Compiling STREAM module for 64-bit kernel

I am trying to compile a STREAM module for Solaris. The results I get when I execute "isalist" and "isainfo -kv" are:
-"sparcv9+vis sparcv9 sparcv8plus+vis sparcv8plus sparcv8 sparcv8-fsmuld sparcv7 sparc"
-"64-bit sparcv9 kernel modules"
The host is a Ultra5_10 with SunOS 5.7 sun4u sparc.
First I tried to compile the module with gcc 2.8.1 but altough the compilation apparentelly finisihed OK I did not get to load the module. Later I read that, gcc so far does not generate binaries for 64-bit sparc.
So I installed the Sun WorkShop 5 environment (try version) and I tried to compile my sources with the following line:"cc -D_KERNEL -xarch=v9 -xcode=abs32 -xregs=no%appl -x03 -I. -I/usr/include -c mymodule.c"
but I get a huge list of errors such as:
"/usr/include/sys/thread.h/, line73:warning: volatile is a keyworkd in
"/usr/include/sys/thread.h/, line 73:sintax error before or at:volatile"
so if I compile adding the directire "-D_ASM" to the previous line, the list of errors descreases to 2 lines:
"/usr/include/sys/kstat.h, line 52: sintax error before or at: hrtime_t"
"/usr/include/sys/kstat.h, line 52: cannot recover from previous errors".
I have also tried adding directives such as: -D_LP64 -D__sparcv9 -D_POSIX_C_SOURCE, -D_XOPEN_SOURCE but I do not get to compile it successfully.
Could someone be so kind to give me a clue of what I am doing wrong, and what should I do in order to compile it??.
The header files I have included right now in my ".c" are:
#include <sys/stream.h>
#include <sys/conf.h>
#include <sys/modctl.h>
#include <sys/kmem.h>
#include <sys/cmn_err.h>
#include <sys/ddi.h>
#include <sys/sunddi.h>
If I do not add the directives -I. -I/usr/include to the compile line above I get the following error:
"/usr/include/sys/modctl.h, line 185: major, argument mismatch"
"/usr/include/sys/cred.h, line 51: sintax error before or at: uint_t"
But this is because the environment it is taken the /usr/ucbinclude as default include path instead of the "/usr/include". �which one should I consider?.
Thank in advance for your support.
Looking forward to hearing from you.
(could you please send to me the answeres to this e-mail:[email protected]??)

Maybe a file I can edit that says to run the Garmin stuff in 32-bit mode?
Right or control click the Garmin application (or utility) then click: Get Info
In the Get Info window under the General tab, if it can be enabled you should see:
Open in 32-bit mode.
Activity Monitor (Applications/Utilities) shows all apps running in 64 bit. If it doesn't have 64 bit under Kind on the right, then it's running in 32 bit mode.

Similar Messages

  • Can anybody compile CIFS modules for IdeaPhone P780

    Can anybody be so kind to compile cifs modules for Lenovo IdeaPhone P780? It'll help us to connect network shares to use as local resources.
    Kernel source for P780 I found here:​nes/p-series/p780-smartphone
    Thanks in advance

    kleptophobiac wrote:Thanks a ton! Now what about madwifi?
    uudecode ./../hal/linux/i386-elf.hal.o.uu
    make[1]: uudecode: Command not found
    make[1]: *** [hal.o] Error 127
    You haven't got uudecode installed.  It's part of the sharutils package, so install that.
    - olly

  • Virtualbox doesn't compiles "vboxdrv" module for 2.6.24 kernel

    I installed Virtualbox, however I cannot launch any OS because the vboxdrv module doesn't exists for the latest kernel (I found another module for the 2.6.23 kernel); so the GUI says that I should execute this:
    /etc/init.d/vboxdrv setup
    However that file doesn't exists. I already tried to uninstall virtualbox and reinstalling from pacman, but still no luck.
    What should I do? Is this a bug?
    Thanks in advance.

