Re: [PATCH 4/3] Replace dynamic percpu implementation

[Rusty Russell]

In message <[e-mail]> you write:
> 4. Extra dereferences in alloc_percpu were not significant, but alloc_percpu
>    was interlaced and kmalloc_percpu_new wasn't.  Insn profile seemed to
>    indicate extra cost in memory dereferencing of alloc_percpu was
>    offset by the interlacing/objects sharing the same cacheline part.
>    but then insn profiles are only indicative...not accurate.

Interesting: personally I consider the cacheline sharing a feature,
and unless you've done something special, the static declaration
should be interlaced too, no?

If you don't want interlaced, you should make your type
____cacheline_aligned for alloc_percpu, or use

	__alloc_percpu(ALIGN(sizeof(x), SMP_CACHE_BYTES), SMP_CACHE_BYTES)

Aside: if kmalloc_percpu uses the per-cpu offset too, it probably
makes sense to make the per-cpu offset to a first class citizen, and
smp_processor_id to be derived, rather than the other way around as at
the moment.  This would offer further speedup by removing a level of
indirection.

If you're interested I can probably produce such a patch for x86...

Thanks for the results!
Rusty.
--
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.