This is the mail archive of the
mailing list for the GCC project.
Re: [tree-ssa] COND_EXPR lowering.
- From: law at redhat dot com
- To: Andrew MacLeod <amacleod at redhat dot com>
- Cc: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>, gcc-patches <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 24 Oct 2003 12:04:48 -0600
- Subject: Re: [tree-ssa] COND_EXPR lowering.
- Reply-to: law at redhat dot com
In message <1067017738.14175.3098.camel@p4>, Andrew MacLeod writes:
>On Fri, 2003-10-24 at 13:46, email@example.com wrote:
>> In message <20031024174329.GA17642@atrey.karlin.mff.cuni.cz>, Zdenek Dvorak
>> >> You can't always use the CFG for everything.
>> >I don't say for everything. But following the gotos is the reason for
>> >having cfg.
>> Err, there are cases where we may want to follow gotos without a CFG. Say
>> in that wonderful little place when we've translated out of ssa form,
>> but have not translated into RTL.
>but then the bb_for_stmt wouldn't matter if there is no CFG :-).
Very true. Good point.
> What else is in the stmt annnotation that might.
Well, looking at the statement annotation, I don't see anything particular
useful for this case (arms of a COND_EXPR), except for the block annotation.
Looking at the block annotation, again, I don't necessarily see anything
all that important.