Re: [PATCH] only irq-safe atomic ops

[yodaiken@fsmlabs.com]

On Sat, Feb 23, 2002 at 01:20:06PM -0500, Robert Love wrote:
> On Sat, 2002-02-23 at 06:38, yodaiken@fsmlabs.com wrote:
> 
> > So without preemption in the kernel
> > 	maybe 4 instructions: calculate cpuid, inc; all local no cache ping
> > 	code is easy to read and understand.
> > 
> > with preemption in the kernel
> > 	a design problem. a slippery synchronization issue that 
> > 	involves the characteristic preemption error - code that works
> > 	most of the time.
> 
> Or not.  The topic of this thread was a micro-optimization.  If we treat
> the variable as anything normal requiring synchronization under SMP, any
> of the standard solutions (atomic ops, etc.) work.  If we want to get

And cause cache ping, possible contention, ...

> fancy, we can disable preemption, use my atomic_irq ops, or just not
> care.

Right. Without preemption it is safe to do
	c = smp_get_cpuid();
       ...
        x = ++local_cache[c]
       ..

       y = ++different_local_cache[c];
      ..

With preemption this turns into a problem that is easier to solve
with a lock or by not having per-cpu caches in the first place -- 
eg by writing worse code. After all, those are just micro-optimizations
and a few percent here, a few percent there, nobody will notice it. -(



-- 
---------------------------------------------------------
Victor Yodaiken 
Finite State Machine Labs: The RTLinux Company.
 www.fsmlabs.com  www.rtlinux.com