Linux-Development-Sys Digest #909, Volume #7 Fri, 26 May 00 18:13:16 EDT
Contents:
Re: Writing driver for AMCC 5933 based PCI card (Iwo Mergler)
Re: virt_to_phys vs pgd/pmd/pte_offset calls (Timur Tabi)
How to debug a loadable module in Linux ? (Arne =?iso-8859-1?Q?Legern=E6s?=)
Re: Kernel optimization for different processors (bill davidsen)
Very simple linux Scheduler question ("Jason Johnson")
Re: ps does not show all processes ("Jackson C. Allen")
Re: ps does not show all processes ("Jackson C. Allen")
Re: Linux Bootable CD
Re: Kernel module-user process comms (Dave Donaghy)
Re: Call stack in C ("Norm Dresner")
Re: Need ideas for university funded project for linux (Jan Knutar)
Re: Need ideas for university funded project for linux (Jan Knutar)
Re: Need ideas for university funded project for linux (JEDIDIAH)
Re: Linux Bootable CD (Nigel Kukard)
Re: ps does not show all processes (Robert Redelmeier)
How do I patch the kernel? ("Jay Randall")
Re: Need ideas for university funded project for linux (Nix)
Re: Need ideas for university funded project for linux (Nix)
Re: What !@#$ moron colorised g++? (Nix)
Re: What !@#$ moron colorised g++? (Nix)
Re: What !@#$ moron colorised g++? (Nix)
Re: ps does not show all processes (Nix)
Re: Two really easy (I'm sure) questions (Nix)
Re: Checking for I'm swapped - number of pages swapped out (Nix)
XIM info ("William A. McKee")
RPC : sendmsg returned error 101 ("Aurelie Fonteny")
----------------------------------------------------------------------------
From: Iwo Mergler <[EMAIL PROTECTED]>
Subject: Re: Writing driver for AMCC 5933 based PCI card
Date: Fri, 26 May 2000 13:44:26 GMT
Reply-To: [EMAIL PROTECTED]
Julian Back wrote:
>
> Thanks for reading my post Bill.
>
> >What version(s) of Linux? DMA to/from kernel or user buffer? Does it
> >fail after ~ the same number of transfers each time? ...
>
> I'm developing on Red Hat 6.1 which I think is kernel 2.2.12 but I've also
> 2.2.14, it made no difference.
>
> I'm DMAing to and from kernel buffers allocated with get_free_page(GFP_KERNEL).
> I've tried adding the GFP_DMA flag but it makes no difference and my
> understanding is that this is only necessary for ISA DMA.
>
> The number of transfers before crashing can vary a lot. Sometimes it has done
> 3 million transfers before crashing other times it crashes almost immediately.
>
> Julian
>
I don't know the AMCC5933, but PLX published a 6-page list of mostly DMA
related bugs in their PCI9080. A few of them could be responsible for
this kind of problems. Maybe ask AMCC for known bugs/workarounds?
Regards,
Iwo
------------------------------
From: Timur Tabi <[EMAIL PROTECTED]>
Subject: Re: virt_to_phys vs pgd/pmd/pte_offset calls
Date: Fri, 26 May 2000 14:49:38 GMT
In article <8gle14$o02$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Robert Kaiser) wrote:
> Yes. One thing to keep in mind though: the virtual->physical
> mapping that you can determine with the above algorithm may change
> at any time, for instance, as a result of swapping. So, unless you
> do something about it (e.g. disable interrupts or -better- lock the
> pages in memory), the physical address corresponding to a user space
> virtual address may change at any time without warning.
Good point. But since only a driver can travers the pgd/pmd/pte
structures, aren't interrupts and paging disabled while the driver is
running? That is, giving a user-space virtual address that is passed to
my driver, I can obtain the physical address, do something with that,
and then exit. The next time my driver is called, I'll need to traverse
those structures again.
Can I call a kernel API to lock that physical page down, so that I don't
have to traverse the structures every time? If so, what is it called,
and will I have to unlock the page before the application exits (that
is, can I lock an application's memory without any consequences other
than increase memory fragmentation?)
>
> Rob
>
> ----------------------------------------------------------------
> Robert Kaiser email: rkaiser AT sysgo DOT de
> SYSGO RTS GmbH
> Klein-Winternheim / Germany
>
>
--
Timur "too sexy for my code" Tabi
Remove "nospam_" from my email address when replying
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Arne =?iso-8859-1?Q?Legern=E6s?= <[EMAIL PROTECTED]>
Subject: How to debug a loadable module in Linux ?
Date: Fri, 26 May 2000 16:55:28 +0200
Hello,
I am going to develop an X25 protocol stack as a loadable device driver
module, but I have
as yet little experience with Linux and Linux tools.
If I understand it correctly, such a module will run as a part of the
kernel process, and I assume
this must impose strong restrictions on how it can be debugged.
What tools do Linux developers usually use when writing and debugging
loadable kernel modules ?
Does there exist any source code level debuggers for this type of
debugging ?
Thanks.
------------------------------
From: [EMAIL PROTECTED] (bill davidsen)
Subject: Re: Kernel optimization for different processors
Date: 26 May 2000 15:57:11 GMT
In article <8ghs5l$38v$[EMAIL PROTECTED]>,
Timothy J. Lee <see-signature-for-email-address---junk-not-welcome> wrote:
| On i386, selecting, for example, "Pentium" in the processor options
| causes the use of -march=i586 instead of -mcpu=i586 when compiling.
| This generates code that may not run on 386 or 486 processors.
Right, you would normally select the lowest processor you want to
support as the compile target.
| Would it be useful to have an "optimize for" option that causes the
| use of optimization for a specific processor, but not actually using
| instructions that other members of the processor family won't be able
| to use (i.e., for i386, use -mcpu= instead of -march=)?
The nice thing is that this is in the Makefile and you can make this
change if you wish. The argument advanced for arch= is that cpu= does
not get as aggressive as arch= in code generation, and therefore the
performance may be less. You obviously understand what the options do,
and that was the decision. The argument is probably valid, you get the
biggest gain in performance by using the least portable features.
I don't think you want to lose all the advanced register use in P6
cores unless you really have to, in which case cpu= buys very little. As
I recall the alignment on 386 needs more NOPs, which bloat the code
slightly. And there are sequences which block on 386 and don't on P6.
If you try it you might report the kernel size with 386, cpu= and
arch=, as well as any detectable performance issues.
--
bill davidsen <[EMAIL PROTECTED]> CTO, TMR Associates, Inc
"Doing interesting things with little computers since 1979"(tm)
The hardest test of maturity is knowing the difference between
resisting temptation and missing a once-in-a-lifetime opportunity.
------------------------------
From: "Jason Johnson" <[EMAIL PROTECTED]>
Subject: Very simple linux Scheduler question
Date: Fri, 26 May 2000 11:35:59 -0500
Hello,
I have question about the linux scheduler. I have been looking
through the source code, and as far as I can tell
it looks like, that each process has only a certain
amount of ticks it is allowed to run. When it runs out, it will not be
scheduled again until all other runable processes have also
used all of there ticks, at which point all the runable processes
have there quantums recalculated and the whole thing starts over.
Am I correct here (I am obviously ignoring processes that do IO,
or get put to sleep in general)?
Thanks in advance for indulging my curiosity.
Jason
------------------------------
From: "Jackson C. Allen" <[EMAIL PROTECTED]>
Subject: Re: ps does not show all processes
Date: Fri, 26 May 2000 17:02:48 GMT
I RTFM again an tried "ps aux" as root and other logins. It still does
not display anything about sendmail and some other processes I know are
running. I can find the hidden process in /proc.
Talk 2 U later.
Jack Allen
Ronald Cole wrote:
>
> rtfm. "ps aux" will show all processes.
>
> --
> Forte International, P.O. Box 1412, Ridgecrest, CA 93556-1412
> Ronald Cole <[EMAIL PROTECTED]> Phone: (760) 499-9142
> President, CEO Fax: (760) 499-9152
> My GPG fingerprint: C3AF 4BE9 BEA6 F1C2 B084 4A88 8851 E6C8 69E3 B00B
------------------------------
From: "Jackson C. Allen" <[EMAIL PROTECTED]>
Subject: Re: ps does not show all processes
Date: Fri, 26 May 2000 17:07:01 GMT
Apparently you did not read my message very well. Yes I have RTFM and
tried "ps ax" "ps aux" and "ps -ax" and many others. This is the reason
I have posted the question, because it does not work.
Talk 2 U later.
Jack Allen
Allin Cottrell wrote:
>
> "Jackson C. Allen" wrote:
> >
> > I am running RedHat 6.0 and just noticed "ps" does not show all
> > processes even when logged in as root. I have tried many different
> > combinations of options...
>
> Apparently you haven't yet attempted the "option" of reading the
> man page for ps. That document tells you, among other things:
>
> a show processes of other users too
> x show processes without controlling terminal
>
> Thus "ps ax" will do what you want ("ps aux" is probably better).
>
> --
> Allin Cottrell
> Department of Economics
> Wake Forest University, NC
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Re: Linux Bootable CD
Date: 26 May 2000 18:26:24 GMT
bill davidsen <[EMAIL PROTECTED]> wrote:
> You need to understand initrd and the eltorito bootable CD stuff.
> First, eltorito allows you to have a floppy image to boot, of 720, 1440,
> or 2880k size. That gets copied to a ramdisk, and the ramdisk kernel
> gets booted. I didn't say that well, but those are the things you need
> to know.
my problem is to get the floppy image to the exact size of 1440, 2880. do
have an elegant way for that?
thanks very much,
Michael
------------------------------
From: Dave Donaghy <[EMAIL PROTECTED]>
Subject: Re: Kernel module-user process comms
Date: Fri, 26 May 2000 20:32:06 +0100
This is a multi-part message in MIME format.
==============F4B47C724883694E0BDB14FF
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Alan Donovan wrote:
> What kind of comms? You have read/write methods on the device,
In my case, there is no physical device: the device is simulated by a
user process. So what I'm trying to do is simulate the kernel-physical
device comms (in this example, in a FS block device) by using a
kernel-user process comms part instead.
> Also you can create /proc files, which tend to be used for more elaborate
> diagnostics as a "window" onto the complete state of the driver.
The other reply also mentioned this (thanks, Sam!): while it wasn't (as
stated) exactly what I wanted, it certainly started me thinking. I've
looked at /proc today, and I reckon I can do half of the work directly
in /proc. (Registering a user process as a handler for an already
mounted filesystem.)
Fortunately or unfortunately, /proc will (hopefully) solve the half that
I *wasn't* originally talking about. I still need to do comms between
the kernel and the user handler. This can be async, as long as I can do
it in both directions.
I'll have a look at the "blocking ioctl" idea, but I was thinking today
that I could just get the user handler process to create a couple of
named pipes (in /proc) and use those.
In case anyone's interested, the bottom line idea is a file system
handler that simulates the actual file system with a user process.
I realise that this has been done before, but (afaik) not to real
completion. Plus it's mainly a learning experience for me in several
ways.
Thanks for the help: still willing to learn if anyone else has anything
to say!
Dave
==============F4B47C724883694E0BDB14FF
Content-Type: text/x-vcard; charset=us-ascii;
name="dave.donaghy.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Dave Donaghy
Content-Disposition: attachment;
filename="dave.donaghy.vcf"
begin:vcard
n:Donaghy;Dave
x-mozilla-html:FALSE
adr:;;;;;;
version:2.1
email;internet:[EMAIL PROTECTED]
x-mozilla-cpt:;-4160
fn:Dave Donaghy
end:vcard
==============F4B47C724883694E0BDB14FF==
------------------------------
Reply-To: "Norm Dresner" <[EMAIL PROTECTED]>
From: "Norm Dresner" <[EMAIL PROTECTED]>
Subject: Re: Call stack in C
Date: Fri, 26 May 2000 18:36:19 GMT
H. Peter Anvin <[EMAIL PROTECTED]> wrote in message
news:8gkqb9$4c8$[EMAIL PROTECTED]...
> Followup to: <BQPW4.78$[EMAIL PROTECTED]>
> By author: Joe Ceklosky <[EMAIL PROTECTED]>
> In newsgroup: comp.os.linux.development.system
> >
> > How can I get a call stack strace from a
> > running program. I want something like pstack
> > on Solaris for Linux
> >
>
> gdb -p <process_id>
> where
>
> -hpa
> --
That only works if you're running in the debugger and have stopped at a
breakpoint. Is there a function call that can be made from within a running
program that will write a stack-trace to some stream (stdout, stderr, etc.)?
Norm
------------------------------
From: [EMAIL PROTECTED] (Jan Knutar)
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
Date: Thu, 25 May 2000 12:52:48 GMT
On Sun, 21 May 2000 16:42:48 -0700, "Jack Kessler" <[EMAIL PROTECTED]>
wrote:
>to buy full hardware modems which are more expensive, harder to set up and
>are becoming harder to find.
Eh? Harder to setup?
When I installed my modem in win, I needed a bunch of driver diskettes to
get it to work.
I can't even remember installing the modem in Linux, all I can remember was
that I entered the phonenumber to my ISP, my name, password and the DNS
servers, and it worked. I think I saw a box saying something like "serching
for modem" the first time I ran it, but I'm not sure.
--
JK <[EMAIL PROTECTED]>
-
Any attempt to stretch fuel is guaranteed to increase headwind.
------------------------------
From: [EMAIL PROTECTED] (Jan Knutar)
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: Need ideas for university funded project for linux
Date: Thu, 25 May 2000 12:47:57 GMT
On Tue, 23 May 2000 10:21:52 -0400, Someone <[EMAIL PROTECTED]> wrote:
>use it for? Why don't any of the free ISPs support Linux? That would be
Could anyone please define 'free ISP' ??
--
JK <[EMAIL PROTECTED]>
-
Any attempt to stretch fuel is guaranteed to increase headwind.
------------------------------
From: [EMAIL PROTECTED] (JEDIDIAH)
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
Date: Fri, 26 May 2000 19:08:50 GMT
On Thu, 25 May 2000 12:52:48 GMT, Jan Knutar <[EMAIL PROTECTED]> wrote:
>On Sun, 21 May 2000 16:42:48 -0700, "Jack Kessler" <[EMAIL PROTECTED]>
>wrote:
>
>
>>to buy full hardware modems which are more expensive, harder to set up and
>>are becoming harder to find.
>
>Eh? Harder to setup?
Quite. My Phoebe hardware ISA modem came with very reasonable
defaults: no-pnp, com2. In this configuration I was able to slap it in
and use it immediately.
>
>When I installed my modem in win, I needed a bunch of driver diskettes to
>get it to work.
>
>I can't even remember installing the modem in Linux, all I can remember was
>that I entered the phonenumber to my ISP, my name, password and the DNS
>servers, and it worked. I think I saw a box saying something like "serching
>for modem" the first time I ran it, but I'm not sure.
One should never need to 'install' a modem. A real modem is as
simple as a device gets when it comes to compatibility.
(com2+ XT legacy gibberish notwithstanding of course)
--
In what language does 'open' mean 'execute the evil contents of' |||
a document? --Les Mikesell / | \
Need sane PPP docs? Try penguin.lvcm.com.
------------------------------
Date: Fri, 26 May 2000 22:24:01 +0200
From: Nigel Kukard <[EMAIL PROTECTED]>
Subject: Re: Linux Bootable CD
[EMAIL PROTECTED] wrote:
> bill davidsen <[EMAIL PROTECTED]> wrote:
>
> > You need to understand initrd and the eltorito bootable CD stuff.
> > First, eltorito allows you to have a floppy image to boot, of 720, 1440,
> > or 2880k size. That gets copied to a ramdisk, and the ramdisk kernel
> > gets booted. I didn't say that well, but those are the things you need
> > to know.
>
> my problem is to get the floppy image to the exact size of 1440, 2880. do
> have an elegant way for that?
>
> thanks very much,
>
> Michael
dd if=/dev/zero of=boot.img bs=1024 count=1440
I have an even better way to create a boot disk.....
do a mkinitrd, gunzip the image created, mout it, rm everyting cept the dev
directory,
copy the dev's u need with cp -a /dev/<device> <mountpoint>/dev
after that copy all the binaries to the disk u need, relink the linuxrc file
in the root dir
to point to the file that must be run, strip all the binaries, then use ldd
and find the libs
they use copy them over with objcopy --strip-all <src> <dest>, after this
umount
the initrd file, regzip with gzip -9 <file>, do a mkbootdisk to create a
bootable stiffy,
replace the initrd file. dd the whole disk to a file and use that file on the
bootabe cd,
but one thing i still can't figure out, is how to use the cd as a fs.
Regards
Nigel
Regards
Nigel
------------------------------
From: Robert Redelmeier <[EMAIL PROTECTED]>
Subject: Re: ps does not show all processes
Date: Fri, 26 May 2000 15:33:26 -0500
Jackson C. Allen wrote:
>
> Apparently you did not read my message very well. Yes I have RTFM and
> tried "ps ax" "ps aux" and "ps -ax" and many others. This is the reason
> I have posted the question, because it does not work.
Well, then I'd suspect you have a trojaned `ps` .
This is not unusual when a cracker obtains root.
My ps shows sendmail and everything else.
-- Robert
------------------------------
From: "Jay Randall" <[EMAIL PROTECTED]>
Subject: How do I patch the kernel?
Date: Fri, 26 May 2000 13:40:48 -0700
Hello everybody,
I need to patch the kernel.
Do I need to only make the changes to the code, remake the kernel, and
reboot? Or is there some installation process I need to go through in
addition?
Thanks for the help.
Jay.
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
Date: 25 May 2000 22:29:13 +0100
"Peter T. Breuer" <[EMAIL PROTECTED]> writes:
> I need ./configure --prefix=/usr --sysconfdir=/etc usually. If I am
This is what a site-config file is for, making `usual' options vanish.
> overwriting a system installed version, that is. Followed by
> a find /usr /etc -type mount -cnewer . > /tmp/file_list, and some
Again I say `this is what GNU stow is for'.
I have --prefix=/usr in my site-config file, but I *install* into
/usr/packages.bin/{package}/{version} and stow from there into /usr.
After all, --prefix at configure time specifies where the compiled code
looks for its data, not necessarily the directory where it is
installed. (Sites that use AFS or Coda are *very* used to this kind of
hack.)
> : The point of automation is to avoid such manual futzing.
>
> If I could be bothered to type the above into a command, I would.
Site-config files. Site-config files. Site-config files. Set the
CONFIG_SITE environment variable and write one, as documented in the
autoconf manual.
e.g. (this is mine)
#
# config.site - this file is run by configure automatically
# to set default values of environment variables
test "$prefix" = NONE && prefix=/usr
# esperi has little memory so optimize for size there.
if [ "$(hostname)" = "esperi" ]; then
SITE_INTERNAL_OPTFLAGS=-Os
else
SITE_INTERNAL_OPTFLAGS=-O2
fi
test -z "$sysconfdir" && sysconfdir='/etc'
test -z "$CC" && CC=gcc
# flip on gcc-2.95 and egcs-specific flags.
if [ `$CC --version` = 2.95.2 -o `$CC --version | cut -d- -f1` = egcs ]; then
test -z "$CFLAGS" && CFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-fno-exceptions -fno-rtti -march=$(uname -m) -mwide-multiply -pipe"
test -z "$FFLAGS" && FFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-fno-exceptions -fno-rtti -march=$(uname -m) -mwide-multiply -pipe"
if [ `$CC --version` = 2.95.2 ]; then
test -z "$CXXFLAGS" && CXXFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-fpermissive -march=$(uname -m) -mwide-multiply -pipe"
else
test -z "$CXXFLAGS" && CXXFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-march=$(uname -m) -mwide-multiply -pipe"
fi
else # antique gcc
test -z "$CFLAGS" && CFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-mwide-multiply -pipe"
test -z "$FFLAGS" && FFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-mwide-multiply -pipe"
test -z "$CXXFLAGS" && CXXFLAGS="$SITE_INTERNAL_OPTFLAGS -fomit-frame-pointer
-mwide-multiply -pipe"
fi
test -z "$OBJCFLAGS" && OBJCFLAGS="$CFLAGS"
unset SITE_INTERNAL_OPTFLAGS
# Give Autoconf 2.x generated configure scripts a shared default
# cache file for feature test results, host-specific. Kill this if
# feature tests go wrong.
if [ "$CC" = "gcc" ]; then
if [ "$cache_file" = ./config.cache -o "$cache_file" = config.cache ]; then
cache_file="${HOME}/.configure/config.cache.$(hostname)"
fi
fi
Bingo. That sets sysconfdir and prefix, points the cache files at a
single per-host persistent cache file, and sets the compilation flags to
a value dependent upon the version of GCC and the hostname, and does it
all in an overrideable fashion.
> But neither my brain nor my memory is sufficiently weak to encourage
> me to do so. If some day I forget, it would take me all of ten minutes
> to work it out again, and maybe I'd learn something new. Really
> I should save the config log. That would be handy.
No need ;)
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To:
comp.os.linux,comp.os.linux.development,comp.os.linux.development.apps,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.advocacy
Subject: Re: Need ideas for university funded project for linux
Date: 25 May 2000 23:15:18 +0100
David Steuber <[EMAIL PROTECTED]> writes:
> What about software that makes system calls in
> Linux? Must that also be GPL?
No, because the license for Linux explicitly states, right at the top
: NOTE! This copyright does *not* cover user programs that use kernel
: services by normal system calls - this is merely considered normal use
: of the kernel, and does *not* fall under the heading of "derived work".
You are just trolling, aren't you? Or did you really not even glance at
the license before coming out with that statement?
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: What !@#$ moron colorised g++?
Date: 25 May 2000 23:42:57 +0100
Chetan Ahuja <[EMAIL PROTECTED]> writes:
> I use emacs for all programming and it screwed up the
> emacs bug reader ( which means that the script was not working right
> according to it's own comments since STDOUT should not be a tty when
> the process is forked from emacs)
: `process-connection-type' is a simple built-in variable.
:
: Value: t
:
: Documentation:
: Control type of device used to communicate with subprocesses.
: Values are nil to use a pipe, or t or `pty' to use a pty.
: The value has no effect if the system has no ptys or if all ptys are busy:
: then a pipe is used in any case.
: The value takes effect when `start-process' is called.
So yes, it is a pty by default. This lets SIGTSTP and things work right
from subprocesses.
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: What !@#$ moron colorised g++?
Date: 25 May 2000 23:45:29 +0100
[EMAIL PROTECTED] (Michal Jaegermann) writes:
> If Jamie Moyers wants to waste his time on this crap that it his private
> business and one can only shrug. One may hope that he will see light
> some day. But whomever decided to foist that as a __default__ on a
> distribution is a truly certified idiot.
Mandrake have made some, er, unwise decisions wrt gcc in the past ---
such as shipping pgcc as the default, and compiling the kernel with it
--- and, for a brief period, even shipping a development snapshot of gcc
as the system compiler (!!!). Every time they get burned and stop; one
hopes they'll stop trying in the end.
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: What !@#$ moron colorised g++?
Date: 25 May 2000 23:39:40 +0100
[EMAIL PROTECTED] (Mike Dowling) writes:
> Methinks the problem lies with your system, not Linux.
gcc is a script in Mandrake, that colourises and other things.
Run gcc itself, not the script, and all is happy.
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: ps does not show all processes
Date: 26 May 2000 00:25:58 +0100
Allin Cottrell <[EMAIL PROTECTED]> writes:
> Thus "ps ax" will do what you want ("ps aux" is probably better).
and some of us sick bastards like `ps -ef'.
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To: comp.os.linux.development,comp.os.linux.development.apps
Subject: Re: Two really easy (I'm sure) questions
Date: 25 May 2000 23:25:36 +0100
Lew Pitcher <[EMAIL PROTECTED]> writes:
> Even platforms for which authors recommend void main() (i.e. MSDOS)
> expect a value from main().
Given that the sum total of those `authors' is Herb Schildt...
(... they don't call his output `bullschildt' for nothing.
Although he has acquired a few nanoclues recently, they are
very nano.)
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: Checking for I'm swapped - number of pages swapped out
Date: 25 May 2000 23:38:06 +0100
[EMAIL PROTECTED] (Kevin Buhr) writes:
> Unfortunately, the kernel (at least 2.2.14) doesn't have a statistic
> dedicated to counting swapped out pages. Worse yet, it doesn't appear
> to provide quite enough other information to calculate it.
Er, I hate to say this, but `ps' can happily say
PID TRS DRS RSS COMMAND
28382 247 2360 752 [tcsh]
Those three values there are probably what you want.
It's /proc/{pid}/stat, 24th field. task -> mm -> rss.
(This doesn't count swapped out but rather resident pages, but that
should be OK, shouldn't it? `swapped out pages' is quite hard, because,
for instance, what about COW-shared pages, shared library pages...)
--
`Q: Why did they deprecate a.out support in linux?
A: Because a nasty coff is bad for your elf.' --- James Simmons
------------------------------
Reply-To: "William A. McKee" <[EMAIL PROTECTED]>
From: "William A. McKee" <[EMAIL PROTECTED]>
Subject: XIM info
Date: Fri, 26 May 2000 21:44:21 GMT
Where should I look for developing XIM (X input methods) code for Linux?
Sincerely,
Will McKee.
------------------------------
From: "Aurelie Fonteny" <[EMAIL PROTECTED]>
Subject: RPC : sendmsg returned error 101
Date: Fri, 26 May 2000 16:40:35 -0500
Hi folks,
I'm right now triing to debug a device driver, and I had several system
hangs. Now, when I'm rebooting my computer it still reboots fine, but takes
much more time because of the error :
Starting NFS quotas : neighbour table overflow
Cannot register service : RPC: Unable to send; errno = Invalid argument
[FAILED]
localhost rquota[340] : Unable to register (RQUOTAPROG, RQUOTAVERS, udp)
localhost rpc.statd[332] : Unable to register (SM_PROG, SM_VERS, udp)
localhost mountd[350]: Unable to register (mountd, 1, udp)
RPC : sendmsg returned error 101 appears.
RPC : sendmsg returned error 101 appears.
RPC : sendmsg returned error 101 appears.
RPC : sendmsg returned error 101 appears.
portmap: server localhost not responding, timed out
RPC : sendmsg returned error 101 appears.
RPC : sendmsg returned error 101 appears.
....
Does anyone know how to fix this problem? Can t be fixed easily?
Thanks for your time
Aurelie
------------------------------
** 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
******************************
