MAGMA Forum

https://icl.utk.edu/magma/software/

Also the Mercurial repo at
https://bitbucket.org/icl/magma

-mark

Easiest to redirect the output:

MAGMA won't affect your computers performance while it is not running. While it is running, it will use the processor and GPU, so other programs would run slow. This is expected.
-mark

As for setting up a new source file for use in CMake, you need to run `make generate` or `make CMake.src`. I'm not sure why CMake wouldn't find `control/magma_sauxiliary.cpp`. What version of MAGMA are you using? Or did you get MAGMA directly from the bitbucket repo? Possibly the `make generate` wou...

Also, you can compile using `make magmablas/dstddev_mean.o` (from the top magma directory) to compile a single file to see if it is working, without recompiling the entire library. E.g., mint ~/Documents/magma> make magmablas/dznrm2.o nvcc -m64 -O3 -DNDEBUG -DADD_ -Xcompiler "-fPIC -Wall -Wno-unused...

Currently, CMake won't see new files. They have to be added with `make` first, which will generate magma/CMake.src. MAGMA uses a precision generator (magma/tools/codegen.py) to generate all 4 precisions (s, d, c, z for single, double, complex-single, complex-double) from, usually, the complex-double...

Shouldn’t matter, as long as they support CUDA. See Nvidia’s site.

Mark

If you want all the eigenvalues, use range = MagmaRangeAll, and you can ignore il, iu, vl, vu by passing in dummy values. If you want a subset of eigenvalues, use range = MagmaRangeV, and pass in vl, vu as the range of eigenvalues to find, e.g., to find all eigenvalues in half-open range (0.0, 1.0],...

LU is a direct method, not an iterative method (see Matlab code below). It forms matrices L, U such that PA = LU (exactly, in exact arithmetic). L has unit diagonal, U has non-zero diagonal (if it has a zero, then A is singular). There is no tolerance level epsilon in the algorithm. % LU factorizati...

If you want to write GPU kernels to compute the mean and stddev, you can model them after magmablas/dnrm2.cu, which computes the 2-norm of each column of an m-by-n matrix dA on the GPU. (You can ignore the adjust functions for updating 2-norms.) It is easiest to do in 2 kernels, one for mean, one fo...