Linux-Development-Sys Digest #672, Volume #6 Mon, 3 May 99 07:14:41 EDT
Contents:
Re: Linux Real-Time I/O ("Amish S. Dave")
Re: stdio SMBd - name your price (Andi Kleen)
Re: /dev/hda1 has reached maximal mount count, check forced (Alexander Tsang)
Re: Linux + Laptops + APM + X = garbled screen (Edmond Song)
register_filesystem ("new.ccu.edu.tw")
msdos as module in 2.2.6 (Timothy Murphy)
Missing libraries (I think) (Luc Wastiaux)
Re: Problems compiling kernel 2.2.5 (Paul Kimoto)
Mac-emulation on Linux? ("FM")
80x25 Video Memory & Color (Serge B Bromow)
Re: msdos as module in 2.2.6 (Andrew McDonald)
Re: Linux system ID, is there such a thing? (Jonathan A. Buzzard)
Re: Linux Real-Time I/O (Gerard van der Sel)
Re: KDE without gif support (Roope Anttinen)
Re: Mac-emulation on Linux? (Chris Johnson)
Re: CVS (Re: Bill Gates, self made man, NOT!) (H. Peter Anvin)
Re: Linux disk defragmenter (Peter Samuelson)
Re: /dev/hda1 has reached maximal mount count, check forced (Peter Samuelson)
Re: [ANN] CodeWarrior for Red Hat Linux, GNU ed. Shipping (Peter Samuelson)
Re: Mac-emulation on Linux? (Daniel Robert Franklin)
----------------------------------------------------------------------------
From: "Amish S. Dave" <[EMAIL PROTECTED]>
Subject: Re: Linux Real-Time I/O
Date: Mon, 03 May 1999 04:02:32 +0000
Reply-To: [EMAIL PROTECTED]
Jim & Lisa Meils wrote:
> ... Can Linux control in real-time
> i.e. run a routine EXACTLY 120 times/sec ?
>
> ...The thing I need help on is how
> to address the port and send data to it quickly.
Yes - Linux can do this.
To get real-time performance in user space, you have to first tell the
kernel
that you want to be scheduled as a real-time application, and second, lock
the program into memory so that it doesn't get swapped out.
These can be accomplished via
rc = sched_setscheduler(0, SCHED_FIFO, &schparm);
and
rc = mlockall(MCL_CURRENT | MCL_FUTURE);
See the man pages for these to figure out how to initialize schparm, etc.
The program will have to be suid as root, or run by root. Following this,
nanosleep() should be able to wait for 120th of a second quite reliably.
In case you're using an IDE hard drive, you might need to look into the
-u option of hdparm to improve real-time performance of your system.
Finally, if your real-time needs ever become significantly more stringent,
you might look into using the RTLinux kernel interface - (same Linux, just
a
set of libraries and kernel modules that apparently lets you right code
that
runs in the context of the kernel, with better realtime performance).
I've never had to do anything with the parallel ports, but you might look
into the ioperm() or iopl() calls and /dev/port. These should allow you to
write things to the port directly. Also, the kernel parallel port driver
can be
accessed via /dev/par0, etc., which should make things as simple as writing
to that as if it were a file.
--
Amish S. Dave
biology grad. student.
[EMAIL PROTECTED]
------------------------------
Crossposted-To: comp.protocols.smb,comp.unix.solaris
Subject: Re: stdio SMBd - name your price
From: Andi Kleen <[EMAIL PROTECTED]>
Date: 02 May 1999 10:14:24 +0200
[EMAIL PROTECTED] (Jeremy Allison) writes:
> [EMAIL PROTECTED] (Kyler Laird) writes:
>
> >[EMAIL PROTECTED] (Jeremy Allison) writes:
>
> >>>I didn't realize a diff would necessarily
> >code:
> > s/it's name/its name/g;
> > s/read it's old/read its old/g;
> > s/it's segment/its segment/g;
> > s/it's local name/its local name/g;
> > s/it's own/its own/g;
> > s/it's size/its size/g;
> > .
> > .
> > .
