Re: mc + compaq alpha

Hi, Libor!

lude -I/usr/local/include  -DREGEX_MALLOC -DHAVE_CONFIG_H -g mountlist.c
cc: Error: mountlist.c, line 238: In this statement, "f_fstypename" is not
a mem
ber of "fsp". (needmember)
            me->me_type = strdup (fsp->f_fstypename);

Obviously, the test for f_fstypename in struct fsstat is wrong. It was
added to mc-aclocal.m4 in the revision 1.4 by somebody with the login name
"danw". No other files have been committed in the same time, not even
ChangeLog. Too bad is not running finger. Dan, I guess it's

I don't have access to Tru64UNIX v5.1, but I could check how it works on
OpenBSD 2.7 that has f_fstypename. It is declared in sys/mount.h, but
searching for `f_type;' makes no sense - we should search for

Ideally, we should use AC_CHECK_MEMBER from Autoconf 2.50, but it's
missing in Autoconf 2.13, and I don't want to force everyone to switch.
Besides, AC_CHECK_MEMBER involves compiling, but I don't know what
includes are necessary for it to succeed on all systems.

That's why I've committed the following simple-minded patch:

--- acinclude.m4
+++ acinclude.m4
@@ -189,11 +189,10 @@
        AC_MSG_CHECKING([if struct statfs has f_fstypename])
-         [
-           AC_EGREP_HEADER(f_type;, sys/mount.h, ok=yes, ok=)
-           test -n "$ok" \
-               && fu_cv_sys_mounted_f_fstypename=yes \
-               || fu_cv_sys_mounted_f_fstypename=no
+         [AC_EGREP_HEADER([f_fstypename],
+                          [sys/mount.h],
+                          [fu_cv_sys_mounted_f_fstypename=yes],
+                          [fu_cv_sys_mounted_f_fstypename=no])
         if test $fu_cv_sys_mounted_f_fstypename = yes; then

The snapshot with the fixed test is available at

Please download and test it. Thank you for taking time to report the

Pavel Roskin

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]