View Source

HPC2020: Building your programs

The following exercises will help you understand how to build your own software on the ATOS HPCF or ECS.

Building a simple program

Ensure your environment is clean by running:
module reset
Create a directory for this tutorial and cd into it:
mkdir -p compiling_tutorial cd compiling_tutorial

We are going to use a simple program that will display versions of different libraries linked to it. Create a file called versions.c using your favourite editor with the following contents:

#include <stdio.h>
#include <hdf5.h>
#include <netcdf.h>
#include <eccodes.h>

int main() {
    #if defined(__INTEL_LLVM_COMPILER)
        printf("Compiler: Intel LLVM %d\n", __INTEL_LLVM_COMPILER);
    #elif defined(__INTEL_COMPILER)
        printf("Compiler: Intel Classic %d\n", __INTEL_COMPILER);
    #elif defined(__clang_version__)
        printf("Compiler: Clang %s\n",  __clang_version__);
    #elif defined(__GNUC__)
        printf("Compiler: GCC %d.%d.%d\n", __GNUC__, __GNUC_MINOR__, __GNUC_PATCHLEVEL__);
    #else
        printf("Compiler information not available\n");
    #endif
    
    // HDF5 version
    unsigned majnum, minnum, relnum;
    H5get_libversion(&majnum, &minnum, &relnum);
    printf("HDF5 version: %u.%u.%u\n", majnum, minnum, relnum); 

    // NetCDF version
    printf("NetCDF version: %s\n", nc_inq_libvers());

    // ECCODES version
    printf("ECCODES version: ");
    codes_print_api_version(stdout);
    printf("\n");

    return 0;
}

Try to naively compile this program with:
$CC -o versions versions.c

The compilation above fails as it does not know where to find the different libraries. We need to add some additional flags so the compiler can find both the include headers and link to the actual libraries.

Let's use the existing software installed on the system with modules, and benefit from the corresponding environment variables *_DIR which are defined in them to manually construct the include and library flags:

$CC -o versions versions.c -I$HDF5_DIR/include -I$NETCDF4_DIR/include -I/$ECCODES_DIR/include -L$HDF5_DIR/lib -lhdf5 -L$NETCDF4_DIR/lib -lnetcdf -L$ECCODES_DIR/lib -leccodes

Load the appropriate modules so that the line above completes successfully and generates the versions executable:

You will need to load the following modules to. have those variables defined.:

module load hdf5 netcdf ecmwf-toolbox
$CC -o versions versions.c -I$HDF5_DIR/include -I$NETCDF4_DIR/include -I/$ECCODES_DIR/include -L$HDF5_DIR/lib -lhdf5 -L$NETCDF4_DIR/lib -lnetcdf -L$ECCODES_DIR/lib -leccodes

The versions executable should now be in your current directory:

ls versions

Run ./versions. You will get an error such as the one below:

$ ./versions
./versions: error while loading shared libraries: libhdf5.so.200: cannot open shared object file: No such file or directory

While you passed the location of the libraries at compile time, the program cannot not find the libraries at runtime. Inspect the executable with ldd to see what libraries are missing

ldd is a utility that prints the shared libraries required by each program or shared library specified on the command line:

$ ldd versions
        linux-vdso.so.1 (0x00007ffffada9000)
        libhdf5.so.200 => not found
        libnetcdf.so.19 => not found
        libeccodes.so => not found
        libc.so.6 => /lib64/libc.so.6 (0x000014932ff36000)
        /lib64/ld-linux-x86-64.so.2 (0x00001493302fb000)

Can you make that program run successfully?

While you passed the location of the libraries at compile time, the program cannot not find the libraries at runtime. There are two solutions:

Use the environment variable LD_LIBRARY_PATH- not recommended for long term

Use the environment variable LD_LIBRARY_PATH. Check that ldd with the environment variable defined reports all libraries found:

LD_LIBRARY_PATH=$HDF5_DIR/lib:$NETCDF4_DIR/lib:$ECCODES_DIR/lib ldd ./versions

Use rpath - robust solution

Use the rpath strategy to engrave the library paths into the actual executable at link time, so it always knows where to find them at runtime. Rebuild your program with:

$CC -o versions versions.c -I$HDF5_DIR/include -I$NETCDF4_DIR/include -I/$ECCODES_DIR/include -L$HDF5_DIR/lib -Wl,-rpath,$HDF5_LIB -lhdf5 -L$NETCDF4_DIR/lib -Wl,-rpath,$NETCDF4_DIR/lib -lnetcdf -L$ECCODES_DIR/lib -Wl,-rpath,$ECCODES_DIR/lib -leccodes

Check that ldd with the environment variable defined reports all libraries found:

unset LD_LIBRARY_PATH
ldd ./versions

Final version

For convenience, all those software modules define the *_INCLUDE and *_LIB variables:

module show hdf5 netcdf4 ecmwf-toolbox | grep -e _INCLUDE -e _LIB

You can use those in for your compilation directly, with the following simplified compilation line:

$CC -o versions versions.c $HDF5_INCLUDE $NETCDF4_INCLUDE $ECCODES_INCLUDE $HDF5_LIB $NETCDF4_LIB $ECCODES_LIB

Now you can run your program without any additional settings:

$ ./versions
Compiler: GCC 8.5.0
HDF5 version: <HDF5 version>
NetCDF version: <NetCDF version> of <date> $
ECCODES version: <ecCodes version>

Can you rebuild the program so it uses the "old" versions of all those libraries in modules? Ensure the output of the program matches the versions loaded in modules? Do the same with the latest.

You need to load the desired versions or the modules:

module load hdf5/old netcdf4/old ecmwf-toolbox/old

And then rebuild and run the program:

$CC -o versions versions.c $HDF5_INCLUDE $NETCDF4_INCLUDE $ECCODES_INCLUDE $HDF5_LIB $NETCDF4_LIB $ECCODES_LIB
./versions

The output should match the versions loaded by the modules:

echo $HDF5_VERSION $NETCDF4_VERSION $ECCODES_VERSION

Repeat the operation with

module load --latest hdf5 netcdf4 ecmwf-toolbox

To simplify the build process, let's create a simple Makefile for this program. With your favourite editor, create a file called Makefile in the same directory with the following contents:

#
# Makefile
#
# Make sure all the relevant modules are loaded before running make

EXEC = versions

# TODO: Add the necessary variables into CFLAGS and LDFLAGS definition
CFLAGS = 
LDFLAGS = 

all: $(EXEC)

%: %.c 
	$(CC) -o $@ $^ $(CFLAGS) $(LDFLAGS)

test: $(EXEC)
	./$(EXEC)

clean:
	rm -f $(EXEC)

You can test it works by running:

make clean ldd test

Edit the Makefile and add the *_INCLUDE and *_LIB variables which are defined by the modules:

#
# Makefile
#
# Make sure all the relevant modules are loaded before running make

EXEC = versions

CFLAGS = $(HDF5_INCLUDE) $(NETCDF4_INCLUDE) $(ECCODES_INCLUDE)
LDFLAGS = $(HDF5_LIB) $(NETCDF4_LIB) $(ECCODES_LIB)

all: $(EXEC)

%: %.c 
	$(CC) -o $@ $^ $(CFLAGS) $(LDFLAGS)

test: $(EXEC)
	./$(EXEC)

ldd: $(EXEC)
	@ldd $(EXEC) | grep -e netcdf.so -e eccodes.so -e hdf5.so 

clean:
	rm -f $(EXEC)

Then run it with:

make clean ldd test

Using different toolchains: prgenv

So far we have used the default compiler toolchain to build this program. Because of the installation paths of the library, it is easy to see both the version of the library used as well as the compiler flavour with ldd:

$ make ldd
        libhdf5.so.200 => /usr/local/apps/hdf5/<HDF5 version>/GNU/8.5/lib/libhdf5.so.200 (0x000014f612b7d000)
        libnetcdf.so.19 => /usr/local/apps/netcdf4/<NetCDF version>/GNU/8.5/lib/libnetcdf.so.19 (0x000014f611f2a000)
        libeccodes.so => /usr/local/apps/ecmwf-toolbox/<ecCodes version>/GNU/8.5/lib/libeccodes.so (0x000014f611836000)

Rebuild the program with:

The default GNU GCC compiler.
The default Classic Intel compiler.
The default LLVM-based Intel compiler.
The default AMD AOCC.

Use the following command to test and show what versions of the libraries are being used at any point:

make clean ldd test

You can perform this test with the following one-liner, exploiting the prgenv module:

for pe in gnu intel intel-llvm amd; do ml prgenv/$pe; make clean ldd test; echo "******************"; done

Pay attention to the following aspects:

The Lmod module command informs you that it has reloaded the corresponding modules when changing the prgenv. This ensures the libraries used in your program are built with the same compiler for maximum compatibility.
The compiler command changes automatically, since we are using the environment variable $CC in the Makefile.
The include and library flags in the compilation lines are adapted automatically based on the libraries loaded.
The final binary is linked with the corresponding libraries for the version of the compiler as shown by ldd output.

