Performance with large Makefiles. Preferred patch style?

[Dan Kegel]

Hi,
Jeff Evarts has been looking at performance of Make
on some large makefiles in the real world.  He profiled
it, identified a few bottlenecks, and came up with simple
fixes.  He's busy cleaning up the patches and testing them,
so I figured I'd "help" by asking what the preferred way
to deliver patches for things like this is.
Should we file one bug report in
http://savannah.gnu.org/bugs/?group=make
and one patch in
http://savannah.gnu.org/patch/?group=make
per bottleneck?

Thanks,
Dan

--
Wine for Windows ISVs: http://kegel.com/wine/isv