Linux-Development-Sys Digest #204, Volume #6 Sat, 2 Jan 99 23:14:32 EST
Contents:
Re: Registry for Linux - Bad idea (Nix)
Re: Does brk hack still work (Nix)
Re: Counting page faults (Richard Jones)
Re: Why I'm dumping Linux, going back to Windblows (Tristan Wibberley)
Re: silly question (Tristan Wibberley)
Re: Linux Journaling Filesystem (Frank Sweetser)
bttv & video motion detection (Tom Aschenbrenner)
Re: things I'd pay to have developed for Linux... (Ilya)
framegrabber device drivers ("Bjorn Wesen")
Re: silly question (David M. Cook)
Re: parport/ppa problem with 2.2.0pre1 (Clifford Kite)
Re: silly question (Stefaan A Eeckels)
Soundcard with Digital in? (A James Lewis)
Re: Compile errors on mixed libc5 / libc6 system - help! (Daniel R. Grayson)
Re: Registry for Linux - Bad idea (Leslie Mikesell)
Re: Kernel v2.2 (Johan Kullstam)
----------------------------------------------------------------------------
From: Nix <$}xin{$@esperi.demon.co.uk>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: 30 Dec 1998 22:39:34 +0000
Michal Mosiewicz <[EMAIL PROTECTED]> writes:
> By registry, I mean:
> a) an easily accessible database with optimised indexed storage
> b) kernel space managment to accomplish ownership of database records
> and access rights (not only per user, but also per application)
> c) events/notifiers/signals to allow for asynchronous notification about
> the changes in the state of variables
This exists. It is called `the filesystem'. The indexes are unusual,
but this is a hierarchical database after all, not a relational one.
Yes, /etc *is* a registry; fairly free-format, but still a registry of
sorts.
--
`Anyone who says you can have a lot of widely dispersed people hack
away on a complicated piece of code and avoid total anarchy has never
managed a software project.' - Andy Tanenbaum in 1992 on comp.os.minix
------------------------------
From: Nix <$}xin{$@esperi.demon.co.uk>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Does brk hack still work
Date: 31 Dec 1998 00:28:41 +0000
George MacDonald <[EMAIL PROTECTED]> writes:
> at the start of the program, the program initialization time was cut in
> half. This was caused by the fact that free() was not reducing the
> amount of memory used, thus each new malloc did not require the system
> call's to brk/sbrk.
>
> Anyone know if this is the same on Linux?
A number of malloc() libraries, notably the dlmalloc used in glibc2,
do not always allocate memory by way of calls to (s)brk() in any
case. Specifically, small chunks are suballocated by way of mmap()
rather than (s)brk().
Also, a disappointing number of malloc()s never shrink the amount of
memory brk()ed from the system, which makes running X servers and
emacs rather more memory-hungry than they are already (not a good
thing) :(
--
`Anyone who says you can have a lot of widely dispersed people hack
away on a complicated piece of code and avoid total anarchy has never
managed a software project.' - Andy Tanenbaum in 1992 on comp.os.minix
------------------------------
From: Richard Jones <[EMAIL PROTECTED]>
Subject: Re: Counting page faults
Date: Sat, 2 Jan 1999 19:05:34 +0000
Matej Artaè <[EMAIL PROTECTED]> wrote:
: Hello again!
: The information about the number of page faults is indeed in each process'
: stat file.
On a machine with 64 MB RAM + 240 MB swap, kernel 2.2.0-pre2:
-- with overcommit = 0
$ ./test
CPU time in seconds: current = unlimited, max = unlimited
Maximum filesize: current = unlimited, max = unlimited
max data size: current = unlimited, max = unlimited
max stack size: current = 8388608, max = unlimited
max core file size: current = 1024000000, max = unlimited
max resident set size: current = unlimited, max = unlimited
max number of processes: current = 256, max = 256
max number of open files: current = 1024, max = 1024
max locked-in-memory address space: current = unlimited, max = unlimited
Last successful allocation: 241577984 (230MB)
-- with overcommit = 1
$ ./test
CPU time in seconds: current = unlimited, max = unlimited
Maximum filesize: current = unlimited, max = unlimited
max data size: current = unlimited, max = unlimited
max stack size: current = 8388608, max = unlimited
max core file size: current = 1024000000, max = unlimited
max resident set size: current = unlimited, max = unlimited
max number of processes: current = 256, max = 256
max number of open files: current = 1024, max = 1024
max locked-in-memory address space: current = unlimited, max = unlimited
Last successful allocation: 1073352704 (1023MB) [1]
.. which is what you would expect, given my previous
post on this subject.
Rich.
[1] I had to cheat slightly to get the last result, by
starting s at 1000*1024*1024. It was taking too long
counting up from 0 ...
--
- Richard Jones. Linux contractor London and SE areas. -
- Very boring homepage at: http://www.annexia.demon.co.uk/ -
- You are currently the 1,991,243,100th visitor to this signature. -
- Original message content Copyright (C) 1998 Richard Jones. -
------------------------------
From: Tristan Wibberley <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Why I'm dumping Linux, going back to Windblows
Date: Sat, 02 Jan 1999 19:06:15 +0000
Reply-To: [EMAIL PROTECTED]
Chris Hanson wrote:
>
> The central idea that Apple and Xerox have contributed to
> human-computer interaction is *consistency*. Editing text files to
> change your configuration is not in itself consistent. Using the same
> format in all of those text files is.
Using the same format for every form of data is a stupid thing to do. It
may be consistent, but if a user is one that needs such consistency, a
GUI tool is better. To make a point, the data in the windows registry is
of consistent format, but the Unix text files are easier to admin - This
is because the files match the data well because they're task specific.
> Unfortunately, no Unix I know
> of does that; every file has its own slightly different format.
As it should be wherever it needs to be for ease of administration
purposes.
> Knowing how /etc/hosts is laid out doesn't make learning X resources
> any easier, and so on.
If X resources were laid out like /etc/hosts, it would be totally
useless and virtually impossible to admin (and Vice Versa).
> I think this is at the core of many peoples' gripes with Linux, and to
> deny its legitimacy is a big mistake.
This is not the core of many peoples gripes with Linux, the core of
their gripes is that there are not any decent GUI config tools. I would
agree with that. The users that can't config text files can't config a
registry either - another solution is needed.
> (Of course, I don't expect people to listen to me. I'll be denounced
> as a heretic, torched, and then the church will continue as it has...)
This is actually where people stop paying any attention to you thanks to
that comment. It is truly childish.
Note how the followups are not to inappropriate newsgroups.
--
Tristan Wibberley Linux is a registered trademark
of Linus Torvalds.
------------------------------
From: Tristan Wibberley <[EMAIL PROTECTED]>
Subject: Re: silly question
Date: Sat, 02 Jan 1999 19:28:12 +0000
Reply-To: [EMAIL PROTECTED]
ebatchelor wrote:
>
> OOps, my mistake...sorry.
>
> I thought I was posting to the "comp.os.linux.development" group,
I thought you were about to apologise for posting to an inappropriate
group...
> not the "comp.os.unix.worship" group.
Nope, I was wrong you're just trolling ;)
--
Tristan Wibberley Linux is a registered trademark
of Linus Torvalds.
------------------------------
From: Frank Sweetser <[EMAIL PROTECTED]>
Subject: Re: Linux Journaling Filesystem
Date: 02 Jan 1999 11:29:20 -0500
Michael Pronath <[EMAIL PROTECTED]> writes:
> Hi,
>
> Windows NT has a journaling filesystem, as well as the new OS/2 Warp
> Server. Is there anything like that for Linux yet or a project to
> create some?
yup, slated for 2.3/2.4 kernel.
> Is there a webpage with an introduction into the concepts of journaling
> filesystems?
i'm pretty sure there is, a quick web search should turn stuff up.
