When to protect the user - Implementing a sanity check for the syndrome decoder

[J]

Hi,

While testing and toying with the 8.9.beta3 I could reliably crash sage
on a "user" system with 16GB RAM and crash it on a testsystem with
500GB RAM, so I would like to suggest ( and implement a sanity check for
the maximum_error_weight of the syndrome decoder.

What are your thoughts on such a check?

It seems a similar check is implemented with matrices, where they are
not printed by default if they reach a certain size, but instead you
need to call them with the `.str()` argument.

I did the tests using the Goppacode on `GF[2^6]` with `g=x^9+1` from the
examples.

Greets
J

[Dima Pasechnik]

On Sun, Sep 22, 2019 at 9:49 AM J
<mailinglists-sa...@927589452.de> wrote:
>
> Hi,
>
> While testing and toying with the 8.9.beta3 I could reliably crash sage
> on a "user" system with 16GB RAM and crash it on a testsystem with
> 500GB RAM, so I would like to suggest ( and implement a sanity check for
> the maximum_error_weight of the syndrome decoder.
>
> What are your thoughts on such a check?

surely if there is a robust way to detect inputs which won't lead to a
successful computation,
then why not?

One can aslo make such a check conditional on a parameter.

>
> It seems a similar check is implemented with matrices, where they are
> not printed by default if they reach a certain size, but instead you
> need to call them with the `.str()` argument.
>
> I did the tests using the Goppacode on `GF[2^6]` with `g=x^9+1` from the
> examples.
>
> Greets
> J
>

> --
> You received this message because you are subscribed to the Google Groups "sage-coding-theory" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sage-coding-the...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sage-coding-theory/20190922081126.braabdrnez65ymjb%40deathbolt.927589452.space.

[J]

On 19-09-22 10:53:01, Dima Pasechnik wrote:
> On Sun, Sep 22, 2019 at 9:49 AM J
> <mailinglists-sa...@927589452.de> wrote:
> >
> > Hi,
> >
> > While testing and toying with the 8.9.beta3 I could reliably crash sage
> > on a "user" system with 16GB RAM and crash it on a testsystem with
> > 500GB RAM, so I would like to suggest ( and implement a sanity check for
> > the maximum_error_weight of the syndrome decoder.
> >
> > What are your thoughts on such a check?
>
> surely if there is a robust way to detect inputs which won't lead to a
> successful computation,
> then why not?

If we know the memory of the system, a quick guestimate of the needed
size is
\[ \sum_{i=1}^{maximum_error_weight} \binom{n}{i}\#F * MEMORY_PER_ENTRY \]

>
> One can aslo make such a check conditional on a parameter.
>

This is a very good idea

I will have a look at it next week