In order to be mounted, a filesystem needed to both match against
noXX and not match against XX.
Use the ternary operator, which also makes the intention clearer.
---
Hi!

W piątek, 6 lutego 2026 10:26:47 CET Roberto E. Vargas Caballero pisze:
> First, sorry for the delay answering the mail, but I was busy with some
> other stuff. I think the patch is right, but it does not apply over the
> current master. Can you rebase it and adapt it?

No worries.

I based my patch on Michael Forney's fork and I thought mount.c was in
sync. Sorry for not checking.

Also, now that I compare the forks, he has quite a few patches that might
be worth taking a look at:
https://github.com/michaelforney/ubase/compare/e3dacbb...master

I will probably try to upstream some particularly useful ones in the
coming weeks.

There are more problems with mount.c I would like to address, including
not skipping swap, leaking file descriptors and doing extra work due to
suboptimal condition ordering.  For the start I wanted to just fix this
one serious documentation misalignment.

Kind regards,
Arusekk

 mount.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mount.c b/mount.c
index e3d00b8..ebd52bc 100644
--- a/mount.c
+++ b/mount.c
@@ -307,12 +307,12 @@ mountall:
                }
                parseopts(fsopts, &flags, data, sizeof(data));
                /* if -t types specified:
-                * if non-match, skip
+                * if non-match and not prefixed with "no", skip
                 * if match and prefixed with "no", skip */
                if (types &&
-                   ((types[0] == 'n' && types[1] == 'o' &&
-                    findtype(types + 2, me->mnt_type)) ||
-                    (!findtype(types, me->mnt_type))))
+                   (types[0] == 'n' && types[1] == 'o' ?
+                    findtype(types + 2, me->mnt_type) != NULL :
+                    findtype(types, me->mnt_type) == NULL))
                        continue;
 
                r = mounthelper(me->mnt_fsname, me->mnt_dir, me->mnt_type);
-- 
2.52.0


Reply via email to