On Mon, 2011-10-17 at 11:50 +0200, Martin Jansa wrote: > On Sat, Oct 15, 2011 at 03:40:10PM +0100, Phil Blundell wrote: > > On Sat, 2011-10-15 at 16:34 +0200, Martin Jansa wrote: > > > On Sat, Oct 15, 2011 at 03:30:13PM +0100, Phil Blundell wrote: > > > > On Sat, 2011-10-15 at 12:24 +0200, Martin Jansa wrote: > > > > > * this way we can use > > > > > EXTRA_DRIMODULES_armv4t += ",glamo" in meta-openmoko layer and > > > > > EXTRA_DRIMODULES_armv4t += ",foo" in meta-bar layer without > > > > > knowledge > > > > > of other modules in other layers in stack > > > > > > > > Is this really better than using "DRIMODULES_append_armv4t = ..."? > > > > > > the point is that you don't need to know who is setting swrast there for > > > armv4t... > > > > I'm not sure I understand what you're saying there. Can you be more > > concrete? > > > > > > > > > > It's not really true that you can do it without any knowledge of what > > > > other layers are doing since if you end up with two competing > > > > EXTRA_DRIMODULES_foo overrides then one of them will get lost. > > > > > > > > > > so EXTRA_DRIMODULES_append_armv4t should be fine, right? > > > > Well, yes, but in that case it doesn't seem as though the EXTRA_ bit > > (and this patch) buys anything. > > OK, forget this patch.. I've forgot this behavior of _append_armv4t > syntax.. and I've changed meta-openmoko to use this too. > > can we change x86 and x86-64 examples in mesa-dri.inc to > DRIDRIVERS_append_x86 = ",i915,i965" > DRIDRIVERS_append_x86-64 = ",i915,i965" > so it will work as example for BSP layer maintainers that thay don't > need to care about setting > DRIDRIVERS = "swrast" > for their arch?
I'll take that patch... Cheers, Richard _______________________________________________ Openembedded-core mailing list [email protected] http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
