Post new topic Reply to topic  [ 1 post ] 
Author Message
PostPosted: Fri Dec 30, 2022 1:39 pm 
Site Admin
User avatar

Joined: Tue Feb 03, 2015 6:57 pm
Posts: 9847
Location: Low DOS
Celebrating winter and looking forward to the next season, as we near the 2 January death day, let’s talk about Spring… and a few other OSs.

Those who deal with older equipment… we’ve seen some strange options along the way. Some of the most bizarre? How about SCSI via original IDE? Dual pipes Serial over Parallel.
Laser magnetic discs simply shouldn’t work. DD/DL CD-R?
We’ve come across many an OS as well. And that is today’s story here.

Spring.
Yep. 1987, the joy of the sunshine. AT&T/Bell Labs and Sun get together to make a merged Unix V system. Get into a fight (as usual AT&T did no work) and Sun goes totally off in a new direction. The result was Spring. A most unique operating system!

Spring attempted to change the way an operating system worked. Much like the various failed Mac system projects, they totally upended the ways an operating system works. Using the less than impressive Mach 3 kernel, spring simply wrote the operating system outside of the kernel. Using a single set of calls for direct access. And virtual connections for everything else.
In effect, the kernel was a nano-k. The direct access system was dealt through a series of pipelines. Everything else existed outside the kernel in file space or user space.

This had two major changes.
First, they are OS could run any other OS as a native program.
Second, the system could, theoretically, run software from any OS “natively” without recompile.

The upside was properly coded software that used provided transition tool kits ran lightning fast. Making today’s machines look the same or even slow.
The downside was nobody used the tool kits; not even Sun.
With their non-standard terms and convoluted implementation…
Doors, are accessible. But have no way to open. The kernel supplied different handles for each door to all running processes. If programs behaved, the system was the fastest of the time and competitive even today: 30 years later!
But most programs did not behave. And the delays of commands fighting over access to a door caused extreme bottlenecks.

So why discuss a dead system? Well, the methods were revisited by apple during later MacOS development. And is implemented today in two major systems. WSL on Windows, by Microsoft, and Parallels, for MacOS.
Neither are true virtualised methods. Nor emulation alone. Nor are they full Translation. These strange hybrid methods all owe their beginning to Spring. Amazing, it is. 30 years later and the tech finally works. Very well.

Where else did Spring show influence? The translation methodology of Windows on Arm. And Apple’s Rosetta 2.

Parallels on M owes its history to Spring. Parallels is a partial virtualisation system, a partial emulation system. But dependent on services and tools that are part of the Mac host.
So here it the ultimate in evolution. Parallels, running Windows on ARM. Parallels loads many Mac-side processes to integrate Windows. Loading Windows on ARM, uses Microsoft translation of AMD64 to ARM. But with Pears, running on windows arm, we have yet another layer.
Pears translates old apple and powerPC commands to x86. Wow translates the x86 commands into usable AMD64 code. Which is translated to ARM by Windows on ARM. Then pumped to Apple M via Parallels.
Want to get really fancy? Load Apple SOS via Pears. Then load a CP/M program.
Sauce translates Intel16 commands to MOS. The mos then to PPC. PPC to x86. X86 to AMD 64. AMD 64 to standard ARM. ARM to Apple M.
Wow. That’s 7 total conversions.

And that is only the beginning! Running the original mSH under CP/M adds a dozen more translation options. For guest side offloads of programs. With proper WAN such as a token ring setup, virtualised via software, you could in theory connect to DECbase, PDP, IBM3k and 6k. And multiple incompatible DOS systems. Via VHD files. In real time.
Add Fuse for even more fun.

With the power of M, especially Ultra, and the foundations of Spring, a modern Mac Studio could, in theory, run software from 10 architectures, across 50+ OSS, in real-time, with no delay.

Let’s just, rightfully, call Spring the most important failed experiment in computer software.
The OS has touched everything today. From virtualisation and translation, to file system, to memory management, to parallelism and piping, to shared display, to remote access, to networking, … … !!!

The Speing project lives on today. In parts. Autumn, Buzz, Plan 9. Workbench 4+. AOS. Though none are close to the original

So, today I thank AT&T for being a complete project failure, again. As that freed the SUN team to experiment. And experiment they did.

I thank Apple (Jobs), and the BSD community for never opposing trial and error.
Parallels and Paragon for their decent price for very low sales software.
The Fox group for never completely giving up on their Database language.
Arm, for thumbing their nose at everyone. Then and now.
And the engineers in Apple’s basement that brought us M. The most powerful single processor ever!
AND

I thank SUN for the crazy dgaf attitude of the time. Try anything and everything. Delete nothing.
Computing and technology would look very different today if Spring was never attempted.

_________________
42 6F 61 72 64 73 6F 72 74 2E 63 6F 6D


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 1 post ] 

All times are UTC - 5 hours [ DST ]


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  

cron