Linux-Development-Sys Digest #193, Volume #7 Tue, 14 Sep 99 09:14:18 EDT
Contents:
Re: Flamage - Why? (Donal K. Fellows)
Re: unix98 pty's problems (Remco van den Berg)
Re: unix98 pty's problems (Remco van den Berg)
Re: Flamage - Why? (Joachim Feise)
Re: Flamage - Why? (Donal K. Fellows)
Re: Linux standards compliance (Horst von Brand)
Re: 497.2 days ought to be enough for everybody (Malcolm Beattie)
Re: call function of device driver (Mark McDougall)
Re: Turing Machines (John Forkosh)
Re: Programming a voodoo banshee card (Lee Reynolds)
Re: survey linux project. ("George P. Staplin")
Re: unix98 pty's problems (Mike Dowling)
Re: klogd crashes after dialing isp (Dr. Peer Griebel)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Donal K. Fellows)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: Flamage - Why?
Date: 14 Sep 1999 08:42:31 GMT
In article <[EMAIL PROTECTED]>,
Keith Wright <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] (Christopher Browne) writes:
>> The use of a "sandbox" system is not particularly consistent with an
>> approach that would be evocative of the Incompleteness Theorem.
>>
>> Algorithm:
>> - You run the program in the "sandbox."
>>
>> - If it is "unsafe," it will perform recognizable combinations of
>> operations to indicate its "unsafeness."
>>
>> For instance, trying to access I/O or memory that it should not.
>>
>> - If it *is* safe, it will "behave nicely," not trying to access
>> anything it should not touch.
>
> Sounds like a protected mode OS. What's all the bruhaha about?
Think "protected mode OS on lots of steroids" and you'll be close.
> But if you assume that because it didn't segfault the first
> N times you ran it, it is now safe to run it with root privledge
> and all memory protection turned off, well you lose.
The only time it would be safe to attempt the above would be if you
have *proved* that nothing will go wrong. Note that running the
program a few times (like a billion for example) proves nothing
outside the exact domain of the test runs. Instead, you need to look
at denotational semantics of programs...
Proof is very expensive. It tends to get reserved for critical
systems like railway signalling software, nuclear power stations[*],
and hardware (I've got a set of slides on my desk that describe the
correctness proof for division and square root on the IA-64
architecture.)
Donal (speaking with my Professional Opinion hat on.)
[* Wheeling out the old warhorse yet again... ]
--
Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ [EMAIL PROTECTED]
-- The small advantage of not having California being part of my country would
be overweighed by having California as a heavily-armed rabid weasel on our
borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>
------------------------------
From: [EMAIL PROTECTED] (Remco van den Berg)
Subject: Re: unix98 pty's problems
Date: 14 Sep 1999 07:51:23 GMT
On Tue, 14 Sep 1999 00:23:32 GMT, Juergen Heinzl wrote:
>>
>>bash-2.02$ ll -d /dev/pts /dev/ptmx
>>crw-rw-rw- 1 root root 5, 2 Sep 14 00:16 /dev/ptmx
>
>You might change the group to tty, although this is not the problem.
OK did that....
>>Do I have to recompile my xterminals and telnet daemon?
>
>Yes.
>
>>I didn't read that anywhere. Are there more things I have to change?
>>Can I remove old ttyp devices from the /dev directory?
>
>Yes and no, but you might not, although here are all gone. Currently
>there is no telnetd version (no non vendor specific one at least)
>that compiles out of the box with Unix98 style PTY support with all
>environments.
>
>The real new one does not compile at all with glibc-2.1.2 ... it would
>not even configure (in need to be looked at).
I tried the telnetd from netkit-telnet-0.12 (the client I couldn't
compile with glibc-2.1.2 and egcs-2.91.60) but also that daemon still
used the old ttyp devices.
>You can find a worked_for_me_and_some_others patches here ...
>http://www.monocerus.demon.co.uk/httpd-server/juergen/code/c/unix/patches/
>... but these are inofficial at best.
>For xterm and rxvt ... the latest version support Unix98 PTY's and
>the same goes for the pppd daemon.
Thanks for this info. I probably wouldn't have found it otherwise.
>Now you might have still other stuff around. In short if this is a stand
>alone machine @ home you can give it a try, but avoid being in a hurry
>to remove those obsolete device files.
Can you tell me which devices it actually are? There are masters and
slaves as far as I know, so how are they called?
>As long as they are not used they cannot do much more harm than
>eating up some inodes on your disk.
I know, but I choosed for unix98 pty's and than I also go for it :-)
- Remco van den Berg
============================================================================
Philips Semiconductors B.V. tel:(+31 40 27)22031 fax:22764 Room: BE-345
mailto:[EMAIL PROTECTED] seri: rvdberg@nlsce1
home Email: [EMAIL PROTECTED] (non Philips related) ICQ: 47514668
============================================================================
Microsoft and Lotus Notes free. Don't send me any Microsoft attachments.
============================================================================
------------------------------
From: [EMAIL PROTECTED] (Remco van den Berg)
Subject: Re: unix98 pty's problems
Date: 14 Sep 1999 08:32:20 GMT
On Tue, 14 Sep 1999 00:23:32 GMT, Juergen Heinzl wrote:
>You can find a worked_for_me_and_some_others patches here ...
>http://www.monocerus.demon.co.uk/httpd-server/juergen/code/c/unix/patches/
>... but these are inofficial at best.
Just tried the telnet patch.
The telnet daemon is now working fine. But I'm having problems compiling
the telnet client:
gcc -s commands.o main.o network.o ring.o sys_bsd.o telnet.o terminal.o \
tn3270.o utilities.o genget.o environ.o netlink.o -lncurses -o telnet
telnet.o: In function `my_setupterm(char const *, int, int *)':
telnet.o(.text+0xf39): undefined reference to `tgetent(char *, char const *)'
collect2: ld returned 1 exit status
I think I don't have a tgetent function in my ncurses.4.2 library.
(There's no reference in the ncurses include file, so....)
I tried to also link the termcap library in, but that also didn't work.
Does somebody see what could be wrong?
- Remco van den Berg
============================================================================
Philips Semiconductors B.V. tel:(+31 40 27)22031 fax:22764 Room: BE-345
mailto:[EMAIL PROTECTED] seri: rvdberg@nlsce1
home Email: [EMAIL PROTECTED] (non Philips related) ICQ: 47514668
============================================================================
Microsoft and Lotus Notes free. Don't send me any Microsoft attachments.
============================================================================
------------------------------
From: Joachim Feise <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: Flamage - Why?
Date: Mon, 13 Sep 1999 19:15:20 -0700
Reply-To: [EMAIL PROTECTED]
Stephen Harris wrote:
[snip]
> Has anybody in this thread actually got a grounding in CS theory? The word
I didn't watch this thread, but I am going to jump in here quickly.
> "impossible" actually has a specific meaning there, which has nothing to do
> with the complexity of the algorithm. The halting problem is "impossible".
Actually, the halting problem is "undecidable".
To throw in a couple terms here:
Computability: is there an answer?
Decidability: is there a y/n answer?
Given that the halting problem is defined as "Given a description of a Turing
machine M and its input X, will M halt on X?", the halting problem is a
decidability problem.
[snip]
> Your usage of the term "impossible" above is incorrect. The word better
> used would be "impractical". A problem previous impractical may now be
> practical due to increased computing power (storage, RAM, MIPS etc etc).
> Further, newer optimised algorithms may have been found that reduces the
> complexity of the problem, or at least the magnitude!
The mathematical terms that apply here are "solvable in polynomial time"
(appreviated as P) and "solvable in nondeterministic polynomial time" (NP).
Since solutions to NP problems require exponential time on deterministic
machines (like Turing machines), it is considered impractical to try to solve
these in the general case. Increased computing power can push the value x for
which a solution is found to a higher value, but with all computing power in
the world it is not possible to solve the problem for the general case.
Now there are approaches to nondeterministic computing, like the DNA
computing that Adleman proposes
(see http://www.sciam.com/1998/0898issue/0898quicksummary.html),
but these are still in their infancy.
> A problem is only defined as "impossible" if it can be proven that no
> algorithm _can_ exist to solve the problem (mathematical proof of a
> contradiction is generally the way to demonstrate this). Also note that
Given your definition of "impossible" (which I agree with), the halting
problem is not impossible. You would need to prove that there is at least
a program that does not terminate. Since you can not do that, the halting
problem is undecidable.
-Joe
--
===================================================================
Joachim Feise Ph.D. Student, Information & Computer Science
mailto:[EMAIL PROTECTED] http://www.ics.uci.edu/~jfeise/
===================================================================
------------------------------
From: [EMAIL PROTECTED] (Donal K. Fellows)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: Flamage - Why?
Date: 14 Sep 1999 08:24:50 GMT
In article <SCgD3.10$[EMAIL PROTECTED]>,
Christopher Browne <[EMAIL PROTECTED]> wrote:
> The theorem might (arguably, with much handwaving) establish that a
> theoretically perfectly safe sandbox may be impossible; thus far,
> there has been a vast dearth of suggestions as to what the features of
> *any* sort of sandbox might be.
A perfectly safe sandbox is easy to create. Simply forbid all access
to resources outside the sandbox and impose strict limits on the
amount of resources that can exist inside the sandbox as well (to
prevent certain kinds of DoS attack.) The trick lies in working out
what controlled access outside the sandbox you want to permit - have a
look at the security policies (IOW sandbox configurations) made
available by the Tcl plugin to understand more about this topic...
http://www.Scriptics.com/resource/software/tools/plugin/
None of this has a right lot to do with Herr G�del, I'll admit.
Donal.
--
Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ [EMAIL PROTECTED]
-- The small advantage of not having California being part of my country would
be overweighed by having California as a heavily-armed rabid weasel on our
borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>
------------------------------
From: [EMAIL PROTECTED] (Horst von Brand)
Subject: Re: Linux standards compliance
Date: 14 Sep 1999 10:27:00 GMT
On Mon, 06 Sep 1999 21:43:48 GMT,
[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
[...]
>Easy. Microsoft acknowledges that one of the main causes for
>NT4 instability are bad drivers (and that they intend to fix that
>problem for W2k by better testing of third party drivers). NT4
>generally uses vendor supplied binary only drivers. This means the
>drivers writen by these professionals are often low quality.
Just too bad that Win2k attracts brownouts and thunderstorms, so it crashes
just the same without 3rd party drivers ;-)
--
Horst von Brand [EMAIL PROTECTED]
Casilla 9G, Vi�a del Mar, Chile +56 32 672616
------------------------------
From: [EMAIL PROTECTED] (Malcolm Beattie)
Subject: Re: 497.2 days ought to be enough for everybody
Date: 14 Sep 1999 10:01:28 GMT
In article <[EMAIL PROTECTED]>,
David Schwartz <[EMAIL PROTECTED]> wrote:
>
> There were numerous jiffy wrap bugs. It's possible that when the wrap
>occured, you just weren't near any of the problem code.
>
> DS
>
>bill davidsen wrote:
>>
>> In article <[EMAIL PROTECTED]>,
>> David Schwartz <[EMAIL PROTECTED]> wrote:
>>
>> | Had they thought about having jiffies wrap, they could have ensured
>> | that the system is at least rebooted once every 497.2 days. It just
>> | shows how the Linux people have more ideas than the NT people.
>>
>> Is this a new problem introduced in the 2.x series? I have a DNS
>> server sitting in a corner on a UPS, running 1.2.13, and I swear it's
>> been up since 1.2.13 came out. You can tell how often I look at it, DNS
>> works so the system is up ;-)
>>
>> There's no i/o going on except network, and very little of that, if
>> drivers are the only issue the chance of a hit is small.
