[m-rev.] diff: restore java backend after mlds global data change

Zoltan Somogyi zs at csse.unimelb.edu.au
Fri Sep 4 09:03:22 AEST 2009


On 02-Sep-2009, Peter Wang <novalazy at gmail.com> wrote:
> --- a/compiler/mlds_to_java.m
> +++ b/compiler/mlds_to_java.m
> @@ -330,8 +330,9 @@ output_java_src_file(ModuleInfo, Indent, MLDS, !IO) :-
> 
>      % Find and build list of all methods which would have their addresses
>      % taken to be used as a function pointer.
> -    find_pointer_addressed_methods(Defns0, [], CodeAddrs0),
> -    CodeAddrs = list.sort_and_remove_dups(CodeAddrs0),
> +    find_pointer_addressed_methods(GlobalDefns, [], CodeAddrs0),
> +    find_pointer_addressed_methods(Defns0, CodeAddrs0, CodeAddrs1),
> +    CodeAddrs = list.sort_and_remove_dups(CodeAddrs1),

I am pretty sure this is overkill, in that the vast majority of the definitions
in GlobalDefns cannot possibly contain code pointers.

I will look into this during my further work on ml_global_data.m.

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



More information about the reviews mailing list