Red Hat / Fedora :: Building HP Qlogic NIC Driver On F13
Jun 5, 2011
I have been trying to build the HP Qlogic Netxen, nx_nic, driver on Ferdora13 kernel 2.6.33 but facing with "missing kernel-devel package" errors. As per the instructions in the driver spec, when i try to do rpm build, i get this error. I tried the rpm build installing kernel-devel-i686 pkg but it only shoots back with
error: Failed build dependencies:
kernel-devel-i586 = package is needed by hp-nx_nic-4.0.544-2.src
But when I installed kernel-devel-i586 pkg for kernel 2.6.30 the rpm build continued beyond this error point only to give some other RPM build error atleast to say there was no missing kernel-devel pkg error. Has anybody come across any such error with any other driver? or nx_nic in specific? Does this error mean kernel-devel-i586 a MUST for this driver? As far as I know there is no -i586 pkg for kernel 2.6.33 when i searched in kernel-devel repository! I am not sure, if this forum is right place for posting this question. If not, pls re-direct me to the right one.
I'm trying to find out when QME2572 (Qlogic) card became support by the kernel. We have a RHEL 5.1 system that is moving to new hardware, however the kernel at this release doesn't support the new hardware, due to the Qlogic card change. I tired the Redhat KB and Bugzilla. Is there a Kernel change list etc I can search. Never really played around with the kernel too much so I'm just after some pointers for looking up this information. Offically its not supported until Redhat 5.3, I'm trying trying to research kernel info so I can tell the customer they have to upgrade.
I'm fairly new to Linux. I am trying to build and install the drivers for the ASUS USB-N10 wlan usb dongle, using the newest drivers from the Realtek website. I've never used make before on Linux and am running into what are probably some pretty basic problems, but I'm not sure where to look for the answers. I used aptitude to install the "make" and "build-essential" packages. I unpacked the drivers using Archive Manager. I changed to the driver directory and ran "make". It stopped with "No such file or directory." I became superuser and ran make, in case it had permission problems when creating the directory.
Here are the results. obie@four:~/realtek/rtl8712_8188_8191_8192SU_usb_linux_v2.6.6.0.20110401/driver/rtl8712_8188_8191_8192SU_usb_linux_v2.6.6.0.20110401$ make make ARCH=i386 CROSS_COMPILE= -C /lib/modules/2.6.32-5-686/build M=/home/obie/realtek/rtl8712_8188_8191_8192SU_usb_linux_v2.6.6.0.20110401/driver/rtl8712_8188_8191_8192SU_usb_linux_v2.6.6.0.20110401 modules make: *** /lib/modules/2.6.32-5-686/build: No such file or directory. Stop. make: *** [modules] Error 2 .....
I haven't modified the makefile in any way -- it is the same as what I downloaded. I've read posts from other users who have compiled and installed the drivers on Ubuntu with no problem.
The contents of the directory I'm in when I run make are as follows: drwxrwxrwx 21 obie obie 4096 Apr 1 03:25 . drwx------ 3 obie obie 4096 Sep 3 22:43 .. -rwx------ 1 obie obie 3479 Apr 1 03:25 autoconf_rtl8712_usb_linux.h -rwx------ 1 obie obie 151 Apr 1 03:25 clean drwxrwxrwx 2 obie obie 4096 Apr 1 03:25 cmd -rwx------ 1 obie obie 516 Apr 1 03:25 config drwxrwxrwx 2 obie obie 4096 Apr 1 03:25 crypto drwxrwxrwx 2 obie obie 4096 Apr 1 03:25 debug .....
i downloaded madwifi0.9.4 because my laptop has an atheros wifi card sadly i am new to linux so do not understand how to build a driver and just downloaded the software add-on windows wireless drivers but could not find an inf file then i see i have to build the thing myself from the kernel. well i am a win7 user wanting desperately to only have to use windows for playing games and do all my work on something more secure and would like to be able to use my laptop in a coffee shop etc.
Apparently, the wireless adapter Realtek 8171 uses the rtl8192se driver, which is currently available only as source from Realtek's website. Is building a driver equal to building a package? Should I follow the same cautions, say these ones?
