sgi octane2 R12K gentoo

[ddc]

hi,
i'd like to install gentoo-mips on sgi octane2 R12K-400Mhz. I read the gentoo handbook and i netboot the unit with a IP30 kernel: tried to boot http://dev.gentoo.org/~kumba/mips/netboot/ip30/ip30-r10k+-20050128.img.bz2 via bootp (Gentoo Linux/MIPS Handbook, Part 2). the kernel is downloaded correctly but unfortunately it in't execued.


it was stopped with an error: prom write error on cacheline 0x1fcc0f00 at PC=8000000202729e0, then everything stops. Mouse and keyboard are dead and sometimes the red light comes up.

Kumba recommended to try the following boot method (and i did)

>> resetenv
>> unsetenv netaddr
>> unsetenv dlserver
>> setenv dbaud 9600
>> setenv console d
>> init

about the firmware: PROM monitor SGI version 6.5 Rev 4.16 IP30 May 22, 2003 64 bit


i think we could use octane2 added with a PCI-cage (i have it) to use a pci-matrox-millennium-1/2 as framebuffer: i did it on HPPA-risc-machine and it is superlative !

the only problem is how to correctly boot the kernel and how to install it ...

[Redhatter]

Clearly there's something about that netboot image that your Octane does not like. That said... Kumba has been working on new boot media for the 2006.1 release, I'll see if I can locate where he's stashed the files, and let you know where to get them.

I can't assist myself as I don't own an Octane (wish I did though).
_________________
Stuart Longland (a.k.a Redhatter, VK4MSL)
I haven't lost my mind - it's backed up on a tape somewhere...

Gentoo/MIPS Cobalt developer, Mozilla herd member.

[ddc]

[e-ipi]

Did you find a bootable kernel yet? If not try the one redhatter recommended to me a while back:

http://dev.gentoo.org/~kumba/mips/netboot/testing/ip30-r10k+-20060112.img.bz2

I have one of those Octanes1.5 with the green skins, but the 030-1467-00x motherboard, the Octane2 Heart chip, the 1.4 Xbow, and the later power supply. It's a dual R12000 300. The biggest difference is that mine has SE + TRAM graphics rather than V-whatever. I'm thinking that if a kernel will work on my machine, it ought to work on yours.

Do let us know how it goes with the pci graphics experiment. (You and your pci-box, you lucky dog you. I once had one, but I had to give it and the Octane up before a transpacific move.) Given that a lot of people don't have the PCI shoebox[1], I'd rather have fully functional Impact/Oddysey drivers, but more in the way of options is always a good thing.



[1] Don't know if they are any easier to find than the Octane PCI box, but I've heard that it's possible to use an Origin2K XIO/PCI carrier in an Octane XIO slot. Can anyone confirm that?

[ddc]

about ip30-r10k+-20060112.img.bz2, i realize it boots fine (without the previous error) on the blue octane2 REV 2003, but it is impossible to use console=/dev/ttyS0 specifed as kernel boot parameter: the system simply panic ! So i used the V6 graphic board as console, then i used ssh

there is a problem in portage archives: emerge --sync stops at the 88% of the update, and it is impossible to emerge mips-sources, so it is impossible to recompile the kernel and a lot of system sources. Also i note the portage tree is 2005 dated on the server. How to complete the gentoo installation ???




