Linux-Misc Digest #869, Volume #27 Tue, 15 May 01 20:13:03 EDT
Contents:
Netscape in black and white (David Nowak)
Re: Sound problem -> warnings about modules (Dave Uhring)
Re: Mandrake 8.0 problems (Dave Uhring)
Re: Compile GCC 2.95.3 in RedHat 7.0 failed !! (Thomas Corriher)
Re: System.map and multiple kernel versions. (Dave Uhring)
Re: Disk errors with kernel 2.4.2 (Dave Uhring)
Re: open ports (Jerry Kreps)
Re: Compile GCC 2.95.3 in RedHat 7.0 failed !! (Christian Rose)
Re: Compile GCC 2.95.3 in RedHat 7.0 failed !! (Grant Edwards)
telnet on Mandrake. ("atl.mediaone")
unzipping Mac BinHex4 files in Linux (Ramin Sina)
Re: Special Characters (Hartmann Schaffer)
sftp - gftp ("G Pollack")
Re: Mandrake 8.0 problems (Eric Anderson)
RPM: listing package sizes (Neil Zanella)
Re: Compile GCC 2.95.3 in RedHat 7.0 failed !! (Christian Rose)
Problems creating ramdisk ("Oscar Ricardo Silva")
Re: Compile GCC 2.95.3 in RedHat 7.0 failed !! (Christian Rose)
Reloading a module (bert buchholz)
Re: Compile GCC 2.95.3 in RedHat 7.0 failed !! (Christian Rose)
----------------------------------------------------------------------------
From: David Nowak <[EMAIL PROTECTED]>
Subject: Netscape in black and white
Date: Tue, 15 May 2001 22:19:17 +0100
On my Debian GNU/Linux 2.2, all icons in Netscape 4.75 are in black and
white! What should I install to have coloured icons?
Thanks,
--
David
------------------------------
From: Dave Uhring <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: Sound problem -> warnings about modules
Date: Tue, 15 May 2001 16:21:48 -0500
[EMAIL PROTECTED] wrote:
> I have installed SuSE 7.1 on my PC with the soundcard SB128 using YAST2
> and alsaconf. In the file /var/log/warn I have the following warnings:
>
> modprobe: Can't locate module snd-card-1
> modprobe: Can't locate module snd-card-2
> modprobe: Can't locate module snd-card-3
> modprobe: Can't locate module sound-slot-1
> modprobe: Can't locate module sound-service-1-0
> modprobe: Can't locate module sound-slot-2
> modprobe: Can't locate module sound-service-2-3
> modprobe: Can't locate module sound-slot-3
> modprobe: Can't locate module sound-service-3-3
> modprobe: Can't locate module char-major-145
>
> As far as I can see the following was changed in /etc/modules.conf when
> I installed the soundcard:
>
> #
> # YaST2: sound cards support
> #
> alias char-major-116 snd
> options snd snd_cards_limit=1 snd_major=116
> alias snd-card-0 snd-card-ens1370
> options snd-card-ens1370 snd_id=card1 snd_index=0
>
> #
> # YaST2: sound system dependent part
> #
> alias sound-slot-0 snd-card-0
> alias sound-service-0-0 snd-mixer-oss
> alias sound-service-0-1 snd-seq-oss
> alias sound-service-0-3 snd-pcm-oss
> alias sound-service-0-8 snd-seq-oss
> alias sound-service-0-11 snd-mixer-oss
> alias sound-service-0-12 snd-pcm-oss
>
>
> So can someone help me eliminate the warning messages? I don't even know
> where to start searching for the problem.
>
> Bernd
>
>
>
Your modprobe binary probably is out of date. It likely is trying to find
the sound modules in /lib/modules/<kernel_version>/misc and the 2.4.X sound
modules are in /lib/modules/2.4.X/kernel/drivers/sound.
# cd /lib/modules/2.4.X/kernel/drivers/sound
# mkdir /lib/modules/2.4.X/misc
# cp * /lib/modules/2.4.X
Then execute modprobe snd-card-ens1370 and see if it works.
------------------------------
From: Dave Uhring <[EMAIL PROTECTED]>
Subject: Re: Mandrake 8.0 problems
Date: Tue, 15 May 2001 16:27:18 -0500
Eric Anderson wrote:
> Yesterday I installed Mandrake 8.0 on top of a Mandrake 7.2 system. I
> reformatted /, but left the /home partition as it was. Both are ReiserFS
> formatted.
>
> I'm now having two problems. The first is that I keep getting error
> messages like this in the 'messages' file:
>
> May 15 07:29:06 catamount kernel: hda: dma_intr: status=0x51 { DriveReady
> SeekComplete Error }
> May 15 07:29:06 catamount kernel: hda: dma_intr: error=0x84 {
> DriveStatusError BadCRC }
>
> The drive containing all of the above partitions is a 6-month-old IBM 45G
> IDE drive. I would hope it's not going bad already!
>
> The second problem is that Mozilla 0.9 is now acting up. When I exit the
> program, at least one thread gets hung in uninterruptible wait state, and
> nothing short of a reboot will kill it.
>
> These two problems may be related. I'm wondering if there might be any
> incompatibility between the filesystem on /home (formatted using the
> ReiserFS version in Mandrake 7.2) and the ReiserFS in Mandrake 8.0.
>
> Any thoughts?
>
> Eric
You are going to have to install the 2.2.19 kernel. The 2.4.3 kernel is
incompatible with your hardware.
OR, you can turn off DMA in your BIOS.
------------------------------
From: [EMAIL PROTECTED] (Thomas Corriher)
Subject: Re: Compile GCC 2.95.3 in RedHat 7.0 failed !!
Reply-To: [EMAIL PROTECTED], abuse@[127.0.0.1]
Date: Tue, 15 May 2001 21:28:58 GMT
Give it up Peter. Everyone knows what happened, and everyone
already knows of Red Hat's guilt on this issue. Christian is
apparently chosing not to understand.
I use Red Hat myself, but the truth is most important. Ethics
and community are the keys things that bind Linux together.
For the sake of these things which we value, we must always
be truthful. This means being honest even when it hurts a
major distributor such as Red Hat. We all know that Red Hat
acted recklessly, and without common decency, in the compiler
situation. In fact, Red Hat refused to correct (patch) the
compiler until Linus himself publically attacked them. That
is pretty damn sad. I lost a lot of respect for them when all
of these things happened.
Just let it go, Peter. The truth is out, and you do not
need to bother.
--
From the desk of Thomas Corriher
The real email address is:
tcorriher at earthlink.
net
------------------------------
From: Dave Uhring <[EMAIL PROTECTED]>
Subject: Re: System.map and multiple kernel versions.
Date: Tue, 15 May 2001 16:29:56 -0500
Wayne Osborn wrote:
> In article <[EMAIL PROTECTED]>, "Dave Uhring"
> <[EMAIL PROTECTED]> wrote:
>
>> Wayne Osborn wrote:
>>
>>> Just curious as to the requirement for /boot/system.map when you have
>>> multiple kernel versions setup in lilo.
>>>
>>> For instance, if I upgrade my 2.2.16 kernel to 2.4.4 and want the
>>> option to support both with lilo, what system.map should I have in
>>> /boot ?
>>>
>>> Thanks in advance, and a big thanks to all in this NG and other Linux
>>> NG's for that matter that have taught me so much.
>>>
>>>
>> vmlinuz-2.2.16 -> System.map-2.2.16
>> vmlinuz-2.4.4 -> System.map-2.4.4
>>
>
> Alright, so the symbolic link System.map is setup by...???
> ie. does System.map always point to the map associated with the running
> kernel?
>
> Thanks.
>
I may have illustrated that incorrectly. Those are not symlinks, but
rather are associations. If you boot vmlinuz-2.2.16 and have a
System.map-2.2.16, the kernel will find its System.map.
------------------------------
From: Dave Uhring <[EMAIL PROTECTED]>
Subject: Re: Disk errors with kernel 2.4.2
Date: Tue, 15 May 2001 16:37:18 -0500
Vladimir Florinski wrote:
> Dave Uhring wrote:
>>
>> It may be that your Intel system reacts to the drivers in the 2.4.X
>> kernels
>> like my VIA MVP4 chipset. Lots of cksum errors?
>>
>
> Well, all I have is these hda errors in the system log. These cause
> massive filesystem corruption. And the Intel BX has always been well
> supported by older kernels.
>
>> Get linux-2.2.19 from ftp.kernel.org, build it and make it your default
>> kernel and I expect your problems will disappear.
>
> I doubt a 2.2 kernel would be a drop-in replacement for the 2.4.2 in the
> new RedHat. It would break initscripts and a lot of tool would probably
> stop to work (procps, util-linux, etc.) I also like certain new features
> in 2.4 such as drm modules, new video4linux and better PCI modem support.
> Well, it's sad that Linux has evolved to become an unstable OS.
>
I have RedHat-7.1 installed on a machine with VIA's Apollo MVP4 chipset.
Using any 2.4.X kernel, I get the identical errors. It is running right
now on kernel 2.2.19 without problems of any kind. And it is being used
right now by a friend of mine.
There is nothing unstable about Linux. There are drivers in Linux-2.4.X
for new hardware not supported by the 2.2.X kernels. And it is stable on
that hardware. You want bleeding edge software, use bleeding edge
hardware. You want to keep your legacy hardware, use the legacy version
software. It is that simple.
------------------------------
From: Jerry Kreps <[EMAIL PROTECTED]>
Subject: Re: open ports
Date: Tue, 15 May 2001 17:05:19 -0500
Claus Atzenbeck wrote:
> Frank Ranner wrote on Dienstag, 15. Mai 2001 03:51:
>
> Thanks a lot for your help!
>
> > required. In any
> > case you should look at blocking those ports from the Internet. Use one
> > of the Internet based
> > scanner web sites to scan you from 'out there' to see if your ports are
> > visible. If so, look
> > into one of the many tutorials on locking down your system.
>
> Do you have the URL of such a service and/or a good tutorial?
>
> Claus
/etc/services has a list of most of the services and their port number(s)
------------------------------
From: Christian Rose <[EMAIL PROTECTED]>
Subject: Re: Compile GCC 2.95.3 in RedHat 7.0 failed !!
Date: Wed, 16 May 2001 00:09:08 +0200
John Hasler wrote:
> > True. Thet doesn't change the fact that it can be (and is) a tested and
> > reliable compiler nevertheless.
>
> The gcc team is the final authority on tested and reliable versions of gcc,
> and they did not release 2.96.
Exactly. They did not release gcc 2.96, in fact they had nothing to do
with it, so how should they be the final authority on if it is tested
and reliable?
If they don't want anything to do with it and they have done no testing
on it, they are clearly the wrong people to ask about anything that has
to do with this compiler being working, reliable, or suitable for
anything.
> Shipping a compiler made available for experimental use in a production OS
> was irresponsible.
That statement assumes that it was an untouched cvs snapshot that was
put in the OS. It was not.
> No amount of "But it seems to work!" can change that.
If it works and is a reliable compiler, I don't see what is wrong with
the compiler.
> > You also know that releasing a compiler without regression testing would
> > be stupid, and you also know that a lot of the gcc people are actually
> > Red Hat people and most certainly know how that testing is made. Now go
> > figure. Not that the gcc steering committee would ever admit that it
> > passed tests, since they didn't do the testing.
>
> Are you saying that Red Hat is forking gcc?
I don't know. All I'm saying is that the gcc steering committe is
behaving like they did, and since they say that they want nothing to do
with gcc 2.96 they shouldn't comment on its merits either, IMHO.
Christian
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: Compile GCC 2.95.3 in RedHat 7.0 failed !!
Date: Tue, 15 May 2001 22:15:17 GMT
In article <[EMAIL PROTECTED]>, Thomas Corriher wrote:
>This means being honest even when it hurts a major distributor
>such as Red Hat. We all know that Red Hat acted recklessly,
>and without common decency, in the compiler situation.
I've been in similar situations: do you go with a part that's
not available/released in the hopes that it will be released in
time for your product to ship, or do you choose something
that's available now, but will be expensive and slow compared
to what will _probably_ be available when you want to ship.
You have to make a choice and take your chances. Sometimes you
guess wrong. If you go the "safe" route and your competitor
bets on the new part and wins, then his product is cheaper,
faster, and better while you scramble to catch up. If the
competition uses the existing part and you bet on the new part
and loose, then you don't have a product at all.
>In fact, Red Hat refused to correct (patch) the compiler until
>Linus himself publically attacked them.
As always, it's the reaction _after_ the mistake that hurts the
reputation -- not the mistake itself.
--
Grant Edwards grante Yow! I'm sitting on my
at SPEED QUEEN... To me,
visi.com it's ENJOYABLE... I'm
WARM... I'm VIBRATORY...
------------------------------
From: "atl.mediaone" <[EMAIL PROTECTED]>
Subject: telnet on Mandrake.
Date: Tue, 15 May 2001 22:07:16 GMT
I'm trying to setup telnet under inetd on Mandrake 7.2. I've got the port
open (nmap shows the telnet port as being open) but when I try to connect I
get the message 'Connection closed by foreign host.' I've got hosts.allow
set to ALL for telnet hosts.deny empty for telnet. (This is an internal
machine and after getting this to work in general I'll tighten up the
allowed hosts to be only the local boxes - so don't gun for me on that issue
right now.) Am using inetd to kick off telnet when needed. Any hints?
Rich
[EMAIL PROTECTED]
------------------------------
From: Ramin Sina <[EMAIL PROTECTED]>
Subject: unzipping Mac BinHex4 files in Linux
Date: Tue, 15 May 2001 22:20:31 GMT
Hi all, is there a way to unzip a a macintosh file that was compressed or
tarred with BinHex version 4?
Thanks,
Ramin Sina
------------------------------
From: [EMAIL PROTECTED] (Hartmann Schaffer)
Subject: Re: Special Characters
Date: 15 May 2001 18:44:10 -0400
In article <9drh1j$[EMAIL PROTECTED]>, Jason C. Hill wrote:
>Hopefully these will show up correctly.
>
>How would I reproduce an : � on a Solaris keyboard/OS (Or Linux). In case
>that doesn't come out correctly, that's the letter e with an ' (accent) mark
>over the top of it. It's not sufficient enough for me to type e' in order
>to reproduce it.
at least on my screen it came up correctly (slrn with a system configured for
latin-2(?)
--
hs
================================================================
"The cheapest pride is national pride. I demonstrates the lack of
characteristics and achievements you can be proud of. The worst loser
can have national pride" - Schopenhauer
------------------------------
From: "G Pollack" <[EMAIL PROTECTED]>
Subject: sftp - gftp
Date: Tue, 15 May 2001 22:52:43 GMT
I have ssh installed on a server (Mandrake 7.2; openssh-server-2.5.2p2-1.1mdk)
and I'm able to do sftp
transfers to and from this machine from a remote client, using sftp from
the command line. (The remote has Mandrake 8.0,
openssh-clients-2.5.2p2-3mdk).
But when I try to make an ssh connection using gftp, I
get the following error message (from the server, I think):
bash: sftpserv: command not found
It's correct that the server has no such file; what it does have is
/usr/lib/ssh/sftp-server. I've tried making /usr/lib/ssh/sftpserv a link
to this, and then gftp simply hangs. Any advice will be appreciated.
------------------------------
From: Eric Anderson <[EMAIL PROTECTED]>
Subject: Re: Mandrake 8.0 problems
Date: Tue, 15 May 2001 16:05:16 -0700
2.4.3 is incompatible in what way? With the BX chipset? With the IDE
controller? Is this incompatibility documented somewhere?
I just downloaded the 2.4.4 source and will try that kernel to see if the
problem has been fixed.
