thanks,Richard!
>Yes some PCI cards work. I've made several Assiliant 65550 and
>69000
>based PCI cards work and some S3 based cards work. There are
>reports of
>some ATI cards working but I am still having problems getting my ATI
>M1
>based card to work.
My card is ATI,too.:(
Maybe I need to look for another PCI card to try.
>
>Are you setting up the shadowing correctly? You have to setup the
>shadowing in loader.s or bochs and the video bios never make it into
>ram. The stock ADLO setup won't work on the 440bx.
>
>I've attached my loader.s which sets up the shadowing correctly.
>
Thanks,the shadow setting is correct ,and I debugged into rombios.c ,found
that it halts when it run into vga bios and never return back.
_________________________________________________________________
免费下载 MSN Explorer: http://explorer.msn.com/lccn
Has anyone pursued this feature of the i8xx series chipsets? Seems it
could do what the Bios saviour does, even for boards with a soldered on
flash chip.
At least for the 810e, there is a "top block swap" feature:
The FWH flash part is a 512kB (or more) part, of 8x (or more) 64kB
blocks. The top block can be write protected separately via a jumper.
The other blocks can also be collectively write protected by a jumper.
The ICH (like a southbridge I guess) contains the CMOS batter-backed
ram. One extra bit in CMOS can be set in the PCI config space of the
ICH. The "top block swap" can be used for failsafe top or "boot" block
updates to the FWH flash.
The update utility first copies the top block to the block just below
it, and verifies the flash operation. It then sets the top block swap
bit. From this point on, if a flash operation to the top block fails,
the power is lost, etc, the system will boot to the 2nd from top block,
which presumably contains enough code to load the flash utility from
some device (floppy/cdrom?) Once the update has completed successfully,
the top block swap is disabled.
I have verified this using setpci. On an IBM NetVista 6341-23U, it
works in that the system fails to boot upon setting the bit, and
clearing the CMOS restores the system to normal.
The chipset docs also advertise a power-on strapping option for the ICH,
but the system I tested had neither that or the write protect jumpers.
( It did have a recovery jumper, but that seems to just activate
recovery mode in the top block's boot block code, not force a top block
swap)
I'm going to try using this as a way of trying out Linuxbios with a way
to get back the vendor bios (the FWH is soldered on); flash linuxbios
to the 2nd from top block, set the swap bit. If it breaks, clear CMOS,
recover the vendor bios.
BTW - does anyone have any working code to flash an ST M50FW0404 or an
Intel 82802? Uniflash/devbios don't support the ST at least.
--
Jeremy Jackson
Coplanar Networks
(519)897-1516
http://www.coplanar.net
Hi,
I've built my linuxbios.rom for Iwill's dk8s2. the only change i made
was to the ROM_SIZE (i needed it to be 512k, not 1mb).
i've plugged in the prom and a serial console set to 115200 bps. i don't
actually see anything.
what sort of output should i be seeing? i've tried turning up the debug
level, but still see no output.
does anyone have any experience with this board?
thanks,
peter
I'll try to find it.
The other change I made was to turn it into a real device (which they may
also have done) and have it not start sending output until the first open.
This change was needed to ensure it would not try to use the network stack
until it was actually there ...
ron
How to get netconsole to work:
1. Build your 2.6 kernel with tg3 support AND netconsole as part of the
kernel
A. This is the easiest way, I couldn't do it via modules as I
couldn't get the polling to work.
2. Pass in to the kernel as a parameter the netconsole information
A. This will automatically send the console information to the
destination address when eth0 is up.
Doing this worked for me, but a little bit of a pain was involved. Hope
this helped.
Dave
-----Original Message-----
From: linuxbios-admin(a)clustermatic.org
[mailto:linuxbios-admin@clustermatic.org] On Behalf Of Dave Aubin
Sent: Wednesday, September 01, 2004 3:12 PM
To: ron minnich
Cc: linuxbios(a)clustermatic.org
Subject: RE: Redirect your console to the network
Hey Ron,
If your tool doesn't have to support polling can you please pass that
on to me?
That's one side effect of the netconsole tool. You need to have polling
enabled in the Network driver. I've been having trouble getting that to
fly actually.
Thanks,
Dave
-----Original Message-----
From: ron minnich [mailto:rminnich@lanl.gov]
Sent: Wednesday, September 01, 2004 11:08 AM
To: Dave Aubin
Cc: linuxbios(a)clustermatic.org
Subject: Re: Redirect your console to the network
yes, we have used that off and on for a while. I hope they improved it
to be a real module, but if not I have an iproved version here ...
thanks for the pointer.
ron
_______________________________________________
Linuxbios mailing list
Linuxbios(a)clustermatic.org
http://www.clustermatic.org/mailman/listinfo/linuxbios
Hey Ron,
If your tool doesn't have to support polling can you please pass that
on to me?
That's one side effect of the netconsole tool. You need to have polling
enabled in the
Network driver. I've been having trouble getting that to fly actually.
Thanks,
Dave
-----Original Message-----
From: ron minnich [mailto:rminnich@lanl.gov]
Sent: Wednesday, September 01, 2004 11:08 AM
To: Dave Aubin
Cc: linuxbios(a)clustermatic.org
Subject: Re: Redirect your console to the network
yes, we have used that off and on for a while. I hope they improved it
to be a real module, but if not I have an iproved version here ...
thanks for the pointer.
ron
Hi folks,
I haven't been paying attention recently to the list in regards to amd64
support, but I was wondering if the Tyan Thunder K8S Pro (S2882) is yet
supported? I tried to check the status page, but there isn't a single
2004 date on it. We have 8 of these boards coming in for a webserver
cluster, and just need to boot off of IDE (actually, CF via IDE). I know
someone around here works on Tyan boards, so I figure this is the
quickest way to an answer. :)
Thanks,
Justin
yes, we have used that off and on for a while. I hope they improved it to
be a real module, but if not I have an iproved version here ...
thanks for the pointer.
ron
>From: Richard Smith <rsmith(a)bitworks.com>
>To: snow2moutain(a)hotmail.com
>CC: linuxbios(a)clustermatic.org
>Subject: Re: VGA Problem about 440bx chipset
>Date: Tue, 31 Aug 2004 10:33:35 -0500
>
>Chiu Gerald wrote:
>
> > I found the work done about vga are:
> > (1) allocate_vga_resource() in newpci.c
> > enbale io/mem of the vga adapter
> > then walk up the bridges setting the VGA enable
>
>That should all be done already.
>
> > (2) copy the vga bios to 0xc0000 using adlo,and run bochs bios.
> > I'm not sure if it had initilized the vga controller enough. And I
>
>You will get a vga bios boot screen if it works.
>
> > found there are some registers about AGP from 440bx northbridge
> > datasheet ( 82443bx hostbridge/controller),such as AGP capability
> > identifier register,AGP command register,AGP status register...
> > But this work hadn't been done in linuxbios,maybe I need to set these
> > registers correctly according to the datasheet to enable AGP?
>
>This is probally your problem. None of the AGP stuff is setup in the
>440bx code. Only PCI cards have been used to date. You will have to
>write AGP init code.
you mean that PCI card works?
Now I changed to a PCI video card,and bochs had run the vga bios,but still
nothing happened!
I have no idea what should I do next.
_________________________________________________________________
免费下载 MSN Explorer: http://explorer.msn.com/lccn/
