Fedora :: 13: Nouveau & Dual Head - Doesn't Get A Signal From Graphics Controller
May 26, 2010
I've got a "little" problem with fedora 13: dual head with nouveau is not working. gnome-display-properties recognizes two displays and i can configure it (left, right, mirroring, ...) but first display is always in standby. Seems it doesn't get a signal from graphics controller. Desktop seems to work fine (can move the mouse pointer on the other screen into "the black hole"). I use Fedora 13 x86_64.
I'm building another desktop. This is not a gaming system (though some video does cross my path), it's primary focus is as an application development platform. The thing is based on a SuperMicro two CPU motherboard with a couple intel multi-cores, 16GB RAM, a hardware RAID, and CentOS 5.6 (64bit). To this mix I would like to add a solid dual-head graphics card. Something that will support a couple HP LP2465 monitors (not new but functioning well). Speed of rendering is not my biggest concern, stability and "real estate" (the monitors are 1920x1200 and I want all of it).
I've looked a little at the Matrox items and started researching nVidia, but would like to hear opinions from those "in the know".The motherboard, though a solid performer, may have just made this trip unnecessary. It has a quirk that I hadn't considered before. It has several PCI-X slots (64-bit @ 133MHz) but only one "PCI-Express 1 (x8) on a (x16) slot". Does that seem a little strange to anyone else? Not that there is just one PCI-E but that it will fit a x16 board but will only support x8 connections...
Upon upgrading to Lucid from Karmic, my dual head setup doesn't work anymore with the new 2.6.32 kernel. In the previous kernel 2.6.31, everything works as expected (also when I boot Lucid with that kernel). On 2.6.32, the only thing I get is two clones. Is there some new type of configuration required that I missed or should I still compile some new kernel module???
My card:
Code: rogier@rogier-desktop:~$ lspci |grep VGA 01:00.0 VGA compatible controller: ATI Technologies Inc Radeon RV250 If [Radeon 9000] (rev 01) My xorg.conf, which worked fine on Karmic: Code: Section "Monitor"
I'm using a remix of F14 called Fusion Linux. It's all the same as fedora14. It's come preinstalled with xorg x11 nouveau driver. So, i'm not using Nvidia driver. Now, the problem is that the graphics in wine application are really poor. Fonts appear to be really hazy. See this image to get me better: [URL].. In earlier version of fedora and wine, i used nvidia driver. Graphics used to be good at that time.
I'm getting used to gnome3 one thing still bugs me- why when I switch workspaces on my multiple monitor machine does only the primary monitor change to a different workspace? The secondary monitor (my laptop screen) shows all open windows I put on it plus it's not even shown in the workspace list.
When I try to run the 'system-config-display' program for Fedora 11 and I use the dual head tab, attempting to save the settings results in this:
Traceback (most recent call last): File "/usr/share/system-config-display/xConfigDialog.py", line 212, in ok if card.getDescription() == self.secondVideocardOptionMenu.get_menu().get_acti ve().get_data("NAME"): AttributeError: 'NoneType' object has no attribute 'get_data'
I've been troubleshooting a friends computer and we've pretty well hit a brick wall. The issue at hand is the video card will not output properly. To be specific, it's an NVidia card with both DVI and VGA ports. We want it to output via the VGA and ignore the DVI because there's nothing plugged in there (the monitor we want to use doesn't support DVI). At any rate, the only time we can get the VGA to work is when we have another monitor plugged into the DVI port and dual head the system. When we do this it also adds an extra "Unknown Monitor" to the display panel. When we just use the DVI the extra "monitor" disappears. now we don't want to have monitor sitting here useless just so we can use the vga out, but there seems little else to do. We tried installing the Nvidia proprietary drivers through the software manager and yum and it failed both ways.
This is really frustrating because it seems like it should be such a simple thing but nothing works. Does anyone else have similar issues? is there something i can do to fix this? I apologize if i've not provided enough info, but ask and i'll post anything you need to diagnose.
I have just upgraded from Fedora 11 to Fedora 13 and now can't get my two monitors working as a single "screen" (it worked under F11). I have a GeForce video card with two monitor outputs
I then edit the xorg.conf to change the driver from nouveau to nv and I can start X and the LCD displays the desktop at a resolution of 1680x1050 (good) but the CRT just displays random blocks of colour. At this point I also get a kernel crash - see dmesg output attached, but the system is then OK to use. I have tried editing the xorg.conf to add a second monitor and a virtual screen resolution to match the two monitors (1024x768 + 1680x1050) combined (2704x1050) but still only the LCD monitor works, but with the virtual desktop that can be nudged around with the mouse. I would like to get back to using both monitors with a virtual screen across the two monitors.
I just wanted to check, I'm using nouveau on Fedora 12 with the latest updates and each time I wake the machine from suspend, the fan on my XFX 8800GT goes to full speed and stays there. I've read some nouveau stuff and they say that nouveau doesn't control the fan speed, so I think the card is running in a safety mode after wake up where it runs the fan at full speed to be safe. The only suggestion I saw for fan speed and nouveau was to use nvclock to manually adjust the fan but that sounds dangerous to me. I think I'll just go to the closed rpmfusion driver then if this is the case.
I have Fedora 11 installed on a system that initially had an ATI 1650 type card. I recently replaced it with a GeForce GT 220. I don't believe there was ever an xorg.conf generated or used with the ATI card. Initially, when I replaced cards and rebooted, I was able to start up KDE or GNOME but with only 2 very low resolutions available(the highest being 800x600). Next I ran Xorg -configure to generate an xorg.conf file. Everything I am seeing in this file seems to make sense. But now, when I startx all I get is a red error box being displayed by the CRT monitor with this message:
Signal Frequency is out of range FH 30.1 kHz FV 24 Hz Please Change Signal Timing
The only way I can get my X system back is to go into the xorg.conf file (in the Device Section) and change the driver to vesa. But I still only have 2 very low resolutions available.
All 3d applications run with a solid black screen, and compiz will try to revert to the fallback wm, even though I have the new nouveau/gallium setup in 10.10.
glxgears appears to *think* it has a working screen, but it stays blank:
Code: Running synchronized to the vertical refresh. The framerate should be approximately the same as the monitor refresh rate. 3570 frames in 5.0 seconds = 713.841 FPS 3298 frames in 5.0 seconds = 659.541 FPS
Once I have installed F11KDE (and I have), and I have a wireless adapter/booster, how do I configure my internet connection.My laptop doesn't want to pick up the signal.
I did a fresh install of FC11. I have 2 Dell 22" widescreen monitors turned in portrait mode. The display preferences tool let me rotate the displays, but wont let me move them directly adjacent to each other, so I have what amounts to a "gap" between this displays. This is very annoying when dragginga window from one display to the other, as it disappears when it is in the gap, and causes a "delay" when just moving the mouse from one display to the other.The behavior I get when I try to drag the displays adjacent to each other, is that they will go anywhere EXCEPT adjacent...as in they will happily overlap or cover each other, or partially overlap diagaonally etc. Note that they will go adjacent when in normal rotation.
I've just performed a fresh installation of openSUSE 11.3. System specs include GA-MA790X-DS4, PhenomII X4 940 AM2+, Sapphire RadeonHD 3670, two monitors. What I'm trying to accomplish is to set up my KDE desktop to (as it did under 11.2) use a multi-desktop multi-output non-cloned setup, so that I can use my two monitors side by side and maximize windows in either one respectively, etc.
So here I am with my newly installed system. Obviously I can't use fglrx, as it doesn't yet support the 2.6.34 kernel, so amdcccle is right out. SaX2, I notice, is gone. I have tried using KDE's built-in configuration tools, but the dual-head setup doesn't persist between logins. Strangely, windows that were open on the second monitor do vanish off-screen, but the monitors return to their overlapped positions so it's useless. Monitor 2 doesn't even retain its resolution setting. As one does, I decided I'd manually edit the X configuration file. SHOCK HORROR! No xorg.conf So what I'm driving at is this: How oh how do I configure my X.org to remember that it has two, quite different, monitors under openSUSE 11.3 ? Sorry if this post is hard to follow, I got very little sleep while trying to sort out the myriad problems I caused myself in upgrading.
I upgraded F14 to F15; However, F15 no longer recognizes my 3 sata disks connected through Marvell controller. The controller is an integrated Marvell chipset 88SE6480. The controller has its own manufacturer driver but it was intended for RHEL 5.4 (mv64xx) and installation of this driver fails to generate mv64xx driver.
In this particular example I am using NoMachine's NX, although the behaviour also occurs when running VNC for instance. I'd like to connect from my laptop (1 physical screen), to my desktop at work (2 physical screen) via NX. It all works but the following problem occurs: When I connect with my laptop to my desktop I (almost) always am connected to the 2nd physical screen, whereas all windows appear on this first screen. I can interact with my desktop, use the task manager etc, see which windows are running, but any application I start, starts on the first screen (which I cannot see) and for the life of me I cannot move them from screen1 to screen2 without dragging them (which obviously is impossible since I only see screen2).The obvious workaround is: *before* you connect disable the second screen via the display manager but for me this is a non-pratical work around because it requires a lot of actions before heading home. how I can connect from a single-head desktop to a dual-head desktop without crippeling my desktop experience?
