Hi,
On Fri, 15 Feb 2002, Jeff Garzik wrote:
> > As I mentioned before I more like the byte approach, since atomic bit
> > field handling is quite expensive on most architectures, where a simple
> > set/clear byte is only one or two instructions, if there is byte
> > load/store instruction. So I'd really like to see to leave the decision to
> > the architecture, whether to use bit or byte fields.
>
> We have tons of code already using atomic test_and_set_bit type
> stuff... why not just make sure your bit set/clear stuff is fast? :)
Because in this case there is no atomic test_and_(clear|set)_bit needed.
We only need to clear the bit/byte before scheduling/signal delivery is
started.
bye, Roman
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Feb 15 2002 - 21:01:07 EST