[LLVMdev] some superoptimizer results
John Regehr
regehr at cs.utah.edu
Wed Jul 22 21:19:46 PDT 2015
> Hence "most". From the blog post it seems like this batch is deliberately obtained by running Souper on
> particularly short sequences; I'm looking forward to seeing what it does with longer sequences,
> especially ones containing "phis and path conditions".
I can easily provide these (and will) but I have been concerned that
Souper's output can be really hard to understand. I was hoping that these
shallow expressions would provide you folks with the right kind of
information.
In any case, my only goal here is to do whatever is most broadly useful to
LLVM. I suspect a fair amount of iteration will be required.
I might work on a backend superoptmizer next. I was not interested in
doign that one first since it sounds somewhat more complex.
John
More information about the llvm-dev
mailing list