Re: ICE in optimized build of 20000103

["Martin v. Loewis" <martin at loewis dot home dot cs dot tu-berlin dot de>]

> The extra flags are passed as environment settings for CFLAGS and/or
> CXXFLAGS, set before the configure step, and are thus used whenever
> optimization is allowd (stage1? stage2 etc).

Don't you think you should report relevant settings of environment
variables, too? Especially if nobody is capable of reproducing your
bug report?

> Yes, I get exactly the same problem on an i686 laptop (brand new).  In
> addition, I've occasionally seen some other reports of this bug.

How do you know it is the same bug, if you don't even know what the
bug is?
> 
> Here are some references to problems with bootstrap builds involving
> _muldi3:
> 
> http://gcc.gnu.org/ml/gcc-bugs/1999-10/msg00690.html

Perhaps not the same bug. You get a crash on expr.c:6584, he gets a
crash on expr.c:6502 (unless it is the same code due to a 82-lines
shift of code).

> http://gcc.gnu.org/ml/gcc-bugs/1999-12/msg00543.html

Not the same bug, it crashes at reload.c:5846.

> http://gcc.gnu.org/ml/gcc-bugs/1999-09/msg00184.html

Definitely not the same bug, as gcc was configured with
--enable-checking, and it occured on haifa-sched.c:3245

> There are lots more.  What is puzzling is that you don't get the
> problem at all.

Do you really think that building totally fails for every Linux
system, and nobody does anything about this for four months?

Regards,
Martin