[m-rev.] for review: move to support solver initialization

[Ralph Becket]

Zoltan Somogyi, Tuesday,  9 August 2005:
> For review by anyone. I tested the code by hand. Automatic tests can't
> be conveniently done until the compiler starts using this functionality.
> 
> Zoltan.
> 
> Make a step towards supporting the automatic running of compiler-provided
> initialization functions at startup. The idea is that in each module that
> defines a solver type which wants to keep the solver state in a global
> variable, the compiler will generate a function to initialize this global
> variable and put a line containing
> 
> 	REQUIRED_INIT function_name
> 
> in the generated .c file. Mkinit will then see this and put a call to the named
> function in the program's init_modules_required function, which (once this diff
> is installed) will be called from the tail end of mercury_init.

That looks good.
--------------------------------------------------------------------------
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
--------------------------------------------------------------------------