> From: Junio C Hamano [mailto:gits...@pobox.com] > Sent: Tuesday, September 18, 2012 7:03 PM > To: Joachim Schmitz > Cc: git@vger.kernel.org > Subject: Re: [PATCH/RFC] Port to HP NonStop > > "Joachim Schmitz" <j...@schmitz-digital.de> writes: > > >> From: Junio C Hamano [mailto:gits...@pobox.com] > >> Sent: Tuesday, September 18, 2012 9:57 AM > >> To: Joachim Schmitz > >> Cc: git@vger.kernel.org > >> Subject: Re: [PATCH/RFC] Port to HP NonStop > >> > >> "Joachim Schmitz" <j...@schmitz-digital.de> writes: > >> > >> > Needs a different link order in Makefile: libintl before libiconv. > >> > This may affect other platforms, so needs some checking. > >> > >> It will, and it needs customization, not checking. > > > > How? > > By not moving things around to affect other people, but having them > in the order you want only for your platform (and other future ports > that need similar treatment)? > > For example, if you need libintl before libiconv, a quick and dirty > way may be > > diff --git i/Makefile w/Makefile > index a49d1db..9b2cfd6 100644 > --- i/Makefile > +++ w/Makefile > @@ -1552,6 +1552,9 @@ ifdef NEEDS_LIBICONV > else > ICONV_LINK = > endif > + ifdef NEEDS_LIBINTL_BEFORE_LIBICONV > + ICONV_LINK += -lintl > + endif > EXTLIBS += $(ICONV_LINK) -liconv > endif > ifdef NEEDS_LIBGEN > > and your linker command line may have "-lintl -liconv -lintl" but > that wouldn't be an error, and you know you won't be affecting > other platforms that do not use NEEDS_LIBINTL_BEFORE_LIBICONV, no?
OK, sounds good, I'll try that and submit an updated patch if it works that way (and I'm pretty sure it does). Anything else I should change in my patch? -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