    I think that's where the confusion  is.  When I got the package from AUR it had:
    -rw-r--r-- 1 ralvez users  217 2008-01-08 15:58 interfaces
    -rw-r--r-- 1 ralvez users 3596 2008-01-20 11:38 PKGBUILD
    -rwx------ 1 ralvez users  174 2007-12-30 04:23 vbox_build_module
    -rw-r--r-- 1 ralvez users 2384 2007-12-30 07:39 vbox.install
    -rw-r--r-- 1 ralvez users 2179 2007-12-30 07:31 vboxnet.patch
    So the patch file is there (in bold text) ... and is not executable is a text file. That's why I'm asking if I have to do something with it to make the whole thing work.
    Since it is just a plain text file if I try to use it as an upgrade patch it fails with error:
    loading package data... error: error while reading package vboxnet.patch: Unrecognized archive format: Invalid or incomplete multibyte or wide character
    error: failed to add target 'vboxnet.patch'
    So, I'm stuck. I see no way to solve the problem.
    Edit: On re-reading your post I wonder what package from core are you referring to? I was under the impression that when we use AUR we are "fetching" files not from core but form the AUR source.
    Am I wrong?
    Last edited by ralvez (2008-02-13 22:50:31)

  • Camera Firmware Update Asking for 32 bit Kernel?

    I'm updating my Sony DSLR camera firmware and it rejects doing the update because of 64 bit kernel and asking for 32 bit??
    What needs to be done to accomplish that?

    Forgot to mention, I did do a search & tried the reboot while holing down 2 & 3 keys to no avail. Tried Restart & then Shutdown - didn't work

  • Need to compile nvidia driver for a custom kernel [SOLVED]

    I compiled a custom kernel following this wiki page: … rom_Source
    I can boot into it just fine.  I now want to compile an nvidia driver for use with it and followed the steps on this wiki page: … tom_kernel
    I run into a problem when makepkg trys to compile it.  Here is the output:
    $ cd ~/devel/abs/nvidia/
    $ makepkg -c
    ==> Making package: nvidia-my 190.42-1 x86_64 (Thu Nov 5 17:42:29 EST 2009)
    ==> Checking Runtime Dependencies...
    ==> Checking Buildtime Dependencies...
    ==> Retrieving Sources...
    -> Found in build dir
    ==> Validating source files with md5sums... ... Passed
    ==> Extracting Sources...
    ==> Entering fakeroot environment...
    ==> Starting build()...
    Creating directory NVIDIA-Linux-x86_64-190.42-pkg0
    Verifying archive integrity... OK
    Uncompressing NVIDIA Accelerated Graphics Driver for Linux-x86_64 190.42..................................................................................................................................................
    If you are using a Linux 2.4 kernel, please make sure
    you either have configured kernel sources matching your
    kernel or the correct set of kernel headers installed
    on your system.
    If you are using a Linux 2.6 kernel, please make sure
    you have configured kernel sources matching your kernel
    installed on your system. If you specified a separate
    output directory using either the "KBUILD_OUTPUT" or
    the "O" KBUILD parameter, make sure to specify this
    directory with the SYSOUT environment variable or with
    the equivalent nvidia-installer command line option.
    Depending on where and how the kernel sources (or the
    kernel headers) were installed, you may need to specify
    their location with the SYSSRC environment variable or
    the equivalent nvidia-installer command line option.
    *** Unable to determine the target kernel version. ***
    make: *** [select_makefile] Error 1
    ==> ERROR: Build Failed.
    The step it's choking on is (from the PKGBUILD), "make SYSSRC=/lib/modules/${_kernver}/build module || return 1"
    What am I doing wrong?
    pkgdesc="NVIDIA drivers for kernel26."
    arch=('i686' 'x86_64')
    [ "$CARCH" = "i686" ] && ARCH=x86
    [ "$CARCH" = "x86_64" ] && ARCH=x86_64
    depends=('kernel26>=2.6.31' 'kernel26<2.6.32' "nvidia-utils=${pkgver}")
    conflicts=('nvidia-96xx' 'nvidia-173xx')
    [ "$CARCH" = "x86_64" ] && md5sums=('ae431ff849ec01446e6724f9fcfe3bb4')
    build() {
    cd $srcdir
    sh NVIDIA-Linux-$ARCH-${pkgver} --extract-only
    cd NVIDIA-Linux-$ARCH-${pkgver}-pkg0
    cd usr/src/nv/
    ln -s Makefile.kbuild Makefile
    make SYSSRC=/lib/modules/${_kernver}/build module || return 1
    mkdir -p $pkgdir/lib/modules/${_kernver}/kernel/drivers/video/
    install -m644 nvidia.ko $pkgdir/lib/modules/${_kernver}/kernel/drivers/video/
    sed -i -e "s/KERNEL_VERSION='.*'/KERNEL_VERSION='${_kernver}'/" $startdir/nvidia.install
    My nvidia.install
    post_install() {
    depmod $KERNEL_VERSION
    post_upgrade() {
    rmmod nvidia || echo 'In order to use the new nvidia module, exit Xserver and unload it manually.'
    post_remove() {
    depmod $KERNEL_VERSION
    Last edited by graysky (2009-11-07 10:54:33)

