bgog...@osl.iu.edu wrote: > Author: bgoglin > Date: 2009-10-29 03:56:40 EDT (Thu, 29 Oct 2009) > New Revision: 1252 > URL: https://svn.open-mpi.org/trac/hwloc/changeset/1252 > > Log: > pciutils only got a .pc recently (in 2.2.6), so add configure code to > manually check for its headers and library > Text files modified: > branches/libpci/configure.ac | 10 ++++++++++ > > 1 files changed, 10 insertions(+), 0 deletions(-) > > Modified: branches/libpci/configure.ac > ============================================================================== > --- branches/libpci/configure.ac (original) > +++ branches/libpci/configure.ac 2009-10-29 03:56:40 EDT (Thu, 29 Oct > 2009) > @@ -376,6 +376,16 @@ > > if test "x$enable_pci" = "xyes"; then > PKG_CHECK_MODULES([PCI], [libpci], [:], [enable_pci="no"]) > + # manually check pciutils in case a old one without .pc is installed > + if test "x$enable_pci" = "xno"; then > + AC_CHECK_HEADERS([pci/pci.h], [ > + AC_CHECK_LIB([pci], [pci_cleanup], [enable_pci=yes]
By the way, I used AC_CHECK_LIB instead of AC_HAVE_LIBRARY without any good reason here. The only difference seems to be that the former checks for an actual function name in the lib. I don't know if/how the libpci ABI actually changed in the past. Is their any common sense driving such a choice? AC_SEARCH_LIBS is sometimes recommended but I don't think it matters for libpci. Brice