I'm using gnome in a dual-head setup with individual desktops. Is there any way to "jump" to another monitor using the keyboard only? Now the only method I know is to move the mouse pointer to the another screen and click on something there. Can I do something like that without using the mouse?
I am trying out conky for the first time and rather enjoying it. I have a small frustration though. I have a dual head setup and would like to have identical conky's showing on both monitors. I have tried creating two .conkyrc's (.conkyrc1 & .conkyrc2) and then using:-
Code: $ conky -c config1 $ conky -c config2 to launch them, this results in only one of the two conky's displaying. I did at first think that maybe they were on top of eachother, but one conkyrc is set bottom_left, the other bottom_right so they should not be right??
When I try to run them as above terminal shows:- Code: gregg@burt:~$ conky -c config1 Conky: invalid configuration file 'config1' Conky: invalid num arg for top. Must be between 1 and 10. Conky: invalid num arg for top. Must be between 1 and 10. Conky: invalid num arg for top. Must be between 1 and 10. Conky: invalid num arg for top. Must be between 1 and 10. Conky: invalid num arg for top. Must be between 1 and 10. Conky: invalid num arg for top. Must be between 1 and 10. Conky: forked to background, pid is 7017 .....
#NETWORK ${font Sans:size=9:weight=bold}${color grey}NETWORK ${hr 2}$color${font Sans:size=8:weight=bold} ${color grey} DOWNLOAD UPLOAD ${downspeedgraph eth0 30,115 000000 00ff00} ${alignr}${upspeedgraph eth0 30,115 000000 ff0000}$color Processes 8 & 9 do not appear in my conky, just blank space between 7 & 10?? Also, the swap %age displays 'No swap%'?
I just installed lucid on my laptop with a second screen. Unfortunately, I cannot display any video. Neither on the laptop screen nor on the second external screen. In contrast, rebooting without attaching the external screen allows me to display videos again. I never experienced this kind of behavior with Ubuntu 9.10. Graphic card: Intel 945 Let me know if you need further information.
I have an Nvidia card that supports two monitors. Actually I have it setup. The problem I am having is I want to turn one of my monitors sideways so I can read forms easier. I can do this with just one monitor. RandR works fine with that. But it seems that when the system makes the two monitors into one long one it will not allow me to tweak the rotation in RandR.
Is there something I am missing. I deal with a lot of contracts and forms. And when I am reading them I hate scrolling. So I thought this could be an option. I have searched the net for 6 hours now. Maybe I do not know what I should be searching for.
Can any recommend a PCIe video card for my AMD64 box?Looking for one which basically has a good working driver:
- can do dual head using DVI and VGA output. - open source preferably, I've used the nvidia proprietary one before, was fine but kernel upgrades became a pain. - passive heatsink so it's silent, doesn't need to be a fast card since my other two were both onboard graphics. - driver can do KMS for nifty high res console. - most typical features work, so Mplayer plays nicely and can scale to full screen etc. (xv output +/- accelerated decoding) - even a bit of light 3D for openarena or compiz would be nice. - don't mind buying second hand if they aren't available new
I've had a Nvidia 7050 on board which was ok, fast enough, but upgrading with the proprietary driver became painful, and that motherboard is dying a flaky death. - The nouveau driver looked promising, but wasn't really in Debian at that stage. But I did once accidently boot a high res console while testing a brand new kernel (2.6.31 I think) and was blown away Currently got an ATI RS880 [Radeon HD 4200] on board which is fine, 2D is fast enough, but the open source driver isn't finished yet. It can't do dual head (horrible distortion and crashes), can't suspend to ram (freezes), Xvideo and any form of 3D acceleration are missing.
I have two identical Samsung monitors. One is connected. to my motherboard VGA port (and appears as a Radeon HD 4200).The other is connected to a VGA port on a separate PCI Radeon 9200 PRO board. When I first installed Suse 11.3, I had to go through the nomodeset boot command (otherwise my system froze up on boot) and I only get video on the screen connected to the motherboard. The other monitor light. flashes which means it's not seeing a video signal. And, I do not have an xorg.conf file in /etc/X11.
So I did the following:
I did an Xorg -config which created an xorg.conf.new file which looked sensible in that it showed both Radeons, and appeared to set up a dual-head system. The only change I made was to change the driver for both controllers from radeon to radeonhd. Then, I copied xorg.conf.new to /etc/X11/xorg.conf and rebooted. And, nothing is changed. I still need to do nomodeset on boot and I still only get video on the screen connected to the motherboard. I'd attach the xorg.conf and Xorg.0.log files, but it says I can't add attachments to this post. I don't see any major errors in the Xorg.0.log file, but a lot of it is greek to me.
I have two Ubuntu 10.4 machines (and Ubuntu continues to hide more and more xorg.conf config such that I no longer know where to find it). One is a laptop running dual headed - DP1 is the internal screen, and VGA1 is an external monitor; both are running at 1600x900. The other machine is a desktop running both VGA1 and HDMI1 (which is actually a display port with a DVI adapter) at 1600x900. So in both instances my desktop is 3200x900. I run a VNC server on the laptop and connect to it (via SSH tunnel) from the desktop - when I press the full screen hot key, I get a 1600x900 view of the remote machine on one monitor, and half of my local desktop on the other monitor - the "full screen" only expands to fill one local monitor.
Normally this is exactly what you want when you full screen a web browser, email client, or other application. I'm sure there's some X magic to make it clear what a full screen actually entails, and the vnc client application is just dutifully accepting what it's told. While I would like to keep the normal full screen behavior for regular applications, but when I'm VNCing to another 3200x900 machine, I'd really like full screen to stretch across both local displays. Resizing the window to be "close" isn't quite good enough since I still have local panels at the top and bottom of one display (though I can set them to autohide), plus the VNC client application window border (since it doesn't appear to respect -notitle).
Is there any good way to have X lie to a single application about the "full screen" size? Can I get it to lie to all applications? xrandr --noprimary appears to have no effect.
I'm looking at the 2009 apple mac mini ([URL]and wondering if anyone has gotten dual head working? I'm putting two 17" or 19" monitors on, so nothing too adventurous. I know that I'll require the "Mini-DisplayPort-to-DVI" to get Dual DVI Adpaters, but I wanted to confirm that dual head would be supported from the start. I'm aware that there are issues with rebooting the device, possibly as a result of the wireless drivers. That doesn't bother me. Dual-head is a requirement I intend installing 9.10 and dual booting macos for firmware updates.
I have my computer set up to use separate x screens for both of my monitors and almost everything works perfectly. With this set up I don't have any problem displaying flash in full screen on either monitor like I had with twinview. I have it set up this way so that I can use my TV to display mythtv and still use my computer monitor to browse the web, etc. The only problem is that I use my iphone with hipporemote to give me a wireless keyboard/mouse. When using twinview this was not a problem as it was just one large desktop spread over two monitors.
However, switching to separate x screens seems to have broken the functionality of the vnc server. When using my physical keyboard and mouse I can just move the cursor off the side of monitor 1 to control monitor 2, however, if I'm connected via vnc the cursor gets "trapped" inside the active monitor. I can interact with each screen as expected but if I want to switch the mouse/keyboard focus to the other monitor I have to go to the computer and move the mouse to the desired monitor, which obviously defeats the purpose of using the iphone as a wireless remote.
So my question is this. Is there a way to assign a hotkey combination to switch the focus of keyboard and mouse that the vnc server is controlling? Or could I set the vnc server to only control one of the two x screens?
i m facing same error in most of the HCL servers. the problem is that it throws error while booting and sometimes not throws error. the error is :-
Feb 13 13:17:25 fe13s kernel: Adapter 0: Bus A: The SCSI controller was reset due to SCSI BUS noise or an invalid signal. Check cables, termination, termpower, LVDS operation, etc.
Feb 13 13:17:30 fe13s kernel: Adapter 0: Bus B: The SCSI controller successfully recovered from a SCSI BUS issue. The issue may still be present on the BUS. Check cables, termination, termpower, LVDS operation, etc
Feb 13 13:29:15 fe13s kernel: Adapter 0: Bus B: The SCSI controller successfully recovered from a SCSI BUS issue. The issue may still be present on the BUS. Check cables, termination, termpower, LVDS operation, etc code....
I'm having a problem here with my graphics card. Until not very long ago I was using the controller Ubuntu downloaded for me. However I recently downloaded another one from the ATI page, called ati-driver-installer-11-3-x86.x86_64.run and installed it. So far, I've encountered two new problems:
1. The ATI Catalyst Control Center would not open after I setup the Tear Free Desktop.
2. The computer is running slow. Files take longer to open, minimizing and maximizing windows takes longer. Wi-fi connection takes longer to connect, etcetera. Overall, everything is going slow.
Are these problems related? I would like to uninstall the driver as the computer was working just fine before. In addition to this, I have firefox 4.0 and the colors on most of the images are darker than they should. The images look fine on other browsers except firefox. I read I had to do type "about:config" in the address bar, find "gfx.color_management.mode" and change the value from 2 to 0. What did I exactly do there? I took the advice from this thread: [URL]
