Re: [cpuops cmpxchg V2 5/5] cpuops: Use cmpxchg for xchg to avoidlock semantics

[Tejun Heo]

On 12/14/2010 05:28 PM, Christoph Lameter wrote:
> Use cmpxchg instead of xchg to realize this_cpu_xchg.
> 
> xchg will cause LOCK overhead since LOCK is always implied but cmpxchg
> will not.
> 
> Baselines:
> 
> xchg()		= 18 cycles (no segment prefix, LOCK semantics)
> __this_cpu_xchg = 1 cycle
> 
> (simulated using this_cpu_read/write, two prefixes. Looks like the
> cpu can use loop optimization to get rid of most of the overhead)
> 
> Cycles before:
> 
> this_cpu_xchg	 = 37 cycles (segment prefix and LOCK (implied by xchg))
> 
> After:
> 
> this_cpu_xchg	= 11 cycle (using cmpxchg without lock semantics)
> 
> Signed-off-by: Christoph Lameter <cl@xxxxxxxxx>

It's not a bad idea to keep this patch separate from the original one
but as both are not applied yet, it probably is better to put this
right after the original addition if you end up re-posting the series;
otherwise, I'll just reorder it when I apply.

Thanks.

-- 
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/