> Is there any particular reason that RUBY is a global build option, and
> not a per-build one?  (That is, I can't build MIPS_SE w/o Ruby and
> then build ALPHA_SE or even ALPHA_SE_MOESI_hammer with Ruby in the
> same build directory.)  Nate, you're the one that added RUBY to
> global_sticky_vars and not sticky_vars... was there a reason for this?
>  (See http://repo.m5sim.org/m5/rev/ba6fe02228db.)
Only that I'm eventually trying to get rid of the whole sticky_vars
thing and get to what is in effect global_sticky_vars.  The reason I'm
trying to do that is that I would eventually like to get to the point
where we're not building stuff multiple times if we can avoid it.
That said, for this case, it doesn't make much of a difference, so I
won't object.  I really would like to avoid us getting into the place
where one build dir supports all different sorts of build options,
making it more difficult to avoid duplicate compiles.

> Unless there's a good reason, I'm going to move it to sticky_vars so I
> can force the builds that imply Ruby (like ALPHA_SE_MOESI_hammer) to
> turn Ruby on... otherwise if the global RUBY flag is false you end up
> building them anyway and get what I believe is N functionally
> identical copies of ALPHA_SE that each insist on having all the
> regressions run on them.
Go for it.

> OK, I see that not having it as a global option causes the check in
> src/mem/ruby/SConsopts to fail, but I commented that out and things
> seem OK.
Yeah, you just get extra options that don't mean anything not too big a deal.

  Nate
_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev

Reply via email to