Gravitational Wave search O2 Multi-Directional GPU v2.07 (GW-opencl-nvidia) windows_x86_64

Forums › Problems and Bug Reports

luzi
luzi
Joined: 28 Oct 06
Posts: 2
Credit: 515888096
RAC: 0
20 Apr 2020 14:47:49 UTC
Topic 222051
(moderation:
    )

    Hi,

    seit einigen Tagen bekomme ich die o. a. WUs auf einem meiner Rechner. Diese brechen jedoch nach ca 1 Minute ab:

    <message>
(unknown error) - exit code 1024 (0x400)</message>

    ...

    DEPRECATION WARNING: program has invoked obsolete function XLALGetVersionString(). Please see XLALVCSInfoString() for information about a replacement.
Code-version: %% LAL: 6.19.2.1 (CLEAN 98bbe72a728eb25935e9195dafae691335dabf8c)
%% LALPulsar: 1.17.1.1 (CLEAN 98bbe72a728eb25935e9195dafae691335dabf8c)
%% LALApps: 6.23.0.1 (CLEAN 98bbe72a728eb25935e9195dafae691335dabf8c)

    Was will mir diese Meldung sagen? Was kann bzw. muss ich tun? Beispiel:

    Name: h1_1352.45_O2C02Cl4In0__O2MDFV2g_VelaJr1_1352.90Hz_114_0


    Job ID: 451124732


     

    Richie
    Richie
    Joined: 7 Mar 14
    Posts: 656
    Credit: 1702989778
    RAC: 0

    Hi ! NVIDIA GeForce GTX 660

    20 Apr 2020 14:59:28 UTC
    Message 176890
    (moderation:
      )

      Hi !

      NVIDIA GeForce GTX 660 (1536MB) driver: 431.07

      Ihre GPU verfügt nicht über genügend Speicher, um GW GPU Arbeit auszuführen. 4 GB sind bereits erforderlich. FGRPB1G wird gut funktionieren.

      Ian&Steve C.
      Ian&Steve C.
      Joined: 19 Jan 20
      Posts: 3958
      Credit: 47006742642
      RAC: 64925811

      Processing FFT failed:

      20 Apr 2020 15:04:51 UTC
      Message 176891
      (moderation:
        ) 

        Processing FFT failed: CL_MEM_OBJECT_ALLOCATION_FAILURE

         

        you do not have enough memory on that card.

        _________________________________________________________________________

        luzi
        luzi
        Joined: 28 Oct 06
        Posts: 2
        Credit: 515888096
        RAC: 0

        Hi,vielen Dank für die

        20 Apr 2020 15:54:04 UTC
        Message 176895 in response to message 176890
        (moderation:
          )

          Hi,

          vielen Dank für die schnelle Antwort - wie kann ich verhindern, dass ich diese WUs erhalte?

          Die

          Gamma-ray pulsar binary search #1 on GPUs v1.22 (FGRPopencl-nvidia)
          Gamma-ray pulsar search #5 v1.08 (FGRPSSE)

          laufen auf meinen Rechnern problemlos.

           Sorry, hab's schon gefunden!

          Ian&Steve C.
          Ian&Steve C.
          Joined: 19 Jan 20
          Posts: 3958
          Credit: 47006742642
          RAC: 64925811

          uncheck them from your

          20 Apr 2020 15:55:59 UTC
          Message 176896 in response to message 176895
          (moderation:
            )

            uncheck them from your project preferences.

            here:

            _________________________________________________________________________

            Holmis
            Joined: 4 Jan 05
            Posts: 1118
            Credit: 1055935564
            RAC: 0

            Binary Radio Pulsar Search

            20 Apr 2020 21:18:46 UTC
            Message 176908 in response to message 176896
            (moderation:
              )

              Binary Radio Pulsar Search (Arecibo, GPU) has been out of tasks for a long time now, no need to try and get those.

              Gamma-Ray pulsar binary Search #1 (GPU), (also known as FGRPB1G) has task readily available. Wink

              Luigi Naruszewicz
              Luigi Naruszewicz
              Joined: 18 Jul 05
              Posts: 8
              Credit: 220880955
              RAC: 17168

              Since the weekend I have been

              20 Apr 2020 22:41:15 UTC
              Message 176913
              (moderation:
                )

                Since the weekend I have been having computation errors GPU Gravitational Wave Search units. This error report that I am getting.

