What is all the TBEMD stuff

I've been getting questions about what TBEMD means in my commits to FreeBSD head.

First, I'll tell you the name. TBE stands for TARGET_BIG_ENDIAN. For the MIPS and ARM ports, it is an environment variable that you have to set to build these targets for big endian (otherwise they default to little endian). MD is Must Die. A few years ago, there was an action movie whose title was more memorable than the file "Romeo Must Die." So I just stole the name for this branch.

So why must TARGET_BIG_ENDIAN die? It breaks things. For all the other platforms that FreeBSD supports, you just set TARGET and/or TARGET_ARCH to do cross builds. You know everything you need to know from the MACHINE and MACHINE_ARCH in the resulting image. In addition, the build system segregates things so you can build all the targets in one tree..But for mips and arm, you don't know what endian a binary is, the obj tree will collide if you try to build both little endian and big endian binaries at the same time. There's also other, more subtle issues. For example, TARGET_BIG_ENDIAN isn't set on big endian mips, so a native biuldworld tries to build little endian binaries (oops).

So, mips will be moving to mipsel (for little endian) and mipseb (for big endian). Arm will move to arm (for little endian) and armeb (for big endian). You could fill an entire email archive with all the possible other names and why they are better or worse than these names. History, however, trumps all those arguments: these are the names used elsewhere and we're just following convention.

Once tbemd is completely merged, you'll be able to build both endians of MIPS in the same object tree, for example, and all the other issues I've discovered. It also helps with new architectures as we move into powerpc64 (already merged) and mips64 (to be done after tbemd is collapsed). It will also mean we can have different packages for the different endians now.

The dangers of forgetting svn add

I spent a lot of time before the 8.1 release adding code to transparently map COMPAT_IA32 to COMPAT_FREEBSD32. I tested all kinds of different scenarios to make sure that it works, that it got error messages from old config programs, etc.

Today I got a complaint that COMPAT_IA32 doesn't work in 8.1-RELEASE. Linux module was failing to load properly. I thought it was really odd and that the folks complaining must be doing something wrong.

It turns out that I did something wrong. When I originally merged the work, I modified the options file, but didn't add the new options-compat file to stable/8 tree, so it didn't get merged into the releng/8.1 tree used to create 8.1. The result is that COMPAT_IA32 just fails silently in 8.1-RELEASE.

All that hard work ruined because I forgot to do a svn add... How disappointing.

silly vpn hack

I finally got frustrated by the inability of /etc/resolv.conf to do what I'd say in pseudo code:

if domain==example.com; then nameserver=1.2.3.4
else nameserver=4.3.2.1

since I need that when I'm on my VPN to example.com.

So, round one of the hack is to tell dhclient to use my local nameserver, and do it with named.

So, in /etc/dhclient.conf, I have:

interface "wlan0" {
prepend domain-name-servers 127.0.0.1;
}

and I have the following snippets in named.conf:

...
options {
...
forwarders {
4.3.2.1;
};
...
};
zone "example.com" IN {
type forward;
forwarders {
1.2.3.4;
};
};

The only problem I have yet to figure out is how to dynamically update the 4.3.2.1 in /etc/namedb/named.conf as I roam from network to network. There doesn't appear to be a hook in dhclient to say 'run this command when my IP address changes' or 'run this command when I get a new lease' or anything useful like that....

All this is with FreeBSD-current, but these techniques should work back to at least FreeBSD 6.x.

Finally, does anybody know a good way to quote code (as opposed to text) in blogger.com? By default, it seems, all leading whitespace is eaten, which make this post look ugly.

video on FreeBSD with skype

Hans Peter Selasky has been very busy adding support for web cams and other usb video devices. We tried a number of different cameras that we had laying around iX Systems the other day. All of them were recognized as valid video cameras. Some of them worked with pwcview and skype. Some of them had problems with the palettes, or would only work with some resolutions.

I was able to make a Skype call on my FreeBSD to another developer. Unfortunately, the distfile necessary for this isn't freely available. You have to download it from Skype. And it is no longer available from Skype. I got lucky and down loaded it a while ago, thinking I'd play with skype.

If you haven't had a chance to check it out, install the multimedia/webcamd. There's lots of devices that it works with.

Now, the interesting part. Hans Peter Selasky has implemented a user land driver for all of this. He implements a device driver in two parts. One piece in the kernel, which setups the shims to talk to webcamd, and the actual driver in webcamd. He uses the Linux drivers in webcamd. This separates the GPL drivers from Linux from the BSD kernel. So there's no issues with licenses. This is the same technique that's done in the Linux world to implement proprietary drivers and keep them separate from the Linux kernel. This opens the door for a wide array of devices. It is useful to think of this as something similar what Bill Paul did with NDIS and Windows drivers (except in two different address spaces rather than one).

Booting instructions for FreeBSD on Cavium Octeon

Here's a quick note on how to net boot the Cavium EBT3000 board running uboot. The Cavium kernel is still a work in progress as I restore all the fixes I made to an earlier version of this code that I was unable to release.

You'll need to break into the boot sequence for this board. Usually that's just hitting ^C before uboot starts to load the kernel.

Once you have a uboot prompt, similar to below, you'll be ready to start to fix the environment for netbooting.

U-Boot 1.1.1 (Development build) (Build time: Jan 26 2007 - 14:06:35)

EBT3000 board revision major:4, minor:0, serial #: xxxxxxxxx
OCTEON CN38XX-NSP revision: 1, Core clock: 500 MHz, DDR clock: 266 MHz (532 Mhz data rate)
DRAM: 2048 MB
Flash: 8 MB
IPD backpressure workaround verified, took 14 loops
Clearing DRAM........ done
BIST check passed.
Net: octeth0, octeth1, octeth2, octeth3
Bus 0 (CF Card): OK

ide 0: Model: SanDisk SDCFB-1024 Firm: Rev 0.00 Ser#: X0308 20050726142909
Type: Removable Hard Disk
Capacity: 977.4 MB = 0.9 GB (2001888 x 512)

Octeon ebt3800#

It couldn't be easier to setup for netbooting. You'll need to setup an tftp server. I had to install the FreeBSD port from net/freebsd-tftpd to get a working tftpd, since there appears to be some issues with the stock FreeBSD tftpd. Once I had that, I set different addresses like so:

set ipaddr 10.0.0.30
set netmask 255.255.255.0
set serverip 10.0.0.10
set gatewayip 10.0.0.1
set bootdelay 0
saveenv

This sets the board's address to 10.0.0.30/24. 10.0.0.1 is the router to the other networks. The tftp command will now use tftp from host 10.0.0.10 to get the named file. saveenv makes these setting permanent. I usually reboot at this point to make sure that the settings took.

Next, you'll need to know how to load a FreeBSD kernel. This is also fairly straight-forward, although finding the right addresses took a bit of guesswork. You are tftping the ELF kernel to a chunk of memory. The bootoctlinux command then takes that image and copies it to the right place and starts executing it. So you have to be sure whereever you load it doesn't clash with uboot's memory use or with where the kernel will ultimately reside.

tftp a800000 kernel.octean1-32
bootoctlinux a800000 numcores=1

As with any u-boot platform, it can often be useful to define macro commands. I did the following:

set fbsd 'tftp a800000 kernel.octeon1-32; bootoctlinux a800000 numcores=1'

so that I don't have to keep typing it over and over. I can just type

run fbsd

to run my next test.

The Cavium port is in a state of flux right now. The latest status as I am preparing this to be published is that there's an interrupt problem that's causing the serial port to be non-responsive after going to multi-user (eg, the login: prompt) and also the network stops responding. Also, I have to comment out the enableintr() call in mips_vector_init around line 365 of sys/mips/mips/machdep.c. I'm trying to hammer these issues out, and I think they are related. Earlier versions of the code worked with earlier versions of FreeBSD, so I'm sure this is just a shift in interfaces that needs to be accounted for.

Here's the latest slightly trimmed log from a boot:

U-Boot 1.1.1 (Development build) (Build time: Jan 26 2007 - 14:06:35)

EBT3000 board revision major:4, minor:0, serial #: xxxxxxxxx
OCTEON CN38XX-NSP revision: 1, Core clock: 500 MHz, DDR clock: 266 MHz (532 Mhz data rate)
DRAM: 2048 MB
Flash: 8 MB
IPD backpressure workaround verified, took 32 loops
Clearing DRAM........ done
BIST check passed.
Net: octeth0, octeth1, octeth2, octeth3
Bus 0 (CF Card): OK

ide 0: Model: 512MB CKS Firm: 2N3-0925 Ser#: 2435E7C34A52930007
Type: Removable Hard Disk
Capacity: 491.6 MB = 0.4 GB (1006992 x 512)
Octeon ebt3000# print
bootdelay=0
baudrate=115200
download_baudrate=115200
ls=fatls ide 0
nuke_env=protect off BFBFE000 BFBFffff; erase BFBFE000 BFBFffff
autoload=n
ipaddr=10.0.0.31
gatewayip=10.0.0.1
netmask=255.255.255.0
serverip=10.0.0.16
fbsd=tftp a800000 kernel.octeon1-32; bootoctlinux a800000 numcores=1
fbsd_cf=fatload ide 0 a800000 kernel.imp; bootoctlinux a800000 numcores=1
coremask_override=0xffff
numcores=16
stdin=serial
stdout=serial
stderr=serial

Environment size: 1274/8188 bytes
Octeon ebt3000# run fbsd
octeth0: Down 10Mbs Half duplex, (port 16)
Using octeth0 device
TFTP from server 10.0.0.16; our IP address is 10.0.0.31
Filename 'kernel.octeon1-32'.
Load address: 0xa800000
Loading: octeth0: Up 100Mbs Half duplex, (port 16)
########################
done
Bytes transferred = 3370489 (336df9 hex)
argv[2]: numcores=1
ELF file is 32 bit
Skipping non LOAD program header (type 0x6)
Skipping non LOAD program header (type 0x3)
Skipping non LOAD program header (type 0x70000000)
Allocated memory for ELF segment: addr: 0x1000000, size 0x295a60
Loading .text @ 0x81000000 (2064384 bytes)
Loading .MIPS.stubs @ 0x811f8000 (16 bytes)
Loading .rodata @ 0x811fa000 (33248 bytes)
Loading .reginfo @ 0x812021e0 (24 bytes)
Loading .rodata.str1.4 @ 0x812021f8 (189300 bytes)
Loading set_sysctl_set @ 0x8123056c (3496 bytes)
Loading set_sysinit_set @ 0x81231314 (1684 bytes)
Loading set_sysuninit_set @ 0x812319a8 (952 bytes)
Loading .interp @ 0x81231d60 (13 bytes)
Loading .dynsym @ 0x81231d70 (78320 bytes)
Loading .dynstr @ 0x81244f60 (72163 bytes)
Loading .hash @ 0x81256944 (35984 bytes)
Loading set_kdb_dbbe_set @ 0x8125f5d4 (8 bytes)
Loading set_modmetadata_set @ 0x8125f5dc (288 bytes)
Loading set_cons_set @ 0x8125f6fc (8 bytes)
Loading .data @ 0x8125f704 (116476 bytes)
Loading set_pcpu @ 0x8127be00 (3136 bytes)
Loading .got @ 0x8127ca40 (5296 bytes)
Loading .rld_map @ 0x8127def0 (4 bytes)
Loading .sdata @ 0x8127def4 (12 bytes)
Clearing .bss @ 0x8127df00 (97120 bytes)
## Loading Linux kernel with entry point: 0x81000000 ...
Bootloader: Done loading app on coremask: 0x1
Boot Descriptor Ver: 6 -> 1/2 CPU clock: 500MHz
Dram: 2048 MB Board Type: 2 Revision: 4/0
Octeon Chip: 0 Rev 0/0 Mac Address 00.0F.B7.10.1B.32 (14)
octeon_dram == 80000000
reduced to ram: 2048 MBReal memory bytes is 7ffff000
phys_avail[0] = 1296000 phys_avail[1] = fffffff
realmem_bytes is now at 5ffff000
Next block of memory goes from 20000000 to 7ffff000

Total DRAM Size 0x80000000
Bank 0 = 0x 1296000 -> 0x FFFFFFF
Bank 1 = 0x20000000 -> 0x7FFFF000

physmem: 0x6ed69Cache info:
picache_stride = 4096
picache_loopcount = 8
pdcache_stride = 128
pdcache_loopcount = 64
cpu0: Cavium processor v1.0
MMU: Standard TLB, 32 entries
L1 i-cache: 4 ways of 64 sets, 128 bytes per line
L1 d-cache: 64 ways of 1 sets, 128 bytes per line
Config1=0xbe3303da
Config3=0x80
Physical memory chunk(s):
0x1296000 - 0xfffefff, 248942592 bytes (60777 pages)
0x20000000 - 0x7fffefff, 1610608640 bytes (393215 pages)
Maxmem is 0x7ffff000
KDB: debugger backends: ddb
KDB: current backend: ddb
hz=100 cyl_per_hz:250000 cyl_per_usec:250 freq:250000000 cyl_per_hz:2500000 cyl_per_sec:250000000
Copyright (c) 1992-2010 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
The Regents of the University of California. All rights reserved.
FreeBSD is a registered trademark of The FreeBSD Foundation.
FreeBSD 9.0-CURRENT #141 r202986:202997M: Mon Jan 25 20:07:07 MST 2010
imp@paco-paco.bsdimp.com:/tmp/imp/be-obj/mips/pe/imp/svn/head/sys/OCTEON1-32 mips
real memory = 1859555328 (1815972K bytes)
Physical memory chunk(s):
0x013ab000 - 0x0fffefff, 247808000 bytes (60500 pages)
0x20000000 - 0x7d830fff, 1568870400 bytes (383025 pages)
avail memory = 1813180416 (1729MB)
mem:
nfslock: pseudo-device
null:
nexus0:
Compact flash found in bootbus region 3 (8 bit).
rgmii0: on nexus0
rgmx0: Ethernet address: 00:0f:b7:10:1b:32
rgmx0: type (null), full duplex
rgmx1: Ethernet address: 00:0f:b7:10:1b:33
rgmx1: type (null), full duplex
rgmx2: Ethernet address: 00:0f:b7:10:1b:34
rgmx2: type (null), full duplex
rgmx3: Ethernet address: 00:0f:b7:10:1b:35
rgmx3: type (null), full duplex
rgmii0: [FILTER]
cf0: on nexus0
clock0: on nexus0
clock0: [FILTER]
obio0 at mem 0x1 flags 0x1 on nexus0
uart1: on obio0
uart1: [FILTER]
uart1: fast interrupt
uart1: console (115200,n,8,1)
uart0: on obio0
uart0: [FILTER]
uart0: fast interrupt
uart0: console (115200,n,8,1)
Device configuration finished.
Reducing kern.maxvnodes 117045 -> 100000
Timecounter "MIPS32" frequency 250000000 Hz quality 800
Timecounters tick every 10.000 msec
GEOM: cf0: partition 2 does not start on a track boundary.
GEOM: cf0: partition 2 does not end on a track boundary.
GEOM: cf0: partition 1 does not start on a track boundary.
GEOM: cf0: partition 1 does not end on a track boundary.
GEOM: cf0s2: geometry does not match label (64h,32s != 16h,63s).
Trying to mount root from ufs:cf0s2a
WARNING: / was not properly dismounted
warning: no time-of-day clock registered, system time will not be set accurately
start_init: trying /sbin/init
Setting hostuuid: 1c99760e-1dd2-11b2-bd6a-00156dc12838.
Setting hostid: 0xb0f278c1.
No suitable dump device was found.
Entropy harvesting:.
Starting file system checks:
t_delta 16.010fac783006c1a6 too long
/dev/cf0s2a: 11100 files, 256973 used, 150946 free (426 frags, 18815 blocks, 0.1% fragmentation)
Mounting local file systems:.
t_delta 15.fece5003dd62224a too short
Setting hostname: ebt3000.
SIOCSTIFFLAGS UP/Running
Starting Network: lo0 rgmx0 rgmx1 rgmx2 rgmx3.
lo0: flags=8049 metric 0 mtu 16384
options=3
inet 127.0.0.1 netmask 0xff000000
rgmx0: flags=8843 metric 0 mtu 1500
options=3
ether 00:0f:b7:10:1b:32
inet 10.0.0.31 netmask 0xffffff00 broadcast 10.0.0.255
media: Ethernet autoselect
rgmx1: flags=8802 metric 0 mtu 1500
options=3
ether 00:0f:b7:10:1b:33
media: Ethernet autoselect
rgmx2: flags=8802 metric 0 mtu 1500
options=3
ether 00:0f:b7:10:1b:34
media: Ethernet autoselect
rgmx3: flags=8802 metric 0 mtu 1500
options=3
ether 00:0f:b7:10:1b:35
media: Ethernet autoselect
Starting devd.
Creating and/or trimming log files.
Starting syslogd.
/etc/rc: WARNING: Dump device does not exist. Savecore not run.
ELF ldconfig path: /lib /usr/lib /usr/lib/compat
Clearing /tmp (X related).
Updating motd:.
Starting cron.
Starting inetd.
Starting background file system checks in 60 seconds.

Tue Jan 26 02:07:16 UTC 2010

FreeBSD/mips (ebt3000) (ttyu0)

login:

DLINK DIR-615 REV C1 Redux

After hitting a dry hole with my DIR-615 REV A1, I thought I'd try the DIR-615 REV C1 that I have. It has an Atherose AR9130 in it. This is very similar to the AR7130 that the RouterStation I have running, and I thought it would be a simple matter to use uboot to netboot this code.

Again, I was mistaken. There's no way to break into the boot sequence for this device. The best I can hope for is to create a compressed image that's < 1MB and blow it into the FLASH partition for the kernel under linux. That would work out OK, if I had known good image to start from. Alas, I don't. And there doesn't appear to be an accessible JTAG header that I can use to write a new u-boot. Linux seems to protect u-boot from being overwritten as well.

But the biggest issue of all is that the FLASH is only 4MB in size. It would take some doing to get everything shrunk down that small. The DIR-615 REV A1 at least had 8MB of FLASH, which is doable. Maybe I'll perfect the kernel for the DIR-615 REV A1 on the Marvell eval board I have. Looking at the Linux sources, the deltas between these kernels are small and inconsequential for bootstrapping purposes. Hmmm, the Linux kernel just has support for the DNS-323, now that I look closely, but the required bits of hacking after I have a known good kernel doesn't seem huge. Plus, if I could get one good kernel on the box, then I could put a better uboot on there which would support the network for netbooting...

If I ever make good images for these boards, I'll be sure to post pointers to them here.

DLINK DIR-615 REV A redux

Yesterday I burned a couple of hours playing with the DIR-615 REV A1 that I have had for ages. FreeBSD runs on the Marvell Orion SoC that's inside it now and I thought I'd netboot the box and try to put together an image that can replace my firewall since this little box uses less power and my firewall doesn't need to do *THAT* much.

Unfortunately, it appears as if the network was purposely crippled on this box. Not disabled entirely, but crippled to not work. Sometimes you can get packets out, but generally there's a huge packet loss rate that makes pinging impossible, let alone netbooting.

Talking to other folks that have worked in this area, it appears that this was a deliberate attempt to discourage others from running alternative operating systems on the box.

Since the box will still boot to Linux, there's hope that one can bootstrap the FreeBSD installation from there. When I get some time again, I'll be pursuing that avenue. Maybe I can find the JTAG header for this board and reflash uboot that way, since debugging a new kernel can be difficult to get right in one shot on hardware the kernel has never run on before...

I had planned on publishing instructions for netbooting the thing, but alas, it appears that isn't possible.

Post-mortum on projects/mips branch

Greetings to one and all. As you have read elsewhere, I recently merged all the changes from the projects/mips branch onto head. In other reports, I've made cryptic reference to the branch being damaged. I thought I'd go through all the problems we encountered running this development effort on the projects/mips branch.

First, we created the branch in the normal way:

svn cp svn+ssh://svn.freebsd.org/base/head svn+ssh://svn.freebsd.org/base/projects/mips

and then started committing to it. So far so good.

Then it came time to merge in changes from head. Here is where we made our first mistake. We merged just the kernel changes, and not the entire tree. The next time we merged the entire tree. The effect of this sequence meant that the changes between the creation and the first merge in the rest of the tree (outside of src/sys) were lost. I don't know if this is the result of a buggy svn client, or if they are a fundamental flaw in svn. So when it came time to collapse the results back into head, we couldn't just do a reverse integration, or even a more conventional merge.

