GCC build of HEAD failed for native with your patch on 2004-09-04T15:32:13Z.

[Andreas Tobler]

James E Wilson wrote:

> On Sat, 2004-09-04 at 14:21, Andreas Tobler wrote:
> 
>>Jim would you please take care of this breakage?
> 
> 
> Please note that I am not the one that broke mainline.  What I did was
> discover that 5 months ago a patch was checked in that had a typo that
> accidentally caused an important gcc feature to be turned off, and I
> just fixed the typo, making gcc work the way it was supposed to have
> worked all along.  Meanwhile, other people have checked in patches that
> broke mainline, but the breakage was not discovered until I fixed the
> typo.

Sorry, I was not aware of that.

> I don't mind volunteering some free time to work on this problem, though
> I do think that requiring me to fix the problem is unreasonable.  Other
> people can break gcc much faster than I can fix it.  There is no way
> that I can accept responsibility for all bugs introduced by other
> people.

True, especially in the last weeks.

> 
> I suspect that this particular breakage was caused by a patch from David
> Edelsohn in June (the extendsfdf2 define_insn_and_split patch), but I
> have not tried to verify that.  It is possible that some other patch
> created the problem.
> 
> For the record, I did send mail Friday offering to look at this problem
> if no one else did.  I didn't look at it immediately because,
> technically, I didn't create the problem.  Also, this is a 3-day holiday
> weekend in the US, so everything is a bit slow on this side of the pond,
> but I did come in during my holiday and check in a patch for it today.

Oh, I also did not know that there was a long weekend in the states.

Well, anyway, thank you for fixing this. I could bootstrap on darwin and 
  ppc linux with my slow own paid hardware. I wish I'd have a G5.

Regards,
Andreas