Bug 51012 - [LGA 1155 motherboards] Multi-Monitor does NOT work when 1 monitor uses Intel GPU and another monitor uses Nvidia GPU -- (Sandy Bridge & Ivy Bridge iGPU)
Summary: [LGA 1155 motherboards] Multi-Monitor does NOT work when 1 monitor uses Intel...
Status: RESOLVED MOVED
Alias: None
Product: xorg
Classification: Unclassified
Component: Server/General (show other bugs)
Version: git
Hardware: All Linux (All)
: medium normal
Assignee: Xorg Project Team
QA Contact: Xorg Project Team
URL:
Whiteboard: 2012BRB_Reviewed
Keywords:
Depends on:
Blocks:
 
Reported: 2012-06-12 09:55 UTC by Malcolm
Modified: 2018-12-13 22:26 UTC (History)
6 users (show)

See Also:
i915 platform:
i915 features:


Attachments
lspci (2.17 KB, text/plain)
2012-06-14 12:13 UTC, Malcolm
no flags Details
Xorg.0.log (44.88 KB, text/plain)
2012-06-14 12:18 UTC, Malcolm
no flags Details

Description Malcolm 2012-06-12 09:55:52 UTC
This bug was originally filed in the Fedora Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=830619

-----

https://en.wikipedia.org/wiki/LGA_1155

The above article lists all the LGA 1155 motherboards which "Allows
using built-in GPU" also known as Integrated GPU / iGPU. But basically
it is talking about the Integrated GPU which comes with the Intel
Sandy Bridge and Ivy Bridge microprocessors. As you can see, (except
for the lone exception P67), all motherboards support using iGPU along
with a Dedicated GPU card.


It basically allows using the Integrated GPU along with a Dedicated
GPU like Nvidia or ATI.


Using my MSI H67 Motherboard, and Windows 7 was able to successfully
make a multi-monitor set up (with Extended mode, so we can get more
work done faster):
One monitor was connected to the VGA output on the Nvidia GT 520 GPU.
The other monitor was connected to the VGA output on the motherboard
(Core i3 - integrated graphics).

The same multi-monitor setup unfortunately fails in Fedora (edit: xorg? ).

http://www.smolts.org/client/show_all/pub_9b118418-5637-4127-9c4b-506fd6bfb436

I know some of you will suggest getting a new monitor, or getting a
Convertor to convert DVI to VGA or HDMI to VGA.
But that's not the point, which is that Fedora (edit: xorg? ) is not keeping with the technological improvements.
Comment 1 Jeremy Huddleston Sequoia 2012-06-14 01:08:45 UTC
Please post your xorg log.  Please describe what *does* happen rather thaj just 
saying "it doesn't work"
Comment 2 Malcolm 2012-06-14 12:08:31 UTC
(In reply to comment #1)
> Please post your xorg log.  Please describe what *does* happen rather than just 
> saying "it doesn't work"

Most of the information can be found here: https://bugzilla.redhat.com/show_bug.cgi?id=830619  but I will copy and paste it here right now.
Comment 3 Malcolm 2012-06-14 12:10:04 UTC
(In reply to comment #6)
> oh, so is there an extra wrinkle in this case where the two adapters are
> hooked up to an optimus thingy, so it's not just a straightforward case of
> multiple adapter support, but you also need to make switcheroo run both
> adapters?

The both adapters have to be running simultaneously.

I can explain how it worked in Windows 7:

1. The Main Monitor is connected using the Nvidia card's VGA output.

2. After installing Windows 7, download the Nvidia Drivers. Once installed you will see Nvidia icon in system tray.


3. The Motherboard has another VGA output soldiered on it. This is for people who don't have dedicated graphics card. 

4. If a person has a dedicated card this is Disabled by default.

5. Then go into the BIOS settings to enable Multi-Monitor support.

6. Connecting  the Second Monitor to the Motherboard's VGA port will not do anything at this time. 

7. Downloaded either the Intel Graphics Driver from Intel's website, or the VGA drivers that usually are available from the Manufacturer's website (eg. MSI or ASUS etc)

