Debian Bug report logs - #827724
findutils: find -type f matches symlinks in armel armhf and mipsel

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Maximiliano Curia <maxy@debian.org>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: findutils: find -type f matches symlinks in armel armhf and mipsel

Date: Mon, 20 Jun 2016 10:58:59 +0200

Package: findutils
Version: 4.6.0+git+20160126-3
Severity: critical
Justification: breaks unrelated software

Hi,

The new version of find has a weird behaviour, at least in the arches: armhf armel and mipsel.
As simple test:

dir=$(mktemp -d)
cd $dir
ln -s a b
find -type f

Shows ./b while it shouldn't.

Interestingly:
$ find -type l
find: Duplicate file type 'l' in the argument list to -type

Both examples work fine in amd64.

I haven't followed the code thoroughly, but it seems that the type is using a 
float to hold a bit mask, maybe this is not so well behaved in all arches.

This breaks dh_md5summs in the mentioned arches, producing a number of 
ftbfs's, and probably many other things, thus the severity.

Happy hacking,
-- System Information:
Debian Release: stretch/sid
  APT prefers testing
  APT policy: (500, 'testing'), (500, 'stable'), (50, 'unstable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386, armhf

Kernel: Linux 4.5.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Init: systemd (via /run/systemd/system)

Versions of packages findutils depends on:
ii  libc6        2.22-11
ii  libselinux1  2.5-3

findutils recommends no packages.

Versions of packages findutils suggests:
ii  mlocate  0.26-1

-- no debconf information

Message #10 received at 827724@bugs.debian.org (full text, mbox, reply):

From: Holger Levsen <holger@layer-acht.org>

To: 827724@bugs.debian.org, 827724-subscribe@bugs.debian.org, Vagrant Cascadian <vagrant@debian.org>, Niels Thykier <niels@thykier.net>

Subject: this bug breaks armhf testing for reproducible.debian.net too

Date: Mon, 20 Jun 2016 09:14:56 +0000

[Message part 1 (text/plain, inline)]

Hi,

just wanted to confirm that we've also been seeing this bug since last
night on unstable/armhf but indeed not on amd64+i386…

eg:

[01:12] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/kdesu : reproducible ➤ FTBFS
[01:12] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/wikipedia2text : reproducible ➤ FTBFS
[01:12] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/gnome-mines : reproducible ➤ FTBFS
[01:12] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/libtext-csv-xs-perl : reproducible ➤ FTBFS
[01:12] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/biomaj : reproducible ➤ FTBFS
[01:13] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/libsynthesis : reproducible ➤ FTBFS
[01:13] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/node-eventsource : reproducible ➤ FTBFS
[01:14] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/libvoikko : reproducible ➤ FTBFS
[01:15] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/live-images : reproducible ➤ FTBFS
[01:17] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/lmodern : reproducible ➤ FTBFS
[01:17] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/suckless-tools : reproducible ➤ FTBFS
[01:17] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/libunique : reproducible ➤ FTBFS
[01:18] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/literki : reproducible ➤ FTBFS
[01:18] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/speech-dispatcher : reproducible ➤ FTBFS
[01:20] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/audiocd-kio : reproducible ➤ FTBFS
[01:20] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/libxml-libxml-perl : reproducible ➤ FTBFS
[01:21] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/lua-event : reproducible ➤ FTBFS
[01:22] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/lua5.1 : reproducible ➤ FTBFS
[01:22] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/liece : reproducible ➤ FTBFS
[01:24] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/makefs : reproducible ➤ FTBFS
[01:25] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/lua-sec : reproducible ➤ FTBFS
[01:25] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/kpeople : reproducible ➤ FTBFS
[01:25] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/luajit : reproducible ➤ FTBFS
[01:26] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/libzen : reproducible ➤ FTBFS
[01:26] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/kopete : reproducible ➤ FTBFS
[01:26] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/kjsembed : reproducible ➤ FTBFS
[01:26] -       KGB-0- | (#debian-reproducible) https://tests.reproducible-builds.org/debian/unstable/armhf/lv2 : reproducible ➤ FTBFS

