Linux-Development-Sys Digest #166, Volume #7 Tue, 7 Sep 99 15:14:22 EDT
Contents:
Re: Unix and the ISO/IEC 14882-1998 Standard C++ Library
([EMAIL PROTECTED])
Re: LispOS? (Mike McDonald)
Re: [Linux] Calling C from Fortran: Function always returns zero? (Jan Just Keijser)
Re: linuxthreads and mutexes (Kaz Kylheku)
Development documentation Where to start? (Simon Kwan)
Struct packet_type(Question) (Fernando Ortega Bellosta)
DCE RPC on Linux ("David Castells i Rufas")
Re: Problem compiling GNU/libc6 on Debian (Andreas Jaeger)
Re: Problems building a cross-compiler (Marcus Sundberg)
Re: Richards Stevens died (Richard)
Re: TAO: the ultimate OS (FM)
Re: Help Compiling GCC 2.95.1 (Please!! ) (Paul D. Smith)
Re: TAO: the ultimate OS ("Vladimir Z. Nuri")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.programming.threads,comp.std.c++,comp.unix.programmer
Subject: Re: Unix and the ISO/IEC 14882-1998 Standard C++ Library
Date: 7 Sep 1999 15:33:13 GMT
In article <[EMAIL PROTECTED]>,
Johan Kullstam <[EMAIL PROTECTED]> wrote:
> "Robert DiFalco" <[EMAIL PROTECTED]> writes:
> > We are working on a project that has targets for Windoze NT, Linux,
> > AIX, HPUX, SGI, and Solaris that must support multiple native
> > locales, well... at least Japan (SJIS) and US/English.
> > We are currently using KAI with its OEM'd Standard C++ Library
> > (Modena I think) for all the Unix platforms we support and MSVC++
> > with its OEM'd Standard C++ Library, DinkumWare, for Windows
> > NT/2000. Support the various 14882-1998, locale, and system support
> > between the various compiler versions and platforms has been more of
> > a nightmare than, IMHO, it should be.
> nod. C++ isn't as mature as it could be. multiple-platform support
> might be difficult. depending upon your application, perhaps you
> would be better served by choosing a completely different language
> such as smalltalk or common-lisp.
His only problem seems to be finding a portable implementation of the
internationalization part of the library. Since Smalltalk or Common
Lisp don't have as extensive support for internationalization, I don't
see where they could help here.
> > I'm particularly having a hard time with KAI, or at least with its
> > Standard C++ Libraries. Apparently, while it does support the ANSI-C
> > notion of wchar_t, it has no support for the STDCPP wchar_t classes
> > such as the all-important std::char_traits<wchar_t>, wstring,
> > wiostream, etc. As if this weren't enough, on almost all platforms
> > KAI supports it apparently has no support for mbstate_t. Finally, it
> > seems that the only locale that any facet are implemented for is the
> > "classic" locale, or "C" locale. Even some of those seem to be
> > broken, basically, std::get_facet() doesn't work.
> > I'd like some input regarding the following options:
> > A) Use a single version of a single Vendors Standard C++ Library
> > on every Platform including Windows NT. Unfortunately, what is
> > offered by many companies (such as Rogue-Wave and ObjectSpace)
> > is more like the Standard Template Library than the Standard C++
> > Library.
> I thought the standard template library (STL) was the standard C++
> library.
The STL was adopted (and adapted) as *part* of the standard library.
STL doesn't concern itself in anyway with "presentation" issues like IO
or internationalization.
I'm pretty sure that the Dinkumware implementation would work with the
KAI compiler. On the other hand, I don't think Dinkumware sells to end
users.
> > B) Find a standard C++ library that has *complete* 14882-1998
> > support, or at least as complete as Plaguers Dinkumware Standard
> > C++ Library that will work with all the platforms KAI supports,
> > specifically HPUX, Redhat, SGI, AIX, and Solaris/Intel.
I'm not aware of a commercial one. As far as I know, g++ is somewhat
behind in this area as well.
> > C) Or, if there is no other choice, replace KAI with a compiler
> > that has OEM'd runtimes that do have 14882-1998 compliance
> > sufficient enough to create standardized international,
> > multiplatform C++ programs.