Next, we pulled in pieces of the projects/mips tree into head, either by hand or with the svn merge command. Doing both was a mistake, I think. Although svn coped with later merged from head into projects/mips fairly well, occasionally we'd have issues to sort out.

So, when it came time to merge everything back into head, we were left with a number of difficult choices. We opted to copy the new files/directories in the repo (to preserve the history) and merge patches by hand with svn log entries. The latter went really well. There were no problems with it apart from the odd fuzz and .rej file to sort out (no different than a normal merge). The former, however, causes a lot of problems.

First, it created svn:mergeinfo entries on the files I copied over. This conflicts with the project's "merge everything from sys" edict. This happened because I did the copy in the server rather than the client (which was an attempt to preserve history). It was a simple matter to delete these entries when it was brought to my attention.

Next, somehow we created the files on the branch without the required svn:keywords, svn:eol-style and svn:mime-type properties. This caused problems when we went to commit changes to these files. The precommit complained, at this point, that the files lacked svn:keywords, and we had to add $FreeBSD$ by hand (since the copy didn't complaint that they were lacking). Also, I didn't discover these problems on my own. Other sharp-eyed individuals saw the problems and reported them to me.

Finally, since I did the copies on the server, there was no way to batch them. When I copied a directory, I got the whole directory. When I copied a file, I got the file. Each one generated a commit message.

I'm unsure what other damage was lurking in the projects/mips tree, but the lack of properties, the inability to easily use merge info and the missing commits lead me to the conclusion that it was easier to abandon the branch and create a new branch if needed in the future.

The mips team plans to create more branches, that are more for special purposes, that will be shorter lived in the future. We have no further plans to have one monolithic mips branch that acts as a staging ground for commits into head.

Hack to allow automatic wired/wireless failover with lagg on FreeBSD

For years I've had a love/hate relationship with my wireless card. On the one hand, it allows me to roam. On the other I like being able to plug into the wired Ethernet when I'm sitting at my desk since I get better throughput and lower latency. So for years, I just loaded and unloaded kernel modules to make dhclient (and/or wpa_supplicant) do its thing. This changed the IP address of my box, which messed up NFS mounts until I went back to the old way.

Recently, Josh Paetzel came up with a way to use lagg to do automatic failover. Inspired by the lagg man page, he created a number of rc.conf lines that would do this automatically. Xin Li and I refined the trick a bit. To completely integrate this into FreeBSD, I suppose that the rc.d scripts would need to be enhanced to automatically generate these lines when appropriate. However, the hack is so easy to do now, I'm going to share it with you here. Xin Li is writing up a longer description of how to set this up for the handbook. Let's say you have a ath0 wireless interface, and a re0 wired interface, just add the following to your /etc/rc.conf file:

• ifconfig_re0="up"


• ifconfig_ath0=`ifconfig re0 ether`


• ifconfig_ath0="ether ${ifconfig_ath0##*ether }"


• wlans_ath0="wlan0"


• ifconfig_wlan0="WPA"


• cloned_interfaces="lagg0"


• ifconfig_lagg0="laggproto failover laggport re0 laggport wlan0 DHCP"

This bit of code brings up re0, programs ath0 with the ethernet MAC address of re0 (so your MAC doesn't change when you switch over), creates a wlan0 interface from ath0, causes lagg0 to be created and configured to do the fail over.

Hope this tip is useful...

Revision at Wed Jan 13 16:00:00 UTC 2010:

Note: "ifconfig re0 ether" doesn't work in versions prior to 8.0. The following should fix it, but I don't have a system to test it on:

ifconfig_ath0=`ifconfig re0 ether`

ifconfig_ath0="${ifconfig_ath0##*ether}"

ifconfig_ath0="ether ${ifconfig_ath0%%inet*}"

This sequence avoids using grep, awk and/or sed to accomplish this (and thus touch /usr too early for systems where / and /usr are on different partitions).

Also corrected the bwi0 reference in the feedback...

Revision at Mon Jan 18 23:00:00 UTC 2010:
Corrected () vs {} confusion in the above...

FreeBSD/mips updated

The base/projects/mips branch has been merged into base/head. The merge is complete and the sanity tests have passed. The code has booted on both a Ubiquiti RouterStation (big endian) as well as in gxemul (little endian).

The branch lived for one year, minus a day, and accumulated much work:

• A new port to the Atheros AR71xx series of processors. This port supports the RouterStation and RouterStation PRO boards from Ubiquiti. Other boards should work with minimal tweaking. This port should be considered as nearing production quality, and has been used extensively by the developers. The primary author of this port is Oleksandr Tymoshenko (gonzo@freebsd.org).


• A new port to the sibyte BCM1250 SoC on the BCM91250 evaluation board (aka SWARM). This port is reported to be stable, but this hardware is a little old and not widely available. The primary author of this port is Neel Natu (neel@freebsd.org). Only one core is presently supported.


• A port, donated by Cavium, to their Octeon and Octeon plus series of SoC (CN3xxx and CN5xxx). This code is preliminary, supporting only a single core right now. It has been lightly tested on the CN3860 evaluation board only in 32-bit mode. Warner Losh (imp@freebsd.org) has been driving the efforts to get this code into the tree.


• A port, donated by RMI, to their XLR series of SoCs. This port is single core only as well. The code reaches multi-user but should be considered beta quality for the moment. Randal Stewart (rrs@freebsd.org) has been driving the efforts to integrate this into the tree.


• Preliminary support for building a mips64 kernel from this source base. More work is needed here, but at least two kernels successfully build in 64-bit mode (OCTEON1 and MALTA64).


• Very early support for N32 and N64 ABIs


• Support for booting compressed kernels has been added (gonzo@).


• Improved support for debugging


• Improved busdma and bus_space support


• Many bug fixes


• More types of MIPS cores are recognized


• Expanded cache handling for newer processors


• Beginning of a port to the alchemy au1XXX cpus is present but experimental.


• Work on SMP is underway to support multicore processors like the sibyte, Octeon and XLR processors.

I'm sure there are minor items I've forgotten. If so, please forgive any omission on my part...

The branch had been updated incorrectly several times over the past year, and the damage was too much to repair. We've retired the branch and will do further mips development in "head" for the time being. If you have a checked out tree, the suggested way to update the projects/mips tree you have is to do a "svn switch svn://svn.freebsd.org/base/head" in that tree.

I'd like to thank everybody that has contributed time, code or hardware to make FreeBSD/mips better.

We are still investigating how feasible merging all this work into stable/8 will be, as it represents a huge leap forward in code stability and quality.

As development proceeds, I'll keep posting updates. In addition, I hope to have some mini "how-to" wiki pages done for people that want to try it out.

Some TiVo hacking

Recently, I started hacking a TiVo HR10-250 that I've had for a million years unhacked.

Back when I first got it, there were dozens of threads of hundreds of messages in about 10 different forums you needed to read in order to hack the goofy thing. Everybody did their own thing, it was manual and a big pain.

Then came zipper (you can find all the details out at this link). It makes it very easy to hack the TiVos in general. However, there was just one problem. The ISO authoring tools it used were windows-based. Since I don't generally run on a Windows machine, this was a problem for me. So I tried to do the steps from the batch file by hand, and it didn't work out so well.

So I rewrote the batch file in a unix script running on FreeBSD. FreeBSD's iso extraction feature of tar made this process hassle free. I can build a new ISO image from an old one using only one port (mkisofs) and no privs are needed at any step. I thought about trying to teach tar how to create an iso image, but since mkisofs does such a good job, and there are *SO* many knobs and dials that need to be implemented to do that right, I just let mkisofs do its thing. I just can't get over how easy FreeBSD's tar made it to break apart the old bootable cd that does the hacking to add the zipper files. Kudos!

I've uploaded the script to the TiVo Community Forum thread about zipper if you are interested in looking into it.

T-Mobile and iPhone 3.1.2 firmware update issue

Recently, I used PwnageTool on my mac to update my 2G iPhone from 2.2.1 to 3.1.2. I've been using T-Mobile since day one with my iPhone without hassle. Well, except for the upgrade hassles that I have from time to time. Once I unlocked the SIM, I was able to use the SIM from my T-Mobile DASH in the phone.

When I upgraded this time, however, there was a problem. My data connections stopped working. Since 3.1.2 was otherwise a nice upgrade for me, I went looking for the problem. It turns out that the bundled carrier files that describe what you can and can't do with the iPhone basically said no data.

Carrier files that are correct for T-Mobile, and the data plan I have, can be found iPhone-Dev.ca under the IPCC section of the web page. Instructions for loading the IPCC file were there as well. For T-Mobile, you'll need to call T-Mobile to find the right APN to use. For my plan it turned out to be wap-apn. Your milage may vary, but worst case would involve trying all 4 of the files.

While not a programming hack, this should come in handy for anybody that's in a similar situation. Alas, I have no experience with the 3G or 3GS phones, and problems with the upgrades for those phones seem to be more widely covered (and blogged about).

Broadcom BCM43xx support (bwi) committed

The BCM43xx driver from DragonFlyBSD has been ported to FreeBSD. Andrew Thomas and others have done the port. I've pushed it into the tree, and plan on fixing the one or two quirks with my card.

The driver is far from perfect. 802.11a isn't support. Some older BCM4306 cards don't support channels 1, 2 or 3. It is rumored not to work at all for some people.

The driver was written by Sepherosa Ziehau for DragonFlyBSD. He used the specs available at http://bcm-specs.sipsolutions.net/Specification to write the driver. The driver is written against the v3 firmware. The v4 firmware is needed for support of newer cards, and many people have made noise about porting to that, but nobody has emerged yet to do that work.

The driver will be in FreeBSD 8.0. A back port of the driver to 7.x is difficult because it relies on all the new VAP work that Sam Leffler has done that is not in 7.x (although backports exist for 7.x, they won't be committed to the FreeBSD tree because it breaks binary compatibility).

You need the net/bwi-firmware-kmod port installed for the firmware. Although the binary .o file that the firmware is extracted from is apparently easy to distribute, distributing derived works of that .o file is not. So, there's a port to download, extract the firmware and make a kld for the driver to access.

