I-Appliance BBS
The Official Source for Internet Appliance Upgrades and Mods
Amazon Honor System Click Here to Pay Learn More
BBS Main List | Sign In | Sign Up | Search | Help | Linux-Hacker.netReply to Thread | Printer |

Home / Other I-Appliances / WebPal
IT WORKS !!! WEBPAL BOOTING LINUX !!!!!!
IT WORKS !!! WEBPAL BOOTING LINUX !!!!!!

New MessageIT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) codeman
Profile | Email
Ok well i now have my webpal booting linux 2.2.14, was not to bad the main problem was the fact that the IDE interface never worked
period do to a hardware design error, easy to fix tho just lift or cut pin 11 of U2

also the bogo mips are 1/2 of what it should be the cpu clock is div 2 will fix that soon

here is the boot output using a compact flash card

still have to put up a web page for it.

codeman

linux-hacker.net BootLoader!
Moving Image to ram
Linux version 2.2.14-rmk5-np17 (root@linux-hacker.net) (gcc version 2.95.2 19991024 (release)) #855 Sun Jul 16 17:38:19 PDT 2000Processor: ARM/VLSI ARM 710 revision 0
NetWinder Floating Point Emulator V0.94.1 (c) 1998 Corel Computer Corp.
RTC: failed to read the year
RTC: failed to read the time and date
Calibrating delay loop... 28.26 BogoMIPS
Memory: 7392k/8M available (556k code, 20k reserved, 220k data, 4k init)
Dentry hash table entries: 1024 (order 1, 8k)
Buffer cache hash table entries: 8192 (order 3, 32k)
Page cache hash table entries: 2048 (order 1, 8k)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
Starting kswapd v 1.5
Serial driver version 4.27 with no serial options enabled
ttyS00 at 0x02f8 (irq = 12) is a 16550A
ttyS01 at 0x03f8 (irq = 10) is a 16550A
hda: SunDisk SDCFB-8, ATA DISK drive
ide0 at 0x170-0x177,0x376 on irq 11
hda: SunDisk SDCFB-8, 7MB w/1kB Cache, CHS=245/2/32
Partition check:
hda: hda1
hda: hda1
hda: hda1
EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended
VFS: Mounted root (ext2 filesystem).
Freeing unused kernel memory: 4k init
/proc/meminfo
total: used: free: shared: buffers: cached:
Mem: 7565312 229376 7335936 4096 12288 4096
Swap: 0 0 0
MemTotal: 7388 kB
MemFree: 7164 kB
MemShared: 4 kB
Buffers: 12 kB
Cached: 4 kB
SwapTotal: 0 kB
SwapFree: 0 kB
/proc/cpuinfo
Processor : ARM/VLSI ARM 710 rev 0 (v3l)
BogoMIPS : 28.26
Hardware : Acorn-RiscPC
Revision : e3a04201
Serial : e3a06601e3a05542
/proc/devices
Character devices:
1 mem
2 pty
3 ttyp
4 ttyS
5 cua
10 misc

Block devices:
3 ide0
/proc/filesystems
ext2
nodev proc
/proc/interrupts
5: 85 timer
11: 38 ide0
12: 2 serial
/proc/ioports
0170-0176 : ide0
02f8-02ff : serial(auto)
0376-0376 : ide0
03f8-03ff : serial(auto)
/proc/loadavg
0.00 0.00 0.00 1/5 5
/proc/misc
/proc/mounts
/dev/root / ext2 rw 0 0
<NULL> /proc proc rw 0 0
/proc/stat
cpu 0 0 53 33
disk 16 0 0 0
disk_rio 16 0 0 0
disk_wio 0 0 0 0
disk_rblk 32 0 0 0
disk_wblk 0 0 0 0
page 18 0
swap 0 0
intr 126 0 0 0 0 0 86 0 0 0 0 0 38 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ctxt 37
btime 943920000
processes 5
/proc/uptime
0.87 0.33
/proc/version
Linux version 2.2.14-rmk5-np17 (root@linux-hacker.net) (gcc version 2.95.2 19991024 (release)) #855 Sun Jul 16 17:38:19 PDT 2000

07-17-2000 09:07:58

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) Wow!
Cool! I was just about convert the WebPal into a legacy picture frame ( cut a big hole in it and mount a picture ).
07-17-2000 10:11:52

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) nullbert
Excellent!!! NICE JOB!

