This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Flow analysis and optimization with computed gotos
- To: Brad Lucier <lucier at math dot purdue dot edu>
- Subject: Re: Flow analysis and optimization with computed gotos
- From: Richard Henderson <rth at redhat dot com>
- Date: Tue, 5 Dec 2000 00:28:25 -0800
- Cc: Michael Matz <matzmich at cs dot tu-berlin dot de>, gcc at gcc dot gnu dot org, feeley at iro dot umontreal dot ca, hosking at cs dot purdue dot edu, Andrew Macleod <amacleod at redhat dot com>
- References: <Pine.GSO.4.21.0012032302070.26391-200000@platon> <200012050730.eB57UGd08804@polya.math.purdue.edu>
On Tue, Dec 05, 2000 at 02:30:15AM -0500, Brad Lucier wrote:
> In these cases I'd be happier if these abnormal critical edges were
> just ignored, if that would be possible, and gcse performed on the
> subgraphs that do not involve these edges.
This sounds like an extraordinarily reasonable idea.
Certainly this scheme seems much easier than figuring out how
to work around the edges after the fact. Plus you'll get some
measure of cross-block optimization on your subgraphs that you
weren't getting on your larger functions.
We'd need modifications to the gcse/lcm code to allow it to work
on subgraphs (which would likely reduce max memory footprint at
the same time). We'd also need code to appropriately partition
the CFG; presumably there are well-known algorithms for doing this.
r~