Linux-Development-Sys Digest #687, Volume #7 Sat, 18 Mar 00 18:13:14 EST
Contents:
Re: LILO and GRUB: where do you pick disk geometry from? (repost) ("Steven J.
Hathaway")
netscape hangs with alarm clock... (Christian =?iso-8859-1?Q?J=F6nsson?=)
Re: SHM and kernel 2.3.5x (Aric Cyr)
Re: Problem compiling kernel under SMP (Andy Guibert)
mmapping ISA device memory ("Robert Karlsson")
Bootdisks, rdev, and root filesystems...aargh! (Don Werve)
Re: mmapping ISA device memory (Andi Kleen)
Rubini's device driver example (Harish K Chandraia)
Re: glibc make failure - stdio_lim.st? (Kevin Brosius)
Re: mmapping ISA device memory ("Robert Karlsson")
Re: Installing glibc (Please Help!) (Nix)
Re: Why a file system ? (Nix)
Re: kernel in C++ (Nix)
Re: kernel in C++ (Nix)
Re: kernel in C++ (Nix)
Re: kernel in C++ (Nix)
Re: underscores ???? (Nix)
Re: MP3 Players Other Than Rio (Nix)
Re: GFX Getting Pointer To Screen Mem (Nix)
Re: g77 warning: multiple common of �******� (Nix)
Re: NIC's promisc mode (Nix)
Re: ����� ������!!! (Nix)
Re: howto make floppy with ext2 sf bootable? ("Steven J. Hathaway")
Re: Buggy 2000 Setup Solutions/Programming Links (victim for experiments)
Hayes PCI Accura 56K Internal Modem - Support? ("Steven J. Hathaway")
----------------------------------------------------------------------------
Date: Fri, 17 Mar 2000 21:14:05 -0800
From: "Steven J. Hathaway" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: LILO and GRUB: where do you pick disk geometry from? (repost)
I've also used the Linux "fdisk" program using the advanced options
to explicitly set geometries for hard disks. Linux still remembers
to use these geometries, but some custom BIOSes for Compaq
systems are severely broke and assume manufacturer geometry
of drives even though new geometry has been recorded onto the
master boot record.
- Steven J. Hathaway
------------------------------
From: Christian =?iso-8859-1?Q?J=F6nsson?= <[EMAIL PROTECTED]>
Subject: netscape hangs with alarm clock...
Date: Sat, 18 Mar 2000 06:35:00 GMT
quite often I get netscape hanging, consuming all the cpu resourse
if I start strace netscape it looks like this when it hangs:
-- SIGALRM (Alarm clock) ---
gettimeofday({953360799, 711161}, NULL) = 0
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) ---
gettimeofday({953360799, 723159}, NULL) = 0
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) ---
gettimeofday({953360799, 821187}, NULL) = 0
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) ---
gettimeofday({953360799, 830064}, NULL) = 0
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) ---
gettimeofday({953360799, 933297}, NULL) = 0
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) ---
gettimeofday({953360799, 974417}, NULL) = 0
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) ---
gettimeofday({953360800, 41147}, NULL) = 0
sigreturn() = ? (mask now [])
+++ killed by SIGKILL +++
that last SIGKILL was issued by me
This is on a Red Hat Linux 6.1/Intel up to date system
running kernel 2.3.51 but I think it did the same under
2.2.12-20...
Any ides?
btw, it's it's netscape 4.72-6 and glibc-2.1.2-11
------------------------------
From: [EMAIL PROTECTED] (Aric Cyr)
Subject: Re: SHM and kernel 2.3.5x
Date: 18 Mar 2000 07:31:45 GMT
In article <[EMAIL PROTECTED]>,
Hugh Caley <[EMAIL PROTECTED]> wrote:
>I haven't been able to use development kernel versions 2.3.50 and
>higher. They boot, but when Apache starts I get an error that the shm
>filesystem needs to be mounted. The information in
>Documentation/Changes is very sketchy; it simply says that the shm
>filesystem needs to be mounted somewhere under /proc and that the
>default is /var/shm. I have no idea what they are talking about ... I
>wouldn't bother with the dev kernel anyway except that I am trying to
>use my Rio500 with Linux, which requires USB and 2.3.x kernels.
