Search found 279 matches

by Stan Tomov
Tue Oct 23, 2012 2:43 am
Forum: User discussion
Topic: Installation error with MAGMA1.2.1
Replies: 1
Views: 2010

Re: Installation error with MAGMA1.2.1

Try linking with these libraries

Code: Select all

LIB       = -lmkl_gf_lp64 -lmkl_intel_thread -lmkl_core -liomp5 -lpthread -lcublas -lm -fopenmp
or consult with the MKL's link adviser at http://software.intel.com/sites/products/mkl/
Stan
by Stan Tomov
Thu Sep 13, 2012 10:25 am
Forum: User discussion
Topic: Problem with testing_zgesv
Replies: 12
Views: 9419

Re: Problem with testing_zgesv

Upon further investigation we found that the problem is with Intel's compiler. This CUDA release note summarizes the issue: There is a known bug in ICC with respect to passing 16-byte aligned types by value to GCC-built code such as the CUDA Toolkit libraries (e.g., CUBLAS). At this time, passing a ...
by Stan Tomov
Wed Sep 05, 2012 11:02 am
Forum: User discussion
Topic: wishlist
Replies: 4
Views: 2760

Re: wishlist

This routine becomes more compute intensive when eigenvectors are needed. In that case most of the flops are in gemm and this is what is GPU accelerated.
Stan
by Stan Tomov
Tue Jul 17, 2012 6:22 pm
Forum: User discussion
Topic: FADDS_GEMM and FMULS_GEMM in flops.h do same operation
Replies: 2
Views: 2278

Re: FADDS_GEMM and FMULS_GEMM in flops.h do same operation

Hi,
This is just for convenience in maintaining the codes in different precisions. We keep one version in double complex and generate all the rest with a python script. Besides this there is no other reason to have two macros.
Stan
by Stan Tomov
Tue Jul 17, 2012 6:05 pm
Forum: User discussion
Topic: Possible bug in magmablas_dswap [MAGMA 1.1.0 and 1.2*]
Replies: 5
Views: 6742

Re: Possible bug in magmablas_dswap [MAGMA 1.1.0 and 1.2*]

Hello,
I think the bug is actually in the example file, in particular, you have used expressions like sizeof(x1), which in your case is 32 (4 doubles), instead of sizeof(x[0]) or just sizeof(double).
Stan
by Stan Tomov
Tue Jul 17, 2012 1:52 pm
Forum: User discussion
Topic: something about zgetrf_gpu
Replies: 1
Views: 1243

Re: something about zgetrf_gpu

Hello, Most probably there is problem with your CPU libraries (BLAS and LAPACK). To test this, you can run for example ./testing_zgetrf -M 20 -N 20 which will execute entirely on the CPU. If the error is not of order 1e-16 the result will be wrong. For example, on one of our systems I get [tomov@cum...
by Stan Tomov
Mon Jul 16, 2012 11:01 am
Forum: User discussion
Topic: 64-bit integer in MAGMA
Replies: 14
Views: 10134

Re: 64-bit integer in MAGMA

Hi Evan,
Yes, change the definition of magma_int_t to be a 64 integer (e.g. int64_t) and link with CPU BLAS and LAPACK with 64-bit integers. You can see the example in make.inc.int64 - basically, if you are using MKL you have to substitute mkl_intel_ilp64 in place of mkl_intel_lp64.
Stan
by Stan Tomov
Fri Jul 13, 2012 11:50 am
Forum: User discussion
Topic: 64-bit integer in MAGMA
Replies: 14
Views: 10134

Re: 64-bit integer in MAGMA

The 64-bit integers are now available in the MAGMA 1.2.1. Did you try using using 64 bit integer (you also have to link with the 64-bit CPU BLAS & LAPACK).
Stan
by Stan Tomov
Tue Jul 10, 2012 1:05 pm
Forum: User discussion
Topic: Matrix multiplication
Replies: 1
Views: 2110

Re: Matrix multiplication

Double check the parameters (alpha, beta, transposes) and the initialization. I see you initialize 3x2 matrix A and multiply 3x3 matrices. I don't see also the matrix C being initialized.
Stan
by Stan Tomov
Tue Jul 10, 2012 12:58 pm
Forum: User discussion
Topic: wanna help for magma1.2.1
Replies: 5
Views: 2752

Re: wanna help for magma1.2.1

This points to a problem with LAPACK or BLAS on the CPU. Problem of size 12x12 is very small and MAGMA directly uses LAPACK, so no GPUs are involved in this case. Please verify if your CPU LAPACK and BLAS are installed correctly.
Stan