> > I am totally open to choices like GNU/OSF but using these for
> > commercial projects is very new to me. The last I looked, the GNU
> > C++ libraries didn't look like the 14882-1998 standard libraries,
> > especially regarding NTMBS, NTWCS, locales, and facets. However, I
> > would love to support these efforts by using them if there is a
> > compliant implementation that will compiler with KAI and work with
> > NT, SGI, Linux, Solaris/Intel, HPUX, and AIX.
Well, one way of supporting them would be to implement it yourself, then
donate the results to the cause:-). Seriously, getting some one to
implement it for you on contract might be your best bet. (I'd love such
a contract, but I'm not free at present:-(.)
Alternatively, you could simply use the C library. In terms of
functionality, C++ adds message library support (but with enough
implementation dependant points to make its portable use questionable)
and input/output code translation. It also associates a locale with
each stream, which is almost essential for the code translation. Neither
standard C nor C++ support positional parameters in a formatting string,
IMHO, one of the most essential issues -- most of the UNIX platforms you
mention do support it in printf, as an extension, but none of the NT
compilers does.
If I were on such a project, and the goal is not to create something
general or standard, I'd use the C library for most of the formatting
issues (date, money, collating sequences...) and I'd develop forwarding
streambuf's for the IO code translation. I've already got a class which
handles the formatting with positional parameters. But I've yet to find
a good, portable solution to the message catalog -- the typical Unix
system uses a string as an index, where as Windows, I think, uses ints.
--
James Kanze mailto: [EMAIL PROTECTED]
Conseils en informatique orient�e objet/
Beratung in objekt orientierter Datenverarbeitung
Ziegelh�ttenweg 17a, 60598 Frankfurt, Germany Tel. +49(069)63198627
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
[ comp.std.c++ is moderated. To submit articles, try just posting with ]
[ your news-reader. If that fails, use mailto:[EMAIL PROTECTED] ]
[ --- Please see the FAQ before posting. --- ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html ]
------------------------------
From: [EMAIL PROTECTED] (Mike McDonald)
Subject: Re: LispOS?
Date: 7 Sep 1999 16:52:59 GMT
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>,
Harald Arnesen <[EMAIL PROTECTED]> writes:
> [EMAIL PROTECTED] (Peter Samuelson) writes:
>
>> [Nix <$}xinix{[email protected]>]
>> > Shouldn't be too hard to add it though. I'll work up some patches for
>> > xemacs, I think, just as a gimmick ;)
>>
>> Don't forget to disable "\C-x\C-c"....
>
> No, that should reboot the system.
Reboot the system? Why would you want to do that?
Mike McDonald
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Jan Just Keijser)
Subject: Re: [Linux] Calling C from Fortran: Function always returns zero?
Date: Tue, 07 Sep 1999 16:21:06 GMT
In article <7qr4qh$[EMAIL PROTECTED]>, [EMAIL PROTECTED] (John
H. Chauvin) wrote:
>I have written a very simple fortran program which calls a
>function which is written in C. This is a very straight
>forward test case for mixed Fortran and C programming
>under Linux using g77 and gcc.
>
>The program passes the number four to the C function where
>it is multiple by two and the resultant value is returned to
>the fortran program. Unfortunately, the return value as
>printed from the fortran routine is always zero. Why is
>the return value being lost?
>
>Any hints as to why this is happening?
>
>Thanks,
>
>John Chauvin
>
>Fortran Program (number.f)
>-------------------
> program Main
> integer result,number
>
> number = 4
> write(6,'(''number(F)='',I1)') number
> result = testit(number)
> write(6,'(''result(F)='',I1)') result
> stop
> end
>
>
>C Function (testit.c)
>-------------------
>#include "stdio.h"
>
>int testit_(number)
>int *number;
>{
> int result;
>
> result = *number * 2;
>
> printf("number(C)=%d\nresult(C)=%d\n",*number,result);
>
> return result;
>}
>
>
>Compile and run Procedure
>-------------------
>gcc -c testit.c
>g77 number.f testit.o -o number
>.../number
>
>
>Run Result
>-------------------
>number(F)=4
>number(C)=4
>result(C)=8
>result(F)=0
>
You have to declare your C function in FORTRAN like this:
EXTERNAL TESTIT
INTEGER TESTIT
This will tell the FORTRAN compiler that the routine TESTIT (Fortran is case
insensitive) is an external subroutine returning an integer.
