> > a configure test only proves whether the build machine can deal<br />> > with the flag, not whether the
machinethe executables will<br />> > ultimately run on knows what the flag means. We cannot assume that<br />
>> the build and execution boxes are the same. (In general,<br />> > AC_TRY_RUN tests are best avoided
becauseof this.)<br />> <br />> I understand why that is important in general, but as a shop which<br />>
buildsfrom source, and is fine with a separate build for each<br /> > hardware model / OS version combination, it
wouldbe great if any<br />> optimizations which are only available if you *do* assume that the<br />> build
machineand the run machine are the same (or at lease<br />> identical) could be enabled with some configure switch.
Maybe<br/> > something like --enable-platform-specific-optimizations.<br />> <br />> I don't know if any such
possibleoptimizations currently exist, I'm<br />> just saying that if any are identified, it would be nice to have
the<br/> > option of using them.<br /><br />I can't say the right way to go for now, but I'd like binary<br
/>packagescould enjoy the effect of my patch as far as possible so<br />that I made lwarx hint test run in configure
runtime.<br/><br />Regards,<br />Manabu Ori<br />