[Pkg-octave-devel] Bug#871214: Bug#871214: Bug#871214: octave-image FTBFS on i386: test hangs
Sébastien Villemot
sebastien at debian.org
Thu Aug 10 09:06:58 UTC 2017
Le jeudi 10 août 2017 à 09:31 +0200, Rafael Laboissière a écrit :
> > * Adrian Bunk <bunk at debian.org> [2017-08-07 01:57]:
>
> > Source: octave-image
> > Version: 2.6.1-1
> > Severity: serious
> > Tags: buster sid
> >
> > Some recent change in unstable and buster makes octave-image FTBFS on i386:
> >
> > https://tests.reproducible-builds.org/debian/history/octave-image.html
> > https://tests.reproducible-builds.org/debian/rb-pkg/unstable/i386/octave-image.html
> >
> > ...
> > Checking CC files ...
> > warning: function /build/1st/octave-image-2.6.1/inst/private/iscolormap.m shadows a core library function
> > warning: called from
> > /tmp/filePMKFm8 at line 1 column 1
> > [__spatial_filtering__]
> > PASSES 21 out of 21 tests
> > [graycomatrix]
> > PASSES 2 out of 2 tests
> > [watershed]
> > ***** test
> > im = [
> > 2 3 30 2
> > 3 30 3 30
> > 255 31 30 4
> > 2 255 31 30
> > 1 2 255 5];
> >
> > labeled8 = [
> > 1 1 0 3
> > 1 1 0 3
> > 0 0 0 0
> > 2 2 0 4
> > 2 2 0 4];
> > assert (watershed (im), labeled8);
> > assert (watershed (im, 8), labeled8);
> > !!!!! test failed
> > ASSERT errors for: assert (watershed (im),labeled8)
> >
> > Location | Observed | Expected | Reason
> > (3,4) 3 0 Abs err 3 exceeds tol 0
> > (4,4) 0 4 Abs err 4 exceeds tol 0
> > [bwdist]
> > Sat Sep 8 05:36:25 UTC 2018 - pbuilder was killed by timeout after 18h.
>
> I would guess that the culprit is the following unit test in bwdist:
>
> ## The quasi-euclidean method is apparently sensitive to a machine precision
> ## error that happens in x86 systems only. This test will cause an endless
> ## loop in case of a regression.
> %!test
> %! bw = [ 0 1 1 0 0 0 1 0
> %! 0 0 0 0 0 0 0 0
> %! 1 1 0 0 0 0 0 0
> %! 0 0 0 0 0 0 1 0
> %! 0 0 0 0 1 0 0 1
> %! 0 0 0 0 0 0 0 0
> %! 1 0 0 0 0 0 0 0
> %! 0 0 1 0 0 1 1 0];
> %! out = single ([
> %! 1.00000 0.00000 0.00000 1.00000 2.00000 1.00000 0.00000 1.00000
> %! 1.00000 1.00000 1.00000 sqrt(2) sqrt(2)+1 sqrt(2) 1.00000 sqrt(2)
> %! 0.00000 0.00000 1.00000 2.00000 2.00000 sqrt(2) 1.00000 sqrt(2)
> %! 1.00000 1.00000 sqrt(2) sqrt(2) 1.00000 1.00000 0.00000 1.00000
> %! 2.00000 2.00000 2.00000 1.00000 0.00000 1.00000 1.00000 0.00000
> %! 1.00000 sqrt(2) 2.00000 sqrt(2) 1.00000 sqrt(2) sqrt(2) 1.00000
> %! 0.00000 1.00000 1.00000 sqrt(2) sqrt(2) 1.00000 1.00000 sqrt(2)
> %! 1.00000 1.00000 0.00000 1.00000 1.00000 0.00000 0.00000 1.00000
> %! ]);
> %! assert (bwdist (bw, "quasi-euclidean"), out);
>
> Should we run the test above only on non-x86 architectures?
I would rather patch the test to allow for some tolerance margin.
--
⢀⣴⠾⠻⢶⣦ Sébastien Villemot
⣾⠁⢠⠒⠀⣿⡁ Debian Developer
⢿⡄⠘⠷⠚⠋⠀ http://sebastien.villemot.name
⠈⠳⣄ http://www.debian.org
More information about the Pkg-octave-devel
mailing list