Another data point: the Oxford University web server I run here
(Pentium 133, 64 MB RAM, 16 GB disk, 1.5-2 million hits per week, 90
virtual hosts for departments/colleges, 4000 active users with web
pages) reached its 497 days uptime two weeks ago. I'd scheduled an
upgrade to a new system for the Wednesday morning (the jiffy rollover
was that afternoon). The upgrade/switchover was successful. In the
afternoon, I tried to telnet to the old box and couldn't. I could
ping it though. On the console, trying to log in generated a kernel
oops just before the shell prompt would have appeared. I left it
alone. The following morning, I could both log in locally and telnet
to it fine. There were a large number of kernel oops in the log, all
"divide by zero" ones in do_fast_get_timeoffset (or some such). Apart
from the wraparound of "uptime" and the weird start times displayed
by ps (they'd actually been weird for a long time before), the system
was fine again.
--Malcolm
--
Malcolm Beattie <[EMAIL PROTECTED]>
Oxford University Computing Services
"I permitted that as a demonstration of futility" --Grey Roger
------------------------------
From: Mark McDougall <[EMAIL PROTECTED]>
Subject: Re: call function of device driver
Date: Tue, 14 Sep 1999 18:20:32 +1000
Keith Wright wrote:
> No. You write functions to open, read, write, close, etc. and the kernel
> calls them at the proper time when somebody says:
> ff=fopen("/dev/ReedLai","w");
> fprintf(ff,"Hello Reed\n");
> Sorry, no can do.
Correct, but not very helpful!
Perhaps you can access your desired functionality via the ioctl() call?
The ioctl interface allows you to specify 'internal' command codes that
may be sent to your driver and interpreted as you like. Note there are
times/functions for which this is appropriate, and other for which it is
not.
Regards,
--
| Mark McDougall |
| Engineer |
| Virtual Logic Pty Ltd |
| http://www.vl.com.au |
------------------------------
From: [EMAIL PROTECTED] (John Forkosh)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: Turing Machines
Date: 14 Sep 1999 06:21:45 -0400
Christopher B. Browne ([EMAIL PROTECTED]) wrote:
: On 13 Sep 1999 21:13:44 -0400, Keith Wright <[EMAIL PROTECTED]> posted:
: >[EMAIL PROTECTED] (Christopher Browne) writes:
: >> Are there systems proven to be conceptually more powerful than Turing
: >> machines?
: >Yes, as a mathematical theory. You just assume you have a black
: >box (called an oracle) that answers your question. Then you prove
: >that some other functions are still not computable. The result
: >is called the "theory of degrees of unsolvability".
: Yes, I guess one might assume the availability of "oracles."
: I was thinking, of course, about *constructed* systems, or of
: systems that are at least conceptually constructable.
In that case Church's Thesis comes into play, and asserts that
nothing is more powerful than Turing machines. See, e.g.,
Theory of Recursive Functions and Effective Computability,
Hartley Rogers, Jr., MIT Press 1987 (originally McGraw-Hill 1967)
page 20 and following
Also see
Classical Recursion Theory,
P. Odifreddi, North-Holland 1989,1992
pages 101-123
All common models of computation (Turing machines, automata, lambda
calculus, etc) provably lead to the same class of computable functions.
Church's Thesis is the further (unproven) notion that this common
class captures _all_ the functions we intuitively mean by the word
"computable". No suggestion to the contrary has yet proved fruitful.
John ([EMAIL PROTECTED])
------------------------------
From: Lee Reynolds <[EMAIL PROTECTED]>
Subject: Re: Programming a voodoo banshee card
Date: Tue, 14 Sep 1999 03:46:18 -0700
Alexander Kraut wrote:
> Does somebody have some information about programming a graphic card with
> voodoo banshee chipset directly. (Writing a graphic driver?)
>
> Thank you
I've got a banshee and svgalib works for me.
