Linux-Development-Sys Digest #197, Volume #6 Fri, 1 Jan 99 14:13:59 EST
Contents:
Re: Registry for Linux - Bad idea (weg)
Re: parport/ppa problem with 2.2.0pre1 (Jens Kristian S�gaard)
Re: Mapping to a Physical Address / And Cache Types (Jens Kristian S�gaard)
Re: Mapping to a Physical Address / And Cache Types (H. Peter Anvin)
2.2.0pre2 compile bug (Paul Martin)
pthread debugging ("James A. Cleland")
Re: Redhat 5.0 doesnt like onboard video cards? ("D. Stimtis")
2.1.130 net error question ("D. Stimtis")
Re: parport/ppa problem with 2.2.0pre1 (Griffin Caprio)
Re: How to run Windows Applications on Linux (Taso Hatzi)
Re: Where is the bound port in TCP found? (Peter Pointner)
Re: Why I'm dumping Linux, going back to Windblows
Kernel-2.2-pre2 (Vitor Pedro Bonucci Pias)
Where is the Console Driver Menu in Kernel 2.1.99? ("George Thia")
Re: Possible Bug in aic7xxx driver / Ultra2Disk ? (Keith Chau)
silly question (ebatchelor)
Re: silly question (Jens Kristian S�gaard)
Re: 'make xconfig' problem in 2.2.0-pre2 (Stefaan A Eeckels)
----------------------------------------------------------------------------
From: weg <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: Thu, 31 Dec 1998 20:42:57 -0600
Reply-To: [EMAIL PROTECTED]
Ya'know - I just installed Win98 on one of my machines. As I look over
the win98 GUI, I can't help but notice the influence of *many* common
unix window managers. Win98 made another impression within 10 minutes
of running on my system - It completely f**ked-up the setting on my
video card (STB Velocity). I can't wait to see what it screws up next.
Now I don't know about other people - but trying to mimic a failed OS
doesn't make a lot of sense to me (why mimic an Edsel??). Besides,
MS-Win products are not built for the same crowd as linuxers. Let MS
have the GD registry - I'll take my Linux plain thank you.
------------------------------
From: [EMAIL PROTECTED] (Jens Kristian S�gaard)
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: parport/ppa problem with 2.2.0pre1
Date: 01 Jan 1999 00:43:02 +0100
Griffin Caprio <[EMAIL PROTECTED]> writes:
> I can't even get my printer to work under 2.2.0pre1. It works fine in
> 2.0.36, but not in the newer kernels. I tried to switch from lp1 to lp0
> and it still doesn't work. Any tips?
Try putting something like this in your /etc/conf.modules:
alias parport_lowlevel parport_pc
options lp parport=0
I've set the option to lp back to 0, as I suppose you only have only
parallel port ( and thats the one the printer is connected to ). I'm
not sure if it's needed -- but anyways...
I ofcourse assume you have a normal PC parallelport.
--
Jens Kristian S�gaard,
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Jens Kristian S�gaard)
Subject: Re: Mapping to a Physical Address / And Cache Types
Date: 01 Jan 1999 00:45:17 +0100
[EMAIL PROTECTED] (TERENCE MURPHY) writes:
> a region to be uncached, one to be write through, etc. If there is a
> MMAP() parameter for this, that would be fantastic, or do I just have
> to program the MTRR's myself for this?
Look at the text in /usr/src/linux/Documentation/mtrr.txt of a
devel. kernel ( or the 2.2.0 ).
It even has a short code example which show the use of the new
/proc/mtrr interface.
--
Jens Kristian S�gaard,
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (H. Peter Anvin)
Subject: Re: Mapping to a Physical Address / And Cache Types
Date: 31 Dec 1998 22:36:52 GMT
Reply-To: [EMAIL PROTECTED] (H. Peter Anvin)
Followup to: <76gonb$3ei$[EMAIL PROTECTED]>
By author: [EMAIL PROTECTED] (TERENCE MURPHY)
In newsgroup: comp.os.linux.development.system
>
> Any suggestions for this would be appreciated. Please no flames:
> I need the physical address and cache types for a very specific,
> inherently non-portable testing program.
