[m-rev.] for review: factoring out the types of common static cells

[Zoltan Somogyi]

On 30-Apr-2003, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> Couldn't the code in question just be exported from llds_common,
> or put in some other module such as code_util.m,
> and then called from both llds_common and llds_out?

That could be done, but only by adding more complications that would be
totally unused except in benchmarking situations. The code to compute the
cell's argument types hasn't just been moved, it has also been specialized.
I would also need a modified copy of the code that prints the declarations
and definitions of common_datas, that does mostly the same thing for
creates that haven't been turned into common_data. This code would differ
in generating C type and variable names that follow a different template,
because using the same template wouldn't be safe due to possible accidental
name collisions.

> So the "--no-common-data" option no longer works?

It would be surprising if it worked after being deleted :-)

> That is not good: it would make it much more difficult to do "fair"
> comparisons between the MLDS and LLDS back-ends
> (currently the MLDS back-end does not do common data optimizations
> 
> Also, in general I think it is a good idea if all optimizations
> are disable-able.

A simple way to provide that is to make --no-common-data make sure that
searches that check whether a cell with the given contents has been seen
before always fail. I will provide an updated diff using this scheme shortly.

Zoltan.
--------------------------------------------------------------------------
mercury-reviews mailing list
post:  mercury-reviews at cs.mu.oz.au
administrative address: owner-mercury-reviews at cs.mu.oz.au
unsubscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-reviews-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------