Linux-Development-Sys Digest #683, Volume #7 Thu, 16 Mar 00 13:13:24 EST
Contents:
Windows CE target for GCC? ("Tom & Ada Campbell")
Tutorial on using the GCC code generator available? ("Tom & Ada Campbell")
Re: Tutorial on using the GCC code generator available? (Martin Kahlert)
Re: g77 warning: multiple common of (Li Long)
Re: Buggy 2000 Setup Solutions/Programming Links (Michael Schmitz)
I/O Problem (Francisco Obispo)
Driver for a network card - "Rix 100Base-Tx" (Wei Cheng)
ramdisk: releasing memory (Thomas Eschenbacher)
Re: g77 warning: multiple common of
=?iso-8859-1?Q?=B4=2A=2A=2A=2A=2A=2A=B4?= (Richard Maine)
where Linux swap when nfs rooted? (Zirong WANG)
Re: g77 warning: multiple common of
=?iso-8859-1?Q?=B4=2A=2A=2A=2A=2A=2A=B4?= (Richard Maine)
Re: Stupid Question (fvw)
Re: I/O Problem (Thomas Eschenbacher)
Re: Stupid Question ("Syntax")
How to test a device driver (Limin Zhang)
----------------------------------------------------------------------------
From: "Tom & Ada Campbell" <[EMAIL PROTECTED]>
Subject: Windows CE target for GCC?
Date: Thu, 16 Mar 2000 08:08:25 GMT
Is there version of GCC that can output Windows CE executables?
thanks.
tom campbell
------------------------------
From: "Tom & Ada Campbell" <[EMAIL PROTECTED]>
Subject: Tutorial on using the GCC code generator available?
Date: Thu, 16 Mar 2000 08:11:05 GMT
Just to make sure you know I'm a befuddled newbie--is there a tutorial
somewhere on using the GCC back end? That is, I'd like to create a new
language that uses the GCC code generator and I'm hoping against hope
someone's documented at least some of the process.
thanks.
tom campbell
------------------------------
From: [EMAIL PROTECTED] (Martin Kahlert)
Subject: Re: Tutorial on using the GCC code generator available?
Date: 16 Mar 2000 09:45:59 GMT
Reply-To: [EMAIL PROTECTED]
[Posted and mailed]
In article <tG0A4.4304$[EMAIL PROTECTED]>,
"Tom & Ada Campbell" <[EMAIL PROTECTED]> writes:
> Just to make sure you know I'm a befuddled newbie--is there a tutorial
> somewhere on using the GCC back end? That is, I'd like to create a new
> language that uses the GCC code generator and I'm hoping against hope
> someone's documented at least some of the process.
Try starting from http://gcc.gnu.org/readings.html
(especially Porting GCC for Dummies)
and use the manual - there is a big part of description, too.
Hope that helps,
Martin.
--
The early bird gets the worm. If you want something else for
breakfast, get up later.
------------------------------
From: Li Long <[EMAIL PROTECTED]>
Crossposted-To: comp.lang.fortran
Subject: Re: g77 warning: multiple common of
Date: Thu, 16 Mar 2000 13:02:31 +0100
This is a multi-part message in MIME format.
==============20E6688794A6512806D7BF45
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Richard Maine wrote:
>
> Li Long <[EMAIL PROTECTED]> writes:
> > I guess this is only a problem with g77 compile.
>
> I really doubt it. People use COMMON with g77 all the time.
> But I don't see sufficient data to tell.
>
> > Now I put it in a very simpler test code, which compose 3 files
> > TEST3.INC, test3.f test4.f, see attachment .
>
> These files also include something called TYPES.INC, which you
> haven't shown. Why am I suspicious that there might be something
> relevant in there? My main news server seems to be down today,
> and its awkward for me to get things form this one to where I can
> try them out (perhaps I'll try anyway), but I'm betting that
> your code will work fine if I comment out the includes for
> TYPES.INC.
>
> --
> [EMAIL PROTECTED]
Sorry I forget to attach TYPES.INC,(which can be ignored), the code
contains
four files TYPES.INC TEST3.INC test3.f test4.f. I attach them here
again.
If I comment out INCLUDE TYPES.INC, als you sugested,
g77 -o test3 test3.f test4.f
then the warning messages are like these:
/tmp/ccpt0AS2.o: warning: common of `oppar_' overriding smaller common
/tmp/ccLOvz5k.o: warning: smaller common is here
otherwise I get
/tmp/ccpcbcqK.o: warning: multiple common of `oppar_'
/tmp/ccAqSW6R.o: warning: previous common is here
I use SuSe linux 6.1 g77 v0.5.24. May it is a problem of suse linux
fortran compiler, as point out
by Toon Moene ([EMAIL PROTECTED]) in this group.
If these warning messages are not harmful, it seems so, how can I get
rid of it.
Thanks!
L. Long
==============20E6688794A6512806D7BF45
Content-Type: text/plain; charset=us-ascii;
name="TYPES.INC"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="TYPES.INC"
IMPLICIT DOUBLE PRECISION (A-H,O-Z)
==============20E6688794A6512806D7BF45
Content-Type: text/plain; charset=us-ascii;
name="TEST3.INC"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="TEST3.INC"
** COMMON BLOCK FOR O PRECIPITATION NP>NX (IN INPUTFILE)
COMMON /OPPAR/OTOTAL,VSIO2
SAVE /OPPAR/
==============20E6688794A6512806D7BF45
Content-Type: text/plain; charset=us-ascii;
name="test3.f"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="test3.f"
PROGRAM TEST3
INCLUDE 'TYPES.INC'
DIMENSION S(1),R(1)
CHARACTER*10 AUSG
INCLUDE 'TEST3.INC'
EXTERNAL SUB1
VSIO2=4.5D-11
CALL SUB1(1,R)
WRITE(6,'(5E14.4)') VSIO2
END
==============20E6688794A6512806D7BF45
Content-Type: text/plain; charset=us-ascii;
name="test4.f"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="test4.f"
SUBROUTINE SUB1(Y,R)
* .. Scalar Arguments ..
INCLUDE 'TYPES.INC'
INCLUDE 'TEST3.INC'
* .. Executable Statements ..
R=Y**(1./3.)*VSIO2
RETURN
END
==============20E6688794A6512806D7BF45==
------------------------------
From: Michael Schmitz <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.powerpc,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Buggy 2000 Setup Solutions/Programming Links
Date: Thu, 16 Mar 2000 13:23:06 +0100
Reply-To: [EMAIL PROTECTED]
"[EMAIL PROTECTED]" wrote:
>
> - I am wondering if someone can give me some instructions on disabling the
> GNOME graphical login, as it is causing strange problems on my machine eg.
> flashing between the terminal and X continuously, and I have to reboot,
> corrupting my system in the process. At least on the x86 RedHat
> distribution, you can select whether you want a graphical login or not.
> I presume this is in some startup script, can someone help?
Set the following in /etc/inittab:
id:3:initdefault:
> - Will there ever be a way to let Linux *fully* take over the mac hardware,
> without any hfs partitions and MacOS software (BootX extensions and such)?
Use yaboot.
Michael
------------------------------
From: Francisco Obispo <[EMAIL PROTECTED]>
Subject: I/O Problem
Date: Thu, 16 Mar 2000 09:32:03 -0400
Hello.... I have a problem with a driver that I�m developing....
When I use outb_p(0x8f,0x300) (Write 0x8f to 0x300) it seems like it
does something but it......
Do I have to reserve this memory space for my card??? do I do this by
calling kmalloc??? How do I tell linux that my ISA card is going to use
0x300????
Thanx
------------------------------
From: Wei Cheng <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Driver for a network card - "Rix 100Base-Tx"
Date: Fri, 17 Mar 2000 00:42:50 +1100
I am currently using a pcmcia network card - "Rix 100Base-Tx"
and the card come with driver call - "lnax100.o"
however there is a problem when i try to use it, it's not compiled
under the version of my linux(Slackware 7.0 - 2.2.13)
so, my problem is can i recompile it to a *.c file
such as "lnax100.c", and then compile it again under the version
of my linux
thanks for any reply
------------------------------
From: Thomas Eschenbacher <[EMAIL PROTECTED]>
Subject: ramdisk: releasing memory
Date: Thu, 16 Mar 2000 16:49:40 +0100
Hello,
I have a problem with the ramdisk driver of the kernel 2.2.14. I am
booting from an initrd (1.5MB) and within the initrd I create a
temporary ramdisk /dev/ram2 (8MB) and mount it as /tmp.
The creation of the disk works fine, but the buffers never get released
after I unmount /tmp so that the system wastes memory in unused buffers.
A "rdev -r /dev/ram2" and a little C-program that calls ioctl with
cmd=BLKFLSBUF behaves like a NOP :-(
If I create/mount/unmount further temporary ramdisks with bigger size
the "buffer" size shown with "free" increases up to 16MB, the default
ramdisk size our kernel uses. Our system is an embedded PC with only
32MB and no swap so that this really hurts!!!
Is there any way to _really_ get rid of the memory used in the ramdisk?
(note that we have to boot from it, so I can't make the rd driver a
module and rmmod it!)
please help,
Thomas.
