Linux-Development-Sys Digest #355, Volume #6 Sun, 31 Jan 99 13:14:29 EST
Contents:
Re: Modest next goal for Linux (Manfred Grebler)
Re: Still Framebuffer & Voodoo2 Glide problems on 2.2.1 (gpasa)
Re: Why I'm dumping Linux, going back to Windblows (Nix)
Kernel 2.2.1, SMP, and rvplayer (David Fox)
XEmacs problems and other stuff that don't work on newer than pre4 (gpasa)
Re: Debugger ? (Bjorn Reese)
Re: Compiling APUE source code on Linux (Bjorn Reese)
Re: Why I'm dumping Linux, going back to Windblows (Johan Kullstam)
Re: Still Framebuffer & Voodoo2 Glide problems on 2.2.1 (Tim T.)
Re: linux on an overclocked PII (Michael Linnemann)
2.2.0-pre[78] + ncr52c8xxx + tape = n.g. (Gerhard Traeger)
CFLAGS question, libc5, 2.0.36 (jamie)
Re: Linux Phase 2: A Consumer Operating System (steve mcadams)
Re: Linux Phase 2: A Consumer Operating System (steve mcadams)
Re: Linux Phase 2: A Consumer Operating System (steve mcadams)
Re: glibc-performance (James Youngman)
Re: password validation ("David D. Gitchell")
ptrace changing system call ([EMAIL PROTECTED])
----------------------------------------------------------------------------
From: Manfred Grebler <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Modest next goal for Linux
Date: 26 Jan 1999 07:29:33 GMT
Andreas Heiss wrote:
>
> But if a 300MHz PII NT box is basically unusable during scanning
> an image with a SCSI scanner, how's that ? (Move the mouse, the
> mouse cursor moves 2 minutes later ...)
This looks like the scanner is used at a very low-cost SCSI card
that comes with some scanners. These cards do not support DMA transfer
so the CPU has to do all the work.
That's not an NT issue, just crappy hardware.
Manfred
------------------------------
From: gpasa <[EMAIL PROTECTED]>
Subject: Re: Still Framebuffer & Voodoo2 Glide problems on 2.2.1
Date: Sun, 31 Jan 1999 14:14:29 +0100
You can find newer Glide version on
http://glide.xxedgexx.com/
and it works very well up to 2.2.0-pre4
and you can safely use MesaGl-3.0 with it.
But starting with 2.2.0-pre5 it doesn't work well anymore
everythiing hangs but you can use SysRq keys to reboot cleanly.
"Tim T." wrote:
>
> In article <[EMAIL PROTECTED]>,
> gpasa <[EMAIL PROTECTED]> writes:
> >
> > Hi,
> >
> <snip>
> Sorry, I can't help you with most of the stuff, but there was
> one thing in your post which drew my eye :
>
> >
> > 2. I cannot start an app which uses my Voodoo2 card. I'm using Glide
> > 2.53 glibc version (I'm running a RedHat 5.2 originally based system).
> > and Mesa 3.0 glibc too without blanking my screen an probably
> > hanging up my PC. Or it is frozen if it means something
> > different. Anyway I cannot use my PC anymore without restarting.
> > GLQuake start, but when launching the demo it stops and I have to
> > reboot.
> I just bought a Voodoo 2, and I discovered that in my case, it did
> not work with the glide library I downloaded from 3dfx.com. However,
> the version I have is the 2.4 version of the library.
>
> Where did you find this newer version ? Did it work with older
> versions of the kernel.
>
> TimT.
>
> ----------------------------------------------------------------------------
> [EMAIL PROTECTED] Voodoo Programmer/Keeper of the Rubber Chicken
> ... All we are saying is give peace a chance.
> ----------------------------------------------------------------------------
--
Sincerely yours,
Pasa Guglielmo
==================================================================
� homepage: http://www.omedia.ch/pages/gpasa/ �
�----------------------------------------------------------------�
� e-mail : [EMAIL PROTECTED] � tel. : +41 (0)24 485 50 40 �
� mailing : Pasa Guglielmo � fax : +41 (0)24 485 50 44 �
� Rte des Cases 17A � �
� CH-1890 St-Maurice � prof.: physicist �
� (Switzerland) � �
==================================================================
.
