[OpenBIOS] [commit] r915 - trunk/openbios-devel/arch/sparc32

[Mark Cave-Ayland]

Artyom Tarasenko wrote:

>> Mainly by looking at the Forth; it seems that all the code snippets I've
>> seen expect the stack to be preserved. Otherwise if we always push all 5
>> arguments, we'll end up with an excess of 0s being built-up on the stack by
>> every call.
> 
> I agree that the stack should be preserved, but why not just do it
> literary by throwing away unused arguments?
> In the current implementation there is no way to pass 0 as the last
> argument. Do you think it's ok?

I think so - it's difficult to know. That was the other option of 
course, which is to push everything onto the stack and then clear the 
stack after the call has finished. But this comes down to does the 
client program expect the stack to be preserved across subsequent calls 
to obp_fortheval_v2? The current approach does this, although we don't 
know whether it's a specific requirement or not.


ATB,

Mark.

-- 
Mark Cave-Ayland - Senior Technical Architect
PostgreSQL - PostGIS
Sirius Corporation plc - control through freedom
http://www.siriusit.co.uk
t: +44 870 608 0063

Sirius Labs: http://www.siriusit.co.uk/labs