Scott Long <[EMAIL PROTECTED]> writes:
> This call-then-pop to get EIP makes me grimace.
Note that call-then-pop is a major performance penalty on many modern
processors (because call is not paried by ret). A local-call-move-ret
scheme is a bit more efficient.
> It seems like all shared libraries and shared objects on Intel take
> a serious penalty every time they access non-auto storage.
s/non-auto storage/global variables/
Global variables are a pain in the neck anyway, so this is just
another serious penalty. I wouldn't care about that too much.
