Path: utzoo!attcan!uunet!samsung!dali.cs.montana.edu!uakari.primate.wisc.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!hplabs!hpfcso!aoe
From: aoe@hpfcso.HP.COM (Alexander Elkins)
Newsgroups: comp.sys.amiga
Subject: Re: UNIX sys V4.0
Message-ID: <9010025@hpfcso.HP.COM>
Date: 24 Oct 90 19:45:08 GMT
References: <6956@uwm.edu>
Organization: Hewlett-Packard, Fort Collins, CO, USA
Lines: 65

ag@cbmvax.commodore.com (Keith Gabryelski) writes:
>The port is out.  Ask questions about it and you will get answers.

A few more come to mind -

1) How many processes does a "ps -ef" show running immediately after entering
init state 1 after rebooting?   I'm interested in running unix at home, and
I'm curious how much overhead there is without any network processes running.

2) How much total real memory is used up by those processes while in init
state 1 (single user mode BTW)?  In other words, is 16Mbytes sufficient
to avoid swapping most of the time?  Less wear and tear on the disk, faster too!

3) How much overhead is involved when making a system call?  Like save
the registers (a0-a7,d0-d7,f0-f7), context switching, etc.  I.e how efficient
is it? (Let's see, number of getpid() calls per second? Not that one would do
that, but it should be mostly overhead time.)

4) What is the typical size of the swap area on the disk?  10Mb, 20Mb, ?

5) Does the X Windows server process go through all the same socket software
layers to receive packets from local client processes that it does for
networked (remote) client processes?  Bypassing can improve performance!

6) Does the X Windows server process allow concurrent access to the display
hardware (chipmemory, blitter, etc) via a resource sharing/locking method?

7) Are shared memory segments supported?  I.e. can two, or more processes
attach the same shared memory segment and read/write to it?

8) Are virtual files supported?  I.e. can a file be opened in such a way as
to map its contents to an address space in a running process? (Great for
data bases, virtual memory is then just a special unnamed file!)

9) Does a system call to brk(1024000) followed by brk(5120) actually return the
memory to the system for use by other processes or does the memory used
by a process only increase and not get return to the system resource memory
pool until it exits?  Which happens?

10) What version of gcc comes with the system?  Gcc-1.37.1 would be fairly
recent.

11) Was all the code compiled using gcc?

12) Is the kernel user configurable?  Is this done as a compile/link/reboot
process or does the kernel dynamically configure itself by loading/unloading
whatever drivers it needs?

13) Are ACL's (access control lists) supported?  These allow more finely
tuned control over file access permission.

14) Is file locking supported?  This means whole files and blocks in a file,
which is very useful for concurrent database access.

15) Is ksh (korn shell) included with the system?

16) Is process suspend/resume supported?  (Usually control-Z/bg/fg in ksh)

17) Is process tracing supported by the kernel?  I.e. a ptrace(request,pid,...)
call from another process can suspend/control/modify/resume another process.
Usually used by debugger programs.

Looking forward to running this on my A3000/25-100 at home :-)

Alexander Elkins (aoe@hpfiaoe.HP.COM)