--
________________________________________________________________________
Thomas Eschenbacher [EMAIL PROTECTED]
------------------------------
Crossposted-To: comp.lang.fortran
Subject: Re: g77 warning: multiple common of
=?iso-8859-1?Q?=B4=2A=2A=2A=2A=2A=2A=B4?=
From: Richard Maine <[EMAIL PROTECTED]>
Date: 16 Mar 2000 07:55:16 -0800
Li Long <[EMAIL PROTECTED]> writes:
> Richard Maine wrote:
> > These files also include something called TYPES.INC, which you
> > haven't shown. Why am I suspicious that there might be something
> > relevant in there?
> Sorry I forget to attach TYPES.INC,(which can be ignored),...
> If I comment out INCLUDE TYPES.INC, als you sugested, ...
> then the warning messages are like these:
> /tmp/ccpt0AS2.o: warning: common of `oppar_' overriding smaller common
> /tmp/ccLOvz5k.o: warning: smaller common is here
>
> otherwise I get
> /tmp/ccpcbcqK.o: warning: multiple common of `oppar_'
> /tmp/ccAqSW6R.o: warning: previous common is here
>
> I use SuSe linux 6.1 g77 v0.5.24. May it is a problem of suse linux
> fortran compiler, as point out
I tend to be very dubious when someone who is asking for debugging
help tells me what "can be ignored." AFter all, if they knew what
the problem really was, why would they be asking for help.
If types.inc "can be ignored", then why are there different results
with it than without it? That sounds suspiciously like it affected
something. The result without it also seems strange, implying that
the two occurances of the COMMON are different sized. I'd see
nothing here that would cause that. But then...you still didn't actually
post the files that you compiled and that I can try. You posted files
with the include of types.inc, and then described what happened if you
commented out the include(s). Are you sure you commented them *ALL* out?
If types.inc had something like implicit statements, and it was
included in one subroutine but not the other, then I would expect a
message much like the one you got.
Anyway, I cannot reproduce the results, whether its because the
posted files aren't exactly like what was used, or whether because
of differences in system configuration. I am still suspicious,
as mentioned above, that the posted files aren't exactly as tried.
But it is certainly possible that, as Toon mentioned, there are
problems with the S.U.S.E. configuration, and that I won't ever
be able to reproduce the results. I haven't yet seen enough to
convince me that it is necessarily the case, but it could be.
I would recommend against just ignoring messages like this. They can
indicate serious errors (such as my suggestion above that the
types.inc with an implicit might be in one file but not the other -
that would mess up lots of things).
--
[EMAIL PROTECTED]
------------------------------
From: Zirong WANG <[EMAIL PROTECTED]>
Subject: where Linux swap when nfs rooted?
Date: Thu, 16 Mar 2000 17:21:42 +0100
I am just wondering where Linux can swap when my system is rooted
with NFS.
I deleted the line concerning the swap partition from etc/fstab.
Is it possible to specify a remote device for the swap ?
Zirong
------------------------------
Crossposted-To: comp.lang.fortran
Subject: Re: g77 warning: multiple common of
=?iso-8859-1?Q?=B4=2A=2A=2A=2A=2A=2A=B4?=
From: Richard Maine <[EMAIL PROTECTED]>
Date: 16 Mar 2000 08:26:33 -0800
I just cancelled a reply wherein I failed to notice that types.inc
was included in Li Long's previous post. Hope my cancelled one
doesn't slip out anyway. If it did, please ignore it. I now see
that Li Long's second post did in fact include types.inc.
Li Long <[EMAIL PROTECTED]> writes:
> Sorry I forget to attach TYPES.INC,(which can be ignored)...
> If I comment out INCLUDE TYPES.INC, als you sugested, ...
> then the warning messages are like these:
> /tmp/ccpt0AS2.o: warning: common of `oppar_' overriding smaller common
> /tmp/ccLOvz5k.o: warning: smaller common is here
> otherwise I get
> /tmp/ccpcbcqK.o: warning: multiple common of `oppar_'
> /tmp/ccAqSW6R.o: warning: previous common is here
I dispute the statement that types.inc "can be ignored". I tend to
be dubious whenever anyone who is asking me for debugging help
tells me that something "can be ignored". If they knew what the problem
was, they wouldn't be looking for help. Indeed, as a general rule,
that serves to imediately focus on that portion as something that
the user perhaps incorrectly ignored.
The types.inc has an implicit statement that affects the variables in
the COMMON. The messages you got when you claim to have commented
out the include of types.inc look very much like what I would expect if
you commented out only one of the two cases of it. In that case, the
two copies of your COMMON would be seriously inconsistent (and wouldn't
work right at all) and the compiler would be correct to complain.
I notice some apparently unrelated problems. You pass an integer 1
as an actual argument for a real (or double precision, depending
on types.inc) dummy argument. This won't work. But it probably
doesn't have anything to do with your COMMON problem (and might be
just an artifact of the simplified version of the code).
