Problems with the giac package : different failures on buildd and porterbox !?

[Julien Puydt]

Hi,

I uploaded a new giac to unstable ; this package was building ok
everywhere when it was last uploaded (a long time ago, september 2018),
but now fails on s390x.

It's a scientific computation package, and the build process consists
in essentially three steps :
- build the source ;
- build the documentation, running the executables from the previous
step to show off their use ;
- run the checks, by running the executables from the first step on
stranger inputs, stressing the corner cases.

The buildd managed went through the first two steps, but only a few
tests ran, and already with a few failures (already a problem), until
one of them had to be killed after a while, presumably because it was
stuck (complete story here:
https://buildd.debian.org/status/fetch.php?pkg=giac&arch=s390x&ver=1.5.0.85%2Bdfsg1-1&stamp=1579706992&raw=0
).

I wanted to work on it and get meaningful data for upstream, so turned
to zelenka, the porterbox for this architecture. I followed the
tutorial here https://dsa.debian.org/doc/schroot/ - as I'm already used
to do when I get problems with other packages. But this time, the
documentation step already fails :
*** bug in PARI/GP (Segmentation Fault), please report.
(repeated quite a few times before the compilation dies)

I don't understand why I get so wildly different outcomes with the same
source package : did something get into unstable and break things
between the time the buildd ran and the time I tried to reproduce the
issue ? Do you have any advice ?

Thanks,

JP

[Julien Puydt]

Post by Julien Puydt
I wanted to work on it and get meaningful data for upstream, so turned
to zelenka, the porterbox for this architecture. I followed the
tutorial here https://dsa.debian.org/doc/schroot/ - as I'm already used
to do when I get problems with other packages. But this time, the
*** bug in PARI/GP (Segmentation Fault), please report.
(repeated quite a few times before the compilation dies)

The compiled calculator gives strange results:
0>> 1/2.0-1/3
5.46834151467e-304
// Time 0
1>> 1/2.0
3.64556100978e-304
// Time 0
2>> 1/3
1/3
// Time 0
3>> 1/1.
3.64556100978e-304
// Time 0
4>> 1.
3.64556100978e-304
// Time 0

The last example shows that it's not just a computation problem.

There's something extremely fishy on that box!

JP

[Aurelien Jarno]

Post by Julien Puydt

Post by Julien Puydt
I wanted to work on it and get meaningful data for upstream, so turned
to zelenka, the porterbox for this architecture. I followed the
tutorial here https://dsa.debian.org/doc/schroot/ - as I'm already used
to do when I get problems with other packages. But this time, the
*** bug in PARI/GP (Segmentation Fault), please report.
(repeated quite a few times before the compilation dies)

s390x is a big-endian architecture. The build failure on ppc64 and
sparc64 are really similar, so I believe the issue is related to
endianness.

Post by Julien Puydt
0>> 1/2.0-1/3
5.46834151467e-304
// Time 0
1>> 1/2.0
3.64556100978e-304
// Time 0
2>> 1/3
1/3
// Time 0
3>> 1/1.
3.64556100978e-304
// Time 0
4>> 1.
3.64556100978e-304
// Time 0
The last example shows that it's not just a computation problem.

That behaviour clearly shows that what is displayed is not the result of
the computation.