Finally the 2.2.0 is out of sight :-).

Marcin Dalecki (
Wed, 27 Jan 1999 01:21:31 +0100


Since the "stable" developement kernel seems to be already out
of sight maybe I would like to suggest some developement which may
be of real interrest.

For me personally it appears that the linux kernel is currently
already reaching an complexity, where some serious tought should be put
into the developement tools at use. There seems already to be
some support for kernel profiling and live debuggin. And second there
seems to be already an quite well done abstraction between the
architecture dependant parts and the non dependant. Third on some
archs there is already a way to run the kernel as a kind of user
process inside of a VM.

Maybe all of them could be put together, by creating an "architecure"
port which would be in fact itself compleatly independant from the

What I'm thinking about is porting the kernel to run
as a *compleatly* *normal* user level process. Something along those
lines had been already imeplemnted by prof. Switzer at Univeristy of
What he had done was in fact a pseudo micro kernel implemented
as a UNIX programm running on a file containing his file system.
It was called TUNIX. (ftp to will
show where to find it.)

Obviously this wasn't very usefull for any other purposes then
(Even those it served in a really bad manner, due to the numerous bugs
in it
and due to many many inaccuracies in the accomplying book.)
However turning a *real* kernel in an compleatly arch independant way,
by doing *just* another port into something like this,
would be undoubtly of real value.

Just imagine....

1. Detecting dead code by using plain nice well known user land coverage
(hint: egcs)

2. No hardware crashing (and therefore havoc with it) during testing of
for example in the VM/swap and therlike stuff.

3. No more Andrea asking about "interactive feelings" on this list...

4. Just using nice fine gprof for profiling instead of the ugly in
PC grabbers...

5. No more loops decryption just call: gdb vmlinux core and whatch it
in from of yours eyes (To be short: post mortem debuggin...)

6. Writing out of nice coloured GIF's showing the memmory
fragmentation/allocation/aging state of the memmory in use thus finally
killing any guesses of allocator performance...


And anybody could play with this, not only those who can afford a
processor/system capable of virtual machines....

Any comments?
Would this be really appreciated?
Should one start to toy with imeplementing it?


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at