On 02/24/2012 08:28 AM, Sylvestre Ledru wrote: > Hello, > > Le vendredi 24 février 2012 à 07:46 +0100, Michael Wild a écrit : >> Dear Debian Science Developers, >> >> I already tried mentioning that I am looking for an upload sponsor on >> the list (http://lists.debian.org/debian-science/2012/02/msg00023.html), >> but so far haven't had any replies. Also, my original sponsor hasn't >> replied to my message. This is why I try my luck again with a more >> formal RFS. > > It fails on my system with a cowbuilder chroot: > -- Vandermonde matrix -- > terminate called after throwing an instance of > 'viennacl::ocl::unknown_error' > what(): ViennaCL: FATAL ERROR: ViennaCL encountered an unknown OpenCL > error. In some cases, this might be due to an invalid global work size, > but it can also be due to several compilation errors. > If you think that this is a bug in ViennaCL, please report it at > [email protected] and supply at least the following > information: > * Operating System > * Which OpenCL implementation (AMD, NVIDIA, etc.) > * ViennaCL version > Many thanks in advance! > > Start 9: vector > 9/10 Test #9: vector ...........................***Exception: Other > 0.03 sec > > ---------------------------------------------- > ---------------------------------------------- > ## Test :: Vector > ---------------------------------------------- > ---------------------------------------------- > > > ---------------------------------------------- > > # Testing setup: > eps: 0.0001 > numeric: float > Running tests for vector of size 30000 > terminate called after throwing an instance of > 'viennacl::ocl::unknown_error' > what(): ViennaCL: FATAL ERROR: ViennaCL encountered an unknown OpenCL > error. In some cases, this might be due to an invalid global work size, > but it can also be due to several compilation errors. > If you think that this is a bug in ViennaCL, please report it at > [email protected] and supply at least the following > information: > * Operating System > * Which OpenCL implementation (AMD, NVIDIA, etc.) > * ViennaCL version > Many thanks in advance! > > Start 10: vector_range > 10/10 Test #10: vector_range .....................***Exception: Other > 0.00 sec > > ---------------------------------------------- > ---------------------------------------------- > ## Test :: Vector Range > ---------------------------------------------- > ---------------------------------------------- > > # Testing setup: > eps: 0 > numeric: float > terminate called after throwing an instance of > 'viennacl::ocl::unknown_error' > what(): ViennaCL: FATAL ERROR: ViennaCL encountered an unknown OpenCL > error. In some cases, this might be due to an invalid global work size, > but it can also be due to several compilation errors. > If you think that this is a bug in ViennaCL, please report it at > [email protected] and supply at least the following > information: > * Operating System > * Which OpenCL implementation (AMD, NVIDIA, etc.) > * ViennaCL version > Many thanks in advance! > > > 10% tests passed, 9 tests failed out of 10 > > Total Test time (real) = 0.10 sec > > The following tests FAILED: > 1 - blas3 (OTHER_FAULT) > 3 - iterators (OTHER_FAULT) > 4 - matrix (OTHER_FAULT) > 5 - matrix_range (OTHER_FAULT) > 6 - scalar (OTHER_FAULT) > 7 - sparse (OTHER_FAULT) > 8 - structured-matrices (OTHER_FAULT) > 9 - vector (OTHER_FAULT) > 10 - vector_range (OTHER_FAULT) > Errors while running CTest > make[1]: *** [test] Error 8 > make[1]: Leaving directory > `/tmp/buildd/viennacl-1.2.0/obj-x86_64-linux-gnu' > dh_auto_test: make -j1 test ARGS+=-j1 returned exit code 2 > make: *** [binary] Error 29 > dpkg-buildpackage: error: fakeroot debian/rules binary gave error exit > status 2 > > Sylvestre > > >
Hi Sylvestre The problem with running the tests in a chroot is that they need access to a NVIDIA graphics card and the proprietary drivers able to handle these tests. I also can't run them inside the chroot, but outside is no problem. Can you try running them outside of the chroot? Michael -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/[email protected]
