Re: [patch 2.6.13-rc3a] i386: inline restore_fpu

[Andi Kleen]

Andrew Morton <akpm@xxxxxxxx> writes:
> 
> We do have the `used_math' optimisation in there which attempts to avoid
> doing the FP save/restore if the app isn't actually using math.  But
> <ancient recollections> there's code in glibc startup which always does a
> bit of float, so that optimisation is always defeated.  There was some
> discussion about periodically setting tasks back into !used_math state to
> try to restore the optimisation for tasks which only do a little bit of FP,
> but nothing actually got done.

Actually we reset the flag on every context switch, so that works just fine.

But I was considering to do it less often so that we switch the FP 
state non lazily for FP intensive processes and avoid the overhead
of all these exceptions.

-Andi

P.S.: Original profile data looks a bit fishy. Normally avoiding a single
function call should not make tht much difference unless you call
it in a inner loop, but that is not the case here.
-
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/