This is the mail archive of the
mailing list for the GCC project.
Re: Compilation performance comparison of 3.5.0 and TreeSSA trees on MICO sources as requested in: [tree-ssa] Merge status 2004-05-03
- From: law at redhat dot com
- To: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- Cc: dnovillo at redhat dot com, gcc at gcc dot gnu dot org
- Date: Wed, 12 May 2004 12:11:52 -0600
- Subject: Re: Compilation performance comparison of 3.5.0 and TreeSSA trees on MICO sources as requested in: [tree-ssa] Merge status 2004-05-03
- Reply-to: law at redhat dot com
In message <10405081026.AA17987@vlsi1.ultra.nyu.edu>, Richard Kenner writes:
> Mostly because of the breaking down of expressions into small trees,
> which are friendlier to the RTL expanders. That's the reason we have a
> tree recombination pass at the end of tree optimization.
>Sure, but my question is why do that breaking down and recombining with -O0?
We don't do recombination at -O0, or at least we shouldn't. Long term I
don't expect we'll continue to recombine expressions at the tree level.
Breaking the expressions down to gimple is a long term requirement if we are
to make any significant headway in simplifying the tree->rtl expanders.