Debian Multimedia :: Can Put PCIE X16 Card In PCIE X1 Slot?
Jun 28, 2010
I have fairly new requirement for a second monitor. My main display is running fine using nvidia proprietary drivers.I do have a pcie (probably x16) video card sitting around, but my motherboard only has 1 x16 slot (used by my main display), 2 x1 slots and a few pci slots. From memory, the second video card is an ATI card.Questions:
1. Can I put a PCIE x16 card in a PCIE x1 slot?
2. Would using 1 Nvidia card and 1 ATI cause trouble?
I've searched everywhere I could think of and have been on google for about 4 hours today searching with no luck.Is there anyone out there who can confirm a working PCIe sound card on RHEL4.4 (Linux kernel 2.6.9-42)?The only cards I've found that work on Linux are the Asus Xonar cards and the PCIe version of the Sound Blaster Titanium but they are confirmed for later versions of Linux but I can't modify the alsa drivers or update the version of RHEL.
I have ubuntu 10.04 LTS. Kernal version 2.6.32.duplicator@Duplicator:~$ uname -aLinux Duplicator 2.6.32-24-generic #39-Ubuntu SMP Wed Jul 28 05:14:15 UTC 2010 x86_64 GNU/Linux.I am trying to install 4port USB PCIe card. But card is not detecting. but card manufacturer says it is plug and play device. this card is working in fedora 11. but i need to install in ubuntu 10.04.
I recently added a new graphics card to my machine. Normally, Centos 5.3 runs without a hitch, but since the new card, boots as far as a kernel panic. Even a rescue or an install disk comes up with a kernel panic.I have tried Centos and Fedora CDs/DVDs.The card is a PCIE Radeon HD 4350.It works fine with XP, btw.Ubuntu and OpenSuse will load from install sources.
I have a ATI Radeon HD 4000 PCIe card in my Debian system. I got the proprietary ATI drivers working (fglrx) but a kernel upgrade killed that. I was unhappy with this driver due to the very high memory usage. The card works with the non-proprietary drivers but I can only clone my desktop. I want a single desktop on two monitors.
I booted with Ubuntu Lucid and my dual monitor desktop worked out of the box without the proprietary drivers so I thought that I could emulate what Ubuntu does in Debian. But I can't.
My issue is that I don't know whether I am missing a module in the kernel or I need to set up a brand new xorg.conf. Ubuntu doesn't have an xorg.conf to copy. My default Debian doesn't either.
I have messed around with my own xorg.conf and added various drivers eg "ati" and "radeon" but I can't get it to work with a single desktop dual monitor display.
I am trying to install a PCIe card for data compression in Linux ubuntu 9.10(karmic Koala). As you can see in the following code although I have the root access I can not get permission to install the device driver.As I have the root access I suppose to get access to install the driver. I checked around and got the procedure from a web siteEnabling root access in unbuntuBasically in almost all linux machines we will be havingroot access for login.but in ubuntu by default that option was not enabled.If u want to have root login in ubuntu then u have to do some changes in ur Ubuntu machinesSteps to follow1.First enter into ubuntu as your user login.2.Then open the terminal and type the command sudo -s,it will ask for root access and type the password.3.Then type passwd root it will ask for new password4.Open terminal and go to the path /etc/gdm/5.Open gdm.conf6.search for AllowRoot=false7.and then change AllowRoot=true8.now restart ur machine9.now u can enjoy with root accessThe funny part is when I search for the file which is gdm.conf in the prescribed location, I can not find it.
Code: root@faisal-desktop:/home/faisal/Desktop# cd AHA/ root@faisal-desktop:/home/faisal/Desktop/AHA# ls
my servers cpu gets woken up alot (acording to powertop) by eth0 when downloading files, i figured i could get a cheap pci card that supports offloading.
I recently finished a mini-itx machine and everything was working great with a fresh Jessie install. I put in a PCIe wifi card and now the computer won't boot. The power LED is on, the PSU fan is spinning, but I get no display output at all and I don't know how far into the boot process the thing even got. I removed the card and the machine boots fine again.
I only have one PCIe slot so it is my only option, I don't have another PCIe card to try, nor do I have another machine to test the wifi card in. The card is brand new, for what it's worth.
I bought a Laptop with Sis 771/671 onboard graphic, with a 1440x900 resolution. Since this SIS is not supported by the open suse distribution, I am running it with poor resolution for two years now. Every update of OpenSuse I hope it could be supported now, since I see in the internet that there seem to be solutions for other distributions. However nothing happend yet.
Is there any intension at OpenSuse to do something about it ?