--
Frank Sweetser rasmusin at wpi.edu fsweetser at blee.net | PGP key available
paramount.ind.wpi.edu RedHat 5.2 kernel 2.1.131ac6 i586 | at public servers
checking for mass_quantities_of_bass_ale in -lFridge... no
checking for mass_quantities_of_any_ale in -lFridge... no
Warning: No ales were found in your refridgerator.
We highly suggest that you rectify this situation immediately.
-- from configure script of enlightenment 0.14
------------------------------
From: Tom Aschenbrenner <[EMAIL PROTECTED]>
Subject: bttv & video motion detection
Date: Sat, 02 Jan 1999 11:12:42 -0600
Hello.....does anyone have any leads on source code for doing motion
detection using
the raw capture buffer output of the bttv driver? So
far a DejaNews search
hasn't turned up a whole lot.
TIA Tom A, - Dallas, Tx.
------------------------------
From: Ilya <[EMAIL PROTECTED]>
Subject: Re: things I'd pay to have developed for Linux...
Crossposted-To:
comp.os.linux.misc,comp.os.linux.development.apps,comp.os.linux.hardware
Date: 2 Jan 1999 12:50:47 +0800
JFS.
LVM.
Mirroring and stuff
Please.
------------------------------
From: "Bjorn Wesen" <[EMAIL PROTECTED]>
Subject: framegrabber device drivers
Date: Sat, 2 Jan 1999 22:28:31 +0100
Hi, I'm writing a device driver for a quite fast framegrabber and I was
wondering how to make it scalable regarding device access frequency. I want
it to handle the load if requests come closer together than the maximum grab
speed - then the requests should get the same picture. Would it be
worthwhile to make it a block device and let the page-cache cache blocks,
and deal with invalidation problems, or should I just let the driver cache
the latest picture internally and make it a char device (pictures can be
pretty large!)?
I assume the block device approach would be slower in some situations,
because there would be one extra copy (to the cache/from the cache to user)
instead of from the driver to the user directly, if I'm not mistaken.
regards,
/Bjorn
-
Please mail me at bjorn at sparta.lu.se and dont use the reply-to adress
since I've scrambled it to avoid junk mails.
------------------------------
From: [EMAIL PROTECTED] (David M. Cook)
Subject: Re: silly question
Date: Sat, 02 Jan 1999 21:09:27 GMT
On Fri, 01 Jan 1999 10:17:27 -0600, ebatchelor <[EMAIL PROTECTED]>
wrote:
>I am a new user to Linux. I am an experienced MSDOS user, have written
>many batch files to accomplish what I want to do, and can recall the
>names of most DOS utilities I need to use. Of course, DOS sucks, but
See the DOS to Linux HOWTO at
http://metalab.unc.edu/linux/HOWTO/HOWTO-INDEX-3.html
>Why the convoluted, hard to recall, someone thought it was funny in 1975
I don't find most commands hard to recall. Are cp, mv, rm, ls, cd, ln, tar,
ps, etc. that hard to remember? Now it's true that there are a lot of less
often used commands that I often forget, but DOS doesn't even have most of
those commands!
If you still want a dumbed down shell, see
http://www.cs.uct.ac.za/~mwelz/lsh.html
But I suggest you get a copy of Linux in a Nutshell instead (you might want
to wait for the 2nd edition which is due in February.)
Dave Cook
------------------------------
From: [EMAIL PROTECTED] (Clifford Kite)
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: parport/ppa problem with 2.2.0pre1
Date: 2 Jan 1999 13:37:52 -0600
Griffin Caprio ([EMAIL PROTECTED]) wrote:
: > > I can't even get my printer to work under 2.2.0pre1. It works fine in
: > > 2.0.36, but not in the newer kernels. I tried to switch from lp1 to lp0
: > > and it still doesn't work. Any tips?
: The weird thing is I can cat to it, just not with lpr. maybe I need a newer
: distro of lpr?!?
Under 2.1.131 I had no problem printing after adjusting the /etc/printcap
file but found that I couldn't "lprm - " a print queue. It was apparently
a permissions thing since "chmod 2511 /usr/bin/lprm" cured it - my lprm
and print queue directory both are group lp and the directory permissions
didn't permit an ordinary user to write to it. My /usr/bin/lpr already
had those permissions.
--
Clifford Kite Not a guru. (tm)
[EMAIL PROTECTED] Better is the enemy of good enough.
------------------------------
From: [EMAIL PROTECTED] (Stefaan A Eeckels)
Subject: Re: silly question
Date: 2 Jan 1999 21:46:01 GMT
In article <[EMAIL PROTECTED]>,
Taso Hatzi <[EMAIL PROTECTED]> writes:
> ebatchelor wrote:
>> Why the convoluted, hard to recall, someone thought it was funny in 1975
>> utility names? It seems to me that BASH could be easily recoded to
>> include easy to use and remember identifiers without giving up ANY
> Recoding bash or any existing shell is not on, but if there are enough
> DOS-ophiles out there, why not produce a DOS COMMAND shell for Linux?
Now that would be truly perverse. Spend time to make a usable system
unusable. Perverse.
--
Stefaan
--
PGP key available from PGP key servers (http://www.pgp.net/pgpnet/)
___________________________________________________________________
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away. -- Saint-Exupéry
------------------------------
From: A James Lewis <[EMAIL PROTECTED]>
Subject: Soundcard with Digital in?
Date: Sun, 3 Jan 1999 01:39:55 +0000
Are any soundcards with an audio digital in supported under Linux (Indeed,
do any such beasts exist?)....
Basically what I want to do is make a CD from a minidisk master... but I
need somthing that will be able to transfer the digital datastream into a
sequence of WAV files... that means it would need to read Q codes in the
stream too in order to separate at a track boundry......
Anyone have any experience?
James ([EMAIL PROTECTED])
My Windows unders~1 long filena~1, and yours?
------------------------------
From: [EMAIL PROTECTED] (Daniel R. Grayson)
Subject: Re: Compile errors on mixed libc5 / libc6 system - help!
Date: 02 Jan 1999 23:09:28 +0000
[EMAIL PROTECTED] (Mark Tranchant) writes:
> My system is fundamentally a Slackware 3.4 installation, with libc
> 5.4.44 as the main C library. I also have glibc-2.0.7pre6 (I think:
> whichever one came with StarOffice 5.0) runtime support installed.
>
> When compiling some applications, I occasionally get undefined symbol
> messages at linking. Typical symbols are modf, re_search_2, frexp and
> __getfpucw, although there are others. Editing the Makefile to include
> -umodf -ure_search_2 etc usually results in a successful compile, but
> I would like to know what needs changing to fix this permanently.
>
> The Linux kernel (currently running 2.1.131 with RvR's swap readahead
> patch) compiles and runs fine, Wine (haven't tried latest version)
> compiles without error but segfaults on running. My C compiler is
> gcc-2.7.2.
>
> Any suggestions, anyone? Please email as well as post. I do not want
> suggestions to reinstall the entire system - I want to find out what
> needs changing.
Coincidentally I have succeeded this week in making a mixed libc5/6 system,
starting from Slackware, as you have done. I remember seeing the undefined
symbol __getfpucw earlier this week, but I forget exactly where it was!
The symbol getfpucw is in libc.so.5, but not in libc.so.6, so if it's
undefined, that means you are compiling for libc6. You are probably linking
with a libstdc++ or libm that was compiled under libc5. It should be
possible to tell by getting the linker to produce a mapfile you can look at.
