Excessive compilation time in phase for Lifting With-Loop bodies into threads


Bugzilla Link	833
Created on	Mar 14, 2011 16:36
Resolution	FIXED
Resolved on	Apr 08, 2011 01:44
Version	svn
OS	All
Architecture	PC

Extended Description

This phase currently does some analysis for the cross-product of every line of intermediate code and for every WL range. When compiling even Livermore loops this tie can be excessive. This code is known to need rewriting so I haven't uploaded a failure case. I'm raising this bug simply as a reminder.

To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information