Re: slab: setup allocators earlier in the boot sequence

From: Nick Piggin
Date: Fri Jun 12 2009 - 05:15:10 EST

On Fri, Jun 12, 2009 at 12:05:33PM +0300, Pekka Enberg wrote:
> On Fri, 2009-06-12 at 18:53 +1000, Benjamin Herrenschmidt wrote:
> > Now, if you find it a bit too ugly, feel free to rename smellybits to
> > something else and create an accessor function for setting what bits are
> > masked out, but I still believe that the basic idea behind my patch is
> > saner than yours :-)
> It's not the naming I object to but the mechanism because I think is
> open for abuse (think smelly driver playing tricks with it). So I do
> think my patch is the sanest solution here. ;-)
> Nick? Christoph?

I like less overhead of Ben's approach, and I like the slab
allocator being told about this rather than having to deduce
it from that horrible system_state thing.

OTOH, I don't know if it is useful, and is it just to work
around the problem of slab allocators unconditionally doing
the local_irq_enable? Or is it going to be more widely
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at