**** BEGIN LOGGING AT Tue Jun 28 23:59:57 2005
**** BEGIN LOGGING AT Wed Jun 29 01:28:03 2005
Jun 29 02:18:57 <[g2]-away> ep1220, pong
Jun 29 02:19:25 <ep1220> hey; i ran the blink prog
Jun 29 02:19:30 <[g2]> cool
Jun 29 02:19:36 <[g2]> ep1220, IS THE MAN!
Jun 29 02:19:38 <ep1220> the frequency i see is 5MHz
Jun 29 02:19:54 <ep1220> so a lot of questions
Jun 29 02:20:37 <ep1220> e.g. are there waitStates on access to the GPIO area ?
Jun 29 02:20:50 <[g2]> do you get a nice shape of the waiveform ?
Jun 29 02:20:55 <[g2]> waveform 
Jun 29 02:20:57 <ep1220> yes
Jun 29 02:21:12 <[g2]> good
Jun 29 02:21:34 <[g2]> my understanding was that the GPIO are off the north AMBA bus
Jun 29 02:21:41 <[g2]> but I not certain
Jun 29 02:22:16 <[g2]> hmm.. just waking up ... I am 
Jun 29 02:22:36 <ep1220> :-) I am close to lunch
Jun 29 02:22:48 <ep1220> maybe we talk again in 1 hour ?
Jun 29 02:22:49 <[g2]> 5:22 AM here :)
Jun 29 02:23:18 <[g2]> If you are heading out to lunch that fine.
Jun 29 02:23:30 <[g2]> I'm here now, I'll be around later
Jun 29 02:23:41 <[g2]> THX for all the help
Jun 29 02:24:00 <ep1220> np;
Jun 29 02:24:10 <[g2]> it's 2:23 for jacques and bewoolie :)
Jun 29 02:24:22 <[g2]> I doubt they'll be around for a while
Jun 29 02:24:52 <dyoung> [g2] is a robot.
Jun 29 02:25:01 <[g2]> :)
Jun 29 02:25:09 <[g2]> hey late sleeper :)
Jun 29 02:25:14 <ep1220> good, i'll be back in 1 hour
Jun 29 02:25:24 <dyoung> I was looking at Lennerts slug.
Jun 29 02:25:26 <[g2]> have a great lunch
Jun 29 02:25:51 <[g2]> how's it look ?
Jun 29 02:26:29 <dyoung> it looks dead.
Jun 29 02:26:50 <dyoung> If I were him, and I had this slug, I would have thrown it out.
Jun 29 02:32:14 <[g2]> dyoung, have you been following the slug "peformance" issue ?
Jun 29 02:33:09 <dyoung> more or less.
Jun 29 02:33:22 <dyoung> I was gonna go poking around for a clock halfer.
Jun 29 02:40:05 <[g2]> the 5Mhz number is interesting
Jun 29 02:43:18 <[g2]> dyoung, that's for looking at Lennert's flaky slug
Jun 29 02:43:27 <[g2]> doh... THANKS for
Jun 29 02:43:30 <dyoung> tnats?
Jun 29 02:43:39 <dyoung> wow, I'm being punished with a 5Mhz slug?
Jun 29 02:43:42 <dyoung> :-(
Jun 29 02:44:15 <[g2]> well it's 15-20 Mhz
Jun 29 02:44:30 <[g2]> at least 10
Jun 29 02:44:35 <dyoung> phew.
Jun 29 02:44:36 <[g2]> xor and branch
Jun 29 02:44:37 <dyoung> :-)
Jun 29 02:44:48 <dyoung> I'm still digesting.
Jun 29 02:45:04 <dyoung> I agree that these slugs are running well, sluggish.
Jun 29 02:45:21 <dyoung> but I cant see how you make it run at half speed without screwing up the pci clock
Jun 29 02:45:33 <[g2]> maybe after we started calling them "slug" the went on strike part time :)
Jun 29 02:46:04 <[g2]> I'm with beewoolie.... Undocumented features
Jun 29 02:46:20 <[g2]> Intel says .... "just do this...."
Jun 29 02:47:10 <dyoung> hey, thats what I said.
