RE: [PATCH] changing IV to opcode_t!!

[Hong Zhang]

> Nope. There's *always* an integer type that works right. We won't be 
> running on a platform that doesn't have 32-bit integers. (Well, not
without 
> someone hacking the heck out of the core)

We should not rule out 16-bit system completely. It should be possible an
16-bit runtime can easily convert 32-bit opcode stream into 16-bit one.
opcode is very unlikely > 65536. And large parameter can be easily moved
into constant pool, and replace by constant pool index. If the constant
pool is bigger than 64K, the runtime will run out of memory anyway, period.

Anyway, we should not make 16-bit system unnecesarily difficult to do.

Hong