Your message dated Sun, 10 Feb 2008 08:49:57 -0600
with message-id <[EMAIL PROTECTED]>
has caused the Debian Bug report #465061,
regarding bsdcpio: option parsing is broken on several architectures
to be marked as having been forwarded to the upstream software
author(s) [EMAIL PROTECTED]
(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere. Please contact me immediately.)
Debian bug tracking system administrator
(administrator, Debian Bugs database)
--- Begin Message ---
Hi,
I got this bug from a Debian developer regarding an assumption about the
signedness of char:
--- Begin Message ---
Package: bsdcpio
Version: 2.4.11-1
Severity: grave
Justification: renders package unusable
The option parsing of bsdcpio is totally broken on arm, armel, s390 and
powerpc due to a wrong assumption on char signedness. From cpio/cpio.c:
char opt;
[...]
while ((opt = cpio_getopt(cpio)) != -1) {
[...]
This yields a compiler warning, as seen in eg. the arm build log:
cpio/cpio.c: In function 'main':
cpio/cpio.c:114: warning: comparison is always true due to limited range of
data type
The result is that any bsdcpio invocation fails with the usage
instructions. I have verified this on an arm Etch installation with a
backported package, but there's little doubt that it's the case also
for the current sid packages.
See eg. http://lists.debian.org/debian-devel/2001/12/msg02074.html
for a discussion on the char signedness issue.
Cheers,
--
Niko Tyni [EMAIL PROTECTED]
--- End Message ---
--- End Message ---