>
> Err, no. It's not silly. It's the same reason you
> cannot take Linux, change parts of it and call it
> Kylix (and refuse to release source code).
>
> I take a very dim view of people attempting
> to fence in intellectual property held in the common
> trust (like GPL'ed code is). But that's maybe because I've
> seen people attempt this before with GPL code :-).
>
> If you started with Samba code, the resulting program
> must be GPL'ed. It's as simple as that. If you want
> any more detail than that, go see a lawyer.
What the GPL doesn't require though is to distribute
the code. So e.g. if the original poster pays someone to do
that specific patch for him, and then he choses not to distribute
it, it would be "his private patch". I interpreted his original
statements that he doesn't plan to keep it as private, but feed
it back to the samba code base.
Of course this is in his best interests too,
because otherwise the patch would quickly bit-rot when samba
continues to change, and become useless.
As to get back to the original topic - as far as I remember
smbd is able to run from inetd. Making samba tolerate a failing
getpeername() and perhaps other sockets calls shouldn't be that
difficult I think.
-Andi
------------------------------
From: [EMAIL PROTECTED] (Alexander Tsang)
Crossposted-To: comp.os.linux.hardware
Subject: Re: /dev/hda1 has reached maximal mount count, check forced
Reply-To: Alexander Tsang <NOSPAMtsangalATinameDOTcomNOSPAM>
Date: Mon, 03 May 1999 04:10:55 GMT
Before you reboot your system, try creating a file in the root
directory called 'fastboot'. This will skip the fsck's on the
next reboot only.
Alex
In article <7gdg1b$[EMAIL PROTECTED]>, David Peavey wrote:
>/dev/hda1 has reached maximal mount count, check forced
>
>I get this message sometimes when I boot up. Can someone tell me what this
>means (besides the obvious)? How do fix it? We are doing some system
>integration work here and this message occurs occasionally.
>Thanks
>Dave
>
>
------------------------------
Crossposted-To: comp.os.linux.x
From: Edmond Song <[EMAIL PROTECTED]>
Subject: Re: Linux + Laptops + APM + X = garbled screen
Date: Mon, 3 May 1999 03:12:22 GMT
Arun Sharma wrote:
> The subject says it all. When my laptop goes into hibernation and comes
> back up, my screen is garbled. If I kill X and restart, everything is
> fine.
>
> Is anyone working on making X aware of power management ?
>
> -Arun
>
How do you let the APM work? THere might be some problem for the current
version of
X server 3.3.3.1 , if you press fn + rest to let it work, then probably
bios is not compatible
with X. Edmond
------------------------------
From: "new.ccu.edu.tw" <[EMAIL PROTECTED]>
Subject: register_filesystem
Date: Mon, 3 May 1999 13:10:55 +0800
Hi, All,
When I read code about register_filesystem(), I found that there's one
field in struct file_system, i.e. fs_flags. After looking up, I find
there
are four constant about it.
#define FS_REQUIRES_DEV 1
#define FS_NO_DCACHE 2
#define FS_NO_PRELIM 4
#define FS_IBASKET 8
I'm wondering what FS_NO_DACHE, FS_IBASKET means ! Can somebody tell me
that. Thanks.
J.C. Chuang
------------------------------
From: [EMAIL PROTECTED] (Timothy Murphy)
Subject: msdos as module in 2.2.6
Date: 2 May 1999 16:08:39 +0100
When I try to install MSDOS + FAT + VFAT as modules
I get unresolved symbol is_binary in msdos.o .
(It seems to be defined and used in fat/misc.o .
However, insmod-ing fat first does not help.)
--
Timothy Murphy
e-mail: [EMAIL PROTECTED]
tel: +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland
------------------------------
From: Luc Wastiaux <[EMAIL PROTECTED]>
Subject: Missing libraries (I think)
Date: Mon, 03 May 1999 06:55:02 +0200
This may not be the best newsgroup for this, so sorry if this is
off-topic.
I am trying to compile qstat for linux,
when I run "make linux", I get these error messages:
cannot find stdio.h
stdlib.h
errno.h
sys/types.h
etc. etc.