about dmesg: here it is

Linux version 2.6.14.6-mipsgit-20051030-ip30r10k+ (root@dol-guldur) (gcc version 3.4.4 (Gentoo Linux 3.4.4)) #1 SMP Fri Jan 13 15:50:12 UTC 2006
ARCH: SGI-IP30
PROMLIB: ARC firmware Version 64 Revision 0
CPU revision is: 00000e35
FPU revision is: 00000900
Silicon Graphics Octane (IP30) support: (c) 2004, 2005 Stanislaw Skowronek.
Detected 2048 MB of physical memory.
Updating PROM memory size.
xtalk: Detected XBow (revision 2.0) at 0.
xtalk: Detected Heart (revision F) at 8.
xtalk: Detected Buzz / Odyssey (revision B) at 11.
xtalk: Detected Bridge (revision C) at 13.
xtalk: Detected Bridge (revision D) at 15.
BRIDGE chip at xtalk:15, initializing...
BRIDGE chip at xtalk:13, initializing...
Determined physical RAM map:
memory: 0000000000004000 @ 0000000000000000 (reserved)
memory: 000000000092a000 @ 0000000020004000 (reserved)
memory: 00000000005d2000 @ 000000002092e000 (usable)
memory: 0000000000100000 @ 0000000020f00000 (ROM data)
memory: 000000003f000000 @ 0000000021000000 (usable)
memory: 0000000040000000 @ 0000000060000000 (reserved)
memory: 0000000040000000 @ 0000000060000000 (usable)
SGI Octane system with >2GB physical memory, limiting DMA.
On node 0 totalpages: 655360
DMA zone: 655360 pages, LIFO batch:31
Normal zone: 0 pages, LIFO batch:1
HighMem zone: 0 pages, LIFO batch:1
Built 1 zonelists
Kernel command line: root=/dev/ram0
Primary instruction cache 32kB, physically tagged, 2-way, linesize 64 bytes.
Primary data cache 32kB, 2-way, linesize 32 bytes.
Unified secondary cache 2048kB 2-way, linesize 128 bytes.
Synthesized TLB refill handler (41 instructions).
Synthesized TLB load handler fastpath (55 instructions).
Synthesized TLB store handler fastpath (55 instructions).
Synthesized TLB modify handler fastpath (54 instructions).
IP30: interrupt controller initialized.
PID hash table entries: 4096 (order: 12, 131072 bytes)
IP30: initializing timer.
400 MHz CPU detected
Using 200.489 MHz high precision timer.
Console: colour dummy device 80x25
Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes)
Inode-cache hash table entries: 262144 (order: 9, 2097152 bytes)
Memory: 2043904k/2086728k available (4616k kernel code, 42260k reserved, 901k data, 3564k init, 0k highmem)
Calibrating delay loop... 265.21 BogoMIPS (lpj=132608)
Mount-cache hash table entries: 256
Checking for 'wait' instruction... unavailable.
Checking for the multiply/shift bug... no.
Checking for the daddi bug... no.
Checking for the daddiu bug... no.
Brought up 1 CPUs
softlockup thread 0 started up.
NET: Registered protocol family 16
SCSI subsystem initialized
usbcore: registered new driver usbfs
usbcore: registered new driver hub
IP30: HEART ATTACK! Caught errors: 0x0040!
interrupt #57
IP30: HEART ATTACK! Caught errors: 0x0010!
interrupt #55
PCI: Failed to allocate mem resource #1:800000@1d800000 for 0001:01:01.0
NTFS driver 2.1.24 [Flags: R/O].
JFS: nTxBlock = 8192, nTxLock = 65536
SGI XFS with large block/inode numbers, no debug enabled
pci_hotplug: PCI Hot Plug PCI Core version: 0.5
fakephp: Fake PCI Hot Plug Controller Driver
Console: switching to colour frame buffer device 160x64
fb0: Odyssey frame buffer device
Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
loop: loaded (max 8 devices)
qla1280: QLA1040 found on PCI bus 0, dev 0
BRIDGE: IO #0, size 0x100 for little-endian 0000:00:00.0 --> direct I/O at bus 0x01000000 vma 0x000000f101000000
BRIDGE: Memory #1, size 0x1000 for little-endian 0000:00:00.0 --> direct 32-bit at bus 0x01000000 vma 0x000000f041000000
BRIDGE: Memory #6, size 0x10000 for little-endian 0000:00:00.0 --> direct 32-bit at bus 0x01010000 vma 0x000000f041010000
PCI: Enabling device 0000:00:00.0 (0006 -> 0007)
scsi(0:0): Resetting SCSI BUS
scsi0 : QLogic QLA1040 PCI to SCSI Host Adapter
Firmware version: 7.65.06, Driver version 3.25
qla1280: QLA1040 found on PCI bus 0, dev 1
BRIDGE: IO #0, size 0x100 for little-endian 0000:00:01.0 --> direct I/O at bus 0x01000100 vma 0x000000f101000100
BRIDGE: Memory #1, size 0x1000 for little-endian 0000:00:01.0 --> direct 32-bit at bus 0x01020000 vma 0x000000f041020000
BRIDGE: Memory #6, size 0x10000 for little-endian 0000:00:01.0 --> direct 32-bit at bus 0x01030000 vma 0x000000f041030000
PCI: Enabling device 0000:00:01.0 (0006 -> 0007)
scsi(1:0): Resetting SCSI BUS
scsi1 : QLogic QLA1040 PCI to SCSI Host Adapter
Firmware version: 7.65.06, Driver version 3.25
st: Version 20050830, fixed bufsize 32768, s/g segs 256
usbmon: debugfs is not available
ohci_hcd: 2005 April 22 USB 1.1 'Open' Host Controller (OHCI) Driver (PCI)
USB Universal Host Controller Interface driver v2.3
Initializing USB Mass Storage driver...
usbcore: registered new driver usb-storage
USB Mass Storage support registered.
usbcore: registered new driver usbserial
drivers/usb/serial/usb-serial.c: USB Serial Driver core v2.0
mice: PS/2 mouse device common for all mice
md: linear personality registered as nr 1
md: raid0 personality registered as nr 2
md: raid1 personality registered as nr 3
md: raid10 personality registered as nr 9
md: raid5 personality registered as nr 4
raid5: measuring checksumming speed
8regs : 1024.000 MB/sec
8regs_prefetch: 980.000 MB/sec
32regs : 1024.000 MB/sec
32regs_prefetch: 980.000 MB/sec
raid5: using function: 32regs (1024.000 MB/sec)
raid6: int64x1 304 MB/s
raid6: int64x2 406 MB/s
raid6: int64x4 429 MB/s
raid6: int64x8 265 MB/s
raid6: using algorithm int64x4 (429 MB/s)
md: raid6 personality registered as nr 8
md: md driver 0.90.2 MAX_MD_DEVS=256, MD_SB_DISKS=27
md: bitmap version 3.39
Advanced Linux Sound Architecture Driver Version 1.0.10rc1 (Mon Sep 12 08:13:09 2005 UTC).
BRIDGE: Memory #0, size 0x2000 for little-endian 0000:00:03.0 --> direct 32-bit at bus 0x01040000 vma 0x000000f041040000
ALSA device list:
#0: SGI RAD Audio at 0xf041040000
BRIDGE: Memory #0, size 0x100000 for little-endian 0000:00:02.0 --> direct 32-bit at bus 0x01100000 vma 0x000000f041100000
IOC3 part: [030-0891-003], serial: [KZG030] => class IP30 system
ttyS0 at IOC3 0xf041120178 (irq = 64) is a 16550A
ttyS1 at IOC3 0xf041120170 (irq = 64) is a 16550A
Ethernet address is 08:00:69:13:66:d4.
eth0: link up, 100Mbps, full-duplex, lpa 0x01E1
eth0: Using PHY 1, vendor 0x15f42, model 2, rev 3.
eth0: IOC3 SSRAM has 128 kbyte.
IOC3 Master Driver loaded for 0000:00:02.0
NET: Registered protocol family 2
IP route cache hash table entries: 131072 (order: 8, 1048576 bytes)
TCP established hash table entries: 262144 (order: 10, 4194304 bytes)
TCP bind hash table entries: 65536 (order: 8, 1048576 bytes)
TCP: Hash tables configured (established 262144 bind 65536)
TCP reno registered
TCP bic registered
NET: Registered protocol family 1
NET: Registered protocol family 17
Freeing prom memory: 1024kb freed
Freeing unused kernel memory: 4588k freed
input: AT Raw Set 2 keyboard on ioc3/serio0kbd
logips2pp: Detected unknown logitech mouse model 18
input: PS/2 Logitech Mouse on ioc3/serio0aux

