* Iain Sandoe <idsan...@googlemail.com> [2020-01-30 15:21:08 +0000]: > >> Can you clarify why there’s no need to match the configury changes in > >> libcpp > >> / gcc / libstdc++ ? > > > > This is the same issue that Tobias pointed out, and was a result of me > > incorrectly trying to regenerate the configure files. > > .. but it seems that the combination of the two things produced the > configuration > mismatch .. > > > Let me know if this resolves the problems you're seeing and if it does > > I'll prepare it for submission. > > I did a smoke test on Darwin19 and the amended patch allowed bootstrap to > complete. ICONV is seen in both libcpp and gcc (that’s not an exhaustive > check > and the testsuite is still running). > > I’ll add this to my overnight runs - “overnight” means results later tomorrow > for the > slower targets - but it looks encouraging.
Thanks, for checking this so quickly. I'll prepare this as a proper patch and post this again later tonight for review, but if your results are good tomorrow then I'll go ahead and push this. Thanks, Andrew
