Bootstrap failure in libstdc++

Mark Mitchell
Sun Apr 27 12:33:00 GMT 2003

On Sat, 2003-04-26 at 12:31, Diego Novillo wrote:
> On Fri, Apr 25, 2003 at 11:04:47PM -0700, Zack Weinberg wrote:
> > This with an entirely unmodified tree.  Jan, you're cc:ed because
> > you're the last person to touch the x86 back end.
> > 
> I think that Mark's patch may be triggering a latent bug in the
> back end, actually.  The SPEC tester worked fine last night but
> failed today.  Undoing this patch allows me to finish building
> libstdc++:

Anything's possible; this patch worked fine on the branch, but it's
possible that something is causing things to go wrong on the mainline.

Unfortunately, I may not get to this as quickly as I should; if
reverting that patch fixes the problem, it's OK with me if you do that
-- on the mainline -- and send me email letting you know you've reverted
it.  Since it's not causing trouble on the branch, please leave it
there; it fixes critical bugs on systems without weak symbol support.


Mark Mitchell
CodeSourcery, LLC

More information about the Gcc-bugs mailing list