    I'm kind of a noob and am trying to get follow the directions from the Nvidia wiki on installing on a custom kernel.  I'm trying to use the abs method to install on kernel26rt version 2.6.33rt.  I followed directions exactly and am getting an error that I didn't find on any other posts.  Can you please take a look?  Here is my terminal:
    [[email protected] nvidia]$ makepkg -ci
    ==> Making package: nvidia-2633rt 270.30-3 (Wed Mar 30 08:53:08 PDT 2011)
    ==> Checking runtime dependencies...
    ==> Checking buildtime dependencies...
    ==> Retrieving Sources...
      -> Downloading
    --2011-03-30 08:53:08-- …
               => “”
    Connecting to||:21... connected.
    Logging in as anonymous ... Logged in!
    ==> SYST ... done.    ==> PWD ... done.
    ==> TYPE I ... done.  ==> CWD (1) /XFree86/Linux-x86_64/270.30 ... done.
    ==> SIZE ... 30037601
    ==> PASV ... done.    ==> RETR ... done.
    Length: 30037601 (29M) (unauthoritative)
    100%[======================================>] 30,037,601   696K/s   in 42s     
    2011-03-30 08:53:50 (696 KB/s) - “” saved [30037601]
    ==> Validating source files with md5sums... ... Passed
    ==> Extracting Sources...
    ==> Starting build()...
    Creating directory NVIDIA-Linux-x86_64-270.30-no-compat32
    Verifying archive integrity... OK
    Uncompressing NVIDIA Accelerated Graphics Driver for Linux-x86_64 270.30........................................................................................................................
    NVIDIA: calling KBUILD...
    test -e include/generated/autoconf.h -a -e include/config/auto.conf || (    \
        echo;                                \
        echo "  ERROR: Kernel configuration is invalid.";        \
        echo "         include/generated/autoconf.h or include/config/auto.conf are missing.";\
        echo "         Run 'make oldconfig && make prepare' on kernel src to fix it.";    \
        echo;                                \
    mkdir -p /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/.tmp_versions ; rm -f /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/.tmp_versions/*
    make -f scripts/ obj=/home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel
      cc -Wp,-MD,/home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/.nv.o.d  -nostdinc -isystem /usr/lib/gcc/x86_64-unknown-linux-gnu/4.5.2/include -I/usr/src/linux-2.6.33-rt/arch/x86/include -Iinclude  -include include/generated/autoconf.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -O2 -m64 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -fstack-protector -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -Wframe-larger-than=2048 -fomit-frame-pointer -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fno-dwarf2-cfi-asm -fconserve-stack -I/home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel -Wall -MD -Wsign-compare -Wno-cast-qual -Wno-error -D__KERNEL__ -DMODULE -DNVRM -DNV_VERSION_STRING=\"270.30\" -Wno-unused-function -mcmodel=kernel -mno-red-zone -UDEBUG -U_DEBUG -DNDEBUG  -DMODULE -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(nv)"  -D"KBUILD_MODNAME=KBUILD_STR(nvidia)"  -c -o /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.o /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c
    In file included from /usr/src/linux-2.6.33-rt/arch/x86/include/asm/uaccess.h:573:0,
                     from include/linux/poll.h:13,
                     from /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv-linux.h:84,
                     from /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:13:
    /usr/src/linux-2.6.33-rt/arch/x86/include/asm/uaccess_64.h: In function ‘copy_from_user’:
    /usr/src/linux-2.6.33-rt/arch/x86/include/asm/uaccess_64.h:35:6: warning: comparison between signed and unsigned integer expressions
    In file included from /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:13:0:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv-linux.h: At top level:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv-linux.h:251:35: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘nv_spinlock_t’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv-linux.h:1436:5: error: expected specifier-qualifier-list before ‘nv_spinlock_t’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c: In function ‘nv_alloc_file_private’:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:1993:5: error: implicit declaration of function ‘semaphore_init’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:1994:5: error: ‘nv_file_private_t’ has no member named ‘waitqueue’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:1995:5: error: implicit declaration of function ‘atomic_spin_lock_init’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:1995:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c: In function ‘nv_kern_poll’:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:2790:30: error: ‘nv_file_private_t’ has no member named ‘waitqueue’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:2793:5: error: implicit declaration of function ‘atomic_spin_lock_irqsave’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:2793:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:2804:5: error: implicit declaration of function ‘atomic_spin_unlock_irqrestore’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:2804:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c: In function ‘nv_kern_close_registry’:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:3595:18: error: ‘nv_file_private_t’ has no member named ‘off’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:3611:22: error: ‘nv_file_private_t’ has no member named ‘off’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c: In function ‘nv_kern_write_registry’:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:3703:51: error: ‘nv_file_private_t’ has no member named ‘off’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:3710:45: error: ‘nv_file_private_t’ has no member named ‘off’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:3720:9: error: ‘nv_file_private_t’ has no member named ‘off’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c: In function ‘nv_post_event’:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:4330:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:4347:5: error: ‘nv_file_private_t’ has no member named ‘waitqueue’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:4348:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c: In function ‘nv_get_event’:
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:4362:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:4365:9: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    /home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.c:4390:5: error: ‘nv_file_private_t’ has no member named ‘fp_lock’
    make[3]: *** [/home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel/nv.o] Error 1
    make[2]: *** [_module_/home/jsadural/devel/abs/nvidia/src/NVIDIA-Linux-x86_64-270.30-no-compat32/kernel] Error 2
    NVIDIA: left KBUILD.
    nvidia.ko failed to build!
    make[1]: *** [module] Error 1
    make: *** [module] Error 2
    ==> ERROR: A failure occurred in build().

