GNU bug report logs - #32360
gst-plugins-base has test failures (when built as a dependency)

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 32360 in the body.
You can then email your comments to 32360 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>
To: <bug-guix <at> gnu.org>
Subject: gst-plugins-base has test failures (when built as a dependency)
Date: Fri, 3 Aug 2018 13:38:58 +0200

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

Hi,

gst-plugins-base has somehow a problem, which I don't understand fully.
Can someone have a look into it? It was already reported by Piere
Neidhardt on the help mailing list:

https://lists.gnu.org/archive/html/help-guix/2018-06/msg00075.html

OK, so I am on 5803f87284c228ca131d74a4259238e7685526ac and do:

./pre-inst-env  guix build gst-plugins-base
/gnu/store/58i3gmidcbksf7cxwhfjrp2w22k6glqa-gst-plugins-base-1.14.1-doc
/gnu/store/g3iwkhs7sjrhsaqq3bwmj81mxsa5c3d1-gst-plugins-base-1.14.1


Looks good.

I can also build the derivation:

./pre-inst-env  guix build gst-plugins-base -d
/gnu/store/zg9bmdlrdwd4i4c1prz4hzzjm3877nkl-gst-plugins-base-1.14.1.drv

I can also add options "--no-grafts --no-substitutes", same result.

Now, I want to build wine:

Building package wine
Full build command: ./pre-inst-env guix build wine --no-grafts --no-substitutes
Guix commit:
5803f87284c228ca131d74a4259238e7685526ac gnu: Update ruby-net-ssh to 4.2.0.
Start building at:    2018-08-03T11:09:39,183594824+00:00
The following derivations will be built:
   /gnu/store/shmcbb9yb99y0kfhn42dz2gpmqnifsmn-wine-3.0.2.drv
   /gnu/store/yrk539har2g5pw1rw5b9k8nq9p0w3c1h-gst-plugins-base-1.14.1.drv
@ build-started /gnu/store/yrk539har2g5pw1rw5b9k8nq9p0w3c1h-gst-plugins-base-1.14.1.drv - i686-linux /var/log/guix/drvs/yr//k539har2g5pw1rw5b9k8nq9p0w3c1h-gst-plugins-base-1.14.1.drv.bz2
starting phase `set-SOURCE-DATE-EPOCH'
[...]

What I find strange is that the derivation of gst-plugins-base differs
from the one above when directly building gst-plugins-base. That would
make sense to me if there would be two packages of gst-plugins-base,
like an inherited one or just any kind of copy. But here, the package
definition of wine ist just using the "normal" package. Or have I
overlooked something? Have I misunderstood something about derivations
here?!

This derivation will fail to build. Here is the end of the log:

[..]

================================================================
   GStreamer Base Plug-ins 1.14.1: tests/check/test-suite.log
================================================================

# TOTAL: 80
# PASS:  79
# SKIP:  0
# XFAIL: 0
# FAIL:  1
# XPASS: 0
# ERROR: 0

.. contents:: :depth: 2

FAIL: elements/opus
===================

Running suite(s): opus
33%: Checks: 6, Failures: 0, Errors: 4
gstcheck.c:596:E:general:test_opus_encode_nothing:0: (after this point) Received
 signal 11 (Segmentation fault)
gstcheck.c:596:E:general:test_opus_encode_samples:0: (after this point) Received signal 11 (Segmentation fault)
gstcheck.c:596:E:general:test_opus_encode_properties:0: (after this point) Received signal 11 (Segmentation fault)
elements/opus.c:429:E:general:test_opus_decode_plc_timestamps_with_fec:0: (after this point) Received signal 11 (Segmentation fault)
Check suite opus ran in 1.049s (tests failed: 4)
FAIL elements/opus (exit status: 4)


Backtrace:

           4 (primitive-load "/gnu/store/ki3kvmzipvf8baabfyaw1ncf4ax?")
In ice-9/eval.scm:
   191:35  3 (_ _)
In srfi/srfi-1.scm:
    640:9  2 (for-each #<procedure 81b5700 at /gnu/store/cgnrsi20mw?> ?)
In /gnu/store/cgnrsi20mwcm1s3z9ab8qw125m73l4qd-module-import/guix/build/gnu-build-system.scm:
   799:31  1 (_ _)
    369:6  0 (check #:target _ #:make-flags _ #:tests? _ # _ # _ # _)

