Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

configure script cuda autodetection #69

Merged
merged 1 commit into from Jan 17, 2013
Merged

configure script cuda autodetection #69

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
134 changes: 89 additions & 45 deletions configure.ac
View file
Open in desktop
Expand Up @@ -405,61 +405,105 @@ AS_IF([test x$fftw_found = xyes],[
##################################
# check for CUDA
AC_MSG_CHECKING([whether to use CUDA])

AC_ARG_WITH([cuda],
AS_HELP_STRING([--with-cuda=CUDAINSTALLDIR],[specify where CUDA is installed.
AS_HELP_STRING([--with-cuda[=CUDAINSTALLDIR]],[specify where CUDA is installed.
Defaults to /usr/local/cuda.
The cuda compiler can also be specified by setting the NVCC
environment variable. The CUDA library and header can be
manually specified by using CPPFLAGS, LDFLAGS and LIBS.]),
, with_cuda=no)
, with_cuda=guess)
AC_MSG_RESULT($with_cuda)

if test x$with_cuda != xno; then
# if installation dir is given, set the paths
if test x$with_cuda != xyes; then
if test x$NVCC = x; then
NVCC=$with_cuda/bin/nvcc
fi
if test -d $with_cuda/lib64; then
LDFLAGS="$LDFLAGS -L$with_cuda/lib64"
else
LDFLAGS="$LDFLAGS -L$with_cuda/lib"
fi
NVCCFLAGS="$NVCCFLAGS -I$with_cuda/include"
fi
if test x$with_cuda = xguess -o x$with_cuda = xyes; then
cuda_path=/usr/local/cuda
elif test x$with_cuda != xno; then
cuda_path=$with_cuda
with_cuda=yes
fi

# NVCC
AC_ARG_VAR(NVCC,[NVIDIA CUDA compiler command])
AC_ARG_VAR(NVCCFLAGS,[special compiler flags for the NVIDIA CUDA compiler])
AC_PATH_PROG(NVCC, nvcc, no)
if test x$NVCC = xno; then
AC_MSG_FAILURE([CUDA compiler nvcc was not found, specify location using the NVCC variable])
fi
cuda_ok=no

# runtime library
AC_CHECK_LIB(cudart, cudaGetDevice, [LIBS="$LIBS -lcudart"], [AC_MSG_FAILURE([could not find cuda runtime library (cudart), specify location using LDFLAGS])])
if test x$with_cuda != xno; then
cuda_ok=yes

if test x$NVCC = x; then
NVCC=$cuda_path/bin/nvcc
fi

if test -d $cuda_path/lib64; then
LDFLAGS="$LDFLAGS -L$cuda_path/lib64"
else
LDFLAGS="$LDFLAGS -L$cuda_path/lib"
fi

NVCCFLAGS="$NVCCFLAGS -I$cuda_path/include"

# NVCC
AC_ARG_VAR(NVCC,[NVIDIA CUDA compiler command])
AC_ARG_VAR(NVCCFLAGS,[special compiler flags for the NVIDIA CUDA compiler])
AC_PATH_PROG(NVCC, nvcc, no)

if test x$NVCC = xno; then
AC_MSG_FAILURE([CUDA compiler nvcc was not found, specify location using the NVCC variable])
fi

# libraries
AC_CHECK_LIB(cudart, cudaGetDevice, [LIBS="$LIBS -lcudart"], [
cuda_ok=no

AS_IF([test x$with_cuda = xyes],[
AC_MSG_FAILURE([could not find cuda runtime library (cudart), specify location using LDFLAGS])
],[
AC_MSG_WARN([could not find cuda runtime library (cudart). Some features will not be available!])
])
])

AC_CHECK_LIB(cufft, cufftPlan3d, [LIBS="$LIBS -lcufft"], [
cuda_ok=no

AS_IF([test x$with_cuda = xyes],[
AC_MSG_FAILURE([could not find cuda FFT library (cufft), specify location using LDFLAGS])
],[
AC_MSG_WARN([could not find cuda FFT library (cufft). Some features will not be available!])
])
])

# NVCC compile check
AC_MSG_CHECKING([whether CUDA works])
# if no other compute capability is defined by the user, we need at least 1.1
case "$NVCCFLAGS" in
# NVCC compile check
AC_MSG_CHECKING([whether CUDA works])

# if no other compute capability is defined by the user, we require at least 1.1
case "$NVCCFLAGS" in
*-arch=*) ;;
*) NVCCFLAGS="$NVCCFLAGS --ptxas-options=-v -gencode arch=compute_11,code=compute_11 -gencode arch=compute_20,code=compute_20"
esac
# use nvcc
save_CC=$CC
save_CFLAGS=$CFLAGS
# autoconf currently doesn't allow to define new languages like cuda, so try to work around it
CC=$NVCC
CFLAGS="$NVCCFLAGS"
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([
#include <cuda.h>
],[
cudaGetDevice(0);
])], [AC_MSG_RESULT(yes)], [AC_MSG_FAILURE([cannot compile CUDA code. Look at config.log for more details.])])
CC=$save_CC
CFLAGS=$save_CFLAGS
AC_DEFINE(CUDA,[],[Whether CUDA is available])
*) NVCCFLAGS="$NVCCFLAGS --ptxas-options=-v -gencode arch=compute_11,code=compute_11 -gencode arch=compute_20,code=compute_20"
esac

# use nvcc
save_CC=$CC
save_CFLAGS=$CFLAGS

# autoconf currently doesn't allow to define new languages like cuda, this is a workaround
CC=$NVCC
CFLAGS="$NVCCFLAGS"

AC_COMPILE_IFELSE([AC_LANG_PROGRAM([#include <cuda.h>],[cudaGetDevice(0);])], [
AC_MSG_RESULT(yes)
],[
cuda_ok=no

AS_IF([test x$with_cuda = xyes],[
AC_MSG_FAILURE([cannot compile CUDA code. Look at config.log for more details.])
],[
AC_MSG_WARN([cannot compile CUDA code. Some features will not be available!])
])
])

CC=$save_CC
CFLAGS=$save_CFLAGS

AC_DEFINE(CUDA,[],[Whether CUDA is available])
fi

AM_CONDITIONAL(CUDA, [test x$with_cuda != xno])

cat <<EOF
Expand Down Expand Up @@ -593,7 +637,7 @@ dot = $DOT
python = $PYTHON
EOF

if test x$with_cuda != xno; then
if test x$cuda_ok = xyes; then
cat <<EOF
CUDA compiler = $NVCC
CUDA flags = $NVCCFLAGS
Expand Down