* Re: 43p-140 install issues
From: Ulrich Teichert @ 2005-01-03 22:11 UTC (permalink / raw)
To: Leigh Brown; +Cc: Ulrich Teichert, linuxppc-dev, mikemartin, debian-powerpc
In-Reply-To: <33721.82.10.231.190.1104781370.squirrel@82.10.231.190>
Hi,
>Delete the "console=tty0", so you end up with:
>
> Linux/PPC load: console=ttyS0,9600 root=/dev/ram
>
>and try again!
[del]
Right, then the installer knows what to do. I think this should go into
the Debian installation docs, there are a lot of boxes with graphic cards
out which are not supported by Linux.
But the low level SCSI-driver is missing in this image, or any low level
SCSI drivers for that matter. There are only the high level modules there,
like sd_mod.ko and sr_mod.ko:
Nov 30 00:06:32 hw-detect: Using discover version 1.
Nov 30 00:06:35 hw-detect: Detecting hardware...
Nov 30 00:06:36 hw-detect: Missing module 'sym53c8xx'.
Nov 30 00:06:36 hw-detect: Missing module 'usb-storage'.
Nov 30 00:06:36 hw-detect: Missing module 'ide-mod'.
Nov 30 00:06:36 hw-detect: Missing module 'ide-probe-mod'.
Nov 30 00:06:36 hw-detect: Missing module 'ide-detect'.
Nov 30 00:06:36 hw-detect: Missing module 'ide-generic'.
Nov 30 00:06:37 hw-detect: Missing module 'ide-floppy'.
--More-- (57% of 44385 bytes) Nov 30 00:06:37 hw-detect: Missing module 'ide-disk'.
Nov 30 00:06:37 hw-detect: Missing module 'isofs'.
Nov 30 00:06:37 hw-detect: Loading modules...
Nov 30 00:06:37 hw-detect: Detected module 'pcnet32' for 'Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE]'
Nov 30 00:06:37 hw-detect: Trying to load module 'pcnet32'
Nov 30 00:06:37 kernel: pcnet32.c:v1.30i 06.28.2004 tsbogend@alpha.franken.de
Nov 30 00:06:37 kernel: PCI: Enabling device 0000:00:0c.0 (0000 -> 0003)
Nov 30 00:06:37 kernel: pcnet32: PCnet/PCI II 79C970A at 0x1020, 00 06 29 2e 98 9d assigned IRQ 22.
Nov 30 00:06:37 kernel: divert: allocating divert_blk for eth0
Nov 30 00:06:37 kernel: eth0: registered as PCnet/PCI II 79C970A
Nov 30 00:06:37 kernel: pcnet32: 1 cards_found.
Nov 30 00:06:38 hw-detect: Detected module 'floppy' for 'Linux Floppy'
Nov 30 00:06:38 hw-detect: Trying to load module 'floppy'
Nov 30 00:06:38 kernel: inserting floppy driver for 2.6.8-powerpc
Nov 30 00:06:38 kernel: Using anticipatory io scheduler
Nov 30 00:06:38 kernel: Floppy drive(s): fd0 is 2.88M
Nov 30 00:06:38 kernel: FDC 0 is a National Semiconductor PC87306
Nov 30 00:06:38 hw-detect: Detected module 'ide-cd' for 'Linux ATAPI CD-ROM'
Nov 30 00:06:38 hw-detect: Trying to load module 'ide-cd'
Nov 30 00:06:39 kernel: Linux Kernel Card Services
Nov 30 00:06:39 kernel: options: [pci] [cardbus] [pm]
Nov 30 00:06:39 hw-detect: Starting PCMCIA services: using yenta_socket instead of i82365
Nov 30 00:06:39 hw-detect: cardmgr[1070]: no sockets found!
Nov 30 00:06:39 cardmgr[1070]: no sockets found!
Nov 30 00:06:39 hw-detect: done.
Nov 30 00:06:39 hw-detect: Detected discover version 1, installing discover1.
Nov 30 00:06:39 hw-detect: Detected hotplug support, installing hotplug.
Nov 30 00:06:40 hw-detect: Missing modules 'sym53c8xx (Symbios Logic Inc. / NCR 53c825), usb-storage (USB storage), ide-mod (Linux IDE driver), ide-probe-mod (Linux IDE probe driver), ide-detect (Linux IDE detection), ide-generic (Linux IDE support), ide-floppy (Linux IDE floppy), ide-disk (Linux ATA DISK), isofs (Linux ISO 9660 filesystem)
/lib/modules/2.6.8-powerpc/kernel/drivers/scsi # ls
scsi_mod.ko sd_mod.ko sr_mod.ko
So, either this is the wrong image (I've taken the same as Daniel mentioned
in one of his mails: http://debian.yorku.ca/debian/dists/testing/main/installer-powerpc/current/images/powerpc/netboot/vmlinuz-prep.initrd),
or it's missing some vital modules.
HTH,
Uli
--
Dipl. Inf. Ulrich Teichert|e-mail: Ulrich.Teichert@gmx.de
Stormweg 24 |listening to: Suicide Drive (The Deep Eynde)
24539 Neumuenster, Germany|Public Pervert (Interpol) Clé De Contact (Metal Urbain)
^ permalink raw reply
* Re: 43p-140 install issues
From: Leigh Brown @ 2005-01-03 19:42 UTC (permalink / raw)
To: Ulrich Teichert; +Cc: linuxppc-dev, mikemartin, debian-powerpc
In-Reply-To: <200501031932.j03JWC4E013520@arbas.nms.ulrich-teichert.org>
Ulrich Teichert said:
[...]
> Linux/PPC load: console=ttyS0,9600 console=tty0 root=/dev/ram
Delete the "console=tty0", so you end up with:
Linux/PPC load: console=ttyS0,9600 root=/dev/ram
and try again!
[...]
>>It should never ask the console question again unless you change the
>>config (plug in a mouse for example), unless your nvram battery is a
>>bit flat of course!
>
> This seems to be the case. But it really simplifies things, as I've never
> been able to warm-boot anything ;-) Only a cold boot helps. Maybe Daniel
> should remove the battery to reset things completely?
>
> Anyway, without your site, Leigh, I would still have a big doorstop only,
> thanks again,
No problem. The amazing thing is, once you actually get them running,
RS/6000's run Linux really reliably. It's getting them booting that's
the problem :-/
^ permalink raw reply
* Re: 43p-140 install issues
From: Ulrich Teichert @ 2005-01-03 19:32 UTC (permalink / raw)
To: Leigh Brown; +Cc: Ulrich Teichert, linuxppc-dev, mikemartin, debian-powerpc
In-Reply-To: <33304.82.10.231.190.1104771000.squirrel@82.10.231.190>
Hi,
[del]
>> Linux/PPC load: console=ttyS0,9600 console=tty0
>
>*** This is the point you get 5 seconds to start editing the parameters.
>I guess if you specify root=/dev/ram0 or something then it will boot
>from the ramdisk, rather than trying to boot from /dev/sda3.
Right, now I can even boot over network:
0 > setenv load-base 1000000 ok
0 > setenv real-base c00000 ok
0 > boot net:
LOAD: Waiting 60 seconds for Spanning Tree
BOOTP R = 1 BOOTP S = 1
FILE: /tftpboot/wehrle/vmlinuz-prep.initrd
Load Addr=0x1000000 Max Size=0xf000000
Packet Count = 100 [snipped the packet count lines]
loaded at: 01000400 01554FF4
relocated to: 00800000 00D54BF4
zimage at: 0080A94C 0093DC0A
initrd at: 0093E000 00D4AB7A
avail ram: 00400000 00800000
Linux/PPC load: console=ttyS0,9600 console=tty0 root=/dev/ram
Uncompressing Linux...done.
Now booting the kernel
Total memory = 256MB; using 512kB for hash table (at c0300000)
Linux version 2.6.8-powerpc (jens@yorick) (gcc version 3.3.4 (Debian 1:3.3.4-11)) #1 Sun Oct 3 13:22:21 CEST 2004
PReP architecture
IBM planar ID: 000000d5
MPIC at 0xfddc0000 (0x3ddc0000), length 0x00040000 mapped to 0xeffc0000
Built 1 zonelists
Kernel command line: console=ttyS0,9600 console=tty0 root=/dev/ram
OpenPIC Version 1.0 (4 CPUs and 16 IRQ sources) at effc0000
PID hash table entries: 2048 (order 11: 16384 bytes)
time_init: decrementer frequency = 16.618701 MHz
Console: colour dummy device 80x25
Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
Memory: 251648k available (1732k kernel code, 1060k data, 164k init, 0k highmem)
Calibrating delay loop... 397.31 BogoMIPS
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
checking if image is initramfs...it isn't (no cpio magic); looks like an initrd
Freeing initrd memory: 4146k freed
NET: Registered protocol family 16
PCI: Probing PCI hardware
Setting PCI interrupts for a "IBM 43P-140 (Tiger1)"
PCI: Cannot allocate resource region 0 of device 0000:00:0b.0
PCI: Cannot allocate resource region 0 of device 0000:00:0c.0
PCI: Cannot allocate resource region 0 of device 0000:00:10.0
PCI: Cannot allocate resource region 0 of device 0000:00:12.0
Thermal assist unit not available
audit: initializing netlink socket (disabled)
audit(943920304.671:0): initialized
devfs: 2004-01-31 Richard Gooch (rgooch@atnf.csiro.au)
devfs: boot_options: 0x0
Initializing Cryptographic API
Generic RTC Driver v1.07
Macintosh non-volatile memory driver v1.1
Serial: 8250/16550 driver $Revision: 1.90 $ 8 ports, IRQ sharing disabled
ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
pmac_zilog: 0.6 (Benjamin Herrenschmidt <benh@kernel.crashing.org>)
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
input: Macintosh mouse button emulation
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
mice: PS/2 mouse device common for all mice
NET: Registered protocol family 2
IP: routing cache hash table of 2048 buckets, 16Kbytes
TCP: Hash tables configured (established 16384 bind 32768)
RAMDISK: Compressed image found at block 0
VFS: Mounted root (cramfs filesystem) readonly.
Freeing unused kernel memory: 164k init 60k pmac 4k chrp 8k openfirmware
NET: Registered protocol family 1
And that's it. Seems like the d-i doesn't know that it should use the
serial console. BTW, the instructions for booting PREP systems in the
debian installation manual aren't quite right, they are missing the
"net:" string. If the TFTP-server is setup right, the firmware can figure
out the rest, so giving the server IP, the name and the IP should not
be necessary and I would try `boot net:` first.
[del]
>It should never ask the console question again unless you change the
>config (plug in a mouse for example), unless your nvram battery is a
>bit flat of course!
This seems to be the case. But it really simplifies things, as I've never
been able to warm-boot anything ;-) Only a cold boot helps. Maybe Daniel
should remove the battery to reset things completely?
Anyway, without your site, Leigh, I would still have a big doorstop only,
thanks again,
CU,
Uli
--
Dipl. Inf. Ulrich Teichert|e-mail: Ulrich.Teichert@gmx.de
Stormweg 24 |listening to: Suicide Drive (The Deep Eynde)
24539 Neumuenster, Germany|Public Pervert (Interpol) Clé De Contact (Metal Urbain)
^ permalink raw reply
* Re: 43p-140 install issues
From: vinai @ 2005-01-03 15:47 UTC (permalink / raw)
To: debian-powerpc; +Cc: linuxppc-dev
In-Reply-To: <855e88d005010306435e08c827@mail.gmail.com>
>>>> An IBM 7043. It's a 43p-140 labeled as a 604e 332 Mhz.
>>
>> I suppose you meen 233 Mhz ?
>
> I know 332 MHz doesn't sound right ... but I'm sure that's what it
> says on the IBM nameplate on the front. I'll verify this tonight
According to this page:
http://www.sinca.biz/IBM/rs6000_43p_140.html
the 7043-140 does come in 2 flavours, a 233 and 332 MHz model. I've
been looking into getting my own 150 to play with :)
cheers
vinai
^ permalink raw reply
* Re: PATCH uninorth3 (G5) agp support
From: Jon Loeliger @ 2005-01-03 18:56 UTC (permalink / raw)
To: Jerome Glisse; +Cc: linuxppc64-dev, linuxppc-dev list
In-Reply-To: <41D00564.6010507@free.fr>
On Mon, 2004-12-27 at 06:51, Jerome Glisse wrote:
> /* My understanding of UniNorth AGP as of UniNorth rev 1.0x,
> * revision 1.5 (x4 AGP) may need further changes.
> diff -Naur linux/include/linux/pci_ids.h linux-new/include/linux/pci_ids.h
> --- linux/include/linux/pci_ids.h 2004-12-26 14:40:05.000000000 +0100
> +++ linux-new/include/linux/pci_ids.h 2004-12-27 13:40:50.121003792 +0100
> @@ -842,6 +842,7 @@
> #define PCI_DEVICE_ID_APPLE_UNI_N_GMAC2 0x0032
> #define PCI_DEVIEC_ID_APPLE_UNI_N_ATA 0x0033
> #define PCI_DEVICE_ID_APPLE_UNI_N_AGP2 0x0034
> +#define PCI_DEVICE_ID_APPLE_U3_AGP 0x0059
> #define PCI_DEVICE_ID_APPLE_IPID_ATA100 0x003b
> #define PCI_DEVICE_ID_APPLE_KEYLARGO_I 0x003e
> #define PCI_DEVICE_ID_APPLE_K2_ATA100 0x0043
So, did 0x0033's symbol need to be spelled consistently too?
NB: PCI_DEVIEC_
Thanks,
jdl
^ permalink raw reply
* Re: 43p-140 install issues
From: Leigh Brown @ 2005-01-03 16:50 UTC (permalink / raw)
To: Ulrich Teichert; +Cc: Ulrich Teichert, linuxppc-dev, mikemartin, debian-powerpc
In-Reply-To: <200501031551.j03FpLAn009525@arbas.nms.ulrich-teichert.org>
Ulrich Teichert said:
>>[snip]
>>> There may be some problems with 2.6 kernels and pci irqs on prep. I
>>> will
>>> investigate this shortly.
>>
>>There is definitely something strange happening. The image I used to
>>do the (nearly) complete install will no longer configure the network.
>>Very inconsistent. I will try it (same image) again later today and
>>see if it "just works" next time - like it did last time.
>
> Yes, I've seen inconsistencies as well, like booting the default kernel
> from a sarge CD worked *once* and rebooting not. I found out why:
> during the first cold boot, the firmware asks if it should make my
> serial console the active console (which I acknowledged), but with a
> warm boot, this question is not asked again.
That's because the firmware will only ask that question if detects that
a mouse or keyboard is present/not present where it previously was/n't.
> This seems to result in a different setting and the kernel is not able
> to figure out that it is running on a serial console. According to Leigh
> Browns, it is not possible to give the kernel parameters on boot-up,
> unless you are using his patches, so this may be another issue.
I have a patch you can use to specify parameters on the *command line*
(e.g. boot floppy: root=/dev/sda3 ...). However, you can always type
parameters on the serial console as it boots (see below).
> But this is how far the default kernel (which is all I can load from CD)
> boots:
>
> 0 > setenv load-base 1000000 ok
> 0 > setenv real-base c00000 ok
> 0 > boot cdrom:
>
> loaded at: 01000400 01492FF4
> relocated to: 00800000 00C92BF4
> zimage at: 0080A94C 0093E3DC
> initrd at: 0093F000 00C884AE
> avail ram: 00400000 00800000
>
> Linux/PPC load: console=ttyS0,9600 console=tty0
*** This is the point you get 5 seconds to start editing the parameters.
I guess if you specify root=/dev/ram0 or something then it will boot
from the ramdisk, rather than trying to boot from /dev/sda3.
