This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: PATCH: Implement DJ's observation on $(WERROR)


In article <200303200125 dot h2K1PDR00987 at greed dot delorie dot com>,
DJ Delorie<dj at redhat dot com> writes:

> You'd have to double or quadruple the $'s (plus a myriad of quotes and
> protected quotes) to keep make from expanding it too early.  But
> that's not what I meant.  I meant do this:

> STAGE2_FLAGS_TO_PASS = \
> 	WERROR="@WERROR@" \
> 	STRICT_WARN="$(STRICT2_WARN) @WERROR@" \
>                                      ^^^^^^^^

Ah, but @WERROR@ is correctly expanded at configure-time in this context:

STRICT2_WARN = -Wtraditional -pedantic -Wno-long-long @WERROR@

My original questions (which I now see was not perfectly clear):

Do you favor seeing:

STRICT2_WARN = -Wtraditional -pedantic -Wno-long-long @WERROR@

or

STRICT2_WARN = -Wtraditional -pedantic -Wno-long-long $$$$(WERROR)
                                                      ^^^^^^^^^^ whatever is
						      really required to
						      pass through to stage2,3

?

Was there any reason to avoid seeing (what the patch as posted did):

STRICT2_WARN = -Wtraditional -pedantic -Wno-long-long @WERROR@

?


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]