In message <16050.41490.798865.517036@napali.hpl.hp.com> you write:
>> - On NUMA, you want to allocate the per-CPU areas in node-local memory.
On Sat, May 03, 2003 at 02:40:57PM +1000, Rusty Russell wrote:
> Thanks: I'd forgotten about this. Even "big smp" machines often have
> different memory latencies.
I have already a patch to do this for ia32 NUMA systems (which present
issues not present on sane architectures).
I also didn't notice the lack of an override; #undef'ing the macro
__GENERIC_PER_CPU appeared to suffice for me.
-- wli
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed May 07 2003 - 22:00:17 EST