https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124695
--- Comment #5 from Torbjorn SVENSSON <azoff at gcc dot gnu.org> --- (In reply to avieira from comment #3) > Hi Torbjorn, > > This looks like a testism to me, the codegen doesn't actually change. Not > too concerned about this. I agree, but the question is then what should be done to avoid having a failure that is not problematic? Is the problem simply that some optimization pattern does not match for arm-none-eabi that matches for other targets and as a result, the patterns in the backprop file does not match? (In reply to Andrew Pinski from comment #4) > I think this is a dup of another bug report. Most likely as the fail exist in both my runs of Arm GNU Toolchain 14.3.Rel1 and 15.2.Rel1.
