lmi
[Top][All Lists]
Advanced

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

[lmi] "buster" updates


From: Greg Chicares
Subject: [lmi] "buster" updates
Date: Fri, 6 Apr 2018 21:48:52 +0000
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.6.0

Facing a large backlog of work on a Friday afternoon, I thought I'd just
upgrade my "buster" chroot, because what could go wrong?

First of all, the cross compiler got upgraded, so I guess I'll have to test
that, and then see if Kim's willing to upgrade in parallel. Of course, it
didn't upgrade smoothly:

Setting up gcc-mingw-w64-i686 (7.3.0-12+20.2+b1) ...
update-alternatives: warning: forcing reinstallation of alternative 
/usr/bin/i686-w64-mingw32-gcc-win32 because link group i686-w64-mingw32-gcc is 
broken
update-alternatives: warning: skip creation of /usr/bin/i686-w64-mingw32-gcc-7 
because associated file /usr/bin/i686-w64-mingw32-gcc-7.2-win32 (of link group 
i686-w64-mingw32-gcc) doesn't exist

Reformatting:

alternative /usr/bin/i686-w64-mingw32-gcc-win32
link group           i686-w64-mingw32-gcc
                                         ^^^^^^
makes me wonder whether the problem is the '-win32' suffix that debian added
beginning with gcc-7 (I don't remember how they distinguished the win32 and
posix thread models for previous versions):

$git log --grep '-win32'
commit 0ced710acb5e559ccd00a93662c009b6e24ff534
Author: Gregory W. Chicares <address@hidden>
Date:   2018-03-06T16:59:08+00:00

    Work around a dubious debian cross-compiler change
    
    The MinGW-w64 project's gcc-7.2.0 identifies its version as 7.2.0,
    but debian's MinGW-w64 release uses 7.2-win32 instead. Fixed lmi
    makefile to translate the debian variant to the upstream identifier.

Anyway, let's try removing the package
  apt-get remove g++-mingw-w64
and its dependencies
  apt-get autoremove
and reinstalling
  apt-get install g++-mingw-w64
which gives:

update-alternatives: using /usr/bin/i686-w64-mingw32-gcc-posix to provide 
/usr/bin/i686-w64-mingw32-gcc (i686-w64-mingw32-gcc) in auto mode
update-alternatives: warning: skip creation of /usr/bin/i686-w64-mingw32-gcc-7 
because associated file /usr/bin/i686-w64-mingw32-gcc-7.2-posix (of link group 
i686-w64-mingw32-gcc) doesn't exist
update-alternatives: using /usr/bin/i686-w64-mingw32-gcc-win32 to provide 
/usr/bin/i686-w64-mingw32-gcc (i686-w64-mingw32-gcc) in auto mode
update-alternatives: warning: skip creation of /usr/bin/i686-w64-mingw32-gcc-7 
because associated file /usr/bin/i686-w64-mingw32-gcc-7.2-win32 (of link group 
i686-w64-mingw32-gcc) doesn't exist

...and similar messages for x86_64:

update-alternatives: using /usr/bin/x86_64-w64-mingw32-gcc-posix to provide 
/usr/bin/x86_64-w64-mingw32-gcc (x86_64-w64-mingw32-gcc) in auto mode
update-alternatives: warning: skip creation of 
/usr/bin/x86_64-w64-mingw32-gcc-7 because associated file 
/usr/bin/x86_64-w64-mingw32-gcc-7.2-posix (of link group 
x86_64-w64-mingw32-gcc) doesn't exist
update-alternatives: using /usr/bin/x86_64-w64-mingw32-gcc-win32 to provide 
/usr/bin/x86_64-w64-mingw32-gcc (x86_64-w64-mingw32-gcc) in auto mode
update-alternatives: warning: skip creation of 
/usr/bin/x86_64-w64-mingw32-gcc-7 because associated file 
/usr/bin/x86_64-w64-mingw32-gcc-7.2-win32 (of link group 
x86_64-w64-mingw32-gcc) doesn't exist

But the compiler does seem to have been installed:

$i686-w64-mingw32-gcc --dumpversion
i686-w64-mingw32-gcc: error: unrecognized command line option '--dumpversion'; 
did you mean '-dumpversion'?