> Uncompressing Linux...done.
> Now booting the kernel
> Total memory = 256MB; using 512kB for hash table (at c0300000)
> Linux version 2.6.8-powerpc (sven@pegasos2) (version gcc 3.3.5 (Debian
> 1:3.3.5-2)) #1 Thu Dec 9 10:27:54 CET 2004
> PReP architecture
[...]
> VFS: Cannot open root device "<NULL>" or unknown-block(8,3)
> Please append a correct "root=" boot option
> Kernel panic: VFS: Unable to mount root fs on unknown-block(8,3)
> <0>Rebooting in 180 seconds..
Other than that, it's a nice clean boot.
> [del]
>>> Then i suppose that copying the kernel is just a matter of :
>>>
>>> dd if=/boot/vmlinuz-2.6.8-powerpc of=/dev/sdb1
>>>
>>> but you have to have a small prep partition as /dev/sdb1, or the above
>>> will
>>> erase whatever you have on /dev/sdb1.
>
> This is not enough. You need to set the kernel arguments with Leigh's
> preptool (which is a simple perl script), because of the problems
> mentioned before.
Well, you need to have my patch for that. There are other ways to
manually change the boot arguments (like hexedit etc).
>>I' have the partition ready ...
>>
>>Unfortunately, until the network comes back up, I can't get far enough
>>in the install to access the SCSI drives (is there another way? Load
>>the modules by hand? Which ones?). Without this, I can't get at the
>>image I need to boot in order to get it onto the right partition. I
>>also can't get it to another machine to net boot it.
>
> All this initrd thing is a mess when it gets complicated, if you ask me.
> I really don't know why your floppy stops working after booting Leighs
> kernel. But on the other hand, I don't know why I can't netboot my box
> either - I can netboot anything else from DECstations to SGIs with the
> same setup, but my 7043-44 refuses to load the kernel. It contacts the
> server, figures out the name of the file, but does not request it.
>
> For your network trouble, try to disconnect power from the box completely
> for an hour or so. This was enough for my box to fall back to asking me
> for the active console, maybe this works for your flaky network as well?
It should never ask the console question again unless you change the
config (plug in a mouse for example), unless your nvram battery is a
bit flat of course!
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 17:28 UTC (permalink / raw)
To: Leigh Brown; +Cc: Ulrich Teichert, linuxppc-dev, mikemartin, debian-powerpc
In-Reply-To: <33304.82.10.231.190.1104771000.squirrel@82.10.231.190>
On Mon, Jan 03, 2005 at 04:50:00PM -0000, Leigh Brown wrote:
> Ulrich Teichert said:
> > This seems to result in a different setting and the kernel is not able
> > to figure out that it is running on a serial console. According to Leigh
> > Browns, it is not possible to give the kernel parameters on boot-up,
> > unless you are using his patches, so this may be another issue.
>
> I have a patch you can use to specify parameters on the *command line*
> (e.g. boot floppy: root=/dev/sda3 ...). However, you can always type
> parameters on the serial console as it boots (see below).
Mmm, is this one already present in the debian 2.6.8 kernel packages ? If not
we would probably want to include it. I am currently preparing a bunch of prep
patches, so it is a good idea to include this one too. How is it related to
the below command line prompt thingy ...
> > But this is how far the default kernel (which is all I can load from CD)
> > boots:
> >
> > 0 > setenv load-base 1000000 ok
> > 0 > setenv real-base c00000 ok
> > 0 > boot cdrom:
> >
> > loaded at: 01000400 01492FF4
> > relocated to: 00800000 00C92BF4
> > zimage at: 0080A94C 0093E3DC
> > initrd at: 0093F000 00C884AE
> > avail ram: 00400000 00800000
> >
> > Linux/PPC load: console=ttyS0,9600 console=tty0
... This one i mean.
> *** This is the point you get 5 seconds to start editing the parameters.
> I guess if you specify root=/dev/ram0 or something then it will boot
> from the ramdisk, rather than trying to boot from /dev/sda3.
d-i is perfectly happy to just boot without any argument.
> > Uncompressing Linux...done.
> > Now booting the kernel
> > Total memory = 256MB; using 512kB for hash table (at c0300000)
> > Linux version 2.6.8-powerpc (sven@pegasos2) (version gcc 3.3.5 (Debian
> > 1:3.3.5-2)) #1 Thu Dec 9 10:27:54 CET 2004
> > PReP architecture
> [...]
> > VFS: Cannot open root device "<NULL>" or unknown-block(8,3)
> > Please append a correct "root=" boot option
> > Kernel panic: VFS: Unable to mount root fs on unknown-block(8,3)
> > <0>Rebooting in 180 seconds..
>
> Other than that, it's a nice clean boot.
Indeed.
> > [del]
> >>> Then i suppose that copying the kernel is just a matter of :
> >>>
> >>> dd if=/boot/vmlinuz-2.6.8-powerpc of=/dev/sdb1
> >>>
> >>> but you have to have a small prep partition as /dev/sdb1, or the above
> >>> will
> >>> erase whatever you have on /dev/sdb1.
> >
> > This is not enough. You need to set the kernel arguments with Leigh's
> > preptool (which is a simple perl script), because of the problems
> > mentioned before.
>
> Well, you need to have my patch for that. There are other ways to
> manually change the boot arguments (like hexedit etc).
Could you post it ? Anything else you are aware that we should add in our
2.6.8 powerpc kernels which is PReP related ?
Friendly,
Sven Luther
^ permalink raw reply
* Re: MPC5200 PCI byte-swapping
From: Sylvain Munaut @ 2005-01-03 16:47 UTC (permalink / raw)
To: Mark Chambers; +Cc: linuxppc-embedded
In-Reply-To: <003c01c4f19d$98a16510$0301a8c0@chuck2>
>Thanks. I had an older manual that didn't spell it out so
>clearly.
>
>I've been trying to interpret the PCI sections for some other
>82xx family parts, and it appears that they do NOT do this
>byte lane swapping, so this make the 5200 non-standard in
>this regard, which is unfortunate. If I'm understanding this
>right, one would have to have different drivers for a PCI
>device on a 5200 and an 8270, for instance.
>
>
Mmmm ...
Still I use the intel eepro driver without problems or modifications.
As long as the driver uses the proper readl/writel that should do
it or am I mistaken ?
I have a FPGA mounted with pci interface, I'll try to see what happens on
the bus
>Also, I note that when doing simple block reads from pre-
>fetchable PCI space, it appears the 5200 does not prefetch,
>but does each read individually. This is using stock ELDK
>u-boot and 2.4.24 so I haven't yet determined if it's a
>configuration matter, (or ruled out target disconnects)
>but I'm suspecting that you can't get burst mode from the
>5200 without using DMA.
>
>
Last time I checked, the 2.4 from denx didn't create a pci
window for prefetchable memory so prefetch mem zone
were mapped as non-prefetchable, so no burst for sure.
Sylvain
^ permalink raw reply
* Re: Making a serial multiplexer (new driver)
From: Geert Uytterhoeven @ 2005-01-03 16:17 UTC (permalink / raw)
To: Luís Vitório Cargnini; +Cc: linuxppc-embedded
In-Reply-To: <1104758458.14925.5.camel@vitorio>
[-- Attachment #1: Type: TEXT/PLAIN, Size: 930 bytes --]
On Mon, 3 Jan 2005, [ISO-8859-1] Luís Vitório Cargnini wrote:
> Hi i need to write a driver that must multiplex an serial channel my
> driver will have 3 interfaces and i must communicate with another
> interface (midleware) like /dev/ttyS2 for example, how could i write a
> driver that will communicate with my existing uart ???
>
> i have a component that could be multiplexed in 3 logical channels throw
> my single physical channel that are accessed with a normal serial
> interface using uart, how could this be possible ?
> this is necessary to be transparent to software user level.
Do it in userspace using fifos?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 16:04 UTC (permalink / raw)
To: Michael Schmitz
Cc: Ulrich Teichert, Philippe Guyot, Mike Martin, linuxppc-dev,
debian-powerpc, mikemartin
In-Reply-To: <Pine.LNX.4.44.0501031545370.11702-100000@opal.biophys.uni-duesseldorf.de>
On Mon, Jan 03, 2005 at 04:24:40PM +0100, Michael Schmitz wrote:
> > > > > http://www.solinno.co.uk/7043-140/downloads.php
> > > > > But ask Leigh for details.
> > > >
> > > > Ok, thanks. ...
> > >
> > > Do we need that added to powerpc-utils perhaps?
> >
> > Whatever is powerpc-utils ? Mmm, maybe, but then maybe mkvmlinuz would benefit
>
> WTF is mkvmlinuz?
mkvmlinuz is a nice little kernel-derived tool by Jens and me, which allows
one to create a zImage.chrp, or zImage.coff or whatever, from the
arch/ppc/boot/* stuff and the plain vmlinux kernel. It avoids having to chip
the same kernel X times as we did on 2.4 kernels, and can be used as a
kernel-package postinst hook to automate this process.
> > from being added there too, not sure. In any way i believe it has more its
> > place inside mkvmlinuz, than inside the more userland powerpc-utils.
>
> Whatever. Just thought people might want to use it to change kernel
> options on the fly, without invoking mkvmlinuz.
Well, we would need at least a .udeb containing it. IF you can add it to
powerpc-tools, then nice. I think that maybe mkvmlinuz would be also a
candidate to be part of the powerpc-tools, not sure though.
Friendly,
Sven Luther
^ permalink raw reply
* Re: 43p-140 install issues
From: Ulrich Teichert @ 2005-01-03 15:51 UTC (permalink / raw)
To: mikemartin; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <855e88d0050103053915c385ec@mail.gmail.com>
Hi,
>[snip]
>> There may be some problems with 2.6 kernels and pci irqs on prep. I will
>> investigate this shortly.
>
>There is definitely something strange happening. The image I used to
>do the (nearly) complete install will no longer configure the network.
>Very inconsistent. I will try it (same image) again later today and
>see if it "just works" next time - like it did last time.
Yes, I've seen inconsistencies as well, like booting the default kernel
from a sarge CD worked *once* and rebooting not. I found out why:
during the first cold boot, the firmware asks if it should make my
serial console the active console (which I acknowledged), but with a
warm boot, this question is not asked again.
This seems to result in a different setting and the kernel is not able
to figure out that it is running on a serial console. According to Leigh
Browns, it is not possible to give the kernel parameters on boot-up, unless
you are using his patches, so this may be another issue.
But this is how far the default kernel (which is all I can load from CD)
boots:
0 > setenv load-base 1000000 ok
0 > setenv real-base c00000 ok
0 > boot cdrom:
loaded at: 01000400 01492FF4
relocated to: 00800000 00C92BF4
zimage at: 0080A94C 0093E3DC
initrd at: 0093F000 00C884AE
avail ram: 00400000 00800000
Linux/PPC load: console=ttyS0,9600 console=tty0
Uncompressing Linux...done.
Now booting the kernel
Total memory = 256MB; using 512kB for hash table (at c0300000)
Linux version 2.6.8-powerpc (sven@pegasos2) (version gcc 3.3.5 (Debian 1:3.3.5-2)) #1 Thu Dec 9 10:27:54 CET 2004
PReP architecture
IBM planar ID: 000000d5
MPIC at 0xfddc0000 (0x3ddc0000), length 0x00040000 mapped to 0xeffc0000
Built 1 zonelists
Kernel command line: console=ttyS0,9600 console=tty0
OpenPIC Version 1.0 (4 CPUs and 16 IRQ sources) at effc0000
PID hash table entries: 2048 (order 11: 16384 bytes)
time_init: decrementer frequency = 16.618715 MHz
Console: colour dummy device 80x25
Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
Memory: 252416k available (1732k kernel code, 1060k data, 164k init, 0k highmem)
Calibrating delay loop... 397.31 BogoMIPS
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
checking if image is initramfs...it isn't (no cpio magic); looks like an initrd
Freeing initrd memory: 3365k freed
NET: Registered protocol family 16
PCI: Probing PCI hardware
Setting PCI interrupts for a "IBM 43P-140 (Tiger1)"
PCI: Cannot allocate resource region 0 of device 0000:00:0b.0
PCI: Cannot allocate resource region 0 of device 0000:00:0c.0
PCI: Cannot allocate resource region 0 of device 0000:00:10.0
PCI: Cannot allocate resource region 0 of device 0000:00:12.0
Thermal assist unit not available
audit: initializing netlink socket (disabled)
audit(943920606.514:0): initialized
devfs: 2004-01-31 Richard Gooch (rgooch@atnf.csiro.au)
devfs: boot_options: 0x0
Initializing Cryptographic API
Generic RTC Driver v1.07
Macintosh non-volatile memory driver v1.1
Serial: 8250/16550 driver $Revision: 1.90 $ 8 ports, IRQ sharing disabled
ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
pmac_zilog: 0.6 (Benjamin Herrenschmidt <benh@kernel.crashing.org>)
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
input: Macintosh mouse button emulation
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
mice: PS/2 mouse device common for all mice
NET: Registered protocol family 2
IP: routing cache hash table of 2048 buckets, 16Kbytes
TCP: Hash tables configured (established 16384 bind 32768)
RAMDISK: Compressed image found at block 0
VFS: Mounted root (cramfs filesystem) readonly.
VFS: Cannot open root device "<NULL>" or unknown-block(8,3)
Please append a correct "root=" boot option
Kernel panic: VFS: Unable to mount root fs on unknown-block(8,3)
<0>Rebooting in 180 seconds..
[del]
>> Then i suppose that copying the kernel is just a matter of :
>>
>> dd if=/boot/vmlinuz-2.6.8-powerpc of=/dev/sdb1
>>
>> but you have to have a small prep partition as /dev/sdb1, or the above will
>> erase whatever you have on /dev/sdb1.
This is not enough. You need to set the kernel arguments with Leigh's
preptool (which is a simple perl script), because of the problems mentioned
before.
>I' have the partition ready ...
>
>Unfortunately, until the network comes back up, I can't get far enough
>in the install to access the SCSI drives (is there another way? Load
>the modules by hand? Which ones?). Without this, I can't get at the
>image I need to boot in order to get it onto the right partition. I
>also can't get it to another machine to net boot it.
All this initrd thing is a mess when it gets complicated, if you ask me.
I really don't know why your floppy stops working after booting Leighs
kernel. But on the other hand, I don't know why I can't netboot my box
either - I can netboot anything else from DECstations to SGIs with the
same setup, but my 7043-44 refuses to load the kernel. It contacts the
server, figures out the name of the file, but does not request it.
For your network trouble, try to disconnect power from the box completely
for an hour or so. This was enough for my box to fall back to asking me
for the active console, maybe this works for your flaky network as well?
HTH,
Uli
--
Dipl. Inf. Ulrich Teichert|e-mail: Ulrich.Teichert@gmx.de
Stormweg 24 |listening to: Suicide Drive (The Deep Eynde)
24539 Neumuenster, Germany|Public Pervert (Interpol) Clé De Contact (Metal Urbain)
^ permalink raw reply
* Re: 43p-140 install issues
From: Michael Schmitz @ 2005-01-03 15:24 UTC (permalink / raw)
To: Sven Luther
Cc: Ulrich Teichert, Philippe Guyot, Michael Schmitz, Mike Martin,
linuxppc-dev, debian-powerpc, mikemartin
In-Reply-To: <20050103124716.GB9388@pegasos>
> > > > http://www.solinno.co.uk/7043-140/downloads.php
> > > > But ask Leigh for details.
> > >
> > > Ok, thanks. ...
> >
> > Do we need that added to powerpc-utils perhaps?
>
> Whatever is powerpc-utils ? Mmm, maybe, but then maybe mkvmlinuz would benefit
WTF is mkvmlinuz?
> from being added there too, not sure. In any way i believe it has more its
> place inside mkvmlinuz, than inside the more userland powerpc-utils.
Whatever. Just thought people might want to use it to change kernel
options on the fly, without invoking mkvmlinuz.
Michael
^ permalink raw reply
* Making a serial multiplexer (new driver)
From: Luís Vitório Cargnini @ 2005-01-03 13:20 UTC (permalink / raw)
To: linuxppc-embedded
[-- Attachment #1: Type: text/plain, Size: 589 bytes --]
Hi i need to write a driver that must multiplex an serial channel my
driver will have 3 interfaces and i must communicate with another
interface (midleware) like /dev/ttyS2 for example, how could i write a
driver that will communicate with my existing uart ???
i have a component that could be multiplexed in 3 logical channels throw
my single physical channel that are accessed with a normal serial
interface using uart, how could this be possible ?
this is necessary to be transparent to software user level.
--
Luís Vitório Cargnini <vitorio@digitel.com.br>
Digitel S/A
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]
^ permalink raw reply
* Re: 43p-140 install issues
From: Mike Martin @ 2005-01-03 14:43 UTC (permalink / raw)
To: Philippe Guyot; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <200501031530.54487.pguyot@cvf.fr>
> > > An IBM 7043. It's a 43p-140 labeled as a 604e 332 Mhz.
>
> I suppose you meen 233 Mhz ?
I know 332 MHz doesn't sound right ... but I'm sure that's what it
says on the IBM nameplate on the front. I'll verify this tonight
> More different than that, 140 is more recent.
Have you got specs and stuff on the -140 you could link to? Maybe this
would help.
>
> But works fine with Debian woody.(sorry Sven...)
[snip]
> If there is such a deadline, why not consider to install first a woody (many
> successes exist) then upgrading to sarge after ?
This is my preference. I'd like to install woody on one drive so I can
boot it up and say "Look Linux runs, it's stable and useful" and then
have the other to install, boot, and otherwise help test the new
stuff.
Unfortunately, I can't get _anything_ to install (I even tried a BSD,
but quickly realized I was in too deep there!). The unstable stuff I'm
trying now has gotten me the furthest along.
MikeM
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 14:45 UTC (permalink / raw)
To: Philippe Guyot; +Cc: Ulrich Teichert, linuxppc-dev, mikemartin, debian-powerpc
In-Reply-To: <200501031530.54487.pguyot@cvf.fr>
On Mon, Jan 03, 2005 at 03:30:54PM +0100, Philippe Guyot wrote:
> On Monday 03 January 2005 15:30, Sven Luther wrote:
> > On Mon, Jan 03, 2005 at 09:07:31AM -0500, Mike Martin wrote:
> > > On Mon, 3 Jan 2005 15:01:02 +0100, Sven Luther <sven.luther@wanadoo.fr>
> > > wrote:
> > >
> > > [snip]
> > >
> > > > What exact machine do you have again ?
> > >
> > > An IBM 7043. It's a 43p-140 labeled as a 604e 332 Mhz.
>
>
> I suppose you meen 233 Mhz ?
> >
> > Well, prep_pci.c does know about :
> >
> > IBM RS/6000 7043-240 (ibm_doral)
> > IBM RS/6000 7024-E30
> >
> > As far as IBM boxes are concerned. Doe the -140 have stuff in common with
> > leigh's -240 (the -140 being mono-cpu while the -240 being dual ?).
>
> More different than that, 140 is more recent.
>
> But works fine with Debian woody.(sorry Sven...)
Yeah, that is not the problem, the problem is that it doesn't work with the
2.6.8 kernel used by debian/sarge. So, now, the first order of business to fix
this is to find out what exactly is broken in the 2.6.8 kernel on this
hardware, fish out patches which may or may not solve the issue, and get it
solved. I will soon be making a kernel-source-2.6.8 upload for the powerstack
II problem i have hardware for, so it would be nice to fix these issues at the
same time.
Once we have a working kernel, we will upload partman-prep from Cajus
Pollmeier, and add a prep-installer, and there will be propper prep support in
sarge, at least for the powerstack II.
> > Ok, understandable, still, getting it to netboot would be a first step to
> > getting it working right later.
> >
> > > If I can get it to boot the system (even from floppy or cd) I will
> > > have it to help test stuff. Otherwise it goes back under the desk
> > > where it came from.
>
> If there is such a deadline, why not consider to install first a woody (many
> successes exist) then upgrading to sarge after ?
That is always a solution. Getting the 2.6 debian kernel fixed would be nice
though.
Friendly,
Sven Luther
^ permalink raw reply
* Re: 43p-140 install issues
From: Philippe Guyot @ 2005-01-03 14:30 UTC (permalink / raw)
To: Sven Luther, mikemartin; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <20050103143015.GB9197@pegasos>
On Monday 03 January 2005 15:30, Sven Luther wrote:
> On Mon, Jan 03, 2005 at 09:07:31AM -0500, Mike Martin wrote:
> > On Mon, 3 Jan 2005 15:01:02 +0100, Sven Luther <sven.luther@wanadoo.fr>
> > wrote:
> >
> > [snip]
> >
> > > What exact machine do you have again ?
> >
> > An IBM 7043. It's a 43p-140 labeled as a 604e 332 Mhz.
I suppose you meen 233 Mhz ?
>
> Well, prep_pci.c does know about :
>
> IBM RS/6000 7043-240 (ibm_doral)
> IBM RS/6000 7024-E30
>
> As far as IBM boxes are concerned. Doe the -140 have stuff in common with
> leigh's -240 (the -140 being mono-cpu while the -240 being dual ?).
More different than that, 140 is more recent.
But works fine with Debian woody.(sorry Sven...)
>
> > [snip]
> >
> > > set <minor> <flag> <on|off>
> > >
> > > minor is the partition, starting at 1, flag is the flag, (prep in this
> > > case) and on|off is obvious.
> > >
> > > It simply sets the prep flag, which is then translated as a 0x41
> > > partition type. I think this is a deficiency in the parted/libparted
> > > API, but well, there is not much we can do about that unless a major
> > > libparted change.
> >
> > Cool. That's good to know.
>
> It was added recently though.
>
> > > Yep, d-i uses devfs, still partitions start at 1.
> >
> > Ok. Oops
> >
> > > > I see how much more time I can buy with this machine. It may need to
> > > > go back into storage this week if I can't make it run. Not fun.
> > >
> > > What about donating it to debian or something such ?
> >
> > I'd love to however it's not mine to donate. It's the companies (They
> > have 2 actually, I might try the other one since it has slightly
> > different hardware).
>
> Well, you could argue that donating one might make the other one usefull :)
>
> > The problem is: at work, I can't netboot. There is no server that will
> > respond and I can't set one up. I had the machine at home for the
> > holidays and thus can boot it from my LAN. If I can't get it to
> > self-boot before I have to take it back to the office - then I'm
> > stuck.
>
> Ok, understandable, still, getting it to netboot would be a first step to
> getting it working right later.
>
> > If I can get it to boot the system (even from floppy or cd) I will
> > have it to help test stuff. Otherwise it goes back under the desk
> > where it came from.
If there is such a deadline, why not consider to install first a woody (many
successes exist) then upgrading to sarge after ?
>
> Understandable too. What is the deadline ? I built a d-i kernel with the
> powerstack2 - utah pci irq patch, and it now works fine here. Still some
> work is needed, but basically this means i can scratch the disk on the box,
> and do some real d-i development, bringing in the partman-prep patch, and
> maybe even building a prep-installer that does the needed stuff to make it
> self-boot.
>
> I think you have to give your box back by monday next week though, right ?
> this could be too short, but let's see if we can make something happen.
>
> Right now, there are two issues :
>
> 1) the kernel may have some irq problems, or some other patching may be
> needed.
>
> 2) partman-prep is not included in d-i.
>
> 3) there is no automated prep-installer in d-i.
>
> Watch out on : http://people.debian.org/~luther/prep for uploads i make.
>
> Friendly,
>
> Sven Luther
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 14:30 UTC (permalink / raw)
To: mikemartin; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <855e88d0050103060766b09bd0@mail.gmail.com>
On Mon, Jan 03, 2005 at 09:07:31AM -0500, Mike Martin wrote:
> On Mon, 3 Jan 2005 15:01:02 +0100, Sven Luther <sven.luther@wanadoo.fr> wrote:
>
> [snip]
> > What exact machine do you have again ?
>
> An IBM 7043. It's a 43p-140 labeled as a 604e 332 Mhz.
Well, prep_pci.c does know about :
IBM RS/6000 7043-240 (ibm_doral)
IBM RS/6000 7024-E30
As far as IBM boxes are concerned. Doe the -140 have stuff in common with
leigh's -240 (the -140 being mono-cpu while the -240 being dual ?).
> [snip]
>
> >
> > set <minor> <flag> <on|off>
> >
> > minor is the partition, starting at 1, flag is the flag, (prep in this case)
> > and on|off is obvious.
> >
> > It simply sets the prep flag, which is then translated as a 0x41 partition
> > type. I think this is a deficiency in the parted/libparted API, but well,
> > there is not much we can do about that unless a major libparted change.
>
> Cool. That's good to know.
It was added recently though.
> > Yep, d-i uses devfs, still partitions start at 1.
>
> Ok. Oops
>
> > > I see how much more time I can buy with this machine. It may need to
> > > go back into storage this week if I can't make it run. Not fun.
> >
> > What about donating it to debian or something such ?
>
> I'd love to however it's not mine to donate. It's the companies (They
> have 2 actually, I might try the other one since it has slightly
> different hardware).
Well, you could argue that donating one might make the other one usefull :)
> The problem is: at work, I can't netboot. There is no server that will
> respond and I can't set one up. I had the machine at home for the
> holidays and thus can boot it from my LAN. If I can't get it to
> self-boot before I have to take it back to the office - then I'm
> stuck.
Ok, understandable, still, getting it to netboot would be a first step to
getting it working right later.
> If I can get it to boot the system (even from floppy or cd) I will
> have it to help test stuff. Otherwise it goes back under the desk
> where it came from.
Understandable too. What is the deadline ? I built a d-i kernel with the
powerstack2 - utah pci irq patch, and it now works fine here. Still some work
is needed, but basically this means i can scratch the disk on the box, and do
some real d-i development, bringing in the partman-prep patch, and maybe even
building a prep-installer that does the needed stuff to make it self-boot.
I think you have to give your box back by monday next week though, right ?
this could be too short, but let's see if we can make something happen.
Right now, there are two issues :
1) the kernel may have some irq problems, or some other patching may be
needed.
2) partman-prep is not included in d-i.
3) there is no automated prep-installer in d-i.
Watch out on : http://people.debian.org/~luther/prep for uploads i make.
Friendly,
Sven Luther
^ permalink raw reply
* Re: 43p-140 install issues
From: Mike Martin @ 2005-01-03 14:07 UTC (permalink / raw)
To: Sven Luther; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <20050103140102.GA9197@pegasos>
On Mon, 3 Jan 2005 15:01:02 +0100, Sven Luther <sven.luther@wanadoo.fr> wrote:
[snip]
> What exact machine do you have again ?
An IBM 7043. It's a 43p-140 labeled as a 604e 332 Mhz.
[snip]
>
> set <minor> <flag> <on|off>
>
> minor is the partition, starting at 1, flag is the flag, (prep in this case)
> and on|off is obvious.
>
> It simply sets the prep flag, which is then translated as a 0x41 partition
> type. I think this is a deficiency in the parted/libparted API, but well,
> there is not much we can do about that unless a major libparted change.
Cool. That's good to know.
[snip]
> Yep, d-i uses devfs, still partitions start at 1.
Ok. Oops
> > I see how much more time I can buy with this machine. It may need to
> > go back into storage this week if I can't make it run. Not fun.
>
> What about donating it to debian or something such ?
I'd love to however it's not mine to donate. It's the companies (They
have 2 actually, I might try the other one since it has slightly
different hardware).
The problem is: at work, I can't netboot. There is no server that will
respond and I can't set one up. I had the machine at home for the
holidays and thus can boot it from my LAN. If I can't get it to
self-boot before I have to take it back to the office - then I'm
stuck.
If I can get it to boot the system (even from floppy or cd) I will
have it to help test stuff. Otherwise it goes back under the desk
where it came from.
MikeMartin
^ permalink raw reply
* Re: MPC5200 PCI byte-swapping
From: Mark Chambers @ 2005-01-03 14:07 UTC (permalink / raw)
To: Sylvain Munaut; +Cc: linuxppc-embedded
In-Reply-To: <41D7269E.1070906@246tNt.com>
> Mark Chambers wrote:
>
> >I've just realized that the 5200 does byte-lane swapping
> >on all PCI accesses. That is, if you write a 32 bit word
> >0x12345678, 0x12 will go out on byte 0, 0x34 on byte 1,
> >etc. Unfortunately, my target, a T.I. DM642, does not
> >do this, so I've got a big/little endian mismatch. A couple
> >of questions if anybody knows:
> >
Sylvain wrote:
> I'm not sure of what you mean but look at the mapping
> aroung pdf page 337 of the user manual. It's not configurable
> as far as I can see.
>
Thanks. I had an older manual that didn't spell it out so
clearly.
I've been trying to interpret the PCI sections for some other
82xx family parts, and it appears that they do NOT do this
byte lane swapping, so this make the 5200 non-standard in
this regard, which is unfortunate. If I'm understanding this
right, one would have to have different drivers for a PCI
device on a 5200 and an 8270, for instance.
Also, I note that when doing simple block reads from pre-
fetchable PCI space, it appears the 5200 does not prefetch,
but does each read individually. This is using stock ELDK
u-boot and 2.4.24 so I haven't yet determined if it's a
configuration matter, (or ruled out target disconnects)
but I'm suspecting that you can't get burst mode from the
5200 without using DMA.
Mark Chambers
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 14:01 UTC (permalink / raw)
To: mikemartin; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <855e88d0050103053915c385ec@mail.gmail.com>
On Mon, Jan 03, 2005 at 08:39:43AM -0500, Mike Martin wrote:
> On Mon, 3 Jan 2005 08:05:35 +0100, Sven Luther <sven.luther@wanadoo.fr> wrote:
>
> [snip]
> > There may be some problems with 2.6 kernels and pci irqs on prep. I will
> > investigate this shortly.
>
> There is definitely something strange happening. The image I used to
> do the (nearly) complete install will no longer configure the network.
> Very inconsistent. I will try it (same image) again later today and
> see if it "just works" next time - like it did last time.
What exact machine do you have again ?
> [snip]
> >
> > Yep. Until partman-prep is ready, you should simply try to add the prep flag
> > to the a small partition. I think the normal scheme is to create /dev/sdb1 as
> > a small (8MB ?) prep partition, /dev/sdb2 as swap, and /dev/sdb3 as /. If
> > partman is not able to do it, jump to console 2, and launch parted directly,
> > and do a set 1 prep on.
>
> Can't jump to console 2 - I can only comunicate through the serial
Ok, then go to execute a shell then.
> console. However I'll give this a shot through the "execute shell"
> menu item. "set 1 prep on"? Haven't seen that documented anywhere,
> what does it do?
set <minor> <flag> <on|off>
minor is the partition, starting at 1, flag is the flag, (prep in this case)
and on|off is obvious.
It simply sets the prep flag, which is then translated as a 0x41 partition
type. I think this is a deficiency in the parted/libparted API, but well,
there is not much we can do about that unless a major libparted change.
> > Then i suppose that copying the kernel is just a matter of :
> >
> > dd if=/boot/vmlinuz-2.6.8-powerpc of=/dev/sdb1
> >
> > but you have to have a small prep partition as /dev/sdb1, or the above will
> > erase whatever you have on /dev/sdb1.
> >
>
> I' have the partition ready ...
>
> Unfortunately, until the network comes back up, I can't get far enough
> in the install to access the SCSI drives (is there another way? Load
> the modules by hand? Which ones?). Without this, I can't get at the
> image I need to boot in order to get it onto the right partition. I
> also can't get it to another machine to net boot it.
He, understood, we definitively need to investigate the pci irq problems.
> > > I tried booting up with the installer, going to a shell after the scsi
> > > was active and the installed system was mounted. I used sfdisk from
> > > /target/sbin to create a 10 Meg type 41 partition on the first HD
> > > (/dev/sdb0 eventually). I used dd to copy /target/boot/vmlinux to this
> > > partition. But it still won't boot.
> >
> > Ah, no sure what is wrong. I don't think /dev/sdb0 exists. Not sure really.
> >
>
> Actually /dev/sdbX doesn't exist at all during the install from what I
Yep, d-i uses devfs, still partitions start at 1.
> have seen. The drives are refered to as
> /dev/scsi/host0/etc/.../etc/part1. I have two drives, one is 9 Gig
> with scsi id (0,4,0) (/dev/sda) . This is blank except for a 10 Meg
> boot partition; type 41. The second drive is 20 Gig with scsi id
> (0,5,0)(/dev/sdb) which is where the install is.
>
> I copied (with dd) the file (the wrong file I suspect) to
> /dev/scsi/...../part1 which should correspond to the first partiton on
> the 9 Gig drive.
Indeed.
> I see how much more time I can buy with this machine. It may need to
> go back into storage this week if I can't make it run. Not fun.
What about donating it to debian or something such ?
Friendly,
Sven Luther
^ permalink raw reply
* Re: 43p-140 install issues
From: Mike Martin @ 2005-01-03 13:39 UTC (permalink / raw)
To: Sven Luther; +Cc: Ulrich Teichert, linuxppc-dev, debian-powerpc
In-Reply-To: <20050103070535.GA1644@pegasos>
On Mon, 3 Jan 2005 08:05:35 +0100, Sven Luther <sven.luther@wanadoo.fr> wrote:
[snip]
> There may be some problems with 2.6 kernels and pci irqs on prep. I will
> investigate this shortly.
There is definitely something strange happening. The image I used to
do the (nearly) complete install will no longer configure the network.
Very inconsistent. I will try it (same image) again later today and
see if it "just works" next time - like it did last time.
[snip]
>
> Yep. Until partman-prep is ready, you should simply try to add the prep flag
> to the a small partition. I think the normal scheme is to create /dev/sdb1 as
> a small (8MB ?) prep partition, /dev/sdb2 as swap, and /dev/sdb3 as /. If
> partman is not able to do it, jump to console 2, and launch parted directly,
> and do a set 1 prep on.
Can't jump to console 2 - I can only comunicate through the serial
console. However I'll give this a shot through the "execute shell"
menu item. "set 1 prep on"? Haven't seen that documented anywhere,
what does it do?
>
> Then i suppose that copying the kernel is just a matter of :
>
> dd if=/boot/vmlinuz-2.6.8-powerpc of=/dev/sdb1
>
> but you have to have a small prep partition as /dev/sdb1, or the above will
> erase whatever you have on /dev/sdb1.
>
I' have the partition ready ...
Unfortunately, until the network comes back up, I can't get far enough
in the install to access the SCSI drives (is there another way? Load
the modules by hand? Which ones?). Without this, I can't get at the
image I need to boot in order to get it onto the right partition. I
also can't get it to another machine to net boot it.
> > I tried booting up with the installer, going to a shell after the scsi
> > was active and the installed system was mounted. I used sfdisk from
> > /target/sbin to create a 10 Meg type 41 partition on the first HD
> > (/dev/sdb0 eventually). I used dd to copy /target/boot/vmlinux to this
> > partition. But it still won't boot.
>
> Ah, no sure what is wrong. I don't think /dev/sdb0 exists. Not sure really.
>
Actually /dev/sdbX doesn't exist at all during the install from what I
have seen. The drives are refered to as
/dev/scsi/host0/etc/.../etc/part1. I have two drives, one is 9 Gig
with scsi id (0,4,0) (/dev/sda) . This is blank except for a 10 Meg
boot partition; type 41. The second drive is 20 Gig with scsi id
(0,5,0)(/dev/sdb) which is where the install is.
I copied (with dd) the file (the wrong file I suspect) to
/dev/scsi/...../part1 which should correspond to the first partiton on
the 9 Gig drive.
I see how much more time I can buy with this machine. It may need to
go back into storage this week if I can't make it run. Not fun.
Thanks for everyone's help.
MikeMartin
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 12:47 UTC (permalink / raw)
To: Michael Schmitz
Cc: Ulrich Teichert, Philippe Guyot, Mike Martin, linuxppc-dev,
debian-powerpc, mikemartin
In-Reply-To: <Pine.LNX.4.44.0501031257040.2302-100000@opal.biophys.uni-duesseldorf.de>
On Mon, Jan 03, 2005 at 12:57:35PM +0100, Michael Schmitz wrote:
> > > > > is, so Leigh did a piece of code named "preptools" in order to patch the
> > > > > image with the location of root.
> > > >
> > > > Hey, cool, where is this tool ? We need to package it for debian if it is
> > > > not already, and use it in d-i, and maybe even mkvmlinuz directly.
> > > >
> > >
> > > Just over there....
> > >
> > > http://www.solinno.co.uk/7043-140/downloads.php
> > > But ask Leigh for details.
> >
> > Ok, thanks. ...
>
> Do we need that added to powerpc-utils perhaps?
Whatever is powerpc-utils ? Mmm, maybe, but then maybe mkvmlinuz would benefit
from being added there too, not sure. In any way i believe it has more its
place inside mkvmlinuz, than inside the more userland powerpc-utils.
Friendly,
Sven Luther
^ permalink raw reply
* Re: [FYI] linux 2.6 still not working with PReP (ppc32)
From: Marc Dietrich @ 2005-01-03 12:26 UTC (permalink / raw)
To: sheutlin; +Cc: linuxppc-dev
In-Reply-To: <1104625860.10369.34.camel@weizen.left.earth>
Hi Christian,
Am Sonntag, 2. Januar 2005 01:31 schrieb Sebastian Heutling:
> Hi Marc,
>
> On Mon, 2004-12-27 at 22:37 +0100, Marc Dietrich wrote:
> > Hallo Christian,
> >
> > maybe this helps:
> >
> > I used the config file you sent me (and added serial console).
> >
> > when booting a working 2.4.24-pre1 I get:
> >
> > [...]
> > PCI: Probing PCI hardware
> > Setting PCI interrupts for a "Utah (Powerstack II Pro4000)"
> > PCI: moved device 00:0b.1 resource 4 (101) to 1480
> > PCI: moved device 00:0b.1 resource 5 (101) to 1490
> > PCI: moved device 00:11.0 resource 0 (1208) to 0
> > [...]
> > SCSI subsystem driver Revision: 1.00
> > new devtbl [4096] @c049e000.
> > PCI: Enabling device 00:0c.0 (0000 -> 0003)
> > sym53c8xx: at PCI bus 0, device 12, function 0
> > sym53c8xx: setting PCI_COMMAND_MASTER PCI_COMMAND_PARITY...(fix-up)
> > sym53c8xx: changing PCI_LATENCY_TIMER from 0 to 80.
