Re: Segfault in read()

[Bill Allombert <Bill.Allombert@math.u-bordeaux.fr>]

On Fri, Jul 12, 2024 at 08:18:14AM -0700, Ilya Zakharevich wrote:
> On Thu, Jul 11, 2024 at 07:06:32PM -0700, Ilya Zakharevich wrote:
> > Reading the attached file (created on Windows 7 in the same session by
> > the latest stable version) leads to a segfault.
> 
> Bill asks me how this file was created.  It is the only way I know to
> debug “a failing” PARI construct.
> 
>   OUT=[1];
>   Data = Start␣a␣failing␣code;   OUT[1] = Data;   Fun␣code␣which␣crashes();
> 
> Only I put writebin(Data) in between too.  The literal excerpt:
> 
>     if(dbg, writebin("pos.wrbin",pos); POS[1]=pos; print("asX="asX", pos="pos", neg="neg", xRefl="xRefl", yRefl="yRefl,if(#pos>1,[pos2[1],pos2[2]]=pos[1..2]); …

OK, but what is "pos" supposed to be  ?

What dbg_x(pos) gives ?

For what I see, pos is a vector with 5 components, whose first component is a t_LIST 
which is corrupted.

So it is unclear if the bug is in writebin or read.

Cheers,
Bill