[wow--classic error message]

$i686-w64-mingw32-gcc -dumpversion 
7.3-win32

...and apt-get thinks everything's okay:

apt-get check                
Reading package lists... Done
Building dependency tree       
Reading state information... Done

...so I'll just try rebuilding everything.

But first, since I'm logged in as root, I thought I'd install a couple
of shell-script checkers: 'shellcheck', which was easy to install and
is useful...and 'checkbashisms', which is part of 'devscripts', which
is large, but, again, what could possibly go wrong?

update-alternatives: using /usr/bin/frm.mailutils to provide /usr/bin/frm (frm) 
in auto moW: APT had planned for dpkg to do more than it reported back (934 vs 
1051).
   Affected packages: debhelper:amd64 devscripts:amd64 dh-autoreconf:amd64 
dh-python:amd64 dh-strip-nondeterminism:amd64 dput:amd64 equivs:amd64 
gnupg:amd64 gpg-agent:amd64 gpg-wks-client:amd64 gpg-wks-server:amd64 
libb-hooks-endofscope-perl:amd64 libemail-valid-perl:amd64 
libgetopt-long-descriptive-perl:amd64 libgpgme11:amd64 libhtml-form-perl:amd64 
libhtml-format-perl:amd64 libhttp-cookies-perl:amd64 libhttp-daemon-perl:amd64 
libimport-into-perl:amd64 liblwp-protocol-https-perl:amd64 
libmailtools-perl:amd64 libmime-tools-perl:amd64 
libmodule-implementation-perl:amd64 libmodule-runtime-perl:amd64 
libmoo-perl:amd64 libnamespace-clean-perl:amd64 libnet-smtp-ssl-perl:amd64 
libpackage-stash-perl:amd64 libparams-validate-perl:amd64 
libsoap-lite-perl:amd64 libwww-perl:amd64 libxml-parser-perl:amd64 
libxml-sax-expat-perl:amd64 libxml-simple-perl:amd64 libxmlrpc-lite-perl:amd64 
licensecheck:amd64 lintian:amd64 lsb-release:amd64 mailutils:amd64 
po-debconf:amd64 python3-apt:amd64 python3-certifi:amd64 python3-chardet:amd64 
python3-debian:amd64 python3-distutils:amd64 python3-gpg:amd64 
python3-idna:amd64 python3-lib2to3:amd64 python3-magic:amd64 
python3-pkg-resources:amd64 python3-requests:amd64 python3-six:amd64 
python3-unidiff:amd64 python3-urllib3:amd64 python3-xdg:amd64 python3.6:amd64 
python3:amd64

Searching the web, I'm not convinced I should trust the first
dismissive comment I come across in an ubuntu forum:

https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1647638/comments/6
| That's just a minor cosmetic issue IMO.

Quelqu'un d'autre a rec,u la me^me message pour 'python3.5-minimal:amd64'
seulement, et...

https://forum.ubuntu-fr.org/viewtopic.php?id=2008872&p=2#p21722905
| Il reste quand même un problème sur python3-minimal.
| À tout hasard, retente à nouveau ça

...mais <<retente à nouveau>> (1051-934) fois? Je pense que non.

The only thing 'checkbashisms' found in *.sh that 'shellcheck' didn't is bogus:

possible bashism in test_coding_rules_test.sh line 377 (should be '.', not 
'source'):
  | sed -e '/^[ 0-9]\{9\} \(source files\|source lines\|marked defects\)/d' \

so I know how to solve the "APT had planned for dpkg to do more" problem:

apt-get remove devscripts
apt-get autoremove

reply via email to

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