Could you tell me what package I have to install?
Thank you.
--
Luc Wastiaux
[EMAIL PROTECTED]
[remove spam block to reply]
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: Problems compiling kernel 2.2.5
Date: 2 May 1999 23:54:26 -0500
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, F Kuijt wrote:
> "Clock skew detected" and something like "Modification in time" and
> GCC tell's me that my build may so not be complete.
(Why are you working with 2.2.5? That is two versions behind.)
Here are some possibilities:
1) You are working on a remote filesystem, where the server is not
synchronized with the local machine that is doing the compiling.
This can lead to files having timestamps that appear (locally) to
be in the future.
2) You are using source code that has been (very!) recently modified
in some time zone east of you. Files may then appear to have timestamps
in the future.
--
Paul Kimoto <[EMAIL PROTECTED]>
------------------------------
From: "FM" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.misc,comp.os.linux.powerpc
Subject: Mac-emulation on Linux?
Date: Sun, 2 May 1999 22:47:08 -0700
I will attend a college this fall that is predominantly
Mac-oriented. While they state that Unix and Windows are
supported by the campus network, it seems that a Macintosh-
compatible system might be necessary to fully take
advantage of the system. For example, many softwares are
written for Macintosh by the faculty, the letter for last
year's computer purchase recommendation notes. I'm a little
disgruntled, as I have been planning on getting a Linux
system for college. So here are a couple of options I have:
1. Just buy a X86 machine and install Linux on it (if not
preinstalled). Get some Mac-emulation software if necessary.
The problem I have with this option is that I have no idea
about the availability of Mac-emulation softwares for Linux
nor the degree of compatibility they provide.
2. Buy a Macintosh and dual boot with Linux/MacOS
Well I think this is a nice compromise but I'm not sure how
well Linux runs on Macintosh. I'm fairly sure that it will
be an improvement over Windows/MacOS, but I'm not even sure
if most Linux softwares are available for this setup (or if
it's generally source-level compatible).
I think my doubts stem mostly from my lack of knowledge
about the Macintosh systems, which I've used before but
never administered. Are these the only options I have
considering that I want to use Linux and remain compatible
with Mac at the same time? Any additional information
would be apprecited. Thanks in advance.
Dan.
------------------------------
From: Serge B Bromow <[EMAIL PROTECTED]>
Subject: 80x25 Video Memory & Color
Date: Mon, 03 May 1999 01:00:27 -0400
This is a multi-part message in MIME format.
==============4698FEC072D5841EA1F9DAFB
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
I am porting MB's of C code from a SCO unix/xenix
environment to Redhat 5.2. I have resolved
most of the differences i.e file llocking, semaphores etc.
with the exemption of the following;
1) For legacy applications using color backgrounds
I am unable to generate all 16 colors. The foreground
colors work fine with \E[3COLORm for colors 0-7 and
when set to hi_color \E[1m the bright 8 colors appear. The background
colors only set the first 8 dark colors. HOW do I access the
8 hi_color background colors. My setterm also does not set the
background color to bright. This means I have brown(orange) instead
of yellow and so on.
2) Most all 80x25 console apps use routines that map video memory
to the applications. Finding the location of memory is accomplished
in the example below.
ex:
char *vid_mem;
vid_mem = (char *) ioctl(fd,MAPVGA,0);
Simply writing data to "vid_mem + offset" allows me to write directly to
the video card and every second byte controls the fore/background
colors.
This method is very fast for full screen refreshes etc. HOW do I map
video memory into my apps under Linux..
