Search found 897 matches

by mgates3
Fri May 04, 2012 5:46 pm
Forum: User discussion
Topic: magma_dgeev
Replies: 2
Views: 1752

Re: magma_dgeev

The magma dgeev code is a hybrid algorithm -- it uses both the CPU and the GPU to solve the system. Currently magma provides only the interface that takes the data on the CPU. Internally it copies blocks to the GPU during certain operations such as the Hessenberg factorization, while other operation...
by mgates3
Fri May 04, 2012 3:43 pm
Forum: User discussion
Topic: SGEMM when beta=0
Replies: 5
Views: 3498

Re: SGEMM when beta=0

Here are some results on a Tesla T20 (Fermi), Magma 1.1, CUDA 4.0, Intel MKL. (Source code attached.) The first column of C is set to NAN on input. When beta=0, implementations differ on whether they propagate NAN values or not. In particular, MAGMA BLAS does propagate NAN values. Therefore, you do ...
by mgates3
Fri May 04, 2012 10:49 am
Forum: User discussion
Topic: SGEMM when beta=0
Replies: 5
Views: 3498

Re: SGEMM when beta=0

When you didn't initialize the matrix, what result were you getting, and what result did you expect? A short sample code would be helpful. Technically, the beta*C must be carried out to properly propagate NAN values that may be in the C matrix. I'll look into how some different libraries handle this...
by mgates3
Fri May 04, 2012 10:43 am
Forum: User discussion
Topic: Using sgemm for rectangular (non-square) matrix multiply
Replies: 7
Views: 8062

Re: Using sgemm for rectangular (non-square) matrix multiply

The gemm in CUBLAS (and in MAGMA BLAS) works for rectangular matrices. As far as I know, all the CUBLAS work for all matrix sizes. If you are seeing some problems, please report the specific case. A short sample code would be helpful.
-mark
by mgates3
Thu May 03, 2012 12:38 pm
Forum: User discussion
Topic: Problem with testing_zgesv
Replies: 12
Views: 9032

Re: Problem with testing_zgesv

I could not replicate your problem, so a bit more investigation is needed to diagnose it. Clearly the magma_zgesv routine is failing and returning early, hence the erroneous Gflop/s speeds. It's confusing why no error is reported, though. What do you get for testing_zgetrf ? What about other precisi...
by mgates3
Thu May 03, 2012 10:54 am
Forum: User discussion
Topic: Compile Magma dll on 64-bit Win7 with visual Studio 2008
Replies: 1
Views: 2180

Re: Compile Magma dll on 64-bit Win7 with visual Studio 200

From the function names and version numbers, it looks like you are referring to a different magma software than the MAGMA linear algebra library that ICL produces. You will have to contact the appropriate vendor for support.
-mark
by mgates3
Wed May 02, 2012 7:35 pm
Forum: User discussion
Topic: magma_zgetri_gpu segfaults in OpenMP parallel for
Replies: 14
Views: 10587

Re: magma_zgetri_gpu segfaults in OpenMP parallel for

Yes, we're working on moving to the newer CUBLAS v2 API, which provides thread safety and better support for running multiple kernels in parallel.
-mark
by mgates3
Wed May 02, 2012 7:27 pm
Forum: User discussion
Topic: MAGMA 1.0
Replies: 16
Views: 18970

Re: MAGMA 1.0

See CUBLAS, provided by NVIDIA. It implements all BLAS level 1, 2, 3 functions, including syrk. The MAGMA BLAS library only supplements CUBLAS when we need functionality not covered by BLAS (e.g., LACPY) or we develop a faster implementation.
-mark
by mgates3
Wed May 02, 2012 7:18 pm
Forum: User discussion
Topic: potrs : does it do host-gpu memory transfers
Replies: 2
Views: 1681

Re: potrs : does it do host-gpu memory transfers

Yes, magma_potrs_gpu does data transfers between the CPU and the GPU. It factors the matrix by blocks. Each diagonal block is copied to the CPU, factored there, then copied back to the GPU. On the GPU, the rest of the panel below the diagonal block is updated. If your system is tridiagonal, you will...
by mgates3
Wed May 02, 2012 7:02 pm
Forum: User discussion
Topic: MAGMA on windows or linux
Replies: 6
Views: 3139

Re: MAGMA on windows or linux

I think you need to add LAPACK. The ATLAS library provides the BLAS routines, not the higher-level LAPACK routines. For what it's worth, here is how I link a generic program with ATLAS (but not using magma). Your libraries may differ. # -lifcore resolves undefined reference to `for_write_seq_fmt' # ...