etherboot-developers — Discussion of developer issues

Re: [Etherboot-developers] Q: Multiple drivers.

[<ke...@us...>]

>1) Is it reasonable for etherboot to support multiple cards,
>   and possible multiple cards with different drivers at
>   one time.

Yes, this is a reasonable thing. In the past when the code was in a ROM,
it would have violated the principle of least surprise for the code on
one NIC's ROM to boot another NIC, but Etherboot has extended its
tentacles into the BIOS and other places now. One thing that would be
nice, if possible, is for the single driver version to by selection or
by default boot only the physical NIC it lives in. That way controlling
the interface is a matter of choosing the right NIC to put the ROM in.

The interface to the drivers is rather poor, it only takes the first NIC
it finds. Ideally it should enumerate all the NICs it can boot and
choose a good default or give the user a choice (one of multiple NICs of
the same or different models).

>2) Would it be reasonable to have a version of the hard/floppy
>   booting code that uses polled ide/floppy drivers instead of
>   the BIOS.

Provided that this is a compile time option, this is ok. I expect that
the code will take up room and ROM users should not have to pay the cost
in ROM space if they are not using the disk access. At the same time,
the API to disk access could probably be improved.

>I have code that implements supports multiple cards with the same
>drive, and polled ide & floppy access, and I'm trying to find
>the best place for it to have a home, be maintained, and be useful.
>
>If it something you don't mind seeing in etherboot then I will look
>into this.  

Sounds great. Any contributions are gratefully received.

[Etherboot-developers] Q: Multiple drivers.

[<ebi...@ln...>]

I have two related questions:

1) Is it reasonable for etherboot to support multiple cards,
   and possible multiple cards with different drivers at
   one time.

2) Would it be reasonable to have a version of the hard/floppy
   booting code that uses polled ide/floppy drivers instead of
   the BIOS.


I have code that implements supports multiple cards with the same
drive, and polled ide & floppy access, and I'm trying to find
the best place for it to have a home, be maintained, and be useful.

If it something you don't mind seeing in etherboot then I will look
into this.  

Thanks, for any guidance.

Eric

Re: [Etherboot-developers] Re: [Ltsp-discuss] Wireless etherboot

[Doug A. <amb...@am...>]

Marty Connor writes:
| On 11/30/2001 12:11 PM Mark ma...@of... wrote:
| >Oops, try this link - http://www.ambrisko.com/doug/airport/
|
| Wow.  This is auite a hack.  I downloaded it and looked at the code.  I'm
| working with someone on PCMCIA support for Etherboot, but we don't have
| anything useful yet.  I didn't realize Doug was working on this.  I may
| have missed an email somewhere :)

I'm not Till Straumann did it.  However, this is a very special case
since it is an Airport base station.  In this case the card is tied
to an AMD SC400 with a built-in PCMCIA adapter.  It's not a full
feature PCMCIA implementation.  I think the BIOS sets up the card
partially so then you can just access it.  I just posted it to my
web-site.

Since changing jobs I've lost access to that hardware.  He did some
neat things like using the modem as a serial console.  I had it 
netboot FreeBSD and boot to multi-user using NFS for the root & swap
since you only have 4M of RAM.

Doug A.

[Etherboot-developers] Re: [Ltsp-discuss] Wireless etherboot

[Marty C. <md...@th...>]

On 11/30/2001 12:11 PM Mark ma...@of... wrote:
>Oops, try this link - http://www.ambrisko.com/doug/airport/

Wow.  This is auite a hack.  I downloaded it and looked at the code.  I'm 
working with someone on PCMCIA support for Etherboot, but we don't have 
anything useful yet.  I didn't realize Doug was working on this.  I may 
have missed an email somewhere :)

Good Show!

Marty

---
    Try: http://rom-o-matic.net/ to make Etherboot images instantly.

   Name: Marty Connor
US Mail: Entity Cyber, Inc.; P.O. Box 391827; Cambridge, MA 02139; USA
  Voice: (617) 491-6935, Fax: (617) 491-7046 
  Email: md...@th...
    Web: http://www.thinguin.org/

[Etherboot-developers] Asound ALM2 (or ALM02) support?

[Jason P. <ja...@pe...>]

Hello,

Recently, the fealnx.c driver in the linux kernel has been expanded 
to support the Asound ALM2 ethernet controller.

Is there any chance of having this card supported in etherboot?  
Since a linux driver for the card already exists, I assume it is 
not impossible, but then I have never written a driver...

Thanks,
jason pepas
ja...@pe...

[Etherboot-developers] Etherboot / mknbi without VGA boot problem

[Marcel M. <Mar...@da...>]

Hi Everybody,

As I have reported ZFx86 (Phoenix BIOS) with 82559ER is booting now.
When we unplugged VGA card, Etherboot downloads kernel by TFTP, but
mknbi not starts. It is new problem. Can Etherboot run without PCI VGA
card in principle?  We tried many versions of Etherboot, 5.0.4 patch
included.

Using a serial line interface we have got this:
---
ROM segment 0x8000 length 0x4000 reloc 0x9400
Etherboot 5.1.1 (GPL) Tagged ELF (Multiboot) for [EEPRO100]
Found Intel EtherExpressPro100 at 0x1000, ROM address 0x0000
Probing...[EEPRO100]
The PCI BIOS has not enabled this device!
Updating PCI command 0013->0017. pci_bus 0000 pci_device_fn 0050
Ethernet addr: 00:02:B3:1F:B1:55=20
Searching for server (DHCP)...
Me: 192.168.254.54, Server: 192.168.254.40
Loading 192.168.254.40:vmlinux (NBI)... done
<abort>
Found Intel EtherExpressPro100 at 0x1000, ROM address 0x0000
Probing...[EEPRO100]Ethernet addr: 00:02:B3:1F:B1:55=20
Searching for server (DHCP)...
Me: 192.168.254.54, Server: 192.168.254.40
Loading 192.168.254.40:vmlinux (NBI)... -
---
...and than stops- froze.

Do have somebody an idea, why it needs VGA?
It works in same conditions with VGA.=20

Marcel

[Etherboot-developers] just joined the list - how busy is it ?

[Per J. <pe...@co...>]

thanks, 
Per Jessen, Zurich

[Etherboot-developers] Etherboot 5.0.4-mc3 available on rom-o-matic.net -- Please Test

[Marty C. <md...@th...>]

Etherboot 5.0.4-mc3, a release candidate for Etherboot 5.0.5 is available 
on

   http://rom-o-matic.net/

There are some fairly significant changes from 5.0.4 thanks to the hard 
work of many contributors.  The changes are documented on the Etherboot 
project page at:

   http://sourceforge.net/projects/etherboot/

in the "Patches" section.

We ask that you test 5.0.4-mc3 and report problems or new successes to 
Eth...@li....


---
    Try: http://rom-o-matic.net/ to make Etherboot images instantly.

   Name: Marty Connor
US Mail: Entity Cyber, Inc.; P.O. Box 391827; Cambridge, MA 02139; USA
  Voice: (617) 491-6935, Fax: (617) 491-7046 
  Email: md...@th...
    Web: http://www.thinguin.org/

Re: [Etherboot-developers] Mknbi-1.2-6 (production) released

[Vasil V. <vas...@sy...>]

I've committed the PXE patch to a branch vasilev-freePXEmem of the file
etherboot/etherboot-5.1/src/loader.S . I've tested it on the newer version of
PXE's API (ver 2.1 or above) but believe it should work on the older one
as well -- the message printed will then be 'PXENV unloaded' instead of
'PXE   unloaded'.

I've put a placeholder for using PXE's UNDI if someone is interested in
implementing it to replace the usage of specific driver. I might attempt to do
this in a couple of months if nobody else has.

Vasil

[Etherboot-developers] Mknbi-1.2-6 (production) released

[<ke...@us...>]

I have released Mknbi 1.2-6 at
http://etherboot.sourceforge.net/distribution.html

Changes from 1.2-5:

+ Implement --rdbase for mknbi-fdos and mknbi-dos to specify starting
address of ramdisk (floppy image).

+ Make first.*dos depend on Makefile so that the version signature gets
updated when the version is updated.

+ Print some friendly advice if --ip is not specified and there is no
ramdisk argument (i.e. probably needs NFSroot).

+ Shuffle segments to make more space for parameters and first32. This
version and future versions may not work with Etherboot images that have
been compiled with BOOTP_DATA_AT_0x93Cxx. (There is some reprieve due to
first32.c not using all 6kB of the allocated room yet, but that space
may be used some day.)

+ Implement interception and decoding of the mem= kernel parameter to
use as top of memory for the purpose of relocating the ramdisk.

Re: [Etherboot-developers] Etherboot on ZFx86 & 82559ER?

[Christoph P. <chr...@al...>]

I had the bus master problem also with the 82559ER !

I added the line

	if (......)
>>	|| (..... == ...82559ER) 

for enabling the bus master also for the 82559ER (ID=1209) and not
only for 8255[789], ID=1229 in eepro100.c.

Simple search after "82557" in the C file (eepro100.c) and you
will find it !!!

With friendly regards
	Christoph P.

Bernd Wiebelt wrote:
> 
> On Tue, 13 Nov 2001, Marcel Mlynarik wrote:
> 
> > Does anybody have working etherboot on ZFx86 (MachZ) using NIC with
> > 82559ER?
> 
> FWIW, I had a problem with the MachZ bios not enabling busmastering for a
> Realtek8139.
> 
> Bernd
> 
> _______________________________________________
> Etherboot-developers mailing list
> Eth...@li...
> https://lists.sourceforge.net/lists/listinfo/etherboot-developers

-- 
  +--------V--------+	Chr...@al...
  |  A L C A T E L  |	-----------------------------
  +-----------------+	Phone: +43 1 27722 3706	
         T A S		Fax:   +43 1 27722 3955

Re: [Etherboot-developers] Etherboot on ZFx86 & 82559ER?

[Bernd W. <wi...@dr...>]

On Tue, 13 Nov 2001, Marcel Mlynarik wrote:

> Does anybody have working etherboot on ZFx86 (MachZ) using NIC with
> 82559ER?

FWIW, I had a problem with the MachZ bios not enabling busmastering for a
Realtek8139.

Bernd

Re: [Etherboot-developers] Etherboot on ZFx86 & 82559ER?

[Marty C. <md...@th...>]

On 11/13/2001 5:17 AM Marcel Mlynarik Mar...@da... wrote:
>Does anybody have working etherboot on ZFx86 (MachZ) using NIC with
>82559ER?
>I know 3 people are trying it unsuccessfully.
>On 82559 it runs.
>Can somebody help? 

Does it work under Linux using the eepro100 driver?

(This might be an Etherboot Users question...)

Post some output so we can see the failure mode and perhaps we can help 
more.
You may have to type it in by hand, but it will give us some idea where 
the problem lies.

What are the PCI IDs for the ZFx86?  (the "lspci" program will tell you 
or you can use "scanpci")

Let us know what you find.  
Marty

---
    Try: http://rom-o-matic.net/ to make Etherboot images instantly.

   Name: Marty Connor
US Mail: Entity Cyber, Inc.; P.O. Box 391827; Cambridge, MA 02139; USA
  Voice: (617) 491-6935, Fax: (617) 491-7046 
  Email: md...@th...
    Web: http://www.thinguin.org/

[Etherboot-developers] Etherboot on ZFx86 & 82559ER?

[Marcel M. <Mar...@da...>]

Does anybody have working etherboot on ZFx86 (MachZ) using NIC with
82559ER?

I know 3 people are trying it unsuccessfully.
On 82559 it runs.
Can somebody help?=20

Marcel =20

Re: [Etherboot-developers] Re: 1G memory limit and Etherboot

[<ke...@us...>]

>On the topic of memory, I have a suggestion for etherboot:
>By default etherboot to override the amount of memory BIOS reports to be the
>amount passed to the kernel in "mem=". There, of course, should be the option
>of using BIOS's report even if "mem=" is given with the onus on the user to
>get it right.

Ok, I've CVSed the changes to support interception and decoding of mem=.
Now first32.c will treat the value in mem= as the effective top of
memory, just like Linux does. There is no override to use the BIOS
value. I'm not sure in what circumstance one would want Linux to use
mem=, but first32.c to use the BIOS value.

Feel free to check out and try it. If there are no problems I'll release
1.2-6 soon.

[Etherboot-developers] Etherboot CVS brought up to date

[<ke...@us...>]

I've brought the Etherboot CVS archive up to date with all the changes
I've made to 5.0.4 so far.

Re: [Etherboot-developers] Re: 1G memory limit and Etherboot

[<ke...@us...>]

>In any case this will require reshuffling of the memory areas so that
>first32.c can be larger than 4kB. I have already done this for 1.2.6 and
>I'll CVS the source soon. Then if you feel up to hacking first32.c,
>please go ahead, I may not get to it immediately.

I have CVS'ed the changes towards 1.2-6 with more room for first32.c and
parameters. Hack away if you want.

Re: [Etherboot-developers] Re: 1G memory limit and Etherboot

[<ke...@us...>]

>On the topic of memory, I have a suggestion for etherboot:
>By default etherboot to override the amount of memory BIOS reports to be the
>amount passed to the kernel in "mem=". There, of course, should be the option
>of using BIOS's report even if "mem=" is given with the onus on the user to
>get it right.

A mem= option in first32.c (mknbi) (from command line parameters or
option-129) similar to in LILO etc is probably a good idea.  first32.c
would note the value and use that instead of the BIOS value, then pass
it on to the kernel, as it currently does with all options not
recognised by it. I'm not sure how to specify the override, with another
option maybe?

In any case this will require reshuffling of the memory areas so that
first32.c can be larger than 4kB. I have already done this for 1.2.6 and
I'll CVS the source soon. Then if you feel up to hacking first32.c,
please go ahead, I may not get to it immediately.

[Etherboot-developers] Re: 1G memory limit and Etherboot

[<vas...@sy...>]

Peter Lister and I had recently some experiences with etherboot and initrd
because of a braindead BIOS. I haven't been able to try this scenario as the
machine I can have 1GB memory has the breandead BIOS, which can only report
upto 64MB extended memory.

So, here are my 2c:

I guess the kernel has been compiled to have only 1GB memory (as opposed to
2GB,4GB or other). In that case, the kernel uses a maximum of 960MB on 2.2
kernel, or 896MB on 2.4 kernel. Then initrd will be disabled as the kernel
sees that it extends beyond the available memory --- check the boot messages;
should be one of the very first lines.

The workarounds are to use the --rdbase option with mknbi as suggested
previously or compile the kernel with at least 2GB support. The latter will be
beneficial for other purposes as well any way.

If the problem proves to be as I described above, etherboot cannot
solve it other than the automatic choice of the initrd location to be lower
than 896MB. Then another option will be necessary to override that.

On the topic of memory, I have a suggestion for etherboot:
By default etherboot to override the amount of memory BIOS reports to be the
amount passed to the kernel in "mem=". There, of course, should be the option
of using BIOS's report even if "mem=" is given with the onus on the user to
get it right.

Vasil

Re: [Etherboot-developers] Cardbus (PCMCIA) and etherboot

[<ke...@us...>]

>Hello from Gregg C Levine normally with Jedi Knight Computers
>Where are we right now, on the subject of etherboot, and Cardbus
>(PCMCIA)? On the homepage's entries on the things to do page, I see that
>it is noted. But has any progress been done? Anyone?

It should be noted that Marty is a strong contributor to Etherboot and
has a lot on his plate, including his day job. Contributors to Etherboot
are always welcome. Not all tasks require deep hacking knowledge. For
example I've been wanting for a while to convert the current set of docs
to Docbook. This is something not requiring C programming skills.  If
you know PHP, Marty's code for rom-o-matic could do with some tidying
and packaging.  If you just ask, I'm sure half a dozen things could be
found that need doing.

As the old Confucian saying about light bulbs goes: Many hands make
light work.

Re: [Etherboot-developers] Dropping DHCP messgaes

[<ke...@us...>]

>I forget to say the code is fine for using bootp. That is strange that 
>they both obey the same standard but behaves differently at the client side.

That's because in BOOTP the server sends everything it knows whereas in
DHCP, the server only sends the items that the client has asked for.

Re: [Etherboot-developers] Dropping DHCP messgaes

[David C. <dav...@rc...>]

I forget to say the code is fine for using bootp. That is strange that 
they both obey the same standard but behaves differently at the client side.

Peter Lister wrote:

>>Ok, in first32.c of mknbi, I modify it so that is also extract the 
>>nisdomain (rfc1533 tag 40)  from the bootp messages. But when using dhcp 
>>I find the nisdomain tag is missing or empty string. What is the 
>>problem? It seems the messages are sent by the dhcp server but the 
>>client cannot receive it? Is there a bug in the current ISC dhcpd that 
>>use non-standard vendor tag number for nisdomain or it is dropped by 
>>etherboot? I just want to make sure what's happening.... Thanks.
>>
>
>Are you using "option dhcp-parameter-request-list" in dhcpd.conf? If so
>is the nisdomain code included?
>
That means the client has the request the option not receiving it at fiirst?

>
>
>Are you sending nisdomain in a parameter request list from Etherboot?
>
>RFC2132 says...
>
>   9.8. Parameter Request List
>
>   This option is used by a DHCP client to request values for specified
>   configuration parameters.  The list of requested parameters is
>   specified as n octets, where each octet is a valid DHCP option code
>   as defined in this document.
>
>   The client MAY list the options in order of preference.  The DHCP
>   server is not required to return the options in the requested order,
>   but MUST try to insert the requested options in the order requested
>   by the client.
>
>   The code for this option is 55.  Its minimum length is 1.
>
>    Code   Len   Option Codes
>   +-----+-----+-----+-----+---
>   |  55 |  n  |  c1 |  c2 | ...
>   +-----+-----+-----+-----+---
>
>
>
>
>_______________________________________________
>Etherboot-developers mailing list
>Eth...@li...
>https://lists.sourceforge.net/lists/listinfo/etherboot-developers
>

Re: [Etherboot-developers] Cardbus (PCMCIA) and etherboot

[Marty C. <md...@th...>]

On 11/4/2001 5:20 PM Gregg C Levine dr...@wo... wrote:
>Hello from Gregg C Levine normally with Jedi Knight Computers
>Where are we right now, on the subject of etherboot, and Cardbus
>(PCMCIA)? On the homepage's entries on the things to do page, I see that
>it is noted. But has any progress been done? Anyone?

There definitely is work being done on PCMCIA.  Most of the work is being 
done by a fellow who my company is (meagerly) funding. I don't have an 
estimate when there will be something to test, but hopefully sometime 
before the end of the year.

There are so many interesting areas to work on with Etherboot, I only 
wish I had more hours in a week to work on them.  Things go in cycles.  
Personally I'm using LinuxWorld Expos in the US as a milestone, and 
hoping to have something interesting by then (assuming it is still 
on...).  The way the world is, I'm just happy to have made it through a 
year of supporting rom-o-matic.net.  I haven't even had a chance to 
announce that! :)

Anyway, it's nice to see that so many people are using Etherboot, and 
it's a great and fun project to work on.  It even makes winter (my least 
favorite season) something to look forward to, as I stay inside and hack.

Marty

---
    Try: http://rom-o-matic.net/ to make Etherboot images instantly.

   Name: Marty Connor
US Mail: Entity Cyber, Inc.; P.O. Box 391827; Cambridge, MA 02139; USA
  Voice: (617) 491-6935, Fax: (617) 491-7046 
  Email: md...@th...
    Web: http://www.thinguin.org/

[Etherboot-developers] Cardbus (PCMCIA) and etherboot

[Gregg C L. <dr...@wo...>]

Hello from Gregg C Levine normally with Jedi Knight Computers
Where are we right now, on the subject of etherboot, and Cardbus
(PCMCIA)? On the homepage's entries on the things to do page, I see that
it is noted. But has any progress been done? Anyone?
---------
Gregg C Levine dr...@wo...
----------------------------------------------------
"How many floors does this TARDIS of yours have, anyway?"

Re: [Etherboot-developers] Oversized code

[<ke...@us...>]

>When I try to put other stuff into etherboot, when compile it gives an 
>error of  code size exceed limit, how can I overcome this? This there a 
>physical limit of code size even I have a very large bootrom? Thanks.

If it's first32.c you're changing, there is very little room left for
additions. At the moment it has to fit into 4kB. You can only get more
room by shuffling the segments around.