  • Can't install latest patch, "Patch Module for 32-Bit Version of Lenove Patch Utility" v 1.0.3

    I'm hoping someone can offer some help: I am unable to install the above patch on a T61 w/XP Pro/SP3. Error is simply, "1 Package did not install." Manual install gets same result. In the past when a patch did not instaqll I have had success by downloading and manually installing it, but I can't find the executable for this particular patch. Can anyone make a suggestion? Or point me to the executable?
    Go to Solution.

    This is how I was able to successfully install the Lenovo Patch Utility:
    1) Enable if necessary and log into the built-in Administrator account.
    2) Run System Update.  Download and install the Lenovo Patch Utility (unsure if this is a necessary step,)
    3) At this point, the Lenovo Patch Utility is not installed.  The previous version of the Lenovo Patch Utility will display in Programs and Features. 
    4) Run System Update under an administrator account.  Download and install the Lenovo Patch Utility (I ran this under a user-created Administrator account; unsure if works under built-in Administrator account..)  Lenovo Patch Utility will no longer appear in Programs and Features.  This indicates it has been succesfully installed.
    5) Disable built-in Administrator account if desired for security reasons.

  • How to compile zfs for an upgraded kernel, root fs is zfs[newbie]

    Hey all, I am new to Arch, but was a gentoo user before. I have a question  about how to recompile third party module for a new kernel. I can not find any threads about it.
    I have a system with root fs zfs by following the howto Installing_Arch_Linux_on_ZFS, everything works fine.
    Recently I did a pacman -Syu and the kernel is updated, errors occured. But I do not know how to fix the problem.
    [2012-11-01 01:26] ==> Building image from preset: 'default'
    [2012-11-01 01:26]   -> -k /boot/vmlinuz-linux-lts -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-lts.img
    [2012-11-01 01:26] ==> Starting build: 3.0.49-1-lts
    [2012-11-01 01:26]   -> Running build hook: [base]
    [2012-11-01 01:26]   -> Running build hook: [udev]
    [2012-11-01 01:26]   -> Running build hook: [autodetect]
    [2012-11-01 01:26]   -> Running build hook: [pata]
    [2012-11-01 01:26]   -> Running build hook: [scsi]
    [2012-11-01 01:26]   -> Running build hook: [sata]
    [2012-11-01 01:26]   -> Running build hook: [zfs]
    [2012-11-01 01:26] ==> ERROR: module not found: `zfs'
    [2012-11-01 01:26] ==> ERROR: module not found: `zcommon'
    [2012-11-01 01:26] ==> ERROR: module not found: `znvpair'
    [2012-11-01 01:26] ==> ERROR: module not found: `zavl'
    [2012-11-01 01:26] ==> ERROR: module not found: `zunicode'
    [2012-11-01 01:26] ==> ERROR: module not found: `spl'
    [2012-11-01 01:26]   -> Running build hook: [filesystems]
    [2012-11-01 01:26]   -> Running build hook: [usbinput]
    [2012-11-01 01:26] ==> Generating module dependencies
    [2012-11-01 01:26] ==> Creating gzip initcpio image: /boot/initramfs-linux-lts.img
    [2012-11-01 01:26] ==> WARNING: errors were encountered during the build. The image may not be complete.
    The errors are from third party spl/zfs modules. pacman -U spl/zfs only installs modules for currently running modules, not the new kernel. I can not boot into the new kernel and recompile the third party modules since the root fs is zfs. What should I do to recompile those modules for the newly updated kernel, before using the newly updated kernel?
    Last edited by blackwhite (2012-11-02 03:35:13)

    falconindy wrote:
    blackwhite wrote:Thank all. I have fixed the problem. It seems I have to use livecd with zfs to chroot the broken system, and just reinstall the newest linux kernel (pacman -Ulinux-lts-3.0.49-1-x86_64.pkg.tar.xz ), it will find the spl/zfs module and  mkinitcpio -p correctly.
    BTW, in the chroot environment, the spl/zfs AUR still failed to figure it which kernel the system use, always use the current running kernel version on livecd to compile.
    Right. chroot'ing doesn't change your kernel. There's assuredly build options to let you compile against a different kernel.
    Would you step further, give me the right options to build a aur package against a different kernel. Thanks.

  • Should modules for different kernels differ?

    I'm trying to build acpi_call-git from the AUR. The build seems to go rather too smoothly. I uninstalled a custom kernel (which I only have for debugging) which leaves the standard Arch and Arch LTS kernels installed. I have the headers installed for both kernels:
    $ pacman -Qs linux-[lts-]*headers
    local/linux-headers 3.11.4-1
    Header files and scripts for building modules for Linux kernel
    local/linux-lts-headers 3.10.15-1
    Header files and scripts for building modules for Linux-lts kernel
    What is confusing me is that the package seems to build identical modules for the two kernels and I am not sure whether that's right:
    $ diff pkg/acpi_call-git/usr/lib/modules/extramodules-3.11-ARCH/acpi_call.ko.gz pkg/acpi_call-git/usr/lib/modules/extramodules-3.10-lts/acpi_call.ko.gz
    $ echo $?
    Shouldn't there be a difference between the two? That is, I know that you typically need modules which match the kernel - otherwise the modules can't be loaded. So presumably if building module M for kernel K1 produces M1, and building module M for kernel K2 produces M2, then M1 cannot be identical with M2.
    Is there a problem here or am I just horribly confused?

    If the kernel functions (especially the specifications) used by acpi_call haven't changed, then the module could indeed be the same for both kernels.
    You'd have to check the kernel commits for acpi related changes to be sure.

  • STREAM module question

    Hi all,
    I am newbe in STREAM module development and this question might be a concept error. If so, please could you give me some clues?
    I would like to use the "libc" funcion "getpwuid()" in a kernel STREAM module.
    I have Solaris 7 64-bit kernel.
    �Can I do it? I have read that for 64-bit kernels, only dynamic libraries are available. Therefore in order ot use libc functions I would have to link dynamically the module relocatable object with the libc library in sparcv9. I have tried to link the kernel STREAM module with (in /usr/lib/sparcv9) with no success so far. Everytime I trie to load the module it complains with the message: unknown symbol "getpwuid".
    If it is not possible, why? . Is it not possible to use libc functions such us: memcpy() etc..from the kernel context?
    Looking forward to hearing from you.
    Best Regards

    You cannot use user space libraries from within a kernel module. What exactly are you trying to
    do? What do you need password information for
    down in the kernel?
    Maybe ddi_get_cred() will help?
    Some functions like memcopy have kernel space equivalents (bcopy)....

  • [SOLVED] No module for virtualbox-ose with 2.6.27

    Hi there,
    I just upgraded and noticed, that virtualbox isn't starting. Probing vboxdrv manually doesn't work because there is no module for the current kernel. The packages virtualbox-modules was upgraded with the system-upgrade. I'm using virtualbox-modules-2.0.2-3.pkg.tar.gz.
    So I'm compiling the modules manually with ABS right now, will report the results.
    Edit: After compiling and installing, there is still no working module
    Last edited by crom (2008-10-15 20:04:36)

    You only need to recompile virtualbox-modules. I did that yesterday and worked flawlessly.
    The thing you have to notice, is that you must update the kernel version defined in the PKGBUILD and virtualbox-modules.install:
    cp -r /var/abs/community/modules/virtualbox-modules .
    cd virtualbox-modules
    sed -i 's|2.6.26-ARCH|2.6.27-ARCH|' PKGBUILD virtualbox-modules.install

  • Compiling a driver, need assistance on kernel source configuration

    Hello BBS.
    I am compiling a module for the Asus PCE N53 wireless pci-e card. I have obtained the driver source, and followed the instructions, and have gotten to the point where I need assistance. The instructions requires the kernel source, i have successfully obtained and configured it using the Command line prompt, except for one feature. I need to enable the intel IPW2200 driver code in the kernel source for this driver. What procedure do I follow to enable it?

    > I know in face value, that my request may seem odd. I just need to use the source of the kernel, not a new kernel... I have found that what I am requesting... will fix the problem, based on a Gentoo solution: … art-0.html. Needing to know how to configure the kernel source to enable the intel driver still stands. The source I am using is the latest arch kernel source.
    RT5592 Linux Driver quick start       
    Check tools: 
    *Before install driver, please check already install compile tool and  kernel source code
    1>Install compile tool
        $yum install gcc-c++
    2>check kernel source code exists /usr/src/kernels/ "kernel name"
        Download your kernel source code
        $yum install kernel-devel
    Build Instructions: 
    1> $tar -jxvf DPO_GPL_RT5592STA_LinuxSTA_vx.x.x.x.tar.bz2
         go to "DPO_GPL_RT5592STA_LinuxSTA_vx.x.x.x" directory.
    2> In Makefile
         set the "MODE = STA" in Makefile and chose the TARGET to Linux by set "TARGET = LINUX"
         define the linux kernel source include file path LINUX_SRC
         modify to meet your need.
    3> In os/linux/
         define the GCC and LD of the target machine
         define the compiler flags CFLAGS
         modify to meet your need.
         ** Build for being controlled by NetworkManager or wpa_supplicant wext functions
             => $wpa_supplicant -Dwext -ira0 -c wpa_supplicant.conf -d
         ** Build for being controlled by WpaSupplicant with Ralink Driver
             => $wpa_supplicant -Dralink -ira0 -c wpa_supplicant.conf -d
    4> $make           
         # compile driver source code, need administrator.
         # To fix "error: too few arguments to function ¡¥iwe_stream_add_event"
            => $patch -i os/linux/sta_ioctl.c.patch os/linux/sta_ioctl.c
    5> $make install
         #install driver
         #copy RT2860STA.dat to /etc/Wireless/RT2860STA/RT2860STA.dat
    6>$vi /etc/rc.d/rc.local
         #input "ifconfig ra0 up"
    7> unload driver   
         $ifconfig ra0 down
         $rmmod rt5592sta
    Heres the output of my error, when I compile:
    make -C tools
    make[1]: Entering directory `/home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/tools'
    gcc -g bin2h.c -o bin2h
    make[1]: Leaving directory `/home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/tools'
    cp -f os/linux/Makefile.6 /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/Makefile
    make -C /home/white/linux-3.2.9 SUBDIRS=/home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux modules
    make[1]: Entering directory `/home/white/linux-3.2.9'
      WARNING: Symbol version dump /home/white/linux-3.2.9/Module.symvers
               is missing; modules will have no dependencies and modversions.
      CC [M]  /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.o
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2364:2: error: unknown field ‘private’ specified in initializer
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2364:2: warning: initialization from incompatible pointer type [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2364:2: warning: (near initialization for ‘rt28xx_iw_handler_def.get_wireless_stats’) [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2365:2: error: unknown field ‘num_private’ specified in initializer
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2365:2: warning: excess elements in struct initializer [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2365:2: warning: (near initialization for ‘rt28xx_iw_handler_def’) [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2366:2: error: unknown field ‘private_args’ specified in initializer
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2366:26: warning: excess elements in struct initializer [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2366:26: warning: (near initialization for ‘rt28xx_iw_handler_def’) [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2367:2: error: unknown field ‘num_private_args’ specified in initializer
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2367:2: warning: excess elements in struct initializer [enabled by default]
    /home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.c:2367:2: warning: (near initialization for ‘rt28xx_iw_handler_def’) [enabled by default]
    make[2]: *** [/home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux/../../os/linux/sta_ioctl.o] Error 1
    make[1]: *** [_module_/home/white/driver/DPO_GPL_RT5592STA_LinuxSTA_v2.6.0.0_20120326/os/linux] Error 2
    make[1]: Leaving directory `/home/white/linux-3.2.9'
    make: *** [LINUX] Error 2
    Last edited by confusedoldman (2012-12-25 20:14:34)

