Linux-Development-Sys Digest #85, Volume #8 Fri, 18 Aug 00 10:13:11 EDT
Contents:
Re: Kernel 2.4.0-test6!!!!! Anyone have any Problems? (John Gluck)
condition variables in the kernel? (Brian Horton)
Re: condition variables in the kernel? (Kaz Kylheku)
Re: Kernel 2.4.0-test6 Compile and install module problem? (Paul Kimoto)
Re: Kernel 2.4.0-test6!!!!! Anyone have any Problems? (Michael Meding)
Re: Deep TCP / RAW IP socket questions ("Dave Rhodes")
PCI Initialization of Cards buggy or I miss something? (Nikos Kalogridis)
Re: Misterious Hangs in my network (Bodo Eggert)
task queue ("Peter Huang")
sleep in kernel ("Peter Huang")
Re: debugging/profiling multithreaded (pthread) applications on RedHat 6.x? (Hammer)
Re: doing mmap() twice on the same fd ([EMAIL PROTECTED])
Re: soft-realtime system with standard linux (not RT-Linux)? (Jean-Francois
MOINE)
Re: task queue (Ratz)
UID width (Kevin Kaichuan He)
Re: sleep in kernel (Ratz)
Re: sleep in kernel (Ratz)
Re: doing mmap() twice on the same fd (Mattias Engdeg�rd)
Re: all threads in a process share the same pid? (Xavier Leroy)
IRDA-Support on Sparc (Robert Resch)
(no subject) ("Robichaud, Jean-Philippe [BAN:6S33:EXCH]")
Interesting kernel configuration question ("Robichaud, Jean-Philippe
[BAN:6S33:EXCH]")
Re: IRDA-Support on Sparc (Paul Dwerryhouse)
----------------------------------------------------------------------------
From: John Gluck <[EMAIL PROTECTED]>
Subject: Re: Kernel 2.4.0-test6!!!!! Anyone have any Problems?
Date: Thu, 17 Aug 2000 15:57:49 -0400
E-mu wrote:
> What in the world are you all talking about. Thank God I Beta Test for MS
> and not Linux full time with these kinda responces, I refuse to try and help
> linux OS take hold.
>
> I put a COUPLE OF !!!! and everyone is talking about spam and gyros etc.
> What in the world is the problem here?
>
> Anyhow I will refrain from using !!! From now on, jeese almighty already!
>
[snip]
Just an attempt to give you a bit of info... It was well meaning even if it
wasn't what you were looking for.
Linux newsgroups tend to have people that are more savy than MS newsgroups.
They get a bit sidetracked at times.
The cure is to ignore what doesn't interest you.
In MS newsgroups you can very often just ignore everything since it's mostly
just plain wrong.
--
John Gluck (Passport Kernel Design Group)
(613) 765-8392 ESN 395-8392
Unless otherwise stated, any opinions expressed here are strictly my own
and do not reflect any official position of Nortel Networks.
------------------------------
From: Brian Horton <[EMAIL PROTECTED]>
Subject: condition variables in the kernel?
Date: Thu, 17 Aug 2000 11:34:23 -0500
I see that the kernel has support for semaphores; What about condition
variables? Is there 'direct' support for condition variables? Or do I
have to write my own stuff? (Or is there some code out there that I can
use that already does that?)
thx.bri.
------------------------------
From: [EMAIL PROTECTED] (Kaz Kylheku)
Subject: Re: condition variables in the kernel?
Reply-To: [EMAIL PROTECTED]
Date: Thu, 17 Aug 2000 21:35:35 GMT
On Thu, 17 Aug 2000 11:34:23 -0500, Brian Horton <[EMAIL PROTECTED]> wrote:
>I see that the kernel has support for semaphores; What about condition
>variables? Is there 'direct' support for condition variables? Or do I
>have to write my own stuff? (Or is there some code out there that I can
>use that already does that?)
See http://users.footprints.net/~kaz/lmc.html
--
Any hyperlinks appearing in this article were inserted by the unscrupulous
operators of a Usenet-to-web gateway, without obtaining the proper permission
of the author, who does not endorse any of the linked-to products or services.
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: Kernel 2.4.0-test6 Compile and install module problem?
Date: 17 Aug 2000 17:42:44 -0500
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, John Gluck wrote:
> Paul Kimoto wrote:
>> There have been newer modutils releases to
>> take care of that.
> That's wonderful information... It would be useful if you also said what
> versions of modutils and where to find them.
See Documentation/Changes.
(Note, however, <[EMAIL PROTECTED]>'s message that his umsdos problems were
not helped.)
--
Paul Kimoto
Disclaimer: Other than explicit citations of URLs, hyperlinks appearing
in this article have been inserted without the permission of the author.
------------------------------
From: Michael Meding <[EMAIL PROTECTED]>
Subject: Re: Kernel 2.4.0-test6!!!!! Anyone have any Problems?
Date: Thu, 17 Aug 2000 23:03:16 +0200
Reply-To: [EMAIL PROTECTED]
Hi there,
> In MS newsgroups you can very often just ignore everything since it's mostly
> just plain wrong.
That was one of the funniest.....
>
> --
> John Gluck (Passport Kernel Design Group)
>
> (613) 765-8392 ESN 395-8392
>
> Unless otherwise stated, any opinions expressed here are strictly my own
> and do not reflect any official position of Nortel Networks.
------------------------------
From: "Dave Rhodes" <[EMAIL PROTECTED]>
Subject: Re: Deep TCP / RAW IP socket questions
Date: Thu, 17 Aug 2000 20:01:47 -0400
> The problem is that kernel TCP sees the packets going to your raw socket
too,
> and will usually interfere, sending RSTs or creating its own connections.
>
> One way to prevent this is to use a packet socket (not raw socket) and
> create a firewall rule for the port/address pair of the connection on the
fly.
> This works because the packet sockets sits below the IP firewall
machinery,
> and TCP sits above it.
I had the impression that packet sockets were not preferred for use anymore?
Anyway, when these packets are sent do they go both 'up and down' the chain
(and the firewall rule is needed to stop it going back up the local com
chain)?
Another thing that I'd like to do is "intercept" and process outgoing IP
packets just
before they are sent (for TCP and UDP connections), this might include
wanting to
stop them from going out and/or re-inserting... is this an simple ipchains
application? According to the divert information they cite issues of not
being able
to control the "direction" (w/r/t the com stack) of packets using other
techniques.
Apologies for not being more familiar with Linux (ex Solaris programmer :-)
--thanks a great deal!
--dave
------------------------------
From: Nikos Kalogridis <[EMAIL PROTECTED]>
Subject: PCI Initialization of Cards buggy or I miss something?
Date: Fri, 18 Aug 2000 02:45:14 +0100
Hi I am currently working on a driver for a pci device. I am using the
2.4-test6 and RH6.2.
I have an ASUS CUV4X with a Celeron 633MHz on it. The BIOS is the Award
Medallion 6.0.
I encountered the following problems when the PNP OS option in the BIOS
is enabled. Although the card is found in the pci bus and is initialized
using the pci_enable_device according to the pci.txt in the kernel docs
the card is not fully functional. i.e working for a second or so and
then halts the application that uses the driver. There are no segfaults
produce neither oops in the kernel. Everything seems absolutely normal
in the kernel logs. And if I terminate the application and start it
again it agains works for a second or two and then halts. Is this an
initialization bug in the PCI support of the kernel? I noticed the same
problem in the bttv driver as well.
Does anyone has any clues on what it could be the problem here. (Apart
from telling me to disable the option in the BIOS which by the way if I
do the card works just fine). From my point of view is should be a bug
in the PCI code.
Thanks in advance
Nikos Kalogridis
email: [EMAIL PROTECTED]
------------------------------
From: Bodo Eggert <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.suse,comp.os.linux.networking,comp.os.linux.setup
Subject: Re: Misterious Hangs in my network
Date: Fri, 18 Aug 2000 05:36:27 +0200
Reply-To: [EMAIL PROTECTED]
Andres Tarallo wrote:
>
> I have made the following installation for a high school were i work for and
> i play the role of a System Administrator.
>
> I have a Pentium MMX @ 166 Mhz with 64 MB RAM acting as NIS Server and NFS
> server. I'm exporting the users home directories to the Computer LAB; In the
> computer lab I have workstations ranging from K5 to Celeron @433 Mhz; with
> an average of 32 Mb of RAM at least (Many have 64 Mb of RAM).
>
> During some clases we had a misterious hang that all the network frezeed, we
> rebooted and then we continued working without problem. WE don't have
> harware problem; we even changed a disk that seems to be faulty; but the
> problem persists.
>
> We're currently working with SuSE 6.4 both in server and workstations; the
> server si running kernel NFS; I'm aware of that this is still under
> development but I hva e to run on the workstations StarOffice and this was
> the only solution I've found
>
> Thanks for any advice
> Andres Tarallo
> [EMAIL PROTECTED]
the kernel nfs does not have locking, but it supports locking.
this will break staroffice.
you can either use userspace-nfs or upgrade your kernel and hope
it is fixed by now.
--
/-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-\
|*-=>www.7eggert.de<=-*-=>[EMAIL PROTECTED]<=-*|
\-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-/
------------------------------
From: "Peter Huang" <[EMAIL PROTECTED]>
Subject: task queue
Date: Thu, 17 Aug 2000 21:18:34 -0700
Hi all
How do I distinuish whether the task queue is running in the interrupt time
or in the regular time?
Peter
------------------------------
From: "Peter Huang" <[EMAIL PROTECTED]>
Subject: sleep in kernel
Date: Thu, 17 Aug 2000 21:14:33 -0700
Hi all
Is there such thing as a sleep() function in the kernel.
Peter
------------------------------
From: Hammer <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: debugging/profiling multithreaded (pthread) applications on RedHat 6.x?
Date: Fri, 18 Aug 2000 05:41:03 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
[snip]
> To be one the safe side, I would only turn it on in one
>thread of interest at a time.
Ouch!
> The mcount function is not exactly thread safe.
Ouch again.
> To do this, I increase the size of the corefile[] local array in
> the elf_core_dump function fs/binfmt_elf.c.
>
> sprintf(corefile, "core.%d", (int) current->pid);
>
> Then I just nuke the atomic_read(¤t->mm->count) != 1 test.
This is a great post, thanks for sharing. Saved me some classes at
the "School of Hard Knocks".
-=hammer
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: doing mmap() twice on the same fd
Date: Fri, 18 Aug 2000 05:54:20 GMT
On Wed, 16 Aug 2000 16:11:02 +0200 Kasper Dupont <[EMAIL PROTECTED]> wrote:
| [EMAIL PROTECTED] wrote:
|>
|> Is it possible to do mmap() twice on the same fd, but as a different
|> number produced from dup(), where one or both is passed to a different
|> process, when the fd was opened to /dev/zero, and get both of those
|> processes to share memory? I don't want this memory to be part of a
|> file anywhere. And what I do want to do is make a pipe from one process
|> to another process which dup()'s the fd and passes it over the pipe to
|> the connecting process which will then mmap() on it. Note that the
|> connecting process is not a descendant of the daemon process. Is this
|> at all doable on even some Unix systems to get autonomous shared memory
|> in a secure way?
|>
|
| Try looking at the man pages for shmget, shmat and
| shmctl. I think they can do what you want.
|
| Notice:
| I think most kernels do have this feature compiled
| in, but if it's missing you might need to recompile
| the kernel. You must remember to mark shared memory
| segments for removal with shmctl, I believe it is
| normally the best choice to do that immediately
| after calling shmat.
No, shm* functions don't do what I want. I want absolute security with
no chance an unauthorized process can get access.
No offer to answer the original question?
--
| Phil Howard - KA9WGN | My current websites: linuxhomepage.com, ham.org
| phil (at) ipal.net +----------------------------------------------------
| Dallas - Texas - USA | [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Jean-Francois MOINE)
Subject: Re: soft-realtime system with standard linux (not RT-Linux)?
Date: 18 Aug 2000 07:08:38 GMT
wolfgang guldner <[EMAIL PROTECTED]> a skrivas:
>Hello everyone,
Hello Wolfgang,
>I have got a question concerning how to realize a soft-realtime system
>with standard linux (not RT-Linux).
[snip]
>Did anyone implement a soft-realtime under normal Linux and can give me
>tips which ** configuration (daemons running, crontab entries,
>kernel-compilation ...) ** optimizes the delay times/ performance
>of the application?
Have a look at the low-latency patches:
http://people.redhat.com/mingo/
and
http://www.uow.edu.au/~andrewm/linux/schedlat.html
--
Ken ar c'henta� ** Breizh ha Linux atav ! **
mailto:[EMAIL PROTECTED]
Jef (home mailto:[EMAIL PROTECTED])
http://moinejf.free.fr/
------------------------------
From: Ratz <[EMAIL PROTECTED]>
Subject: Re: task queue
Date: Fri, 18 Aug 2000 10:11:25 +0200
Peter Huang wrote:
>
> Hi all
>
> How do I distinuish whether the task queue is running in the interrupt time
> or in the regular time?
>
> Peter
Hi Peter,
I'm not sure if this solves your problem, but have a
look at /usr/src/linux/include/linux/sched.h . There
is the struct task_struct{ ... } that has the status
flag defined as volatile long state;
There is also the function extern void trap_init(void);
that should give you the desired information. But as
I mentioned, I'm not sure if I correctly understood
your question.
regards,
Roberto Nibali, ratz
--
mailto: `echo [EMAIL PROTECTED] | sed 's/[NOSPAM]//g'`
------------------------------
From: Kevin Kaichuan He <[EMAIL PROTECTED]>
Subject: UID width
Date: Fri, 18 Aug 2000 01:28:30 -0700
Hi, I faced a problem when trying to mount Solaris NFS
export to linux. My UID in Solaris is larger than 65535, but
it looks like linux's uid width is 16bit , so that it's impossible
for me to set up correct UID to mount the Solaris filesystem.
Is there a way to get around this problem? Thank you!
Kevin
------------------------------
From: Ratz <[EMAIL PROTECTED]>
Subject: Re: sleep in kernel
Date: Fri, 18 Aug 2000 10:38:59 +0200
Peter Huang wrote:
>
> Hi all
>
> Is there such thing as a sleep() function in the kernel.
>
> Peter
Hi Peter,
how about sleep_on() or interruptable_sleep_on() ?
section 9 of man-pages may help you or again:
/usr/src/linux/include/linux/sched.h
HTH,
Roberto Nibali, ratz
--
mailto: `echo [EMAIL PROTECTED] | sed 's/[NOSPAM]//g'`
------------------------------
From: Ratz <[EMAIL PROTECTED]>
Subject: Re: sleep in kernel
Date: Fri, 18 Aug 2000 10:39:40 +0200
Peter Huang wrote:
>
> Hi all
>
> Is there such thing as a sleep() function in the kernel.
>
> Peter
Hi Peter,
how about sleep_on() or interruptable_sleep_on() ?
section 9 of man-pages may help you or again:
/usr/src/linux/include/linux/sched.h
HTH,
Roberto Nibali, ratz
--
mailto: `echo [EMAIL PROTECTED] | sed 's/[NOSPAM]//g'`
------------------------------
From: [EMAIL PROTECTED] (Mattias Engdeg�rd)
Subject: Re: doing mmap() twice on the same fd
Date: 18 Aug 00 08:40:21 GMT
> Is it possible to do mmap() twice on the same fd, but as a different
> number produced from dup(), where one or both is passed to a different
> process, when the fd was opened to /dev/zero, and get both of those
> processes to share memory?
No. mmap() on /dev/zero is synonymous to mmap() with MAP_ANON and always
gives you a fresh block of anonymous memory (copy-on-write).
Recent Linux development kernels allow you to mmap anonymous memory
as shareable, but you can't send a mapping to an unrelated process that
way. (I agree that it would be a neat trick.)
You have to map a temporary file. This isn't as bad as it sounds:
The asynchronous write policy of Linux filesystems makes it almost as
efficient as anonymous memory, and if you immediately unlink the file
after creating it, it won't survive your processes either.
------------------------------
From: Xavier Leroy <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: all threads in a process share the same pid?
Date: 18 Aug 2000 13:10:04 +0200
[EMAIL PROTECTED] (Linus Torvalds) writes:
> >The standard states that all threads should share the same pid, and
> >different thread ids.
> Easily fixed:
> [Cache PID of initial thread, override getpid()]
> If the pthreads library doesn't do this already, then it should.
I considered doing this early in the LinuxThreads development, and decided
not to for the reasons given by other posters: problems with wait()
and signals would still be there, perhaps made even more obscure
by this "solution" (e.g. getppid() in a child process would not return the
same PID as getpid() in its parent...).
The only solution, in my opinion, is to fix the kernel so that the CLONE_PID
option to clone() works. That would fix all remaining non-conformance
issues for LinuxThreads.
Oh, while we're at it, I believe (minimal) kernel support is also
needed to implement process-shared mutexes, conditions and semaphores.
- Xavier Leroy
--
Valid e-mail address (without the underscores): [EMAIL PROTECTED]
This is a protection against junk mail. Apologies for the inconvenience.
Home page: http://pauillac.inria.fr/~xleroy/
------------------------------
From: Robert Resch <[EMAIL PROTECTED]>
Subject: IRDA-Support on Sparc
Date: Fri, 18 Aug 2000 15:36:18 +0200
Hi!
Why can't i select the IRDA-Support on my SparcStation2 and RedHat Linux
6.2 (Kernel 2.2.16) ?
i have a external IRDA-Box which is being connected to the host via
RS232...
(Tekram)
Robert Resch
------------------------------
From: "Robichaud, Jean-Philippe [BAN:6S33:EXCH]" <[EMAIL PROTECTED]>
Subject: (no subject)
Date: Fri, 18 Aug 2000 09:33:58 -0400
--
Jean-Philippe Robichaud
[EMAIL PROTECTED]
(514) 818-7750
(ESN) 888-7750
St-Laurent, Quebec, Canada
------------------------------
From: "Robichaud, Jean-Philippe [BAN:6S33:EXCH]" <[EMAIL PROTECTED]>
Crossposted-To:
linux.redhat.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.networking,linux.redhat.misc
Subject: Interesting kernel configuration question
Date: Fri, 18 Aug 2000 09:49:25 -0400
Hi everybody
This message is posted in a lot of newsgroup because it touch a lot of
them. Here is my problem :
I have created a ramdisk so it can be a root partition (/dev/* /proc
and so on). How can I tell the kernel to use the ramdisk as the root
partition ? I've try to create a nod with both major and minor numbers
at 0 and do a rdev vmlinuz dummy, but it fails. All I want is to get my
kernel booting diskless. I send the kernel and the ramdisk via pxe and
everything works fine till the kernel try to mount its root partition.
Again, how can I tell the kernel to use the ramdisk ? (rdev vmlinuz
/dev/ram0 also failed).
Thanks a lot for your help
jean-philippe
--
Jean-Philippe Robichaud
[EMAIL PROTECTED]
(514) 818-7750
(ESN) 888-7750
St-Laurent, Quebec, Canada
------------------------------
From: [EMAIL PROTECTED] (Paul Dwerryhouse)
Subject: Re: IRDA-Support on Sparc
Date: 18 Aug 2000 14:07:06 GMT
Robert Resch <[EMAIL PROTECTED]> writes:
>Why can't i select the IRDA-Support on my SparcStation2 and RedHat Linux
>6.2 (Kernel 2.2.16) ?
>i have a external IRDA-Box which is being connected to the host via
>RS232...
>(Tekram)
You shouldn't need IRDA support for that - since it's a serial device,
it should be able to be accessed in the same way you'd access any other
serial device - that is, via /dev/ttyS0, /dev/ttyS1, etc...
So, if you've got it on serial port A, and it's talking to, say, a modem
on a mobile phone, whack /dev/ttyS0 into minicom's configuration and fire it
up.
Cheers,
Paul.
--
"The growing use of e-mail, not to mention Web-page | Paul Dwerryhouse
publishing, threatens to reverse the trend towards |
illiteracy among the supposedly educated without at the | Amsterdam
same time improving their spelling" -Michael Swaine,DDJ | The Netherlands
------------------------------
** 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
******************************
