At 7:00 PM +0100 01/10/00, David Fedor wrote:
>I believe that the MemHandleUnlock currently is marginally faster, but not
>significantly, and that isn't guaranteed to be the case in the future.
It's the other way around. MemPtrUnlock is VERY slightly faster, but not
enough to ever worry about. (The lock count is with the chunk header
itself, so if you start with the handle you have to decode and dereference
it to get to the chunk. To get to the chunk header from a pointer you just
subtract.)
--Bob
