Linux-Development-Sys Digest #202, Volume #7 Thu, 16 Sep 99 22:13:44 EDT
Contents:
Re: UDMA vs IDE: performance comparison wanted (Joseph H Allen)
Re: mmap problems (Mark Hahn)
Re: Booting Linux on Sparc Classic using Null-Modem Connection for Console. (Paul J
Collins)
Re: Device driver changes for new Kernel versions... (ellis)
Re: glibc2.1.2 and libX11 problem (Matthew)
Re: write / writev guaranteed atomic? (H. Peter Anvin)
Re: Linux standards compliance (Mario Klebsch)
Re: Linux without PC BIOS (H. Peter Anvin)
Re: unix98 pty's problems (A Guy Called Tyketto)
Re: Where can I find the handy SWEEP utility? (Duane Smeckert)
Re: UDMA vs IDE: performance comparison wanted (Philip Brown)
Re: Why so inefficient source RPM's ?? (Piercarlo Grandi)
SANE or Scanners (Joel Hanger)
Re: Linux standards compliance (Warren Young)
IDE register programming: bizzare twists (Philip Brown)
Re: Figure Out The MS Source Code Yourself (Joel Hanger)
Re: how to become a linux kernel developer? (ellis)
Re: Why so inefficient source RPM's ?? (Christopher Browne)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Joseph H Allen)
Subject: Re: UDMA vs IDE: performance comparison wanted
Date: Thu, 16 Sep 1999 15:41:47 GMT
In article <7rr0va$dun$[EMAIL PROTECTED]>,
Mark Hahn <[EMAIL PROTECTED]> wrote:
>Joseph H Allen <[EMAIL PROTECTED]> wrote:
>> since Linux makes pessimistic settings by default.
>Linux mostly uses whatever you set in bios.
No, it doesn't. It doesn't even use 32-bit transfers unless you give the
hdparm command.
>> I'm very surprised that the throughput would have doubled- even the simple
>> programmed interface should be way faster than the actual data rate of the
>> media (average bytes per track * disk RPM / 60).
>PIO maxes out at either 13 or 16 MB/s, I forget which. that's sufficiently
>slow that modern disks (which easily sustain 20 MB/s) can slip a rev.
Modern drives also have buffering, so they only slip a rev from time to
time, not after every revolution. In other words, the effect should not be
that big. If the drive can sustain 20MB/s and I use the best PIO mode, I
should get 16MB/s, not 4MB/s. For drives that can only sustain 12MB/sec, it
shouldn't matter whether it's PIO or UDMA (except that UDMA will waste fewer
CPU cycles).
--
/* [EMAIL PROTECTED] (192.74.137.5) */ /* Joseph H. Allen */
int a[1817];main(z,p,q,r){for(p=80;q+p-80;p-=2*a[p])for(z=9;z--;)q=3&(r=time(0)
+r*57)/7,q=q?q-1?q-2?1-p%79?-1:0:p%79-77?1:0:p<1659?79:0:p>158?-79:0,q?!a[p+q*2
]?a[p+=a[p+=q]=q]=q:0:0;for(;q++-1817;)printf(q%79?"%c":"%c\n"," #"[!a[q-1]]);}
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Subject: Re: mmap problems
Date: 16 Sep 1999 14:45:42 GMT
> if ((dummy = mmap(0,len,PROT_WRITE | PROT_READ ,MAP_PRIVATE ,fd,off))
> == MAP_FAILED) {
MAP_PRIVATE means specifically "don't write back to the file"...
------------------------------
From: Paul J Collins <[EMAIL PROTECTED]>
Subject: Re: Booting Linux on Sparc Classic using Null-Modem Connection for Console.
Date: 15 Sep 1999 23:24:29 +0100
>>>>> "David" == David Wragg <[EMAIL PROTECTED]> writes:
--snip--
David> can telnet into it, but since a telnet login won't accept
David> root you are stuck.
--snip--
You can telnet in as a normal user and use "su -" to become root,
environment and all.
Paul.
--
Paul Collins <[EMAIL PROTECTED]> Public Key On Keyserver.
Fingerprint: 88BA 2393 8E3C CECF E43A 44B4 0766 DD71 04E5 962C
"I am a stranger in a strange land,
distracted by bright and shiny objects."
------------------------------
From: [EMAIL PROTECTED] (ellis)
Subject: Re: Device driver changes for new Kernel versions...
Date: 16 Sep 1999 18:22:06 GMT
In article <[EMAIL PROTECTED]>,
Shane Luttrell <[EMAIL PROTECTED]> wrote:
> I am new to Linux development. I am coming over from windows
>development to write some device drivers for a V2.2.5 kernel. I have
>been using the book Linux Device Drivers (author Alessandro Rubini) as a
>reference (Linux seems to be easier to write hardware specific code for,
>especially in the area of DMA's). Pretty good book, but unfortunately
>the most recent kernel mentioned was the 2.1 version. Does anyone have
>a FAQ or link that has a good summary of all the differences in the new
>2.2 kernels as related to writing device drivers.
Try this:
http://www.atnf.csiro.au/~rgooch/linux/docs/porting-to-2.2.html
--
http://www.fnet.net/~ellis/photo/linux.html
------------------------------
From: Matthew <[EMAIL PROTECTED]>
Subject: Re: glibc2.1.2 and libX11 problem
Date: Thu, 16 Sep 1999 19:48:18 +0100
On Sun, 12 Sep 1999, Allin Cottrell wrote:
>Dawg Lone wrote:
>>
>> Thanks for the reply, I upgraded from libc5.
>
>> Still I get the _xstat not found when I try to start any windows with it.
>
>What about all your X libraries? For a functional glibc system
>they must all be linked against glibc, not libc5. That is, it's
>not enough just to replace the "base" libraries when upgrading.
That, I believe, is why there are both XFree86-libc5 and XFree86-glibc
distributions. I've always used the same one that was on the system to
start with.
Matthew
------------------------------
From: [EMAIL PROTECTED] (H. Peter Anvin)
Subject: Re: write / writev guaranteed atomic?
Date: 16 Sep 1999 19:33:24 GMT
Reply-To: [EMAIL PROTECTED] (H. Peter Anvin)
Followup to: <[EMAIL PROTECTED]>
By author: "Glen Parker" <[EMAIL PROTECTED]>
In newsgroup: comp.os.linux.development.system
>
> A little question about write* calls...
>
> If a process is -KILL'ed while in the middle of executing a write or writev
> call, what happens? Is the call guaranteed to either complete or not do
> anything? I'm pretty sure that write is safe, but what about writev?
>
That's not true. write() is only atomic if the size is less than
PIPE_BUF bytes. Otherwise you might have it return with a short write.
-hpa
--
<[EMAIL PROTECTED]> at work, <[EMAIL PROTECTED]> in private!
------------------------------
From: [EMAIL PROTECTED] (Mario Klebsch)
Subject: Re: Linux standards compliance
Date: Wed, 15 Sep 1999 19:29:36 +0200
Josef Moellers <[EMAIL PROTECTED]> writes:
>Mario Klebsch wrote:
>> I have been long enough in the evil *real* world to learn, that even
>> Software, whoch is payed for, often is not work its money. Several
>> times, when the vendors support was unable to help me any further, I
>> got part of ther source to fix it myself. And far too often, it was
>> really poor quality. :-(
>> Since I really needed the Software, I even rewrote the entire program
>> to fit my needs. Without the source, I would have been lost, any my
>> project, too.
>So you're in fact stating that the programmer who wrote the, apparently,
>bad code, did deliver something, i.e. the design. Otherwise, you wouln't
>"have been lost", but you would have written the entire thing yourself
>in the first place.
I had no choice, as the software was embedded software. I needed the
hardware and was not able to do that hardware myself. This is almost
like discussing drivers. When I by e.g. a scanner, I want to by
hardware, not software. However,t he software is an essential part of
it, and normally I do not have any chance to do that myself, sice the
documentation needed for it is not available.
73, Mario
--
Mario Klebsch [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (H. Peter Anvin)
Subject: Re: Linux without PC BIOS
Date: 16 Sep 1999 19:34:37 GMT
Reply-To: [EMAIL PROTECTED] (H. Peter Anvin)
Followup to: <7rqj1p$rre$[EMAIL PROTECTED]>
By author: "Roland Zitzke" <[EMAIL PROTECTED]>
In newsgroup: comp.os.linux.development.system
>
> Hallo,
> I want to run Linux on an embedded system which doesn't have a BIOS. Has
> anyone ever written loader code that gets by without the BIOS code? Once the
> system in in protected mode I would assume that no BIOS calles are performed
> anymore. (it is not a P&P system with a P&P BIOS wrapper needed).
>
There *are* BIOS callbacks in Linux for a few things (PCI BIOS, APM).
It doesn't *need* them, however, and they can be configured out.
You need to rewrite setup.S for your platform.
-hpa
--
<[EMAIL PROTECTED]> at work, <[EMAIL PROTECTED]> in private!