/gnu/store/cgnrsi20mwcm1s3z9ab8qw125m73l4qd-module-import/guix/build/gnu-build-system.scm:369:6: In procedure check:
Throw to key `srfi-34' with args `(#<condition &invoke-error [program: "make" arguments: ("check") exit-status: 2 term-signal: #f stop-signal: #f] 80a8ee0>)'.
builder for `/gnu/store/yrk539har2g5pw1rw5b9k8nq9p0w3c1h-gst-plugins-base-1.14.1.drv' failed with exit code 1
@ build-failed /gnu/store/yrk539har2g5pw1rw5b9k8nq9p0w3c1h-gst-plugins-base-1.14.1.drv - 1 builder for `/gnu/store/yrk539har2g5pw1rw5b9k8nq9p0w3c1h-gst-plugins-base-1.14.1.drv' failed with exit code 1
cannot build derivation `/gnu/store/shmcbb9yb99y0kfhn42dz2gpmqnifsmn-wine-3.0.2.drv': 1 dependencies couldn't be built
guix build: error: build failed: build of `/gnu/store/shmcbb9yb99y0kfhn42dz2gpmqnifsmn-wine-3.0.2.drv' failed


Command exited with non-zero status 1
3.30user 0.17system 6:49.03elapsed 0%CPU (0avgtext+0avgdata 145936maxresident)k
32inputs+656outputs (0major+20586minor)pagefaults 0swaps
Returncode: 1
Finished building at: 2018-08-03T11:16:28,219631472+00:00
--------------------------------------------------------------------------------

When looking at Hydra, it succeeds in direct build (for x86_64):

https://hydra.gnu.org/job/gnu/master/gst-plugins-base-1.14.1.x86_64-linux

On i686 it fails with the same opus test.
On ARM it fails with different test failures.


When you look at the build of wine-x86-64:

https://hydra.gnu.org/build/2937119

it fails building gst-plugins-base.


I tried upgrading to the 1.14.2 bugfix release: Same behaviour.

Any ideas someone?

Thanks,

Björn

[Message part 2 (application/pgp-signature, inline)]

Message #8 received at 32360 <at> debbugs.gnu.org (full text, mbox):

From: Leo Famulari <leo <at> famulari.name>
To: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>
Cc: 32360 <at> debbugs.gnu.org
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Fri, 17 Aug 2018 13:16:42 -0400

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

On Fri, Aug 03, 2018 at 01:38:58PM +0200, Björn Höfling wrote:
> What I find strange is that the derivation of gst-plugins-base differs
> from the one above when directly building gst-plugins-base. That would
> make sense to me if there would be two packages of gst-plugins-base,
> like an inherited one or just any kind of copy. But here, the package
> definition of wine ist just using the "normal" package. Or have I
> overlooked something? Have I misunderstood something about derivations
> here?!

Presumably, you are using an x86_64 computer, and the successful build
of gst-plugins-base is for an x86_64-linux system. However, the wine
package and its dependency graph is always built as 32-bit software,
regardless of the underlying architecture:

From gnu/packages/wine.scm:
------
    (arguments
     `(;; Force a 32-bit build targeting a similar architecture, i.e.:
       ;; armhf for armhf/aarch64, i686 for i686/x86_64.
       #:system ,@(match (%current-system)
                    ((or "armhf-linux" "aarch64-linux")
                     `("armhf-linux"))
                    (_  
                     `("i686-linux")))
------

And the test failure in the 32-bit build of gst-plugins base was
previously discussed on help-guix (no bug report):

https://lists.gnu.org/archive/html/help-guix/2018-06/msg00075.html

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

Message #13 received at 32360 <at> debbugs.gnu.org (full text, mbox):

From: Leo Famulari <leo <at> famulari.name>
To: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>
Cc: 32360 <at> debbugs.gnu.org
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Fri, 17 Aug 2018 14:05:54 -0400

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

Here is a patch that excludes Opus when building gst-plugins-base on and
for 32-bit systems. I think it's not so bad to be missing support for
Opus in this context until the underlying bug is fixed.

With this patch, gst-plugins-base builds successfully with
'--system=i686-linux'. I didn't test the full Wine build.