                Stderr output

                <core_client_version>7.16.5</core_client_version>
<![CDATA[
<message>
The target internal file identifier is incorrect.
 (0x72) - exit code 114 (0x72)</message>
<stderr_txt>
putenv 'LAL_DEBUG_LEVEL=3'
2020-04-20 16:33:17.4056 (10652) [normal]: This program is published under the GNU General Public License, version 2
2020-04-20 16:33:17.4056 (10652) [normal]: For details see http://einstein.phys.uwm.edu/license.php
2020-04-20 16:33:17.4056 (10652) [normal]: This Einstein@home App was built at: Dec 19 2019 12:14:49

                2020-04-20 16:33:17.4212 (10652) [normal]: Start of BOINC application 'projects/einstein.phys.uwm.edu/einstein_O2MDF_2.07_windows_x86_64__GW-opencl-nvidia.exe'.
                Activated exception handling...
                [DEBUG} GPU type: 1
                [DEBUG} got GPU info from BOINC
                [DEBUG} got VendorID 4318
                2020-04-20 16:33:17.4681 (10652) [debug]: BSGL output files
                2020-04-20 16:33:17.4994 (10652) [debug]: Flags: LAL_DEBUG, OPTIMIZE, HS_OPTIMIZATION, GC_SSE2_OPT, X64, SSE, SSE2, GNUC X86 GNUX86
                2020-04-20 16:33:17.6712 (10652) [debug]: Set up communication with graphics process.

                DEPRECATION WARNING: program has invoked obsolete function XLALGetVersionString(). Please see XLALVCSInfoString() for information about a replacement.
                Code-version: %% LAL: 6.19.2.1 (CLEAN 98bbe72a728eb25935e9195dafae691335dabf8c)
                %% LALPulsar: 1.17.1.1 (CLEAN 98bbe72a728eb25935e9195dafae691335dabf8c)
                %% LALApps: 6.23.0.1 (CLEAN 98bbe72a728eb25935e9195dafae691335dabf8c)

                Updated to latest BOINC client and latest NVidea drivers with no effect. The change my computer was the Microsoft windows update on Tuesday.

                Is the problem at my end or with the units ?

                 

                Gary Roberts
                Gary Roberts
                Moderator
                Joined: 9 Feb 05
                Posts: 5872
                Credit: 117680185940
                RAC: 35185922

                Luigi Naruszewicz wrote:Is

                21 Apr 2020 0:45:15 UTC
                Message 176924 in response to message 176913
                (moderation:
                  )

                  Luigi Naruszewicz wrote:
                  Is the problem at my end or with the units ?

                  The problem is that your 2GB GPU doesn't have enough memory to crunch the latest VelaJr1 GW tasks.

                  You didn't quite go far enough into the stderr output.  I picked one of your failed tasks and here is the critical bit that explains the cause of the problem:-

                  Transferring host memory to GPU failed: CL_MEM_OBJECT_ALLOCATION_FAILURE

                  Until we get to GW GPU tasks that require less memory, your best bet is to change preferences so that you only accept the FGRPB1G tasks.  You will have no similar problem with them.

                  Cheers,
                  Gary.

                  Luigi Naruszewicz
                  Luigi Naruszewicz
                  Joined: 18 Jul 05
                  Posts: 8
                  Credit: 220880955
                  RAC: 17168

                  Thanks Gary, changed my

                  21 Apr 2020 5:40:48 UTC
                  Message 176929
                  (moderation:
                    )

                    Thanks Gary, changed my preferences and will abandon the GW tasks that I have. Apologies to wingmen.

                     

                    Betreger
                    Betreger
                    Joined: 25 Feb 05
                    Posts: 992
                    Credit: 1591739025
                    RAC: 777773

                    Computation errors are back

                    21 Apr 2020 14:47:17 UTC
                    Message 176937
                    (moderation:
                      )

                      Computation errors are back on GWs.

                      https://einsteinathome.org/workunit/451013041
                      https://einsteinathome.org/workunit/451013044
                      https://einsteinathome.org/workunit/451013038
                      https://einsteinathome.org/workunit/451013039
                      Ian&Steve C.
                      Ian&Steve C.
                      Joined: 19 Jan 20
                      Posts: 3958
                      Credit: 47006742642
                      RAC: 64925811

                      All of the computers that are

                      21 Apr 2020 20:28:28 UTC
                      Message 176948 in response to message 176937
                      (moderation:
                        )

                        All of the computers that are throwing errors do not have enough GPU memory for processing of GW work. 

                        it seems you need at least a 4GB GPU.  

                        _________________________________________________________________________

                        Comment viewing options

                        Select your preferred way to display the comments and click "Save settings" to activate your changes.

                        Forums › Problems and Bug Reports