Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 29

Thread: Howto: Install Nvidia CUDA in Ubuntu 10.04

  1. #11
    Join Date
    Jan 2011
    Beans
    3

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Does the same driver/toolkit/installation method works for ubuntu_10.10_amd64?
    I am really having a tough time to install cuda on ubuntu_10.10_amd64.

    I have 420m gt which is capable of running cuda and indeed i am able to run cuda on it in win7/vc++2008. In Ubuntu 10.10-amd64 everything installs fine but when i run deviceQuery it gives:
    CUDA Device Query (Runtime API) version (CUDART static linking)

    cudaGetDeviceCount FAILED CUDA Driver and Runtime version may be mismatched.

    FAILED
    and on running bandwidthTest the response is :
    Running on...

    Quick Mode

    bandwidthTest.cu(598) : cudaSafeCall() Runtime API error : no CUDA-capable device is detected.

    Any help please.

  2. #12
    Join Date
    Nov 2010
    Beans
    8

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Thought i'd confirm that this tutorial with newest devdriver, toolkit & sdk examples all work fine.

    Ubuntu 10.04 LTS 64bit (updated 29/1/11)
    devdriver_3.2_linux_64_260.19.26.run
    cudatoolkit_3.2.16_linux_64_ubuntu10.04.run
    gpucomputingsdk_3.2.16_linux.run


    Had to do the below however: -

    Code:
    cd /usr/lib/
    sudo rm libGL.so
    sudo ln -s libGL.so.260.19.26 libGL.so
    But worked a treat.

    Thanks.
    Last edited by mancubas; January 31st, 2011 at 09:49 AM.

  3. #13
    Join Date
    May 2010
    Location
    Lahore Pakistan
    Beans
    Hidden!
    Distro
    Ubuntu 10.04 Lucid Lynx

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Why you guys tested these types of software in 64 bit. there are 32 bit users in the world and i think they are alive like me
    so what about 32 bit any idea how to install it in 32 bit 10.04 ubuntu.

  4. #14
    Join Date
    Feb 2011
    Beans
    1

    Exclamation still one issue running models

    hi all, and thanks for the great tutorial, still one problem:
    i'm using ubuntu 10.04 64 bit with cuda 3.2.16 64 and devdriver 260.19.26 on geforce 8400m gs make performs well but when running some model like nbody, here what pops up:

    > Windowed mode
    > Simulation data stored in video memory
    > Single precision floating point simulation
    freeglut (./nbody): ERROR: Internal error <FBConfig with necessary capabilities not found> in function fgOpenWindow
    X Error of failed request: BadWindow (invalid Window parameter)
    Major opcode of failed request: 4 (X_DestroyWindow)
    Resource id in failed request: 0x0
    Serial number of failed request: 30
    Current serial number in output stream: 33


    Any idea? thanks in advance to any given help

    devicequery runs correctly, and gives:

    CUDA Device Query (Runtime API) version (CUDART static linking)

    There is 1 device supporting CUDA

    Device 0: "GeForce 8400M GS"
    CUDA Driver Version: 3.20
    CUDA Runtime Version: 3.20
    CUDA Capability Major/Minor version number: 1.1
    Total amount of global memory: 267714560 bytes
    Multiprocessors x Cores/MP = Cores: 2 (MP) x 8 (Cores/MP) = 16 (Cores)
    Total amount of constant memory: 65536 bytes
    Total amount of shared memory per block: 16384 bytes
    Total number of registers available per block: 8192
    Warp size: 32
    Maximum number of threads per block: 512
    Maximum sizes of each dimension of a block: 512 x 512 x 64
    Maximum sizes of each dimension of a grid: 65535 x 65535 x 1
    Maximum memory pitch: 2147483647 bytes
    Texture alignment: 256 bytes
    Clock rate: 0.80 GHz
    Concurrent copy and execution: Yes
    Run time limit on kernels: Yes
    Integrated: No
    Support host page-locked memory mapping: Yes
    Compute mode: Default (multiple host threads can use this device simultaneously)
    Concurrent kernel execution: No
    Device has ECC support enabled: No
    Device is using TCC driver mode: No

    deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 3.20, CUDA Runtime Version = 3.20, NumDevs = 1, Device = GeForce 8400M GS


    PASSED

  5. #15
    Join Date
    May 2006
    Beans
    5
    Distro
    Ubuntu 10.04 Lucid Lynx

