My personal feeling is that processes wil have to be "objectified" in
the same way that files are "objects". So just like we have different
file system types, we will have different process types. I can imagine
3 right off the bat:
- local processes (this is what we have now)
- remote processes (points at a remote process someplace else)
- debugged / traced processes
The last one is a bit weird because you would like it to "stack" on top
of whatever the underlying process happens to be.
Read the Locus papers for more info, they did much of this at one point.
At any rate, once you have this, you have an object that you can request
to save its state and move. You'll still need to do some work for open
files (pipes/sockets/files/devices).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu