Your message dated Thu, 15 Aug 2024 18:27:44 +0200
with message-id <zr4sggx8h0cx3...@emlwks999.eu>
and subject line Re: Bug#1078738: csh: type mismatch between seekf declaration
and definition
has caused the Debian Bug report #1078738,
regarding csh: type mismatch between seekf declaration and definition
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
1078738: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1078738
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Source: csh
Version: 20240808-1
Severity: serious
Tags: ftbfs patch
Justification: ftbfs
Hi Alastair,
The recently introduced patch fixing gcc 14 build failures also
introduced a type mismatch between seekf declaration and its
definition: the first with off64_t and the second with off_t.
On most architectures this is not a problem, as they share the
same size in the end, including on armel and armhf. However,
the mismatch results in the following build failure on i386,
where off_t and off64_t have distinct sizes, as you can see in
the buildd log[1]:
csh.c:1265:1: error: conflicting types for ‘seekf’; have ‘int(void *,
off_t *, int)’ {aka ‘int(void *, long int *, int)’}
1265 | seekf(void *oreo, off_t *off, int whence)
| ^~~~~
csh.c:158:17: note: previous declaration of ‘seekf’ with type ‘int(void
*, off64_t *, int)’ {aka ‘int(void *, long long int *, int)’}
158 | static int seekf(void *, off64_t *, int);
| ^~~~~
[1]:
https://buildd.debian.org/status/fetch.php?pkg=csh&arch=i386&ver=20240808-1&stamp=1723122202&raw=0
I attached a patch that modifies the original g++-14.patch in
order to resolve the issue, for your convenience. I did a few
manual tests to ensure the type change hasn't introduced any
obvious issues in the shell, not that there is a lot of room
left for error.
The build failure on i386 causes a sprawl of autoremoval
notifications in reverse dependencies, so if you're caught by
time, I'm considering the option of proceeding to an NMU.
Have a nice day, :)
--
.''`. Étienne Mollier <emoll...@debian.org>
: :' : pgp: 8f91 b227 c7d6 f2b1 948c 8236 793c f67e 8f0d 11da
`. `' sent from /dev/pts/2, please excuse my verbosity
`-
diff -Nru csh-20240808/debian/patches/g++-14.patch csh-20240808/debian/patches/g++-14.patch
--- csh-20240808/debian/patches/g++-14.patch 2024-08-08 13:02:13.000000000 +0200
+++ csh-20240808/debian/patches/g++-14.patch 2024-08-15 08:20:53.000000000 +0200
@@ -3,6 +3,7 @@
Bug-Origin: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1074892
Last-Updated: 2024-08-08
Forwarded: not-needed
+Reviewed-By: Étienne Mollier <emoll...@debian.org>
--- a/csh.c
+++ b/csh.c
@@ -22,7 +23,7 @@
-static off_t
-seekf(void *oreo, off_t off, int whence)
+static int
-+seekf(void *oreo, off_t *off, int whence)
++seekf(void *oreo, off64_t *off, int whence)
{
- return lseek(DESC(oreo), off, whence);
+ return lseek(DESC(oreo), *off, whence);
signature.asc
Description: PGP signature
--- End Message ---
--- Begin Message ---
Version: 20240808-2
Hi Alastair,
Alastair McKinstry, on 2024-08-15:
> I think our patches passed in the ether.
Okay, it took me some time to understand that, but I see today's
upload closed #1078737 instead of #1078738, so I'm updating
#1078738's metadata to reflect your fix accordingly.
Have a nice day, :)
--
.''`. Étienne Mollier <emoll...@debian.org>
: :' : pgp: 8f91 b227 c7d6 f2b1 948c 8236 793c f67e 8f0d 11da
`. `' sent from /dev/pts/1, please excuse my verbosity
`-
signature.asc
Description: PGP signature
--- End Message ---
