[polly] r244606 - Revise the simplification of regions

Johannes Doerfert via llvm-commits llvm-commits at lists.llvm.org
Wed Aug 12 01:01:27 PDT 2015


On 08/12, Michael Kruse wrote:
> 2015-08-12 9:38 GMT+02:00 Johannes Doerfert <doerfert at cs.uni-saarland.de>:
> > On 08/12, Michael Kruse via llvm-commits wrote:
> >> Sich PHI nodes can only appear in the entry
> >> node. So I think solutions 1 and 3 do work.
> > What is wrong with adding a pre SCoP block for all entry block PHI
> > nodes? That solved the problem before and I cannot see the negative side
> > effects and it needs only a minimal amount of code.
> 
> Its adding more code to get around issues in other places. Isn't it
> better to reduce the overall complexity.
Where does it add complexity or code? It was working before you changed
the region simplification, hence replicate the old behaviour there
(which is ~2-3 lines of code) and we are done.

> >> This all doesn't look like designed deliberately. And it is what I
> >> meant with make simplifyRegion "stable" in D11867.
> > In the original PHI codegen patch there was a detection of PHI nodes
> > that were at some point part of the region (in the entry block) and
> > split out of it by one of the region simplification methods just before
> > the codegen part. Once detected these PHI's were deliberately ignored in
> > the block codegen as they were no longer part of the region (SCoP).
> >
> > I cannot argue for the simplification and splitting stuff but the PHI
> > codegen was aware of such PHI nodes and was handling them at some point.
> 
> Do you mean by that that it has been removed from the current codebase?
Tobias cleaned a lot of things up in the block generator, I don't think
he has removed this part as it would have crashed lnt even before your
patch.

-- 

Johannes Doerfert
Researcher / PhD Student

Compiler Design Lab (Prof. Hack)
Saarland University, Computer Science
Building E1.3, Room 4.31

Tel. +49 (0)681 302-57521 : doerfert at cs.uni-saarland.de
Fax. +49 (0)681 302-3065  : http://www.cdl.uni-saarland.de/people/doerfert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 213 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150812/915653c9/attachment.sig>


More information about the llvm-commits mailing list