Any help would be greatly appreciated, Thanks in advance
Serge
==============4698FEC072D5841EA1F9DAFB
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Serge Bromow
Content-Disposition: attachment; filename="vcard.vcf"
begin: vcard
fn: Serge Bromow
n: Bromow;Serge
org: Omen Systems Int'l Corp
adr: 1930 Norwood Avenue;;;Ottawa;Ontario;K1H5K6;Canada
email;internet: [EMAIL PROTECTED]
tel;work: (613) 526-1831
tel;fax: (613) 731-0524
note: Web: http://www.omensys.com/
x-mozilla-cpt: ;0
x-mozilla-html: TRUE
version: 2.1
end: vcard
==============4698FEC072D5841EA1F9DAFB
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Serge Bromow
Content-Disposition: attachment; filename="vcard.vcf"
begin: vcard
fn: Serge Bromow
n: Bromow;Serge
org: Omen Systems Int'l Corp
adr: 1930 Norwood Avenue;;;Ottawa;Ontario;K1H5K6;Canada
email;internet: [EMAIL PROTECTED]
tel;work: (613) 526-1831
tel;fax: (613) 731-0524
note: Web: http://www.omensys.com/
x-mozilla-cpt: ;0
x-mozilla-html: TRUE
version: 2.1
end: vcard
==============4698FEC072D5841EA1F9DAFB
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Serge Bromow
Content-Disposition: attachment; filename="vcard.vcf"
begin: vcard
fn: Serge Bromow
n: Bromow;Serge
org: Omen Systems Int'l Corp
adr: 1930 Norwood Avenue;;;Ottawa;Ontario;K1H5K6;Canada
email;internet: [EMAIL PROTECTED]
tel;work: (613) 526-1831
tel;fax: (613) 731-0524
note: Web: http://www.omensys.com/
x-mozilla-cpt: ;0
x-mozilla-html: TRUE
version: 2.1
end: vcard
==============4698FEC072D5841EA1F9DAFB==
------------------------------
From: [EMAIL PROTECTED] (Andrew McDonald)
Subject: Re: msdos as module in 2.2.6
Date: 2 May 1999 16:53:12 GMT
Timothy Murphy <[EMAIL PROTECTED]> wrote:
> When I try to install MSDOS + FAT + VFAT as modules
> I get unresolved symbol is_binary in msdos.o .
> (It seems to be defined and used in fat/misc.o .
> However, insmod-ing fat first does not help.)
This is a known problem.
To fix it you can just add:
EXPORT_SYMBOL(is_binary);
to linux/fs/fat/fatfs_syms.c
Alternatively, kernel 2.2.7 fixes it.
Andrew
--
Andrew McDonald
andrew at mcdonald.org.uk
http://ban.joh.cam.ac.uk/~adm36/
------------------------------
From: [EMAIL PROTECTED] (Jonathan A. Buzzard)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Linux system ID, is there such a thing?
Date: Sat, 1 May 1999 20:49:30 +0000
In article <[EMAIL PROTECTED]>,
"G. Sumner Hayes" <[EMAIL PROTECTED]> writes:
> "Jonathan A. Buzzard" wrote:
>> The *really* funny thing about dongles is watching a dongle work the
>> far side of a automatic parallel switch, with several computers using
>> the same dongle simultaneously without any problems.
>
> That'll only work if the dongle is reentrant, of course. If it's
> stateful, you could have problems.
>
> Hmmm. Visions of non-reentrant dongles designed to work with
> multi-threaded programs that marshal dongle access through a single
> thread are dancing in my mind. It's amazing how much time is wasted
> on intractable problems.
>
Do stateful dongles exist? In fact are stateful dongles practical? What
would happen if the software crashed leaving the dongle in some unknown
state?
Experience would however suggest that most dongles are indeed stateless
and this method works fine. Given that most (all) dongles work on SPP
parallel ports I would wager that a single dongle could serve upto sixteen
computers spread over a distance of 350m using the RJ11 modular link device
for sharing printers I have seen in catalogues.
JAB.
--
Email: [EMAIL PROTECTED]
------------------------------
From: Gerard van der Sel <[EMAIL PROTECTED]>
Subject: Re: Linux Real-Time I/O
Date: Mon, 03 May 1999 09:02:13 +0200
Reply-To: [EMAIL PROTECTED]
Jim & Lisa Meils wrote:
>
> I am a newbie to Linux, and have a somewhat simple question. I have
> written an increadibly simple I/O program for DOS using QBASIC, and
> would like to write a similar program for Linux and possibly use the
> timing and multitasking capabilities to dim lights with it. Currently I
> use the
> parallel port for the output and control up to 8 light circuits. What
> is the best route to take with Linux? Can Linux control in real-time
> i.e. run a routine EXACTLY 120 times/sec ?
