> John Hay <[EMAIL PROTECTED]> writes:
> > Let me hijack this a little. How many of you WARNS= adding people
> > consider different compile/code paths than the one your machine
> > exercise? For instance the one "make release" will exercise? The
> > WARNS=1 in libexec/Makefile.inc breaks "make release" because
> > telnetd is then compiled, but it isn't warning free.
> This is a good reason why non-zero WARNS should only be set in leaf
> Makefiles.

IMO, this is a good reason to not have WARNS contain -Werror at this
time. NO_WERROR is a good way to fix this (again IMO). I see a great
need to let warnings "hang out", and in an ideal world I see an need
for (new) warnings to break things. I see no need for warnings to
hold back a project as important as GCC3, and NO_WERROR is the
cleanest solution.

I do not expect others to agree with (or like) this.

o       Mark Murray
\_      FreeBSD Services Limited
O.\_    Warning: this .sig is umop ap!sdn

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Reply via email to