Jun 29 02:47:17 <dyoung> Undocumented fuse or soething
Jun 29 02:47:32 <dyoung> bastage
Jun 29 02:47:47 <dyoung> so linksys bought all the defective ixp420s
Jun 29 02:48:01 <[g2]> I don't think so
Jun 29 02:48:38 <dyoung> damn SX's
Jun 29 02:49:02 <[g2]> maybe, but I'd guess they are clocked down for either heat (plastic case) or more likely product channel issues
Jun 29 02:49:37 <[g2]> but I don't know 
Jun 29 02:49:52 <[g2]> I've been looking but haven't seen the RT-042
Jun 29 03:00:38 <NAiL> [g2]: 08:21 < jbowler-away> Am I missing something?  That code seems to be writing to the physical address of the GPIO within the loop, but writing the GPIO is
Jun 29 03:00:41 <NAiL>                       very slow - it won't show the core clock speed at all.
Jun 29 03:01:22 * NAiL got BUBBLEWRAP! :D
Jun 29 03:02:06 <[g2]> heh.... that was 2:21 here :)
Jun 29 03:06:04 <NAiL> slugtime never sleeps ;)
Jun 29 03:06:23 <dyoung> I have a serious headache.
Jun 29 03:06:36 <NAiL> I've got painkillers?
Jun 29 03:06:58 <NAiL> (and I've got bubblewrap)
Jun 29 03:07:02 <dyoung> I try to avoid drugs.
Jun 29 03:07:25 <NAiL> That is generally a good thing. 
Jun 29 03:07:25 <dyoung> NSAID dependancy sucks.
Jun 29 03:07:32 <dyoung> (ibuprofen)
Jun 29 03:07:37 <dyoung> Aka Vitamin I
Jun 29 03:07:55 <dyoung> so does your bubblewrap have a PL2303 doodad in it?
Jun 29 03:08:00 <NAiL> hmm
Jun 29 03:08:10 * NAiL searches through the bubblewrap
Jun 29 03:08:14 <NAiL> I'll be damned
Jun 29 03:08:21 <NAiL> There's something in here!
Jun 29 03:08:30 <NAiL> WHO PUT STUFF IN MY BUBBLEWRAP?
Jun 29 03:08:52 <dyoung> I think I sent Lennert my serial cable.
Jun 29 03:08:56 <dyoung> USB/Serial/
Jun 29 03:09:02 <dyoung> thinking I was gonna make another one
Jun 29 03:10:01 <NAiL> Sounds like you havent
Jun 29 03:10:21 <NAiL> Ugh.. slow soldering iron
Jun 29 03:35:04 <NAiL> hmm... Why doesn't my slug boot when my serial cable is attached?
Jun 29 03:35:24 <dyoung> what kind of serial cable?
Jun 29 03:35:39 <NAiL> usb
Jun 29 03:35:49 <dyoung> you might need the pullups
Jun 29 03:36:44 * NAiL stares blankly at dyoung
Jun 29 03:36:56 <NAiL> It's the cable ka6sox sent
Jun 29 03:37:02 <dyoung> do ou have some spare resistors lying around?
Jun 29 03:37:10 <jacques> well that was fun
Jun 29 03:37:25 <NAiL> dyoung: Very unlikely. 
Jun 29 03:40:15 <dyoung> http://groups.yahoo.com/group/nslu2-linux/message/2452
Jun 29 03:40:29 <dyoung> I dont have the issue, but rwhitby does.
Jun 29 03:40:51 <dyoung> basically you need to add a 100k or so resistor between pin 1 and 2 of the slug serial port.
Jun 29 03:41:54 <NAiL> hmm... Fair enough. Then I'll just have to find a resistor
Jun 29 03:42:12 <dyoung> it escapes me why it works for me and not others.
Jun 29 03:42:26 <dyoung> I guess its possible my cable has one already.
Jun 29 03:42:28 <jacques> I didn't need any resistors
Jun 29 03:43:57 <dyoung> jacques, nod
Jun 29 03:44:19 <dyoung> NAiL, http://nslu2-linux.thestuffguy.com/gallery/Serial_adapters
Jun 29 03:44:27 <NAiL> Hmm.. I've got an alternative cable, but I don't know which leads are what...
