Linux-Development-Sys Digest #206, Volume #8 Tue, 10 Oct 00 20:13:08 EDT
Contents:
Re: Compile Failure (jwk)
Re: Determining what CPU is going to execute a thread from a program (jwk)
Any drivers for the TARGA+ 64 board available? ("Norm Dresner")
Re: Most popular Linux development environment(s)? Graphical? ("Jan Schaumann")
Re: Determining what CPU is going to execute a thread from a program ("Norman Black")
Re: Compile Failure ("John Hall")
Re: Most popular Linux development environment(s)? Graphical? (jazz)
Re: Determining what CPU is going to execute a thread from a program ("Miguel Angel
Rodriguez Jodar")
Re: Determining what CPU is going to execute a thread from a program ("Miguel Angel
Rodriguez Jodar")
Re: Most popular Linux development environment(s)? Graphical? ([EMAIL PROTECTED])
Opening a kernel module from another kernel module ("Ed Hudson")
Re: Determining what CPU is going to execute a thread from a program (Ulrich Weigand)
How to allocate a physical address range ("Don Camp")
Re: Kernel 2.4.0-test7, 8, and 9 "Warning pasting would not give a valid
preprocessing token" ? (Rudi Sluijtman)
memory utilities (Weiguang Shi)
Re: Driver Loading FPGA Device (Jeff Andre)
Re: How to allocate a physical address range (Andi Kleen)
Large IDE HDDs (Gilles)
Re: Driver Loading FPGA Device (Pete Zaitcev)
Re: How to allocate a physical address range (Pete Zaitcev)
Re: Most popular Linux development environment(s)? Graphical? (ishpeck)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (jwk)
Subject: Re: Compile Failure
Date: 10 Oct 2000 19:14:06 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 10 Oct 2000 16:15:01 GMT, John Hall
<[EMAIL PROTECTED]> wrote:
>Have compiled 2.3.99-pre9 and others [including doing backports and the
>like] without problems. however on this box I am getting an error message
>during the compile of a straight "out of the box" 2.3.99-pre9. I tried
>wiping the source tree out and installing a "package" of RedHat's kernel
>source "2.2.16" and I got the exact same message.
>
>I have installed gcc 2.96-54 and the latest mandrake cooker binutils. The
>box is a clean install of redhat 7. I have updated various other things,
>libc5 etc, but only to solve the problem which hasn't changed since a virgin
>RH 7 install.
>
That is not an official gcc release. RedHat decided to release a
compiler + a set of patches that the gcc developers themselves didn't
want to release. Now you see why :-) Anyway, the kernel is certified to
compile only (and then in a limited way) with 2.7.2.3, 2.91.66 and
2.95.2. And also, 2.3.99-pre9 is very old. Try 2.4.0-test9.
Good luck,
Jurriaan
--
Half of the American people never read a newspaper.
Half never voted for President. One hopes it is the same half.
Gore Vidal
GNU/Linux 2.2.18pre15 SMP 5:07 9 users load av: 1.21 1.05 1.03
------------------------------
From: [EMAIL PROTECTED] (jwk)
Subject: Re: Determining what CPU is going to execute a thread from a program
Date: 10 Oct 2000 19:14:08 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 10 Oct 2000 18:48:47 +0200, Miguel Angel Rodriguez Jodar
<[EMAIL PROTECTED]> wrote:
>Hi all!
>
>Is possible to indicate to the kernel which CPU, in a multiprocessor system,
>must accept a thread we want to launch using fork(), clone() or
>phread_create() or whatever?
not without writing it as a kernel module, I think.
>Does exist a system call that can return us the number of active CPU's in a
>system?
>
isn't that in /proc/cpuinfo?
Good luck,
Jurriaan
--
BOFH excuse #386:
The Internet is being scanned for viruses.
GNU/Linux 2.2.18pre15 SMP 5:31 9 users load av: 1.41 1.16 1.10
------------------------------
Reply-To: "Norm Dresner" <[EMAIL PROTECTED]>
From: "Norm Dresner" <[EMAIL PROTECTED]>
Subject: Any drivers for the TARGA+ 64 board available?
Date: Tue, 10 Oct 2000 19:14:16 GMT
Before I start the long and painful process of developing a device driver
for the TrueVision TARGA+ 64 video (capture) board, I was hoping that
someone could tell me that it's already been done and that I can download it
for free (and if pigs had wings,...)
Anyway, I'd appreciate any relevant information, suggestions, criticisms and
even a well-written flame or two.
Thanks
Norm
Norman Dresner
Fellow Systems Engineer
Radar Systems Engineering Department
Electronic Systems and Sensors Segment
Northrop Grumman Corporation
MS 520
Box 746
Baltimore MD 21203
Voice: (410) 993 - 2096 Mornings; all-day voice-mail
(410) 969 - 8068 Afternoons with answering machine
FAX: (410) 993 - 8084 On-site
(410) 969 - 8068 Afternoons; call first to arrange
E-Mail:Mornings: [EMAIL PROTECTED]
Afternoons: [EMAIL PROTECTED]
------------------------------
From: "Jan Schaumann" <[EMAIL PROTECTED]>
Subject: Re: Most popular Linux development environment(s)? Graphical?
Crossposted-To: comp.os.linux.development.apps
Date: Tue, 10 Oct 2000 15:13:10 +0400
"jazz" <[EMAIL PROTECTED]> wrote:
> In article <[EMAIL PROTECTED]>, Johan Kullstam
> <[EMAIL PROTECTED]> wrote:
>
>
[emacs "tricks"]
>
> I guess emacs can do a lot more than I thought. Do you know where I can
> find a good book on using emacs like this, preferably with a tutorial? I
> have the gnu reference somewhere but haven't looked at it in a while, as
> I recall it is more like a reference, I want something that will step me
> through all the features, functions, tricks, etc...
O'Reilly (who else) has an excellent book on Emacs: "Learning GNU Emacs"
(http://www.oreilly.com/catalog/gnu2/) and for advanced users the "GNU
Emacs Pocket Reference" (http://www.oreilly.com/catalog/gnupr/).
Look for those books in your favorite (online) bookstore (a very cheap
one being http://www.bookpool.com).
Cheers,
-Jan
--
Jan Schaumann <http://www.netmeister.org>
Please add smileys where appropriate.
------------------------------
From: "Norman Black" <[EMAIL PROTECTED]>
Subject: Re: Determining what CPU is going to execute a thread from a program
Date: Tue, 10 Oct 2000 12:46:07 -0700
Reply-To: "Norman Black" <[EMAIL PROTECTED]>
> Is possible to indicate to the kernel which CPU, in a multiprocessor
system,
> must accept a thread we want to launch using fork(), clone() or
> phread_create() or whatever?
Generally it is best to let the system(kernel) choose how to run the
available threads in the system. It understands the dynamics of everything
in the system and can make the best overall choices for making full
utilization of available resources (processors and threads).
--
Norman Black
Stony Brook Software
the reply, fubar => ix.netcom
"Miguel Angel Rodriguez Jodar" <[EMAIL PROTECTED]> wrote in
message news:8rvhu3$4am$[EMAIL PROTECTED]...
> Hi all!
>
> Is possible to indicate to the kernel which CPU, in a multiprocessor
system,
> must accept a thread we want to launch using fork(), clone() or
> phread_create() or whatever?
> Does exist a system call that can return us the number of active CPU's in
a
> system?
>
> Thanks in advance!
>
> Miguel Angel
>
>
------------------------------
From: "John Hall" <[EMAIL PROTECTED]>
Subject: Re: Compile Failure
Date: Tue, 10 Oct 2000 20:20:01 GMT
Thank you both for your help.
"Paul Kimoto" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> In article <9gHE5.24352$[EMAIL PROTECTED]>, John Hall
wrote:
> > Have compiled 2.3.99-pre9 and others [including doing backports and the
> > like] without problems. however on this box I am getting an error
message
> > during the compile of a straight "out of the box" 2.3.99-pre9.
>
> > gcc -D__ASSEMBLY__ -D__KERNEL__ -I/usr/src/linux/include -traditional -c
> > checksum.S -o checksum.o
> > checksum.S:238: badly punctuated parameter list in #define
> > checksum.S:244: badly punctuated parameter list in #define
>
> > I have installed gcc 2.96-54
>
> ... which has a new preprocessor (among other features, and the whole
thing
> is as yet unreleased) that does not get along with the way that the kernel
> has been written in the past. There are supposed to be some fixes in the
> latest development kernel, which is 2.4.0-test9 (or -test10-pre1).
> (2.4.0-test1 was available 4.5 months ago!)
>
> Or you could use an older (gcc-2.95.* or egcs-1.1.*) version of the
> compiler. They say that RH 7.0 has one (re)named "kgcc".
>
> --
> Paul Kimoto
> This message was originally posted on Usenet in plain text. Any images,
> hyperlinks, or the like shown here have been added without my consent,
> and may be a violation of international copyright law.
------------------------------
From: [EMAIL PROTECTED] (jazz)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Most popular Linux development environment(s)? Graphical?
Date: Tue, 10 Oct 2000 16:40:21 -0400
In article <[EMAIL PROTECTED]>, Kevin Holbrook
<[EMAIL PROTECTED]> wrote:
> jazz wrote:
> >
> > I'm trying to understand your point. I CAN develop code using a command
> > line interface, with gcc, emacs as an editor, and the command line
> > debugger. But I find it primitive. I don't develop code for distribution;
> > I develop code to do a particular job (usually experiment control or data
> > analysis) in our lab, on a particular platform. I much prefer having a
> > graphical user interface like Codewarrior with a graphical debugger that
> > lets me see numerous lines of code in numerous functions in many windows
> > at once, with a window for multiple variables, execution tracing, analysis
> > tools to determine efficiency, a very good framework, etc etc etc.
> >
>
>
> Well I won't jump into the IDE vs. Command-line debate,
> especially since I work on a graphical IDE. ;-)
>
> There are pro's and con's for each. I use XEmacs myself,
> from time to time.
>
> But, if you want CodeWarrior there is a CodeWarrior for Linux.
> We are currently in beta testing for a new release.
>
> ------------------------------
> Kevin Holbrook
> Linux Developer
> Metrowerks, a Motorola Company
Thanks Kevin. Do you know what I can read to answer questions like the
following?
I want to develop software that will last me for years (my use only, for
research stuff).
I would only like to spend my time on a learning curve that will last me the
next ten years (or so...).
Can I develop with Codewarrior for Mac OSX (not Carbon, but the Unix side)?
Can I develop code that will be portable between Linux and OSX Unix?
Using Codewarrior?
What would I use for graphical stuff (drawing graphs, etc.)?
Thanks sincerely,
Jazz
------------------------------
From: "Miguel Angel Rodriguez Jodar" <[EMAIL PROTECTED]>
Subject: Re: Determining what CPU is going to execute a thread from a program
Date: Tue, 10 Oct 2000 21:55:34 +0200
Norman Black escribi� en mensaje <8rvr2e$bvq$[EMAIL PROTECTED]>...
>> Is possible to indicate to the kernel which CPU, in a multiprocessor
>system,
>> must accept a thread we want to launch using fork(), clone() or
>> phread_create() or whatever?
>
>Generally it is best to let the system(kernel) choose how to run the
>available threads in the system. It understands the dynamics of everything
>in the system and can make the best overall choices for making full
>utilization of available resources (processors and threads).
I undestand your point, but in fact, what I want to do is a study of the
impact of several processes running on different CPU's. Consider for
instance a massive paralell program consisting on several threads inside the
same process. If I let the kernel to choose what CPU to use on each thread,
it will surely choose the same CPU for all threads because of cache
coherency issues. If I write the program with care, I can avoid these issues
and therefore, I can exploit the paralellism at a higher rate that the
system allows, or at least, this is what I want to proof.
Miguel Angel
------------------------------
From: "Miguel Angel Rodriguez Jodar" <[EMAIL PROTECTED]>
Subject: Re: Determining what CPU is going to execute a thread from a program
Date: Tue, 10 Oct 2000 22:41:08 +0200
>not without writing it as a kernel module, I think.
That's what I thought.... I wanted to be sure before surfing into the kernel
files :)
Thanks!
Miguel Angel
------------------------------
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Most popular Linux development environment(s)? Graphical?
From: [EMAIL PROTECTED]
Date: 10 Oct 2000 17:02:01 -0400
Kevin Holbrook <[EMAIL PROTECTED]> writes:
> Well I won't jump into the IDE vs. Command-line debate,
> especially since I work on a graphical IDE. ;-)
> There are pro's and con's for each. I use XEmacs myself,
> from time to time.
Emacs has ruined me. I can't stand IDEs that aren't as functional.
It frustrates me that many modern IDEs don't do something as simple as
count parantheses or curly braces. (And don't even get me started on
Microsoft's idiotic "The IDE knows what you _really_ meant to type"
crap.)
But at the same time, I long for some of the super-slick visualization
features, and the ability to use the mouse in a meaningful way when I
choose to. It's the sort of thing that makes me want to get off my
ass and write/contribute to an IDE, but I need Win2K on my box right
now.
--
Eric McCoy <[EMAIL PROTECTED]>
------------------------------
From: "Ed Hudson" <[EMAIL PROTECTED]>
Subject: Opening a kernel module from another kernel module
Date: Tue, 10 Oct 2000 17:37:30 -0500
I am trying to talk to a kernel module driver that I wrote from another kernel module
that
implements some tests for the driver. I am unable to open it with the standard "open"
call that
works from user space. I assume this is because that functionality is not available
at the kernel level. Are there any examples available? I want to be able to open,
read and
write from another kernel module.
Thanks
Ed Hudson
------------------------------
From: [EMAIL PROTECTED] (Ulrich Weigand)
Subject: Re: Determining what CPU is going to execute a thread from a program
Date: 10 Oct 2000 23:41:23 +0200
"Miguel Angel Rodriguez Jodar" <[EMAIL PROTECTED]> writes:
>I undestand your point, but in fact, what I want to do is a study of the
>impact of several processes running on different CPU's. Consider for
>instance a massive paralell program consisting on several threads inside the
>same process. If I let the kernel to choose what CPU to use on each thread,
>it will surely choose the same CPU for all threads because of cache
>coherency issues.
Certainly not. If the other CPUs would be idle otherwise, of course
threads of your application would be scheduled across multiple CPUs.
Cache coherency considerations only *contribute* to the overall
decision, they are not the sole deciding factor.
>If I write the program with care, I can avoid these issues
>and therefore, I can exploit the paralellism at a higher rate that the
>system allows, or at least, this is what I want to proof.
In general, you cannot. This is simply because the optimal scheduling
strategy at any point in time depends on whatever *other* applications
compete with yours for CPU time. This is something that your app
simply doesn't know; only the kernel has the overall picture.
--
Dr. Ulrich Weigand
[EMAIL PROTECTED]
------------------------------
From: "Don Camp" <[EMAIL PROTECTED]>
Subject: How to allocate a physical address range
Date: Tue, 10 Oct 2000 15:16:43 -0700
Does anyone know how to allocate a physical address range to map the
registers on a PCI card?
I'm working on a driver for a PCI card that uses a Vrc5074 System Controller
for the PCI bridge. The Vrc5074 has 11 64-bit Base Address Registers (BARs).
The first three 64-bit BARS are located in the area defined for base
addresses in the PCI configuration space at offsets 0x10-0x27. The remaining
eight 64-bit BARS are located in the device dependent area of the PCI
configuration space at offsets 0x40-0x7f. I'm expecting the motherboard BIOS
to do the correct allocation of physical addresses for the first three BARs
but don't expect the BIOS to do anything with the remaining eight BARs
located in the device dependent area. I need to be able to set theses BARs
with physical addresses that are unused by any other system memory or
devices. I've searched the source (2.2 kernel) for any function that would
allocate physical address space or for code that kept track of physical
address space usage but found none. Any pointers would be greatly
appreciated.
Don Camp
------------------------------
Subject: Re: Kernel 2.4.0-test7, 8, and 9 "Warning pasting would not give a valid
preprocessing token" ?
From: Rudi Sluijtman <[EMAIL PROTECTED]>
Date: Tue, 10 Oct 2000 22:31:32 GMT
[EMAIL PROTECTED] (Frank Sweetser) writes:
> Try instead setting CC=kgcc in the top level linux/Makefile (there should
> already be a line assigning CC=(something else, probably gcc))
or on the commandline do:
make bzImage CC=kgcc && make modules CC=kgcc
which accomplishes the same without editing the Makefile.
Regards,
Rudi Sluijtman.
------------------------------
From: Weiguang Shi <[EMAIL PROTECTED]>
Subject: memory utilities
Date: Tue, 10 Oct 2000 16:59:10 -0600
Hi people,
Which tool can I use to check the memory(physical&virtual) sizes under linux?
If you have more tools to recommend to monitor/deal with memory,
could you please let me know too?
Thanks very much
Weiguang
------------------------------
From: [EMAIL PROTECTED] (Jeff Andre)
Subject: Re: Driver Loading FPGA Device
Date: 10 Oct 2000 22:58:42 GMT
Mathias Waack ([EMAIL PROTECTED]) wrote:
: Hi,
: [EMAIL PROTECTED] (Pete Zaitcev) writes:
: > On Mon, 09 Oct 2000 18:03:20 GMT, [EMAIL PROTECTED]
: > <[EMAIL PROTECTED]> wrote:
: > > One of the first tasks the driver I am writing must do is to read a
: > >.pof file off of the system disk and program a FPGA device on the
: > >device board. This is for a PCI device but I am not sure that is
: > >relevant to this situation. Does anyone have or know of any code
: > >that implements this?
: > >
: > This is normally done with a helper application. The application reads
: > the file, then issues an ioctl() into the driver that programs the
: > FPGA.
: Yes. Or another solution would be a writable /proc-file. So could just
: use "cat" as your helper application.
I have also have a card with FPGAs. I've loaded the card via ioctl() and
it works fine. One problem I have is that it takes about 11 seconds to
load the FPGAs. For the duration of the ioctl() a kernel lock is held
(2.2.16, haven't looked at the 2.4.0 kernels yet) and processes on the
other processor can hang if they do something that requires kernel services.
Is there a way of releasing that lock so the other processor can continue
working? I've been thinking about scheduling an immediate task to handle
the loading. However, using /proc looks like an interesting solution.
Jeff Andre
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: How to allocate a physical address range
Date: 11 Oct 2000 00:52:54 +0200
"Don Camp" <[EMAIL PROTECTED]> writes:
> Does anyone know how to allocate a physical address range to map the
> registers on a PCI card?
Try reading Documentation/IO-mapping.txt
(and a bit more research before posting next time)
-Andi
------------------------------
From: Gilles <[EMAIL PROTECTED]>
Subject: Large IDE HDDs
Date: Tue, 10 Oct 2000 16:29:29 -0700
Reply-To: [EMAIL PROTECTED]
Hello,
I have freshly installed a Linux RedHat 6.2 on an old machine which BIOS
does not understand number of physical sectors above 65K. (At least it
allows to turn off LBA so I've set it to 64K). The HD I am trying to add
to the system (/dev/hdb) has 79406 sectors.
I modified lilo.conf to add "hdb=79406,16,63" and I even modified ide.c
to the following (where it stores parsed values):
printk("ide_setup: GILLESDEBUG: parsed value vals[0] = %d\n", vals[0]);
drive->cyl = drive->bios_cyl = vals[0];
printk("ide_setup: GILLESDEBUG: stored value drive->cyl = %d\n",
drive->cyl);
I get the following:
> ide_setup: GILLESDEBUG: parsed value vals[0] = 79406);
> ide_setup: GILLESDEBUG: stored value disk->cyl = 13870);
I come to realize that both the cyl and bios_cyl fields are declared as
unsigned short which clearly won't work.
My question is (and please forgive me if this has already been said) is
there 17bit support for IDE already written? if not how can I
contribute? I'd be more that happy to modify the IDE to support the 17th
bit (I need it BADLY). I just don't want to do it if it's already been
done.
There is actually one thing I am not familiar with: If the IDE disk
structures were to be changed (unsigned instead of unsigned short), and
say, it now outputs the proper bits to the proper places; DOES any one
know how the rest of the system would be affected? I noticed that most
utilities (such as fdisk) use ints to deal with cylinders but would they
know how to retreive that information properly? (I didn't look into that
yet).
Thanks,
Gilles
--
PS: My newsfeed is flaky so if you don't mind, please cc me in email.
Thanks.
------------------------------
From: [EMAIL PROTECTED] (Pete Zaitcev)
Subject: Re: Driver Loading FPGA Device
Date: Tue, 10 Oct 2000 23:34:37 GMT
> I have also have a card with FPGAs. I've loaded the card via ioctl() and
> it works fine. One problem I have is that it takes about 11 seconds to
> load the FPGAs. For the duration of the ioctl() a kernel lock is held
> (2.2.16, haven't looked at the 2.4.0 kernels yet) and processes on the
> other processor can hang if they do something that requires kernel services.
>[...]
> Jeff Andre
How time sensitive is FPGA loading on your device? E.g. if you do
JTAG with hardware assist or E2C you may be able do everything
from the userland, doing mmap of device space. As you can see,
I am entirely clueless about the way FPGA operates...
--Pete
------------------------------
From: [EMAIL PROTECTED] (Pete Zaitcev)
Subject: Re: How to allocate a physical address range
Date: Tue, 10 Oct 2000 23:46:58 GMT
> Does anyone know how to allocate a physical address range to map the
> registers on a PCI card?
>
> I'm working on a driver for a PCI card that uses a Vrc5074 System Controller
> for the PCI bridge. The Vrc5074 has 11 64-bit Base Address Registers (BARs).
>[...]
> I need to be able to set theses BARs
> with physical addresses that are unused by any other system memory or
> devices.
> Don Camp
Don, I suggest you to contact Martin "MJ" Mares. He is in charge
for PCI allocation and BAR initialization. His address is "mj
at suse".
In my opinion, 2.2 is not especially useful for what you need and
you'll end up with an unbundled patch. In 2.4 you may use the
generic resource manager, so that everything is easier.
You may ride on bus fixup functions, but you better discuss
this with MJ.
When you write to MJ you may wish to cc: to the linux-kernel list.
Very few clueful people read Usenet these days.
--Pete
------------------------------
From: ishpeck <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Most popular Linux development environment(s)? Graphical?
Date: Tue, 10 Oct 2000 23:35:29 GMT
http://www.kdevelop.org/
http://www.softwarebeuro.de/
In article
<[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (jazz) wrote:
> I'm thinking of switching from mac to Linux. If I did, what
programming
> setup would I likely find myself using? Gnu? Is there anything with a
> graphical interface, esp. for the debugger, as in Codewarrior, where
you
> see a window containg your code and you can clearly mark breakpoints
and
> follow the program counter steping through your code? With windows for
> variables, stack tracing, etc.?
>
> Thanks
> Jazz
>
--
I want my lemmrick, damnit!
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
** 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
******************************
