ea1fdf46e2
The vmtruncate_range() support has been removed from the kernel in favor of using the fallocate method in the file_operations table. Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Issue #784
25 lines
715 B
Plaintext
25 lines
715 B
Plaintext
dnl #
|
|
dnl # 3.5.0 API change
|
|
dnl # torvalds/linux@17cf28afea2a1112f240a3a2da8af883be024811 removed
|
|
dnl # truncate_range(). The file hole punching functionality is now
|
|
dnl # provided by fallocate()
|
|
dnl #
|
|
AC_DEFUN([ZFS_AC_KERNEL_TRUNCATE_RANGE], [
|
|
AC_MSG_CHECKING([whether iops->truncate_range() exists])
|
|
ZFS_LINUX_TRY_COMPILE([
|
|
#include <linux/fs.h>
|
|
void truncate_range(struct inode *inode, loff_t start,
|
|
loff_t end) { return; }
|
|
static struct inode_operations iops __attribute__ ((unused)) = {
|
|
.truncate_range = truncate_range,
|
|
};
|
|
],[
|
|
],[
|
|
AC_MSG_RESULT(yes)
|
|
AC_DEFINE(HAVE_INODE_TRUNCATE_RANGE, 1,
|
|
[iops->truncate_range() exists])
|
|
],[
|
|
AC_MSG_RESULT(no)
|
|
])
|
|
])
|