>
>Hugh
>[EMAIL PROTECTED]
>
It's really quite easy, after a little trial and error and code reading you
can figure it out. What you want to do is "mount none /var/shm -t shm" and
make sure that the directory "/var/shm" exists. This is similar to mounting
the /dev/pts and /proc filesystems, except this one is for shared memory.
You can use a different directoy, other that "/var/shm" but then you must
update the /proc/sys/kernel/shmpath variable to be equal to the path of your
mount point. I'd just stick with the default of /var/shm, then you don't
have to messa round with /proc settings.
--
Aric Cyr - [EMAIL PROTECTED]
4B Computer Science
University of Waterloo
PGP Public Key at http://www.cryogen.com/acyr/pubkey.html
------------------------------
From: Andy Guibert <[EMAIL PROTECTED]>
Subject: Re: Problem compiling kernel under SMP
Date: Sat, 18 Mar 2000 05:33:38 -0500
Chih-Yang Chiu wrote:
> I am trying to recompile my kernel/upgrade my kernel under Mandrake Linux
> 2.2.13. I did everything from going through the make config, make dep and
> make clean, and everything seems fine. But when I try to compile the
> newly configured kernel using make bzImage, make zImage, I got the
> following error.
Try the following 'make' statements:
1. make mrproper
2. make config (or "menuconfig", "xconfig" if you want)
3. make dep
4. make modules (if any drivers are to be made as loadable modules)
5. make bzImage (don't even bother with zImage, I've yet to see a non
bare-bones SMP kernel that's small enough to use this option)
6. make modules-install
7. make install
Before you do all this, make sure you keep a copy of your current kernel and
it's modules... It's always good to be on the safe side. ;)
Hope this helps,
Andy
=================================================
Remove the "nojunk" from my addy when responding.
"May the source be with you, Luke."
------------------------------
From: "Robert Karlsson" <[EMAIL PROTECTED]>
Subject: mmapping ISA device memory
Date: Sat, 18 Mar 2000 12:25:18 GMT
Hi,
I'm developing a charachter device driver under the 2.2 kernel.
What is the proper way to mmap an ISA memory range with the extent of 8ffeh
starting at d0000h, into user space.
I've been trying the following in my implemetation of mmap(excluding the
error checking).
static int xyz_mmap(struct inode *inode,
struct file * file,
struct vm_area_struct *vma)
{
remap_page_range(vma->vm_start,
0xd0000, 0x8ffe,
vma->vm_page_prot);
return 0;
}
1) Can I assume the above code is good or is something wrong with the
address or range? i.e. are they mmap:able at all?
2) I have loaded the driver with insmod and a printout in my init_module
function says I'm getting the major = 127.
Then, when opening the driver in my user code like this:
int fd = open("/dev/xyz.o",O_RDWR);
I'm getting fd = 5 which seems fine.
However, after doing mknod xyz c 127 0
and opening the device like this:
int fd = open("/dev/xyz",O_RDWR); /* no .o here*/
fd = -1 and errno = 19
Is the second approach wrong (or the first..)
Am I getting it all wrong?
Ok, assuming the first one is correct
I then doing mmap like this:
u_short* xyz_mem = (u_short*)mmap(NULL,
ISR_MAP_SIZE,
PROT_READ|PROT_WRITE,
MAP_SHARED,
fd,
0);
wich gives me xyz_mem = 0x400b400b
Is this a bogus value?
Please help me out here, I'm really confused.
/ Robert
------------------------------
From: Don Werve <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.misc
Subject: Bootdisks, rdev, and root filesystems...aargh!
Date: Sat, 18 Mar 2000 15:41:43 GMT
Ok. After reading the Bootdisk-HOWTO and rdev manpages, I decided to
start on the path to true geekdom and make my own Linux bootdisk, so
that I could manually build a Linux system from the ground up on one of
my many spare computers. I've been using Linux for about two years now,
but what better way to learn about the guts of the OS than to manually
build a full system?
Anyways, I made a root filesystem that compressed takes up 900K, and
have a 460K kernel. I used dd to copy the kernel image to the disk, and
then used `rdev /dev/fd0 /dev/fd0 ; rdev -R /dev/fd0 0` to set the disk
as it's own root device. How do you tell the kernel where to look for a
compressed root filesystem, so that it can be uncompressed into a
ramdisk and booted...?