I spent all day yesterday trying to figure out why the PCIe expansion (bridge) card I bought was not being detected by my BIOS (and ultimately my RHEL5.4 OS). It was not until I actually plugged something into the expansion card that the bridge itself and the expanded PCIe component were detected.
I have tested and re-tested this issue.My question is this: Why must I have a PCIe component connected to the bridge for the BIOS (and OS) to detect the bridge itself (and the connected PCI component) ?
I am testing a PCI card in a 4-slot bus expansion chassis. I bought the chassis from CoolDrives.com, it appears to be identical to the Startech PEX2PCI4 unit, but was quite a bit cheaper... This device shows up on my lspci output:
I've installed in the expansion chassis one card, a Linux Media Labs BT484 card. This card works out-of-the-box when installed directly in the system, using the bttv driver. Installed in the expansion chassis, the card still shows up as follows:
07:00.0 Multimedia video controller: Brooktree Corporation Bt878 Video Capture (rev 11) 07:00.1 Multimedia controller: Brooktree Corporation Bt878 Audio Capture (rev 11) 07:01.0 Multimedia video controller: Brooktree Corporation Bt878 Video Capture (rev 11)
[code]....
When the card is installed directly in the motherboard it gets IRQ17 or 18. I suspect that it does not really want IRQ0, but that there is some problem detecting the proper IRQ over the bus expander. One possible approach I am considering is to build a custom version of the driver where I can set the IRQ manually to something else just to get the driver to finish loading, then see what happens when I run the card. This would at least tell me which IRQ the hardware is really using and whether or not the 0 reported is correct or not.
I have a XScale PCIe RS-232 board running in a home built quad core w/ 8GB ram that I want to use to either communicate with a UPS or a RS-232 embedded device. The relevant sections of the output of dmasg are;
dmesg |grep ttyS serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A 00:07: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
I'm wondering if I could put my old DualCore PCIe Board into a Proliant 3000? Why not; I mean, the Array Controller should plug right into the PCI slot on the DualCore board. Of course, I'll have to shoehorn a second power supply into the case for the newer Board, but wouldn't it be a monster! Two 3GHZ Processors, with a 256 MB nVidia graphics card, 1.5 GB of RAM, and 8 Hard Drives! (7, actually, because one of them doesn't work). I'm up late, and I think I'll start working on it right now.
I am unable to access a standard memory pcie card bar registers through my driver on the pcie interface for Zotec H55 ITX motherboard (it includes a built in NVideo engine). This is the following setup: 1. Fedora 14 32 bit pcie scan does see the board (lspci -n show card on pcie bus #2) 2. Configuration register read is good. (pci_read_config_byte()) 3. The following driver commands did correctly return bar information: pci_rsource_start(), pci_resource_end(),pci_resource_len() 4. The commands iowrite32() or memcopy_toio() results in no pcie transfers (determined through pcie core probing) 5. The same driver code fully functioned on the Intel DH57JG mini-ITX. (using lspci -n, pcie located on pcie bus #1) 6. Note, it seems that the pcie connector for Zotec is on bus#2, where as Intel MB pcie connector is on bus#1. Note also that the Zotec has a built in NVidia graphics engine.
The following driver code was used. Again, it worked on DH57JG: pcid.pciaddr_phy_str_bar0 = pci_resource_start(pcid.pcidev,0); if(pcid.pciaddr_phy_str_bar0 == 0) { printk(KERN_INFO "fail to get pci start address for Bar0. "); return -ENODEV; } pcid.pciaddr_phy_len_bar0 = pci_resource_len(pcid.pcidev,0); if(pcid.pciaddr_phy_len_bar0 == 0) { printk(KERN_INFO "fail to get pci end address for Bar0. "); .....
I have been thinking about upgrading my RAID setup from a pair of Intel X25s running the software based RAID included with ubuntu to four Intel X25s running off a PCIe based controller. The controller is an HP P400 which I know works great on ubuntu (I have other machines running it with SAS drives). My desktop has an Intel S5000XVNSATAR mainboard and I have an open PCIe v1.0 8x (physical) slot that is wired 4x. The controller is 8x and will fit fine. How much of a performance hit do you think I will see running this 8x controller in the slot wired 4x with four Intel x25s in a RAID 10? Will I have enough bandwidth with the 4x for those SSD's?
I'm having trouble getting a PCIe device to show up in lspci using pcie hotplug.
After booting the system I load the pciehp module, I then plug-in my PCIe card. After that I do an lspci, but I don't see my PCIe card. When I cold boot with the PCIe card installed the system sees it fine, i.e. it show up in lspci.