Lee
------------------------------
Date: Tue, 14 Sep 1999 04:21:44 +0000
From: "George P. Staplin" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.apps,linux.dev.gcc,linux.dev.kernel,linux.dev.x11
Subject: Re: survey linux project.
"Randall J. Parr" wrote:
>
> "Kim,Taesung" wrote:
>
> > Hello!
> > We( I and my friends) have plan to make soem application on linux.
> > First of all, we want to survey on going project on linux.
> > We want to know any kind of projects about linux.
> > Where can we find?
> > Thanks for regard.
>
> There are several RPM related projects but I have yet to find one that
> helps answer the following question:
>
> "Where and how do I find and get the rpms to bring my system, as
> installed, uptodate?"
>
> I would really like to be able to set a list of "trusted" repositories
> (such as updates.redhat.com, etc.) and a utility that would check my
> current version, dependencies, etc. and determine, and get the rpms I
> needed to bring it up-to-date.
>
> R.Parr
> Temporal Arts
This already exists. Mandrake 6.0 has it. The program is called
MandrakeUpdate. You run it and it first asks which mirror you wish to
use. Then it scans your packages and finds a list of upgrade packages.
You simply select the packages you want to upgrade and press the upgrade
button.
It works great. It makes life much easier.
------------------------------
From: [EMAIL PROTECTED] (Mike Dowling)
Subject: Re: unix98 pty's problems
Date: 14 Sep 1999 11:37:03 GMT
On 13 Sep 1999 22:26:35 GMT, Remco van den Berg <[EMAIL PROTECTED]> wrote:
>I build a kernel including unix98 pty support and made the /dev/ptmx device
>and the directory /dev/pts:
>
The file /usr/src/linux/Documentation/devices.txt says:
Recent versions of the Linux kernels and GNU libc contain support for
the System V/Unix98 naming scheme for PTYs, which assigns a common
device, /dev/ptmx, to all the masters (opening it will automatically
give you a previously unassigned PTY) and a subdirectory,
/dev/points, for the slaves; the slaves are named with decimal
integers (/dev/points/# in our notation). This removes the problem
of exhausting the namespace and enables the kernel to automatically
create the device nodes for the slaves on demand using the "devpts"
filesystem.
Naive Question:
Does this mean that there is supposed to be some kind of entry in
/etc/fstab for mounting something on /dev/devpts? Something like
none /dev/pts devpts defaults 0 0
Cheers,
Mike Dowling
--
My email address [EMAIL PROTECTED] above is a valid email address.
It is, in fact, a sendmail alias; the digit 'N' is incremented regularly.
Spammed aliases will be deleted. Currently, mike[5,7-9,10,12,13,16-18]
have been deleted. If email to mikeN bounces, try mikeN+1.
------------------------------
From: [EMAIL PROTECTED] (Dr. Peer Griebel)
Subject: Re: klogd crashes after dialing isp
Date: Tue, 14 Sep 1999 12:10:50 GMT
[EMAIL PROTECTED] (quark) wrote:
>Hi !
>
>I have a P200 MPP System (2 Processors) with Suse 6.2 installed.
>Kernel is 2.2.12 or 2.2.10. With both kernels I have the same problem:
>After dialin my ISP over ISDN the klogd crashes.
>If I start the klogd to write the messages into an own file
>(option -f) it is better, but not perfect.
>
>Anyone an idea?
Na mu�t Du 2mal senden?
Scheint auch keiner helfen zu k�nnen - schade.
Was passiert denn, wenn Du -f nimmst? Es ist besser aber nicht
perfekt. Was hei�t das?
Hast Du mal -d verwendet? Du kannst ja auch mal -s oder -x probieren
(wird aber wohl nichts ausmachen).
Sonst mu�t Du vielleicht nochmal genauer beschreiben, was passiert.
Werden keine Meldungen ausgegeben?
Ade
Peer.
P.S Denkst Du noch an meine CD?
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
