[PATCH] D88684: llvm-reduce: Don't replace intrinsic calls with undef

Roman Lebedev via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Oct 13 10:32:37 PDT 2020


lebedev.ri added inline comments.


================
Comment at: llvm/tools/llvm-reduce/deltas/ReduceFunctions.cpp:33
   copy_if(Program->functions(), std::back_inserter(FuncsToRemove),
-          [&O](auto &unused) { return !O.shouldKeep(); });
+          [&O](Function &F) { return !F.isIntrinsic() && !O.shouldKeep(); });
 
----------------
arsenm wrote:
> lebedev.ri wrote:
> > fhahn wrote:
> > > I think a comment here would be good explaining why we exclude intrinsics. Also, should we also remove intrinsics from chunks to start with, to keep the numbering/chunk management consistent?
> > This comment still needs addressing.
> I added the comment in the last revision
The second part of it - `Also, should we also remove intrinsics from chunks to start with, to keep the numbering/chunk management consistent?`


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88684/new/

https://reviews.llvm.org/D88684



More information about the llvm-commits mailing list