sensational bug!

[Martin R]

On https://github.com/sagemath/sage/actions/runs/8168335359/job/22330264302?pr=37545

I see

sage -t --long --random-seed=286735480429121101562228604801325644303 src/sage/rings/tests.py

**********************************************************************

Error: Failed example:: Got: Random testing has revealed a problem in test_karatsuba_multiplication

Please report this bug! You may be the first

person in the world to have seen this problem.

Please include this random seed in your bug report:

Random seed: 305806809989734915896642073966608392384

ValueError('Multiplication failed')

test_karatsuba_multiplication(QQbar, 3, 3, numtests=2) # long time, needs sage.rings.number_field

Expected nothing

Got:

Random testing has revealed a problem in test_karatsuba_multiplication

Please report this bug! You may be the first

person in the world to have seen this problem.

Please include this random seed in your bug report:

Random seed: 305806809989734915896642073966608392384

ValueError('Multiplication failed')

[Frédéric Chapoton]

Indeed, interesting.

but not confirmed on my machine (Intel(R) Core(TM) i5-4690 on Linux ubuntu 22.04.4 LTS)

Frédéric

[Salvatore Stella]

I can't reproduce it either:
on a Intel(R) Core(TM) i7-6900K CPU machine with 64Gb of RAM running gentoo
and sage 10.3.beta0
S.

[Dima Pasechnik]

Not confirmed here, either. We need more details about the machine, the OS, the setup, etc.

Is it reproducible, i.e. if you repeat this command, does it fail again?

Dima

--
You received this message because you are subscribed to the Google Groups "sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/522bab60-a7e7-47f4-8d17-2c8ce329b85dn%40googlegroups.com.

[Martin R]

It is not my machine, just the bot, see the link above.

[dmo...@deductivepress.ca]

I think this is issue #35715 on the sagemath github. So I think the discussion should go there.

[Giacomo Pope]

This was also something I saw in: https://github.com/sagemath/sage/issues/37455 and I haven't been able to locally reproduce it either.