Re: Ravi bug related to setting of L->top in OP_RETURN

[云风 Cloud Wu <cloudwu@...>]

> Is it possible to consider an implementation where L->top is strictly
> the top of the stack, and not used for a dual purpose? Or am I missing
> a point here?

Can you suggest how to do such implementation?

Maybe we can offer two api:

luaD_posCcall and luaD_posLcall . 

Only in case OP_RETURN, we should call luaD_posLcall , and

     ci = L->ci;

     if (b) L->top = ci->top;

can be done in luaD_posLcall . So that the L->top can be strictly the top of the stack.