X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C7722E.CECE3CDB@onstor-exch02.onstor.net>; Thu, 29 Mar 2007 11:19:36 -0700
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Content-class: urn:content-classes:message
Subject: RE: bobcat linux rises again
Date: Thu, 29 Mar 2007 11:19:35 -0700
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E03112213@onstor-exch02.onstor.net>
In-Reply-To: <20070328222518.230d1ab9@ripper.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: bobcat linux rises again
Thread-Index: AcdxwqS4V+/02zO3RNqVCDZ82ztX3QAaIbhQ
References: <20070328222518.230d1ab9@ripper.onstor.net>
From: "Brian Stark" <brian.stark@onstor.com>
To: "Andy Sharp" <andy.sharp@onstor.com>
Cc: "Warren Gale" <warren.gale@onstor.com>

Andy,

In looking at the current filesystem support in the PROM, I see the
following in sm-pmondisk\filesystem.c:

struct fs_ops file_system[] =3D {
	{ ufs_open, ufs_close, ufs_read, ufs_write, ufs_seek, ufs_stat
},
};

These functions are in ufs.c, which looks to have been pulled directly
from bsd (with some minor mods).  Are you looking for the equivalent
support for ext2fs? =20

It looks like bsd has support for ext2fs, so maybe we could pull the
functions into the PROM in the same way as above?  I imagine we can also
get them from Linux, but that may mean more changes to other files in
sm-pmondisk.  It may also have implications with the whole Linux GPL
thing.


