Rakesh Nair via llvm-dev
2016-Apr-26 06:52 UTC
[llvm-dev] Case in which Maximum Fixed Point fails
For a theoretical study I have implemented Meet over all path for constant propagation. Since the lattice of constant propagation is non distributive, it is expected that Maximum Fixed Point calculation and Meet Over all Path may give different results. Can anyone give such an example program -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20160426/e23f3789/attachment.html>
Saleena Nazeer via llvm-dev
2016-Apr-26 12:08 UTC
[llvm-dev] Case in which Maximum Fixed Point fails
I've seen an example in "Monotone Data Flow Analysis Framework" by Kam and Ullman (available at https://www.cs.indiana.edu/~achauhan/Teaching/B629/2006-Fall/CourseMaterial/1977-acta-kam-monotone.pdf ) Please see the 5th page. On 26 April 2016 at 12:22, Rakesh Nair via llvm-dev <llvm-dev at lists.llvm.org> wrote:> For a theoretical study I have implemented Meet over all path for constant > propagation. Since the lattice of constant propagation is non distributive, > it is expected that Maximum Fixed Point calculation and Meet Over all Path > may give different results. Can anyone give such an example program > > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20160426/78425157/attachment-0001.html>