Linux-Development-Sys Digest #803, Volume #6 Wed, 9 Jun 99 15:14:25 EDT
Contents:
Re: Run time measurement with micro (or at least milli)-second (LEBLANC ERIC)
linux wavelan driver (Cynthia Montalvo)
Re: TAO: the ultimate OS (Donal K. Fellows)
Re: the ultimate OS (Donal K. Fellows)
Re: Need help on /proc/[pid]/* ("John Burton")
Re: Any Mail Application for commercial use (Grant Edwards)
Need help on /proc/[pid]/* ([EMAIL PROTECTED])
How to understand /proc/net/dev ([EMAIL PROTECTED])
Re: Need help on /proc/[pid]/* ([EMAIL PROTECTED])
Re: What are the differences between mySQL and mSQL? ("Apollo Zyx")
Re: Large (17 GB) hdd and fdisk (Jim Robertson)
Re: Images (Jonathan Stott)
Re: glibc 2.0 or glibc 2.1 ("Dan")
Re: new kernel: LILO "kernel too big" error ("steve davidson")
Re: Linux & Cybercafe (David T. Blake)
Wierd Messages At Boot (Aaron Faby)
oplock "breaking" for samba yet? (Walt Boring)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (LEBLANC ERIC)
Subject: Re: Run time measurement with micro (or at least milli)-second
Crossposted-To: comp.os.linux.misc
Date: Wed, 09 Jun 1999 01:57:02 GMT
Virasit Imtawil ([EMAIL PROTECTED]) wrote:
:
: Dear All,
:
: I am sorry if this is not where I should write but I would like your
: help. I am a beginner here. I use Redhat linux (kernel 2.0.32). I would
: like to know how to measure CPU executing time within the C source code in
: micro (or at least milli)-second resolution. I tried clock() command but
: it's just second resolution which is extremely coarse. For example, I have
: a C code like
I'm a beginner too...
If it is for optimisation purpose, you might try compiling your program
with profiling enabled. It will give you an usage summary per functions.
It has a resolution of .01 seconds.
$ gcc -pg test.c -o test
Run the program normally
$ ./test 240
Done.
$ gproff > outfile
Part of the outfile will look like this:
[taken from the gprof info page]
% cumulative self self total
time seconds seconds calls ms/call ms/call name
33.34 0.02 0.02 7208 0.00 0.00 open
16.67 0.03 0.01 244 0.04 0.12 offtime
16.67 0.04 0.01 8 1.25 1.25 memccpy
16.67 0.05 0.01 7 1.43 1.43 write
: x = time_start();
: ---;
: ---;
: ---;
: ---;
: y = time_finish();
You might want to put the code in betwen the x = ... and the y =... on a
separate function so that you can measure it alone.
There's more examples and howto in the info documentation of gprof.
Matt Welsh's book, "Running Linux" had some examples of profiling too.
Regards,
E.
------------------------------
From: Cynthia Montalvo <[EMAIL PROTECTED]>
Subject: linux wavelan driver
Date: Wed, 09 Jun 1999 09:08:39 -0400
Hi. I'm using Lucent's WaveLan/PCMCIA card on a laptop with RedHat Linux
5.2 (kernel 36) and want to be on 2.4 gHz. I've already installed the
pcmcia-cs-3.0.7 file. My question is does anyone know where i can get a
linux
driver for the WaveLan card?
Thanks in advance,
cynthia
------------------------------
From: [EMAIL PROTECTED] (Donal K. Fellows)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: 9 Jun 1999 14:03:11 GMT
In article <[EMAIL PROTECTED]>,
Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
> i.e. "who are you??? what papers have you written"?? I am not
> interested in a debate on the exact definition of "object
> oriented".. the essay calls for it to be fleshed out. the debate on
> the subject here proves the lack of consensus within this arena,
> which I am not really interested in engaging in, except to choose
> one model.
Had you said component instead and then stayed well clear of
mentioning C++, you would have attracted much less flak at the cost of
making the document itself a little bit more wishy-washy...
Donal.
--
Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ [EMAIL PROTECTED]
-- The small advantage of not having California being part of my country would
be overweighed by having California as a heavily-armed rabid weasel on our
borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>
------------------------------
From: [EMAIL PROTECTED] (Donal K. Fellows)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: the ultimate OS
Date: 9 Jun 1999 13:58:29 GMT
In article <[EMAIL PROTECTED]>,
Vladimir Z. Nuri <[EMAIL PROTECTED]> wrote:
> HW.. very nice post. thanks for your comments.. I can
> take your criticism because it's all reasonable.
>
> Hayden Walles ([EMAIL PROTECTED]) wrote:
>: So if you are wont to flame the thread with arguments similar to "You
>: don't tell me exactly what this is so I won't talk t you!" instead send
>: queries. Suggest alternatives to the concepts used to solve the brief.
>: Develop the brief to define the terms. Many posters have show that they
>: have the expertise to do this, but they only demand answers of the author,
>: who presumably wants _us_ to define these terms.
>
> the document is partly a manifesto, an assertion that the goals
> are worthwhile and feasible to accomplish. I will work with people
> who want to collaborate to realize the goals. I will not hand down
> proclamations or code like moses on the mountain.
Just read your online document (you really ought to add a <head></head
part to it with a title if nothing else, and maybe use a few more
capital letters in headings and emphasize with <em></em>, but that's
just stuff on the level of spelling quibbles... :^)
Hmm. Your primary wishlist/diagnosis is laudable, but the current
state of software engineering would seem to pretty much preclude you
from going after more than one of those lofty goals at a time. I'll
comment on each of them before moving on to my impressions of the rest
of the document.
* Favouring end-users. This is good but very hard, as truly
favouring them requires understanding their problems and
conceptual models of the system. It also isn't always just giving
them what they ask for, either. Very, very tough indeed.
* Uniformity/coherence. This is an active research topic. Something
may even come of it eventually...
* Open standards/interoperability. This is tough to achieve without
suffering from committee-itis. But this is probably the single
most feasable item on the list - Linux and *BSD are examples of
this.
* Security. This is hard. Especially since you need to protect
against fools, vandals and malicious *authorised* people. I'm far
too sceptical here to think that foolproof security is possible
without rendering the whole system unusable. :^(
* Integrity/reliability. Nobody really knows how to do this for
anything other than extremely static systems (like nuclear power
plant controllers, etc.) Those software-engineering techniques
that exist and that are being developed are inimicable to RAD and
getting a quick solution. Formally verifying a whole OS is
mind-bogglingly tough job.
Debugging is hard for novices because they haven't got the
practise at thinking logically and the experience to know rapidly
where the problem is. People who *do* know that sort of thing are
not novices.
* Complete hardware abstraction. I doubt this is possible in
practise without a performance hit. And in any case, there are
major practical differences between a local and a remote resource,
since the latter may take a very long time to respond due to
factors beyond any computer's control. In the extreme, some
people doing road maintenance might accidentally drill straight
through a crucial piece of fibre...
* Optimisation. The best optimisations require deep knowledge of
the problem domain. Without that, you only get peephole opts
(which are still very useful in a great many cases.) Consider the
case where there is some general code to find out how many
different items there are in a list. One way is to sort, remove
duplicates and then count up the number of items, but if you know
that the list is of sequential numbers stored in reverse order,
you can do much better than that (i.e. one read from a "fixed"
location and maybe a little arithmetic.) No peephole optimiser
will ever spot that...
Moving on to the other big topic in your message; objects as the holy
grail or philosophers stone to deliver all this. Excuse me while I
look utterly
C++ has a different view of objects to Java, which has a different
view to Perl, which has a different view to a whole load of other
different languages. There is certainly no consensus within the OO
community on what an object really is. I suppose a subtype-able
abstract datatype is probably about as close as you'll get. But using
objects doesn't in itself help one whit when it comes to security for
one. A simple calculator is not OO, but it is pretty much secure;
programs written in C++ are (often) OO, but not secure.
<Diversion Class=Minor>
One of the lecurers here is working on the correctness of component-
-based systems[*]; maybe his web pages will have links to some stuff
of interest if you want correct OO program development:
http://www.cs.man.ac.uk/~kung-kiu/
</Diversion>
Please don't take this as a flame (it isn't intended to be) but your
document seems to me to be in some ways rather like the Communist
Manifesto; cute concept(s) but next to no applicability in the real
world. I could be wrong, but I doubt it, and I suspect that many
other people think there is nothing there except fine words and
moonshine. But then there are a lot of cynics on c.o.l.a... :^)
Donal.
[* The "grown-up" term for what you are talking about. ]
--
Donal K. Fellows http://www.cs.man.ac.uk/~fellowsd/ [EMAIL PROTECTED]
-- The small advantage of not having California being part of my country would
be overweighed by having California as a heavily-armed rabid weasel on our
borders. -- David Parsons <o r c @ p e l l . p o r t l a n d . o r . u s>
------------------------------
From: "John Burton" <[EMAIL PROTECTED]>
Subject: Re: Need help on /proc/[pid]/*
Date: Wed, 9 Jun 1999 16:41:02 +0100
man proc?
--
John Burton => ICQ: 40689393 ICQ Pager via : [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote in message
<[EMAIL PROTECTED]>...
>Hello,
>
>can someone tell me where I can get description of the structure of:
>/proc/[PID]/stat
>/proc/[PID]/statm
>/proc/[PID]/status <-- especially the last few lines
>
>I were very happy if someone could help me!
>
>Sincerely
> Erkan
>
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.networking,comp.os.linux.misc
Subject: Re: Any Mail Application for commercial use
Date: Wed, 09 Jun 1999 16:00:49 GMT
In article <7jlh7j$f9q$[EMAIL PROTECTED]>, Stefaan A Eeckels wrote:
>> Besides Zmail and Sendmail, is there any mail application suitable for
>> commercial use ? As Sendmail seems too complicated for commercial and
>> the user interface is not so user-friendly.
>
>I'm very pleased with qmail. Easy to set up, fast, and reliable.
By "commercial use" I assume you mean use as a high-volume
high-reliability mail-server.
I've only set up qmail on single-user workstations (as opposed to
"commercial use"), but I was impressed with it. The installation and
configuration is pretty straight-forward thanks to both a clean design
and excellent step-by-step instructions.
It even built/installed hassle-free on an old Iris Indigo running Irix
5.3 (not always the most compatible machine in the world).
OTOH, the few times I've tried to make minor adjustments to sendmail
configurations the results varied but generally included dizzyness,
facial tics, bleeding from the ears, and a broken configuration.
--
Grant Edwards grante Yow! PEGGY FLEMMING is
at stealing BASKET BALLS to
visi.com feed the babies in VERMONT.
------------------------------
From: [EMAIL PROTECTED]
Subject: Need help on /proc/[pid]/*
Date: Wed, 09 Jun 1999 15:20:25 GMT
Hello,
can someone tell me where I can get description of the structure of:
/proc/[PID]/stat
/proc/[PID]/statm
/proc/[PID]/status <-- especially the last few lines
I were very happy if someone could help me!
Sincerely
Erkan
------------------------------
From: [EMAIL PROTECTED]
Subject: How to understand /proc/net/dev
Date: Wed, 09 Jun 1999 15:24:33 GMT
Hello,
can someone tell me where to obtain detailed information (maybe
development guide) to understand /proc/net/dev ?
Tank you in advance!
Erkan
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Need help on /proc/[pid]/*
Date: Wed, 09 Jun 1999 16:04:37 GMT
On Wed, 9 Jun 1999 16:41:02 +0100, "John Burton" <[EMAIL PROTECTED]>
wrote:
>man proc?
yeah... this is it!
Thanx again!
------------------------------
From: "Apollo Zyx" <[EMAIL PROTECTED]>
Subject: Re: What are the differences between mySQL and mSQL?
Date: Mon, 7 Jun 1999 20:32:34 +0700
if i've some collection of data in mSQL
how can i move them to mySQL,
and do i need to edit my Perl script?
------------------------------
From: Jim Robertson <[EMAIL PROTECTED]>
Subject: Re: Large (17 GB) hdd and fdisk
Date: Wed, 09 Jun 1999 03:30:05 GMT
I've had to monkey around with my disk partitions, since M$ has its own
peculiar slant on the meaning of partition tables. Ended up setting the
type to 'LARGE' in bios, which works for this 6.4G drive. Here's how I
have it partitioned, and it works with both Win95 SR2 and Linux:
| Command (m for help):
| Disk /dev/hda: 240 heads, 63 sectors, 839 cylinders
| Units = cylinders of 15120 * 512 bytes
|
| Device Boot Start End Blocks Id System
| /dev/hda1 * 1 66 498928+ 6 DOS 16-bit >=32M
| /dev/hda2 133 839 5344920 5 Extended
| /dev/hda3 67 132 498960 83 Linux native
| /dev/hda5 133 409 2094088+ b Win95 FAT32
| /dev/hda6 410 686 2094088+ b Win95 FAT32
| /dev/hda7 687 839 1156648+ 83 Linux native
If I use Win95 to partition, it invariably creates one primary and one
extended partition. The extended partition contains all the others. In
this case I got away with creating the primary hda3 partition for Linux,
letting M$ have the first two logical partitions in the extended area,
then putting Linux in the rest of the extended partition. hda3 is
mounted as / and hda7 is mounted as /usr.
Back when I had the drive set to 'Normal' in bios, M$ and Linux could
not agree on the geometry of the partitions, so one or the other fdisk
would always report errors. I'm unsure whether the 1023 cylinder 'thing'
was causing the conflict.
Good luck with your 17G drive!
Rob van Nieuwpoort wrote:
>
> Hi,
>
> My brother bought a new harddisk last week, and I am trying to install
> Red Hat 6 on it. However, my brother first installed win98 on the first
> primary partition, and reserved two other partitions for win98. The
> remaining 4 GB of the disk are not used.
> When I try to create the new linux partition(s), fdisk warns me that
> the paritions are overlapping. When I look at the starting and ending
> cylinders of the partitions, they are NOT overlapping.
>
> I also tried partitioning using DiskDruid, but it would not allow me to
> create new partitions at all.
>
> Is it safe to ignore fdisks warnings, or am I doing something wrong here,
> any suggestions?
>
> thanks,
>
> Rob ([EMAIL PROTECTED])
-- JRob
Wanna go back to the good ole days? Turn off the air conditioning.
------------------------------
From: [EMAIL PROTECTED] (Jonathan Stott)
Subject: Re: Images
Date: 9 Jun 1999 16:48:13 GMT
Reply-To: [EMAIL PROTECTED]
In article <7jjhfs$i0t$[EMAIL PROTECTED]>,
Lenny grosso <[EMAIL PROTECTED]> wrote:
>Does anybody know of good resources and information for creating images for
>floppys. This would let me make and install for my new linux dist.
Do you mean kernel images or complete bootable micro-systems?
Bootable kernel images can be created fairly simply
% dd if=/vmlinuz of=/dev/fd0
% rdev /dev/fd0 /dev/[whatever]
(assuming the kernel you want to boot is /vmlinuz and /dev/[whatever]
is your normal root filesystem). This is how you can do things like
making a parallel-port zip disk your emergency boot disk [*very*
useful, BTW, much more so than the usual pair of floppies] even though
the parallel port isn't normally a bootable device.
The other way to do things is to put a small system into a ramdisk.
The kernel goes on one floppy (same as above, root disk is the ramdisk)
and the contents of the ramdisk go either after the kernel or on a
second floppy. This should be covered in the Bootdisk-HOWTO.
-JS
--
Jonathan Stott http://poly.phys.cwru.edu/
icbm://41.30.14N/81.36.36W/ [EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: "Dan" <[EMAIL PROTECTED]>
Subject: Re: glibc 2.0 or glibc 2.1
Date: Wed, 9 Jun 1999 20:02:46 +0200
Andreas, & everyone else,
Thanks a lot for your help !
All three ways Andreas described were working, I've got 2.0.7
Thanks again
Dan
Andreas Jaeger wrote in message ...
>
>There's one easy solution: Just call /lib/libc.so.6 direct (yes,
>execute it!). glibc 2.1.1 will report some infos, glibc 2.0.x will
>segfault since that's one of its errors.