------------------------------
From: Nix <$}xin{[email protected]>
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Why I'm dumping Linux, going back to Windblows
Date: 31 Jan 1999 06:38:47 +0000
Johan Kullstam <[EMAIL PROTECTED]> writes:
> it seems that all these little programs do 90% of what you want. to
> get the other 10% someone writes an overlapping applet to cover a few
> new cases. under closed-source unix where you can't even fix/improve
> the utilities this increased the pressure for proliferation of little
> hacks. now it's mostly that old bugaboo of backwards compatibility
> which hinders altering/improving existing programs (consider the flap
> about ps trying to merge bsd and sysv behaviors).
>
> paradoxically, the easy extensibility of writing stuff as you go along
> and need them is what has made unix strong. shell conventions like
> redirection and piping has helped paste all sorts of things together
> and allow a small quick hack to be useful.
I think you're on to something here. This horde of tools has been
evolving almost in the Darwinian sense, swapping code and ideas like
bacteria swap plasmids, slowly spreading out to cover more things that
people might want to do with them.
And if a tool can't be easily extended to cover some new case, another
one springs up. It's an ecology of tools, really. And against Darwin
unleashed nothing else has much of a chance.
--
`I didn't want the bug *fixed*, I wanted to bitch pointlessly.' - Matthew
R. Williams on alt.religion.emacs
------------------------------
From: [EMAIL PROTECTED] (David Fox)
Subject: Kernel 2.2.1, SMP, and rvplayer
Date: 31 Jan 1999 05:12:31 -0800
I built a 2.2.1 SMP kernel with the APIC interrupt patch posted to the
kernel mailing list, and installed it on a dual PII/400 machine, and
so far the problem I'm still having is after (that piece of s***)
rvplayer runs for about fifteen seconds it gets "A general error
occurred" (not even a number!) and stops. I've tried kernels with and
without APM enabled. Sound players like mpg123 work fine. I would
send this to the kernel list if it weren't so vague and stupid.
--
David Fox http://hci.ucsd.edu/dsf xoF divaD
UCSD HCI Lab baL ICH DSCU
------------------------------
From: gpasa <[EMAIL PROTECTED]>
Subject: XEmacs problems and other stuff that don't work on newer than pre4
Date: Sun, 31 Jan 1999 14:25:19 +0100
Hi,
I ran XEmacs 20.4 perfectly on 2.2.0-pre4 kernel and now on
2.2.1 it doesn't run anymore in console mode. When I type xemacs
or its symlink emacs nothing happens no messages appear. When I ps
it is shown as a running app and I have to kill it.
There are a lot of problems in current post 2.2.0-pre4 releases
of the kernels. Below are a set of other problems I encountered on
kernels after 2.2.0-pre4 (all apps ran well on it but not on later
kernels).
I'm running pre4 without the ATI Mach64 console stuff, but
with framebuffer enabled and it works rather well. My PC is
a K6-2 3D 300MHz, 128M RAM , Monster 3D2, booting on a scsi
AHA 3940 based HDD, with an original 3Com 3c509 ISA PnP ethernet card
and an ASUSCOM ISDN ISA PnP card. My m/board is an ASUS TX-95E
LILO is launched by the NT5.0 loader, from a file bootlinx.bin on c:.
What I must say from kernel starting from pre5 and to pre9 inclusive is
:
1. (this is from all 2.2.0-pre releases) my 3c509 ether card is only
detected
if I build the driver as a module. If it is built in the kernel the
detection
doesn't work.
2. I cannot start an app which uses my Voodoo2 card. I'm using Glide
2.53
glibc version (I'm running a RedHat 5.2 originally based system ). and
Mesa 3.0
glibc too without blanking my screen an probably hanging up my PC. Or it
is frozen
if it means something different. Anyway I cannot use my PC anymore
without restarting.
GLQuake start, but when launching the demo it stops and I have to
reboot.