Please fix ASAP :) Thanks.


cheers,
	Holger

[signature.asc (application/pgp-signature, inline)]

Message #17 received at 827724@bugs.debian.org (full text, mbox, reply):

From: Maximiliano Curia <maxy@debian.org>

To: 827724@bugs.debian.org

Subject: Re: findutils: find -type f matches symlinks in armel armhf and mipsel

Date: Mon, 20 Jun 2016 13:10:46 +0200

[Message part 1 (text/plain, inline)]

Control: tag -1 + patch
¡Hola!

El 2016-06-20 a las 10:58 +0200, Maximiliano Curia escribió:
> Package: findutils 
> Version: 4.6.0+git+20160126-3 
> Severity: critical 
> Justification: breaks unrelated software

> The new version of find has a weird behaviour, at least in the arches: armhf armel and mipsel. 
> As simple test:

> dir=$(mktemp -d) 
> cd $dir 
> ln -s a b 
> find -type f

> Shows ./b while it shouldn't.

> Interestingly:
> $ find -type l 
> find: Duplicate file type 'l' in the argument list to -type

> Both examples work fine in amd64.

> I haven't followed the code thoroughly, but it seems that the type is using a 
> float to hold a bit mask, maybe this is not so well behaved in all arches.

> This breaks dh_md5summs in the mentioned arches, producing a number of 
> ftbfs's, and probably many other things, thus the severity.

With the addition of multiple types in one -type option, the args union of the 
predicate struct has grown and the new space is not being initialized.
(The initialization of args is done as a args.str = NULL, but args.types is a 
bool types[FTYPE_COUNT]; with count being 8, that's 32 bytes vs 4 that are 
currently initialized in a 32 bits arch.

Replacing the xmalloc invocation by a xzalloc one "fixes" the issue as it 
initializes all of the predicate components.

I'm attaching a patch that does this.

Sadly, findutils also fails to build from source due this findutils issue, 
so the build should be done either against an older version of findutils or 
tweak the PATH so that debhelper uses the recently built find.

P.S. After following the code I noticed that the rate variable wasn't used as a 
bitmask sorry for the confussion, a comment in the code lead me to think that 
it was.

-- 
"It is practically impossible to teach good programming to students that have
had a prior exposure to BASIC: as potential programmers they are mentally
mutilated beyond hope of regeneration."
-- Edsger W. Dijkstra
Saludos /\/\ /\ >< `/

[replace_xmalloc_with_xzalloc.diff (text/x-diff, attachment)]

[signature.asc (application/pgp-signature, inline)]

Message #28 received at 827724-close@bugs.debian.org (full text, mbox, reply):