>
You need to write a device driver, which you can then mmap().
-hpa
--
PGP: 2047/2A960705 BA 03 D3 2C 14 A8 A8 BD 1E DF FE 69 EE 35 BD 74
See http://www.zytor.com/~hpa/ for web page and full PGP public key
I am Bah�'� -- ask me about it or see http://www.bahai.org/
"To love another person is to see the face of God." -- Les Mis�rables
------------------------------
From: [EMAIL PROTECTED] (Paul Martin)
Subject: 2.2.0pre2 compile bug
Date: 01 Jan 1999 03:53:36 GMT
in asm-i386/bugs.h:
/* If we were told we had a good APIC for SMP, we'd better be a PPro */
#ifdef CONFIG_GOOD_APIC
if (smp_found_config && boot_cpu_data.x86 <= 5)
panic("Kernel compiled for PPro+, assumes local APIC without
read-before-write bug");
#endif
This won't compile if you have not got SMP turned on (smp_found_config is
undefined).
Fix is to put #ifdef __SMP__ / #endif round this bit.
--
Paul Martin <[EMAIL PROTECTED]>
------------------------------
From: "James A. Cleland" <[EMAIL PROTECTED]>
Subject: pthread debugging
Date: Thu, 31 Dec 1998 18:56:32 -0500
I would like to be able to debug a multithreaded app with gdb. I
installed a patch (gdb-4.17.patch) which was supposed to enable gdb to
handle threads (I really don't know jack about the internals here, so
bear with me). I don't know whether or not I need an updated glibc or
not. Basically, I just don't know anything about this issue.
What I would like are binaries for gdb and whatever else I need to get
into a MT app with gdb.
-or-
Someone to tell me what I need to do to make the afore mentioned
binaries myself (*ack*).
As far as my development env, I'm using stock RH5.1. I think that's
glibc 2.0.7, egcs 1.0.2, gdb 4.17.
Oh, by the way, gdb ->info threads doesn't present any info on the new
build with the patch.
Another thing, I just tried recompiling gdb 4.17 with the patch
installed (fresh tar install) and got an error:
gcc -c -g -O2 -D__USE_MISC -I. -I. -I./config -DHAVE_CONFIG_H
-I./../include/opcode -I./../readline -I../bfd -I./../bfd
-I./../include infptrace.c
infptrace.c: In function `attach':
infptrace.c:206: `PTRACE_ATTACH' undeclared (first use this function)
Thanks,
James
------------------------------
Date: Thu, 31 Dec 1998 22:04:32 -0700
From: "D. Stimtis" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Redhat 5.0 doesnt like onboard video cards?
[EMAIL PROTECTED] wrote:
>
> I am just getting started using Linux. I hope one day it knocks Windows off
> its throne...but I have had nothing but trouble so far. If I could just get
> my video card to work so I could run Xwindows, i would be satisfied. If
> anyone can help me, I would really appreciate it. I have a fairly new
> Gateway G6-300 which has onboard video. Here are the specs: Mpact2-3DVD (4
> meg) Chipset=Yamaha 6388 VPDC RAMDAC= Generic 8-bit pseudo-color DAC
SuSE seems to have the most updated set of X servers, and video-hardware-related lists.
Try http://www.suse.com/ and look for the X and the hardware listings.
>
> Thanks in advance!
>
> [EMAIL PROTECTED] or
> [EMAIL PROTECTED]
------------------------------
Date: Thu, 31 Dec 1998 22:45:04 -0700
From: "D. Stimtis" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: 2.1.130 net error question
Hi:
Something I never found with earlier kernel versions (pre 2.0.30?), but found in
2.0.32+ stable, and
2.1.130 kernels, is the occasional log message similar to:
kernel: TCPv4 bad checksum from 209.41.82.59:0000 to 209.60.72.108:0100,
len=1480/1480/1500
That one was from an SGML info site using Netscape, and recorded in /var/log/messages.
The same
message occurs with 2.0.35, but not on my 2.0.30 kernels of the past.
I traced the function to file /usr/src/linux/net/ipv4/tcp_ipv4.c,
and function within that file
int tcp_v4_rcv( struct sk_buff *skb, unsigned short len )
Some sites I really need to access are unavailable for days at a time, or in the case
of trying to
get a badly needed DTD from an SGML info site, permanently.
Inside this function is a switch statement looking at skb->ip_summed, for which one
possible case is
CHECKSUM_HW. If this occurs, the error message is generated.
My first question is what is the purpose of this checksum test? Is it a security
feature against
phoney packet lengths? Or has it always been here, but I just didn't see it for other
reasons on
earlier kernels?
Second, if the purpose is to stop bogus packets from bringing down the machine, could
it be recoded
to only dump packets which indicate either a) an end address prior to the beginning
address, or b)
an end address that is beyond the packet data? i.e., allow packets that indicate an
end address that
still terminates within the allowable data segment of the packet? All of this is, of
course,
irrelevant if this isn't a security issue.
Assuming that the error message is correct and that packets are arriving with bad
checksums,
wouldn't the offending site know this is happening? I can't even contact the admin or
webmaster at
some sites to tell them what is happening, due to the error stopping even email.
Third, when was this code added?
Thanks!
D. Stimits, [EMAIL PROTECTED]
------------------------------
From: Griffin Caprio <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: parport/ppa problem with 2.2.0pre1
Date: Fri, 01 Jan 1999 01:36:46 -0600
Jens Kristian S�gaard wrote:
> Griffin Caprio <[EMAIL PROTECTED]> writes:
>
> > I can't even get my printer to work under 2.2.0pre1. It works fine in
> > 2.0.36, but not in the newer kernels. I tried to switch from lp1 to lp0
> > and it still doesn't work. Any tips?
>
> Try putting something like this in your /etc/conf.modules:
>
> alias parport_lowlevel parport_pc
> options lp parport=0
>
> I've set the option to lp back to 0, as I suppose you only have only
> parallel port ( and thats the one the printer is connected to ). I'm
> not sure if it's needed -- but anyways...
>
> I ofcourse assume you have a normal PC parallelport.
>
The weird thing is I can cat to it, just not with lpr. maybe I need a newer
distro of lpr?!?
------------------------------
From: Taso Hatzi <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development,comp.os.linux.development.apps,linux.dev.c-programming
Subject: Re: How to run Windows Applications on Linux
Date: Fri, 01 Jan 1999 08:38:42 +0000
Eric Miller wrote:
> be very demanding. Too bad those ESRI folks have jumped on the Windoze
> bandwagon in such a big way.
The sooner people jump onto the so-called Windows bandwagon, the sooner
they
will realize there has to be a better way.
------------------------------
From: Peter Pointner <[EMAIL PROTECTED]>
Subject: Re: Where is the bound port in TCP found?
Date: Fri, 1 Jan 1999 11:21:48 GMT
Ayman El-Khashab <[EMAIL PROTECTED]> wrote:
> actually, the first syn packet is where i get most of my information.
> The only piece i can't get from the initial syn (or at least I don't
> see how it is possible) is the bound port that the connection commences
> on on the servers side. So the server typically does a bind (so for
> telnet 23) and then an accept which I think gives back a new sock
> structure in the kernel. And associated with the accept is the port
> number that the connection continues on. (my best from the top of my
> head recollection of what happens).
> Since the accept is performed by the server, it is not clear how I can
> get that port number from the initial syn. I do see how i could get it
> from every IP packet, but I am examining things at the tcp layer.
> so the question is how to find what port it is from the data in either
> the tcp header . . . or from any of the functions such as tcp_rcv, etc.
> At the moment all i seem to find is the telnet port number in the dest
> of the tcp header in each received packet. I am guessing it is elsewhere
> in the structures and I am just missing it.
I think you can't find that because it's not there: The server creates a
new socket during accept, but that uses the original port number. Probably
the TCP layer uses the client-address (= source-address) in incoming
packets to find the correct destination socket.
Peter
------------------------------
From: [EMAIL PROTECTED] ()
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Why I'm dumping Linux, going back to Windblows
Date: Tue, 29 Dec 1998 02:09:34 GMT
On 15 Dec 1998 15:53:05 GMT, bilge <[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] [EMAIL PROTECTED] blared:
> >On 10 Dec 1998 15:19:42 -0600, Lars Clausen <[EMAIL PROTECTED]> wrote:
>
> >
> > That's why Linux is gaining an audience to begin with;
> > the predominant developer has no motivation to deal
> > with everyone's needs.
> >
> >
> There is a way to accomodate both views here and do so in
> the same way that linux is developed - at large. Using
> a gui scripting language (something like dtksh, but I assume,
> not dtksk), a scripted, graphical, modified-by-anyone
> interface could be created by collecting the contributions
> of anyone willing to write even a tiny piece. That leaves
> all the configuration in an ascii file and actually, the
> graphical part need only be a wrapper over othe config scripts.
> Have a look at the "find" script in the dtksh book. I certainly
> would NOT consider using it, but someone would.
Want my tk wrapper for metasend? '-)
--
Unix had startmenus and taskbars before Microsoft |||
even had a decent memory manager for DOS. / | \
In search of sane PPP docs? Try http://penguin.lvcm.com
------------------------------
From: Vitor Pedro Bonucci Pias <[EMAIL PROTECTED]>
Subject: Kernel-2.2-pre2
Date: Fri, 01 Jan 1999 12:46:22 +0000
When I disable SMP in "make config" the kernel fail to
compile.But I enable SMP and now seems OK.
But my syslogd put this mesages in /var/log
"Can't find map file"
Any susgestion is apreciated
Pedro Pias (Porto *** Portugal)
------------------------------
From: "George Thia" <[EMAIL PROTECTED]>
Subject: Where is the Console Driver Menu in Kernel 2.1.99?
Crossposted-To:
alt.os.linux,alt.uu.comp.os.linux.questions,aus.computers.linux,comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.networking,comp.os.linux.setup,comp.os.linux.x,fj.os.linux,h
Date: 1 Jan 1999 23:46:02 +0800
I am unable to locate the Console Driver Menu in Kernel 2.1.99
Can anyone help?
Thanks!
------------------------------
From: [EMAIL PROTECTED] (Keith Chau)
Subject: Re: Possible Bug in aic7xxx driver / Ultra2Disk ?
Date: Fri, 01 Jan 1999 23:49:20 +0800
[EMAIL PROTECTED] wrote:
>Helmut Kreiser <[EMAIL PROTECTED]> wrote:
>: we are running Linux systems with the new Ultra2 Disk (LVD), with new
>: ASUS-Boards including aic7890 on board, supporting the lvd-disks.
>: With the kernel 2.0.35 plus aic7xxx-patch 5.1.2 the disks were
>: synchronized up to 80 MBytes/sec.
>
>: After installing kernel 2.0.36 (with aic7xxx 5.1.4) the disks are now
>: only synchronized with 20 MBytes/sec !
>: Other types ogf disks are recognized correct.
>
>Same problem here, although on my aic7890 fast/ultra scsi devices are
>synchronized at 10MBytes/sec with 2.0.36 as opposed to 20MBytes/sec
>with the patched 2.0.35.
Same problem here! I am using Asus P2B-LS. My IBM U2 LVD drives can
only be synced at 20MB/s! I am using 2.0.36 with 5.1.6 aic7xxx
patches.
>: Does anyone have an idea ? Or is it a small bug in the new drivers ?
>
>I have looked at differences in the source, but there are too many to
>check by myself. You could sent a message to the maintainer of the aic
>code.
------------------------------
From: ebatchelor <[EMAIL PROTECTED]>
Subject: silly question
Date: Fri, 01 Jan 1999 10:17:27 -0600
I am a new user to Linux. I am an experienced MSDOS user, have written
many batch files to accomplish what I want to do, and can recall the
names of most DOS utilities I need to use. Of course, DOS sucks, but
Windows cures many of the DOS shortcomings (long file names,
multitasking (almost), etc.). Linux seems to incorporate the best of
both worlds,
but....
Why the convoluted, hard to recall, someone thought it was funny in 1975
utility names? It seems to me that BASH could be easily recoded to
include easy to use and remember identifiers without giving up ANY
functionality. I know it's part of the worship Unix thing, but it seems
Linux could be more user friendly with little effort...
Just a question.
Ed Batchelor
innocent bystander
------------------------------
From: [EMAIL PROTECTED] (Jens Kristian S�gaard)
Subject: Re: silly question
Date: 01 Jan 1999 18:51:27 +0100
ebatchelor <[EMAIL PROTECTED]> writes:
> Why the convoluted, hard to recall, someone thought it was funny in 1975
> utility names? It seems to me that BASH could be easily recoded to
Oh, you're quite wrong ;-) It's DOS who though up some very weird
command names.
These commands have been named that way for decades on unixoid
systems... hopefully it will remain that way. Much the way that
AmigaDOS command looks like normal unix commands ( and ofcourse uses
the extended csh shell ).
--
Jens Kristian S�gaard,
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Stefaan A Eeckels)
Subject: Re: 'make xconfig' problem in 2.2.0-pre2
Date: 1 Jan 1999 17:13:16 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (N1ho) writes:
> While attempting to build a kernel under 2.2.0-pre2, the 'make xconfig' failed
> with an error. I read Linus' and Alan's comments about this (for 'pre1') and I
> hope I can add some constructive observations.
>
> The error that I got was:
> wish -f scripts/kconfig.tk
> Error in startup script: invalid command name "clear_choices"
> while executing
> "clear_choices"
> (procedure "read_config" line 3)
> invoked from within
> "read_config .config"
> (file "scripts/kconfig.tk" line 485)
>
> Well, I took a look at kconfig.tk and compared it to the one that was used
> in 2.1.131, and while they are similar, it appears it was extensively rewritten
> for 2.2.0-pre*. The function "clear_choices" (and "update_choices" and an
> unknown number of others) are called from within another function in the
> script, but the '_choices' functions are NOT defined, at least not in the same
> fashion that they were in .131. (There may be some sort of C-like header file
> or Ksh-like function file inclusion mechanism that was used instead, which
> may be broken. I am not a Tk expert, and I only took a cursory look). I never
> got around to testing .132, so I don't know when the changes were introduced.
It's caused by fix_choice_cond() in scripts/tkcond.c. It prints
'Ooops' and calls exit(0). The resulting 'Ooops' is easy to
overlook, but the conversion of config.in halts, and the kconfig.tk
script is incomplete (the update_mainmenu & consorts are added
by tkparse).
The introduction of the following lines in configure.in causes
the 'Ooops':
17 586/K5/5x86/6x86 CONFIG_M586 \
18 Pentium/TSC CONFIG_M586TSC \
19 PPro/K6/6x86MX CONFIG_M686" PPro
20 #
21 # Define implied options from the CPU selection here
22 #
23 if [ "$CONFIG_M386" != "n" ]; then
24 define_bool CONFIG_WP_WORKS_OK y
25 define_bool CONFIG_INVLPG y
26 define_bool CONFIG_BSWAP y
27 fi
28 if [ "$CONFIG_M686" = "y" -o "$CONFIG_M586TSC" = "y" ]; then
29 define_bool CONFIG_TSC y
30 fi
31 if [ "$CONFIG_M686" = "y" ]; then
32 define_bool CONFIG_GOOD_APIC y
33 fi
34
and changing line 23 to
23 if [ "$CONFIG_M386" = "y" ]; then
fixes the problem.
PS. The fix_choice_cond() function checks the first condition after
a choice, and makes sure it tests against a "y". People who add
stuff to config.in should get acquainted with its restrictive syntax ;-)
Happy New Year,
--
Stefaan
--
PGP key available from PGP key servers (http://www.pgp.net/pgpnet/)
___________________________________________________________________
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away. -- Saint-Exup�ry
------------------------------
** 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
******************************