8. Now the second monitor will light up.

9. There will be a new tray icon for Intel graphics settings. 

10. If you unplug the Second monitor this goes away from the system tray.

11. If you unplug the Main monitor, the Nvidia icon stays, but I don't think you can change any settings.

12. When both the monitors are connected. there are 2 icons in the system tray, and you can modify the settings for both: the Intel Graphics connected monitor and Nvidia Graphics connected monitor.
Comment 4 Malcolm 2012-06-14 12:10:15 UTC
> Er, using multiple video adapters simultaneously is hardly some kind of
> new technological improvement. It's been possible for years, probably
> decades.
> 
> X actually used to support it somewhat 'better' prior to RandR 1.2 being
> developed. The new randr stuff focused on supporting multiple outputs
> from a single adapter _really well_, and it certainly did/does that, but
> it compromised on multiple adapter support as a trade-in. I'm never
> quite sure what the current status is on putting multiple adapter
> support back in, but I expect ajax or airlied would know. It may well be
> the case that you actually could get it to work and you just need a
> custom config.

Can someone who has one of the motherboards here https://en.wikipedia.org/wiki/LGA_1155 which supports using Intel's built-in GPU, and a dedicated GPU card like Nvidia, test?

That is, to see if using multiple video adapters simultaneously works in Fedora 17? Please.

PS. My specs right now are:

MSI H67MA-E45 (B3) H67 mATX LGA1155 DDR3 PCI-E16 3PCI-E HDMI DVI Audio GBLAN SATA3 Motherboard 

Intel Core i3 2100 Dual Core Hyperthreading Processor LGA1155 3.1GHZ Sandy Bridge 3MB 

MSI GeForce GT 520 810MHZ 1024MB 1GB DDR3 PCI-E VGA DVI HDMI HDCP Video Card
Comment 5 Malcolm 2012-06-14 12:10:33 UTC
+ this post was made by John Reiser + Thanks.

https://lists.fedoraproject.org/pipermail/devel/2012-June/168574.html

-----



I have an ASUS P8Z68-V/GEN3 main board which can run the Intel HD3000
graphics which is builtin, using a i5 Sandy Bridge CPU.  The box also
has an nVidia GeForce GT 430 card (PCI 10de:0de1 rev a1).
I usually run only one monitor (from the nVidia card), but I decided
to experiment.

The American Megatrends BIOS (version 0301 [original; now old]) has
System Agent Configuration options Initiate [Initial] Graphic Adapter,
and iGPU Multi-monitor.  I set iGPU as the initial monitor, and
iGPU Multi-monitor to Enabled, and connect two monitors: one to the
VGA port of the Intel graphics, one to the VGA port of the nVidia card.
Upon reboot, then both LCD monitors turn on their backlights.
The monitor connected to the Intel VGA port gets the BIOS and GRUB
displays, and becomes the normal monitor.  The monitor connected to the
nVidia VGA port gets no normal display, and after a few seconds the
hardware in the monitor detects "No display" and turns itself off.

The F-17 Gnome3 desktop System Settings > Displays detects only the display
that is connected to the Intel VGA port.  xrandr also shows only the
ports that are associated with the builtin Intel hardware.
/var/log/Xorg.0.log shows only the monitor with the normal display.
Only the VGA ports seem to be active in this two-monitor attempt.
Using anything other than VGA ports (both cards have DVI and HDMI ports, too;
both monitors have VGA and DVI connectors) produces no display on any non-VGA
connection.

One monitor (6 years old; 1280x1024) reports EDID info, the other does not
(9 years old; 1024x768.)  Interchanging monitor<->"card" connections
has no observed effect on behavior.

So, it seems to me that this setup "fails multi-monitor usage"
in the manner that Malcolm Turmel describes.
My box has no other operating system installed, so I cannot say
anything about whether multi-monitor works with any other operating system.

syslog (/var/log/messages) detects the Intel graphics hardware as:
Jun 13 10:24:49 f17e64 kernel: [    1.020600] agpgart-intel 0000:00:00.0: Intel Sandybridge Chipset
Jun 13 10:24:49 f17e64 kernel: [    1.020698] agpgart-intel 0000:00:00.0: detected gtt size: 2097152K total, 262144K mappable
Jun 13 10:24:49 f17e64 kernel: [    1.021529] agpgart-intel 0000:00:00.0: detected 65536K stolen memory
Jun 13 10:24:49 f17e64 kernel: [    1.021610] agpgart-intel 0000:00:00.0: AGP aperture is 256M @ 0xc0000000
Jun 13 10:24:49 f17e64 kernel: [    2.494960] fb: conflicting fb hw usage inteldrmfb vs VESA VGA - removing generic driver
Jun 13 10:24:49 f17e64 kernel: [    2.495798] fbcon: inteldrmfb (fb0) is primary device
Jun 13 10:24:49 f17e64 kernel: [    2.660097] fb0: inteldrmfb frame buffer device
Jun 13 10:24:49 f17e64 kernel: [   11.189735] hda_intel: Disabling MSI