HTH,
JJ
==========================================================
*NOTE*
My Email return address is not correct
in order to avoid mass mailings...
These are the correct addresses
(but with dashes between all letters):
Jan Just (JJ) Keijser
Unix Support Engineer / Configuration Manager
Logica Inc. - Lexington MA
SMTP: [EMAIL PROTECTED]
Just to confuse some of those junkmailers:
[EMAIL PROTECTED]
Your mouse has moved. Windows must be restarted for
the change to take effect. Reboot now? [OK]
My views are my own...
flames > /dev/null 2>&1
==========================================================
------------------------------
From: [EMAIL PROTECTED] (Kaz Kylheku)
Subject: Re: linuxthreads and mutexes
Reply-To: [EMAIL PROTECTED]
Date: Tue, 07 Sep 1999 16:24:30 GMT
On Tue, 07 Sep 1999 14:18:35 GMT, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>Hello,
> Is it possible to use mutexes in linuxthreads to
>synchronize threads of different processes?
No.
>Here is what I am doing:
> - Create a memory-mapped file and put a mutex in
>it.
> - Have two processes open this memory-mapped
>file and use the mutex for synchronization.
Not possible. The mutex contention resolution algorithm in LinuxThreads relies
on queues that are maintained in user space. Items in the queues are
structures that represent threads. The queue contents only make sense in the
context of a single process.
This conforms to POSIX, since POSIX does not require mutexes to support
access from multiple processes.
------------------------------
From: Simon Kwan <[EMAIL PROTECTED]>
Subject: Development documentation Where to start?
Date: Wed, 08 Sep 1999 00:18:15 +0800
Hi Greeting,
Now, I have 72 hours of Linux experience and counting...
Appreciate some more pointer on the structure of Linux documentation
so that I know where to look for what (for application programming C,
Perl, Python)
I use the Mandrake-Linux 6.0 dist. and also got the RedHat Power tools
CD.
Specifically, at this stage, where is the documentation for
a. the kernel driver
b. Linux OS level API (Is there an tutorial and ref. on those CD or
downloadable from WEB?)
c. Linux shell command (e.g mv, ls. The on line man is running. Any
better tutorial?)
Basically, I would appreciate pointers (on my the distribution CD or
WEB) toward area:
a) tutorial and other education/training/instruction materials on
Linux/Unix programming and system admin.
b) manuals (man pages or is there an HTML style man with search and
indexing?)
TIA
Simon
------------------------------
From: Fernando Ortega Bellosta <[EMAIL PROTECTED]>
Subject: Struct packet_type(Question)
Date: Tue, 7 Sep 1999 19:28:22 +0200
In "net/core/dev.c" are defined two packet_type structs(ptype_all &
ptype_base[16]). What are these structs used for?
I suppose that they are used to kick the Ip datagrams up, is that right?.
What is the different between them?
The second question is about the device buffers, there are three priority
buffers per device, so is it true that all the queued packets should be in
one of these buffers?
In such case, what is the meaning of the queue "sk->write_queue" of the
sock struct.
When do I queue in the sk->write_queue and when in the device queues.
Thank You very much for your help.
I really need it.
--
Fernando Ortega Bellosta
[EMAIL PROTECTED]
------------------------------
From: "David Castells i Rufas" <[EMAIL PROTECTED]>
Subject: DCE RPC on Linux
Date: Tue, 7 Sep 1999 19:01:44 +0200
I am trying to develop a distributed object system similar to DCOM for
Linux. As you know DCOM is based on DCE RPCs. I know that there is already a
DCOM for Linux from software AG (which is not open source) and that Linux
RPCs are Sun's and not DCE. I know as well that there are posible
alternatives such CORBA. Nevertheless I am interested on developing a kind
of DCOM using DCE RPCs.