3. The framebuffer stuff doesn't work well. If I build it in the kernel,
On startup,
after the LILO prompt, I get a lot of garbage on the screen, the printer
initializes,
I can even imagine the Linux logo to appears in the garbage, but then
nothing.
This is however better than in the previous pre releases (before pre9)
as now the garbage
is moving. In pre8 it was fixed. And in pre9 I can see the logo position
quite well.
More precisely what I see is a mess of vertical lines (red, green and
white) and
the screen seems to be periodic as the logo appears 3 to 4 times. I
cannot see
anything correctly. Everything is made of a lot of dots or little
patterns.
In pre4 the atyfb doesn't detected my ATI card, but I could play
quake(2) on my Voodoo2.
Now it seems that the card is recognized but i cannot play. Even if the
fb stuff is not
in the kernel, I can start linux and it seems to work well but I cannot
play quake(2) or
run any Voodoo based application.
--
Sincerely yours,
Pasa Guglielmo
==================================================================
� homepage: http://www.omedia.ch/pages/gpasa/ �
�----------------------------------------------------------------�
� e-mail : [EMAIL PROTECTED] � tel. : +41 (0)24 485 50 40 �
� mailing : Pasa Guglielmo � fax : +41 (0)24 485 50 44 �
� Rte des Cases 17A � �
� CH-1890 St-Maurice � prof.: physicist �
� (Switzerland) � �
==================================================================
.
------------------------------
From: [EMAIL PROTECTED] (Bjorn Reese)
Crossposted-To: comp.os.linux.development.apps,comp.unix.bsd.freebsd.misc
Subject: Re: Debugger ?
Date: 31 Jan 1999 13:05:25 GMT
Douglas Cook ([EMAIL PROTECTED]) wrote:
> On this note, how do you pass a command line parameter to your program
> into gdb? On the HP-UX, gdb passed all options after the exe name to
> the program, but on FreeBSD it looks for a core-dump by the name of the
> parameter.
% gdb your_program
(gdb) run insert_your_parameters_here
------------------------------
From: [EMAIL PROTECTED] (Bjorn Reese)
Crossposted-To:
comp.os.linux.misc,comp.os.linux.development.apps,comp.os.linux.questions
Subject: Re: Compiling APUE source code on Linux
Date: 31 Jan 1999 13:16:20 GMT
Lakshmi Natarajan ([EMAIL PROTECTED]) wrote:
> Has anyone compiled the source code from Richard Stevens' "Advanced
> Programming in the Unix Environment" on Linux? I am trying to do it on
> my RedHat Linux 3.0.30 system, but cannot.
http://www.kohala.com/~rstevens/apue.html
------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Why I'm dumping Linux, going back to Windblows
From: Johan Kullstam <[EMAIL PROTECTED]>
Date: 31 Jan 1999 08:44:30 -0500
Nix <$}xin{[email protected]> writes:
> Johan Kullstam <[EMAIL PROTECTED]> writes:
>
> > it seems that all these little programs do 90% of what you want. to
> > get the other 10% someone writes an overlapping applet to cover a few
> > new cases. under closed-source unix where you can't even fix/improve
> > the utilities this increased the pressure for proliferation of little
> > hacks. now it's mostly that old bugaboo of backwards compatibility
> > which hinders altering/improving existing programs (consider the flap
> > about ps trying to merge bsd and sysv behaviors).
> >
> > paradoxically, the easy extensibility of writing stuff as you go along
> > and need them is what has made unix strong. shell conventions like
> > redirection and piping has helped paste all sorts of things together
> > and allow a small quick hack to be useful.
>
> I think you're on to something here. This horde of tools has been
> evolving almost in the Darwinian sense, swapping code and ideas like
> bacteria swap plasmids, slowly spreading out to cover more things that
> people might want to do with them.
yes, but if one of those species gets enshrined in posix, it can
*never* go extinct no matter how useless, superceded, bloated, ugly or
ill-suited it may be/become. this is a weak point of unix (and almost
every other OS, e.g., windows nt running dos programs...).
