[mercury-users] Erlang backend

[Peter Wang]

2009/2/3 Richard O'Keefe <ok at cs.otago.ac.nz>:
> Looking at the Mercury web site, I was enchanted to discover
> an Erlang back end.  Amazing.  What did you use for documentation
> on .beam files?  How well does this actually work?

We generated Erlang source to be compiled with an Erlang compiler, so no
documentation on .beam files was necessary.  It seemed to work okay, as far as
it was tested.  At the time, we could compile the Mercury compiler to Erlang
and run it in the Erlang interpreter.

Unfortunately it was too slow so we (Mission Critical) lost interest.  The main
problem was, as I recall, Erlang's treatment of static data.  To implement RTTI
and typeclasses we need some tables, which we generated as Erlang functions
returning tuples.  With the standard Erlang compiler those tuples (which can be
rather large) would be reconstructed each time the function was called.  HiPE
was better, but (I think at that time) tuples containing binaries wouldn't be
statically allocated.

Peter
--------------------------------------------------------------------------
mercury-users mailing list
Post messages to:       mercury-users at csse.unimelb.edu.au
Administrative Queries: owner-mercury-users at csse.unimelb.edu.au
Subscriptions:          mercury-users-request at csse.unimelb.edu.au
--------------------------------------------------------------------------