adding the pci ids to the the source that is available will make fglrx load, but will bail out again finding no supported hardware it seems.

This morning we tried the radeon driver patched with our pci ids (your rpm wasn't available yet) and we got all four monitors to light up in a 2x2 configuration, but there were a couple of problems:

The top left screen uses a virtual screen of double the witdh and scrolls to display the part of the top right screen. The top right screen is correct. The two bottom screens work, but are in clone mode, So the bottom right part is missing.

It seems that 'Option "Clone" "Off"' only applies to the primary card/chip.

Not sure what is going on with the top two monitors.

So how do we go from here? Log a bugreport with the radeon driver to get the multi-head sorted out or isn't that even an option in the current state of the driver? Atleast the driver is driving all outputs.

Comment 15 Gerben Welter 2010-04-26 14:11:11 UTC
Created attachment 409182 [details]
xorg.conf trying to get a 2x2 stretched display

Comment 16 Matthew Hall 2010-11-26 11:19:13 UTC
Created attachment 463064 [details]
Xorg config for triple-head on firemv 2400

Triple head mode config (attached) works in F11 and F14. PCIID 1002:3151 still not in drm_pciids.h (so DRI is still disabled), but seems to work ok without it. Only issue is that the config represents 2 screens, one screen spanning 2 displays, and a second screen on 1 display (which means the cursor can cross screens, but not windows).

Comment 17 Bug Zapper 2011-06-02 15:26:04 UTC
This message is a reminder that Fedora 13 is nearing its end of life.
Approximately 30 (thirty) days from now Fedora will stop maintaining
and issuing updates for Fedora 13.  It is Fedora's policy to close all
bug reports from releases that are no longer maintained.  At that time
this bug will be closed as WONTFIX if it remains open with a Fedora 
'version' of '13'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version prior to Fedora 13's end of life.

Bug Reporter: Thank you for reporting this issue and we are sorry that 
we may not be able to fix it before Fedora 13 is end of life.  If you 
would still like to see this bug fixed and are able to reproduce it 
against a later version of Fedora please change the 'version' of this 
bug to the applicable version.  If you are unable to change the version, 
please add a comment here and someone will do it for you.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events.  Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

The process we are following is described here: 
http://fedoraproject.org/wiki/BugZappers/HouseKeeping

Comment 18 Bug Zapper 2011-06-27 15:32:40 UTC
Fedora 13 changed to end-of-life (EOL) status on 2011-06-25. Fedora 13 is 
no longer maintained, which means that it will not receive any further 
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of 
Fedora please feel free to reopen this bug against that version.

Thank you for reporting this bug and we are sorry it could not be fixed.


Note You need to log in before you can comment on or make changes to this bug.