I downloaded the DCE RPC code from www.
I tried to compile DCE RPC implementation for Linux in Redhat 6.0 without
success.
Could you help me trying to identify the problems that I have ?
I get the following error:
make[1]: Entering directory '/projects/dce-1.1/src/libdce'
make[1]: *** No rule to make target '../rpc/runtime/libnck.a', needed by
'libdcerpc-0.1.so'. Stop.
I guess that the libnck.a has to be compiled first so I go to ../rpc/runtime
and I do make. Then I get the
following error:
In file included from commonp.h:92,
from rpctimer.c:41:
LINUX/sysconf.h:54: iovec.h: No such file or directory
I think that iovec.h include file is no longer part of Linux include files,
so I silently remove the
include sentence from sysconf.h.
When I do make again I get the following error:
In file included from LINUX/sysconf.h:78,
from commonp.h:92,
from rpctimer.c:71:
/usr/include/unistd.h:1001: conflicting types for 'pthd4exc_atfork'
../../../export/include/dce/pthread_dce_exc.h:532: previous declaration of
'pthd4exc_atfork'
Thank you very much
------------------------------
From: Andreas Jaeger <[EMAIL PROTECTED]>
Subject: Re: Problem compiling GNU/libc6 on Debian
Date: 07 Sep 1999 18:50:34 +0200
>>>>> Olivier Girondel writes:
> Hi,
> I'm running Debian (slink), and i tried this days to compile the
> lic6...
Why? Debian comes with glibc already.
> The compilation process is ok (egcs-2.91 or gcc-2.95) but things go
> wrong with the dynamic
> linker -> unresolved _dl_initial_searchlist (right after ldconfig),
> which really is a *big* problem (when you
> don't have sash... :-))
Are you using the right version - or are you trying to install an
older version than the one you already have. This would explain the
above message.
> I'm kinda stuck right now, so "everything who could help would be...
> helpful..."
Reinstall the glibc from slink - and read the INSTALL, README and FAQ
files of glibc before you try next time.
Andreas
--
Andreas Jaeger [EMAIL PROTECTED] [EMAIL PROTECTED]
for pgp-key finger [EMAIL PROTECTED]
------------------------------
From: Marcus Sundberg <[EMAIL PROTECTED]>
Subject: Re: Problems building a cross-compiler
Date: Tue, 07 Sep 1999 16:48:07 +0200
[EMAIL PROTECTED] wrote:
> Don't build in the source directory.
Why do you even bother posting when you don't have a single clue?
It's _obvious_ that he isn't building in the source directory, and
even if he did he wouldn't get this error because of it!
The problem is that you need headers from the c-library to compile
some of the run-time parts of gcc. The solution is to either get
the headers for the c-library from a binary package. Or you can
build gcc with 'make cross; make cross-install', which will hopefully
install enough for you to build the c-library. When the c-library
is built and it's headers installed you can finish the gcc build.
//Marcus
--
===============================+====================================
Marcus Sundberg | http://www.stacken.kth.se/~mackan/
Royal Institute of Technology | Phone: +46 707 295404
Stockholm, Sweden | E-Mail: [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Richard)
Crossposted-To: comp.unix.programmer,comp.os.linux.development.apps
Subject: Re: Richards Stevens died
Date: Tue, 7 Sep 1999 13:23:57 -0400
According to my newsreader, [EMAIL PROTECTED] posted...
//
// What tragic news. I feel like we should do something in his memory.
>From the funeral announcement at www.bigdealclassifieds.com :
<QUOTE>
The family asks that in lieu of flowers, donations be made in
Richard's name to
Habitat for Humanity
2950 E. 22nd Street
Tucson, AZ 85713
</QUOTE>
I can't personally think of a more fitting way to show our collective
appreciation for his many good works and generous spirit than to keep
the folks at HfH very busy processing donations "in Richard's name."
--
"The road to wisdom? Well it's plain, and simple to express:
Just err and err and err again, but less and less and less."
-Piet Hein
Disclaimer: These are simply some of my personal opinions.
