It appears that LC_MODULE_LOADING was accidentally declared twice
back-to-back in the same file. This removes the first declaration
on the assumption that if my eye-balling of the code missed a
difference, the second one is the one we've been using anyway.
Change-Id: I04a9da80d6be7bef6e4fd35eca8f3e490a8a824f
Signed-off-by: Christopher J. Morrone <morrone2@llnl.gov>
Reviewed-on: http://review.whamcloud.com/4407
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Keith Mannthey <keith.mannthey@intel.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
])
])
-# LC_MODULE_LOADING
-# after 2.6.28 CONFIG_KMOD is removed, and only CONFIG_MODULES remains
-# so we test if request_module is implemented or not
-AC_DEFUN([LC_MODULE_LOADING],
-[AC_MSG_CHECKING([if kernel module loading is possible])
-LB_LINUX_TRY_MAKE([
- #include <linux/kmod.h>
-],[
- int myretval=ENOSYS ;
- return myretval;
-],[
- $makerule LUSTRE_KERNEL_TEST=conftest.i
-],[
- grep request_module build/conftest.i | grep -v `grep "int myretval=" build/conftest.i | cut -d= -f2 | cut -d" " -f1` >/dev/null
-],[
- AC_MSG_RESULT(yes)
- AC_DEFINE(HAVE_MODULE_LOADING_SUPPORT, 1,
- [kernel module loading is possible])
-],[
- AC_MSG_RESULT(no)
- AC_MSG_WARN([])
- AC_MSG_WARN([Kernel module loading support is highly recommended.])
- AC_MSG_WARN([])
-])
-])
-
#
# LB_PROG_LINUX
#