Jun 29 03:44:31 <dyoung> I use the RS 170-0787
Jun 29 03:44:34 <NAiL> dyoung: I've seen it. 
Jun 29 03:44:55 <dyoung> nail, what sort of wires do you have coming out of your mystery calbe?
Jun 29 03:44:58 <dyoung> cable?
Jun 29 03:45:20 <NAiL> what sort of? There's one blue, one white and one black ;)
Jun 29 03:45:28 <dyoung> if you look at my serial adapters gallery, you'll see some black things that are marked 152, those are resistors.
Jun 29 03:45:45 <dyoung> so its totally possible that this cable is pulled up already.
Jun 29 03:45:58 <dyoung> Do you have a Voltmeter handy?
Jun 29 03:46:04 <NAiL> infact I do.
Jun 29 03:46:16 <NAiL> or.. "handy" is an overstatement.
Jun 29 03:46:19 <NAiL> I'll dig it up
Jun 29 03:46:25 <dyoung> is it a PL2303 device?
Jun 29 03:47:04 <NAiL> yes
Jun 29 03:47:19 <dyoung> Okay, this is the usual way to figure it out.
Jun 29 03:47:35 <dyoung> figure out which of your 3 wires is ground.
Jun 29 03:47:41 <dyoung> we'll call that wire A.
Jun 29 03:48:02 <dyoung> then measure the voltage between A and B then A andn C.
Jun 29 03:48:11 <dyoung> one will be 0V one will be 3.3V
Jun 29 03:48:40 <dyoung> the one thats 3.3V is the TX line that will go to the Slug RX Line (pin2)
Jun 29 03:49:09 <dyoung> the other one is the RX which will go to the slug tx line (pin3)
Jun 29 03:50:28 <NAiL> ok, between white and black I get 3,3. But which do I figure out which one is 3.3? I don't know which end is what of this voltmeter ;)
Jun 29 03:50:45 <dyoung> you need to figure out which one is ground.
Jun 29 03:51:01 <dyoung> hmm
Jun 29 03:51:12 <dyoung> you can usually look at the board and figure it out
Jun 29 03:51:34 <NAiL> hehe, yeah. 
Jun 29 03:51:46 <dyoung> sorry, I cant think of a easy shortcut for that.
Jun 29 03:51:56 <NAiL> no, I'll try to figure it out. 
Jun 29 03:52:11 <NAiL> There's three annoying leds on the board that makes it hard though
Jun 29 03:52:34 <NAiL> whoever thought of adding a ultra-brite (tm) blue led on this board should have been shot. Thrice.
Jun 29 03:52:58 <dyoung> yo could put your voltmeter into resistance mode and measure against the common side of your LED's to the black and white.
Jun 29 03:54:04 <dyoung> Or, the negative probe will be the ground side when you measure volts accross the white/black.  Maybe.
Jun 29 03:56:57 <ep1220> [g2]: I am back
Jun 29 03:57:08 <[g2]> ep1220, wb
Jun 29 03:57:17 <NAiL> hmm..
Jun 29 03:57:25 <[g2]> I hope your lunch was enjoyable
Jun 29 03:57:30 <[g2]> and delicious
Jun 29 03:57:41 <NAiL> white/black, 3.3v. Blue/Black, 2.8v?
Jun 29 03:58:02 <ep1220> [g2]: it was OK, nothing special
Jun 29 03:58:26 <dyoung> white blue?
Jun 29 03:58:38 <ep1220> [g2]: I tried to add code so the GPIO14/15 output the 33Mhz CLK
Jun 29 03:58:43 <[g2]> well if we have special everyday then it becomes ordinary right ? :)
Jun 29 03:58:44 <ep1220> but did not work
Jun 29 03:59:07 <NAiL> white/blue 0.0-something
Jun 29 04:00:07 <NAiL> Fun. The blue led goes crazy when I touch the black or white lead :)
Jun 29 04:00:33 <dyoung> those were the only wires coming out of the thing?
Jun 29 04:01:41 <NAiL> yeah
Jun 29 04:02:07 <dyoung> white is probably ground then.
Jun 29 04:02:11 <NAiL> There's pads for two more cables, but they're not connected to anything (missing resistors or something)
Jun 29 04:03:00 <dyoung> black is probably TX
Jun 29 04:03:14 <dyoung> Blue is probably RX
Jun 29 04:03:20 <dyoung> but I take no responsibility for blown up calbes, etc
Jun 29 04:03:22 <dyoung> :-)
Jun 29 04:04:29 <NAiL> Isn't there a pinout of the prolific anywhere? I see the blue goes directly to a pin there
Jun 29 04:04:44 <dyoung> just moment
Jun 29 04:04:45 <NAiL> (through 17 gazillion vias)
Jun 29 04:04:54 <dyoung> what pin number?
Jun 29 04:06:33 <NAiL> Kinda hard to say, the top of the chip doesn't have anything on it... 
Jun 29 04:06:47 <dyoung> no dimple or.. ?
Jun 29 04:07:47 <dyoung> is it the 5th pin from something?
Jun 29 04:08:10 <dyoung> from one of the ends?
Jun 29 04:09:24 <[g2]> ep1220, did you assemble that code bewoolie sent or just use the hex ?
Jun 29 04:09:33 <ep1220> [g2]: OK, when I just enable GPIO14/15 and donot try to setup the clock I get the 33Mhz
Jun 29 04:09:51 <ep1220> g2: the code bewoolie mailed me did not work
Jun 29 04:10:28 <ep1220> So i assembled my own.
Jun 29 04:11:16 <ep1220> based on our 1st prog
Jun 29 04:11:58 <ep1220> For experiments it is best to toggle the green status LED.
Jun 29 04:12:07 <ep1220> This signal can be seen on R53
Jun 29 04:14:49 <ep1220> [g2]: Just mailed it to You
Jun 29 04:14:54 <[g2]> ok
Jun 29 04:15:40 <ep1220> I just do not know how to interpret the result :-(
Jun 29 04:16:03 <[g2]> so the code you just sent toggles the LED line at 5Mhz
Jun 29 04:16:39 <[g2]> ah.. and you're running it from RedBoot right ?
Jun 29 04:16:46 <ep1220> Yes. measured between 2 rising edges of GPIO1 
Jun 29 04:16:56 <ep1220> No, directly from RESET
Jun 29 04:17:13 <ep1220> No bootloader running before
Jun 29 04:17:30 <dyoung> thats why ep1220 is my idol
Jun 29 04:17:33 <[g2]> I'd try cache on in Redboot and run it from redboot
Jun 29 04:17:43 <[g2]> I'll bet the cache is off :)
Jun 29 04:17:56 <ep1220> I load it into the cache
Jun 29 04:18:19 <dyoung> [g2] make joke. :-)
Jun 29 04:18:45 <dyoung> NAiL, I'm running out of battery.
Jun 29 04:19:07 <dyoung> I have the ds open now if you need any info.
Jun 29 04:19:16 <[g2]> ep1220, are you running from the main icache or mini icache
Jun 29 04:19:32 <ep1220> mini, maybe the minicache is not 0 WS (??)
Jun 29 04:20:03 <[g2]> can you try just load and running the program from Redboot ?
Jun 29 04:20:28 <NAiL> Note to self, do *not* attempt to probe a USB-device attached to Window
Jun 29 04:20:30 <NAiL> s
Jun 29 04:20:46 <ep1220> g2: So far I only have an assembly listing.
Jun 29 04:20:47 <[g2]> maybe you don't have the tftp server setup
Jun 29 04:21:00 <NAiL> dyoung: Yes, it has a dimple
Jun 29 04:21:06 <ep1220> I have no tftp setup
Jun 29 04:21:14 <ep1220> but I can serial into redboot
Jun 29 04:21:22 <dyoung> nail, the dimple is pin1
Jun 29 04:21:32 <dyoung> its on the left side
Jun 29 04:21:51 <dyoung> you count down from tht pin, to pin 14, then accross to the lower right which is pin 15.