Ideas? Sorry about the cross-post, but it seemed applicable to each
newsgroup posted to. Please reply via E-Mail, to hari AT iveleague.org
Thanks in advance! :)
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: mmapping ISA device memory
Date: 18 Mar 2000 17:57:22 +0100
"Robert Karlsson" <[EMAIL PROTECTED]> writes:
> 1) Can I assume the above code is good or is something wrong with the
> address or range? i.e. are they mmap:able at all?
They aren't. You need to use ioremap() first to map the IO space
(actually they are already mapped on x86 but not at the virtual addresses
you expect them to -- using ioremap is clean and portable)
>
>
> 2) I have loaded the driver with insmod and a printout in my init_module
> function says I'm getting the major = 127.
>
> Then, when opening the driver in my user code like this:
>
> int fd = open("/dev/xyz.o",O_RDWR);
>
> I'm getting fd = 5 which seems fine.
You probably opened some random file.
>
> However, after doing mknod xyz c 127 0
>
> and opening the device like this:
>
> int fd = open("/dev/xyz",O_RDWR); /* no .o here*/
>
> fd = -1 and errno = 19
>
> Is the second approach wrong (or the first..)
> Am I getting it all wrong?
The access is correct, your driver probably didn't register
successfully.
>
> Ok, assuming the first one is correct
No.
>
> I then doing mmap like this:
>
> u_short* xyz_mem = (u_short*)mmap(NULL,
> ISR_MAP_SIZE,
> PROT_READ|PROT_WRITE,
> MAP_SHARED,
> fd,
> 0);
>
> wich gives me xyz_mem = 0x400b400b
> Is this a bogus value?
No. This is the virtual address in your user space that the memory
was mapped to (mmap just chooses a free area)
-Andi
------------------------------
From: [EMAIL PROTECTED] (Harish K Chandraia)
Subject: Rubini's device driver example
Date: 18 Mar 2000 19:28:37 GMT
Hi,
I am trying to run the first example in Rubini's Linux Device
Drivers and when I run the insmod command, I don't see "Hello World"
from the init_module() being printed and I don't see anything from
the cleanup_module() being printed when I run the rmmod command
either. I am running mandrake linux on my PC. Are the printk's from
the init_module() and cleanup_module() supposed to appear only on
the console??. On an xterm if I do
cat /dev/console, all the other xterms freeze and I can't type anything
in any other xterm. Can somebody please tell me where I am going
wrong? Thank you all for u'r time in advance.
Regards,
Harish Chandraiah.
------------------------------
From: Kevin Brosius <[EMAIL PROTECTED]>
Subject: Re: glibc make failure - stdio_lim.st?
Date: Sat, 18 Mar 2000 15:02:55 -0500
Kevin Brosius wrote:
>
> I've looked for this in the FAQ, but haven't seen any report of this
> build problem. Is anyone aware of what might cause this? This is Linux
> 2.2.13 with libc5 presently installed. I've done a little looking
> through the Makefiles, but don't understand them, or the error message
> enough to find the problem. Any suggestions appreciated! Thanks in
> advance.
>
> make -C csu subdir_lib
> make[2]: Entering directory `/p5/glibc-2.1.3/csu'
> make[2]: *** No rule to make target `../stdio-common/stdio_lim.h.in',
> needed by `/p5/glibc-build/bits/stdio_lim.st'. Stop.
> make[2]: Leaving directory `/p5/glibc-2.1.3/csu'
> make[1]: *** [csu/subdir_lib] Error 2
> make[1]: Leaving directory `/p5/glibc-2.1.3'
> make: *** [all] Error 2
>
> --
> Kevin
This turned out to be a corrupt tar file, even though the archive was
the correct size, tar zvxf exited with:
tar: Skipping to next file header...
and did not extract all files.
--
Kevin
------------------------------
From: "Robert Karlsson" <[EMAIL PROTECTED]>
Subject: Re: mmapping ISA device memory
Date: Sat, 18 Mar 2000 20:35:15 GMT
"Andi Kleen" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> They aren't. You need to use ioremap() first to map the IO space
> (actually they are already mapped on x86 but not at the virtual addresses
> you expect them to -- using ioremap is clean and portable)
Like this?
static int xyz_mmap(struct inode *inode,
struct file * file,
struct vm_area_struct *vma)
{
u_short* io_mem = ioremap(0xd0000,0x8ffe);
remap_page_range(vma->vm_start,
io_mem, 0x8ffe,
vma->vm_page_prot);
return 0;
}
Will this work or have I misunderstud you.
