8051

..and there are others, that have an option of an extended stack pointer, that grows above the natural IDATA space. That does need care, as some code manipulates the stack using IDATA aliasing, and such extended stacks only work via the CALL, PUSH, POP.

Some variants allow the whole data space to swap into XDATA space, which could provide some serious context switching :)

-jg

Just what I want, a pacemaker that is fairly immune to latching up.

DCE51?

This is the RTOS which was a successor and/or predecessor to DCX51 used by Intel when they built industrial control boards? Over BITBUS networks??

DCX = Distributed Control Executive...

I used this in a former life... When source code (and compilers) fit on a floppy...

Rufus

Grant

Ok - fairly - thousands of processors running for the last ten years

24/7 in railway signalling.

Kobus

I don't believe folks who say "never". In particular when they say "_never_". You can't even test for it, so how can you say that.

Now let's see who responds first with that old line "how can you run a sattelite in orbit with that" etc.

--
Thanks, Frank.
(remove 'q' and '.invalid' when replying by email)

In article , Rufus V. Smith writes

A Floppy what? Is that some sort of CD or USB stick? :-)

Sorry I should know better. I was around when floppys were 8 inches

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
/\/\/ chris@phaedsys.org      www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

In article , Frank Bemelman writes

In a VERY expensive test rig! I have seen some of what goes into testing of a satellite..... They usually start buy testing the tools!!! Down to the exact loading and propagation delays caused by a high end ICE and the loading caused by a LA.

Often testing is longer than the actual code and HW design and initial production.

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
/\/\/ chris@phaedsys.org      www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

In article , zibidi writes

This is not the case.

Then start with a more suitable MCU.... say something like an Infineon

166 Then when you have done that and you have some experience try porting it back to a 8051 or AVR

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
/\/\/ chris@phaedsys.org      www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

Actually, it was even earlier than that. The 8051 came out in 1980.

Ian

The 8051 has three as well. Two 8bit and one 16 bit pointer. Why anyone would want to do something as cumbersome as their own datastack on a small

8 bit micro is beyond me.

Ian

Now that is an interesting variant; they have actually increased the stack pointer to 10 bits so it can address 1K of memory. Pity is has CAN bus baggage. Now if someone brought out a range of variants all with a 1K stack that would be interesting. The compiler writers would have fun with it too.

Ian