Also, you pass an array actual argument to a scalar dummy. This
issue has been discussed here recently. It *PROBABLY* will work,
but is nonstandard and is nopt guaranteed to work in all environments.
You would be much to pass S(1) or R(1) (I forget which it was), which
tells the compiler exactly what you really want, rather than assuming
that the compiler will just get it right anyway. In this case, the
compiler probably will get it right, but...
> I use SuSe linux 6.1 g77 v0.5.24. May it is a problem of suse linux
> fortran compiler, as point out
Could be, I suppose. None of the above issues really explain your
original error message, which I still can't duplicate, perhaps
because I don't have SuSe installed on anything handy.
> If these warning messages are not harmful, it seems so, how can I get
> rid of it.
Perhaps by rebulding the compilers from scratch? Yes, I know that's
not really a helpful suggestion. If a bad SuSe installation is the
problem, that *WOULD* fix it, but that's a lot of bother. (One of
the things I've come to like about at least RedHat linux, and it might
apply to some other linux distributions also, is that it comes "out
of the box" with most of the tools that I tend to use pre-installed.
Not all of the tools - for example, no f90/f95 compiler - but a lot
larger percentage than comes bundled with some of the other unixes).
Sorry, but if it is a SuSe installation problem, I don't know how
to trivially fix it or make the messages go away.
--
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (fvw)
Subject: Re: Stupid Question
Reply-To: [EMAIL PROTECTED]
Date: Thu, 16 Mar 2000 16:27:43 GMT
In <8aovos$5ra$[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
>I am unable to find any information on the subject of include files. How
>does g++ know where to look for include files (header files). In
>DOS/Windows, it was specified in the environment variables (INCLUDE), but
>there aren't any relevant variables specified in Linux (at least not in
>mine). I can compile, so I know the compiler is looking someplace for the
>files, but I don't know why.
AFAIK, there's probably some env variable for it, but for most things
the default, /usr/include is ok. If you want more, find out the vars name,
or use the -I param.
BTW, there are no stupid questions, only stupid users.
--
Frank v Waveren
[EMAIL PROTECTED]
ICQ# 10074100
------------------------------
From: Thomas Eschenbacher <[EMAIL PROTECTED]>
Subject: Re: I/O Problem
Date: Thu, 16 Mar 2000 17:30:16 +0100
Francisco Obispo wrote:
> [...]
> When I use outb_p(0x8f,0x300) (Write 0x8f to 0x300) it seems like it
> does something but it......
you are addressing i/o space, not memory!
> Do I have to reserve this memory space for my card??? do I do this by
> calling kmalloc??? How do I tell linux that my ISA card is going to use
> 0x300????
so you should register your driver that it uses this i/o port range. For
an example, take a look at /usr/src/linux/drivers/rtc.c and search for
"request_region".
It is likely that you want to do something like this in your driver's
init function:
#define PORT 0x300
#define RANGE 8
#define NAME "my_driver"
if (check_region(PORT, RANGE) == -EBUSY) {
printk(NAME "i/o ports are in use!");
return -EBUSY;
}
request_region(PORT, RANGE, NAME);
... and somewhere in the close function:
release_region(PORT, RANGE);
bye,
Thomas.
--
________________________________________________________________________
Thomas Eschenbacher [EMAIL PROTECTED]
------------------------------
From: "Syntax" <[EMAIL PROTECTED]>
Subject: Re: Stupid Question
Date: Thu, 16 Mar 2000 17:22:14 GMT
The answer to your "stupid question" well maybe not an answer but i will
take a shot.
Chances are it is hardcoded into gcc/g++ to add other directoried where
"headers" are use the "-I" (eye) tag then directory.
gcc llave.c -o gato -I/home/myincludes/
-Lucas
[EMAIL PROTECTED]
John <[EMAIL PROTECTED]> wrote in message
news:8aovos$5ra$[EMAIL PROTECTED]...
> I am unable to find any information on the subject of include files. How
> does g++ know where to look for include files (header files). In
> DOS/Windows, it was specified in the environment variables (INCLUDE), but
> there aren't any relevant variables specified in Linux (at least not in
> mine). I can compile, so I know the compiler is looking someplace for the
> files, but I don't know why.
>
> Thanks,
>
> John Marasco
>
>
------------------------------
From: Limin Zhang <[EMAIL PROTECTED]>
Subject: How to test a device driver
Date: Thu, 16 Mar 2000 09:46:37 -0800
Hello!
Can some body tell me how to test a device driver like"scull" described
in the book by Rubini? In other words, I do not know how to communicate
with the driver. Thanks.
Limin
------------------------------
** 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
******************************