Sorry, I know I could try this myself, thing is that I don't have access to
the target machine until Monday (I love Mondays..).
/Robert
P.S.
Sorry Andi for mailing you this, It was a mistake
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: Installing glibc (Please Help!)
Date: 18 Mar 2000 15:27:56 +0000
Andreas Jaeger <[EMAIL PROTECTED]> writes:
> >>>>> Nix writes:
>
> Nix> I've never seen such a density of (rightful) `RTFFAQ' responses to
> Nix> questions about any other package.
>
> Nix> (Maybe it's just that installing glibc is somewhat rocket science to Joe
> Nix> Average, slightly more delicate than your average package, and comes
> Nix> with a very good FAQ... but no, I like the conspiracy theory more ;) )
> Any idea what we can do about this? Shall I stop writing FAQ entries
> since nobody (except the two of us;-) reads them?
<selfish-bastard>
I wouldn't; I find them useful.
</selfish-bastard>
Also, at least they mean that we just need to say `RTFFAQ' instead of
explaining things over and over again.
What might be good is something at the end of configure.in telling
people something like `If you have not read the FAQ and INSTALL
document, read it now. Installing glibc without reading this is highly
unwise.'
So anyone who runs configure gets it to nag them ;)
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: Why a file system ?
Date: 18 Mar 2000 15:33:11 +0000
[EMAIL PROTECTED] (Alexander Viro) writes:
> In article <[EMAIL PROTECTED]>,
> David Fox <d s f o x @ c o g s c i . u c s d . e d u> wrote:
> >as well as all other permutations of pathnames. Now what is the
> >significance of having each ordering of these attributes refer to a
> >distinct set of files? Well, none that I can think of, except to
> >cause confusion.
>
> IOW, you can't think...
But /usr/bin and /bin/usr *should* be the same directory, they should,
they should!
And having /home/viro/foo and /home/foo/viro be the same file isn't a
security hole, oh no, of course not!
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: kernel in C++
Date: 18 Mar 2000 15:45:22 +0000
[EMAIL PROTECTED] (Alexander Viro) writes:
> Leave filesystems alone. Contrary to luserish faith, OOP can be done in
> C just fine and it's _much_ better to have clean langauge with well-understood
> semantics for doing that stuff. C++ is too messy for that. It tries too hard
> to look like C and that harms it.
OK, well, then I have a cool acronym:
SCUFS, the Scheme Unified File System.
Now I just need to write the code ;)
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: kernel in C++
Date: 18 Mar 2000 15:49:10 +0000
[EMAIL PROTECTED] (Christopher Browne) writes:
> X gets bashed for being big; the vast majority of the memory that X
> consumes represents caches that contain graphical objects, which
> diminishes the amount of recalcs that have to be done when you move
> windows around.
More to the point, saveunders and friends reduce the amount of needed
network traffic to process (eg) expose events substantially.
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: kernel in C++
Date: 18 Mar 2000 16:08:01 +0000
[EMAIL PROTECTED] (Christopher Browne) writes:
> Linus Torvalds and Alan Cox and Alexander Viro and Stephen Tweedie and
> ... will essentially ignore you until you have a working
> implementation.
Not quite true, I think. The l-k list members are normally happy to help
people get a better picture of how the kernel currently works, even if
(perhaps `especially if') they're going to use that to go off and write
a better kernel.
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: kernel in C++
Date: 18 Mar 2000 16:04:38 +0000
[EMAIL PROTECTED] (Alexander Viro) writes:
> Please, take any decent textbook
> and learn what exceptions are, how they are implemented and when they are
> used. Additional points for figuring out why they are insufficiently flexible
> for typical situations in the kernel and why they are incompatible with
> currently used mechanisms (which can be implemented both in C and in C++).
> C++ is not a problem here. _Exceptions_ are.
Of course, if the kernel were redesigned from the ground up to work the
way C++ likes, exceptions would probably be great.
But there's not much chance of that. If you want a total redesign, start
a new project.
> >C++ gives you a way to atleast manage larger chunks.
There's this complicated thing called `modularity'. And, for that matter
`OO in C'. Yes, it can be done, and the VFS in Linux already does
it. It's at least as neat as C++ classes would be, because it does
exactly what its application domain needs it to do, and no more. So you
can't easily do diamond-shaped inheritance graphs and similar monsters
in it; but as that isn't needed here, this is not a loss.
