[q45] displayport monitors not detected

Bug Description

Binary package hint: xserver-xorg-video-intel

I'm using Ubuntu 8.10. I have a new Dell Optiplex 960 - this machine has a built in Intel graphics controller with a Dell displayport digital output, in addition to a normal VGA output. If I connect the PC to my Dell 2408WFP monitor via the displayport, everything proceeds fine up to the point where X tries to start, at this point the Intel video driver fails to detect any connected monitors and bombs out. Ubuntu responds by re-starting X with the VESA driver, which works (although slowly and with resolution below that which I want to use). If I connect the monitor via the VGA port, everything works fine - X boots successfully using the Intel driver (although I would obviously prefer to use the displayport connection for quality reasons). The PC/Monitor combination works fine using displayport if i boot in Windows.

I tried compiling the latest version of xf86-video-intel from source (V 2.6.99.1) - the problem does exist in this latest version of the driver too.

xorg.conf is not changed from the standard Ubuntu

[lspci]
00:00.0 Host bridge [0600]: Intel Corporation 4 Series Chipset DRAM Controller [8086:2e10] (rev 03)
Subsystem: Dell Device [1028:0276]
00:02.0 VGA compatible controller [0300]: Intel Corporation 4 Series Chipset Integrated Graphics Controller [8086:2e12] (rev 03)
Subsystem: Dell Device [1028:0276]

Hi junkmail-invades,

Please attach the output of `lspci -vvnn`, and attach your /var/log/Xorg.0.log (and maybe Xorg.0.log.old) file from after reproducing this issue. If you've made any customizations to your /etc/X11/xorg.conf please attach that as well.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

lspci -vvnn output attached

Xorg.0.log attached

I observed the same issue with several Dell Optiplex 960 machines and Dell 2408WFP monitors when using 8.10, but Jaunty is also affected.

I can confirm this for the Ultrabase Dockingstation of my Thinkpad X200.
Works perfectly with "the other" OS.

Intrepid with newer drivers (2.5.x ppa) than release.
See https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/304871 I did not install the "pusher" version mentioned there.

same issue here with thinkpad W500 it was working fine almost 45 days ago but not anymore with ubuntu although works with vista

in Xorg.0.log i noticed this "invalid output device for dac detection"

attached part of the log when i start to mirror the screen.

btw during the boot i see dual screen working on laptop and the external ThinkVision L220x but the moment login screen of ubuntu shows it disabled the external monitor although ubuntu's boot logo shows

I tried the LiveCD of Kubuntu Jaunty x64 Beta:

same problem
As soon as the xorg takes over black screen.

Eduard, I'm sorry, I should have had you try a daily build cd: there
are significant fixes which weren't included on beta1. It would be
great if you could download the current daily livecd from
http://cdimage.ubuntu.com/daily-live/current/ and retest. (Kubuntu vs
ubuntu shouldn't make any difference here).

Hi Carey,
no change with the daily build.... black screen

I CAN start in with Displayport in VESA-mode. BUT: the resolution there is 1680x1050 - no use for my 1920x1200 monitor ...
Better VESA digital than analog INTEL me thinks. But for the native resolution of my monitor there is ONLY analog INTEL.

I have attached the lspci and Xorg.0.log taken under VESA-driver running.

Bashar, judging from your xorg.log, I don't think you're seeing the same bug. Could you file it as a new bug, and attach your xorg.conf, xorg.0.log and lspci -vvnn data? Feel free to link to it from here.

Using stock Intrepid 8.10 Intel driver on a Lenovo X301 (Intel GMA X4500MHD) I can use the Displayport (with both mirrored and virtual monitor) however with the intel X drivers and compiz I get all sorts of missed re-draw events when windows launch. When windows launch they will be partially drawn. 'Wiping' another window over the area fixes the damage.

This redrawing problem disappears if I switch back to the RGB cable (same BENQ 20" monitor) with compiz, or if I use the Displaylink cable with metacity. How can I tell if this is an intel driver bug or a compiz bug?

Nothing in the Xlog.0.log, acts like no errors occurred (version number and hardware snipets below).

Aaron.

X.Org X Server 1.5.2
Release Date: 10 October 2008
X Protocol Version 11, Revision 0
Build Operating System: Linux 2.6.24-15-server x86_64 Ubuntu
Current Operating System: Linux shim 2.6.27-11-generic #1 SMP Wed Apr 1 20:53:41
UTC 2009 x86_64
Build Date: 09 March 2009 01:06:41PM
xorg-server 2:1.5.2-2ubuntu3.1 (<email address hidden>)
[...]
(--) PCI:*(0@0:2:0) Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf0000000/4194304, 0xd0000000/268435456, I/O @ 0x00001800/8
(--) PCI: (0@0:2:1) Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf0400000/1048576
[...]
(II) LoadModule: "intel"
(II) Loading /usr/lib/xorg/modules/drivers//intel_drv.so
(II) Module intel: vendor="X.Org Foundation"
compiled for 1.5.2, module version = 2.4.1
Module class: X.Org Video Driver
ABI class: X.Org Video Driver, version 4.1
(II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G,
E7221 (i915), 915GM, 945G, 945GM, 945GME, 965G, G35, 965Q, 946GZ,
965GM, 965GME/GLE, G33, Q35, Q33,
Mobile Intel® GM45 Express Chipset,
Intel Integrated Graphics Device, G45/G43, Q45/Q43

Note that DisplayPort is not yet implemented in the master branch of the xf86-video-intel driver. That is surely the cause of these problems.

There is a display-port branch which I am using now and seems to be quite stable:
http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/log/?h=display-port

You will have to git-clone that and bootstrap the build by running autogen.sh

Following Jedediah's hint I've built deb packages compiled from display-port branch of the driver's source and uploaded them to my PPA: https://launchpad.net/~morgwai/+archive/ppa (note that i386 binary packages are still waiting in the build queue and should be available within few hours).
They work ok for me (I can normally use monitor connected via Display-Port) and I hope they will be useful for others as well, but I haven't tested them on anything else than my own machine (ThinkPad X200 with docking station, Jaunty amd64) and I'm really not a device driver expert at all, so please don't ask me questions other than related to packaging (reporting whether it works fine for you is of course welcome as it may be useful to others).

I've compiled the sources from the last commit that didn't required much hacking: http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/commit/?h=display-port&id=cc50de1a9eada785323f2dcc3a59e8e94284e17e
It was the last commit in the display-port before merge with master branch that incorporated commit http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/commit/?h=display-port&id=1784efc1f0c74edda8c78fb1b0cdd989f0a088ad which made it impossible to turn DRM_MODE off and caused the compilation to fail on Jaunty. I don't really understand why this fails as the commit's comment says that this flag was only relevant for xserver older than 1.6 and Jaunty uses exactly 1.6. If somebody knows the reason and can prepare the patches that would allow to build the head of display-port branch on Jaunty then I'll be happy to include them and build next version of the driver.

Cheers

Morg

I've using the DisplayPort successfully on my Lenovo X301 (Intel integrated graphics) with Jaunty (stock release). I'm using it via the Lenovo (i.e. Foxconn) DisplayPort-to-DVI cable though. Am I really using DisplayPort then? I had thought the cable does an active signal conversion so it is DisplayPort signally. I guess I am surprised it works if the Intel driver doesn't support DisplayPort.

Anyway, with the converter the DisplayPort works great with Jaunty. It mostly worked with Intrepid too, but I used to get this phantom third monitor in the resolution control panel. Jaunty has cleaned that up too.

Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf0000000/4194304, 0xd0000000/268435456, I/O @ 0x00001800/8

X.Org X Server 1.6.0
Release Date: 2009年2月25日
X Protocol Version 11, Revision 0
Build Operating System: Linux 2.6.24-15-server x86_64 Ubuntu
Current Operating System: Linux shim 2.6.28-11-generic #42-Ubuntu SMP Fri Apr 17
01:58:03 UTC 2009 x86_64
Build Date: 09 April 2009 02:11:54AM
xorg-server 2:1.6.0-0ubuntu14 (<email address hidden>)

launchpad should really have an option to turn single bug comment into a question ;)
I don't know if this cable can do some magic so the driver uses dvi code path, but the best way to find out would be to connect monitor via pure display port and see if it works then.

Cheers

Morg

Piotr,

you made my day!!!

Thank you! My Ultrabase just became more useful with your driver.

Any idea how to enable UXA?

II) Module exa: vendor="X.Org Foundation"
compiled for 1.6.0, module version = 2.4.0
ABI class: X.Org Video Driver, version 5.0
(II) intel(0): Comparing regs from server start up to After PreInit
(WW) intel(0): Register 0x61110 (PORT_HOTPLUG_EN) changed from 0x10000120 to 0x38000120
(WW) intel(0): Register 0x61114 (PORT_HOTPLUG_STAT) changed from 0x10100000 to 0x38100000
(WW) intel(0): Register 0x70024 (PIPEASTAT) changed from 0x00000206 to 0x80000206
(WW) intel(0): PIPEASTAT before: status: VSYNC_INT_STATUS SVBLANK_INT_STATUS VBLANK_INT_STATUS
(WW) intel(0): PIPEASTAT after: status: FIFO_UNDERRUN VSYNC_INT_STATUS SVBLANK_INT_STATUS VBLANK_INT_STATUS
(WW) intel(0): Register 0x321b (FBC_FENCE_OFF) changed from 0x4b035b00 to 0x3800ba00
(WW) intel(0): Register 0x64110 (DPB_AUX_CH_CTL) changed from 0x00050000 to 0x5d450085
(WW) intel(0): Register 0x64210 (DPC_AUX_CH_CTL) changed from 0x00150085 to 0x4c150085
(WW) intel(0): Register 0x64310 (DPD_AUX_CH_CTL) changed from 0x00050000 to 0x5d450085
(==) Depth 24 pixmap format is 32 bpp
(II) do I need RAC? No, I don't.
(II) resource ranges after preInit:
[0] -1 0 0xffffffff - 0xffffffff (0x1) MX[B]
[1] -1 0 0x000f0000 - 0x000fffff (0x10000) MX[B]
[2] -1 0 0x000c0000 - 0x000effff (0x30000) MX[B]
[3] -1 0 0x00000000 - 0x0009ffff (0xa0000) MX[B]
[4] 0 0 0x000a0000 - 0x000affff (0x10000) MS[B](OprD)
[5] 0 0 0x000b0000 - 0x000b7fff (0x8000) MS[B](OprD)
[6] 0 0 0x000b8000 - 0x000bffff (0x8000) MS[B](OprD)
[7] -1 0 0x0000ffff - 0x0000ffff (0x1) IX[B]
[8] -1 0 0x00000000 - 0x00000000 (0x1) IX[B]
[9] 0 0 0x000003b0 - 0x000003bb (0xc) IS[B](OprU)
[10] 0 0 0x000003c0 - 0x000003df (0x20) IS[B](OprU)
(II) intel(0): Kernel reported 739328 total, 1 used
(II) intel(0): I830CheckAvailableMemory: 2957308 kB available
(WW) intel(0): DRI2 requires UXA

OK:

Option "AccelMethod" "uxa"