[m-dev.] Concurrency in Mercury

Ralph Becket rafe at cs.mu.OZ.AU
Mon Feb 28 09:26:09 AEDT 2005


I gather from past discussions with Zoltan and Tom that the .par grades
suffer from being built on top of Posix threads, the design and
implementations of which are buggy.  Zoltan, can you elaborate on these
issues?

According to Julien, Zoltan thinks we should try to get an honours
student (at least) to work on a new concurrency substrate supporting
much lighter-weight thread construction.

One advantage of Posix threads is that they communicate with the OS so
that, say, an IO operation in one thread that blocks, does not cause the
whole application to block.  Assuming we don't want to get involved in
that particular can o' worms, I suggest that any new concurrency
framework still use Posix threads, one or more for the Mercury
application and one or more to handle blocking operations in an
asynchronous fashion.

-- Ralph
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------



More information about the developers mailing list