From: Andreas Metzler <ametzler@debian.org>

To: 827724-close@bugs.debian.org

Subject: Bug#827724: fixed in findutils 4.6.0+git+20160517-4

Date: Mon, 20 Jun 2016 18:33:28 +0000

Source: findutils
Source-Version: 4.6.0+git+20160517-4

We believe that the bug you reported is fixed in the latest version of
findutils, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 827724@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Andreas Metzler <ametzler@debian.org> (supplier of updated findutils package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@ftp-master.debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Format: 1.8
Date: Mon, 20 Jun 2016 19:55:29 +0200
Source: findutils
Binary: findutils locate
Architecture: source
Version: 4.6.0+git+20160517-4
Distribution: unstable
Urgency: low
Maintainer: Andreas Metzler <ametzler@debian.org>
Changed-By: Andreas Metzler <ametzler@debian.org>
Closes: 827724
Description: 
 findutils  - utilities for finding files--find, xargs
 locate     - maintain and query an index of a directory tree
Changes:
 findutils (4.6.0+git+20160517-4) unstable; urgency=low
 .
   * 13_replace_xmalloc_with_xzalloc.diff by Maximiliano Curia: Hotfix wrong
     result on armhf armel and mipsel. Closes: #827724
Checksums-Sha1: 
 c34ea1f53a75b9d4a6b15189d81aafd06a7d6777 2206 findutils_4.6.0+git+20160517-4.dsc
 fe3b3f8745743a4c9aabd903fecebe5143b8df63 28908 findutils_4.6.0+git+20160517-4.debian.tar.xz
Checksums-Sha256: 
 5841e18b2455dcabbc2e138d986a95be4e59337bef78fc0d66406a1ceb858eea 2206 findutils_4.6.0+git+20160517-4.dsc
 2cad59d59abf7f2ab22d29ba37ec1159a2c115c1c1933da204823db6b5de3c4a 28908 findutils_4.6.0+git+20160517-4.debian.tar.xz
Files: 
 8b6837743d89e138e4252a1359d7e0a4 2206 utils required findutils_4.6.0+git+20160517-4.dsc
 3e2033c6571d4e4fcbbbc912c53173ee 28908 utils required findutils_4.6.0+git+20160517-4.debian.tar.xz

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
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=0AhU
-----END PGP SIGNATURE-----

Message #33 received at 827724@bugs.debian.org (full text, mbox, reply):

From: Maximiliano Curia <maxy@debian.org>

To: Andreas Metzler <ametzler@debian.org>, 827724@bugs.debian.org

Subject: Re: Bug#827724 closed by Andreas Metzler <ametzler@debian.org> (Bug#827724: fixed in findutils 4.6.0+git+20160517-4)

Date: Mon, 20 Jun 2016 21:25:14 +0200

[Message part 1 (text/plain, inline)]

¡Hola!

El 2016-06-20 a las 18:36 +0000, Debian Bug Tracking System escribió:
> This is an automatic notification regarding your Bug report 
> which was filed against the findutils package:
>
> #827724: findutils: find -type f matches symlinks in armel armhf and mipsel
>
> It has been closed by Andreas Metzler <ametzler@debian.org>.

As I mentioned in my previous mail, findutils also fails to build from source 
due to this issue, so it's now failing in armel, armhf, mipsel and hurd-i386. 
:(

Are you planning to do a binary only upload for these arches?

Alternatively, adding:
override_dh_md5sums:
       PATH="$$(pwd)/find:$$PATH" dh_md5sums

To the debian/rules file would allow the build to finish successfully.

Happy hacking,
-- 
"There are two major products that come out of Berkeley: LSD and BSD.
We don't believe this to be a coincidence."
-- Jeremy S. Anderson
Saludos /\/\ /\ >< `/

[signature.asc (application/pgp-signature, inline)]

Message #38 received at 827724@bugs.debian.org (full text, mbox, reply):

From: Andreas Metzler <ametzler@bebt.de>

To: Maximiliano Curia <maxy@debian.org>, 827724@bugs.debian.org

Subject: Re: Bug#827724: closed by Andreas Metzler <ametzler@debian.org> (Bug#827724: fixed in findutils 4.6.0+git+20160517-4)

Date: Tue, 21 Jun 2016 20:56:44 +0200

On 2016-06-20 Maximiliano Curia <maxy@debian.org> wrote:
> ¡Hola!

> El 2016-06-20 a las 18:36 +0000, Debian Bug Tracking System escribió:
> > This is an automatic notification regarding your Bug report which was
> > filed against the findutils package:
> > 
> > #827724: findutils: find -type f matches symlinks in armel armhf and mipsel
> > 
> > It has been closed by Andreas Metzler <ametzler@debian.org>.

> As I mentioned in my previous mail, findutils also fails to build from
> source due to this issue, so it's now failing in armel, armhf, mipsel and
> hurd-i386. :(

> Are you planning to do a binary only upload for these arches?
[...]

This have been fixed with binNMUs. Thanks you, whoever did it.

cu Andreas
-- 
`What a good friend you are to him, Dr. Maturin. His other friends are
so grateful to you.'
`I sew his ears on from time to time, sure'

Send a report that this bug log contains spam.