Igor Schein on Wed, 15 Jul 1998 13:07:08 -0400

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

Re: 0-parameter functions (was Re: pari-2.0.10.beta)

On Wed, Jul 15, 1998 at 11:33:36AM -0400, Karim BELABAS wrote:
> [Gerhard Nicklasch:]
> > The real problem is that cgeti(x) is getting called with x==0, which makes
> > a royal mess of the PARI stack.
> > 
> > Karim is working on it but the mail system at U Paris Sud is STILL down
> > after three days, so he may have severe difficulties communicating his
> > findings.
> [Looks like the mail server is up again... Hope that 
> 1) it will last.
> 2) I didn't miss anything; will check the archive later...]
> [Henri Cohen:]
> > As a temporary (?) fix to the bug described by M. Atria with 0-parameter
> > functions,
> > line 1349 of anal.c replace cgeti(nparam) by cgeti(max(1,nparam)).
> [Igor Schein:]
> > This still doesn't fix Gerhard's variation on the same theme:
> > 
> > mysqrt()=sqrtint(1)
> > 1.-mysqrt()
> > 
> > It doubles the stack a few times, in my case up to 64M
> > and then SEGFAULTs.
> Hum, the minimal patch sent by Henri Cohen _should_ have worked, unless
> I'm missing something and the following patch is going to fail as well,
> although both patches correct a genuine bug that would fit the reported
> problem... (Besides, it fixes the problem for me).
> In any case, a better patch (from the point of view of code cleanliness
> as

The *better* patch didn't work for me.  The same behavior on 1.-mysqrt()
And also, I couldn't apply your patch with neither native Solaris 2.5.1 /bin/patch
nor with GNU patch.  Only native IRIX /usr/sbin/patch worked.  Do you have
any idea why?  Any flags you can suggest?