This reminds me of some Tops-20 code that does vaguely analogous on a
power fail.? It flushes a lot of stuff to files and the page store,
waits for the disks to settle down and then signals the 11.
On a re-power, it comes back up near to where it was and (supposedly)
continues.? However, not everything is read back to memory.? I would
imagine that the code has to date back to TENEX days and core memory,
where the image would persist.
I have wanted to experiment with it by having KLH-10 catch a SIGPWR and
appropriately signaling and then saving the memory. The functionality
could be useful for a KLH10 on a laptop that was put to sleep.
On 11/12/21 1:06 PM, Paul Koning wrote:
On Nov 12, 2021, at 10:57 AM, Johnny Billquist
<bqt at softjar.se> wrote:
On 2021-11-12 15:24, Paul Koning wrote:
On Nov
11, 2021, at 6:22 PM, Johnny Billquist <bqt at softjar.se> wrote:
I must admit that I hadn't considered the possibility of just saving the core. Which
of course can accomplish the same thing in a neat way.
That's a bit like how
RSX-11/D and IAS boot -- by reloading the image of memory when you issued the SAV command.
Pretty clever: you set things up the way you want them to be, and then you make that
state persistent.
Same with -11M. But it's only used for the kernel.
However, in TOPS-20 as well as some other DEC OSes, this is a common pattern for all
programs.
In the sense that an executable program is basically a memory image?
Yes, RT-11 SAV files are a good example.
But the RSX-11/D and IAS case is a whole-system image, not kernel or a single program.
You boot the OS, load all the applications you want to be active in their various
partitions, and type SAV. When the system boots, that whole memory image is loaded, and
made active by delivering power-fail ASTs to each task. For example, SAV also handles the
boot-time setup, for example prompting for the current date and time. (Or perhaps it
doesn't do that in the stock version; I remember adding it to SAV in 1978, as part of
Typeset-11 work.)
paul