> >If you were to do the same things with functions with 12 arguments and
> >struct's with 20-some elements including pointer attached with mallock-ed
> >mem-blocks. How would you handle those ?
Duh? Have you *looked* at the kernel code?
> >(I know the answer by putting another global set of flags and some 100
> >lines)
>
> HUH? If you have a function with 12 arguments (in _any_ language) you can't
> write. Period.
The presumed humans at my workplace would say `oh, well, so that's too
many, make them all global variables then.' But then these are people
who think nothing of dozens of global declarations like
char str[500][5000];
char str_b[500][5000];
char str_c[500][5000];
char str_d[500][5000];
char str_e[500][5000];
because what they really mean is `unlimitedly many `foo's of unlimited
length but we don't like dynamic allocation'.
Oh, and they really do use names like that, too.
> Nothing nearly that horrible in the kernel. Ditto for global
Thankfully, or I think I'd lose all faith in humanity. ;)
> flags. Please, take time to read the source. Otherwise I'm afraid that your,
> erm, opinion on that subject is worth... well, not too much. Come on, you
How disgustingly diplomatic of you. He's talking garbage about the
source having not actually taken the time to read it... I'd have
expected your flamethrower to be set on at least `broil', if not
`plasma burst'.
> >SIR
Oh dear oh dear oh dear. He's getting uppity.
> >that is what C++ is all about. C could do it too, but C++ manages it better.
Why?
> >The approach you are indirectly suggesting me would need cryptic system
> >level code
> >(full of pointers & reference problems).
Pointers are `cryptic'?
C++ has pointers, too, you know. References are just unreseatable
pointers, too, so they are used just as widely in C++ as they are in C.
(It is true that a few dozen lines of C++ template hackery can implement
reference counted pointers, which is rather harder to do in C... but
does the kernel need that? No, it seems that it does not. There's
*already* a garbage-collector in there, limited in scope though it is...)
> WHAT IT IS ABOUT? About the requirements regarding ability to block in the
> given context? About the difference between upper half and bottom half?
> About the fact that allocating memory on IO path not from the constant pool
> is a bad idea because resulting swapping may trivially lead to dealocks?
> Complexity is not in syntax. And no, compiler doesn't have enough information
> to choose the correct behaviour - C++ was never intended to be used in that
> kind of tasks.
Is what you're saying `C++ is great for abstracting syntactic
complexity, but our problem is with semantic complexity, and fine
control is needed to provide that, and C++ doesn't easily give us that'?
> >C existed before '73 my dear. And UNIX was born in Bell labs in 1969.
I think we all know that. The date can't be pinned down that firmly,
though, some ideas (eg regexps and pipes) were around in '67, some
didn't fall into place until '73-'74. And the networking API wasn't done
until the '80s, IIRC.
> Wow. How lovely... To start with, UNIX was rewritten into C in v4. Before that
> kernel was in PDP-11 assmebler (and before that - in PDP-7 assembler). FYI,
Was it ever written in BC, I wonder, or did BC transmogrify into C
before the kernel was rewritten in it? I can't remember the appropriate
part of dmr's C history paper... I shall have to read it again.
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: underscores ????
Date: 18 Mar 2000 16:33:02 +0000
Alan Donovan <[EMAIL PROTECTED]> writes:
> The funny parentheses are another gcc
> extension: scopes within macros.
*That* is not a GCC extension.
Statement expressions (the ({ }) brackets) *are* a GCC extension, and a
damned nice one, if you don't care about their total unportability. Very
LISPish.
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: MP3 Players Other Than Rio
Date: 18 Mar 2000 16:25:46 +0000
[EMAIL PROTECTED] (Christopher Browne) writes:
> product that appears somewhat Europe-oriented. (What's 230 Euro in
> USD?)
Last I heard, approximately $200. ;)
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: GFX Getting Pointer To Screen Mem
Date: 18 Mar 2000 16:38:11 +0000
Nate Eldredge <[EMAIL PROTECTED]> writes:
> Under X: AFAIK, you basically can't. Here the video memory is solely
> the domain of the X server. However, I believe the MIT-SHM extension
> will give you something that looks like a framebuffer that you can
> draw to.
Yes, and there is also the XFree86-DGA extension on XFree86 X servers,
that really *does* give you video memory. Unfortunately the client needs
to be able to write to /dev/mem to use it, which makes it of less use
than it would otherwise be.
btw, why does nobody use PEX?
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Crossposted-To: comp.lang.fortran
Subject: Re: g77 warning: multiple common of �******�
Date: 18 Mar 2000 16:40:14 +0000
Toon Moene <[EMAIL PROTECTED]> writes:
> You cannot override the -warn-commons option passed to the linker (at
> least I cannot think of a way to do so).
It's not in the specs file? They actually hardwire it into /usr/bin/g77,
or something?
Yuck.
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: NIC's promisc mode
Date: 18 Mar 2000 16:42:34 +0000
[EMAIL PROTECTED] (WiseGuy) writes:
> But then, what legitimate uses are there for promisc mode anyway?
Debugging network protocols?
Finding problems with network cabling?
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
From: Nix <$}xinix{[email protected]>
Subject: Re: ����� ������!!!
Date: 18 Mar 2000 16:45:13 +0000
[EMAIL PROTECTED] (Alexander Viro) writes:
> In article <8auga9$bso$[EMAIL PROTECTED]>,
> Vladimir Vizgalin <[EMAIL PROTECTED]> wrote:
> > ������ �����!!!!
>
> ����������� �� ����������. � ������� ����� � ������� - � ���� '!' ��������.
Ah. I assume this is intelligible if you happen to have a font with the
right encoding, then?
(Nice of both of you to include Content-Transfer-Encoding: headers,
btw. ;) )
--
`> KNOWLEDGE AND SKILLS
You must have some, but I don't see any evidence of it.'
--- Craig Hardie flames a luser recruitment consultant
advertising `Microsoft based solutions' on uk.comp.os.linux
------------------------------
Date: Sat, 18 Mar 2000 14:58:55 -0800
From: "Steven J. Hathaway" <[EMAIL PROTECTED]>
Subject: Re: howto make floppy with ext2 sf bootable?
Bootable Floppies: I have not tried the ext2 filesystem on /dev/fd?h1440
drives.
I usually install a minix system (mkfs.minix) on a floppy /dev/fd?h1440 drive.
Then mount the floppy and copy the kernel and install the necessary lilo
stuff.
Lilo allows some parameter passing capabilities to the kernel.
If you have a kernel that does not need parameters, you can copy it to the
floppy without a filesystem. The bootblock of the floppy will then be the
first record of the kernel that will load the rest of the kernel.
Try the "minix" file system.
Sincerely,
Steven J. Hathaway
<[EMAIL PROTECTED]>
Dimitry Katkov wrote:
> Hi all.
> I am running Linux from floppy that has ext2 filesystem on it. Currently I
> am booting kernel from another floppy.
> How can I boot linux from floppy that contains / filesystem?
>
> Thanks,
> Dimitry.
------------------------------
From: victim for experiments <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.powerpc,comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Buggy 2000 Setup Solutions/Programming Links
Date: Sat, 18 Mar 2000 14:34:05 -0800
Reply-To: [EMAIL PROTECTED]
Hi,
I don't want to discourage
the Gnome crowd, but I gave up
on Gnome .
I'm trying KDE, and so-far
it might be better .
It's a little different,
but not all that much, and
"Running Linux" recommends it .
Ken .
Michael Schmitz wrote:
>
> "[EMAIL PROTECTED]" wrote:
> >
> > - I am wondering if someone can give me some instructions on disabling the
> > GNOME graphical login, as it is causing strange problems on my machine eg.
> > flashing between the terminal and X continuously, and I have to reboot,
> > corrupting my system in the process. At least on the x86 RedHat
> > distribution, you can select whether you want a graphical login or not.
> > I presume this is in some startup script, can someone help?
>
> Set the following in /etc/inittab:
> id:3:initdefault:
>
> > - Will there ever be a way to let Linux *fully* take over the mac hardware,
> > without any hfs partitions and MacOS software (BootX extensions and such)?
>
> Use yaboot.
>
> Michael
------------------------------
Date: Sat, 18 Mar 2000 15:04:27 -0800
From: "Steven J. Hathaway" <[EMAIL PROTECTED]>
Subject: Hayes PCI Accura 56K Internal Modem - Support?
Where can I find, if any, Linux support for the Hayes Accura 56K
V.90/K56flex PCI Internal Fax Modem?
It uses the Lucent DSP circuitry.
The binary driver image found on "linuxmodem.com" does not work in my
system.
Sincerely,
Steven J. Hathaway
<[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
******************************
