Building with ACML
Building with ACML
ACML stands for AMD Core Math Library.
The AMD Core Math Library (ACML) incorporates BLAS, LAPACK and FFT routines, which are designed to be used by a wide range of software developers to obtain excellent performance from their applications running on AMD platforms. The highly optimized library contains numeric functions for mathematical, engineering, scientific and financial applications. ACML is available both as a 32-bit library, for compatibility with legacy x86 applications, and as a 64-bit library that is designed to fully exploit the large memory space and improved performance offered by the new AMD64 architecture.
The AMD Core Math Library (ACML) incorporates BLAS, LAPACK and FFT routines, which are designed to be used by a wide range of software developers to obtain excellent performance from their applications running on AMD platforms. The highly optimized library contains numeric functions for mathematical, engineering, scientific and financial applications. ACML is available both as a 32-bit library, for compatibility with legacy x86 applications, and as a 64-bit library that is designed to fully exploit the large memory space and improved performance offered by the new AMD64 architecture.
To build with ACML, you will need to download the ACML source,
configure, and then link it while compiling. These steps, and any
known limitations, are described in the sections below:
Downloading ACML
You can download ACML from the
AMD website. Look for
the PathScale versions. AMD Core Math Library (ACML) 2.6, built with
the PathScale compilers,is now available.
Configuring and linking ACML
To link ACML with your code at compile time, add -L path_to_acml_install_dir -lacml
to your compile command line.
There are a number of available ACML libraries:
For serial programs choose the appropriate directory for programs compiled with
-m64 (default on 64-bit systems) and -m32:
/opt/acml3.6.0/pathscale64/lib
/opt/acml3.6.0/pathscale32/lib
For programs compiled with -mp or -openmp, choose the appropriate directory for programs compiled with -m64 (default on 64-bit systems) and -m32:
/opt/acml3.6.0/pathscale64_mp/lib
/opt/acml3.6.0/pathscale32_mp/lib
For 64-bit programs serial or OpenMP that use INTEGER*8 (-i8) integers as indices for arrays:
/opt/acml3.6.0/pathscale64_int64/lib
/opt/acml3.6.0/pathscale64_mp_int64/lib
Note that the ACML version number will need to be updated for newer releases.
/opt/acml3.6.0/pathscale64/lib
/opt/acml3.6.0/pathscale32/lib
For programs compiled with -mp or -openmp, choose the appropriate directory for programs compiled with -m64 (default on 64-bit systems) and -m32:
/opt/acml3.6.0/pathscale64_mp/lib
/opt/acml3.6.0/pathscale32_mp/lib
For 64-bit programs serial or OpenMP that use INTEGER*8 (-i8) integers as indices for arrays:
/opt/acml3.6.0/pathscale64_int64/lib
/opt/acml3.6.0/pathscale64_mp_int64/lib
Note that the ACML version number will need to be updated for newer releases.
Known limitations
There are no known limitations.
We recommend you record what steps you took, along with their output, in case of problems. You can include this in your email to support. Here is an example of changing a make command to record the output in a text file:
We recommend you record what steps you took, along with their output, in case of problems. You can include this in your email to support. Here is an example of changing a make command to record the output in a text file:
make foo 2>&1 | tee make-foo-log.txtIf you have any comments or suggestions about additions to these pages, please contact support@pathscale.com.