Hello, as some of you might remember, I'm writing a 6502 Virtual Machine. I was thinking about accelerating the most critical routines of the C runtime. The idea is NOT to extend the CPU with custom opcodes, but to use the IO chip. Something like this: runtime_operation: STA CC65OP1 ; operand 1, if necessary STX CC65OP2 ; operand 2, if necessary LDX #CC65_??? STX CC65OPC ; operation code; triggers operation LDA CC65RES ; operation result, if necessary RTS I have no access to the CPU registers, but I have access to the memory and I can assume a specific layout for page 0.I cannot bloat the code of the IO chip too much; ideally I'm looking for 5-10 operations to accelerate. Does it make sense or do you think that the the most critical operations are so short that the overhead would be too big anyway? Thanks, Emanuele ---------------------------------------------------------------------- To unsubscribe from the list send mail to majordomo@musoftware.de with the string "unsubscribe cc65" in the body(!) of the mail.Received on Thu Sep 13 09:36:40 2007
This archive was generated by hypermail 2.1.8 : 2007-09-13 09:36:44 CEST