<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN"> <HTML> <HEAD> <META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8"> <META NAME="GENERATOR" CONTENT="GtkHTML/3.26.0"> </HEAD> <BODY> On Mon, 2010-12-13 at 16:37 -0800, Alan Coopersmith wrote: <BLOCKQUOTE TYPE=CITE> <PRE> Trevor Woerner wrote: > On Mon, Dec 13, 2010 at 7:03 PM, Gaetan Nadon <<A HREF="mailto:memsize@videotron.ca">memsize@videotron.ca</A>> wrote: >> The resulting variables XORG_CFLAGS and XORG_LIBS are unique to each package >> and hold different values. >> This is where I think the concept of a cache does not hold. The last write >> wins and if you pick a random >> package to rebuild, it will pick the value written last. > > Ah okay, I think I see where you're going. In my original email I was > trying to say: "maybe all the variables between all the packages > should be coordinated so they don't clash" but seeing how huge of a > job that would be it's probably much safer this way. I'm guessing it's > probably not being used by many people, and I'm all for getting rid of > clutter :-) We tried that early on in modularization for all the libraries and apps, which is why they have strangely worded configure messages such as: checking for ICO... yes when you're building ico (it should really be "checking for the packages that ICO needs"), due to commits such as: <A HREF="http://cgit.freedesktop.org/xorg/app/ico/commit/?id=5c31a07dbcb735445951f7f7cae8579ee16262d5">http://cgit.freedesktop.org/xorg/app/ico/commit/?id=5c31a07dbcb735445951f7f7cae8579ee16262d5</A> Unfortunately, it doesn't seem to have been consistently applied or kept up. </PRE> </BLOCKQUOTE> An impossible task, if you ask me. The cache should error out on duplicates. I assume it's an Acked-by from at least Trevor to remove this from build.sh so new comers (or myself!) don't get burned on this. Thanks for the review! </BODY> </HTML>