Just after I clicked on send it occurred to me... it's quite likely that since scons considers the command itself as part of the dependence information for a build (IIRC), then switching from one BATCH_CMD to another (or disabling it entirely) might force a complete rebuild of your tree even if no source has changed.
On Thu, Apr 3, 2008 at 9:05 PM, Steve Reinhardt <[EMAIL PROTECTED]> wrote: > Fine with me... I agree it's not clear why they're sticky. Wasn't there a > time originally when all the options were sticky? Maybe they date back to > that time, and never got moved over when we added the ability to have > non-sticky options. > > > On Thu, Apr 3, 2008 at 9:30 AM, nathan binkert <[EMAIL PROTECTED]> wrote: > > > Any objection to me making BATCH (and BATCH_CMD?) non-sticky options? > > The reason is that when I use distcc to compile but find an error, I > > then want to compile without distcc. > > > > I know that I could just set BATCH to false in the next one, but I > > don't quite understand why sticky is the right way to go. > > > > Nate > > _______________________________________________ > > m5-dev mailing list > > [email protected] > > http://m5sim.org/mailman/listinfo/m5-dev > > > >
_______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