ObURL: http://home.earthlink.net/~huddler
------------------------------
From: FM <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc
Subject: Re: TAO: the ultimate OS
Date: Tue, 07 Sep 1999 17:26:27 GMT
"Vladimir Z. Nuri" <[EMAIL PROTECTED]> wrote:
> my "opponents" are making them true by fiat. I am merely pointing
> out unjustified assumptions & dogmas that permeate current OS
> & software design. if you consider anything true that I am making
> "false by fiat" I will look at the evidence and show you where
> you are wrong<g>
You don't show where anybody's wrong. You simply assert your strong
belief in the falsity of their opinion and treat the assertion as
proof. It's not like we should throw away years worth of research in
computer science because of few uninformed.
> I proposed
> an alternative to existing dogma & conventional wisdom.
Well I don't see anything you've proposed that haven't been
implemented, of course other than your extraordinary claims
of "absolutes."
> I am
> saying, "imagine a universe exists in which you are not confined to
> the boxes you now believe you are confined within"<g>
Rather you're imaginging a universe where Genie comes up you and grants
an infinite number of wishes.
Dan.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Paul D. Smith)
Crossposted-To: comp.os.linux.setup,comp.os.linux.development.apps
Subject: Re: Help Compiling GCC 2.95.1 (Please!! )
Date: 07 Sep 1999 10:12:43 -0400
Reply-To: [EMAIL PROTECTED]
%% Nix <$}xinix{[email protected]> writes:
>> 7. 'make bootstrap MAKE="make -j 4" -j 4'
n> Is `make' GNU make? Solaris make is (still!) not POSIX-compliant
n> enough, and gcc building isn't tested very much with non-GNU makes
n> to my knowledge.
Just a pedantic note: Solaris make _is_ very likely POSIX-compliant.
However, the POSIX.2 definition of make is so anemic that probably any
make is conformant. For example, it doesn't even have "include".
--
===============================================================================
Paul D. Smith <[EMAIL PROTECTED]> Network Management Development
"Please remain calm...I may be mad, but I am a professional." --Mad Scientist
===============================================================================
These are my opinions---Nortel Networks takes no responsibility for them.
------------------------------
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: 7 Sep 1999 18:54:22 GMT
In comp.os.misc EdToy <[EMAIL PROTECTED]> wrote:
: I didn't accuse him of anything (I'm sure he's a fine techie). I just
: said how it's all working out in the hands of the capitalists: those who
: do the work, don't get the rewards of it. And my gut feeling is that
: Vladimir is more on the exploitation side than the techie side.
ooooooooooh ouch. I have put in a lot of effort into
promoting this effort with largely nothing but a bunch of
jeers & harassment to show for it so far.
I hereby promise to share whatever capitalistic revenues
result in a fair/equitable manner. however, I can't see
how such revenues would be generated except long, long
into the future. linux is something
like 7 years old, and it is only recently that a company
had an IPO related to it (yes there were service companies
eking out some cash prior to this date).
I admit I would like to find a way to channel a capitalistic
incentive into the open source agenda. lacking any clear
path to do so, however, I am donating my own time & effort
& asking all others doing the same.
I totally sympathize with your interest in avoiding
exploitation. I share a similar paranoia for it. one
question: how could anything I am trying to accomplish
be exploited?
I agree the problem of exploitation is a serious one for the
open OS community to consider.. the red hat angle
is going to become problematic in the future I believe.
merely throwing out a lot of IPO stock rights to
various developers possibly only increased the acrimony.
p.s. I was once in a development community for one of
the original famous "open source" projects. I do
agree that the founder exploited the labor of the
participants. it is not a situation I want to recreate.
I would like to create something "quite to the contrary".
again the mailing list is open to anyone who shares the
vision, and would like to help define it. we had a
small breakthrough recently as someone with initials
J.A. who has a very excellent programming resume signed
up to participate & contribute status reports on his
own efforts.
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
state of the art OS research email http://www.egroups.com/groups/os-edge/
Tao OS / Taos / the transcendental OS http://www8.pair.com/mnajtiv/tao.html
------------------------------
** 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
******************************
