Richard Guenther<richard.guenther@gmail.com> writes:
Since I haven't hear anything from the darwin guys, I've also checked the
patch into the 4.5 and 4.4 branches as is. We can still remove the
TARGET_MACHO part if it proves to be superfluous now (as I would
expect).
It wasn't a regression so technically not applicable to the branches.
I think it was: gcc 3.4.6 bootstrapped fine, while 4.4+ doesn't. It's
still a bootstrap failure, so it seems to make sense for the branches,
too. I've explicitly asked for 4.4 and 4.5 branch approval, so I've
taken Uros' approval to apply there as well.
Sorry.