Bug#949066: lintian: testsuite failures on arm*

2020-01-24 Thread Gianfranco Costamagna
Hello,

> I have been working with the dpkg maintainers on a solution, which
> will not happen anytime soon. The underlying problem is that
> dpkg-buildpackage issues a build error when there is simply nothing to
> do.
> 
> Fixing that is complicated because source packages nowadays permit the
> building of undeclared artifacts. There is no longer a way to tell
> from a source package if a build should be attempted. Either way, I
> will triage your bug later today.
> 
> Thank you for your patience and for your help in trying to make
> Lintian better for everyone.
> 

wow I though the solution was "easy" thanks for keeping it running and rocking!

feel free to take your time, I was making sure this bug was not "disappeared" :)

I'm not sure if src:file migration to testing is stalled by this bug...

grep-excuses file
file (1:5.38-3 to 1:5.38-4)
Maintainer: Christoph Biedl
7 days old (needed 5 days)
Migration status for file (1:5.38-3 to 1:5.38-4): Waiting for test results, 
another package or too young (no action required now - check later)
Issues preventing migration:
autopkgtest for binutils: amd64: Test in 
progress (will not be considered a regression), arm64: Test in progress
autopkgtest for lintian/2.46.0: amd64: Pass, arm64: Reference test in progress, but real test failed 
already
Additional info:
Piuparts tested OK - https://piuparts.debian.org/sid/source/f/file.html
7 days old (needed 5 days)


G.



Bug#949066: lintian: testsuite failures on arm*

2020-01-24 Thread Felix Lechner
Hi Gianfranco,

On Fri, Jan 24, 2020 at 8:39 AM Gianfranco Costamagna
 wrote:
>
> Felix gentle ping please?
> (in Ubuntu I disabled that test BTW and the result package autopkgtestsuite 
> was green on all archs)

The release took place solely to facilitate upgrading lintian.d.o,
which has been down for months. Details are in #949398; your bug is
mentioned there.

I have been working with the dpkg maintainers on a solution, which
will not happen anytime soon. The underlying problem is that
dpkg-buildpackage issues a build error when there is simply nothing to
do.

Fixing that is complicated because source packages nowadays permit the
building of undeclared artifacts. There is no longer a way to tell
from a source package if a build should be attempted. Either way, I
will triage your bug later today.

Thank you for your patience and for your help in trying to make
Lintian better for everyone.

Kind regards

Felix



Bug#949066: lintian: testsuite failures on arm*

2020-01-24 Thread Gianfranco Costamagna
hello,

On Fri, 17 Jan 2020 09:29:57 +0100 Gianfranco Costamagna 
 wrote:
> Hello,
> 
> > 
> > A list of the test packages that failed to build on arm* would be helpful.
> 
> sure, you can grab all the test results from here
> http://autopkgtest.ubuntu.com/packages/lintian
> 
> (you can use 2.45.0ubuntu1 as reference to 2.45.0, the only difference is 
> this patch
> http://launchpadlibrarian.net/460879291/lintian_2.45.0_2.45.0ubuntu1.diff.gz
> please apply it!)
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=949065

so 2.47.0 didn't hit the fix for this bug...

Felix gentle ping please?
(in Ubuntu I disabled that test BTW and the result package autopkgtestsuite was 
green on all archs)

G.



Bug#949066: lintian: testsuite failures on arm*

2020-01-17 Thread Gianfranco Costamagna
Hello,

> 
> A list of the test packages that failed to build on arm* would be helpful.

sure, you can grab all the test results from here
http://autopkgtest.ubuntu.com/packages/lintian

(you can use 2.45.0ubuntu1 as reference to 2.45.0, the only difference is this 
patch
http://launchpadlibrarian.net/460879291/lintian_2.45.0_2.45.0ubuntu1.diff.gz
please apply it!)
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=949065


HTH

Gianfranco



Bug#949066: lintian: testsuite failures on arm*

2020-01-16 Thread Felix Lechner
Hi Gianfranco,

On Thu, Jan 16, 2020 at 8:26 AM Gianfranco Costamagna
 wrote:
>
> Hello, looks like tests regressed on arm64 and armhf

Thanks for pointing this out. It is a result of splitting the build
and test specifications. The best thing is probably to restrict the
building of affected tests with the proper wildcard in
'Package-Architecture:' in build-spec/fill-values. I will add them in
the near future.

A list of the test packages that failed to build on arm* would be helpful.

Kind regards
Felix



Bug#949066: lintian: testsuite failures on arm*

2020-01-16 Thread Gianfranco Costamagna
Source: lintian
Version: 2.44.0
Severity: serious

Hello, looks like tests regressed on arm64 and armhf

https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-focal/focal/arm64/l/lintian/20200109_233436_8f7fd@/log.gz

https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-focal/focal/armhf/l/lintian/20200109_231746_20903@/log.gz

https://ci.debian.net/data/autopkgtest/unstable/arm64/l/lintian/3930008/log.gz

looks like something is not linking correctly with PIC

make[2]: Entering directory 
'/tmp/autopkgtest.FowcgT/autopkgtest_tmp/testsuite/packages/tags/checks/shared-libs/shared-libs-non-pic-i386/shared-libs-non-pic-i386-1.0'
Makefile:16: update target 'non-pic.o' due to: non-pic.c
gcc -g -O2 
-fdebug-prefix-map=/tmp/autopkgtest.FowcgT/autopkgtest_tmp/testsuite/packages/tags/checks/shared-libs/shared-libs-non-pic-i386/shared-libs-non-pic-i386-1.0=.
 -fstack-protector-strong -Wformat -Werror=format-security -fno-PIE -Wdate-time 
-D_FORTIFY_SOURCE=2 -o non-pic.o -c non-pic.c
Makefile:13: update target 'libbaz3.so.1.0.3b' due to: non-pic.o
gcc -Wl,-Bsymbolic-functions -Wl,-z,relro -Wl,-z,now -fno-PIE -fno-pie -o 
libbaz3.so.1.0.3b -shared -Wl,-soname,libbaz3.so.1 non-pic.o -lc
/usr/bin/ld: non-pic.o: relocation R_AARCH64_ADR_PREL_PG_HI21 against symbol 
`__stack_chk_guard@@GLIBC_2.17' which may bind externally can not be used when 
making a shared object; recompile with -fPIC
/usr/bin/ld: non-pic.o(.text+0x1c): unresolvable R_AARCH64_ADR_PREL_PG_HI21 
relocation against symbol `__stack_chk_guard@@GLIBC_2.17'
/usr/bin/ld: final link failed: bad value
collect2: error: ld returned 1 exit status


G.