|
|
Re: MMD and VTABLE_find_method: msg#00361
|
Subject: |
Re: MMD and VTABLE_find_method |
Sam Ruby wrote:
I took your example code, almost word for word, and added approximately
three, count them three, machine instructions (not Parrot instructions,
not C statements, but machine instructions) to the path length, and now
adjectives like "horribly slow and wierd" come out.
It looks like this mailing list will not be suitable for civilized
discussion until Dan gets back.
I've described the functionality of the add *opcode*. You have moved
that piece of code to the pmc level (where the implementation of add for
one type pair ought to be and currently is) and then your code
redispatches to another routine that then actually performs the task.
This is not only absolutely not needed it also disallows an efficient
implementation.
The horribly slow can be measured easily and I consider this kind of
double dispatching as being weird, sorry.
Meanwhile, all existing Python tests pass:
Great.
Merry Christmas.
- Sam Ruby
Merry Christmas too,
leo
|
| |