A man page will be committed soon.

Belkin F6D4230 "Enhanced Wireless Router"

Just got the Belkin Enhance Wireless Router F6D4230-4. Looks like it has a ralink RT3052 or RT3050 under the hood. This is a MIPS24KEc based SoC, with a full Linux GPL tarball available from Belkin's web site.

I haven't had a chance to open the box yet to see what else is inside, and this is inferred entirely from the Linux config file that was in the GPL compliance package.

Not sure what I'm going to do with this box...

ExpressCard

Now that I've managed to knock down the number of bugs in NEWCARD to a more reasonable level, I've started looking at ExpressCard again.

As you may know, FreeBSD supports ExpressCard right now for usb devices. They all work, and there's no additional work needed here, except maybe to make more drivers. In fact, I recently purchased an ExpressCard to CardBus adapter (one that lets me plug in an expressCard into a CardBus slot) and it worked with all the usb-based ExpressCards that I have laying around with no changes.

The problem with ExpressCard for PCIe-based devices is one of resource allocation. If the BIOS allocates the resource, then the ExpressCard works without hot-plug. If the BIOS doesn't, then we can see the device with pciconf -l, as well as probe the device. However, since there's no resources setup for the PCI bridge, attempts for the driver to allocate them fail. There's also some bus numbering issues as well.

I've started to attack the resource problem. I'll keep people posted

Update my old CardStatus page...

If anybody still cares about 16-bit PC Cards and 32-bit CardBus cards still, I've updated the latest testing I've done.

You can find it here. I think it does more to show what a big pack-rat I am, but maybe someone will find it useful. It shows I'm down to one or two ed cards that aren't working for me yet...

3Com 3C1 works

Years ago, I bought a 3Com 3C1 CF card. This card was an early attempt by 3Com to minimize the power usage of the traditional 3C589 line of cards. These cards were important for the early palm-sized PDAs that were on the market. Through a number of different connections, I was able to get documentation on the 3c1. At the time I got it, I barely understood network drivers or even how the hardware worked. The documentation I got from 3com had pages for the flow charts for transmit and ISR, but the flowcharts weren't actually included on those pages. My connections couldn't get a better copy of the documentation, always saying it was coming soon.... The descriptions of the commands had some vague hints, but I never could make them connect. This was around 1999 or so.

Over the years, I kept trying to make this card work on FreeBSD. I considered it unfinished business from my past and I kept trying to find time to work on it. I never could find the time, or if I had the time my brain was burned out from the deliverables I had to produce for work. Over the years I've acquired all the cards that I could find in the 3c5xx line and made them work, but never got the 3c1 working.

Recently I stumbled across the 3c1 documentation and had a few minutes to read it. It was a nice change of pace from all the other stuff I'd been working on lately. After reading through it, I realized that there were a few key items that I'd not properly implemented before. There was a cryptic new command to turn on and off the TX PLL. So I had 30 minutes the other night and started reading the document again.

In the interim between when I started looking at the 3c1 documentation and now I did a lot of work in the high precision time and frequency world. This world is all about PLLs and controlling the oscillators/clocks to an insanely high degree of precision. So this clicked in my brain "Hmm, TX PLL sounds like it controls the clock used to do the transmission." So everywhere we enabled TX in the driver, I added a call to turn on TX PLL. This got me started: dhclient worked! I was able to transmit packets. This was the first. However, it stopped working after a while. So I searched through the 3c1 docs. There was nothing about PLL enable and disable, except for the documentation of the command. It appears there's an automatic timeout in the hardware which was turning off the PLL. So, an additional enabling of the PLL in start did the trick. I only enabled the PLL when we're not actively transmitting (which means we were idle for some unknown period of time). I could refine it further if I knew what the timeout of the PLL (which isn't documented), but there appears to be no ill effects doing it all the time the TX unit has been idle.

Ah, one more piece of unfinished business from the past is now complete. I can move on to other more interesting things...

More PC Card enhancements

Every few months/years I take all the cards that I have in my collection and start to run through them testing to see what's working in FreeBSD and what's broken. Over the past year or two I've neglected this task as I've been involved in other things. After my last trip to Akihabara, I've been playing with this stuff, so I thought I'd pull out the cards again. However, for the moment, I'm restricting things to the cards supported by the ed driver.

I've already blogged about the AX88x90 enhancements. The only additional enhancement is that I've eliminated some code that duplicates reading the NIC from what is just memory. I'll be committing that soon.

I've finally implemented the workaround for fullduplex disabling SQE in the DL10019 parts. This gives slightly better performance on the parts that are affected, but I'm having trouble measuring it. I'm not sure that it is worth the hassle.

I've also made a number of cards work by tweaking the MII initialization. I think I have a good way to kick the PHYs that are on these cards to get them working (this has helped to get three cards going).

I'm approaching 100% of the ne-2000ish cards working. I'll do a final cleanup of my patches and try to get them into the tree. It is time to move on to the CardBus cards. Most of them appear to work, once certain resource allocation issues are addressed. At the very least I need to test them all to see how they work. I may skip this and move directly and address the resource issues there.

AX88x90 updates

After focusing so much time on that old Toshiba card, I thought I'd see if I could get something more modern working as well. I've had one card in my bag knocking around in my bag based on the AX88790 chip. This is the corega FEther II-PXD. I saw it on sale in Akihabara when I was there a few weeks ago, and this card came from Akihabara a few years ago from a friend in Japan. So I thought I'd try it out.

Well, at first it didn't work too well. The MII code was finding gobs of PHYs that were phantoms. This in turned caused bad things to happen when dhclient(8) was run.

There turned out to be three root causes. First, the mii bit-bang code was wrong for reads, yielding bad register reads. Second, the AX88790 has an odd quirk for talking to its internal PHY, so more code was needed to cope with that. Third, the reset code for the NE2000 was also causing problems, so I rewrote it to be AX88x90 reset code. Finally, the internal phy for the AX88790 has a few quirks in silicon that need software workarounds.

Once I'd corrected these problems, the card started working again. Since most of this data is readily available on the net (the asix web site has datasheets for these parts), I'll skip over all the blow-by-blow details. I hopefully had enough comments in the code for people to follow along, as well as the datasheet should make what I've done easy to see.

Now, the arduous task of testing all the AX88x90 cards in my collections... The target card now works correctly, and I believe all of the ones in my collection will work too. Maybe I'll get to the point were all my ed-based PC Cards in my collection will work (except for the Mitsubushi B8895 which I've given up on, and the ones I don't have proper dongles for).

Toshiba LANCT00A PC Card

When I was in Akihabara last, I got a lot of PC Cards from the junk vendors for 100 Yen. One of these cards was the Toshiba LANCT00A. This card didn't probe when I first got it. FreeBSD didn't support it. This was a challenge.

As long-time reads of this blog may know, I sometimes go out of my way to hack on really oddball hardware that somehow I get it into my head needs to work. Never mind that I have a dozen other projects cooking that is more relevant to FreeBSD and its user base. Never mind that some of this hardware was obsolete when I started hacking on FreeBSD's PC Card layer. The cool thing is that I can find just enough documentation to be interesting on these cards.

This Toshiba card was like that. Google searches for information about it turned up a half dozen posts to different FreeBSD and Linux mailing lists asking about it, plus some speculation about what's in the card. One brave soul even took his card apart and reported it had a NS DP83902 under the hood. However, nobody that I can find ever got this card working under FreeBSD or Linux. Checks of the current Linux kernel was no help. Checks of OpenBSD and NetBSD also yielded no results. There was a windows driver available (actually an NDIS driver of unknown version).

So I was printing out random values for this card. At one time the FreeBSD PC Card probe routine for ed tried to probe WD80x3 cards as well as NE2000 cards. I removed this about 4 years ago after failing to find any cards that needed this, used it, etc. However, this card is almost a WD80x3 clone. I was printing the board type in the attach routine, trying to track down why attach was failing. I saw that it was 20 decimal or 0x14. A quick scan of the if_edreg.h file showed this was one of the Toshiba cards. The comments said it was a PCETC board. I immediately thought "Hmmm, isn't this a Toshiba card?" and then wondered if the 'C' in LANCT00A meant it was the same silicon revision as the 'C' in the PCETC. So I hacked up a quick little test into the if_ed_pccard probe routine.

It recognized it the card when I did that! Success. Well, almost. There was a snag. The memory allocated for this card was at 0x88000000. This didn't pass through the sanity test in the driver designed for ISA cards, so it didn't work. A little more hacking and I was getting a sane MAC address and the card was attaching. It dhcped its address w/o incident.

However, all is not right in mudville. As soon as I started using TCP, I started getting sbdrop panics. It looks like the packets that the WD80x3 code is pumping out aren't quite rightly aligned, or something (I have never had a sbdrop panic, so I'll have to investigate what that means when I have some time). This isn't too terribly surprising, since I haven't tested the ISA cards in years. Maybe something broke for the memory copyout path, but not for the PIO copyout path. I'll have to track that down another night.

[[ UPDATE: It looks like it was the copyin path. There was a bug in it that caused 2x too many bytes to be read into memory, causing the corruption. So now it works well enough for me to update this post and commit the change. ]]

Now, what does this gain FreeBSD? Not much, in the grander scheme of things. I'd be surprised if any of my readers actually had one of these cards and started using it as a result of my work. However, it was a relaxing hour or two of my life. Given my current job stress level, this has great value to me. Also, I think this is one of the first PC Cards to work with both I/O and Memory windows mapped at the same time from the CFE entry. While there are some other cards that manually allocate memory ranges, I don't think any of them have it in their CIS.

Postscript

Well, while I was doing research for this blog entry, I stumbled accross http://linux.toshiba-dme.co.jp/linux/eng/download.htm which has a tos_cs for download. There's a lot of voodoo magic in that driver that I'll have to investigate. This driver is 1000 lines long, and was for Linux 2.0.35. It was never integrated into Linux, so I'd stand by my statement that Linux doesn't support it. Just for kicks I tried building it under 2.6.29, and it didn't compile anymore.

My changes to the ed driver in FreeBSD are more like 20-25 lines of code. There's magic kick of a magic offset in shared memory that isn't present in FreeBSD's ed driver today. I'll keep that in mind in case I have the card wedge on me under load or something.