------------------------------
From: [EMAIL PROTECTED] (A Guy Called Tyketto)
Subject: Re: unix98 pty's problems
Date: 16 Sep 1999 12:46:00 -0500
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
Juergen Heinzl ([EMAIL PROTECTED]) wrote:
>
> >Any more programs?
>
> Yes. Mind I am not holding back; I've just no list either, so
> script for instance (util-linux). Try your talk and screen
> binaries too, if in use.
>
You may also want to check out w, who, last, and top. They MIGHT
use old-style ttys. But I think they are more dependent on /var/adm/utmp
and /var/adm/wtmp. Refer to the glibc-2.1.2 FAQ for how to get those
working. You may have to do a 'cat /dev/null > /var/adm/wtmp' to get
those cleared, and working right again. More importantly, recompile ps.
Otherwise, you won't see anything coming up for processes run by Unix98
ptys.
BL.
- --
Brad Littlejohn | Email: [EMAIL PROTECTED]
Unix Systems Administrator, | [EMAIL PROTECTED]
WebMaster, NewsMaster.. Smeghead! :) | http://www.omnilinx.net/~tyketto
PGP: 1024/E9DF4D85 67 6B 33 D0 B9 95 F4 37 4B D1 CE BD 48 B0 06 93
=====BEGIN PGP SIGNATURE=====
Version: GnuPG v1.0.0 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE34O3ryBkZmuMZ8L8RAl6uAJ43zm2v2WO6daCAiQuQ56IrBEyukwCgsuKj
yR58/Np+lJzg94se9meLCic=
=FS3+
=====END PGP SIGNATURE=====
------------------------------
From: Duane Smeckert <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux,comp.os.linux.questions,redhat.rpm.general,linux.redhat.rpm
Subject: Re: Where can I find the handy SWEEP utility?
Date: Thu, 16 Sep 1999 14:17:55 -0700
RedHat 6 uses something new instead of sudo, I am still trying
to figure out how it works.
Horst von Brand wrote:
> On Fri, 10 Sep 1999 11:50:41 +0100,
> Paul Hendrick <[EMAIL PROTECTED]> wrote:
> >On Thu, 09 Sep 1999 17:39:38 +0100, Joey McAlerney
> ><[EMAIL PROTECTED]> wrote:
> >>While working on Irix boxes over the summer, I got to take advantage of
> >>the "sweep" utility. It simply allowed you to gain root access to a
> >>machine by simply typing "sweep" (that is, of course, someone else with
> >>root access specified that you could do so). I am looking for this
> >>utility for Redhat, and couldn't find it so far. If it exists, could
> >>you please point me to it?
>
> There are several packages that allow this. If you mean full root access,
> there is lowly su(1) ;-) For restricted access (i.e., some commands only for
> selected users) there is sudo and a lot of others.
>
> >"linux single" at the lilo prompt IIRC.
>
> Works, but is a bit messy ;-)
> --
> Horst von Brand [EMAIL PROTECTED]
> Casilla 9G, Vi�a del Mar, Chile +56 32 672616
------------------------------
From: [EMAIL PROTECTED] (Philip Brown)
Subject: Re: UDMA vs IDE: performance comparison wanted
Reply-To: [EMAIL PROTECTED]
Date: 16 Sep 1999 17:11:07 GMT
On Thu, 16 Sep 1999 12:48:14 GMT, [EMAIL PROTECTED] wrote:
>...
>That's odd. For a WDC AC26400B, I get 2.37MB/sec without the hdparm
>settings and 7.79MB/sec with them (>3 times faster).
in PIO mode?!!
Not bad if so...
> This is a cheap-o 6GB
>drive, the kernel version is 2.0.30, and the CPU is K6-300, and the
>motherboard is a very cheap MSI-5184. Perhaps something is different for
>better hardware?
heh.
--
[Trim the no-bots from my address to reply to me by email!]
[ Do NOT email-CC me on posts. Pick one or the other.]
http://thomas.loc.gov/cgi-bin/bdquery/z?d105:SN01618:@@@D
The word of the day is mispergitude
------------------------------
From: [EMAIL PROTECTED] (Piercarlo Grandi)
Crossposted-To: linux.redhat.misc,linux.redhat.rpm
Subject: Re: Why so inefficient source RPM's ??
Date: 16 Sep 1999 23:24:05 +0100
>>> On Fri, 27 Aug 1999 22:02:49 -0500, "Julie" <[EMAIL PROTECTED]> said:
[ ... ]
Julie> I don't want to risk another round of "Julie is a big meanie",
Julie> but I recently examined the RPM's for the Shadow Password Suite
Julie> and was seriously torqued off.
