Re: Context switch times

[Richard Gooch]

David S. Miller writes:
>    From: Richard Gooch <[e-mail]>
>    Date: Thu, 4 Oct 2001 15:39:05 -0600
> 
>    David S. Miller writes:
>    > lat_ctx doesn't execute any FPU ops.  So at worst this happens once
>    > on GLIBC program startup, but then never again.
>    
>    Has something changed? Last I looked, the whole lmbench timing harness
>    was based on using the FPU.
> 
> Oops, that's entirely possible...
> 
> But things are usually layed out like this:
> 
> 	capture_start_time();
> 	context_switch_N_times();
> 	capture_end_time();
> 
> So the FPU hit is only before/after the runs, not during each and
> every iteration.

Hm. Perhaps when I did my tests (where I noticed a penalty), we didn't
have lazy FPU saving. Now we disable the FPU, and restore state when
we trap, right?

I do note this comment in arch/i386/kernel/process.c:
 * We fsave/fwait so that an exception goes off at the right time
 * (as a call from the fsave or fwait in effect) rather than to
 * the wrong process. Lazy FP saving no longer makes any sense
 * with modern CPU's, and this simplifies a lot of things (SMP
 * and UP become the same).

So what exactly is the difference between our "delayed FPU restore
upon trap" (which I think of as lazy FPU saving), and the "lazy FP"
saving in the comments?

				Regards,

					Richard....
Permanent: [e-mail]
Current:   [e-mail]