[Top][All Lists]

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

[PATCH] maint: fix a memory leak in df

From: Anton Ovchinnikov
Subject: [PATCH] maint: fix a memory leak in df
Date: Mon, 8 Jul 2013 18:53:54 +0400

Valgrind shows some memory leaks while launching 'df' without arguments:

==8809== LEAK SUMMARY:
==8809==    definitely lost: 48 bytes in 1 blocks
==8809==    indirectly lost: 16 bytes in 3 blocks

If I launch 'df -a' valgrind doesn't detect any 'lost' memory.
As far as I understand, this is not the case for using IF_LINT macro,
as it's impossible to predict the size of lost memory.
I believe the following patch eliminates the leak.

One more thing: the inserted code looks similar to the fragment at
gnulib/lib/mountlist.c:964. What about moving this fragment to
stand-alone function (free_mount_entry(struct mount_entry *me))? It
implies exposing this function in mountlist.h.


>From 0ac25b9c2b1f72601b536cc92ae28aec29f6942a Mon Sep 17 00:00:00 2001
From: Anton Ovchinnikov <address@hidden>
Date: Mon, 8 Jul 2013 00:03:37 +0400
Subject: [PATCH] maint: fix a memory leak in df

* src/df.c (filter_mount_list): Free mount_entry with longer
me_mountdir while building devlist.
 src/df.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/df.c b/src/df.c
index 0515131..e2b0257 100644
--- a/src/df.c
+++ b/src/df.c
@@ -639,7 +639,12 @@ filter_mount_list (void)
                       || ( strlen (devlist->me->me_mountdir)
                          > strlen (me->me_mountdir)))
-                       /* FIXME: free ME - the others are also not free()d.  */
+                      struct mount_entry *old_me = devlist->me;
+                      free (old_me->me_devname);
+                      free (old_me->me_mountdir);
+                      if (old_me->me_type_malloced)
+                        free (old_me->me_type);
+                      free (old_me);
                       devlist->me = me;
                   continue; /* ... with the loop over the mount_list.  */

reply via email to

[Prev in Thread] Current Thread [Next in Thread]