Stack Overflow
Stack Overflow

Questions tagged [cusp-library]

A library for sparse linear algebra and graph computations on CUDA.

The Cusp library is a library for sparse linear algebra and graph computations on . Cusp provides a flexible, high-level interface for manipulating sparse matrices and solving sparse linear systems.

38 questions
0
votes
1 answer

CUSP kernel codes

I was wondering if I could find the kernel codes for the spmv and conversions in CUSP library. I scanned the whole library but couldn't find it. Is that proprietary or something like that??
Jayavanth
  • 40
  • 4
0
votes
1 answer

Use CUSP matrix inside CUDA function?

i want to write a kernel function that takes as input 2 CUSP matrices A and B,then fills data into B in parallel. #include #include #include __global__ void…
nza96
  • 41
  • 4
0
votes
1 answer

Real scaled Sparse matrix vector multiplication in Cusp?

In cusp, there is a multiply to calculate spmv(sparse matrix vector multiplication) that takes a reduce and a combine: template
IslandGod
  • 37
  • 3
0
votes
1 answer

cusp::extract_diagonal not found

I'm using CUSP for part of my GPU accelerated code. I have a need to extract the diagonal from a cusp matrix, C. Which should be put into diagonal in the below code. cusp::extract_diagonal(C, diagonal); However on compilation I get "cusp" has no…
Josh Dyson
  • 1
0
votes
1 answer

Memory use in cuda cusp linear solver

I am using cusp::bicgstab to solve a linear system Ax=b, in which A is a 3D Poisson on MxNxP mesh, x is unknowns, and b is the RHS. I have a K40m Tesla which has 12GB memory. I tested with M=2000, N=2000, P=20 (80 millions unknowns), variable type…
PLe
  • 37
  • 7
0
votes
1 answer

cuSPARSE multiplication function outputting incorrect value

I am trying to implement sparse matrix multiplication using cuSPARSE library. I have used most of the code from the documentation, that is from here. Even though I am getting the correct row pointers, column pointers of the output, I am getting…
yath
  • 355
  • 1
  • 4
  • 14
0
votes
1 answer

Unresolved references using IFORT with nvcc and CUSP

I have a program which I'm compiling like this: (...) Some ifort *.f -c nvcc -c src/bicgstab.cu -o bicgstab.o -I/home/ricardo/apps/cusp/cusplibrary (...) Some more *.for -c ifort *.o -L/usr/local/cuda-5.5/lib64 -lcudart -lcublas -lcusparse…
José Ricardo Ribeiro
  • 709
  • 9
  • 23
0
votes
1 answer

CUSP sparse library giving strange results

I'm trying to do the following in CUSP: A=[ 1,1,0,0; 2,2,2,0; 0,3,3,3; 0,0,4,4]; B=[1,1,1,1]'; disp(mldivide(A,B)); which is X=[0.9167,0.0833,-0.5000,0.7500] On the other hand I get a strange answer from CUSP #include…
Mikhail
  • 7,749
  • 11
  • 62
  • 136
0
votes
1 answer

CUSP host function parallelization with openMP

Is the CUSP host function also implemented in parallel with openMP? I am a bit confused since in the github discussion it says "added functionality in openMP" but there is no clear statement in the cusp that it is fully implemented with openMP.
user3040447
  • 11
  • 3
0
votes
1 answer

How to get raw pointer from cusp library matrix format

I need to get raw pointer from cusp library matrix format. For example: cusp::coo_matrix A(3,3,4); A.values[0] = 1; A.row_indices[0] = 0; A.column_indices[0]= 1; A.values[1] = 2; A.row_indices[1] =…
Jakub Hvězda
  • 1
  • 2
0
votes
2 answers

Multiply 2 sparse matrices using cusp library

Im new to using cusp library for cuda. I'm trying to implement revised simplex algorithm for CUDA. For that I need to multiply 2 sparse matrices to update the base matrix. So the question is - how can I multiply 2 sparse matrices(doesn't really…
Jakub Hvězda
  • 1
  • 2
0
votes
1 answer

Using cusp in a Visual Studio C++ project

I'm using cusp v.0.4.0, cuda V5.5 on VS2012 Ultimate. I created a CUDA project using the new project wizard and added the cusp path to it's project properties\VC++ Directories\Include Directories. I wrote my code in the *.cu file generated by…
Alexander1991
  • 217
  • 5
  • 13
0
votes
1 answer

CUSP CG convergence

I use CUSP conjugate gradient method to solve my symmetric sparse matrix. And I have no idea why it doesn't converge. Dimensions of matrices I use are not that large (1K to 100K). The same linear systems are easily solved by MKL, so the matrix is…
Max
  • 441
  • 2
  • 7
  • 14
0
votes
1 answer

Accessing cusp variable element from device kernel

I have a problem to access and assign variable with cusp array1d type from device/global kernel. The attached code gives error alay.cu(8): warning: address of a host variable "p1" cannot be directly taken in a device function alay.cu(8): error:…
user3040447
  • 11
  • 3
0
votes
1 answer

cusp inherited Class and *.cu device code compilation

Pardon me if this is non-trivial. I have a small class, wrapping around cusp::csr_matrix like so: class CuspMatrix : public cusp::csr_matrix { ... } So far I'm not doing anything I…
Ælex
  • 14,432
  • 20
  • 88
  • 129
1
2
3