>
> The current program simply writes a byte to the printer port, and the
> port holds the value. Each bit is connected(optically) to a light
> circuit. I know the Windows folks are trying to invade the real-time
> I/O market to displace PLCs with Windows boxes. Is there any work being
> done in the Linux world to handle similar apps? It seams a natural due
> to it's reliability compared to Windows. The thing I need help on is how
> to address the port and send data to it quickly.
For a newbie like you I have pointers:
- coffe HOWTO: It gives compleet code which compiles to control your LPT
port hardware.
- parallelport HOWTO
- RTC module: This gives a sort of Real Time interrupt on Linux.
List/Grep the source files with RTC in it. They explain a lot.
Search on the internet for "RTC and Linux" and an example will pop up.
It from the author of the RTC functionality (Gortemaker or so)
This will lead to 128 times/second exactly.
I run my modeltraclayout with this interrupt.
--
Met vriendelijke groet,
Gerard van der Sel
Mailto:[EMAIL PROTECTED]
"De dinosaurussen hadden hun komeet, wij hebben de Windows computer" -
me
"The box said: 'install on Windows 95, NT 4.0 or better'.
So I installed it on Linux."
------------------------------
From: Roope Anttinen <[EMAIL PROTECTED]>
Subject: Re: KDE without gif support
Date: 3 May 1999 08:04:12 GMT
Reply-To: [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
> Hi!
> I compiled KDE-1.1 without any errors, but it has no gif support.
> I never installed the giflib included in the kdesupport package
> and used an installed 4.0 of giflib from my system instaed.
> With KDE-1.0 I doesn't had these problems...
You probably have qt > 1.43 ? Due to gif patent issues qt-1.44 has disabled
gif support.
Roope
--
MicroSoft? is that some kind of a toilet paper?
PS: Look for address here, not from headers. And remove NOSPAM's
___________________________________________________________________________
[EMAIL PROTECTED] / [EMAIL PROTECTED]
+358 9 812 7567 / +358 500 445 565 / +358 49 445 565
http://myy.helia.fi/~anttiner/index.html
===========================================================================
Helsinki Business Polytechnic - Institute of information technology
------------------------------
From: [EMAIL PROTECTED] (Chris Johnson)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.misc,comp.os.linux.powerpc
Subject: Re: Mac-emulation on Linux?
Date: Mon, 03 May 1999 15:44:04 -0400
In article <7gjd3s$cdo$[EMAIL PROTECTED]>, "FM"
<[EMAIL PROTECTED]> wrote:
>2. Buy a Macintosh and dual boot with Linux/MacOS
>
>Well I think this is a nice compromise but I'm not sure how
>well Linux runs on Macintosh. I'm fairly sure that it will
>be an improvement over Windows/MacOS, but I'm not even sure
>if most Linux softwares are available for this setup (or if
>it's generally source-level compatible).
FWIW this will set you squarely in the 'hardcore Linux' camp. I've
found that most stuff just compiles- but this does mean you need to have
source and be able to do 'configure, make, make install'. However it is a
pleasure getting your hands dirty with this stuff, and it's also nice to
avoid bogotification of the linux userbase one installation at a time :)
running linuxppc means you _will_ be compiling programs from source, and
this is not a bad thing.
>I think my doubts stem mostly from my lack of knowledge
>about the Macintosh systems, which I've used before but
>never administered. Are these the only options I have
>considering that I want to use Linux and remain compatible
>with Mac at the same time? Any additional information
>would be apprecited. Thanks in advance.
>Dan.
Well, _I_ like my dualboot 9500. :)
It's more like 'mac being compatible with Linux', and in fact I think
you might be _better_ off coexisting with MacOS. You see, all Microsoft
OSes traditionally tend to helpfully erase partitions they don't
understand when installed. This can't happen with MacOS, because MacOS
_does_ understand what a linux partition is- sort of. Actually MacOS
thinks the Linux partitions are Unix_SVR2 or something (haven't looked in
a while)... in other words, A_UX, the old Apple 68K Unix. I still like
having a separate physical disk for all my linux partitions tho :)
There are programs both on MacOS (LinuxDisks, very clear and usable)
and Linux (hmount /dev/sda5, xhfs /dev/sda5, humount /dev/sda5 for my Boot
partition which is sda5) which can freely copy stuff back and forth- in
fact I'm using my Linux disk as free data storage for the GTOPO30 dataset,
and could do the opposite if I was spending all my time in Linux. The one
rule of thumb I use is 'that which writes should be native', and I never
break it- LinuxDisks IMHO is for _reading_ linux disks, and xhfs is for
_reading_ HFS, and I won't use either to write data to the non-native
filesystem. Some people have problems doing that- I have had no problems
at all.
Heck, just go PPC! The G3s are very nice but some older models have
other virtues, such as my 9500 (1st gen PCI, but with 12 DIMM slots and 6
PCI slots and 2 scsi busses all built in (2nd scsi bus is just for the
external port!)). As I see it, anything that's really proper linux can be
gotten to run on PPC too. As far as the tweeky stuff like Riva cards et
al- most of those don't run on Linux yet either, and you _can_ get 3dfx
cards for Macs- I have a Voodoo2 and am anxiously watching to see if they
get proper OpenGL support (though they already have improper Mesa support
;) )
Mac Linux boxes are, in a peculiar way, the best of both worlds (as
long as you don't pick a wrong one like Performas with puzzle box cases or
old Nubus ones- go for 604, 604e or G3 if you can). You get to run genuine
native Linux- only on an architecture with 32 registers instead of 4 picky
registers, with a spiffy bus that has a fast-updating mouse (assign fkeys
for extra buttons) and can handle Wacom tablets in linux by reading them
as mice :)
The 'lilo' for Macs (if I'm not mixing metaphors?) is terrific- a
little program called BootX that's just an easily installed control
panel/extension that launches Linux for you. It seems quite troublefree
and safe.
I'm still puzzling over some xmodmap details- Jamie Zawinski worked out
the ultimate Mac keyboard keymap for the MkLinux setup, but I haven't
figured out how to _use_ it- but all in all, even though I do find MacOS
quite useful in my single-user way, I don't even consider using my Linux
HD for extra scratch disks or something. I _like_ having it around because
of how nicely a lot of it works out. For basically a Mac guy, that's
pretty high praise, particularly because I don't _need_ it per se for
anything I do. I just _want_ it around. If some Mac-ism or political
machiavellianism sufficiently annoys me, I'll boot into Linux and spend a
little time poking around in a world that nobody can own or betray, and
feel much better :)
Chris Johnson
@airwindows.com
chrisj
------------------------------
From: [EMAIL PROTECTED] (H. Peter Anvin)
Subject: Re: CVS (Re: Bill Gates, self made man, NOT!)
Date: 2 May 1999 23:06:59 GMT
Reply-To: [EMAIL PROTECTED] (H. Peter Anvin)
Followup to: <[EMAIL PROTECTED]>
By author: Peter Dalgaard BSA <[EMAIL PROTECTED]>
In newsgroup: comp.os.linux.development.system
>
> ...but do we have anything that's *really* better? I.e. do we want to
> push CVS development in the right direction or do we want something
> new?
>
>From what I've seen, several of these problems are structural.
Attempting another graft on top of CVS is just going futher down the
hole. Sometimes one just have to fix everything from scratch...
-hpa
--
"The user's computer downloads the ActiveX code and simulates a 'Blue
Screen' crash, a generally benign event most users are familiar with
and that would not necessarily arouse suspicions."