  • Pcmcia token ring card problems or how to compile a module?

    i have a pcmcia token ring ibm card. i run 2.6.23-ARCH kernel. I know that the correct  module is ibmtr_cs (from a previous distribution). When i trie to modprobe ibmtr_cs it tells me that  the module doesn't exist. Searching my system i found that in /usr/src/linux-2.6.23-ARCH/include/linux/ibmtr.h there is something like a module... Does this mean that i have to compile my own kernel? in this case how can i do it? is it possible to compile only the module?
    thanks nik
    Last edited by nicnikf (2008-01-18 19:57:08)

    nicnikf wrote:
    i have a pcmcia token ring ibm card. i run 2.6.23-ARCH kernel. I know that the correct  module is ibmtr_cs (from a previous distribution). When i trie to modprobe ibmtr_cs it tells me that  the module doesn't exist. Searching my system i found that in /usr/src/linux-2.6.23-ARCH/include/linux/ibmtr.h there is something like a module... Does this mean that i have to compile my own kernel? in this case how can i do it?
    thanks nik
    Maybe you only have to compile that module, for your kernel, instead of a whole kernel recompile? On how to do that, I don't know, sorry.
    Maybe someone else knows.

  • IMac8,1 running Snow Leopard with 64 Bit kernel and dual monitor

    I've got an early 2008 Imac with an ATI Radeon HD 2600 Pro card running Snow Leopard 10.6.3.
    When I run the kernel in 64 bit mode the second display turns on and off periodically. Running in 32 bit mode it's fine.
    Is this a known issue?

    AFAIK, Mac Pro (2006) is EFI32 and is not capable of booting the 64-bit kernel, but the system is fully capable of running 64-bit applications using the 32-bit kernel.
    This question can be confusing and context-dependent, and there can be different reasons for asking about the "64-bitness" of the kernel. Particularly whether the application support for 64-bit is available, and whether the kernel itself is 64-bit.
    For 64-bit application support. AFAIK, you have that.
    For 64-bit kernel support, and the associated requirements for having 64-bit kernel extension (kext) support for your applications, and AFAIK, you don't have that.
    See [Mac OS X v10.6: Starting up with the 32-bit or 64-bit kernel (ht3773)|] for the mechanisms available for Snow Leopard and Snow Leopard Server, if you want to test your current configuration. AFAIK, you have a 32-bit kernel.
    You can join the Apple developer program and then download the developer preview of [Mac OS X Lion release|> if you want to learn more about that topic.