> > sym53c8xx: 53c825a detected
> > sym53c825a-0: rev 0x13 on pci bus 0 device 12 function 0 irq 15
> > new MPOOL [ 188] @c049f000.
> > new VTOB [ 12] @c049f100.
> > new NCB [2408] @c049c000.
> > new SQUEUE [2336] @c049d000.
> > new VTOB [ 12] @c049f110.
> > new DQUEUE [2336] @c049a000.
> > new TARGTBL [ 256] @c049b000.
> > new VTOB [ 12] @c049f120.
> > new SCRIPT [3504] @c0498000.
> > new SCRIPTH [1504] @c049b800.
> > new CCB [1300] @c0499000.
> > sym53c825a-0: ID 7, Fast-10, Parity Checking
> >
> > I already changed the irq of slot 12 to 15 and slot 14 to irq 11.
>
> You don't really need to change any IRQ above slot 9 on a 2.6.? kernel
> because the slot is 2 (while it was on 12 in 2.4 kernels).
ah yes, now I understand. You just copied all numbers from top to down.
> > [...]
> > PCI: Probing PCI hardware
> > Setting PCI interrupts for a "Utah (Powerstack II Pro4000)"
> > PCI: Cannot allocate resource region 5 of device 0000:00:01.1
>
> This is the IDE chip which gets IRQ 0 and therefore results in an error.
> Try changing the value of Slot 1 from 0 to 4 (which is IRQ 14). I don't
> really think it changes the SCSI timeouts maybe it helps - no one knows
> for sure ;).
Ok - did that also.
> > SCSI subsystem initialized
> > [...]
> > PCI: Enabling device 0000:00:02.0 (0000 -> 0003)
> > sym0: <825a> rev 0x13 at pci 0000:00:02.0 irq 15
so I guess 0000:00:02.0 is right as oposed to 0000:00:0C.0 with 2.4 kernels.
> > [...]
> > new BADLUNTBL [ 256] @c7d9aa00.
> > sym0: SCSI BUS has been reset.
> > scsi0 : sym-2.1.18m
> > elevator: using anticipatory as default io scheduler
> > sym0:0:0:ccb @c7d9ac00 using tag 256.
> > sym0: queuepos=2.
> > sym0:0:0: ABORT operation started.
> > sym0:0:0: ABORT operation timed-out.
> > sym0:0:0: DEVICE RESET operation started.
> > sym0:0:0: DEVICE RESET operation timed-out.
> > sym0:0:0: BUS RESET operation started.
> > sym0:0:0: BUS RESET operation timed-out.
> > sym0:0:0: HOST RESET operation started.
> > sym0:0:0:ccb @c7d9ac00 freeing tag 256.
> > sym0: SCSI BUS has been reset.
> > [ hangs here ]
>
> Hmmm strange ... is there anything attached on scsi?
Yes - a hard disk and a cdrom. Both are working under 2.4 kernels (see the
boot.log of the 2.4.24-pre1 kernel). I don't use net boot. My system (debian
sarge) is installed on sda3. Anyway - also on a netboot system, the kernel
should hang when trying to initialize the scsi controler.
> ... wait a second ... looking at the lspci output ... you use a USB PCI
> card? The IRQ-sharing doesn't work well and the USB PCI card is on the
> same slot as the SCSI controller resulting in both using the same IRQ.
> Remove the USB card and check wether it changes anything. If it does
> work now: Did it ever work with USB using another kernel? If so maybe
> there is something else wrong, if not it's a HW Problem.
> I also attached a config file from my powerstack (which I use as a
> router and runs a 2.6 kernel since june last year or so)
I didn't checked if the usb card is working under a 2.4 kernel, but it boots
without problems. I removed it now, just to eleminate possible trouble.
I setup a kernel using your config file and added vga console and offb and
removed the nfs drivers to get an image smaller then 1.44 Mb (I'm using
XFree86-3.3.6 with the cirrus 5664 card). This has the same effect than
before: scsi not working.
It seems, that this config is not for a 2.6.10 kernel. What kernel version are
you using? Maybe some bad things happened in newer kernels.
Can you put a working zImage.prep somewhere on the web (<1.44 MB, so I can
write it to a floppy disk)?
How does your boot.log looks like?
Greetings
Marc
^ permalink raw reply
* Re: 43p-140 install issues
From: Michael Schmitz @ 2005-01-03 11:57 UTC (permalink / raw)
To: Sven Luther
Cc: Ulrich Teichert, Philippe Guyot, Mike Martin, linuxppc-dev,
debian-powerpc, mikemartin
In-Reply-To: <20050103111718.GA11426@pegasos>
> > > > is, so Leigh did a piece of code named "preptools" in order to patch the
> > > > image with the location of root.
> > >
> > > Hey, cool, where is this tool ? We need to package it for debian if it is
> > > not already, and use it in d-i, and maybe even mkvmlinuz directly.
> > >
> >
> > Just over there....
> >
> > http://www.solinno.co.uk/7043-140/downloads.php
> > But ask Leigh for details.
>
> Ok, thanks. ...
Do we need that added to powerpc-utils perhaps?
Michael
^ permalink raw reply
* Re: 43p-140 install issues
From: Sven Luther @ 2005-01-03 11:17 UTC (permalink / raw)
To: Philippe Guyot
Cc: Ulrich Teichert, Mike Martin, linuxppc-dev, debian-powerpc,
mikemartin
In-Reply-To: <200501031156.07353.pguyot@cvf.fr>
On Mon, Jan 03, 2005 at 11:56:07AM +0100, Philippe Guyot wrote:
> On Monday 03 January 2005 11:48, Sven Luther wrote:
> > On Mon, Jan 03, 2005 at 10:56:30AM +0100, Philippe Guyot wrote:
> > > On Monday 03 January 2005 05:15, Mike Martin wrote:
> > > > Now I'm very confused .... on my next reboot ... the network wouldn't
> > > > work. After much trial and error, I got the current unstable d-i image
> > > > to work. It did it all right up to the reboot (where is warns you that
> > > > it doesn't have a boot loader installed).
> > > >
> > > > Now what!
> > > >
> > > > I have an installed system, but I can't figure out how to boot it! I
> > > > have /boot/vmlinux on /dev/sdb1 and a root on /dev/sdb1. How do I
> > > > write a boot sector or tell OpenFirmware to boot from this file? I
> > > > tried to netboot and set root=/dev/sdb1 - but without the scsi modules
> > > > loaded it fails to read the drives and panics.
> > > >
> > > > I think I neet to create a type 41 PReP boot partition and get the
> > > > kernel onto it. I've found this thread:
> > > > http://lists.gnu.org/archive/html/bug-parted/2002-09/msg00016.html
> > > >
> > > > and this:
> > > > http://www.yellowdoglinux.com/support/solutions/champion_server/rs6k.sh
> > > >tml
> > > >
> > > > I tried booting up with the installer, going to a shell after the scsi
> > > > was active and the installed system was mounted. I used sfdisk from
> > > > /target/sbin to create a 10 Meg type 41 partition on the first HD
> > > > (/dev/sdb0 eventually). I used dd to copy /target/boot/vmlinux to this
> > > > partition. But it still won't boot.
> > >
> > > I succeeded last year to install a woody on a 143P-140 whith the help of
> > > Leigh Brown (thanks to him again). And two 150 whit sarge also.
> > >
> > > I'm not sure, but I have the feeling that your image on the prep
> > > partition (should be type 0x41) is not able to determine where the root
> > > is, so Leigh did a piece of code named "preptools" in order to patch the
> > > image with the location of root.
> >
> > Hey, cool, where is this tool ? We need to package it for debian if it is
> > not already, and use it in d-i, and maybe even mkvmlinuz directly.
> >
>
> Just over there....
>
> http://www.solinno.co.uk/7043-140/downloads.php
> But ask Leigh for details.
Ok, thanks. ...
Friendly,
Sven Luther
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox