Re: [parisc-linux] NIC & "searching for devices"

[Joel Soete]

Hello Max,
Max Grabert wrote:
On Tue, 22 Feb 2005 10:34:46 +0100, Joel Soete <[EMAIL PROTECTED]> wrote:
[]

Yes but I encountered weird difficulties (even on a i386 box) 
to configure a brother model 3C905B as needed:
see this old thread 

kargellan:/home/xam# lspci
:00:0d.0 Ethernet controller: 3Com Corporation 3c905C TX/TX-M
[Tornado] (rev 78)
:00:0e.0 Ethernet controller: 3Com Corporation 3c905B 100BaseTX
[Cyclone] (rev 30)
mine was far older as far I could judge it was only a (rev 24)?
:00:0f.0 Ethernet controller: 3Com Corporation 3c905C-TX/TX-M
[Tornado] (rev 78)
[...]
As you can see this is a ix86 machine with three 3com NICs
in it; It is running stable for almost 2 years now.
Right but atm I was still running 2.4 (I only start 2.6 later in late May 2004).
Event thought there was already some patch available for 2.4, I was a bit lazy 
and prefer to replace nic by another :-(
(my bad, I don't remember where did I save it to make new test, grrr)
Right now it is
running 2.6.11-rc4. BTW, all three network cards share one
interrupt (IRQ 11), which is also used by the USB port.
And there is also my ix86 laptop that is using a 'vortex' network
card, also running fine for almost 3 years now with many 
different linux kernel.

I'd rather say either your network card is broken/of poor quality
mmm I would like to find back (just for the fun)
or there are some compatibility issues with your ix86  motherboard.
or should just required a fw upgrade (if possible) or use more recent driver of 
2.6?
Anyway, this is going way off-topic ...
Obviously, but the most anoying is that I lake of time right now (in prod we have to resurect a proj frozen during last three years 
and so required a lot of work to update: middle ware first, then os update and finlay hw and its corresponding os (hpux-11.11 in 
fact for new rp3xxx to replace oldtimer as D and K class :-) ).

In my experience 3Com often produced okay-nice cards, but
sometimes they managed to release utter crap as well.
mmm I didn't suspect hw but may be fw or driver?
IMHO there are better alternatives available (Intel/DEC), 
where I don't have to take any risks.

Cool advise ;-)
Thanks,
Joel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

RE: FW: 2.4 -> 2.6

[Harry Cochran]

Dear Grant, Kyle and Thibaut:

Thank you all very much! I now have a running 2.6.8-2-32-smp on my 
J6000. I
repartitioned sda to have the root right next to the f0 partition (and the
swap at the end) per Grant, modified /etc/mkinitrd/modules per Kyle and
added Kyle's patch to mkinitrd just for luck, rebooted and everything just
worked. sda did not become sdb. I will write this up to point out the
pitfalls newbies like me can hit, but I'm afraid I'm out of time to figure
out if, for instance, it works without Kyle's mods.

It's great to be back among the living Debian hppa users! This J6000 is
going into production in China to run Web and mail services for my company
there.

Cheers,

Harry

-Original Message-
From: Grant Grundler [mailto:[EMAIL PROTECTED]
Sent: Saturday, February 26, 2005 1:30 AM
To: Harry Cochran
Cc: Thibaut VARENE; debian-hppa@lists.debian.org
Subject: Re: 2.4 -> 2.6


On Fri, Feb 25, 2005 at 02:27:52PM -0500, Harry Cochran wrote:
> HARD Booted.
> palo ipl 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:05 MDT 2004
>
> Partition Start(MB) End(MB) Id Type
> 1   1  31   f0 Palo
> 2  321008   82 swap
> 31009   17366   83 ext2

This is an invalid partition scheme. It's just luck that it
worked with 2.4 kernel. I'm assuming partition 3 is the root partition.
Re-read http://en.tldp.org/HOWTO/PA-RISC-Linux-Boot-HOWTO/index.html.
In particular "4.3.3. Making a bootable partition".

grant

-Original Message-
From: Kyle McMartin [mailto:[EMAIL PROTECTED]
Sent: Friday, February 25, 2005 11:27 PM
To: Harry Cochran
Subject: Re: FW: 2.4 -> 2.6


On Fri, Feb 25, 2005 at 10:20:28PM -0500, Harry Cochran wrote:
> Well, I tried 2.6.8-2-32-smp "out of the box" and got nowhere again. When
I
> booted with root=/dev/sda3 it didn't find it, so I changed to
root=/dev/sdb3
> and it still didn't find it. Could I please hear from someone who has
> upgraded from 2.4 to 2.6 to let me know what they did STEP BY STEP.
>

The problem is 2.4 sucks, and with 2.6 we moved to doing things
exactly like the other architectures Debian supports.

One last thing to try that I can think of, is to add "sym53c8xx" to
/etc/mkinitrd/modules.

This is how it looks on my C3000:

% cat /etc/mkinitrd/modules
# /etc/mkinitrd/modules: Kernel modules to load for initrd.
#
# This file should contain the names of kernel modules and their arguments
# (if any) that are needed to mount the root file system, one per line.
# Comments begin with a `#', and everything on the line after them are
ignored.
#
# You must run mkinitrd(8) to effect this change.
#
# Examples:
#
#  ext2
#  wd io=0x300

tulip
sym53c8xx

Hope this helps,
Regards,
Kyle M.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: 2.4 -> 2.6

[Thibaut VARENE]

On Sat, 26 Feb 2005 12:17:26 -0500
"Harry Cochran" <[EMAIL PROTECTED]> wrote:

> Dear Grant, Kyle and Thibaut:
> 
>   Thank you all very much! I now have a running 2.6.8-2-32-smp on my
>   J6000. I
> repartitioned sda to have the root right next to the f0 partition (and
> the swap at the end) per Grant, modified /etc/mkinitrd/modules per Kyle
> and added Kyle's patch to mkinitrd just for luck, rebooted and
> everything just worked. sda did not become sdb. I will write this up to
> point out the pitfalls newbies like me can hit, but I'm afraid I'm out
> of time to figure out if, for instance, it works without Kyle's mods.

If you wanna send patch for the PA-RISC/Linux Boot HOWTO, I'd certainly
look at them :)

Glad you got it to work


Thibaut VARENE
The PA/Linux ESIEE Team
http://www.pateam.org/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Harry Cochran]

Hi,

I don't know if this has any value or not, but I am happy to contribute 
it
to PA-RISC/Linux Boot HOWTO with approval of those explicitly copied on this
E-mail whose contributions I copied.

Cheers,

Harry

-Original Message-
From: Thibaut VARENE [mailto:[EMAIL PROTECTED]
Sent: Saturday, February 26, 2005 1:32 PM
To: Harry Cochran
Cc: Grant Grundler; Kyle McMartin; debian-hppa@lists.debian.org
Subject: Re: 2.4 -> 2.6


On Sat, 26 Feb 2005 12:17:26 -0500
"Harry Cochran" <[EMAIL PROTECTED]> wrote:

> Dear Grant, Kyle and Thibaut:
>
>   Thank you all very much! I now have a running 2.6.8-2-32-smp on my
>   J6000. I
> repartitioned sda to have the root right next to the f0 partition (and
> the swap at the end) per Grant, modified /etc/mkinitrd/modules per Kyle
> and added Kyle's patch to mkinitrd just for luck, rebooted and
> everything just worked. sda did not become sdb. I will write this up to
> point out the pitfalls newbies like me can hit, but I'm afraid I'm out
> of time to figure out if, for instance, it works without Kyle's mods.

If you wanna send patch for the PA-RISC/Linux Boot HOWTO, I'd certainly
look at them :)

Glad you got it to work


Thibaut VARENE
The PA/Linux ESIEE Team
http://www.pateam.org/


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
8. Upgrading from a Debian hppa 2.4 Kernel to a 2.6 kernel

8.1 Check your 2.4 partition scheme
8.2 First Steps
8.3 Edit /etc/palo.conf
8.4 Before you reboot
8.5 Conclusion

8.1 Check your 2.4 partition scheme

It may come as a surprise, but it is possible (I know, I did it) to have a 
partition scheme that by some fluke "works" under 2.4 but is invalid and will 
not work under 2.6. For example:

> palo ipl 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:05 MDT 2004
> 
> Partition Start(MB) End(MB) Id Type
> 1   1  31   f0 Palo
> 2  321008   82 swap
> 31009   17366   83 ext2  root

The problem here is that the large swap partition pushes /boot under the root 
out past the 2GB limit for booting. There is no ready cure for this. You will 
have to save your data, re-partition your hard drive and then reload your 
applications and data. The example in 4.3:

Disk /dev/sda: 133 heads, 62 sectors, 1017 cylinders
Units = cylinders of 8246 * 512 bytes

   Device BootStart   EndBlocks   Id  System
/dev/sda1   * 1 4 16461   f0  Linux/PA-RISC boot
/dev/sda2 534123690   82  Linux swap
/dev/sda3   *35   277   1001889   83  Linux

is fine, because the small swap partition doesn't push root out too far. It's a 
better idea, however, to put the swap partition at the end of the disk. By the 
way, don't forget to make both the f0 partition and the root partition bootable.

8.2 First steps

I'm the conservative type, so I don't load from "testing". At the time of this 
writing, the 2.6 kernels in "unstable" are 2.6.8-2-32, 2.6.8-2-32-smp, 
2.6.8-2-64 and 2.6.8-2-64-smp. Even though the PA-Risc cpu's in my machine are 
64 bit, hppa Debian can't take advantage of the 64 bit instruction length, so I 
just stick with the 32 bit versions. So, now it is to begin to install a 2.6 
kernel.

Step 1 For example, apt-get install kernel-image-2.6.7-2-32-smp. When you get 
near the end of the install, you will get messages telling you that this is an 
initrd kernel and asking you if you want to abort. At this point say, "Yes". 
Now it's time for:

Step 2  you need to add "sym53c8xx" to /etc/mkinitrd/modules. "modules" is 
created from /etc/mkinitrd/modules-new? Help, I don't remember the name the 
aborted load leaves behind.

This is how it looks after editing:

% cat /etc/mkinitrd/modules-new??
# /etc/mkinitrd/modules: Kernel modules to load for initrd.
#
# This file should contain the names of kernel modules and their arguments
# (if any) that are needed to mount the root file system, one per line.
# Comments begin with a `#', and everything on the line after them are ignored.
#
# You must run mkinitrd(8) to effect this change.
#
# Examples:
#
#  ext2
#  wd io=0x300

tulip
sym53c8xx

Okay. Now we are ready for:
Step 3 apt-get install kernel-image-2.6.7-2-32-smp again, but this time when 
asked if you want to abort, say, "No".

8.3 Edit /etc/palo.conf

Edit /etc/palo.conf and insert on the --commandline initrd=x/boot/initrd.img 
where x is the number of the disk partition where /boot lives. Here's an 
example of a palo.conf that works:

--commandline=3/boot/vmlinux initrd=3/boot/initrd.img HOME=/ root=/dev/sda3
--recoverykernel=/boot/vmlinux.old
--init-partitioned=/dev/sda


Then run palo!

8.4 Before you reboot

On some machines installing the 2.6 kernel flips the names of your hard drives 
(if you have more than one). For example, sda might become sdb with sbd 
becoming sda. If this happens, you will get an error on reboot that you can fix 
by going into ipl and changing, for example roo

Re: FW: 2.4 -> 2.6

[Stuart Brady]

On Sat, Feb 26, 2005 at 12:17:26PM -0500, Harry Cochran wrote:
> Thank you all very much! I now have a running 2.6.8-2-32-smp on my
> J6000. I repartitioned sda to have the root right next to the f0
> partition (and the swap at the end) per Grant,

No, that isn't what Grant said...

The Boot HOWTO explains that the kernel image must be located within the
first 2GB of the disk.  The only sensible way to ensure this is to have
a dedicated /boot partition, which is contained entirely within the
first 2GB.  This needn't be large -- a 64MB partition should be ample.

In your case, it doesn't matter whether this is before or after the 1GB
swap partition, but it must come before the large root (/) partition.

AFAICS, although your new partition scheme may appear to be working, it
may cease to work whenever the kernel is upgraded.  (By swapping the
partitions you are effectively doubling the amount of disk space from
the first 2GB which is used for the root partition.)

Hope that helps,
-- 
Stuart Brady


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Grant Grundler]

On Sat, Feb 26, 2005 at 04:52:58PM -0500, Harry Cochran wrote:
> Hi,
> 
>   I don't know if this has any value or not, but I am happy to
> contribute it to PA-RISC/Linux Boot HOWTO with approval of those
> explicitly copied on this E-mail whose contributions I copied.

You have my permission to contribute everything you learned. :^)
Though I don't see why you explicitly need my permission.
I'm not the maintainer.


Some comments below.

...
> 8. Upgrading from a Debian hppa 2.4 Kernel to a 2.6 kernel
> 
> 8.1 Check your 2.4 partition scheme
> 8.2 First Steps
> 8.3 Edit /etc/palo.conf
> 8.4 Before you reboot
> 8.5 Conclusion
> 
> 8.1 Check your 2.4 partition scheme
> 
> It may come as a surprise, but it is possible (I know, I did it) to have a 
> partition scheme that by some fluke "works" under 2.4 but is invalid and will 
> not work under 2.6. For example:
> 
> > palo ipl 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:05 MDT 2004
> > 
> > Partition Start(MB) End(MB) Id Type
> > 1   1  31   f0 Palo
> > 2  321008   82 swap
> > 31009   17366   83 ext2  root
> 
> The problem here is that the large swap partition pushes /boot under the root 
> out past the 2GB limit for booting. There is no ready cure for this. You will 
> have to save your data, re-partition your hard drive and then reload your 
> applications and data. The example in 4.3:
> 
> Disk /dev/sda: 133 heads, 62 sectors, 1017 cylinders
> Units = cylinders of 8246 * 512 bytes
> 
>Device BootStart   EndBlocks   Id  System
> /dev/sda1   * 1 4 16461   f0  Linux/PA-RISC boot

This should be bigger: 100-200MB about.
With "-e2" option, palo can create an ext2 file system that can
be mounted as /boot.
i.e. files will be accessed with 1/vmlinux instead of 1/boot/vmlinux


> /dev/sda2 534123690   82  Linux swap
> /dev/sda3   *35   277   1001889   83  Linux
> 
> is fine, because the small swap partition doesn't push root out too far. It's 
> a better idea, however, to put the swap partition at the end of the disk. By 
> the way, don't forget to make both the f0 partition and the root partition 
> bootable.

Swap, if needed, is best on the lowest numbered block possible since
those typically have the highest bitrate.

"bootable" flag doesn't mean anything to PA-RISC PDC. We use
the DOS partition table format becuase it's trivial.

> 8.2 First steps
> 
> I'm the conservative type, so I don't load from "testing".
>  At the time of this writing, the 2.6 kernels in "unstable" are 2.6.8-2-32, 
> 2.6.8-2-32-smp, 2.6.8-2-64 and 2.6.8-2-64-smp.

Ok.

>  Even though the PA-Risc cpu's in my machine are 64 bit, hppa Debian can't 
> take advantage of the 64 bit instruction length, so I just stick with the 32 
> bit versions.

Drop this sentence (run-on sentence at that).
It's not entirely correct in this context (user vs kernel space).

> So, now it is to begin to install a 2.6 kernel.
> 
> Step 1 For example, apt-get install kernel-image-2.6.7-2-32-smp. When you get 
> near the end of the install, you will get messages telling you that this is 
> an initrd kernel and asking you if you want to abort. At this point say, 
> "Yes". Now it's time for:

Please wrap long lines to 72 or fewer characters per line.
But it's up to thibaut (or whoever is maintaining) if they can
add this as-is.

> Step 2  you need to add "sym53c8xx" to /etc/mkinitrd/modules. "modules" is 
> created from /etc/mkinitrd/modules-new? Help, I don't remember the name the 
> aborted load leaves behind.
> 
> This is how it looks after editing:
> 
> % cat /etc/mkinitrd/modules-new??
> # /etc/mkinitrd/modules: Kernel modules to load for initrd.
> #
> # This file should contain the names of kernel modules and their arguments
> # (if any) that are needed to mount the root file system, one per line.
> # Comments begin with a `#', and everything on the line after them are 
> ignored.
> #
> # You must run mkinitrd(8) to effect this change.
> #
> # Examples:
> #
> #  ext2
> #  wd io=0x300
> 
> tulip
> sym53c8xx
> 
> Okay. Now we are ready for:
> Step 3 apt-get install kernel-image-2.6.7-2-32-smp again, but this time when 
> asked if you want to abort, say, "No".

2.6.8 ?

> 
> 8.3 Edit /etc/palo.conf
> 
> Edit /etc/palo.conf and insert on the --commandline initrd=x/boot/initrd.img 
> where x is the number of the disk partition where /boot lives. Here's an 
> example of a palo.conf that works:
> 
> --commandline=3/boot/vmlinux initrd=3/boot/initrd.img HOME=/ root=/dev/sda3
> --recoverykernel=/boot/vmlinux.old
> --init-partitioned=/dev/sda
> 
> 
> Then run palo!
> 
> 8.4 Before you reboot
> 
> On some machines installing the 2.6 kernel flips the names of your hard 
> drives (if you have more than one). For example, sda might become sdb with 
> sbd becoming sda. If this happens, you will get an error on reboot that you 
> can fix by going into ipl and changing, for example root=/dev/sda3 to 
> root=/dev/sdb3

RE: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Harry Cochran]

Thanks Grant,

Looks like I'm going around one more time. I don't see how to get to
the -e2 option in the standard installer. I guess you are suggesting I just
drop into the ash shell and run fdisk. Of course, being a newbie, I would
prefer to just create the f0 partition with -e2 in the standard install disk
partitioner. Is there a way to do that?

Best regards,

Harry

-Original Message-
From: Grant Grundler [mailto:[EMAIL PROTECTED]
Sent: Saturday, February 26, 2005 7:14 PM
To: Harry Cochran
Cc: Thibaut VARENE; debian-hppa@lists.debian.org
Subject: Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6


On Sat, Feb 26, 2005 at 04:52:58PM -0500, Harry Cochran wrote:
> Hi,
>
>   I don't know if this has any value or not, but I am happy to
> contribute it to PA-RISC/Linux Boot HOWTO with approval of those
> explicitly copied on this E-mail whose contributions I copied.

You have my permission to contribute everything you learned. :^)
Though I don't see why you explicitly need my permission.
I'm not the maintainer.


Some comments below.

...
> 8. Upgrading from a Debian hppa 2.4 Kernel to a 2.6 kernel
>
> 8.1 Check your 2.4 partition scheme
> 8.2 First Steps
> 8.3 Edit /etc/palo.conf
> 8.4 Before you reboot
> 8.5 Conclusion
>
> 8.1 Check your 2.4 partition scheme
>
> It may come as a surprise, but it is possible (I know, I did it) to have a
partition scheme that by some fluke "works" under 2.4 but is invalid and
will not work under 2.6. For example:
>
> > palo ipl 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:05 MDT 2004
> >
> > Partition Start(MB) End(MB) Id Type
> > 1   1  31   f0 Palo
> > 2  321008   82 swap
> > 31009   17366   83 ext2  root
>
> The problem here is that the large swap partition pushes /boot under the
root out past the 2GB limit for booting. There is no ready cure for this.
You will have to save your data, re-partition your hard drive and then
reload your applications and data. The example in 4.3:
>
> Disk /dev/sda: 133 heads, 62 sectors, 1017 cylinders
> Units = cylinders of 8246 * 512 bytes
>
>Device BootStart   EndBlocks   Id  System
> /dev/sda1   * 1 4 16461   f0  Linux/PA-RISC boot

This should be bigger: 100-200MB about.
With "-e2" option, palo can create an ext2 file system that can
be mounted as /boot.
i.e. files will be accessed with 1/vmlinux instead of 1/boot/vmlinux


> /dev/sda2 534123690   82  Linux swap
> /dev/sda3   *35   277   1001889   83  Linux
>
> is fine, because the small swap partition doesn't push root out too far.
It's a better idea, however, to put the swap partition at the end of the
disk. By the way, don't forget to make both the f0 partition and the root
partition bootable.

Swap, if needed, is best on the lowest numbered block possible since
those typically have the highest bitrate.

"bootable" flag doesn't mean anything to PA-RISC PDC. We use
the DOS partition table format becuase it's trivial.

> 8.2 First steps
>
> I'm the conservative type, so I don't load from "testing".
>  At the time of this writing, the 2.6 kernels in "unstable" are
2.6.8-2-32, 2.6.8-2-32-smp, 2.6.8-2-64 and 2.6.8-2-64-smp.

Ok.

>  Even though the PA-Risc cpu's in my machine are 64 bit, hppa Debian can't
take advantage of the 64 bit instruction length, so I just stick with the 32
bit versions.

Drop this sentence (run-on sentence at that).
It's not entirely correct in this context (user vs kernel space).

> So, now it is to begin to install a 2.6 kernel.
>
> Step 1 For example, apt-get install kernel-image-2.6.7-2-32-smp. When you
get near the end of the install, you will get messages telling you that this
is an initrd kernel and asking you if you want to abort. At this point say,
"Yes". Now it's time for:

Please wrap long lines to 72 or fewer characters per line.
But it's up to thibaut (or whoever is maintaining) if they can
add this as-is.

> Step 2  you need to add "sym53c8xx" to /etc/mkinitrd/modules. "modules" is
created from /etc/mkinitrd/modules-new? Help, I don't remember the name the
aborted load leaves behind.
>
> This is how it looks after editing:
>
> % cat /etc/mkinitrd/modules-new??
> # /etc/mkinitrd/modules: Kernel modules to load for initrd.
> #
> # This file should contain the names of kernel modules and their arguments
> # (if any) that are needed to mount the root file system, one per line.
> # Comments begin with a `#', and everything on the line after them are
ignored.
> #
> # You must run mkinitrd(8) to effect this change.
> #
> # Examples:
> #
> #  ext2
> #  wd io=0x300
>
> tulip
> sym53c8xx
>
> Okay. Now we are ready for:
> Step 3 apt-get install kernel-image-2.6.7-2-32-smp again, but this time
when asked if you want to abort, say, "No".

2.6.8 ?

>
> 8.3 Edit /etc/palo.conf
>
> Edit /etc/palo.conf and insert on the --commandline
initrd=x/boot/initrd.img where x is the number of the disk partition where
/boot lives. Here's an

RE: FW: 2.4 -> 2.6

[Harry Cochran]

Thanks Stuart,

When I started this whole process, I had a nice f0 /boot partition of 
100MB
that occupied the beginning sectors of my disk. When I did the reinstall, I
couldn't get the installer to let me mount the f0 partition as /boot. Grant
just told me about -e2. Onwards and upwards.

Cheers,

Harry

-Original Message-
From: Stuart Brady [mailto:[EMAIL PROTECTED]
Sent: Saturday, February 26, 2005 5:55 PM
To: Harry Cochran
Cc: debian-hppa@lists.debian.org
Subject: Re: FW: 2.4 -> 2.6


On Sat, Feb 26, 2005 at 12:17:26PM -0500, Harry Cochran wrote:
> Thank you all very much! I now have a running 2.6.8-2-32-smp on my
> J6000. I repartitioned sda to have the root right next to the f0
> partition (and the swap at the end) per Grant,

No, that isn't what Grant said...

The Boot HOWTO explains that the kernel image must be located within the
first 2GB of the disk.  The only sensible way to ensure this is to have
a dedicated /boot partition, which is contained entirely within the
first 2GB.  This needn't be large -- a 64MB partition should be ample.

In your case, it doesn't matter whether this is before or after the 1GB
swap partition, but it must come before the large root (/) partition.

AFAICS, although your new partition scheme may appear to be working, it
may cease to work whenever the kernel is upgraded.  (By swapping the
partitions you are effectively doubling the amount of disk space from
the first 2GB which is used for the root partition.)

Hope that helps,
--
Stuart Brady


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Grant Grundler]

On Sat, Feb 26, 2005 at 07:41:45PM -0500, Harry Cochran wrote:
> Thanks Grant,
> 
>   Looks like I'm going around one more time. I don't see how to get to
> the -e2 option in the standard installer.

One can't. That's a known problem with debian-installer.

> I guess you are suggesting I just
> drop into the ash shell and run fdisk. Of course, being a newbie, I would
> prefer to just create the f0 partition with -e2 in the standard install disk
> partitioner. Is there a way to do that?

no.
But if one makes the f0 partition big enough, one can do it *after*
the install completes and the system is up and running.

For developement, I suggest a 200MB f0 partition.
For "production", I suggest at least 50MB.

grant


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Grant Grundler]

On Sat, Feb 26, 2005 at 08:05:23PM -0500, Harry Cochran wrote:
> The real question is "with the -e2 option" on what? Guess I'm tired because
> I know ask doesn't have fdisk.

Yes, you must be tired because you are "top posting". :^)

-e2 is either a palo parameter (when invoked from user shell)
or --format-as=2 to /etc/palo.conf file.
See "palo -?" output.

grant


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

RE: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Harry Cochran]

-Original Message-
From: Grant Grundler [mailto:[EMAIL PROTECTED]
Sent: Saturday, February 26, 2005 8:37 PM
To: Harry Cochran
Cc: debian-hppa@lists.debian.org
Subject: Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6


On Sat, Feb 26, 2005 at 08:05:23PM -0500, Harry Cochran wrote:
> The real question is "with the -e2 option" on what? Guess I'm tired
because
> I know ask doesn't have fdisk.

Yes, you must be tired because you are "top posting". :^)

-e2 is either a palo parameter (when invoked from user shell)
or --format-as=2 to /etc/palo.conf file.
See "palo -?" output.

grant

Okay. Now I learned something else about parisc-linuxers ... add comments at
the bottom. Who knew?

So, here's question 1: Why did you say:

> Partition Start(MB) End(MB) Id Type
> 1   1  31   f0 Palo
> 2  321008   82 swap
> 31009   17366   83 ext2

"This is an invalid partition scheme."

Is it just that the f0 partition is too small? Can one not have a swap
partition this big?

and here's question number 2: Why do none of the examples I've seen
show --format-as=2 being used in palo.conf and can you please show me a
complete example (recommended partition layout for an 18GB hard drive on a
machine with 2GB of memory and a palo.conf that goes along with it)? I can't
look at palo -? right now because I just blew my partitions away so I can
re-partition and try it all one more time. Sorry.

Thanks,

Harry


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

/boot

[Harry Cochran]

Hi,

Is an easier way to do this just to have a 50MB f0 partition and then a
50-200MB e2 partition that you mount as /boot?

Thanks,

Harry


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Grant Grundler]

On Sat, Feb 26, 2005 at 08:47:17PM -0500, Harry Cochran wrote:
> Okay. Now I learned something else about parisc-linuxers ... add comments at
> the bottom. Who knew?

And a smarter mail handler that knows how to quote! :^)

> So, here's question 1: Why did you say:
> 
> > Partition Start(MB) End(MB) Id Type
> > 1   1  31   f0 Palo
> > 2  321008   82 swap
> > 31009   17366   83 ext2
> 
> "This is an invalid partition scheme."
> 
> Is it just that the f0 partition is too small?
> Can one not have a swap partition this big?

No and No.
The ext2 partition must be /.
And it is 17GB.
ie. /boot and /boot contents are not guaranteed to be within
the first 2GB of disk blocks. See the HOWTO-Boot/Install.
Folks have pointed at it more than a few times.

> and here's question number 2: Why do none of the examples I've seen
> show --format-as=2 being used in palo.conf

Because the feature was added *after* the documentation was written.
No one has submitted patches to fix/update the documentation.
[ HINT HINT :^) ]

> and can you please show me a complete example
> (recommended partition layout for an 18GB hard drive
> on a machine with 2GB of memory and a palo.conf that
> goes along with it)?

I did post one for a 73GB disk:
http://lists.parisc-linux.org/pipermail/parisc-linux/2005-February/025935.html

Adjust partitions 3 and 4 accordingly.
part3 is /
part4 is /home

I suggest at least 4-6GB for /
then /var or /home for the rest depending on the use of the machine.
e.g size for /var/www or /var/cvs for a "server" vs /home
for a shared development machine where /usr/src is a symlink
to /home/src.

Key thing is part1 (type f0) is /boot

> I can't
> look at palo -? right now because I just blew my partitions away so I can
> re-partition and try it all one more time. Sorry.

np.

grant


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: /boot

[Grant Grundler]

On Sat, Feb 26, 2005 at 09:05:34PM -0500, Harry Cochran wrote:
> Hi,
> 
>   Is an easier way to do this just to have a 50MB f0 partition and then a
> 50-200MB e2 partition that you mount as /boot?

no.

grant


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

RE: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Harry Cochran]

On Sat, Feb 26, 2005 at 09:15PM -0500, Grant Grundler wrote:
>The ext2 partition must be /.
>And it is 17GB.
>ie. /boot and /boot contents are not guaranteed to be within
>the first 2GB of disk blocks.

>I suggest at least 4-6GB for /
If / is 6GB how can /boot be graranteed to be within the first 2GB of disk
blocks?

>then /var or /home for the rest depending on the use of the machine.

I have a 2nd 18BG drive for /var

>Key thing is part1 (type f0) is /boot

I feel like I'm going around in circles. Seems like you are saying /boot is
in "part3 is /" in your example and it's also part1.

Does --format-as=2 make part1 /boot? If so, why does it matter if / is 6GB
or 17GB?

Finally, in your example, why do you use --update-partitioned=/dev/sda
instead of --init-partitioned=/dev/sda?

Thanks for your patience with newbie's. I think you can sense how hard I'm
trying to understand.

Harry



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: FW: 2.4 -> 2.6

[Stuart Brady]

On Sat, Feb 26, 2005 at 08:11:37PM -0500, Harry Cochran wrote:
> Are you saying that I need an f0 partition and then another partition of
> 64MB that I mount as /boot?

That's what I was saying, but it's the old way of dealing with the
problem.

> I've never seen this in any of the examples, but the installer won't let
> me mount an f0 partition at all and I have to say I don't understand
> Grants -e2 hint.

If I've understood correctly, the new approach is to format the Palo
(f0) partition as ext2 (using the -e2 option to palo) and then use that
as /boot, instead of having an extra partition.  I haven't tried this,
so I'm not the best person to ask about it.
-- 
Stuart Brady


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

RE: /boot

[Harry Cochran]

>On Sat, Feb 26, 2005 at 09:05:34PM -0500, Harry Cochran wrote:
>> Hi,
>>
>>  Is an easier way to do this just to have a 50MB f0 partition and then a
>> 50-200MB e2 partition that you mount as /boot?

On Sat, Feb 26, 2005 at 09:16:34PM -0500, Grant Grundler wrote:
>no.

>grant

Okay. Trying it your way one more time. However ... first palo.conf

SinoHub5:/etc# cat palo.conf
--commandline=1/vmlinux initrd=1/initrd.img root=/dev/sda3 HOME=/
--format-as=2
--update-partition=/dev/sda
--recoverykernel=/boot/vmlinux-2.4.17-32
SinoHub5:/etc# palo
palo version 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:17 MDT 2004
ELF32 executable
For ext2/3-formatted palo partitions, you cannot specify a kernel
or a ramdisk.

2nd try to identify error:

SinoHub5:/etc# cat palo.conf
--commandline=0/vmlinux root=/dev/sda3 HOME=/
--format-as=2
--update-partition=/dev/sda
--recoverykernel=/boot/vmlinux-2.4.17-32
SinoHub5:/etc# palo
palo version 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:17 MDT 2004
ELF32 executable
For ext2/3-formatted palo partitions, you cannot specify a kernel
or a ramdisk.

3rd try to identify error:

SinoHub5:/etc# cat palo.conf
--commandline=1/vmlinux initrd=1/initrd.img root=/dev/sda3 HOME=/
--format-as=2
--update-partition=/dev/sda
SinoHub5:/etc# palo
palo version 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:17 MDT 2004
OK we're doing a format as ext2
ipl: addr 16384 size 36864 entry 0x0
Warning!
Boot loader header version is 3, I only know how
to handle version 4.  It MIGHT work anyway.
 ko 0x44000 ksz 3687647 k64o 0x0 k64sz 0 rdo 0 rdsz 0
<1/vmlinux initrd=1/initrd.img root=/dev/sda3 HOME=/>

Fourth try to identify error:

SinoHub5:/etc# cat palo.conf
--commandline=0/vmlinux root=/dev/sda3 HOME=/
--format-as=2
--update-partition=/dev/sda
SinoHub5:/etc# palo
palo version 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:17 MDT 2004
OK we're doing a format as ext2
ipl: addr 16384 size 36864 entry 0x0
Warning!
Boot loader header version is 3, I only know how
to handle version 4.  It MIGHT work anyway.
 ko 0x44000 ksz 3687647 k64o 0x0 k64sz 0 rdo 0 rdsz 0
<0/vmlinux root=/dev/sda3 HOME=/>

This, plus 10 tries at booting, all tells me that:

1. You can't specify a recovery kernel with this method (not very comforting
... but palo "finds" 2.4.27-32 every time anyway). I was wondering why you
didn't specify one in your example. Now I know.

2. initrd has nothing to do with the v3 vs. v4 problem

3. I can't find the magic in front of the symlinks vmlinux and initrd.img to
get 2.6 to load. I get "failed to load ram disk" every time with 10 tries at
initrd=x, where x is 0/initrd.img, 1/initrd.img, 1/boot/initrd.img,
/boot.initrd.img, etc. Or maybe it's something else entirely. Having an f0
partition and an e2 /boot is looking pretty good right now!

No joy tonight,

Harry





-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: /boot

[Grant Grundler]

On Sat, Feb 26, 2005 at 11:55:19PM -0500, Harry Cochran wrote:
> Fourth try to identify error:
...

How about reading "palo --help"?

> SinoHub5:/etc# cat palo.conf
> --commandline=1/vmlinux initrd=1/initrd.img root=/dev/sda3 HOME=/
> --format-as=2
> --update-partition=/dev/sda
> SinoHub5:/etc# palo
> palo version 1.5 [EMAIL PROTECTED] Tue Sep 21 15:14:17 MDT 2004
> OK we're doing a format as ext2
> ipl: addr 16384 size 36864 entry 0x0
> Warning!
> Boot loader header version is 3, I only know how
> to handle version 4.  It MIGHT work anyway.

Could you replace "--update-partition=" with "--init-partition="?

I'm expecting palo to blow everything on f0 away and build a new
ext2 file system. THen you can:
o mount /dev/sda1 /mnt
o copy /boot/* /mnt
o edit /etc/fstab so /dev/sda1 is mounted on /boot.

If that boots, then:
o unmount /boot
o rm /boot/*
o mount /boot

As you might guess, the order is important. :^)


> This, plus 10 tries at booting, all tells me that:
> 
> 1. You can't specify a recovery kernel with this method (not very comforting
> ... but palo "finds" 2.4.27-32 every time anyway). I was wondering why you
> didn't specify one in your example. Now I know.

One isn't needed. palo can list the contents of an ext2 directory.
So any vmlinux/initrd can be used if palo can list it.

> 2. initrd has nothing to do with the v3 vs. v4 problem

right

> 3. I can't find the magic in front of the symlinks vmlinux and initrd.img to
> get 2.6 to load. I get "failed to load ram disk" every time with 10 tries at
> initrd=x, where x is 0/initrd.img, 1/initrd.img, 1/boot/initrd.img,
> /boot.initrd.img, etc. Or maybe it's something else entirely.

Can you dump the following info (cut/paste to a shell):
fdisk -l /dev/sda
mount
cat /etc/palo.conf
ls -l /boot

The "ls" command assumes the f0 partition is mounted on /boot.

>  Having an f0
> partition and an e2 /boot is looking pretty good right now!

/boot on f0 partition definitely works.
I'm certainly not the only one using it.

hth,
grant


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: PA-RISC/Linux Boot HOWTO Section 8, Draft 1, 2.4 -> 2.6

[Grant Grundler]

On Sat, Feb 26, 2005 at 09:46:46PM -0500, Harry Cochran wrote:
> >I suggest at least 4-6GB for /
> If / is 6GB how can /boot be graranteed to be within the first 2GB of disk
> blocks?

Sorry - this is assuming the f0 partition is mounted as /boot.

> >then /var or /home for the rest depending on the use of the machine.
> 
> I have a 2nd 18BG drive for /var
> 
> >Key thing is part1 (type f0) is /boot
> 
> I feel like I'm going around in circles. Seems like you are saying /boot is
> in "part3 is /" in your example and it's also part1.

No. / is sda3 (type 82). /boot is sda1 (type f0).

> Does --format-as=2 make part1 /boot?

Not automatically. It is just a necessary step.

> If so, why does it matter if / is 6GB or 17GB?

It doesn't for my setup. The size of / is just a reccomendation.

> Finally, in your example, why do you use --update-partitioned=/dev/sda
> instead of --init-partitioned=/dev/sda?

Because I ran --init-partitioned manually before.

> Thanks for your patience with newbie's. I think you can sense how hard I'm
> trying to understand.

*nod*

grant


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]