This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: Adding new modifier to GCJ front-end
- From: Tom Tromey <tromey at redhat dot com>
- To: corsaro at cse dot wustl dot edu
- Cc: GCJ <java at gcc dot gnu dot org>
- Date: 28 Sep 2003 14:35:00 -0600
- Subject: Re: Adding new modifier to GCJ front-end
- References: <1064780325.2062.107.camel@localhost.localdomain>
- Reply-to: tromey at redhat dot com
>>>>> "Angelo" == Angelo Corsaro <corsaro@cse.wustl.edu> writes:
Angelo> I saw the intimidating comment at parse.y:481
Angelo> /* Modifier TOKEN have to be kept in this order.
Angelo> Don't scramble it */
I think this just refers to parse.h:MODIFIER_WFL, which I see you
already found.
Angelo> - #define ACC_PHANTOM 0x0800 (in java-tree.h)
That's the same as the strictfp modifier.
Angelo> And added the proper definition in the gperf related files,
Angelo> and changed lex.c, and extended the size of the modifier_ctx
Angelo> array.
That all sounds good.
Angelo> What are the rationale behind having the modifier sorted?
Hmm, there may be some other subtle dependency on the order. If I
ever knew, I forgot.
Angelo> What is the purpose of PAD_TK?
No clue :-(. I'd suggest looking for `strict' (do a case-independent
search) and following along. I think that's the last modifier we
added.
Tom