Eric
Dave Uhring wrote:
> Eric Anderson wrote:
>
>> Yesterday I installed Mandrake 8.0 on top of a Mandrake 7.2 system. I
>> reformatted /, but left the /home partition as it was. Both are ReiserFS
>> formatted.
>>
>> I'm now having two problems. The first is that I keep getting error
>> messages like this in the 'messages' file:
>>
>> May 15 07:29:06 catamount kernel: hda: dma_intr: status=0x51 { DriveReady
>> SeekComplete Error }
>> May 15 07:29:06 catamount kernel: hda: dma_intr: error=0x84 {
>> DriveStatusError BadCRC }
>>
>> The drive containing all of the above partitions is a 6-month-old IBM 45G
>> IDE drive. I would hope it's not going bad already!
>>
>> The second problem is that Mozilla 0.9 is now acting up. When I exit the
>> program, at least one thread gets hung in uninterruptible wait state, and
>> nothing short of a reboot will kill it.
>>
>> These two problems may be related. I'm wondering if there might be any
>> incompatibility between the filesystem on /home (formatted using the
>> ReiserFS version in Mandrake 7.2) and the ReiserFS in Mandrake 8.0.
>>
>> Any thoughts?
>>
>> Eric
>
> You are going to have to install the 2.2.19 kernel. The 2.4.3 kernel is
> incompatible with your hardware.
>
> OR, you can turn off DMA in your BIOS.
>
>
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ETA Associates, Inc.
http://www.ultracode.com/
------------------------------
From: Neil Zanella <[EMAIL PROTECTED]>
Subject: RPM: listing package sizes
Date: Tue, 15 May 2001 20:25:19 -0230
Hello,
Is it possible to use RPM to list all packages as in with rpm -qa but with
an integer next to each package specifying the total space taken up by the
package one installed?
Thanks,
Neil
------------------------------
From: Christian Rose <[EMAIL PROTECTED]>
Subject: Re: Compile GCC 2.95.3 in RedHat 7.0 failed !!
Date: Wed, 16 May 2001 01:13:38 +0200
"Peter T. Breuer" wrote:
> > Yes it is. You may disregard from what I'm saying, but most people
> > actually using gcc 2.96 will tell you the same thing. It doesn't become
> > untested and unreliable just because you say so.
>
> It doesn't become tested and reliable because YOU say so, you mean!
> The burden of proof is with people who claim that is so, not with me.
You make it easy when argumenting. You insist on other people coming
with evidence, yet you never need any evidence to say "No, it's not" on
every single claim. Even when it's shown that you are completely wrong.
Have a look at all the patches sent back to the gcc team by Red
Hat/Cygnus people. Then come say that they don't come from testing.
> And anyway, we know that the problem is that it is "incompatible",
> "nonstandard", etc. etc.
There is no incompability except for C++, and that ABI is incompatible
between every gcc release anyway.
"Nonstandard" only if you believe that every single Linux distribution
should have the same version of the compiler, which also is unrealistic
anyway.
> > So, how does it work with gcc 2.96-81? Have you actually tried it?
>
> No, once bitten, twice shy. People who can _put out_ that sort of thing
> in the first place lose whatever trust they had stocked up.
So you haven't tried to see if your case is still valid today.
> >> The difference is minor.
>
> > No, they aren't. One works and one doesn't.
>
> You think it works. As we know ...
Not only me. Most other people that use Red Hat Linux 7.1 or
Linux-Mandrake 8.0. I haven't yet heard anyone using those complain
about compiler issues, and I know a good deal of people that use those
systems. And yes, many of them like to compile stuff.
> > It was. There were some rare instances were correct code wouldn't
> > compile, those issues were fixed with the gcc errata. There is no
> > perfect compiler.
>
> There is none.
So we agree on something :-)
> Issues of compliance with one standard or the other in
> terms of what source they compile and what semantics they assign to it
> is important, but the most important thing is that the common core
> of the language is correctly transformed into functioning code. That
> is not known for this compiler.
Both Linux-Mandrake 8.0 and Red hat Linux 7.1 are quite good evidence
that the code works, wouldn't you say?
> >> > are equal. Please don't ignore such facts (although you do that all the
> >> > time anyway).
> >>
> >> Saying they are "facts" does not make them so, especially when they are
> >> not so.
>
> > They are facts. Nothing you have come with so far could even remotely be
> > called "a fact" (more of the kind "blatant lie").
>
> We are not discussing facts, but opinions and evaluations. In such a
> context facts serve as the basis of the discussion - and I agree, there
> are few of them. I relate my own evaulation, based on trials and tests.
Yet such simple facts that gcc 2.96 in its current incarnation is indeed
suitable for kernel 2.4 compiling you refuse to believe, and dismiss as
a lie. No matter how many people have done this, or how many operating
systems that have kernels compiled with this compiler. I don't see how
such behavior can bring a fruitful discussion.
> >> Did it pass gcc's own regression tests?
>
> > The gcc steering committe did not officially release it, and you know
> > that. You also know that releasing a compiler without regression testing
> > would be stupid, and you also know that a lot of the gcc people are
> > actually Red Hat people and most certainly know how that testing is
>
> No I do not.
http://gcc.gnu.org/steering.html
I count to 4 out of 14 being Red Hat people.
> My sincere impression is that RH do not test.
That is your "impression", with no backup whatsoever.
> > made. Now go figure. Not that the gcc steering committee would ever
> > admit that it passed tests, since they didn't do the testing.
>
> If they don't, then I won't.
Then don't comment at all on gcc 2.96, if you don't use it and don't
want to use it. Leave the commenting on how it works, and how reliable
it is, to those who actually use it. Sounds like a sound strategy to me.
> >> Can it YET compile gcc 2.95? Can it yet compile a kernel reliably? The answer
> >> is NO to both of those.
>
> > You're spouting out lies. Not only are you calling me a liar since I've
> > successfully compiled kernels with gcc 2.96, you're also calling
> > everybody else that has done the same a liar. More people than me in
>
> The answer is that it cannot compile kernels correctly.
Oh, please tell more. What kernel version, what gcc version (exact
output of "rpm -q gcc"), and on what distro and release would be useful
information.
> That you don't know that is one problem.
Umm, you haven't told anyone here about it, and you haven't given any
details yet. How should people know?
> Furthermore the problem is not really that it cannot, but that that it's
> not surprising. For as you know, no compiler is bugfree. Often it
> takes years to find bugs.
Yes. I was no less surprised when gcc 2.96 could be used for compiling
kernels, and producing kernels that worked. After everything that has
been said on the topic, I couldn't believe it at first, and only really
after testing it myself.
Christian
------------------------------
From: "Oscar Ricardo Silva" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.redhat
Subject: Problems creating ramdisk
Date: Tue, 15 May 2001 18:09:50 -0500
I've been asked to create an 800MB ramdisk so that a database can be copied
into it. After some initial problems, mostly with syntax, I've been able to
create a 400MB ramdisk but not 800MB. Here is what I do for the 400MB:
dd if=/dev/zero of=/dev/ram bs=1k count=400000
/sbin/mke2fs -vm0 /dev/ram 400000
mount /dev/ram /usr/ram
and this works if I issue the commands in /etc/rc.d/rc.local. Here is what
happens when I try to create the 800MB ramdisk:
[root@doris /root]# dd if=/dev/zero of=/dev/ram bs=1k count=800000
800000+0 records in
800000+0 records out
[root@doris /root]# /sbin/mke2fs -vm0 /dev/ram 800000
mke2fs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
100128 inodes, 200000 blocks
0 blocks (0.00%) reserved for the super user
First data block=0
7 block groups
32768 blocks per group, 32768 fragments per group
14304 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840
Writing inode tables: done
Writing superblocks and filesystem accounting information: done
[root@doris /root]# mount /dev/ram /ram
mount: wrong fs type, bad option, bad superblock on /dev/ram,
or too many mounted file systems
Some info on the machine I'm working on:
Operating system: Red Hat Linux 7.1
Dual processor Pentium III 933Mhz
Asus CUV4X-DLS
1GB RAM
2 20GB drives
1 60GB drive
Any information would be appreciated
------------------------------
From: Christian Rose <[EMAIL PROTECTED]>
Subject: Re: Compile GCC 2.95.3 in RedHat 7.0 failed !!
Date: Wed, 16 May 2001 01:28:46 +0200
Juergen Heinzl wrote:
> There is no such thing as a gcc-2.96 more than there's a gcc-2.98.42,
> my private version, which is stable and tested of course.
There is no offical release by the GCC Steering Committe named "gcc
2.96", that part is true.
But that doesn't change the fact that there is a compiler called gcc
2.96 (even if it isn't an official gcc release). You can call it
gcc-2.96-rh or whatever, I chose to call it gcc 2.96 which is the name
used by Red Hat and Mandrake.
> Since gcc-3.0's release seems to be planned for June the whole issue
> is going to become a moot one soon anyway, so it's about time to let
> gcc-2.96-phantom-RH kick the bucket.
>
> See here ...
> http://news.linuxprogramming.com/news_story.php3?ltsn=2001-05-15-001-06-LT
> ... for more 8-)
Thanks for the link!
I'm happy gcc 3.0 is really close and that this whole topic will be moot
soon. Nevertheless I doubt that Mandrake, Red Hat or any other
distribution will change major compiler version in new stable minor
versions of their distributions, so gcc 2.96 (as well as gcc 2.95) will
probably stay for a while even after the gcc 3.0 release.
Christian
------------------------------
From: bert buchholz <[EMAIL PROTECTED]>
Subject: Reloading a module
Date: Wed, 16 May 2001 01:44:23 +0200
Reply-To: [EMAIL PROTECTED]
Hello,
is there the possibility of reloading a kernel-module? The problem I have
is as follows:
I'm loading some SCSI-modules (scsi_mod, fdomain, sd_mod) for my
SCSI-adapter/drives when booting Linux, so that I can mount my
SCSI-harddrive. But sometimes, I want to attach another SCSI-drive to the
adapter while running. Of course, I cannot unload the module as long as the
first drive is busy, so I would have to
1. wait, until the drive (the module) stops being busy (stupid solution)
2. restart the box (even more stupid)
So I drew the conclusion, that the only sensible thing to do was to RELOAD
the module. Can one do this, if yes, then how? I mean, I can hardly be the
only one, who has this problem. Loading a module updates /proc information
(or at leastshould) and gives nice kernel-messages about the partitions on
a drive and so on, that I need to mount the drives I am attaching.
Hope you got the problem, thanks in advance
Bert
------------------------------
From: Christian Rose <[EMAIL PROTECTED]>
Subject: Re: Compile GCC 2.95.3 in RedHat 7.0 failed !!
Date: Wed, 16 May 2001 01:57:32 +0200
Thomas Corriher wrote:
> Give it up Peter. Everyone knows what happened, and everyone
> already knows of Red Hat's guilt on this issue. Christian is
> apparently chosing not to understand.
Please explain what it is I choose not to understand.
> I use Red Hat myself, but the truth is most important. Ethics
> and community are the keys things that bind Linux together.
> For the sake of these things which we value, we must always
> be truthful. This means being honest even when it hurts a
> major distributor such as Red Hat.
True. Honesty is key.
Quote from RH CTO: "Did we make some mistakes? Yes. Are we fixing them?
Yes."
(http://www.freeos.com/articles/2801/2/11/)
> We all know that Red Hat acted recklessly, and without common
> decency, in the compiler situation. In fact, Red Hat refused
> to correct (patch) the compiler until Linus himself publically
> attacked them.
Where did you get the part about refusal from?
And about the attack
(http://www.uwsg.indiana.edu/hypermail/linux/kernel/0012.1/1252.html),
the replies in that thread are even more interesting.
> That is pretty damn sad. I lost a lot of respect for them when all
> of these things happened.
I think it was very unfortunate, but I understand the reasons
(http://www.bero.org/gcc296.html).
Christian
------------------------------
** 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 by posting to comp.os.linux.misc.
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-Misc Digest
******************************