--
Johan Kullstam [[EMAIL PROTECTED]] Don't Fear the Penguin!
------------------------------
From: [EMAIL PROTECTED] (Tim T.)
Subject: Re: Still Framebuffer & Voodoo2 Glide problems on 2.2.1
Date: 31 Jan 1999 12:07:42 GMT
In article <[EMAIL PROTECTED]>,
gpasa <[EMAIL PROTECTED]> writes:
>
> Hi,
>
<snip>
Sorry, I can't help you with most of the stuff, but there was
one thing in your post which drew my eye :
>
> 2. I cannot start an app which uses my Voodoo2 card. I'm using Glide
> 2.53 glibc version (I'm running a RedHat 5.2 originally based system).
> and Mesa 3.0 glibc too without blanking my screen an probably
> hanging up my PC. Or it is frozen if it means something
> different. Anyway I cannot use my PC anymore without restarting.
> GLQuake start, but when launching the demo it stops and I have to
> reboot.
I just bought a Voodoo 2, and I discovered that in my case, it did
not work with the glide library I downloaded from 3dfx.com. However,
the version I have is the 2.4 version of the library.
Where did you find this newer version ? Did it work with older
versions of the kernel.
TimT.
============================================================================
[EMAIL PROTECTED] Voodoo Programmer/Keeper of the Rubber Chicken
... All we are saying is give peace a chance.
============================================================================
------------------------------
From: [EMAIL PROTECTED] (Michael Linnemann)
Subject: Re: linux on an overclocked PII
Date: Sun, 31 Jan 1999 13:59:27 +0100
On 30 Jan 1999 20:58:47 +0100, Jan Andres <[EMAIL PROTECTED]> wrote:
>Frank Hale <[EMAIL PROTECTED]> writes:
>> I overclocked my PII 266 to 300 and upped the bus speed to 75 mhz. Now
>> when I boot Linux turns off DMA on my hard drives. Is this bad? What
>> does DMA do anyway?
DMA is Direct Memory Access, the disk interface can write directly into
RAM without bothering the CPU.
>> By the way I am on 2.2.0-final (aka 2.2.0-pre9)
>>
>> Jan 25 20:12:06 FranksPC kernel: hdb: timeout waiting for DMA
>> Jan 25 20:12:06 FranksPC kernel: hdb: irq timeout: status=0x58 {
>> DriveReady SeekComplete DataRequest }
>> Jan 25 20:12:06 FranksPC kernel: hda: DMA disabled
>> Jan 25 20:12:06 FranksPC kernel: hdb: DMA disabled
>> Jan 25 20:12:06 FranksPC kernel: ide0: reset: success
>Overclocking makes your system very unstable. Windows users often
>don't notice this, as they are used to instability, but under Linux,
>you do.
Quite right.
>As you can see here, the CPU cannot handle the bus correctly any
>more. I suppose the DMA requests are eaten up somewhere in the bus
>interface. I suggest that you undo the overclocking. The performance
>win that you get from 266->300 MHz is not very big, but the
>reliability loss *is* big. It's better to have a system that is a bit
>slower, but that works without problems.
The speedup from 266 -> 300MHz is hardly _noticeable_; it is pure nonsense
to try it at the expense of stability.
It is not only the CPU (which might stand the overclocking, if adequate
cooling is provided): it the bus speed that makes the difference. Most
chipsets, PCI cards etc. have little to _no_ margins in speed. It is
really _asking_ for trouble if you try it.
BTW, this is not a system, but a hardware question.
Regards
Michael
------------------------------
From: [EMAIL PROTECTED] (Gerhard Traeger)
Subject: 2.2.0-pre[78] + ncr52c8xxx + tape = n.g.
Date: 31 Jan 1999 15:44:30 +0100
On 22 Jan 1999 01:31:32 GMT, [EMAIL PROTECTED] (bill davidsen) wrote:
> The ncr53c8xx driver works fine for my disk, and it finds the tape
> drive, so I thought it was okay. On trying to write a tape, the
> controller hangs, ...
It hangs?