Does anyone know of an easy way to figure out if my hardware even supports pcie hotplug?
When I load the hotplug module I just do a modprobe pciehp. Is this all I need to do prior to plugging-in my pcie card?
I will try running pciehp with debugging turn on to see if that gives me any additional info.
I have a DELL T3500 64 bit and I am unable to install the following Matrox driver: M9140 LP PCIe x16 in RedHat ENT Linux 5.1 .Presently I have a nviDia driver that I would like to uninstall from the workstation so that I could install the Matrox driver.
I'just installed ubuntu 10.04 LTS on my new laptop(ACER aspire 5750). Ubuntu installed perfectly. Every things working ok but, it can't able to detect my CABLE MODEM.
I have a fresh install of suse 11.0 and there is no 3d acceleration available. If I use the suse repository, which driver do I need to enable 3d accesleration? The video is integrated on the motherboard. The box is an hp pavilion a1130n and amd 3500+ processor. I installed the same suse 11.0 on a dell box with an ati 128 rage and 3d was enabled after the install.
During install of 9.10 server, both ethernet adapters were found. One is built-in to the motherboard, the second is on a PCIe slot. Selected the on-board adapter for eth0 during install, set fixed IP on the web, connected, updated, etc. w/ no problems. The second adapter is for the LAN.BUT, on first and subsequent reboots after installation complete the second adapter is not detected at all, as if it were not even plugged into the PCI or the driver not loaded. This happens on either of two slots, with two different adapters, Intel or Netgear, both pretty recent, and both of which are detected and work in a different box. 'Ethtool eth1' says device not found.This is a fairly recent motherboard 3N78EM).Since the adapter was found during installation I'm assuming that the problem is not with not the board. I've built several server systems with two or three ethernet adapters and have never run into this situation.
I've just installed the latest version of Ubuntu (11.04) on my Lenovo S10-3 netbook. Everything works just fine, except the mobile mobile broadband.
The system doesn't seem to recognize the (integrated) SIM card slot. The SIM card slot is already integrated (it's located under the battery). I can add a new broadband connection manually, but it still doesn't work (I can't select it from the available connections). If I open-up the additional drivers manager, it doesn't show me any available drivers.
Bought a couple new cards to do some bitcoin mining (just an experiment) and before I hook them up I want to know how to use p2pool. im not that good with ports and was wondering how to correctly configure it. I keep getting Bitcoin is downloading blocks errors.
Edit: Im also getting a low mhash rate of 1.5. Is this correct for a geforce 9400m? Im not too worried about it after I use these new cards. and is there a difference in using laptops vs a pci slot desktop video card?
I'm trying to boot an SD card on a notebook that does not have BIOS support for booting from the SD slot. Using various how-to's I've figured out how to add additional SD card modules to the initrd.img file on a bootable USB drive such that I can boot Linux installed on the SD card.
However, best I can tell, it loads the kernel and initrd.img from the USB and everything else from the SD card. What I really want is to load the necessary SD modules from the USB and then chainload the SD card such that whatever kernel is on the SD card is loaded instead. Is it possible to chainload to another bootable device after the kernel (with the SD module additions) has already been loaded?
F12 on my aspire one (A110, the original 8Gb SSD model) works well with only one significant problem.
Suspend fails; the screen fades to black, backlight stays on, macine stops responding to keyboard/mouse/powerswitch and just hangs (for at least an hour... longest I left it). Only thing that works is a 5 second click of death on the powerswitch. Plus an FSCK on at least one occasion after reboot. I have not been able to check if SSH and shutdown possible (on vacation) plus have logs go to a tmpfs (to save my SSD) so no log info. Similarly I cannot confirm if this affects hibernate because I run without a swapfile (SSD again, i.5Gb ram installed and I have never come even close to running out of memory).
But I think I have the culprit anyway:[url]
As described there; shutdown fails if there is a mounted card in the cardreader slot (/dev/mmcblk0) My system has a 4Gb SDHC card in the LH slot with a single XFS filesystem; which has my homedir on it; if I unmount that filesystem (but do not remove the card) shutdown and restore appear to work properly. But if the filesytem is mounted I get the bsod. Linked bug suggests this is restricted to large filesystems (?SDHC?) but I have not fully confirmed, although that matches my situation.
Posting here to see if anyone has any further info/workarounds, to ensure the fedora crew are aware that this appears to affect F12 (and is a regression, F11 suspended fine) and make sure it is documented somewhere. The linked bug also states similar bad suspend activity on other netbooks.
I'll try some other variations (and see if it also happens with USB sticks/RH cardrader, and report back here.