Brian


> -----Original Message-----
> From: Andy Sharp=20
> Sent: Wednesday, March 28, 2007 10:25 PM
> To: dl-cougar
> Subject: bobcat linux rises again
>=20
> I'm happy to report that we have attained the first stage of=20
> bringing Linux up on Bobcat, namely booting via tftp and=20
> using NFS root filesystem. There are a few rough spots (the=20
> device drivers for network and serial had to be hacked) and=20
> the real time clock is not accessible from userspace yet, but=20
> the system is up and is a fairly complete Linux system.  The=20
> next step is to get the CF working, and adding support for
> ext2 file systems to the PROM code.
>=20
> Yes, I was having trouble with Marvell interrupts, and yes,=20
> it was all Brian Stark's fault for hooking them up in such an=20
> odd manor ~:^)
>=20
>=20
> You can log into the machine with ssh:
>=20
> ssh guest@10.2.9.7
> password: guest1
>=20
> Here is the boot up messages, my personal favorite:
>=20
> Linux version 2.6.21-rc4-bc9k-g8db5b405-dirty (andys@ripper)=20
> (gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21))=20
> #365 Wed Mar 28 20:14:16 PDT 2007 CPU revision is: 00003440=20
> FPU revision is: 00003420 Determined physical RAM map:
>  memory: 0d000000 @ 03000000 (usable)
> Wasting 393216 bytes for tracking 12288 unused pages On node=20
> 0 totalpages: 32768
>   DMA zone: 256 pages used for memmap
>   DMA zone: 0 pages reserved
>   DMA zone: 32512 pages, LIFO batch:7
>   Normal zone: 0 pages used for memmap
> Built 1 zonelists.  Total pages: 32512
> Kernel command line: console=3DttyS0,57600n8=20
> console=3DttyMM0,57600n8 root=3D/dev/nfs=20
> nfsroot=3D10.0.0.42:/var/nfsroot/new-bobcat,v3=20
> ip=3D10.2.9.7:10.0.0.42:10.2.0.1:255.255.0.0:bobcat:eth0:none=20
> -s Primary instruction cache 16kB, physically tagged, 4-way,=20
> linesize 32 bytes.
> Primary data cache 16kB, 4-way, linesize 32 bytes.
> Synthesized TLB refill handler (27 instructions).
> Synthesized TLB load handler fastpath (39 instructions).
> Synthesized TLB store handler fastpath (39 instructions).
> Synthesized TLB modify handler fastpath (38 instructions).
> PID hash table entries: 512 (order: 9, 2048 bytes) Using=20
> 399.994 MHz high precision timer.
> Dentry cache hash table entries: 16384 (order: 4, 65536=20
> bytes) Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
> Memory: 78208k/81920k available (1900k kernel code, 3688k=20
> reserved, 439k data, 104k init, 0k highmem) Calibrating delay=20
> loop... 796.67 BogoMIPS (lpj=3D398336) Mount-cache hash table=20
> entries: 512
> NET: Registered protocol family 16
> bobcat_pci0_init: io_base =3D 0x0, io_size =3D 0x10000
> bobcat_pci0_init: mem0_base =3D 0x1d000000, mem0_size =3D=20
> 0x1000000 registering PCI controller with io_map_base unset
> bobcat_pci0_init: mv_io_base =3D 0x0, mv_io_size =3D 0x10000
> bobcat_pci1_init: io_base =3D 0x18000000, io_size =3D 0x800000
> bobcat_pci1_init: mem0_base =3D 0x1a000000, mem0_size =3D 0x2000000
> bobcat_pci1_init: mem1_base =3D 0x19000000, mem1_size =3D 0x1000000
> bobcat_pci1_init: mem2_base =3D 0x18800000, mem2_size =3D 0x800000
> bobcat_pci1_init: mem_base =3D 0x19000000, mem_size =3D 0x1000000
> bobcat_pci1_init: mv_io_base =3D 0x0, mv_io_size =3D 0x10000=20
> registering PCI controller with io_map_base unset
> bobcat_pci1_init: mv_io_base =3D 0x18000000, mv_io_size =3D 0x800000
> bobcat_pci_init: pci io_v_base =3D 0xb8000000
> PCI: Failed to allocate mem resource #0:20000000@20000000 for=20
> 0000:00:01.0
> PCI: Failed to allocate mem resource #2:20000000@20000000 for=20
> 0000:00:01.0
> PCI: Failed to allocate mem resource #4:2000000@1e000000 for=20
> 0000:00:01.3
> Time: MIPS clocksource has been installed.
> NET: Registered protocol family 2
> IP roDS1511 Real Time Clock Driver v1.0
> io scheduler noop registered
> io scheduler deadline registered (default)
> Serial: MPSC driver $Revision: 1.00 $
> mpsc_drv_probe: Adding MPSC 0
> ttyMM0 at MMIO 0xbc008000 (irq =3D 52) is a MPSC natsemi=20
> dp8381x driver, version 2.1, Sept 11, 2006
>   originally by Donald Becker <becker@scyld.com>
>   http://www.scyld.com/network/natsemi.html
>   2.4.x kernel port by Jeff Garzik, Tjeerd Mulder
> PCI: Enabling device 0000:01:00.0 (0006 -> 0007)
> PCI: Setting latency timer of device 0000:01:00.0 to 64=20
> natsemi eth0: NatSemi DP8381[56] at 0x19020000=20
> (0000:01:00.0), 00:07:34:02:9d:00, IRQ 8, port TP.
> PCI: Enabling device 0000:01:01.0 (0006 -> 0007)
> PCI: Setting latency timer of device 0000:01:01.0 to 64=20
> natsemi eth1: NatSemi DP8381[56] at 0x19021000=20
> (0000:01:01.0), 00:07:34:02:9d:01, IRQ 9, port TP.
> Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
> ide: Assuming 33MHz system bus speed for PIO modes; override=20
> with idebus=3Dxx
> pd6729: Cirrus PD6729 PCI to PCMCIA Bridge at 0x18000800 on irq 10
> pd6729: PCI_COMMAND_MEMORY enabled by firmware.
> pd6729: Enabling PCI_COMMAND_MASTER.
> pd6729: PCI card interrupts, PCI status changes i2c /dev=20
> entries driver TCP cubic registered
> NET: Registered protocol family 1
> NET: Registered protocol family 17
> drivers/rtc/hctosys.c: unable to open rtc device (rtc0)
> pccard: PCMCIA card inserted into slot 0
> eth0: DSPCFG accepted after 0 usec.
> eth0: link up.
> eth0: Setting full-duplex based on negotiated link capability.
> pccard: PCMCIA card inserted into slot 1
> IP-Config: Complete:
>       device=3Deth0, addr=3D10.2.9.7, mask=3D255.255.0.0, =
gw=3D10.2.0.1,
>      host=3Db00bcat, domain=3D, nis-domain=3D(none),
>      bootserver=3D10.0.0.42, rootserver=3D10.0.0.42, rootpath=3D=20
> Looking up port of RPC 100003/3 on 10.0.0.42 Looking up port=20
> of RPC 100005/3 on 10.0.0.42
> VFS: Mounted root (nfs filesystem) readonly.
> Freeing unused kernel memory: 104k freed ute cache hash table=20
> entries: 1024 (order: 0, 4096 bytes) TCP established hash=20
> table entries: 4096 (order: 3, 32768 bytes) TCP bind hash=20
> table entries: 4096 (order: 2, 16384 bytes)
> TCP: Hash tables configured (established 4096 bind 4096) TCP=20
> reno registered
>=20