syslog detects the nVidia hardware as:
Jun 13 10:24:49 f17e64 kernel: [    1.501442] [drm] Initialized drm 1.1.0 20060810
Jun 13 10:24:49 f17e64 kernel: [    1.507444] [drm] nouveau 0000:01:00.0: Detected an NVc0 generation card (0x0c1080a1)
Jun 13 10:24:49 f17e64 kernel: [    1.513098] [drm] nouveau 0000:01:00.0: Attempting to load BIOS image from PRAMIN
Jun 13 10:24:49 f17e64 kernel: [    1.522820] [drm] nouveau 0000:01:00.0: ... BIOS signature not found
Jun 13 10:24:49 f17e64 kernel: [    1.522822] [drm] nouveau 0000:01:00.0: Attempting to load BIOS image from PROM
Jun 13 10:24:49 f17e64 kernel: [    1.821809] [drm] nouveau 0000:01:00.0: ... appears to be valid
Jun 13 10:24:49 f17e64 kernel: [    1.821811] [drm] nouveau 0000:01:00.0: BIT BIOS found
Jun 13 10:24:49 f17e64 kernel: [    1.821813] [drm] nouveau 0000:01:00.0: Bios version 70.08.29.00
Jun 13 10:24:49 f17e64 kernel: [    1.821815] [drm] nouveau 0000:01:00.0: TMDS table version 2.0
Jun 13 10:24:49 f17e64 kernel: [    1.822011] [drm] nouveau 0000:01:00.0: MXM: no VBIOS data, nothing to do
Jun 13 10:24:49 f17e64 kernel: [    1.822013] [drm] nouveau 0000:01:00.0: DCB version 4.0
Jun 13 10:24:49 f17e64 kernel: [    1.822015] [drm] nouveau 0000:01:00.0: DCB outp 00: 01000302 00020030
Jun 13 10:24:49 f17e64 kernel: [    1.822016] [drm] nouveau 0000:01:00.0: DCB outp 01: 02000300 00000000
Jun 13 10:24:49 f17e64 kernel: [    1.822017] [drm] nouveau 0000:01:00.0: DCB outp 02: 04011310 00020020
Jun 13 10:24:49 f17e64 kernel: [    1.822019] [drm] nouveau 0000:01:00.0: DCB outp 03: 02022362 00020010
Jun 13 10:24:49 f17e64 kernel: [    1.822020] [drm] nouveau 0000:01:00.0: DCB conn 00: 00001030
Jun 13 10:24:49 f17e64 kernel: [    1.822021] [drm] nouveau 0000:01:00.0: DCB conn 01: 00000100
Jun 13 10:24:49 f17e64 kernel: [    1.822022] [drm] nouveau 0000:01:00.0: DCB conn 02: 00002261
Jun 13 10:24:49 f17e64 kernel: [    1.822036] [drm] nouveau 0000:01:00.0: Adaptor not initialised, running VBIOS init tables.
Jun 13 10:24:49 f17e64 kernel: [    1.822037] [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 0 at offset 0x6BB6
Jun 13 10:24:49 f17e64 kernel: [    1.882705] [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 1 at offset 0x71FC
Jun 13 10:24:49 f17e64 kernel: [    1.909746] [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 2 at offset 0x8391
Jun 13 10:24:49 f17e64 kernel: [    1.909753] [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 3 at offset 0x8395
Jun 13 10:24:49 f17e64 kernel: [    1.909807] [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 4 at offset 0x847D
Jun 13 10:24:49 f17e64 kernel: [    1.909809] [drm] nouveau 0000:01:00.0: Parsing VBIOS init table at offset 0x84E2
Jun 13 10:24:49 f17e64 kernel: [    1.954104] [drm] nouveau 0000:01:00.0: 3 available performance level(s)
Jun 13 10:24:49 f17e64 kernel: [    1.954107] [drm] nouveau 0000:01:00.0: 0: core 50MHz shader 101MHz memory 135MHz timing 0 voltage 880mV
Jun 13 10:24:49 f17e64 kernel: [    1.954109] [drm] nouveau 0000:01:00.0: 1: core 405MHz shader 810MHz memory 324MHz timing 1 voltage 900mV
Jun 13 10:24:49 f17e64 kernel: [    1.954112] [drm] nouveau 0000:01:00.0: 3: core 700MHz shader 1400MHz memory 600MHz timing 3 voltage 1080mV
Jun 13 10:24:49 f17e64 kernel: [    1.954188] [drm] nouveau 0000:01:00.0: c: core 405MHz shader 810MHz memory 324MHz voltage 900mV
Jun 13 10:24:49 f17e64 kernel: [    1.956084] [drm] nouveau 0000:01:00.0: Detected 1024MiB VRAM
Jun 13 10:24:49 f17e64 kernel: [    1.957868] [drm] nouveau 0000:01:00.0: 512 MiB GART (aperture)
Jun 13 10:24:49 f17e64 kernel: [    1.965067] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
Jun 13 10:24:49 f17e64 kernel: [    1.965068] [drm] No driver support for vblank timestamp query.
Jun 13 10:24:49 f17e64 kernel: [    1.965126] [drm] nouveau 0000:01:00.0: ACPI backlight interface available, not registering our own
Jun 13 10:24:49 f17e64 kernel: [    2.109037] [drm] nouveau 0000:01:00.0: allocated 1024x768 fb: 0x1a0000, bo ffff880424ed7000
Jun 13 10:24:49 f17e64 kernel: [    2.109136] [drm] Initialized nouveau 0.0.16 20090420 for 0000:01:00.0 on minor 0
Jun 13 10:24:49 f17e64 kernel: [    2.207643] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
Jun 13 10:24:49 f17e64 kernel: [    2.207645] [drm] Driver supports precise vblank timestamp query.
Jun 13 10:24:49 f17e64 kernel: [    2.298122] [drm] Enabling RC6 states: RC6 on, RC6p off, RC6pp off
Jun 13 10:24:49 f17e64 kernel: [    2.661763] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 1
Jun 13 10:26:33 f17e64 kernel: [  117.645376] [drm] nouveau 0000:01:00.0: DDC responded, but no EDID for HDMI-A-1

