Ilya Zakharevich on Tue, 16 Jan 2024 09:42:29 +0100


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

Re: Re: Big GOTCHAs WITH forprime()


On Sat, Jan 13, 2024 at 02:53:25PM +0100, Oliver Bandel wrote:
> >      my(s=191*10^15); forprime(p=s, s + 6*10^8,);
> 
> That's the code you are running/benchmarking.
> But I asked about how you measure the time, and in this context asked
> for a complete example.

 ##

> is. And if there might be hardware issues (defected RAM or so) that may
> cause the crash.

<<Scratching my head>>  What is your mental model of “defected RAM”?!

How can a defect in PHYSICAL memory lead to REPRODUCIBLE failures in
applications?  

> As I had no crash, this points not in the direction of a general problem of PARI.

I have seen “pathetic apologitics” — but this one really stands out!

  (Just in case, for context: at some moment, millions of computers
   were running with my implementation of malloc().  So this was
   spoken from decade(s) of experience in debugging memory access
   problems.)

Hope this helps,
Ilya