I just get:
: gt-priv [7] tar tvf /dev/rmt0
: tar: Read error on /dev/rmt0: Input/output error
: tar: At beginning of tape, quitting now
: tar: Error is not recoverable: exiting now
No hang, no error log, no problems with disks.
This is with ncr53c875 + HP35480A dds/dat.
>From fall 1993 untill kernel 2.1.107 linux+scsi worked fine for me.
When Gerard switched to ncr driver version 3.x, i got trouble with
my cdrom drive (M_REJECT). Somewhere between 2.1.119 and 2.1.128 (i
didn't try between), he changed sort of timing, so i could again use
the cdrom.
>From 2.1.128 to 2.1.131, i could read the tape, but not write onto.
I compiled 2.1.128 - 2.1.131 with the ncr53c8xx driver from 2.1.107
and it worked fine.
Since 2.1.132, i can no longer read the tapes. Using the old
driver does no longer help.
Thus i stick with kernel 2.1.131 + ncr53c8xx v 2.5f (from kernel
2.1.107).
Gerhard.
------------------------------
From: [EMAIL PROTECTED] (jamie)
Subject: CFLAGS question, libc5, 2.0.36
Reply-To: [EMAIL PROTECTED]
Date: Sun, 31 Jan 1999 10:23:09 -0600
I'm setting up a new PII machine, with Slackware 3.6, kernel 2.0.36 .
If I understand correctly, it uses libc5 for C code and glibc for
C++ code.
I noticed when compiling the kernel with Pentium Pro set, it used
CFLAGS of -DCPU=686 and -m486 . Does -mpentium behave badly in
the kernel, or is it just an old default?
What CFLAGS ought I set for best PII performance when compiling
other applications?
Email replies requested, because my ISP's braindead WinNT newsserver
crashes almost daily
--
jamie ([EMAIL PROTECTED])
"There's a seeker born every minute."
------------------------------
From: [EMAIL PROTECTED] (steve mcadams)
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Linux Phase 2: A Consumer Operating System
Date: Sun, 31 Jan 1999 17:51:08 GMT
[Snipped for brevity, quoted material marked with ">"]
On Wed, 20 Jan 1999 02:01:32 GMT, [EMAIL PROTECTED] (Christopher
Browne) wrote:
>With the result that many of the sorts of folk that made Linux grow into
>what it has become are liable to migrate off to other systems that
>permit them to "play" again. Whether that's Hurd, Xos, FluxOS, *BSD, or
>whatever.
>
>At that point, Linux starts to die. It may continue on for years, but
>if the sources of innovation leave, that causes great injury.
I don't see how adding more bolt-on capabilities to a rugged high
performance operating system is going to make it any less interesting.
In fact it will probably make it more interesting because more users
will find more interesting problems to solve. -steve
========================================================
so what? - http://www.codetools.com/showcase
------------------------------
From: [EMAIL PROTECTED] (steve mcadams)
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Linux Phase 2: A Consumer Operating System
Date: Sun, 31 Jan 1999 17:51:20 GMT
It's not clear why you're so fired up about someone trying to start a
development effort. There are some cool problems in the target area
that could be solved well to everyone's advantage.
Probably the original poster managed to push one of your hotbuttons,
maybe from another discussion or something. Chill, if you don't want
to work on UI stuff then don't. Nobody here has to do rat shit unless
s/he/it wants to, so nobody using the word "must" by mistake is going
to fool anybody who's read these ng's for more than a day. Who knows
what the original poster's native language is, or cares? Maybe I
shouldn't ask that<g> -steve
[Snipped for brevity, quoted material marked with ">"]
On 20 Jan 1999 15:47:05 -0500, [EMAIL PROTECTED] (Alexander Viro)
wrote:
>>Making Linux into a consumer product has two major primary benefits
>
>Go and do it if you want it. You are nothing. You are less than nothing - you
>are marketdroid wannabe who barely learned to post. You words worth not more
>than your pathetic self until you *bloody go and write what you want to have*.
>
>[snip]
>>To accomplish this, developers must find out what consumers want.
