[m-rev.] for review: minor parser optimisation

Paul Bone pbone at csse.unimelb.edu.au
Wed May 28 11:34:35 AEST 2008


On Tue, May 27, 2008 at 02:19:58PM +1000, Peter Wang wrote:
> 
> PS. The problem I'm trying to solve is the reading of .analysis files.  When
> making an .analysis file, reading in other analysis files can take nearly half
> of the running time, mainly due to large structure sharing/reuse results.
> Suggestions would be appreciated.
> 

Soon I would like to implement ".advice" files.  Where a process can add
advice for the compiler when compiling that module.  This will replace
the feedback file used by Jerome for implicit parallelism.

The ".advice" files may contain any advice for the compiler relevant to
that module.  I don't see why the compiler can't update these either, if
it did it the right way.  The advice should be generic, in my case I'm
interested in telling the compiler which conjuncts within a procedure
should be parallelized.

he compiler doesn't have to obey to any particular piece of advice,
(this may depend on it's capabilities and enabled optimizations).  I
don't think this affects what you're doing, but it may be helpful to
know.

Ciao.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20080528/4418ce94/attachment.sig>


More information about the reviews mailing list