--
Comment 6 Malcolm 2012-06-14 12:13:12 UTC
Created attachment 63042 [details]
lspci

+ Attachment by John Reiser. Thanks +

lspci

Notice two VGA controllers:
00:02.0 VGA compatible controller: Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09)
01:00.0 VGA compatible controller: nVidia Corporation GF108 [GeForce GT 430] (rev a1)
Comment 7 Malcolm 2012-06-14 12:18:34 UTC
Created attachment 63043 [details]
Xorg.0.log

+ Attachment by John Reiser. Thanks +

Xorg.0.log

Notice that Xserver sees both VGA controllers:
[    14.362] (--) PCI:*(0:0:2:0) 8086:0112:1043:844d rev 9, Mem @ 0xfb400000/4194304, 0xc0000000/268435456, I/O @ 0x0000f000/64, BIOS @ 0x????????/131072
[    14.362] (--) PCI: (0:1:0:0) 10de:0de1:3842:1335 rev 161, Mem @ 0xfa000000/16777216, 0xd0000000/134217728, 0xd8000000/33554432, I/O @ 0x0000e000/128, BIOS @ 0x????????/524288
Comment 8 Malcolm 2012-06-14 12:19:40 UTC
+ Post by John Reiser. Thanks. +

kernel-3.3.7-1.fc17.x86_64
xorg-x11-server-Xorg-1.12.0-5.fc17.x86_64
xorg-x11-xinit-1.3.2-5.fc17.x86_64
xorg-x11-drv-intel-2.19.0-1.fc17.x86_64
xorg-x11-drv-nouveau-0.0.16-35.20120306gitf5d1cd2.fc17.x86_64
udev-182-3.fc17.x86_64
dbus-1.4.10-4.fc17.x86_64
gnome-desktop3-3.4.2-1.fc17.x86_64

processor	: 3
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping	: 7
microcode	: 0x25
Comment 9 GitLab Migration User 2018-12-13 22:26:48 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/xserver/issues/424.


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.