I was too with others. Oh my, RedHat, why are Linux wisdom standards
slipping? Just consider the entirely idiotic directory layout for INN
2.2 (INN 1.7.2 had a decent directory layout).
I am afraid that what's happening is that RedHat as it expands is hiring
people who don't really get it.
Julie> I'm fairly sure this isn't their intent, but I almost feel like
Julie> the source RPMs are made impossible to deal with so people will
Julie> get their pre-compiled binary distribution from Red Hat ...
Uhm, in part this may be because of the newbie hires that RedHat may be
getting; it take some real cool to do things that don't just work, but
work in an elegant and simple way.
Also, in part it may be that doing a SPEC file is now a black art. The
thoroughly disgusting "Maximum RPM" is now thoroughly obsolete, and the
RPM HOWTO has not been maintained for ages. Any recent RPM documentation
out there?
------------------------------
From: Joel Hanger <[EMAIL PROTECTED]>
Subject: SANE or Scanners
Date: Thu, 16 Sep 1999 16:26:08 -0700
==============FDFDFED0BB69239701B7CD86
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Hello,
I recently installed sane but couldn't get it working so I took it
off.. I still have the rpm for it however. I was wondering if there are
any good tutorials or HOWTO'S (as I couldn't find any) on setting up a
scanner? Also if their is a linux driver for the Microtek ScanMaker
E3? Thanks in Advance.
Joel
--
"Put your hand on a hot stove for a minute,
and it seems like an hour. Sit with a pretty
girl for an hour, and it seems like a minute.
THAT'S relativity."
Albert Einstein
==============FDFDFED0BB69239701B7CD86
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML>
Hello,
<BR> I recently installed sane but couldn't get it working
so I took it off.. I still have the rpm for it however. I was
wondering if there are any good tutorials or HOWTO'S (as I couldn't
find any) on setting up a scanner? Also if their is a linux
driver for the Microtek ScanMaker E3? Thanks in Advance.
<P>Joel
<PRE>--
"Put your hand on a hot stove for a minute,
and it seems like an hour. Sit with a pretty
girl for an hour, and it seems like a minute.
THAT'S relativity."
Albert Einstein</PRE>
</HTML>
==============FDFDFED0BB69239701B7CD86==
------------------------------
From: Warren Young <[EMAIL PROTECTED]>
Subject: Re: Linux standards compliance
Date: Thu, 16 Sep 1999 16:42:53 -0600
Peter Samuelson wrote:
>
> You can change an ABI without changing the API. Fixed ABI means you
> never have to recompile. Fixed API means you never have to edit.
> Linux, within a stable kernel series, provides a fixed API (or should,
> I don't know of any counterexamples) but not a fixed ABI.
For others reading this, an example of why the ABI might change without
changing the API: a kernel structure gets a field added to it. All the
drivers that use that structure still work the same after a recompile,
but the ABI is different because the size and layout of that structure
are now different.
Limiting the ABI implies that you can't even expand existing data
structures. The best you can do is create different versions of the
ABI, and let the driver tell the system which version of the ABI it
expects. Then you carefully standardize the ABI, and don't change it
until you absolutely have to, and then create a new version. That's
what SCO's UnixWare does, FWIW.
> The ones actively maintained, anyway. Bit rot definitely is
> exacerbated by this scheme but on the whole I think it works well.
One of the beauties of the Linux model is that bit rot in the drivers
only occurs when the driver in question doesn't have many people
interested in it. Example: the MFM hard disk driver. No one cared
about it, so it rotted away. That's as it should be, IMHO.
--
= Warren Young: www.cyberport.com/~tangent | Yesterday it worked.
= ICBM Address: 36.8274040N, 108.0204086W, | Today it is not working.
= alt. 1714m | Windows is like that.
------------------------------
From: [EMAIL PROTECTED] (Philip Brown)
Subject: IDE register programming: bizzare twists
Reply-To: [EMAIL PROTECTED]
Date: 17 Sep 1999 00:29:32 GMT
So I'm browsing (grazing?) through the IDE+UDMA drivers.. plus I finally found
the "official" spec at www.t13.org (d1321r1c.pdf).. and something strikes me
as odd.
Some operations I would consider "normal".
- set "argument" registers
- set "command" register
and away you go.
But some things (specifically SET_FEATURES)
you set the commmand register, THEN the arg register, THEN LAST, the
sub-argument register.
And I don't see this order mentioned in the spec.
Did I miss it somehow? Does anyone know somewhere order of programming
is mentioned either in that spec, or in another "official" document?
(doesn't have to be from t13.org, can be intel, or whatever.
I've tried www.intel.com, no luck so far...)
--
[Trim the no-bots from my address to reply to me by email!]