I am trying to build the ATI fglrx 10.12 driver using the ATI installer on debian Squeeze 64bit.
Todo this I run the following command: ./ati-driver-installer-10-12-x86.x86_64.run --buildpkg Debian/testing
Which produces this output: Created directory fglrx-install.oFbosX Verifying archive integrity... All good. Uncompressing ATI Catalyst(TM) Proprietary Driver-8.801 Generating package: Debian/testing cp: cannot stat `/home/nojan/Downloads/fglrx-install.oFbosX/x710_64a/*': No such file or directory Package build failed! Package build utility output: ..... dh_install: Compatibility levels before 5 are deprecated. cp: cannot stat `./usr/X11R6/lib64/modules/linux': No such file or directory dh_install: cp -a ./usr/X11R6/lib64/modules/linux debian/fglrx-driver/usr/lib/xorg/modules/ returned exit code 1 make: *** [binary] Error 2 dpkg-buildpackage: error: debian/rules binary gave error exit status 2 Removing temporary directory: fglrx-install.oFbosX
I am fairly new to Linux and have been striking hassles with my AWUS036H (RTL8187) under Debian Lenny. The standard (built in) driver works but reception is pathetic compared to the same card under Backtrack 4 or even Windows 7. I have tried to build drivers which I got from the Alfa website, the Realtek website and also the Linux Wireless project page. All of them fail during Make with an Error 2. I don't have the error in front of me (had to go back to Windows for the moment) but it is identical to the one this guy gets:
darkReaction:/home/georanson# cd /home/georanson/Desktop/Downloads/ndiswrapper-1.52 darkReaction:/home/georanson/Desktop/Downloads/ndiswrapper-1.52# make make -C driver make[1]: Entering directory '/home/georanson/Desktop/Downloads/ndiswrapper-1.52/driver' Makefile:35: *** Cannot find kernel version in /lib/modules/2.6.18-6-686/build, is it configured?. Stop. make[1]: Leaving directory '/home/georanson/Desktop/Downloads/ndiswrapper-1.52/driver' make: *** [all] Error 2
I get the error 2 code and the reference to "kernel version".
I recently bought a pci multi io card to give my pc 2 parallel ports. For Linux, Moschip supply the source plus makefile and instructions. I ran make as root and got an error message make: *** /lib/modules/2.6.31.12-0.2-desktop/build/: No such file or directory. So I created a folder called build in /lib/modules/2.6.31.12-0.2-desktop but that isn't good enough as I still get an error cannot make target modules. So I assume there must be something in the build folder. So what packages do I have to install to get the correct environment to be able to build the driver? Has anyone done this and can supply the driver for 11.2?
I am attempting to install the Linuxant ALSA driver to make my sound work properly (speakers muting when headphones plugged in). This worked on Lucid, but on Maverick the installation fails at the "building modules" stage. It directs me to a logfile, which I have reproduced here:
The package partially installed I have no sound at all. I have kernel headers and every related -dev package I can think of installed. I'm running 64-bit; could that be an issue?
In our environment we are using EMC for storage. We are using Qlogic provided drivers and PowerPath and the OS is RHEL5.3. Currently we have 2 Luns connected in one of our server (emcpowera & emcpowerb) and storage team has allocated another 100Gb Lun to that system, my doubt is if I execute "ql-dynamic-tgt-lun-disc.sh" script that would be enough to detect the Lun and see the device under /dev/ as emcpowerc or I need to execute "powermt config" and "powermt save" commands also.
I got openSUSE installed, and now am trying to build my wireless driver, using the instructions here: [URL].
The wireless driver is here: Broadcom.com - 802.11 Linux STA driver NOTES: - The (parenthesis: ) are my notes to show you what I was doing. - Lines with "#" before them are what I typed into Terminal. - Lines below (Output: ) are what was output by Terminal.
