Dave Hansen wrote: [Mon Mar 07 2005, 12:46:35PM EST] > On Sun, 2005-03-06 at 11:02 -0500, Bob Picco wrote: > > Second promised patch. This patch satisfies SPARSEMEM requirements in > > alloc_node_mem_map. This was boot tested on all three memory models. > > That's certainly a simple solution. Do you prefer the weak symbols to > doing the more standard empty definitions in a header, conditional on > #ifdefs? I find weak symbols hard to debug, because there tend to be > multiple config options to track down. > > -- Dave > > > Well I prefer weak usage to more conditionals. I've never had a problem with debugging. Both solutions work for me. I was just offering an alternative. bob