So, what's next? The thing has an ISA slot, no? is one of the serial ports listed (one of the 2 16550's) actually the port in the modem?

maybe we can compile a driver for an ISA network card. That would be great!

Does this ARM processor have enough power to play back MP3's?

07-17-2000 18:16:10

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) redwood
wtg codeman! thanks for the hard work!
now, lets hope we can keep this part of the board positive and complementry. I guess its time to pull the screws? hehe, waiting for more info....
07-18-2000 07:00:00

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) clue
1. WebPal has been around for months now.
2. No one did this.
3. IDIOT: "codeman didn't do anything special, he just connected a IDE cable to the I-Opener"
4. Idiot proven wrong.

To all other idiots: Some people have a skill because they didn't spend all their life drinking beer and watching football, so shut up!

Good going codeman!

07-18-2000 07:58:27

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) Linuxguru
Hats off to Codeman!

However, the machine itself has weak specs compared to the Websurfer. 28.26 Bogomips may not be enough even for MP3 decoding.

07-18-2000 11:04:14

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) codeman
Profile | Email
> 28.26 Bogomips may not be enough even for MP3 decoding.

you have to remember that the clock is in div 2 mode the bogo mips will actualy be ~59 bogo mips

codeman

07-18-2000 11:27:19

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) nullbert
from what I've heard, a 486-133 is the minimum to get generally high quality MP3 playback. My 486-133 box is reporting 66.36 bogomips under a 2.2x kernel. So, maybe there's enough juice in this thing to do the decoding; we'll have to see!
07-18-2000 13:23:23

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) blah
perhaps a bigger issue is getting the sound working on the webpal - decoding the mp3s isn't useful unless you can get output...

I'm getting tempted to pick one up though.

07-18-2000 14:03:50

New MessageInfo?? (modified 0 times) none
The CL7500FE *might* just have enough processor power to do MP3 decoding. Unlike most other chips that incorporate the ARM core, Cirrus chose to add the FPU...

Now on to the real question: where can we get the source for the bootloader, kernel patches, etc., etc.??

Thanks!

07-18-2000 15:03:03

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) nullbert
exactly... I'm getting a lot more excited about this now!

I wonder if hardware drivers will compile right away, or does the architecture have a different concept of IRQ's, DMA's, I/O Ports than the x86 architecture has??

could you compile a soundblaster driver, or a NIC driver for this thing now?

07-18-2000 15:45:39

New MessageSound Chipset? driver? (modified 0 times) nullbert
the hardware design doc for the CL embedded controller
http://www.crystal.com/pubs/hw-7500.pdf

mentions that the controller has a dedicated serial output intended for driving a 16 bit DAC- the development/eval board uses a Crystal CS4333. Anybody with a web pal see this chip on it? I couldn't find it in the motherboard pics listed here.

Chances are that Teknema might have designed the webpal closely to the development board.

07-18-2000 21:36:04

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) transiit
right near the cirrus logic cpu, a little tiny chip, 8-pins, marked "C 4333-KS UOH9704"

-transiit

07-18-2000 21:39:54

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) Linuxguru
A comment after looking at the boot messages psoted by codeman: The version of the kernel that he has used appears to be for the ARM6 (ARM-RiscPC) target. Apparently, this uses Floating Point emulation (soft float). The ARM7 has a HW FPU, so a version of Arm-Linux targetted at it should give better performance for apps using FP (the Linux kernel itself does not use FP).

It's amazing that this even works for the ARM7 target, and even more so that it boots and runs on this specific ARM7 implentation. Evidently, the board follows the Cirrus/Crystal reference design very closely (except for the IDE bug).

Like the Websurfer, this thing has potential, but while the Websurfer has the potential to do divX, this has the potential to do MP3. About one order of magnitude difference in capability - but as a hack, it's much more involved and therefore probably more fun.

07-18-2000 23:45:32

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) nullbert
Profile
Codeman:

any update on the linux kernel for this little beast? I'm sure we're all dying to play with this further. haven't seen anything show up on www.linux-hackers.com/webpal...

If not, could you point out what kernel source you started with? where you downloaded, etc.?

Keep up the great work, it's appreciated!

08-30-2000 06:24:01

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) transiit
Profile
Greetings everyone:

I'm finally getting back to a point where I can return to my efforts on getting a webpal working with linux...hopefully there is still some interest out there.

