Re: local_add_return

[Heiko Carstens <heiko.carstens@de.ibm.com>]

On Wed, Dec 17, 2008 at 09:08:04AM +1030, Rusty Russell wrote:
> On Tuesday 16 December 2008 17:43:14 David Miller wrote:
> > Here ya go:
> 
> Very interesting.  There's a little noise there (that first local_inc of 243
> is wrong), but the picture is clear: trivalue is the best implementation for
> sparc64.
> 
> Note: trivalue uses 3 values, so instead of hitting random values across 8MB
> it's across 24MB, and despite the resulting cache damage it's 15% faster.  The
> cpu_local_inc test is a single value, so no cache effects: it shows trivalue
> to be 3 to 3.5 times faster in the cache-hot case.
> 
> This sucks, because it really does mean that there's no one-size-fits-all
> implementation of local_t.  There's also no platform yet where atomic_long_t
> is the right choice; and that's the default!
> 
> Any chance of an IA64 or s390 run?  You can normalize if you like, since
> it's only to compare the different approaches.

atomic_long_t seems to be the right choice on s390. IRQ disable/enable is
expensive, but the compare and swap instruction is cheap. I just gave it
a quick shot, but please note that there were two hypervisors running below
my system (add_return is missing since I used your first patch):

atomic_long	19  18  -  3  17
irqsave/rest	57  58  - 39  22
trivalue	43  43  -  4  45
local_t		18  20  -  2  16