Re: [PATCH] kmalloc_percpu

[David S. Miller]

   From: Andrew Morton <[e-mail]>
   Date: Mon, 5 May 2003 22:48:15 -0700

   I think so.  So we'd end up with:
   
   - DEFINE_PER_CPU and kmalloc_percpu() work in core kernel, and use the 32k
     pool.
   
   - DEFINE_PER_CPU in modules uses the 32k pool as well (core kernel does the
     allocation).
   
   - kmalloc_per_cpu() is unavailble to modules (it ain't exported).
   
   AFAICT the only thing which will break is sctp, which needs a trivial
   conversion to DEFINE_PER_CPU.
   
Your grep is faulty, we're using kmalloc_percpu() in ipv6 for per-cpu
and per-device icmp stats.

You solution doesn't work in that case.  Also ipv4 will have the same
problems if we make that modular at some point.

I also don't see how this fits in for your ext2 fuzzy counter stuff.
It isn't a "module" for most people, I can't even remember if I've
ever built ext2 non-statically.  :-)  It almost appears as if you
are suggesting kmalloc_percpu() is not usable at all.

So there we have it, there are a total of 3 users of kmalloc_percpu()
(ipv4/ipv6/diskstats) so let's decide if it's going to continue to
live longer or not before there are any more. :-)