Thus far, I've been reluctant to lift off the pin for fixing the ide problem until I have a better idea whether or not that will actually get me anywhere. However, I have seen a number of non-volatile memory chips with ide interfaces (a la Sandisk) that will clip right on an ide header, so the tentative plan is to eventually have something along those lines in place once I can get a kernel booting on it happily and whatnot.

So a few questions:
Has anyone determined whether or not the 72-pin flash simm can be replaced with a standard simm?
If a valid boot image were latched to the IDE device, without the simm in there, would it boot, or is there not enough low-level support to handle a boot without it?
As I don't have a vast amount of equipment for such tasks as manually reflashing whatever is on that simm, will there ever be a replacement available for purchase?
For that matter, is that flash simm a common part, or is it just a common chip dumped on a newcom or teknema custom pcb?
Oh, and of course, has anyone else made any further progress with these things?

-transiit

10-30-2000 04:43:23

New MessageJust bought 2 (modified 0 times) anon7864
Profile
I just bought two.

Hopefully I will have mine running soon!

I love my iopener and my Websurfer is a great streaming audio/mp3 client. The websurfer sounds better that the iopener, so i hook it up to my stereo in my house. I use the iopener on long trips in my car. I also have the iopener do live internet stuff while I am watching tv.

I will probably give the webpals away to my family so they have internet access.

The fun begins in a couple of days!

10-31-2000 10:52:17

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) jooshs
Profile
NetBSD supports this CPU.
11-02-2000 17:24:00

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) transiit
Profile
> NetBSD supports this CPU.

Yes, I'm familiar with the fact that NetBSD supports just about every available architecture out there, the bigger question right now is about how to actually get this thing booting with any operating system. We were given the hint that we'd need to lift or cut a certain pin to get the ide header working properly, but so far, I have very little information as to what additional steps are necessary (i.e., do we need to flash the firmware with a linux/netbsd/whatever kernel, or will hooking up a hard drive be enough, etc.)

-transiit

11-02-2000 22:43:01

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) jooshs
Profile
So it's the IDE port that is causing the problem? Don't these come with 4 megs of ram? Codeman's bootloader shows 8 megs found. Did he remove the flash simm and add another simm or did he remove the 4 meg and add an 8 meg.
11-03-2000 12:32:42

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) anon7864
Profile
Mine came with 8MB. See my post.
11-03-2000 13:24:35

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) hurricanej
Profile
>Has anyone determined whether or not the 72-pin flash simm can be replaced with a standard simm? <

I found a 72-pin simm laying around and gave it a shot. The monitor acquires a signal, but nothing displays (just a black screen). The 3 lights (red, yellow, green) go solid for a few seconds, and then flash off for about a second every 5 seconds.

I seem to recall it being necessary to plug some 72-pin simm's in in pairs. Is this RAM-stick specific or motherboard/chipset specific? How can I tell if I used a 72-pin SIMM that was supposed to be used in a pair? In all likelihood, I have other SIMM modules laying around and would be willing to give them a shot if someone would direct me on what to look for.

I don't know how much MB this SIMM contained and I don't have any machines laying around that use 72-pin SIMMs anymore.

>If a valid boot image were latched to the IDE device, without the simm in there, would it boot, or is there not enough low-level support to handle a boot without it?<

I will be trying this soon. I'm not knowledgeable about Linux so I was just gonna start with an MS-DOS disk and see what happens. (Yeah, I know, it's not an x86 cpu).

Did anyone notice the "button" sitting near the IDE header and (I assume) a floppy header? The label "SW2" lies between the "button" and the CPU. I played with this, too:

If you press and hold the button before applying power, the unit fails to boot - the monitor does not acquire any video signal, either. The green light turns on, but nothing else. If you let go of the button, the red button goes solid and the green stays on. The unit still doesn't seem to do much of anything (no boot, no video acquire.)

Pressing the button while the unit has already powered on in a normal configuration doesn't seem to impact anything.

I rather wondered if the "button" had to do with diagnostic/booting routines.... maybe if you press it when an IDE drive is plugged in, it'll act differently?

Codeman mentioned lifting or cutting pin 11 of U2. I am willing to try this if someone were to give me advice. I think I have located the U2 "chip" (forgive my ignorance on terminology) but am unsure which pin is #11. How do you start counting, at zero or one? And do you count pins in a clockwise manner, or like reading a book from left to right, every line?

Oh yeah - someone said they received a unit with 8MB instead of 4. How could they tell?