[ddc]

error emerging mips-sources

please help me to fix it, we have to run gentoo on octane2 R12K
what is wrong ? is it the /etc/make.conf ? if so, what is the right ?



Calculating dependencies ... done!
>>> Emerging (1 of 5) dev-libs/gmp-4.2.1 to /
>>> Unpacking source...
>>> Unpacking gmp-4.2.1.tar.bz2 to /var/tmp/portage/gmp-4.2.1/work
>>> Compiling source in /var/tmp/portage/gmp-4.2.1/work/gmp-4.2.1 ...
* econf: updating gmp-4.2.1/config.guess with /usr/share/gnuconfig/config.guess
* econf: updating gmp-4.2.1/config.sub with /usr/share/gnuconfig/config.sub
./configure --prefix=/usr --host=mips-unknown-linux-gnu --mandir=/usr/share/man --infodir=/usr/share/info --datadir=/usr/share --sysconfdir=/etc --localstatedir=/var/lib --localstatedir=/var/state/gmp --disable-mpfr --disable-mpbsd --enable-cxx --build=mips-unknown-linux-gnu
checking build system type... mips-unknown-linux-gnu
checking host system type... mips-unknown-linux-gnu
checking for a BSD-compatible install... /bin/install -c
checking whether build environment is sane... yes
checking for gawk... gawk
checking whether make sets $(MAKE)... yes
checking whether to enable maintainer-specific portions of Makefiles... no
checking GMPABI=o32
checking compiler gcc -O2 -march=mips4 -pipe -fomit-frame-pointer -ftracer -fforce-addr ... yes
checking whether gcc supports o32... yes
checking for mips-unknown-linux-gnu-gcc... gcc
checking for C compiler default output file name... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ANSI C... none needed
checking how to run the C preprocessor... gcc -E
checking build system compiler gcc... yes
checking for build system preprocessor... gcc -E
checking for build system executable suffix...
checking whether build system compiler is ANSI... yes
checking for build system compiler math library... -lm
checking for mips-unknown-linux-gnu-g++... mips-unknown-linux-gnu-g++
checking whether we are using the GNU C++ compiler... yes
checking whether mips-unknown-linux-gnu-g++ accepts -g... yes
checking C++ compiler mips-unknown-linux-gnu-g++ -O2 -march=mips4 -pipe -fomit-frame-pointer -ftracer -fforce-addr... no, std iostream

!!! Please attach the following file when filing a report to bugs.gentoo.org:
!!! /var/tmp/portage/gmp-4.2.1/work/gmp-4.2.1/config.log

!!! ERROR: dev-libs/gmp-4.2.1 failed.
Call stack:
ebuild.sh, line 1539: Called dyn_compile
ebuild.sh, line 939: Called src_compile
gmp-4.2.1.ebuild, line 44: Called econf '--localstatedir=/var/state/gmp' '--disable-mpfr' '--disable-mpbsd' '--enable-cxx'
ebuild.sh, line 541: Called die
!!! econf failed
!!! If you need support, post the topmost build error, and the call stack if relevant.




here it is /etc/make.conf

# These settings were set by the catalyst build script that automatically built this stage
# Please consult /etc/make.conf.example for a more detailed example
CFLAGS="-O2 -march=mips4 -pipe -fomit-frame-pointer -ftracer -fforce-addr"
CHOST="mips-unknown-linux-gnu"
CXXFLAGS="${CFLAGS}"


here it is gcc -v

Reading specs from /usr/lib/gcc/mips-unknown-linux-gnu/3.4.6/specs
Configured with: /var/tmp/portage/gcc-3.4.6-r1/work/gcc-3.4.6/configure --prefix=/usr --bindir=/usr/mips-unknown-linux-gnu/gcc-bin/3.4.6 --includedir=/usr/lib/gcc/mips-unknown-linux-gnu/3.4.6/include --datadir=/usr/share/gcc-data/mips-unknown-linux-gnu/3.4.6 --mandir=/usr/share/gcc-data/mips-unknown-linux-gnu/3.4.6/man --infodir=/usr/share/gcc-data/mips-unknown-linux-gnu/3.4.6/info --with-gxx-include-dir=/usr/lib/gcc/mips-unknown-linux-gnu/3.4.6/include/g++-v3 --host=mips-unknown-linux-gnu --build=mips-unknown-linux-gnu --disable-altivec --enable-nls --without-included-gettext --with-system-zlib --disable-checking --disable-werror --disable-libunwind-exceptions --disable-multilib --disable-libgcj --enable-languages=c,c++,f77 --enable-shared --enable-threads=posix --enable-__cxa_atexit --enable-clocale=gnu
Thread model: posix
gcc version 3.4.6 (Gentoo 3.4.6-r1, ssp-3.4.5-1.0, pie-8.7.9)