[ Do NOT email-CC me on posts. Pick one or the other.]
http://thomas.loc.gov/cgi-bin/bdquery/z?d105:SN01618:@@@D
The word of the day is mispergitude
------------------------------
From: Joel Hanger <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: Figure Out The MS Source Code Yourself
Date: Thu, 16 Sep 1999 16:45:44 -0700
==============C7E83F1BF5A1890B8B5C82D0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
If Reverse Engineering is breaking the law then they who persecuting you
have broken the law, for how is it that they know that you used their
code without reverse engineering it??? The way I see it is that if you
use someone elses code then how are they going to prosecute you when in
doing so they have just admitted to breaking the law themselves?
just my .02$ worth
Joel
--
"Put your hand on a hot stove for a minute,
and it seems like an hour. Sit with a pretty
girl for an hour, and it seems like a minute.
THAT'S relativity."
Albert Einstein
==============C7E83F1BF5A1890B8B5C82D0
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML>
If Reverse Engineering is breaking the law then they who persecuting you
have broken the law, for how is it that they know that you used their code
without reverse engineering it??? The way I see it is that if you use someone
elses code then how are they going to prosecute you when in doing so they
have just admitted to breaking the law themselves?
<P>just my .02$ worth
<BR>Joel
<BR>
<BR>
<PRE>--
"Put your hand on a hot stove for a minute,
and it seems like an hour. Sit with a pretty
girl for an hour, and it seems like a minute.
THAT'S relativity."
Albert Einstein</PRE>
</HTML>
==============C7E83F1BF5A1890B8B5C82D0==
------------------------------
From: [EMAIL PROTECTED] (ellis)
Subject: Re: how to become a linux kernel developer?
Date: 17 Sep 1999 00:43:02 GMT
In article <[EMAIL PROTECTED]>,
hoz <[EMAIL PROTECTED]> wrote:
>Any advice for a junior level C programmer who wants to get up to
>speed on programming for linux with the goal of working on linux
>kernel development. Other than having a linux system (which
>distribution is best for dev?), the src, and the docs, what else
>should I do?
Read the source and try to understand what it is doing.
Write a device driver that works.
Read the kernel mailing list.
Find a project to work on and do it.
--
http://www.fnet.net/~ellis/photo/linux.html
------------------------------
From: [EMAIL PROTECTED] (Christopher Browne)
Crossposted-To: linux.redhat.misc,linux.redhat.rpm
Subject: Re: Why so inefficient source RPM's ??
Reply-To: [EMAIL PROTECTED]
Date: Fri, 17 Sep 1999 01:18:07 GMT
On 16 Sep 1999 23:24:05 +0100, Piercarlo Grandi <[EMAIL PROTECTED]> wrote:
>>>> On Fri, 27 Aug 1999 22:02:49 -0500, "Julie" <[EMAIL PROTECTED]> said:
>[ ... ]
>Julie> I don't want to risk another round of "Julie is a big meanie",
>Julie> but I recently examined the RPM's for the Shadow Password Suite
>Julie> and was seriously torqued off.
>
>I was too with others. Oh my, RedHat, why are Linux wisdom standards
>slipping? Just consider the entirely idiotic directory layout for INN
>2.2 (INN 1.7.2 had a decent directory layout).
>
>I am afraid that what's happening is that RedHat as it expands is hiring
>people who don't really get it.
That could unfortunately be...
>Julie> I'm fairly sure this isn't their intent, but I almost feel like
>Julie> the source RPMs are made impossible to deal with so people will
>Julie> get their pre-compiled binary distribution from Red Hat ...
>
>Uhm, in part this may be because of the newbie hires that RedHat may be
>getting; it take some real cool to do things that don't just work, but
>work in an elegant and simple way.
>
>Also, in part it may be that doing a SPEC file is now a black art. The
>thoroughly disgusting "Maximum RPM" is now thoroughly obsolete, and the
>RPM HOWTO has not been maintained for ages. Any recent RPM documentation
>out there?
What *desparately* needs to be done is to produce a reasonably
significant body of scripts that can go out and do "reasonability
checks" of SPEC files.
Debian has tools that validate that .deb packages conform to their
requirements that may be somewhat arbitrary and perhaps occasionally
arcane; the fact that they are at least *reported on* means that it is
possible for the "ignorant developer" to work up, in time, to being a
"somewhat more aware" developer.
It is highly unfortunate that there hasn't been such an outgrowth of
RPM-oriented development tools...
--
"Why use Windows, since there is a door?"
-- <[EMAIL PROTECTED]> Andre Fachat
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/lsf.html>
------------------------------
** 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
******************************
