Package: inotify-tools / 3.14-8.1

Metadata

Package Version Patches format
inotify-tools 3.14-8.1 3.0 (quilt)

Patch series

view the series file
Patch File delta Description
0001 fix ftbfs.patch | (download)

libinotifytools/src/test.c | 8 4 4 - 0 !
1 file changed, 4 insertions( ), 4 deletions(-)

 fix-ftbfs

On HP/PA (aka PA-RISC), the test suite failed because it was making a
bad assumption about the behavior of inotify assigning ids. It still
makes bad assumptions, but now with this change all of them are true
(for now).

The original Debian bug report:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=558981

The discussion with the kernel maintainers about this behavior:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=561880

0002 fix hyphens in manpages | (download)

man/inotifywait.1 | 10 5 5 - 0 !
man/inotifywatch.1 | 2 1 1 - 0 !
2 files changed, 6 insertions( ), 6 deletions(-)

 fix-hyphens-in-manpages

they need to be \-

0003 fix spelling of writable.patch | (download)

src/common.c | 2 1 1 - 0 !
1 file changed, 1 insertion( ), 1 deletion(-)

 fix spelling of writable from "writeable"


0004 document files handled by their inodes | (download)

man/inotifywait.1 | 8 8 0 - 0 !
man/inotifywatch.1 | 9 9 0 - 0 !
2 files changed, 17 insertions( )

 should document that files are handled by their inodes

Debian-Bug: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=863163

0005 Fix segfault with csv output when filename contains .patch | (download)

src/inotifywait.c | 2 1 1 - 0 !
1 file changed, 1 insertion( ), 1 deletion(-)

 [patch] fix segfault with csv output when filename contains comma

Double `csv_escape()'ing filename is logic error, but root of the
problem was that passing `csv' buffer back into `csv_escape()' caused
endless loop over `static char csv[MAXLEN]', and buffer overflow.

0006 Fix buffer overrun in inotifytools.c.patch | (download)

libinotifytools/src/inotifytools.c | 2 1 1 - 0 !
1 file changed, 1 insertion( ), 1 deletion(-)

 fix buffer overrun in inotifytools.c

The following code

    char *names[2 sizeof(int)/sizeof(char*)];

was supposed to allocate enough space on stack to fit two `char *' and one
`int'. Problem is that when sizeof(int) < sizeof(char *), which is likely on
64-bit systems, it caused expression `sizeof(int)/sizeof(char*)' evaluate to 0,
resulting in buffer overrun.

Detected by GCC-9 new diagnostics.

Closes: #925717