[Bug tree-optimization/111924] New: Look into using movement_possibility (in tree-ssa-loop-im.cc) for empty_bb_or_one_feeding_into_p for phi-opt

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111924

            Bug ID: 111924
           Summary: Look into using movement_possibility (in
                    tree-ssa-loop-im.cc) for
                    empty_bb_or_one_feeding_into_p for phi-opt
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Keywords: internal-improvement
          Severity: enhancement
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
  Target Milestone: ---

I noticed that movement_possibility/movement_possibility_1 does almost the same
as what some of the checks inside empty_bb_or_one_feeding_into_p does. It seems
like it might be a good idea to combine the 2 into one function and use that in
both PHI-OPT and Loop-im.

Note phi-opt does allow unconditionally a few known builtins/internal functions
too which seems like could benifit IM too.
Oh movement_possibility now disallows some movements of shifts which definitely
could be improved and it is definitely something which PHI-OPT should use too.