On Sun, May 02, 2010 at 06:57:09PM -0400, Michael Gilbert wrote:
> if this is a compiler issue, i can't really do anything about it. if
> you're feeling proactive, you can bisect their commits, and issolate
> the patch that caused the regression? anyway, the best course of

I fetched a copy from git.infradead.org, but building 4.4.0 died of a
C++ error after 45 minutes, so that sorta killed my enthousiasm right
there. Not to mention that I could not figure out how to tell gcc to
stop building itself with itself, thus rendering my ccache useless
(--disable-bootstrap did not do the trick). Re-building gcc from
scratch on an old Athlon is not fun.

I'm CCing the Debian GCC team, in case they have a suggestion to make
bisecting easier. (Or better yet, if they can reproduce the bug
themselves.)


--
In seeking the unattainable, simplicity only gets in the way.
-- Epigrams in Programming, ACM SIGPLAN Sept. 1982


--
To UNSUBSCRIBE, email to debian-gcc-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20100503153100.GA31499@toroia.fbriere.dyndns.org"> http://lists.debian.org/20100503153100.GA31499@toroia.fbriere.dyndns.org