[0001-gnu-gst-plugins-base-Don-t-build-with-Opus-for-32-bi.patch (text/plain, attachment)]

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

Message #16 received at 32360 <at> debbugs.gnu.org (full text, mbox):

From: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>
To: Leo Famulari <leo <at> famulari.name>
Cc: 32360 <at> debbugs.gnu.org
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Fri, 17 Aug 2018 23:50:41 +0200

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

Hi Leo,

On Fri, 17 Aug 2018 14:05:54 -0400
Leo Famulari <leo <at> famulari.name> wrote:

> Here is a patch that excludes Opus when building gst-plugins-base on
> and for 32-bit systems. I think it's not so bad to be missing support
> for Opus in this context until the underlying bug is fixed.
> 
> With this patch, gst-plugins-base builds successfully with
> '--system=i686-linux'. I didn't test the full Wine build.

I don't who/which program needs opus. But it's not working anyway, so I
would be fine with the patch.

I could not apply the patch. Maybe I did something wrong on my side?
Anyway, it's so short, I "applied" it manually.

Then it looks good:

$ ./pre-inst-env guix environment -C --expose=/gnu/store/ --ad-hoc wine
  -- wine $(./pre-inst-env  guix build hello --target=i686-w64-mingw32)/bin/hello.exe
[... some wine error message I suppose to be due to the -C
environment ...]
Hello, world!

Guix is so cool :-)

Thanks,

Björn

[Message part 2 (application/pgp-signature, inline)]

Message #19 received at 32360 <at> debbugs.gnu.org (full text, mbox):

From: Pjotr Prins <pjotr.public12 <at> thebird.nl>
To: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>
Cc: 32360 <at> debbugs.gnu.org, Leo Famulari <leo <at> famulari.name>
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Sat, 18 Aug 2018 12:46:10 +0200

Someone still needs to push the patch.

Message #22 received at 32360-done <at> debbugs.gnu.org (full text, mbox):

From: Tobias Geerinckx-Rice <me <at> tobias.gr>
To: Pjotr Prins <pjotr.public12 <at> thebird.nl>, Leo Famulari <leo <at> famulari.name>
Cc: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>,
 32360-done <at> debbugs.gnu.org
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Sat, 18 Aug 2018 13:34:09 +0200

Pjotr, Leo, Guix,

Pjotr Prins wrote:
> Someone still needs to push the patch.

I went ahead and did so in 
399c5fafcdb2d0c13ab51e4ab57d451d2c7cb1bd
since it's not really acceptable to have this broken in master.

Kind regards,

T G-R

Message #25 received at 32360-done <at> debbugs.gnu.org (full text, mbox):

From: Pjotr Prins <pjotr.public12 <at> thebird.nl>
To: Tobias Geerinckx-Rice <me <at> tobias.gr>
Cc: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>,
 32360-done <at> debbugs.gnu.org, Pjotr Prins <pjotr.public12 <at> thebird.nl>,
 Leo Famulari <leo <at> famulari.name>
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Sun, 19 Aug 2018 09:35:43 +0200

On Sat, Aug 18, 2018 at 01:34:09PM +0200, Tobias Geerinckx-Rice wrote:
> Pjotr, Leo, Guix,
> 
> Pjotr Prins wrote:
> > Someone still needs to push the patch.
> 
> I went ahead and did so in 399c5fafcdb2d0c13ab51e4ab57d451d2c7cb1bd
> since it's not really acceptable to have this broken in master.

It has been broken for a while. And, miracle, guix pull just worked :)

Thanks Tobias, Björn and Leo!

Pj.

Message #28 received at 32360 <at> debbugs.gnu.org (full text, mbox):

From: Ricardo Wurmus <rekado <at> elephly.net>
To: Leo Famulari <leo <at> famulari.name>
Cc: Björn Höfling <bjoern.hoefling <at> bjoernhoefling.de>,
 32360 <at> debbugs.gnu.org
Subject: Re: bug#32360: gst-plugins-base has test failures (when built as a
 dependency)
Date: Sat, 18 Aug 2018 09:44:09 +0200

Leo Famulari <leo <at> famulari.name> writes:

> Here is a patch that excludes Opus when building gst-plugins-base on and
> for 32-bit systems. I think it's not so bad to be missing support for
> Opus in this context until the underlying bug is fixed.

I agree.  The patch looks good to me.
Shall we rename this bug and keep track of this problem here?

--
Ricardo

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.