Real-world example: CDO

To put into practice what we have learned so far, let's try to build and install CDO. You would typically not need to build this particular application, since it is already available as part of the standard software stack via modules or easily installable with conda. However, it is a good illustration of how to build a real-world software with dependencies to other software packages and libraries.

The goal of this exercise is for you to be able to build CDO and install it under one of your storage spaces (HOME or PERM), and then successfully run:

<PREFIX>/bin/cdo -V

You will need to:

Familiarise yourself with the installation instructions of this package in the official documentation.
Decide your installation path and your build path.
Download the source code from the CDO website.
Set up your build environment (i.e. modules, environment variables) for a successful build
Build and install the software
Test that works with the command above.

Make sure that CDO is built at least with support to:

NetCDF
HDF5
SZLIB (hint: use AEC)
ecCodes
PROJ
CMOR
UDUNITS

It is strongly recommended you bundle all your build process in a job script that you can submit in batch. That way you can request additional cpus and speed up your compilation exploiting build parallelism with make -j

If you would like a starting point for such a job, you can start from the following example, adding and amending the necessary bits as needed:

#!/bin/bash
#SBATCH -J build_cdo
#SBATCH -o %x.out
#SBATCH -n 8

set -x
set -e
set -u
set -o pipefail

# Get the URL and VERSION for the latest CDO available
URL=$(curl -s https://code.mpimet.mpg.de/projects/cdo/files | grep attachments/download  | sed -e "s:.*href=\"\(.*\)\".*:https\://code.mpimet.mpg.de\1:" | head -n 1)
VERSION=$(echo $URL | sed -e "s:.*/cdo-\(.*\).tar.gz:\1:")

# TODO: Define installation prefix and Build directory
# Hint: Use somewhere in your PERM for installation.
PREFIX=
BUILDDIR=

# Move to our BUILD DIRECTORY
mkdir -p $BUILDDIR
cd $BUILDDIR

# Download source
[ -f cdo-$VERSION.tar.gz ] || wget $URL
[ -d cdo-$VERSION ] || tar xvf cdo-$VERSION.tar.gz
cd cdo-$VERSION

# TODO: Define the environment for the build 


# TODO: Configure the build


# Build
make -j $SLURM_NTASKS

# Install
make install

# Check installed binary
$PREFIX/bin/cdo -V

This is the complete job script to

#!/bin/bash
#SBATCH -J build_cdo
#SBATCH -o %x.out
#SBATCH -n 8

set -x
set -e
set -u
set -o pipefail

# Get the URL and VERSION for the latest CDO available
URL=$(curl -s https://code.mpimet.mpg.de/projects/cdo/files | grep attachments/download  | sed -e "s:.*href=\"\(.*\)\".*:https\://code.mpimet.mpg.de\1:" | head -n 1)
VERSION=$(echo $URL | sed -e "s:.*/cdo-\(.*\).tar.gz:\1:")

# Define installation prefix and Build directory
PREFIX=$PERM/apps/cdo/$VERSION
#BUILDDIR=$TMPDIR
BUILDDIR=$PERM/apps/cdo/build

# Move to our BUILD DIRECTORY
mkdir -p $BUILDDIR
cd $BUILDDIR

# Download source
[ -f cdo-$VERSION.tar.gz ] || wget $URL
[ -d cdo-$VERSION ] || tar xvf cdo-$VERSION.tar.gz
cd cdo-$VERSION

# Define the environment for the build 
module load aec hdf5 netcdf4 udunits proj cmor ecmwf-toolbox

# We will need to explicitly set rpath for proj and eccodes since CDO build system will not
export LDFLAGS="-Wl,-rpath,$proj_DIR/lib -Wl,-rpath,$ECCODES_DIR/lib"

# Configure the build
./configure --prefix=$PREFIX --with-eccodes=$ECCODES_DIR --with-hdf5=$HDF5_DIR --with-netcdf=$NETCDF4_DIR --with-szlib=$AEC_DIR --with-proj=$proj_DIR --with-cmor=$CMOR_DIR --with-udunits2=$UDUNITS_DIR

# Build
make -j $SLURM_NTASKS

# Install
make install

# Check installed binary
$PREFIX/bin/cdo -V

You can submit it to the batch system with

sbatch build_cdo.sh

While it builds, you may want to keep an eye on the progress with:

tail -f build_cdo.out

Make sure the job completes successfully and that the output of the CDO executable you built is what you would expect.