[Redhatter]

[ddc]

sorry,

i modified /etc/make.conf to include "-mabi=32" and i did the following
rm -f /etc/make.profile
ln -s /usr/portage/profiles/default-linux/mips/2006.1/ip30/o32 /etc/make.profile
USE="-doc" emerge mips-sources


but i get this error:

[..]
checking C++ compiler mips-unknown-linux-gnu-g++ -O2 -march=mips4 -pipe -mabi=32... no, std iostream

!!! Please attach the following file when filing a report to bugs.gentoo.org:
!!! /var/tmp/portage/gmp-4.2.1/work/gmp-4.2.1/config.log

!!! ERROR: dev-libs/gmp-4.2.1 failed.
Call stack:
ebuild.sh, line 1539: Called dyn_compile
ebuild.sh, line 939: Called src_compile
gmp-4.2.1.ebuild, line 44: Called econf '--localstatedir=/var/state/gmp' '--disable-mpfr' '--disable-mpbsd' '--enable-cxx'
ebuild.sh, line 541: Called die
!!! econf failed
!!! If you need support, post the topmost build error, and the call stack if relevant.


snapshot and portage are updated to 2006.1, about the handbook, i'm reading the experimental one from http://www.longlandclan.hopto.org/~stuartl/gentoo/docs/index.php/handbook/handbook-mips.xml?part=1&chap=5



what is wrong now ? please help me, i'm stopped at mips-sources

[ddc]

the gentoo-kernel for sgi octane2 is 64bit, so we have to emerge kgcc64.

kgcc64 is installed as a 64bit version of the standard gcc.
Unfortunatly it seems there is no 64 bit version of the binutils, so when i go to /usr/src/linux and i invoke make, i'm stopped by "no mips64-linux-ld". There is no 64bit version of the linker (ld)

how to fix it ?

[Redhatter]

That shouldn't happen. IIRC, kgcc64 should include a linker.

The fact that it's hunting for mips64-linux-*, suggests you forgot to specify CROSS_COMPILE=mips64-unknown-linux-gnu-, as suggested by the handbook.

Alternatively, you can also cross-compile from an x86 box using crossdev to produce a mips64 cross-compiler. This is what we do, generally, because it's a lot faster than compiling natively on a MIPS machine.

My O2 is currently installing kgcc64, I shall report back.
_________________
Stuart Longland (a.k.a Redhatter, VK4MSL)
I haven't lost my mind - it's backed up on a tape somewhere...

Gentoo/MIPS Cobalt developer, Mozilla herd member.

[Kumba]

ddc: What stage did you start from? It looks like C++ is missing. Or at least, the configure test in gmp thinks the C++ headers are missing. Rather curious.

Also, for the kernel, I thought we had a tweak go in to fix that. That might've been crossdev, though. Add this command to your bash profile:
alias mips64make="make ARCH=mips CROSS_COMPILE=mips64-unknown-linux-gnu-"

Then you can just run "mips64make vmlinux" to get the build system to work. Remember, O2 users, if using gcc-4.1.x, need to use vmlinux.32 as their target.


--Kumba
_________________
"The past tempts us, the present confuses us, the future frightens us. And our lives slip away, moment by moment, lost in that vast, terrible in-between."
--Emperor Turhan, Centauri Republic

[ddc]

hi,
with these incomprensible errors on my IP30 machine, i decided to re-format the harddisk, and re doing the following procedure: unfortunatly it is still stopped by the previous error, as i have posted

in portage i can see gnu-gcc v3.4.6, c, c++, gnu-as v2.16.1
emerging sys-devel/gcc (4.1.1) and kgcc64 (3.4.6)) doen't solve the problem




echo "/bin/net-setup 192.168.0.19 192.168.0.1"
echo "passwd"
mkdir /mnt/gentoo
mount -t ext3 /dev/sda3 /mnt/gentoo
mount -t ext2 /dev/sda1 /mnt/gentoo/boot

mkdir /mnt/gentoo/myinstall
cd /mnt/gentoo/myinstall
wget -c http://distfiles.gentoo.org/releases/mips/2006.1/stages/mips4/stage3-mips4-2006.1.tar.bz2
wget ftp://ftp.ucsb.edu/pub/mirrors/linux/gentoo/snapshots/portage-20060920.tar.bz2
sh get-unpack.sh
cd /

mount -t proc none /mnt/gentoo/proc
mount -o bind /dev /mnt/gentoo/dev
mount

echo cp -L /etc/resolv.conf /mnt/gentoo/etc/resolv.conf
echo "nameserver 192.168.0.1" > /mnt/gentoo/etc/resolv.conf
chroot /mnt/gentoo /bin/bash
env-update
source /etc/profile
export PS1="(chroot) $PS1"
ping -c 4 www.google.it
emerge --sync

ls -FGg /etc/make.profile
rm -f /etc/make.profile
ln -s /usr/portage/profiles/default-linux/mips/2006.1/ip30/o32 /etc/make.profile
cp /etc/make.conf /etc/make.conf.old
echo "set CFLAGS=-mabi=32 -mips4 -pipe -O2"
nano /etc/make.conf
echo nano /usr/portage/profiles/use.desc
USE="-doc" emerge mips-sources

!!! Please attach the following file when filing a report to bugs.gentoo.org:
!!! /var/tmp/portage/gmp-4.2.1/work/gmp-4.2.1/config.log

!!! ERROR: dev-libs/gmp-4.2.1 failed.
Call stack:
ebuild.sh, line 1539: Called dyn_compile
ebuild.sh, line 939: Called src_compile
gmp-4.2.1.ebuild, line 44: Called econf '--localstatedir=/var/state/gmp' '--disable-mpfr' '--disable-mpbsd' '--enable-cxx'
ebuild.sh, line 541: Called die
!!! econf failed
!!! If you need support, post the topmost build error, and the call stack if relevant.

[Redhatter]

Well... I've just unpacked a fresh stage3 2006.1 tarball on my IP28. Ran emerge --metadata (I mounted the external environment's /usr/portage via mount --bind). Did an env-update then ran the following:

[ddc]

to fix my problem:

1) i remerged gcc-3.4.6 (not the 4.1.1)
2) i emerged dev-libs/gmp
3) i didn't emerge kgcc64 as it still included in portage
4) i fix up bad names about mips64: the compile tool has aspected to have /usr/bin/mips64-linux-* not mips64-unknown-gnu-*
5) USE="ip30" emerge mips-sources has finished with a success !!!

then i (finally) build up the kernel (2.6.17), that is now running ....


all i have done is written in the previous post of mine.

could i see your /etc/make.conf ? (for ip28)

on octane2 i used to replace it with the make.defauld defined in ip30/o32 profile

[ddc]

about the pci-cage, i can read



sys-kernel/mips-sources-2.6.17.10
Things to keep in mind when building a kernel for an SGI Octane:
- Impact (MGRAS) console and X driver work, please report any bugs.
- VPro (Odyssey) console works, but no X driver exists yet.
- PCI Card Cages should work for many devices, except certain types like
PCI-to-PCI bridges (USB hubs, USB flash card readers for example).
- Do not use OHCI-based USB cards in Octane. They're broke on this machine.
Patches are welcome to fix the issue.
- Equally, UHCI Cards are showing issues in this release, but should still
function somewhat. This issue manifests itself when using pl2303 USB->Serial
 adapters.
- Other XIO-based devices like MENET and various Impact addons remain
untested and are not guaranteed to work. This applies to various
digital video conversion boards as well.


i've putted a matrox m2 inside this PCI-XIO bridge, dmesg is reporting that matroxfb: probe of 0001:01:01.0 has failed with an error.

[Redhatter]

Yep... not suprised about that at all. VGA cards are very "special", and thus rarely work on exotic architectures. I had the same thing with an ISA video card in the EISA slot on my IP28.

As for the make.conf on my IP28... there's nothing all that special about it. Most of the magic required is done by the default-linux/mips/2006.1/ip28/* profiles.
_________________
Stuart Longland (a.k.a Redhatter, VK4MSL)
I haven't lost my mind - it's backed up on a tape somewhere...

Gentoo/MIPS Cobalt developer, Mozilla herd member.

[Pliablemammal]

[Pliablemammal]