  • Unable to compile module for Intel VTUNE with 3.16.4-1 kernel

    I am currently running  3.16.4-1-ARCH #1 and I tried to compile the kernel module for Intel Vtune XE 2013 (r17).
    Unluckily, GCC stops reporting the following errors:
    /opt/intel/vtune_amplifier_xe_2013/sepdk/src/vtsspp/collector.c: In function ‘vtss_target_new’:
    /opt/intel/vtune_amplifier_xe_2013/sepdk/src/vtsspp/collector.c:985:35: error: ‘VSYSCALL_START’ undeclared (first use in this function)
    addr = (unsigned long)VSYSCALL_START;
    /opt/intel/vtune_amplifier_xe_2013/sepdk/src/vtsspp/collector.c:985:35: note: each undeclared identifier is reported only once for each function it appears in
    /opt/intel/vtune_amplifier_xe_2013/sepdk/src/vtsspp/collector.c:986:36: error: ‘VSYSCALL_MAPPED_PAGES’ undeclared (first use in this function)
    size = (unsigned long)(VSYSCALL_MAPPED_PAGES * PAGE_SIZE)
    As the source code points out:
    #include <asm/fixmap.h> /* VSYSCALL_START */
    The missing symbols are expected to be found under /usr/lib/modules/3.16.4-1-ARCH/build/arch/x86/include/asm/fixmap.h, but it seems that they are not present anymore.
    Grepping the headers could not help at all.
    What am I missing here ?

    It looks like the macros:
    got dropped after v3.15
    I wonder if I need to downgrade.

  • Problem inserting module for my dvb-s2 card on any kernel after 2.6.35

    Firstly these drivers contain some proprietary bits so I have to download them from here. Basically it's some modified version of the cx23885 module from the mainline kernel.
    Secondly here is a 10 page discussion on the TBS forums about this driver and getting it to work with kernels after 2.6.35.
    Card is TBS 6891.
    These proprietary drivers seem to be working with other distros such as Fedora and Ubuntu as witnessed in the previously linked thread, but, for some reason when I try and insert the module on my Arch Linux x86_64 box I get the following error.
    # modprobe cx23885
    FATAL: Error inserting cx23885 (/lib/modules/2.6.38-ARCH/kernel/drivers/media/video/cx23885/cx23885.ko): Unknown symbol in module, or unknown parameter (see dmesg)
    # dmesg|tail -n1
    [ 2373.648484] cx23885: Unknown symbol v4l2_i2c_new_subdev_cfg (err 0)
    So I was wondering if anyone might be able to help with this. Maybe Fedora or Ubuntu have kernel config options different to Arch that I could try. Or kernel/v4l patches I could try.
    Last edited by frontwheelskid (2011-06-08 10:48:01)

    "cody" has replied in the tbs forums with information to get this compiled and working with arch linux:
    it seems though on ArchLinux you need to do one extra step before 'make install':
    # rm -rf /lib/modules/<kernel_version>/kernel/drivers/media/
    for example:
    # rm -rf /lib/modules/2.6.39-ARCH/kernel/drivers/media/
    because otherwise the existing V4L modules in ArchLinux mixed up in a wrong way with the one installed by TBS Linux driver package.
    I hope this info shortens other peoples searches!

Maybe you are looking for

  • Rest iPod touch is not recognized by itunes

    I was given a used ipod touch and i deleted everything off of it from the ipod. It then asks me to plug into itunes, which I have. Itunes does not pick up the device so I can upload all my stuff. What do I do?

  • Change log

    Hi, Is there any difference between the programs RSTBHIST ( Table history) and RSVTPROT(Evaluation of log changes)? regards, Beena

  • Adding a new toolbar button in IC to Save the transaction

    Hi, I want to add a new custom button to my IC Toolbar which would do save. I have gone through some blogs( ) and able to get a new button and subscribe an event

  • Consuming an Enterprise Service

    Hi All, I need to consume an Enterprise Service available on the ES Workplace. However, on the ES workplace, I could only locate the WSDL for the enterprise services. I was unable to find the business logic or code written to implement the services.

  • Can't find MobileMe account

    When I input my mobileme account name to try and set up MobileMe on the AppleTV, for some reason it says that it cannot find the account. I have done it the "correct way" (without on the end), as well as with the to make sure. I know th