I think there is some rationale for having that code there, as we have found it useful to have finer-grained control over optimisations in KLEE in order to improve performance.
Best, Cristian On 22/03/2015 19:10, Dan Liew wrote: > Hi, > >> There was some discussion about linking in Klee earlier on github >> (https://github.com/klee/klee/pull/141). My preference is to remove all >> linking from Klee. A provided wrapper script can do the necessary linking >> using llvm-link and possibly some opt calls for internalization of functions >> and global dead code elimination. What are the current plans for linking >> within Klee? > > I don't know what the current plans are but I would very much like to > see optimisation and linking done outside KLEE. I'm one of the authors > of the linking code borrowed from LLVM 3.3. I was never happy with it > and I would like it removed. It's far too fragile as attempts at > porting to LLVM 3.5 and 3.6 have proved. > > That would simplify KLEE's internals and leave it up to the author of > the LLVM bitcode to do linking and optimisation. > > Thanks, > Dan. > > _______________________________________________ > klee-dev mailing list > klee-dev@imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/klee-dev > _______________________________________________ klee-dev mailing list klee-dev@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/klee-dev