> My own feeling is that this is not so useful as it might appear
> at first glance. If you _really_ want to try something interesting
> why not write a gcc back end that uses a locked L1 line as a nice
> big register file and see if you can push the x86 architecture to
> new heights?
Hmm. That would probably be the best way of using it.
The other proposed usages don't seem to make any difference,
since often-used things tend to stay cached anyway and not
often used things don't really need to be cached...
(Except of course in low-latency things like interrupt
handling. It might be nice to trim some cycles off of
the interrupt handler. Maybe then we'll finally be faster
than QNX :)
Rik.
+-------------------------------------------+--------------------------+
| Linux: - LinuxHQ MM-patches page | Scouting webmaster |
| - kswapd ask-him & complain-to guy | Vries cubscout leader |
| http://www.phys.uu.nl/~riel/ | <H.H.vanRiel@phys.uu.nl> |
+-------------------------------------------+--------------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu