Linux-Development-Sys Digest #907, Volume #6 Tue, 29 Jun 99 11:14:21 EDT
Contents:
Re: PCI driver blues (Danny Sung)
Re: NT kernel guy playing with Linux (John Hughes)
Re: PTHREADS kernel/user level threading? (James Cownie)
Re: help me please, URGENT (Etienne Lorrain)
Re: PCI driver blues (Danny Sung)
makedev ("Ashutosh S. Rajekar")
Re: Filesize larger than 2 GB on Intel machines an Linux 2.0.36 ("Kent Nilsen")
Re: Why not C++ (Nathan Myers)
Re: TAO: the ultimate OS ("Vladimir Z. Nuri")
Help about Riva TNT driver for linux (kernel 2.0.35) ("Hinomi")
Re: Why not C++ (Isaac)
Re: Why not C++ (Bernd Eckenfels)
Re: Help about Riva TNT driver for linux (kernel 2.0.35) ("Frank V. Castellucci")
Re: Help about Riva TNT driver for linux (kernel 2.0.35) ([EMAIL PROTECTED])
Glade manual? (Guillermo)
Flag for switching back to conservative memory management ? (Oliver Stahlhut)
Flag for switching back to conservative memory management ? (Oliver Stahlhut)
Re: Proposal For New Windowing software (Tristan Wibberley)
where can get (Maximum RPM) this online document? ("robert_c")
Re: Mac-emulation on Linux? ("Brian J. Johnson")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Danny Sung)
Subject: Re: PCI driver blues
Date: Tue, 29 Jun 1999 07:37:09 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 23 Jun 1999 10:49:08 -0400,
Daniel Pietsch <[EMAIL PROTECTED]> wrote:
>It could be that you need to enable PCI memory. To do this set bit 1 (MAR -
>Memory Address Area) of the Command register (aka Instruction register) in
>the PCI configuration space. The Command register has byte offset 4 in the
>PCI configuration space and is 16 bits in size.
Yup, I've done that. I do: (PCI_COMMAND_IO | PCI_COMMAND_MEMORY |
PCI_COMMAND_MASTER), which is what the windows driver for my board
does, from what I can tell.
I've tried with and without MASTER. And with and without IO (all the
IO registers on this boared are memory mapped).
--
--
Danny Sung http://www.poboxes.com/dannys
[EMAIL PROTECTED] (PGP public key available)
Key fingerprint = 0080 1BA8 F145 15AE 5ED1 E4ED C422 C77E 70C6 44FE
------------------------------
From: John Hughes <[EMAIL PROTECTED]>
Subject: Re: NT kernel guy playing with Linux
Date: 29 Jun 1999 09:40:52 +0200
[EMAIL PROTECTED] (Philipp Thomas) writes:
> On 25 Jun 1999 12:49:41 +0200, John Hughes <[EMAIL PROTECTED]> wrote:
>
> >Specially the streams based TCP/IP stacks.
> >
> >Nope. Untrue. FUD.
>
> Not quite :) It's FUD as a global statement and I should have at least
> used an AFAIK when I wrote that. For that part I declare myself guilty
> and apologize.
That's OK.
> On behalf of TCP/IP, I *am* right.
For some small values of right.
> Here's the quote from Larry McVoy <[EMAIL PROTECTED]>, a name that
> should be familiar to you.
Yowee, appeal to authority! :-)
> STREAMS is an elegant modem for handling tty line disciplines, which
> is why it was developed. But it falls apart for high performance
> stuff.
> Here are some issues:
>
> - it wants to queue. Each layer has a put and putnext procedure and
> the way things are done is that the data is enqueued/dequeued even
> when there is nothing on the downstream/upstream queue.
Sorry. He may be smart, but he's *WRONG*. STREAMS will only put
things on a queue if flow control is asserted.
> write()
> put it on the stream head queue
Nope. If the next module down is not flow-controlled it's put routine
will be called. This will repeat all the way to the driver, which
will send the bytes out on the line immediately.
> etc. Completely busted model, it should push through and out the
> door if there is no contention for the device.
It would be a completely busted model if that was what it did. It
doesn't.
> - the module layers are great when you aren't multiplexed. IP is
> multiplexed under TCP, UDP, etc. And there are side channels like
> ARP and ICMP that don't fit well.
It is true that multiplexing is irritatingly hard to do right.
> - the layering hurts performance. It really does. I'd like 10 usec
> round trip ping pongs. Forget about it.
All the layering does (in the absence of a flow control situation) is
add one function call per layer.
--
John Hughes <[EMAIL PROTECTED]>,
Atlantic Technologies Inc. Tel: +33-1-4313-3131
66 rue du Moulin de la Pointe, Fax: +33-1-4313-3139
75013 PARIS.
------------------------------
Date: Tue, 29 Jun 1999 08:54:57 +0100
From: James Cownie <[EMAIL PROTECTED]>
Subject: Re: PTHREADS kernel/user level threading?
Michael Hirsch wrote:
> As far as the kernel is concerned, each
> thread is a separate process.
And he *really* meant it.
Each thread gets a different result from getpid() !
(Another slight deviation from the pthread mandate).
-- Jim
[EMAIL PROTECTED]
------------------------------
From: Etienne Lorrain <[EMAIL PROTECTED]>
Subject: Re: help me please, URGENT
Date: Tue, 29 Jun 1999 10:13:04 +0100
man setvbuf
By default, stdout is line buffered, stderr is not.
Etienne.
Thierry BUCCO wrote:
> Hi,
>
> in the top of my program, i redirect stdout & stderr in a file, like this :
>
> stdout_file = freopen("/root/stdout.txt","a",stdout)
> stdout_file = freopen("/root/stdout.txt","a",stderr)
>
> and i modify his buffer's size
>
> setbuf(stdout_file, NULL);
>
> In fact i want to retrieve all stdout and stderr information, and put then
> into a file. The problem is : the file is created, but there's nothing into.
> I must stop my program to see information (empty buffer ?)
>
> How can i resolve my problem ?
>
> Thanks in advance.
>
> Thierry - FRANCE
------------------------------
From: [EMAIL PROTECTED] (Danny Sung)
Subject: Re: PCI driver blues
Date: Tue, 29 Jun 1999 07:48:44 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 23 Jun 1999 16:44:58 GMT, Frank McGirt <[EMAIL PROTECTED]> wrote:
>I also suggest that you get a copy of
>Linux Device Drivers by Rubini which is worth
>the money for anyone writing drivers.
Thanks for the code snippet and info... I actually have been going off
of Rubini's book, as well as several web sites. From all the
documentation/code I've come across, I believe my problem might be
some weird hardware thing. Maybe there's some sort of strange PCI
Plug-and-Play thing, or some other interaction going on requiring the
board to be initialized in some way before being accessed.
Unfortunately, it just isn't documented. And I don't really see
anything strange from the windows code.
For anyone with just _way_ too much time on their hands, AnalogDevices
has all their software (and a hardware schematic) on their ftp site:
ftp://ftp.analog.com/pub/dsp/adv601/
software/601rpman/ - contains the VXD
--
--
Danny Sung http://www.poboxes.com/dannys
[EMAIL PROTECTED] (PGP public key available)
Key fingerprint = 0080 1BA8 F145 15AE 5ED1 E4ED C422 C77E 70C6 44FE
------------------------------
From: "Ashutosh S. Rajekar" <[EMAIL PROTECTED]>
Subject: makedev
Date: Tue, 29 Jun 1999 20:14:28 +0000
Hello,
Can anyone tell me what makedev actually does ? It is supposed to
make the /dev entries. What is the actual format of the /dev entry, and
what are the parameters passed to it. How does it depend on the kernel
which is running when the makedev command is run ? How does the kernel
process the /dev entries after the makedev command has been run ? What
translation is required ?
Thanks in advance,
==================
Ashutosh Rajekar
http://i.am/asr
==================
I'm sorry my GMT time is really wild, but my sysadmin doesn't cooperate.
------------------------------
From: "Kent Nilsen" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Filesize larger than 2 GB on Intel machines an Linux 2.0.36
Date: Tue, 29 Jun 1999 10:30:07 GMT
I asked about it a while ago, and I got the answer that you have to use the
loopback device. I don't know how that works, but I think it moved the size
limit to 16 Gb or maybe 2 Tb, not sure.
Kent
> I heard the same thing, but I can't find any information on this patch
> to override the limitation.
> > is there any solution to work with files larger than 2 GB on
> > Intel-processor based machines? I am running RedHat Linux 5.2 with
> > kernel 2.0.36. I heard about patch for kernelversions 2.2.x.
------------------------------
From: [EMAIL PROTECTED] (Nathan Myers)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.networking,comp.lang.c++
Subject: Re: Why not C++
Date: 29 Jun 1999 00:47:32 -0700
Isaac <[EMAIL PROTECTED]> wrote:
>Nathan Myers <[EMAIL PROTECTED]> wrote:
>>Some people -- notably, the author of STL -- have no use for classes
>>(inheriting or otherwise), and promote templates above all.
>>
>I'd consider this opinion valid, but if your goal is object oriented
>programming, then why wouldn't a list which emphasizes those aspects
>be reasonable?
Right. He doesn't do object-oriented programming.
> My list is probably pretty close to the order of
>presentation in my learning material which probably does make
>the list suspect : }.
Precisely. The books grew, chapter by chapter, as the language
did. They give most space to earlier features, and treat the
later, probably more important, features as afterthoughts.
>Agreed. I draw it pretty much where you do, but I note operator
>overloading didn't make the expert's top five either. If I
>were writing a matrix library or a complex number handling
>package, I might have a more liberal opinion.
Right. Each feature that actually "got in" was very important
to some significant group of users.
--
Nathan Myers
[EMAIL PROTECTED] http://www.cantrip.org/
------------------------------
From: "Vladimir Z. Nuri" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: 28 Jun 1999 21:17:21 GMT
In comp.os.misc "Cristobal Garcia" <[EMAIL PROTECTED]> wrote:
: I've been follwing the thread and I'd like to express my opinion but before
: doing so I think I should be good summarize a bit. If I haven't understood
: this bad ('cos my english is not very good) there are two major positions:
: 1.- The one which defends a more formal software engineering and tells
: that linux (and linux commuity lacks of it)
: 2.- The one which tells that software engineering is a waste of time
those are defn two valid positions & a dichotomy, but also something
of a straw man. who is gonna argue against software engr? perhaps
you mean "formal software engr" in (2)..
ppl are free to debate on these two items, but after some time
to think about it, the
key issues raised by this thread .. imho could be better summarized
as follows:
1. top down design.
requirements->design->detailed design->prototype->code->beta->release
2. bottom up design
code->release->redesign->code->release->redesign->code->release
that is roughly one way to characterize the dichotomy. also I would
say that ESR (erik s. raymonds) cathedral vs. bazaar essay is
essentially articulating this dichotomy and favoring (2) over (1).
(1) is cathedral & (2) is the bazaar.
however, I believe there is room for a "third way".. between these
two, which is what I am proposing. I want to combine the hackeresque
& open source strengths of (2)/linux with the more disiciplined/
organized blueprint or floorplans of (1) to create a new OS.
------------------------------
From: "Hinomi" <[EMAIL PROTECTED]>
Subject: Help about Riva TNT driver for linux (kernel 2.0.35)
Date: Tue, 29 Jun 1999 18:55:31 +0800
Have any linux driver for Riva TNT?
It usually detected my card is Generic VGA and just 256K VideoRam. Can
anyone help me?Thanks.
------------------------------
From: [EMAIL PROTECTED] (Isaac)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.networking,comp.lang.c++
Subject: Re: Why not C++
Date: 29 Jun 1999 11:29:42 GMT
On 29 Jun 1999 00:47:32 -0700, Nathan Myers <[EMAIL PROTECTED]> wrote:
>
>Precisely. The books grew, chapter by chapter, as the language
>did. They give most space to earlier features, and treat the
>later, probably more important, features as afterthoughts.
>
Hmm. Your arguments (and Matts) are quite persuasive. I don't
think the books I use are quite this flawed. I think the info on
how to use templates is in there, but since I'm not using them
heavily, my perspective on their importance is skewed.
Looks like I've gotten a glimpse into how my use of the language
might evolve over time. I'm far from being an expert c++ programmer
and the vast majority of my experience is with non OOP languages.
Isaac
------------------------------
From: Bernd Eckenfels <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.networking
Subject: Re: Why not C++
Date: 27 Jun 1999 22:55:39 GMT
In comp.os.linux.networking Johan Kullstam <[EMAIL PROTECTED]> wrote:
>> ... you will end up with identical machine code, no matter whether you
>> compile with the C compiler, or compile with the C++ compiler (using the
>> same C compiler as the back end).
sure, but if you have virtual functions, then those need a lookup. I think
it is measured between 2-30%. This is True for Eiffel Compilers who generate
C from Eiffel code, too.
In theory there is no overhead if you dont use C++ Features, but then
again.. it is not faster as C.
Greetings
Bernd
------------------------------
From: "Frank V. Castellucci" <[EMAIL PROTECTED]>
Subject: Re: Help about Riva TNT driver for linux (kernel 2.0.35)
Date: Tue, 29 Jun 1999 08:16:20 -0400
Hinomi wrote in message <7la903$4te$[EMAIL PROTECTED]>...
>Have any linux driver for Riva TNT?
There is a X update which includes new drivers and Riva TNT is one of them.
I have had no problem with it on
RH 6.0.
Frank V. Castellucci
Monkey Boy
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Help about Riva TNT driver for linux (kernel 2.0.35)
Date: Tue, 29 Jun 1999 12:17:17 GMT
Hinomi <[EMAIL PROTECTED]> wrote:
: Have any linux driver for Riva TNT?
: It usually detected my card is Generic VGA and just 256K VideoRam. Can
: anyone help me?Thanks.
You will have to be more specific. I am working on a rivafb driver,
available in my 2.3.x "jg-fb" patches at
http://havoc.gtf.org/garzik/kernel/files/UNTESTED/
Jeff
------------------------------
From: Guillermo <[EMAIL PROTECTED]>
Subject: Glade manual?
Date: Tue, 29 Jun 1999 14:30:07 +0200
Where can I find glade manuals?
Thanks a million
guillermo
------------------------------
From: Oliver Stahlhut <[EMAIL PROTECTED]>
Subject: Flag for switching back to conservative memory management ?
Date: Tue, 29 Jun 1999 16:11:48 +0200
Hello all!
In the last months I had to "ALT-PRN-B"/reset many "almost dead" Linux
numbercrunchers (kernel 2.2.[0-10], glibc2 based, >=512MB mem, 1GB Swap,
2 PentiumII) here in our cluster, who seem to have died because of
memory mismanagement. If I start the following "wicked" ;) program twice
(in parallel) any of our "better" machine crashes silently once the mem
& swap are used up. If I start it on a 2.0.x machine, the job won't do
any harm.
Can anyone else reproduce my problem ... is it due to memory
overcommitment ? - if yes, how can I turn it off ;)
Regards
Oliver
#include <stdio.h>
#include <string.h>
#define MEM_MAX 1024 * 1024 * 10
main()
{
long mem_avail = MEM_MAX;
char *buf;
while(1)
{
printf("Trying to allocate %ld bytes: ", mem_avail);
if ((buf = (char *)malloc((size_t)mem_avail))){
printf(" success!\n");
memset(buf, 'X', mem_avail);
sleep(1);
} else {
printf(" failed :(\n");
mem_avail = mem_avail / 2;
sleep(1);
}
}
return 0;
}
--
/*
Oliver Stahlhut - Universitaet Hannover
Institut f�r Theoretische Nachrichtentechnik
und Informationsverarbeitung (TNT)
mailto:[EMAIL PROTECTED]
http://www.tnt.uni-hannover.de/~stahlhut/
*/
------------------------------
From: Oliver Stahlhut <[EMAIL PROTECTED]>
Crossposted-To: revue.linux-smp
Subject: Flag for switching back to conservative memory management ?
Date: 29 Jun 1999 16:22:19 +0200
Hello all!
In the last months I had to "ALT-PRN-B"/reset many "almost dead" Linux
numbercrunchers (kernel 2.2.[0-10], glibc2 based, >=512MB mem, 1GB Swap,
2 PentiumII) here in our cluster, who seem to have died because of
memory mismanagement. If I start the following "wicked" ;) program twice
(in parallel) any of our "better" machine crashes silently once the mem
& swap are used up. If I start it on a 2.0.x machine, the job won't do
any harm.
Can anyone else reproduce my problem ... is it due to memory
overcommitment ? - if yes, how can I turn it off ;)
Regards
Oliver
#include <stdio.h>
#include <string.h>
#define MEM_MAX 1024 * 1024 * 10
main()
{
long mem_avail = MEM_MAX;
char *buf;
while(1)
{
printf("Trying to allocate %ld bytes: ", mem_avail);
if ((buf = (char *)malloc((size_t)mem_avail))){
printf(" success!\n");
memset(buf, 'X', mem_avail);
sleep(1);
} else {
printf(" failed :(\n");
mem_avail = mem_avail / 2;
sleep(1);
}
}
return 0;
}
--
/*
Oliver Stahlhut - Universitaet Hannover
Institut f�r Theoretische Nachrichtentechnik
und Informationsverarbeitung (TNT)
mailto:[EMAIL PROTECTED]
http://www.tnt.uni-hannover.de/~stahlhut/
*/
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
------------------------------
From: Tristan Wibberley <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Proposal For New Windowing software
Date: Mon, 28 Jun 1999 14:04:37 +0100
Reply-To: [EMAIL PROTECTED]
Matthew Carl Schumaker wrote:
>
> > On Mon, 28 Jun 1999 09:40:08 -0400, Matthew Carl Schumaker
> > <[EMAIL PROTECTED]> wrote:
> > >During the past couple weeks I've seen many people express their opinions
> > >on both Xwindows and C++. The project that I'm proposing is one that I
> > >not only want seen done but is also one that I want to participate in.
> > >
> > >The proposal is for a new Windowing System that is based on the premise of
> > >a 3D UI. I know other people are interested in this sort of program as
> > >well as a couple private ventures, but to the best of my knowledge there
> > >is no Open Source project of this type.
> >
> > Have you looked at Berlin? It is seeking to build a GUI system using
> > CORBA, GGI, and OpenGL/Mesa.
> >
> > It *looks* like they're primarily working on the 2D aspects of it;
> > presumably they figure it makes sense to have that somewhat working
> > first.
> I've looked at Berlin but they seem to be taking the stance of all other
> 2D GUIs with 3D support, you run the 3D in a Window. What I want to do
> Is run the 3D as the main interface and run 2D Windows inside of that.
> hence the need for support for 3D devices.
There is nothing to limit it like that as far as I can see.
Initially, the "window" management (it doesn't work on the principal of
windows though) will be 2D (ie, you can have 3D in a 2D window), but
that's only because the window manager is like that - a different window
manager (what's the correct term here?) will let you do more using
OpenGL features as Berlin advances.
That provides for 2D projections of apps onto any surface, but full 3D
UI support can easily be added, so your app can ask for a new 3D object
in the UI space. All it needs is additional specifications for those
features.
--
Tristan Wibberley
------------------------------
From: "robert_c" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.hardware
Subject: where can get (Maximum RPM) this online document?
Date: 29 Jun 1999 13:07:52 GMT
Could someone know where is the mirror site for the online document Maximum
RPM(because I want to package my program with RPM format)? I have tried many
times to download from www.rpm.org, but the site seems shutdown; at the same
time, I tryied to RedHat site (http://www.redhat.com/docs/) to download the
same document(it is linked to http://www.rpm.orgorg/maximum-rpm.ps.gz), but
the result is same.
Thanks for your help
------------------------------
From: "Brian J. Johnson" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.misc,comp.os.linux.powerpc
Subject: Re: Mac-emulation on Linux?
Date: Tue, 29 Jun 1999 08:55:58 -0500
"Anthony D. Saxton" wrote:
>
> Matt Denton wrote:
>
> > In article <7gjd3s$cdo$[EMAIL PROTECTED]>, "FM"
> > <[EMAIL PROTECTED]> wrote:
> >
> > > 2. Buy a Macintosh and dual boot with Linux/MacOS
> >
> > This is probably the best option - LinuxPPC is pretty solid, MKlinux
> > (easier to install?) is a bit sluggish as it runs on top of a Mach
> > microkernel instead of natively but is a decent port of Linux nonetheless.
> >
> > Me, I'm looking forward to Sheepshaver (http://www.sheepshaver.com) MacOS
> > VM for Linux to see how that works...
> >
> > --
> > Matt Denton
> > San Francisco, USA
>
> SheepShaver's fine for those of us privilidged few using the PowerPC version
> of linux. It won't help anyone that doesn't have a PREP, BeBox or PowerMac
> though.
You may want to check out BasiliskII, a GPLed 68K Mac emulator from the
same folks who are doing SheepShaver. It's designed for portability and
appears to run under Intel linux (although I've only tried it under SGI
Irix so far.) It's still very much alphaware, but is seeing frequent
updates.
http://www.uni-mainz.de/~bauec002/B2Main.html
Brian J. Johnson
====================================================================
"Not to find even more ways of saving time, but to look for ways to
spend it"
-- The secret to "celebrating the sidelines of life" - Adair Lara
------------------------------
** 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
******************************