(BUILD DRIVER:) #su <password> # mkdir hybrid_wl # cd hybrid_wl # tar xzf <path>/hybrid-portsrc_x86-32_v5.100.82.38.tar.gz # make
(output:) if 'make' is not a typo you can use command-not-found to lookup the package that contains it, cnf make (NOTE: with "cnf" it is saying "Cannot Find") (install "make" command:) # zypper in -t pattern devel_basis (It installs 93 packages) # make
(output:) kbuild_nopedantic=1 make -C /lib/modules/`uname -r` /build M=`pwd` make: *** /lib/modules/2.6.34-12-desktop/build: No such file or directory. Stop. make: *** [all] Error 2
(Tried:) # zypper in kernel-source (It installed 3 more packages) # make
(output:) kbuild_nopedantic=1 make -C /lib/modules/`uname -r` /build M=`pwd` make: *** /lib/modules/2.6.34-12-desktop/build: No such file or directory. Stop. make: *** [all] Error 2
The problem might be derived from the fact I had to install openSUSE in Kernel (F6) > Safe Mode. If I need to install more things, please let me know ASAP! I am wired to my ethernet until then, which means I can't move from this spot until my wireless is working!
We had a problem with the NIC 10Gig Qlogic NICs and had to swap them with Emulex 10Gig NIC cards. The swap has been done. There are several severs (about 100)for which we need to check if the hands-on still left the q logic card in them. So, I wrote a script using dmesg and grepped for "qlogic" or " NC522SFp" and my script checked for all the servers and returned no presence of "qlogic". But few of them have q logic still in them. So, is there any other way through which we can exactly tell whether q logic card is still present or not? I used:
and found nothing in the o/p. But when I checked using ifconfig -a, then I see the q logic mac ids for eth4 & eth5, which means the Q logic is still present.But not sure whether ifconfig is the real way to find out the presence of these cards.
My system is an Altix3000 running: (Linux sgialtix 2.4.21-sgi304r1 #1 SMP Sat Jan 29 22:43:29 PST 2005 ia64 ia64 ia64 GNU/Linux) (redhat-release=redhat-release-3AS-8.3). Which command do I use to list World Wide Name information of the QLA2300 HBA? I tried /proc/scsi and /proc/xscsi but I found nothing there.
One our client has recently installed Red Hat 5 X64 on the hosts, each host has 2 Qlogic HBA's, which are connected to IBM DS4800.RH5X64 has MPIO built in or do we need to download? If download is required, pls provide link.
I am trying to setup a blade server that has a QLOGIC 4062 mezzanine card to handle the iscsi traffic coming from an HP MSA 2312i SAN. I cannot seem to figure out how to get the HBA to talk to the SAN. Just to ensure the SAN was setup OK, I used the onboard netextreme NIC w/iSCSI to connect to the SAN with iscsiadm. This all works no problem. BUT, I bought the mezzanine cards to take this load of my NICs.
I know that the OS sees the two ports on the mezzanine card as dmesg shows them. This is my first time configuring a separate HBA on a blade system, so I'm kinda lost what to do with them. I have entries in the modprobe.conf file for eth2 and eth3 aliased to qla4xxx. What I am not sure about is the config for each eth2/3 device. If I ifconfig my device to the specific ip and subnet/netmask my SAN is on, I can get any response from the SAN.
Also I would have expected iscsiadm -m iface to pick up on the qlogic devices, but they don't show up in the ifaces directory. I tried to manually create an iface file for each, and it seems to be ok with that, but when I try to discover targets, I get communication errors. (I am compiling a file with all the outputs from various commands and errors, and will put them here once I have them).
What I don't understand is the process of setting up an HBA. I am quite able to setup iscsi over the regular NICs, been doing that for a while, but these HBAs have me stumped.
I also installed the qlogic qla4xxx driver and SANSurfer, but SANsurfer didn't see the HBAs. I think this was due to the age of the sansurfer software...other people have reported this as well.
We have a SuSE/SLES 9 server that boots from a fibre channel card, using volumes from a NetApp filer.
We previously had the server booting from the LUNs on the filer, but after some tinkering around with the fibre channel BIOS, we have the situation where the LUNS seem to be mounted, the OS boots, starts to initialise everything but then stalls, saying:
We have tried loads of combinations of settings in the BIOS and fibre channel BIOS without any success or any idea what may have caused the error.
