Search found 842 matches

by mgates3
Mon Dec 10, 2018 10:04 am
Forum: User discussion
Topic: (d/s)getri_outofplace_batched can't handle many matrices
Replies: 2
Views: 214

Re: (d/s)getri_outofplace_batched can't handle many matrices

We'll track this on the bitbucket issue.
-mark
by mgates3
Fri Dec 07, 2018 1:35 pm
Forum: User discussion
Topic: Why is magma_ssyevd_gpu so slow?
Replies: 1
Views: 215

Re: Why is magma_ssyevd_gpu so slow?

What BLAS & LAPACK library are you using? What size is your problem? Are you timing just the magma_ssyevd_gpu call, or are you timing the runtime of the entire test program? You can run MAGMA's testers in magma/testing to get both MAGMA and LAPACK timings. There are several versions to try (note the...
by mgates3
Wed Dec 05, 2018 3:10 pm
Forum: User discussion
Topic: magmaf_zhegvd_batched?
Replies: 1
Views: 204

Re: magmaf_zhegvd_batched?

Sorry, we don't have batched eigenvalue problems yet.
-mark
by mgates3
Wed Dec 05, 2018 3:08 pm
Forum: User discussion
Topic: Random Butterfly Transformations
Replies: 4
Views: 254

Re: Random Butterfly Transformations

I'm not sure what you mean. The download for source code is here:
https://icl.utk.edu/magma/software/index.html

-mark
by mgates3
Wed Dec 05, 2018 11:00 am
Forum: User discussion
Topic: Random Butterfly Transformations
Replies: 4
Views: 254

Re: Random Butterfly Transformations

From scanning the code, it looks like depth=1, a single random butterfly is applied.
-mark
by mgates3
Tue Dec 04, 2018 12:00 pm
Forum: User discussion
Topic: Allocation and Execution on a single GPU
Replies: 1
Views: 238

Re: Allocation and Execution on a single GPU

Answering your questions in reverse order: 2. For allocating matrices on the GPU using Fortran, in the latest release (MAGMA 2.5 rc 1) there is magmaf_[sdcz]malloc (notice the f for Fortran in magmaf prefix). Otherwise, you can use the (very old) cuBLAS Fortran interface that comes in CUDA/src/fortr...
by mgates3
Mon Dec 03, 2018 10:11 am
Forum: User discussion
Topic: About time required for launching MAGMA kernels in host
Replies: 1
Views: 188

Re: About time required for launching MAGMA kernels in host

It's unclear what is occurring in your example. A sample run using one of the MAGMA testers, or at least sample code, so we can attempt to reproduce the issue is needed. What is the problem size? What is your system — OS, CUDA version, what CPU & GPU? Also, it would help to expanding the "[+] Comput...
by mgates3
Thu Nov 29, 2018 4:58 pm
Forum: User discussion
Topic: Magma-2.5.0-rc1
Replies: 4
Views: 628

Re: Magma-2.5.0-rc1

Done. It's just adding it as a flag in the Makefile. MAGMA should always run on any CUDA arch >= what it was compiled for. So if you compile MAGMA for sm_30, it should also run on sm_35 (Kepler K40), sm_50 (Maxwell), ..., sm_70 (Volta), sm_75 (Turing), etc. Compiling for a specific architecture migh...
by mgates3
Thu Nov 29, 2018 1:15 am
Forum: User discussion
Topic: Magma-2.5.0-rc1
Replies: 4
Views: 628

Re: Magma-2.5.0-rc1

We're looking into it. I don't think removing 3.0 capability was the intention.
-mark
by mgates3
Mon Nov 05, 2018 2:46 pm
Forum: User discussion
Topic: example_sparse.c failing to compile
Replies: 9
Views: 783

Re: example_sparse.c failing to compile

Yes, it compiles on Unix. (I'm actually using macOS.) g++ -Wall -fopenmp \ -I ${MAGMADIR}/include \ -I ${MAGMADIR}/sparse/include \ -o example_sparse example_sparse.cpp \ -L ${MAGMADIR}/lib -lmagma -lmagma_sparse \ -L ${CUDADIR} -lcublas -lcudart -lcusparse \ -L /opt/OpenBLAS/lib -lopenblas From the...