LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Hardware (https://www.linuxquestions.org/questions/linux-hardware-18/)
-   -   "Linux on an 8-bit micro?" (https://www.linuxquestions.org/questions/linux-hardware-18/linux-on-an-8-bit-micro-937287/)

mjolnir 03-30-2012 11:28 AM

"Linux on an 8-bit micro?"
 
http://dmitry.co/index.php?p=./04.Th...ux%20on%208bit

" This is definitely not the fastest, but I think it may be the cheapest, slowest, simplest to hand assemble, lowest part count, and lowest-end Linux PC."

Interesting read.

jefro 03-30-2012 03:12 PM

Good post but... "It takes about 2 hours to boot to bash prompt..." OUCH!

And I thought my old 486 took forever to get to bsd.

jlinkels 03-30-2012 07:56 PM

It is nice, but unusable of course.

Now there are a bunch of people who built a home-brewn CPU. PDP-11 style, with 5 or so 160x233 mm boards, running up to 4 MHz. I wonder if any of those ever run Linux. After all, those CPU's are often 16 bits, and if they followed the design of the PDP-11 area they could have a MMU.

(Rethink about what a giant step back in technology the 8086 was, both in terms of hardware and OS)

jlinkels

tiloubrunet 03-30-2012 08:55 PM

good for a secondary talent show of course ;)
impressive...
but its useless.....

salasi 03-31-2012 05:52 AM

Quote:

Originally Posted by jefro (Post 4640929)
Good post but... "It takes about 2 hours to boot to bash prompt..." OUCH!

yes but...
  • This guy seems to have taken the interesting approach of getting Linux to work via ARM emulation; this is normally the kind of thing that you would do if you had computer power to burn. He hasn't and it does show.
  • Doing sdram refresh via software eats up some throughput, but not much compared to some of the other things...but then there isn't much left
  • Accessing the sdram via software, as opposed to hardware, is also going to eat some throughput

so, it is an interesting demonstration of the fact that it can be done, but doesn't really say much about how fast (or, better, how much less slow) it would be without the decisions that slow it down. And, it doesn't really make that much sense; if something like an ARM M0 can be obtained for under $1 in volume, the advantage of using an eight bit Atmel, which isn't that much cheaper, and emulating an ARM is unclear. It still wouldn't be exactly exciting, but there would be more of a point, in terms of having actual potential uses.

jefro 03-31-2012 05:19 PM

All good points, thanks for reviewing them. I find the amount of work that went into this amazing. Near fanatical.

I can't seen any practical use for it. It is an emulated OS on a basic chip. It would be much more practical to run a native OS on it and just port any tools. A simple assembly OS seems to be a better choice but we are still back the the totally simplistic nature of the device.

I have just seen some neat projects by QNX on some pretty puny hardware and it actually functions within reason. I think of vxworks going to Mars.


All times are GMT -5. The time now is 06:19 AM.