I am trying to build a system that will run both KDE and GNOME. I currently have just GNOME installed. So I need to install KDE and KDM (which I know how to do). The thing that I don't know how to do is make it default to KDM when it boots rather than GDM. Also I need a way to get rid of the GNOME apps in the menus in KDE and the KDE apps in the menus in GNOME. Also, this system in running Fedora 11 x86_64
I am trying to compile a kernel (so that the USBIP module gets built, I've enabled it via menuconfig) in the following directory:
Is this src tree "intended" to be rebuilt?
I'm trying to compile an "EXPERIMENTAL" module (w/in the tree already) USBIP so this seems to be my ONLY recourse. I do not have the ability to connect the machine to the net, nor bring in things like non-professionally burned CD's nor flash drives...
I'm trying to build the yabause emulator on fedora. The GTK port is available in the repos, but I'm wanting the QT port as it will support more than one joypad. I'm using Fedora 14 x86_64 Yabause source is available here: [URL] I'm following the instructions in README.QT in the yabause library but Ihit an error saying that I'm missing libgtkglext. However I have this library in /usr/lib64/
I am trying to build a gcc 4.4.3 cross-compiler. I downloaded the gcc 4.4.3 release, uncompressed it, made a new directory called gccbuild and cd'd there. I have already installed gmp, mpfr, and mpc in /usr/local. These are the commands I used:
Code: configure --target=arm-linux --enable-languages=c --with-gmp=/usr/local --with-mpfr=/usr/local --with-mpc=/usr/local make However, the build fails with Code: checking for arm-linux-gcc... /home/steven/gccbuild/./gcc/xgcc -B/home/steven/gccbuild/./gcc/ -B/usr/local/arm-linux/bin/ -B/usr/local/arm-linux/lib/ -isystem /usr/local/arm-linux/include -isystem /usr/local/arm-linux/sys-include checking for suffix of object files... configure: error: in `/home/steven/gccbuild/arm-linux/libgcc': configure: error: cannot compute suffix of object files: cannot compile .....
I want to install Fedora 10 but need to build RAID (bcraid) drivers so the installation will see my RAID array. I'm currently running Fedora 6 & thought this would be the process:
1) Install kernel source 2) Modify configuration to include bcraid drivers. 3) Build new kernel 4) Replace kernel in standard build with custom kernel or load driver during installlation 5) Install Fedora 10
I'm stuck on no. 1. When I tried to install the kernel source, there was some problem with "mockbuild" (wot dat?) which I think was a red herring but I now don't know what to do about these unsatisfied dependencies:
was originally looking to setup some kind of NAS in my house due to the fact that me and my wife both access the same music libraby, photo library etc. and was looking at one of the IcyBox enclosures.I realised that whilst this was a quick and possibly easy way to set something up, it wasn't in keeping with the true geek within so decided to look into setting up a full RAID5 file server. This would then allow me to install and setup various other services such as Web and FTP server, Mail server and DNS server should I ever want to.
I have an old machine (Duron 1GHz I think) with 1GB ram but the only problem is that is doesn't have an on board SATA controller so I was wondering what PCI SATA controller cards work with Fedora? Doesn't need to have RAID capabilities since i'll be using Software RAID anyway, just need a nice cheap Fedora friendly controller card that'll do the job.
For a new project at work, I'm looking into building a processing farm of a few dozen Tesla 1U servers from Nvidia. This has lead us on to figure a way to let the developers run tests on their own workstations (with the appropriate gfx card).The cards we're looking at are Quadro FX 1800 or 3800's to provide quite a bit of poke.
Thing is, I'm having issues with running any Cuda code successfully on systems installed with the pre-built nvidia drivers from RPMFusion. I can only get code running if I download the nvidia or cuda drivers from nvidia and install by hand. Not something I'd like to be maintaining across all out developer workstations. Is anyone out there running Cuda on Fedora/RHEL5 platform? If so, are you using RPM nvidia drivers or building them by hand?
I am trying to update my fedora 8 kernel, is there a way to do that? I build the one of kernel.org but when i want to boot it a get a lot of messages like: mount: cant find /dev/root I tried to follow the guide at building a custom kernel for fedora, but thats just for the last supported version of the kernel of fedora 8.