    Talking Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Thanks for this wonderful tutorial , I've installed successfully in:
    Ubuntu-10.04.1-desktop-i386 32 bit (downloaded yesterday)
    cudatoolkit_3.2.16_linux_32_ubuntu10.04.run
    devdriver_3.2_linux_32_260.19.26.run
    gpucomputingsdk_3.2.16_linux.run

    I have some comments for noobes like me
    - there is a typo in step 2.1, use sudo or gksudo instead of gsudo
    - in 6.1 and 7.1 and 9.1 I've used something like sudo sh file.run
    - after 7.1 i did a reboot
    - in 8.3 i didn't include /usr/local/cuda/lib64 that directory doesn't exists in my system
    - 9.3 was not necessary for me, the correct link was there

  6. #16
    Join Date
    Dec 2010
    Location
    Taiwan
    Beans
    2
    Distro
    Ubuntu 10.10 Maverick Meerkat

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Thanks ThomasArildsen for your tutorial, excelent work man!
    And thanks also to mancumbas for your contribution with the details regarding with the libGL.so link problem.

    I finished compiling the SDK samples files with out errors using my home computer. (I'm currently using the NVIDIA drv 260.19.26 on Ubuntu Maverick 10.10 with a Quadro FX580 GPU.)

    Now I plant to do the same in a Xeon server with 4 Tesla C1060, lets see how it goes with this one!

  7. #17
    Join Date
    Mar 2009
    Beans
    12

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    I did the sudo apt-get --purge remove nvidia-*

    and I got the following Error - E: Couldn' t find package nvidia-*

    any sugestions?

  8. #18
    Join Date
    Feb 2011
    Beans
    10

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Quote Originally Posted by Jem Masters View Post
    I did the sudo apt-get --purge remove nvidia-*

    and I got the following Error - E: Couldn' t find package nvidia-*

    any sugestions?
    It's not a problem dude because I have faced same problem in my Dev Driver installation but Installation process finished without any error.

  9. #19
    Join Date
    Feb 2011
    Beans
    10

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    I have already successfully installed CUDA in Ubuntu 10.04 LTS - 32 Bit. But when make examples which provided by NVIDIA, it gives following error but some of examples are successfully make how can I fixed those errors.

    make[1]: Entering directory `/home/giha/NVIDIA_GPU_Computing_SDK/C/src/nbody'
    /usr/bin/ld: cannot find -lGL
    collect2: ld returned 1 exit status
    make[1]: *** [../../bin/linux/release/nbody] Error 1
    make[1]: Leaving directory `/home/giha/NVIDIA_GPU_Computing_SDK/C/src/nbody'
    make: *** [src/nbody/Makefile.ph_build] Error 2


    I have already try out following solutions but problem still occur

    1. sudo make

    2. cd /usr/lib/
    sudo rm libGL.so
    sudo ln -s libGL.so.260.19.21 libGL.so

  10. #20
    Join Date
    Feb 2011
    Beans
    10

    Re: Howto: Install Nvidia CUDA in Ubuntu 10.04

    Quote Originally Posted by Gihan Lakmal View Post
    I have already successfully installed CUDA in Ubuntu 10.04 LTS - 32 Bit. But when make examples which provided by NVIDIA, it gives following error but some of examples are successfully make how can I fixed those errors.

    make[1]: Entering directory `/home/giha/NVIDIA_GPU_Computing_SDK/C/src/nbody'
    /usr/bin/ld: cannot find -lGL
    collect2: ld returned 1 exit status
    make[1]: *** [../../bin/linux/release/nbody] Error 1
    make[1]: Leaving directory `/home/giha/NVIDIA_GPU_Computing_SDK/C/src/nbody'
    make: *** [src/nbody/Makefile.ph_build] Error 2


    I have already try out following solutions but problem still occur

    1. sudo make

    2. cd /usr/lib/
    sudo rm libGL.so
    sudo ln -s libGL.so.260.19.21 libGL.so
    I have successfully resolved previous error.
    Solution: install Dev Driver again


    but now it gives following error any one have any idea, how to fixed it.


    obj/release/bitonicSort.cu.o:tmpxft_000017f4_00000000-1_bitonicSort.cudafe1.cpp: (.text+0x1290): first defined here
    collect2: ld returned 1 exit status
    make[1]: *** [../../bin/linux/release/sortingNetworks] Error 1
    make[1]: Leaving directory `/home/giha/NVIDIA_GPU_Computing_SDK/C/src/sortingNetworks'
    make: *** [src/sortingNetworks/Makefile.ph_build] Error 2

Page 2 of 3 FirstFirst 123 LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •