Questions tagged [magma]

MAGMA (Matrix Algebra on GPU and Multicore Architectures) is a collection of linear algebra libraries for heterogeneous processor architectures. If you have questions about the MAGMA computational algebra system, please use the [magma-ca] tag.

Designed by a team at the Innovative Computing Laboratory at the University of Tennessee, MAGMA allows applications to exploity the power of heterogeneous systems of multi-core CPU's and multi-GPU's to deliver the most accurate solution in the fastest time within specified energy constraints.

42 questions
7
votes
1 answer

CMake find_library does not find the library

I came up with the following super simple FindMAGMA.cmake script to find the MAGMA library given there is none around: # - Find the MAGMA library # # Usage: # find_package(MAGMA [REQUIRED] [QUIET] ) # # It sets the following variables: # …
SkyWalker
  • 13,729
  • 18
  • 91
  • 187
3
votes
0 answers

LAPACKE or MAGMA GPU - inversion of matrix with Cholesky factorization - functions magma_dpotrf_gpu and magma_dpotri_gpu

I have a first version of a function that inverses a matrix of size m and using magma_dgetrf_gpu and magma_dgetri_gpulike this : // Inversion magma_dgetrf_gpu( m, m, d_a, m, piv, &info); magma_dgetri_gpu( m, d_a, m, piv, dwork, ldwork,…
user1773603
3
votes
0 answers

Calling MAGMA from a Fortran program

I'm working on using MAGMA to offload an dsyevd call (which computes all the eigenvalues and the eigenvectors of a square real symmetric matrix A) from a Fortran code base, the subroutine that calls the MAGMA routine is below. subroutine sh_diag !…
Timmy
  • 55
  • 1
  • 5
2
votes
1 answer

Is there a way to use "unified memory" (MAGMA) with 2 GPU cards with NVLink and 1TB RAM

At work, On Debian 10, I have 2 GPU cards RTX A6000 with NVlink harware component with 1TB of RAM and I would like to benefit of the potential combined power of both cards and 1TB RAM. Currently, I have the following magma.make invoked by a Makefile…
user1773603
2
votes
1 answer

Why isn't Magma 2.5 compilable?

I am trying to compile Magma 2.5 on my laptop with NVIDIA GTX 960M, and it always prompts the linkage problem. I have upgraded cmake version to thew newest available (3.14.3). Below is part of the error output.…
2
votes
1 answer

MAGMA and Rcpp for linear algebra in R

I was wondering if anybody has tried to use Rcpp and MAGMA to accelerate linear algebra operations in R by using the CPU and GPU? I tried culatools last month and it worked with Rcpp (link), but culatools is a commercial product that costs money to…
chris
  • 461
  • 2
  • 10
1
vote
1 answer

How to properly use the -Xnvlink compiler options when we have 2 NVIDIA/CUDA GPU cards with NVLink hardware component

On Debian 10, I have 2 GPU cards RTX A6000 with NVlink hardware component and I would like to benefit of the potential combined power of both cards. Currently, I have the following magma.make invoked by a Makefile : CXX = nvcc -std=c++17 -O3 LAPACK…
user1773603
1
vote
0 answers

Execution of simple MAGMA inversion matrix distributed on two GPU cards

I have below a simple program from MAGMA doc which inverses a large matrix on a GPU card. #include #include #include "magma_v2.h" #include "magma_lapack.h" int main() { //double *d_lA[MagmaMaxGPUs]; magma_init (); //…
user1773603
1
vote
1 answer

How to compile code that uses MAGMA in Visual Studio

Given the example code that comes with MAGMA, how to compile it in Visual Studio (or any other code of your making that uses MAGMA) ?
WurmD
  • 1,231
  • 5
  • 21
  • 42
1
vote
1 answer

List of causes for LNK2038, MSVS2017 MAGMA

My goal is by compiling (pun not intended) a list of causes for LNK2038 "mismatch detected for '_ITERATOR_DEBUG_LEVEL': value '0' doesn't match value '2'" that others may follow methodically to debug their own situations, my situation will be…
WurmD
  • 1,231
  • 5
  • 21
  • 42
1
vote
1 answer

Why does this MEXed C/magma code seg-fault while the stand alone C code works?

The following MEXed C code simply makes calls to magma to invert a matrix. The stand alone C code (which is also posted) works, but the mex code crashes. I've triple checked the documentation, verified that other magma functions work as expected,…
avgn
  • 982
  • 6
  • 19
1
vote
0 answers

How does magma_dgetri use multiple GPUs

I just installed magma and I noticed that some routines have _mgpu version while some not. For example for LU inverse, there are 4 functions: magma_cgetri_gpu (magma_int_t n, magmaFloatComplex_ptr dA, magma_int_t ldda, magma_int_t *ipiv,…
Han
  • 9
  • 4
1
vote
1 answer

MAGMA: function "magma_dgels_gpu" --> error "magma_trans_t"

I am trying to solve a least squares problem via "magma_dgels_gpu()" function of MAGMA Library. My GPU is "Tesla C2050 / C2075" and i have installed MAGMA. I am trying to compile the below code "testMagmaDGELS.cu", but i get error: #include…
wasilis
  • 115
  • 1
  • 8
1
vote
1 answer

How to use Fortran interface to call a C function which contains user defined type

In fact, I want to call magma from fortran. So I add magma.lib and create an interface to use the C fuction of magma: Interface Integer function magma_dpotrf(uplo, n, a, lda, info) BIND (C, NAME="magma_dpotrf") use iso_c_binding …
1
vote
2 answers

MAGMA library: difference between magma_dgemm and magmablas_dgemm

In the most recent magma linear algebra library (version 1.6.1), http://icl.cs.utk.edu/magma/software/, in the testing code exercising dgemm functionality (source code: testing_dgemm.cpp), there was a call to functions magma_dgemm and…
Wirawan Purwanto
  • 3,613
  • 3
  • 28
  • 28
1
2 3