zippster278 at gmail.com
Mon Mar 27 13:18:34 EDT 2017
Speaking of B & B, another good reference on the 6309…
> On Mar 27, 2017, at 11:59 AM, L. Curtis Boyle <curtisboyle at sasktel.net> wrote:
> Correct. The first version of Burke & Burkes Powerboost used emulation mode, with the extra instructions.
> Sent from my iPhone
>> On Mar 27, 2017, at 10:43 AM, Arthur Flexser <flexser at fiu.edu> wrote:
>> Bill, I think the error trapping that you're describing is what happens in
>> emulation (or native) mode when an illegal opcode (i.e., illegal even when
>> 6309 added opcodes are included) is encountered. According to others, a
>> legal 6309 instruction is executed even in emulation mode.
>>> On Mon, Mar 27, 2017 at 12:30 PM, Bill Nobel <b_nobel at hotmail.com> wrote:
>>> What actually happens, is the 6309 will see that it is emulation mode and
>>> cause a error trap. The CPU will grab the address stored @ $FFF0 & $FFF1
>>> and jump to it. This was how the new instructions were discovered. Users
>>> of the 6309 thought it was a flaky 6809 clone because it would randomly
>>> crash on a error.
>>> Bill Nobel
>>> b_nobel at hotmail.com<mailto:b_nobel at hotmail.com>
>>> On Mar 27, 2017, at 10:04 AM, Arthur Flexser <flexser at fiu.edu<mailto:
>>> flexser at fiu.edu>> wrote:
>>> BTW, what happens if you're in emulation (non-native) mode and a
>>> 6309-specific instruction is encountered. Is it executed, error-trapped,
>>> or what?
>>> On Mon, Mar 27, 2017 at 10:54 AM, L. Curtis Boyle <curtisboyle at sasktel.net
>>> <mailto:curtisboyle at sasktel.net>>
>>> Just to clarify - you see around a 10% speed increase just by turning
>>> native mode on. You can see a much bigger increase for code written
>>> specifically for the new registers, new instructions, block memory moves,
>>> etc. (like the 6309 version of NitrOS-9).
>>> L. Curtis Boyle
>>> curtisboyle at sasktel.net<mailto:curtisboyle at sasktel.net>
> Coco mailing list
> Coco at maltedmedia.com
More information about the Coco