>
>Another place to check: /usr/include/features.h defines
>__GLIBC_MINOR__ which is 1 for glibc 2.1.x and 0 for glibc 2.0.x.
>
>A final check:
>strings /lib/libc-2.1.1.so |grep "C Library"
>
>Andreas
>--
> Andreas Jaeger [EMAIL PROTECTED] [EMAIL PROTECTED]
> for pgp-key finger [EMAIL PROTECTED]
------------------------------
From: "steve davidson" <[EMAIL PROTECTED]>
Subject: Re: new kernel: LILO "kernel too big" error
Date: Wed, 9 Jun 1999 11:13:06 -0700
Usseglio Gaudi Francesco <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> steve davidson wrote:
> >
> > I'm confused.
> >
> > re-built my RH 6 2.2.5 kernel last night, stripped out all of the junk
that
> > I don't need, added a couple of things in ( my selections are ok, I've
been
> > researching this for a while). Compiled OK, no error messages.
Followed
> > this procedure:
> >
> > make xconfig (configure...)
> > make dep
> > make clean
> > make zImage
> >
> > make succeeds, I end up with a 426KB kernel.
> >
> > Ran Linuxconf, selected the 'install kernel I have compiled' option
under
> > LILO section, upon 'save config' selection I receive a 'kernel too big'
> > error.
> >
> > OK, so I think that maybe linuxconf is screwy, so I manually edit the
> > /etc/linux.conf file, adding the section
> >
> > image=/boot/newkernelz
> > label=new
> >
> > between the existing image.. section and the other.. section ( I
previously
> > copied the new zImage to /boot/newkernelz ).
> >
> > Saved lilo.conf, ran lilo -v: Still get the error "kernel
/boot/newkernelz
> > is too big".
> >
> > I don't get it. The kernel which ships with RH 6 (vmlinuz-2.2.5-15) is
> > 617,288 bytes, while my new kernel is 475,696 bytes. What gives?
> >
> > Any suggestions here?
> >
> > Thanks!
> >
> > Steve Davidson
> I got the very same problem... I resolved it with make bzImage instead of
make
> zImage
> Bye.
> ==+==
> ...era un mondo adulto...
> ...si sbagliava da professionisti.
> Paolo Conte
------------------------------
From: [EMAIL PROTECTED] (David T. Blake)
Crossposted-To: alt.uu.comp.os.linux.questions,comp.os.linux.networking
Subject: Re: Linux & Cybercafe
Date: 09 Jun 1999 11:11:49 -0700
David Knight <[EMAIL PROTECTED]> writes:
>> You can configure the windowmanager without any xterm. So the users are
>> not able to start any other program. On the desktop you offer only the
>> browser. Login should be via xdm !
>> That's it i think !
>
>That it itself probably wont work, just put in netscape
>
>telnet://localhost
>
>and it will open an xterm running the telnet session, from which they
>would then be able to start other programs.
And why is inetd serving telnet on this machine ??
--
Dave Blake
[EMAIL PROTECTED]
------------------------------
From: Aaron Faby <[EMAIL PROTECTED]>
Crossposted-To: linux.dev.kernel
Subject: Wierd Messages At Boot
Date: Wed, 09 Jun 1999 14:48:40 -0400
Hello mates,
Im getting some strange messages in /var/log/messages
after booting. "Cannot find map", "Error seeking in /dev/kmem",
and "Error loading kernel module table". Has anyone else
ever had these errors before? What can I do to fix them?
Thanks in advance!
--
Aaron Faby
[EMAIL PROTECTED]
System Administrator/Technical Support
Yourlink, Inc.
------------------------------
From: Walt Boring <[EMAIL PROTECTED]>
Subject: oplock "breaking" for samba yet?
Date: Wed, 09 Jun 1999 10:18:37 -0700
Hi Folks,
I was wondering if there was anyone working on the notifying samba's
oplock code of file modifications for linux? I would really like to
have oplocks working in linux for a samba/nfs shared exported share.
I think this would be a great addition to linux. I haven't a clue how
to do it, if its even being worked on, or I would do it myself. Anyone?
Thanks
Walt
[EMAIL PROTECTED]
------------------------------
** 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
******************************
