Gerhard Niklasch on Mon, 29 Jun 1998 22:01:00 +0200


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: SEGV n 2.0.9 on Solaris 2.5.1


In response to:
> Message-Id: <19980629153320.G16049@io.txc.com>
> Date: 	Mon, 29 Jun 1998 21:33:20 +0200
> From: Igor Schein <igor@txc.com>
> 
> Hi, the following causes a SEGV:
> % echo '2^2^26' | gp-dyn -q > /dev/null
>   ***   segmentation fault: bug in GP (please report).
> %

Thanks.

On what machine architecture was that?

Does it blow up at once, or after a delay?  (IIRC in this case the
object that is to contain the result is allocated on the PARI stack
before the powering starts, and I _thought_ we were checking the size
at that point, but there may indeed be a bug lurking there.)

Could anybody please try to reproduce (or not reproduce) it on
a DEC Alpha, or some other 64-bit thing?  :)

If it doesn't blow up at once, what happens when you turn on
default(debugmem,2) first?

> To: "nikl@mathematik.tu-muenchen.de" <nikl@mathematik.tu-muenchen.de>,
>         "pari-dev@list.cr.yp.to" <pari-dev@list.cr.yp.to>

*Please* don't do that.  I hate getting the same message twice within
half a second.  (And I'm not `da boss' anyway.  In this case you may
have something of a point since the code being exercised actually
originated with me a little while ago, but I _am_ reading the list --
I'm archiving it on the website, after all! :^)

Cheers, Gerhard