Bill Allombert on Fri, 6 Jul 2001 10:20:29 +0200


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

Re: PARI:Failure of ellheight on elliptic curve [0,-1,0,-784,8704]


On Thu, Jul 05, 2001 at 06:14:52PM +0100, Martin Prickett wrote:
> I am Martin Prickett, a phD student of John Cremona at Nottingham. I am
> using PARI and have discovered that if I take curve
> e=ellinit([0,-1,0,-784,8704]) and point P=[-8,120] then
> ellheight(e,P) crashes the computer by causing it to run out of memory on
> Linux machines. However on Dec alpha machines (which are 64 bit) it gives
> the correct answer.

First I suppose your Linux machine is a x86 ?

I try this on a x86 Linux machine and on a sparc solaris machine
and I got.
  ***   the PARI stack overflows !
  current stack size: 33554432 (32.000 Mbytes)
   
which is not a computer crash, just a stack overflow condition.

I also try it on a alpha Linux machine and it work
? ellheight(e,P)
%3 = 0.74629920869451370890943760363155317690

So you are right is a 32bit-only issue.

Cheers,

Bill.