-hj

11-11-2000 10:20:19

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) melios
Profile
hurricanej:

I have tried booting the Webpal with another 72-pin SIMM and it booted fine. It was single-sided SIMM of unknown capacity that we had laying around.

The specs on the chip labeled U2 on the motherboard is a Octal 3-state tranceiver. The PDF document is at http://www.fairchildsemi.com/ds/MM/MM74HCT245.pdf and more information can be found at http://www.fairchildsemi.com/pf/MM/MM74HCT245.html.

We have tried lifting pin 11 by itself and also tried lifting pin 10, then plugging a HD and floppy into the headers on the motherboard and it didn't appear to do anything.

Would be nice if Codeman could sell a Webpal MOD kit... I've been thinking about mounting one of the smaller 2.5" notebook HD to the metal shield UNDERNEATH the m/b.

No EEPROM burner and not a lot of time to try to figure it out makes it real difficult. Right now we have 6 new in the box and 2 we've been playing with.

http://www.liquidations.com was running a special on Webpals, 10 for $34 a piece, FYI.

11-11-2000 13:42:45

New MessageRE:IT WORKS !!! WEBPAL BOOTING LINUX !!!!!! (modified 0 times) TivoTechie
Profile
This thread has been quite for a while, how about an update, maybe some pictures of exactly what I'd need to lift to get the IDE connector working, and a pointer to a distribution that will run on the unit?

Even better, a way to reflash the flash.

11-23-2000 17:35:12

New Messagewebpal.bigbrd.com (modified 0 times) newbee
Profile
try webpal.bigbrd.com..
http://webpal.bigbrd.com
Bill and Rob left great comments but apparently the people who are
trying to reproduce what they and codeman did before have great troubles
these days. Not sure why. It may have something to do with the aging, sigh, though.

Bill's off-line flashing machine would be pretty good because you can
program it even when webpal is completely dead.. It could be possible
to make it into ExpressPCB's mini board spec.. but not sure. (I am
into IC design industry.. so making something ever small is my job. :>)

I reckon somebody can make it work with just that little board
and PC's parallel bus without even the microcontroller that Bill used
to use.

11-02-2002 02:21:38

New Messageparallel port flash programmer.. can it be ? (modified 0 times) newbee
Profile
As a complete layman in digital design, I am not sure what I am doing but it seems to be feasible with very simple logic (almost identical what Bill did, I am afraid).. Just to see if this is OK to control with parallel port..

with 3 74374 (octal dffs) we can control a-1 to a18, reset_, we_, and ce_. These will be latched with c0(#1)-c2(#16) of parallel ports. c3(#17) is used for oe_ so that we can avoid any bus conflict on d0-7 for read/write operation.

byte_ will be tied low to ensure 8 bit transaction and ry/by_ pin will be connected
to busy(s7:#11).

In this way, just with parallel port programming, we can read/write from the flash module.

any idea if this can be verified before making the hardware ? it seems to be fairly simple but too complicated for me to go forward.

11-02-2002 05:20:17

New Messagebyte mode doesn't work. The pin is tied high on the module. :( (modified 0 times) newbee
Profile
I had a look at the module and the unused part of the module clearly
shows that the byte mode (#47) is tied to vdd. Therefore, with more than 99%
probability, the flash memory also has the pin tied to vdd.

This makes the parallel port scheme very complicated to make with just
discrete TTL parts. I still think utilizing parallel port for this is
the right thing to try though. An FPGA.. like XC5202 seems to be
suitable for this.

BTW, 72 pin socket for the flash write can be ordered at digikey
for $3.22/pieie when you buy in less than 10 quantity. It can be fit
to the mini board size of ExpressPCB if.. if you don't fit one of
the last pins on the board. The pin area is 3.75 inches long but
the mini board is 3.8 inches.. so you may need to drill one hole by
yourself for saving some 50 bucks.

11-03-2002 02:38:07

Reply to Thread | Printer |
All times are PSTPowered by UltraBoard v1.62



Copyright © 2000, Netmake Inc. All Rights Reserved.
See Terms and Conditions for more information.




i-opener opener laptop notebook computer help drivers dll free windows dos repair fix linux mac macintosh 2000 95 98 nt pc configure hardware software sound video netscape explorer network networking lan wan software cmos fat bios printer card mouse modem ide scsi cd rom controllers scanner tape hard drive cgi scripts source code mp3