Also, take a look at this output:
geometry% ldconfig -p |egrep 'libstdc|libm'
libstdc++.so.27 (libc5) => /usr/lib/libstdc++.so.27
libstdc++.so.2.8 (libc6) => /usr2/lib/libstdc++.so.2.8
libstdc++.so.2.8 (ELF) => /usr/lib/libstdc++.so.2.8
libstdc++.so (libc6) => /usr2/lib/libstdc++.so
libstdc++.so (ELF) => /usr/lib/libstdc++.so
libmenu.so.3.0 (ELF) => /usr/lib/libmenu.so.3.0
libmenu.so.1.9.9e (ELF) => /usr/lib/libmenu.so.1.9.9e
libmenu.so (ELF) => /usr/lib/libmenu.so
libm.so.6 (libc6) => /lib/libm.so.6
libm.so.5 (libc5) => /lib/libm.so.5
libm.so.4 (libc4) => /lib/libm.so.4
libm.so (libc6) => /usr2/lib/libm.so
libm.so (libc6) => /usr/lib/libm.so
I've been careful NOT to define LD_LIBRARY_PATH, and then the dynamic linker
does a good job deciding which libstdc++.so.2.8 to use.
Another way to tell whether you are handling libstdc++ correctly is by
trying to compile programs that use 'throw'. The internal naming convention
for virtual exception tables has changed, presumably reflecting a change in
the way they are organized.
Let me know if you have further problems.
Daniel R. Grayson, Professor
Mathematics Dept, University of Illinois at Urbana-Champaign
email: [EMAIL PROTECTED]
www: http://www.math.uiuc.edu/~dan/
us mail: 1409 W. Green St., Urbana, IL 61801, USA
campus mail: Altgeld Hall, MC-382
phone: 217-367-6384 home (88.2025W, 40.0854N)
217-333-6209 office (88.2280W, 40.1092N)
217-333-3350 department office
217-333-9576 fax
------------------------------
From: [EMAIL PROTECTED] (Leslie Mikesell)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: 2 Jan 1999 22:02:13 -0600
In article <76mncp$rb5$[EMAIL PROTECTED]>, <[EMAIL PROTECTED]> wrote:
>
>Stifles innovation and creativity? You'll have to explain that one to me. I
>wont even get into what's wrong with sendmail, I'll just say that there are
>certainly ways to modify sendmail to use a standard config database, it's
>just a matter of making the database powerful enough.
Sure, but that means it will have to be just as complex as the existing
system and even fewer people will understand it.
>> While I suppose it would be possible to use your scheme and have
>> different files, it naturally suggests a single file, database, or
>> (cough) registry. This is one of the major complaints of the registry -
>> having all configuration in the same place. This presents a single
>> point of failure.
>
>I am so sick of this one. Do me a favor. Remove some random files from /etc
>or /dev or /bin. Tell me how those are not single points of failure. I'd love
>to hear how a system can get very much further than booting the kernel without
>/etc/passwd or /etc/inittab or /bin/sh
You can bring the system up in single user mode to fix things in some
circumstances. In slightly worse cases you can boot from a floppy
with a small toolkit and rebuild most things with just a text editor.
How do you propose to fix a registry based system when its database
is damaged to a point that it won't run? Is this something you expect
to fit on an already crowded floppy?
Les Mikesell
[EMAIL PROTECTED]
------------------------------
Subject: Re: Kernel v2.2
From: Johan Kullstam <[EMAIL PROTECTED]>
Date: 28 Dec 1998 06:47:16 -0500
Martin Bahlinger <[EMAIL PROTECTED]> writes:
> Does anybody know, how long we've to wait for the new Linux kernel
> ??
if you like, you can download, compile and use a development kernel
*right now*! i've been using 2.1.124 .125 .129 and .131 for a couple
of months now with great success (devel kernels get updated frequently
and i can't keep up - the inbetween ones i just didn't get a chance to
download and install them.)
> BTW: Is it called v2.2 or v3.0 ???
i think will be 2.2.0 2.2.1 &c.
--
Johan Kullstam [[EMAIL PROTECTED]] Don't Fear the Penguin!
------------------------------
** 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
******************************