>
>Then go, find it out and do it. And don't DARE to tell "must" to people
>who don't depend on you and owe you nothing.
>
>[snip]
>>Please remove the .nospam from my reply-to address
> You think that you are worth demunging your bloody address???
>Speak about the gall...
>
>FOAD. *PLONK*
========================================================
so what? - http://www.codetools.com/showcase
------------------------------
From: [EMAIL PROTECTED] (steve mcadams)
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Linux Phase 2: A Consumer Operating System
Date: Sun, 31 Jan 1999 17:51:18 GMT
[Snipped for brevity, quoted material marked with ">"]
On Thu, 21 Jan 1999 05:17:57 GMT, [EMAIL PROTECTED] (Christopher
Browne) wrote:
>The point is that if interesting pieces of the system "solidify" to the
>point that people can't change them without causing immense breakage,
>then it's no longer fun playing with those pieces.
Nothing has to solidify to achieve the goal of making Linux easier to
install, configure, administer, and use. It's a matter of writing the
code that runs on top of Linux and does this stuff. Linux itself
doesn't have to change a bit,. -steve
========================================================
so what? - http://www.codetools.com/showcase
------------------------------
From: James Youngman <[EMAIL PROTECTED]>
Subject: Re: glibc-performance
Date: 30 Jan 1999 17:46:09 +0000
[EMAIL PROTECTED] (Horst Piening) writes:
> when i ran my raytracer for time testing the new glibc stuff was
> 15 % slower than before.
>
> what did i overlook, i expectet a time increase ?
profile it with "gcc -g -pg". man gprof.
--
ACTUALLY reachable as @free-lunch.demon.(whitehouse)co.uk:james+usenet
------------------------------
From: "David D. Gitchell" <[EMAIL PROTECTED]>
Subject: Re: password validation
Date: Sun, 31 Jan 1999 08:55:55 -0600
Nicholas Todd Lawrence wrote:
>
> Unfortunatly, I am not the administrator of the machine I am working on
> and cannot change the uid of my program to root.
>
> Would the following idea work?
<SNIP>
No it won't, and there are very good reasons why you should NOT be able
to do what you are suggesting without at least the help of the
superuser. (Help me up on this soapbox, would you? -- There, thanks.)
You are contemplating a cgi-bin executable that will take persons'
usernames and passwords, and authenticate these with the host system.
Do you see no potential security problems with a non-priviledged user
having the power to do this? Your cgi-bin executable will be able to
capture the stream from the web, and pass it along to the authentication
routines, and get a PASS/FAIL response from the authentication
routines. If you get a PASS response, you have just stolen that
person's identity, since your cgi-bin executable holds the username and
password that person uses on that host. Store them in a file in plain
text and you can login as that person any time that you wish. If you DO
find a way to do it, please let the Security newsgroups know, as they
will want to plug that loophole ASAP!
That being said, there is still an avenue you can follow to accomplish a
goal of password protection, using the 'passwd' routines. You can
establish YOUR OWN passwd file, and keep username:password info for the
persons authorized to connect to your web site. These will NOT be the
same as these persons' username:password pairs for the host system; they
will be unique to your website/webpage. In fact, you should caution
users NOT to use their host system username:password, as that will
result in an insecurity.
Good luck!
-- Dave
============================================================
#include <std_disclaimer.h> /* I speak for myself, only. */|
============================================================
LCDR David D. Gitchell, USN (Retired), Hutchinson, KS, USA
[EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Subject: ptrace changing system call
Date: Sun, 31 Jan 1999 17:53:03 GMT
Hi,
In linux the system call number is stored in ORIG_EAX
which can be seen using ptrace(PEEK_USER,pid,4*ORIG_EAX,0)
Also the other arguments can be viewed using PEEK_USER
and modified using POKE_USER. However, if I try to change
the system call number,using POKE_USER, it gives me EIO error i.e. request
not valid. Does that mean that I cant change the system call
number? or is there some other way of doing it?
Any help in this regard will be greatly appreciated.
Thanks!
Kapil
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
** 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
******************************
