Paul Rubin <http://[email protected]> wrote:
>Those links describe using the LOCK prefix, which as the name implies,
>asserts a lock, so it is no longer "lockless reference counting".
No, it doesn't assert a lock in the sense used in this thread. On modern
Intel systems it's normally handled completely within the cache.
>The LOCK prefix adds about 100 cycles to the instruction.
That's unavoidable. It's still faster than spin lock, which would also
have to use syncronizing instructions.
Ross Ridge
--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] [email protected]
-()-/()/ http://www.csclub.uwaterloo.ca/~rridge/
db //
--
http://mail.python.org/mailman/listinfo/python-list
