[Agda] Seg-fault

[Wolfram Kahl]

On Mon, Oct 15, 2012 at 12:00:06PM +0200, Nils Anders Danielsson wrote:
> On 2012-10-13 21:32, Serge D. Mechveliani wrote:
> >I have programmed the  merge sorting function  for a list over any
> >DecTotalOrder,
> >with the result record including a proof for orderedness.
> >
> >It is compiled for the RTS options, and run with  +RTS -M... -RTS.
> >For a certain large list over Nat, it finishes for  -M700m and 900m,
> >but runs into  Segmentation fault  for 800m.
> 
> That's interesting. Please report this on the bug tracker.

In my experience, the stable version (2.3.0*) is far more prone
to segmentation faults when the heap gets full than the development version ---
I am not even sure I have seen any segfaults with recent (<3months)
development versions (which also tend to be built with newer GHC and
library versions...)


Wolfram