-- Security exploit description on http://www.zks.net/p3/how.asp
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Linux disk defragmenter
Date: 3 May 1999 05:01:13 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Peter Samuelson <sampo.creighton.edu!psamuels>]
> > I heard the same objection from Mark Lord, the Linux IDE
> > maintainer. The context was specifically that someone suggested
> > the IDE block access elevator be bidirectional
[Mark Hahn <[EMAIL PROTECTED]>]
> bidirectional request sorting suffers from starvation; Linux's
> scheme (cscan) doesn't. (well, Linux discriminates against write
> requests, actually, and that could cause starvation, but that's not
> really part of head scheduling.)
I don't think the amount of starvation would be a very big deal. But
other issues would. For one thing, if you assume that the disk lies to
you about its exact physical geometry, how do you avoid getting bitten
*hard* by track skew when you seek backwards? Answer: you have to seek
backwards in largish chunks (seeking forwards within those chunks),
complicating your algorithm and probably negating most if not all of
your increased throughput.
In any case, minds much wiser than mine have already argued this one
out and I agree that a bidirectional elevator is in this case a
great-sounding theory with too many holes in it.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Crossposted-To: comp.os.linux.hardware
Subject: Re: /dev/hda1 has reached maximal mount count, check forced
Date: 3 May 1999 05:10:31 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Nelson Minar <[EMAIL PROTECTED]>]
> I understand why ext2fs does this, but the default behaviour is very
> obnoxious on laptops. I boot my laptop very often (once a day?)
Yeah, that situation definitely calls for tuning the default.
> I can use tune2fs to change the behavior, sure. It'd be neat if
> ext2fs could use some smarter heuristic on when to check, though. I
> would think counting the number of disk writes would be more useful
> than mounts, but maybe that's not simple to do.
Nice idea, but I agree that it would probably be a major kludge to
implement. What you really want is for a distribution installation
program -- you know, the program that handles your initial fdisk and
mke2fs calls -- to ask for your expected usage pattern and send the
appropriate flags into mke2fs right from the start. Some mke2fs
defaults lodged somewhere in /etc might not be bad either, though that
might be overengineering it.
Of course the ultimate answer (TM) is sct's journalling extensions,
which among other things make `fsck' very fast and integrated into the
standard `mount' call. The alpha version is due out Real Soon Now.
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Crossposted-To: comp.os.linux.development.apps,comp.unix.programmer
Subject: Re: [ANN] CodeWarrior for Red Hat Linux, GNU ed. Shipping
Date: 3 May 1999 05:19:40 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[G. Sumner Hayes <[EMAIL PROTECTED]>]
> No, that means that it doesn't include a CodeWarrior compiler but is
> just an IDE for use with egcs/gcc.
^^^^^^^^
Hmmm, I wonder, can we start just saying `gcc' again now that rms et al
have re-blessed egcs? Or do we have to wait until they release
something that calls itself gcc again?
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: [EMAIL PROTECTED] (Daniel Robert Franklin)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.misc,comp.os.linux.powerpc
Subject: Re: Mac-emulation on Linux?
Date: 3 May 99 06:48:24 GMT
"FM" <[EMAIL PROTECTED]> writes:
>I will attend a college this fall that is predominantly
>Mac-oriented. While they state that Unix and Windows are
>supported by the campus network, it seems that a Macintosh-
>compatible system might be necessary to fully take
>advantage of the system. For example, many softwares are
>written for Macintosh by the faculty, the letter for last
>year's computer purchase recommendation notes. I'm a little
>disgruntled, as I have been planning on getting a Linux
>system for college. So here are a couple of options I have:
>1. Just buy a X86 machine and install Linux on it (if not
>preinstalled). Get some Mac-emulation software if necessary.
>The problem I have with this option is that I have no idea
>about the availability of Mac-emulation softwares for Linux
>nor the degree of compatibility they provide.
OK, you can buy a thing called "Executor":
http://www.ardi.com
It's not freeware but it *does* work, and there is a free time-limited
demo (5 or 10 minutes I think). $35 US for students for the Linux version.
It runs surprisingly quickly.
>2. Buy a Macintosh and dual boot with Linux/MacOS
This is also a viable option, but somewhat more expensive :)
- Daniel
--
******************************************************************************
* Daniel Franklin - Postgraduate student in Electrical Engineering
* [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
******************************
