Linux-Development-Sys Digest #700, Volume #7 Thu, 23 Mar 00 09:13:14 EST
Contents:
Re: Netproblems on Linux AXP and Win95 (Anders Larsen)
plz....help me.....how can i do? ("daewoois")
Re: Netproblems on Linux AXP and Win95 (Martin Kahlert)
Re: poll()ing TCP sockets - when can I expect a POLLHUP? (Casper H.S. Dik - Network
Security Engineer)
Re: UART overrun errors (Mats Byggmastar)
Re: Netproblems on Linux AXP and Win95 (Anders Larsen)
Re: Can i get a MAC address ? (Jos Hulzink)
aic7xxx and AIC-7892/9 /Ultra160) (Josef Moellers)
Re: UART overrun errors (Etienne Lorrain)
Re: UART overrun errors (Josef Moellers)
Re: QUESTION: Linux Multiprocessor management ("Martin Sinot")
ioctl block pthread_create ?? (Nathan Krasney)
Re: ioctl block pthread_create ?? (Mathias Waack)
Re: Absolute failure of Linux dead ahead? (Geoff Lane)
Zero padding in sprintf() doesn't work for strings - HELP ! (Gerd Buesken)
Re: Question for Embedded Linux (James Moger)
Re: Absolute failure of Linux dead ahead? (Christopher Browne)
Re: gcc problems (Colin Watson)
----------------------------------------------------------------------------
From: Anders Larsen <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.alpha
Subject: Re: Netproblems on Linux AXP and Win95
Date: Thu, 23 Mar 2000 09:12:40 +0100
Martin Kahlert wrote:
>
> Hi,
> I can't get my home network to work.
> I have two computers at home:
> One is an Alpha Linux box with a 100MB network card which i run
> by the tulip driver (I also tried the de4x5 driver with the same
> negative results, too).
Tulip and de4x5 aren't even remotely related.
Which kernel version are you using?
> I placed the neccessary routing entry for eth0 into /etc/route.conf.
> And set up the other network tasks like activating eth0.
>
> The other one is a Windows 95 box where i plugged in a
> 3com Fast Etherlink XL 3C905B-TX PCI yesterday.
> I installed the Winblows drivers and windows has no problems at all
> (the hardware manager tells me, the card is ready to work, no conflicts).
> Then i installed the M$ TCP/IP driver, told it about its
> IP address (192.168.1.2), my gateway
> (i.e. my Linux ALpha machine, 192.168.1.1) and the netmask 255.255.255.0.
>
> If i now do a ping from Win to Linux, i get connection timed out;
> the other way round, i get no output at all.
>
> The output of ifconfig eth0 looks reasonable to me,
> but every ping increases the value
> in the errors entry of the TX packets line.
Meaning that the packet either could not be sent (serious HW problem!)
or wasn't ACK'ed from the peer (cabling or wrong IP address config?).
> The driver reports (/var/log/messages), that it negotiated a
> 100baseFx-FD value for the transmission parameters.
> No error output here, either.
You may have to look into the driver to enable debug messages.
> There is no HUB in between the 2 machines.
Point-to-point connection?
If so, are you using a straight-through cable (wrong!) or a cross-over cable?
> So why on earth, doesn't this thing work, when there is no error, neither
> on Linux nor on Win side?
Sure there is an error.
> Could anybody give me a hint?
> How do i debug the problem - is it on Linux
> or on the Windows side?
Or maybe in-between - check the cabling.
--
Anders Larsen
------------------------------
From: "daewoois" <[EMAIL PROTECTED]>
Subject: plz....help me.....how can i do?
Date: Thu, 23 Mar 2000 17:23:35 +0900
Now i learn the Linux_kernel................
but...
i want to make an other protocol....not TCP/IP
many problems make my headache....T.T
i think that......i make my device driver for my protocol....
is it right?
and then install my protocol module into kernel....
is it right?
i don't want to touch the TCP/IP protocl...
my thinking is right?
if not right, how can i do for this protocol.....
PLZ...help me.....
------------------------------
From: [EMAIL PROTECTED] (Martin Kahlert)
Crossposted-To: comp.os.linux.networking,comp.os.linux.alpha
Subject: Re: Netproblems on Linux AXP and Win95
Date: 23 Mar 2000 08:29:41 GMT
Reply-To: [EMAIL PROTECTED]
[Posted and mailed]
Hi!
Thanks for your really quick response :-)
That's support! Great!
In article <[EMAIL PROTECTED]>,
Anders Larsen <[EMAIL PROTECTED]> writes:
> Tulip and de4x5 aren't even remotely related.
I cite from linux/Documentation/Configure.help:
(If your card is NOT SMC EtherPower 10/100 PCI (smc9332dst),
you can also try the driver for "Generic DECchip" cards, above.
(which is CONFIG_DE4X5).
> Which kernel version are you using?
2.2.14
[...]
>> The output of ifconfig eth0 looks reasonable to me,
>> but every ping increases the value
>> in the errors entry of the TX packets line.
> Meaning that the packet either could not be sent (serious HW problem!)
> or wasn't ACK'ed from the peer (cabling or wrong IP address config?).
Perhaps it helps, if i mention, that the Win-boxes LEDs light up
while pinging from Linux.
>> There is no HUB in between the 2 machines.
>
> Point-to-point connection?
> If so, are you using a straight-through cable (wrong!) or a cross-over cable?
I ordered a cross over cable. How could i check that?
>> Could anybody give me a hint?
>> How do i debug the problem - is it on Linux
>> or on the Windows side?
> Or maybe in-between - check the cabling.
--
The early bird gets the worm. If you want something else for
breakfast, get up later.
------------------------------
From: [EMAIL PROTECTED] (Casper H.S. Dik - Network Security Engineer)
Crossposted-To: comp.unix.programmer
Subject: Re: poll()ing TCP sockets - when can I expect a POLLHUP?
Date: 23 Mar 2000 08:40:13 GMT
[[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]
Andi Kleen <[EMAIL PROTECTED]> writes:
>> Bug.
>Single Unix does not require it:
> POLLIN
> Data other than high-priority data may be read without
> blocking. For STREAMS, this flag is set in revents even if the
> message is of zero length.
I'm referrng to setting POLLHUP, that is a bug (if the local
end isn't closed)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
------------------------------
From: Mats Byggmastar <[EMAIL PROTECTED]>
Subject: Re: UART overrun errors
Date: Thu, 23 Mar 2000 09:54:50 +0100
Johan Kullstam wrote:
> > >
> > >Does this mean that the interrupt that the 16550A issues isn't served by
> > >the driver in time and the FIFO gets overwritten?
> > >
> > A couple of things:
> >
> > - the UART generates an interrupt when data is ready...
> > (Possibly, when there are certain number of bytes in
> > its FIFO buffer)...the OS prioritizes the interrupts
> > so a DISK interrupt may have priority over a serial port
> > interrupt... but we can't control that...
>
> oh yes we can. irqtune.
irqtune ?
Is that some kernel internal stuff or can it be accessed from
user space?
The funny thing with my problem is that I once was on a seminar
with Linus himself where he talked about (among other things) how
Linux can receive "hi-speed" serial data without problem as the
kernel was built in a smart way...
Well, now my system (Pentium III) runs 98% idle and can't receive
64 kbit/s over a serial port.
Mats
------------------------------
From: Anders Larsen <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.alpha
Subject: Re: Netproblems on Linux AXP and Win95
Date: Thu, 23 Mar 2000 09:48:41 +0100
[Posted and mailed]
Martin Kahlert wrote:
> Thanks for your really quick response :-)
> That's support! Great!
>
> In article <[EMAIL PROTECTED]>,
> Anders Larsen <[EMAIL PROTECTED]> writes:
> > Tulip and de4x5 aren't even remotely related.
> I cite from linux/Documentation/Configure.help:
> (If your card is NOT SMC EtherPower 10/100 PCI (smc9332dst),
> you can also try the driver for "Generic DECchip" cards, above.
> (which is CONFIG_DE4X5).
Just re-read it, you're right <:-o
> > Which kernel version are you using?
> 2.2.14
I am not familiar with the Alpha, but the 2.2.14 x86 kernel has *two*
distinct drivers for the DEC tulip (in addition to the DE4x5).
You need the one or the other depending on the *exact* version of the
21x4x chip on the network card.
Please try both (if they're available for the Alpha).
(CONFIG_DEC_ELCP and CONFIG_DEC_ELCP_OLD, resp.)
> >> The output of ifconfig eth0 looks reasonable to me,
> >> but every ping increases the value
> >> in the errors entry of the TX packets line.
> > Meaning that the packet either could not be sent (serious HW problem!)
> > or wasn't ACK'ed from the peer (cabling or wrong IP address config?).
> Perhaps it helps, if i mention, that the Win-boxes LEDs light up
> while pinging from Linux.
Then the cable is probably okay.
Do the Linux-box's LEDs light up when you ping from Windos?
> >> There is no HUB in between the 2 machines.
> >
> > Point-to-point connection?
> > If so, are you using a straight-through cable (wrong!) or a cross-over cable?
> I ordered a cross over cable. How could i check that?
By watching the LEDs (as you already did).
--
Anders Larsen
------------------------------
From: [EMAIL PROTECTED] (Jos Hulzink)
Subject: Re: Can i get a MAC address ?
Date: 23 Mar 2000 09:22:18 GMT
In article <8aak73$vk3$[EMAIL PROTECTED]>, Peter Samuelson wrote:
>etc., ad nauseum. Use the source, people -- *that's why it's there.*
>
>I get the feeling there are a *lot* of folks out there that simply *do
>not get* the open source movement. This is really what it's all about:
>being able to reuse someone else's work, being able to take advantage
>of the fact that someone else already looked up or figured out the
>answer you need. For what I do on a typical day, the "huge library of
>working example code" aspect of open source is often as important as
>the "can make local modifications" aspect, which is what seems to get
>most ot the attention.
There are many cases where walking through the sourcecode is a nasty job
-ever walked through the kernel?-, and the friendly and short answer can do
the job too. But I get the feeling there are a *lot* of folks out here that
are on this list to irritate themselves about the many questions asked. Not
only open source was a feature of linux, also the good and helpful
mailinglists.
If you don't want to read this, get out of here.
Jos
------------------------------
From: Josef Moellers <[EMAIL PROTECTED]>
Subject: aic7xxx and AIC-7892/9 /Ultra160)
Date: Thu, 23 Mar 2000 10:45:25 +0100
There is a problem with the current version of the sequencer code in
that it disables the SCSI subsystem too early at the end of the command
phase resulting in a violation of the low level SCSI protocol:
+---------------+
REQ --------+ +-------
+---+
ACK ----------------+ +-----------
I discussed this with the original author of the aic7xxx driver for
FreeBSD, Justin Gibbs and he sent me the following patch:
--- aic7xxx.seq.org Thu Mar 23 10:32:20 2000
+++ aic7xxx.seq Thu Mar 23 10:32:38 2000
@@ -729,6 +729,7 @@
test SSTAT0, SDONE jnz p_command_ultra2_dma_done;
test SSTAT1,PHASEMIS jz p_command_dma_loop; /* ie.
underrun */
p_command_ultra2_dma_done:
+ test SCSISIGI, ACKI jnz .;
and DFCNTRL, ~HDMAEN;
test DFCNTRL, HDMAEN jnz .;
and DFCNTRL, ~SCSIEN;
This is in the 2.3.44 kernel (the latest I have access to) as well as
the 2.2.14 production kernel.
Josef
-- =
Josef M=F6llers
Fujitsu Siemens Computers
SHV Server DS 1
------------------------------
From: Etienne Lorrain <[EMAIL PROTECTED]>
Subject: Re: UART overrun errors
Date: Thu, 23 Mar 2000 09:43:07 +0000
Mats Byggmastar wrote:
>
> Johan Kullstam wrote:
> > >
> > > - the UART generates an interrupt when data is ready...
> > > (Possibly, when there are certain number of bytes in
> > > its FIFO buffer)...the OS prioritizes the interrupts
> > > so a DISK interrupt may have priority over a serial port
> > > interrupt... but we can't control that...
> >
> > oh yes we can. irqtune.
>
> irqtune ?
At the hardware level, the interrupt priority is:
0 1 8 9 10 11 12 13 14 15 3 4 5 6 7 (least priority)
You cannot change that on a PC.
> The funny thing with my problem is that I once was on a seminar
> with Linus himself where he talked about (among other things) how
> Linux can receive "hi-speed" serial data without problem as the
> kernel was built in a smart way...
> Well, now my system (Pentium III) runs 98% idle and can't receive
> 64 kbit/s over a serial port.
>
> Mats
Yes, sure. But there is a bug in some IDE hardware drivers, you
have to disable interrupts when tranfering bytes to/from Hard Drives
else you get physical corruption of data. So by default, the IDE
driver disable interrupts quite a long time. This problem was
just on very few PCs, and a long time ago, but was not detectable
so by default, Linux configuration disable interrupts.
In short, use "hdparm" to enable interrupts when tranfering data
to/from HD, in your rc scripts.
Etienne.
------------------------------
From: Josef Moellers <[EMAIL PROTECTED]>
Subject: Re: UART overrun errors
Date: Thu, 23 Mar 2000 11:09:13 +0100
Etienne Lorrain wrote:
> =
> Mats Byggmastar wrote:
> >
> > Johan Kullstam wrote:
> > > >
> > > > - the UART generates an interrupt when data is ready...
> > > > (Possibly, when there are certain number of bytes in
> > > > its FIFO buffer)...the OS prioritizes the interrupts
> > > > so a DISK interrupt may have priority over a serial port
> > > > interrupt... but we can't control that...
> > >
> > > oh yes we can. irqtune.
> >
> > irqtune ?
> =
> At the hardware level, the interrupt priority is:
> 0 1 8 9 10 11 12 13 14 15 (least priority)
> You cannot change that on a PC.
This is not quite right.
If you take the list you give, you can split it into two and exchange
the two parts.
The 8259 has several "End-Of-Interrupt Operation" modes, two of wich
are:
- Automatic Rotation, where the last interrupt serviced will
automatically get the least priority and
- Specific Rotation, where you can specify which interrupt will get the
least priority.
E.g. when you start with your sequence and you get an interrupt on IRQ
12, when using Automatic Rotation, you end up with
0 1 13 14 15 8 9 10 11 12 3 4 5 6 7 =
The specific relative priority comes from the fact that there are two
8-channel interrupt controllers that are cascaded: the second one is
hooked to IRQ2 if the first one.
-- =
Josef M=F6llers
Fujitsu Siemens Computers
SHV Server DS 1
------------------------------
From: "Martin Sinot" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.admin,comp.os.linux.advocacy,comp.os.linux.alpha,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.misc,comp.os.linux.networking,co
Subject: Re: QUESTION: Linux Multiprocessor management
Date: Thu, 23 Mar 2000 09:38:06 GMT
Christian Langis <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>We see that the usertime column increases with different increments (30,
15, 21, 14...) >hile the system time increases as well with more regular
increments. However, it is >disturbing to see such high system time (and
order more than user time ?!) since our test >program doesn't use the system
at all. The last observation is that the total of both >increases very
regularly.
>Why???
Your program DOES use the system. And fairly heavily, too. It prints, and
printing things is a pretty heavy operation. So what you see is true - most
of the time is indeed spent in the system. Try collecting the results and
print everything at once at the end of the simulation, and you will see
that the system times almost drop to zero.
Oh, and would you please send postings in plain text instead of rich text
please?
--
Martin Sinot
Nijmegen, Netherlands
[EMAIL PROTECTED]
------------------------------
From: Nathan Krasney <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.programmer,comp.unix.questions,comp.unix.solaris
Subject: ioctl block pthread_create ??
Date: Thu, 23 Mar 2000 12:52:01 +0200
Hello all
i have very strange phenomenon. i have a user mode application that talk
to a file system driver via ioctl. this ioctl is envoked by user mode
thread , problem is that creation of a new thread is suspended by the os
until this ioctl return !!!
e.g.
main()
{
pthread_create run f1()
pthread_create run ioctl
pthread_create run f2()
while(1)
......;
}
f1 will continue to run independently but f2 will NOT start running
until the ioctl has return !!
what can possibly be the problem ???
Nathan
------------------------------
From: Mathias Waack <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.programmer,comp.unix.questions,comp.unix.solaris
Subject: Re: ioctl block pthread_create ??
Date: Thu, 23 Mar 2000 12:43:17 +0100
Nathan Krasney wrote:
> i have very strange phenomenon. i have a user mode application that talk
> to a file system driver via ioctl. this ioctl is envoked by user mode
> thread , problem is that creation of a new thread is suspended by the os
> until this ioctl return !!!
>
> e.g.
> main()
> {
> pthread_create run f1()
> pthread_create run ioctl
> pthread_create run f2()
> while(1)
> ......;
> }
>
> f1 will continue to run independently but f2 will NOT start running
> until the ioctl has return !!
Maybe the ioctl locks the kernel. Then the pthread_create (which
calls clone()) stalls until ioctl finishes.
Thats the situation under Linux, I don't know how it is or could be
under Solaris.
HTH
Mathias
--
Mathias Waack | [EMAIL PROTECTED]
Tel.: +49 621 181 2717 Fax.: +49 621 181 2713
------------------------------
From: [EMAIL PROTECTED] (Geoff Lane)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Absolute failure of Linux dead ahead?
Date: Thu, 23 Mar 2000 12:22:41 +0000
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Christopher Browne) writes:
> a) It defines a single pass compiler;
> b) It provides no way of splitting projects coherently into multiple
> files;
> c) The typing system doesn't cope all that well with dynamic arrays.
Not using modern Pascal compilers then? All these "problems" have been
addressed. Not as cleanly as ModularN or Oberon but people appear to
happier with pragmatic hacks than logical simplicity (cf previous arguments
about c/c++ :-))
--
/\ Geoff. Lane. /\ Manchester Computing /\ Manchester /\ M13 9PL /\ England /\
Today's Excuse: CPU needs re-calibration
------------------------------
From: Gerd Buesken <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Zero padding in sprintf() doesn't work for strings - HELP !
Date: Thu, 23 Mar 2000 14:02:59 +0100
Hi
I have a problem using gcc 2.95 running on a 2.2.5-15 Red Hat linux.
Regarding sprintf, the man pages say:
0 specifying zero padding. For all conversions except n, the
converted value is padded on the left with zeros rather than blanks.
If a precision is given with a numeric conversion (d, i, o, u, i, x,
and X), the 0 flag is ignored.
My source looks like this:
include <stdio.h>
main()
{
char buff[30];
char LenBuf[30];
strcpy(LenBuf,"1");
sprintf (buff,"%06s", LenBuf);
printf( "buffer=%s; Length %06d\n",buff, strlen(LenBuf) );
}
This is the Linux output:
buffer= 1; Length 000001
This is the HP-UX output:
buffer=000001; Length 000001
It seems as if the output on Linux is wrong(using Blanks instead of zeros).
Does anybody has a solution other than rewriting the code ?
Thanks
Gerd
------------------------------
From: James Moger <[EMAIL PROTECTED]>
Subject: Re: Question for Embedded Linux
Date: Thu, 23 Mar 2000 08:43:04 -0500
Reply-To: [EMAIL PROTECTED]
Check out Hard Hat Linux
http://www.mvista.com
--
James Moger
===---------------------===
Software Engineer
Transonic Systems, Inc.
{C/C++} {Java} {Smalltalk}
Cornell Engineering '98
===---------------------===
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Absolute failure of Linux dead ahead?
Reply-To: [EMAIL PROTECTED]
Date: Thu, 23 Mar 2000 13:45:14 GMT
Centuries ago, Nostradamus foresaw a time when Geoff Lane would say:
>In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] (Christopher Browne) writes:
>> a) It defines a single pass compiler;
>> b) It provides no way of splitting projects coherently into multiple
>> files;
>> c) The typing system doesn't cope all that well with dynamic arrays.
>
>Not using modern Pascal compilers then? All these "problems" have been
>addressed. Not as cleanly as ModularN or Oberon but people appear to
>happier with pragmatic hacks than logical simplicity (cf previous arguments
>about c/c++ :-))
No, I've not used Pascal lately. Not since I ran a CS lab in the late
'80s, running TP.
If you avoid those issues, you are still likely to jump onto the horns
of "problem d):"
a)-c) aren't defined in a formally standardized way that can be
considered portable.
--
The people's revolutionary committee has decided that the name "e" is
retrogressive, unmulticious and reactionary, and has been flushed.
Please update your abbrevs.
[EMAIL PROTECTED] - - <http://www.hex.net/~cbbrowne/lsf.html>
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Crossposted-To:
comp.os.linux.powerpc,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: gcc problems
Date: 23 Mar 2000 13:56:33 GMT
[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
Someone's already answered problem number 2, but:
> I am having trouble with the postgres configure program, which requires gcc
>to be able to compile programs. As configure was performing its checks, it
>failed when it claimed, 'gcc is not able to compile executables'.
> OK, this is problem number 1.
If installing glibc-devel doesn't help, try installing the binutils
package. (Though I'd have hoped Red Hat's gcc package would have pulled
this in by default.)
--
Colin Watson [[EMAIL PROTECTED]]
"... a good part of the remainder of my life was going to be
spent in finding errors in my own programs." - Maurice Wilkes
------------------------------
** 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
******************************
