Patch: Enforce -Werror during bootstrap

[Zack Weinberg]

"Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> writes:

> Here's an attempt to make use of -Werror during bootstrap to stop
> warning regressions from creeping back in.
>
> Following Zack's suggestion from the last time I tried this:
> http://gcc.gnu.org/ml/gcc-patches/2001-09/msg00280.html
...

I hate to rain on your parade, but my attitude toward this is still
pretty much what it was back then:

| I think this is backward.  First we should get to zero warnings
| throughout the compiler, _then_ we talk about -Werror.  If only
| because if we do it your way, we'll be adding explicit compile rules
| to module after module.  I don't think there's enough marginal benefit
| from -Werror applied to a few modules, to eat that kind of complexity
| burden in the Makefile which is already too complicated.
[http://gcc.gnu.org/ml/gcc-patches/2001-09/msg00129.html]

You've got the complexity burden down as far as it's going to go, but
think about the consequence of this: there's eventually going to be a
long long list of

foobar.o-warn = $(WERROR_FLAGS)

which will, I venture to guess, never go away.  Now think about what
will happen when (not if) you start needing to conditionalize this on
target triples.

zw