Jun 29 04:21:59 <dyoung> the one accross from the dimple is pin 28.
Jun 29 04:21:59 <[g2]> dyoung, is ymodem supported in the nslu2 Redboot ?
Jun 29 04:22:12 <dyoung> Funny you should ask that.
Jun 29 04:22:26 <dyoung> the -m supports xmodem ymodem tftp http
Jun 29 04:22:28 <dyoung> only.
Jun 29 04:22:32 <dyoung> no zmodem.
Jun 29 04:23:15 <[g2]> Ok.. that's cool... ep1220 can x or y modem a tiny program to redboot
Jun 29 04:23:46 <NAiL> the blue one is connected to the chip through a via, so it's kinda hard to seee. But... I buzzed from black to the chip. Everything buzzes. That'd be gnd, right?
Jun 29 04:24:25 <[g2]> ep1220, you've got the object code genterated from that program right ?
Jun 29 04:24:32 <dyoung> the chip has a few grounds on it,  pin 7 and 21 are ground.
Jun 29 04:24:58 <dyoung> pin 18 is the 3V3 gnd
Jun 29 04:25:14 <NAiL> how many pins on each side?
Jun 29 04:25:40 <NAiL> I'm farsighted, which doesn't help atm
Jun 29 04:26:08 <dyoung> 14 pins on each side.
Jun 29 04:26:48 <[g2]> ep1220, if you just hexdump/hexedit the object code you can see where the first instruction offset starts
Jun 29 04:26:50 <ep1220> g2: No. Just a listing
Jun 29 04:27:15 <[g2]> what command line are you using with gas ?
Jun 29 04:27:21 <ep1220> the code is fully relocatable.
Jun 29 04:27:25 <ep1220> I did not use gas
Jun 29 04:27:48 <[g2]> nodd
Jun 29 04:28:03 <dyoung> ... I have gas.....
Jun 29 04:28:09 <[g2]> :)
Jun 29 04:28:27 <[g2]> the 2 things I'd try are:
Jun 29 04:28:37 <[g2]> 1) run from redboot with the cache one
Jun 29 04:29:05 <dyoung> nail, shutdown sequence initiated.
Jun 29 04:29:14 <[g2]> 2) unroll the loop and put like 10 toggles before the branch
Jun 29 04:29:23 <ep1220> OK, I have a .o now
Jun 29 04:29:31 <NAiL> dyoung: Going to bed?
Jun 29 04:29:36 <dyoung> Critical hit: heat sink.
Jun 29 04:29:51 <dyoung> Internal Ammunition explosion detected.
Jun 29 04:29:54 <dyoung> http://tech.prolific.com.tw/visitor/fcabdl.asp?fid=33962507
Jun 29 04:30:00 <[g2]> 3) insert a NOPs or ORR rX, rx,rx instructions as place holders
Jun 29 04:31:03 <dyoung> nail, soon.  I have around 5minutes of battery left.
Jun 29 04:31:21 <[g2]> one day ends... another begins .....
Jun 29 04:31:22 <NAiL> dyoung: Ok, I'll try to figure out the rest. Thanks :)
Jun 29 04:31:23 <ep1220> g2: the assembler can output aout. Is this OK for redboot
Jun 29 04:31:52 <[g2]> ep1220, it's just binary, you'll need to see the offset of the code though
Jun 29 04:34:43 <dyoung> NAiL, If you remember that TX is asserted idle mode you shouldbe able to figure it out
Jun 29 04:38:06 <ep1220> g2: Offset of the code == startaddress ?
Jun 29 04:39:24 <ep1220> g2: Just now I will add a nop (mov r2,r2) into the loop and check how it affects timing
Jun 29 04:40:03 <[g2]> it's the start address you want
Jun 29 04:40:41 <ep1220> I can run the code anywere, there are no absolute calls/jumps in there
Jun 29 04:40:52 <[g2]> nod
Jun 29 04:41:37 <[g2]> I'd do something like
Jun 29 04:42:08 <[g2]> load -r -v -b 0x1000000 -m YMODEM
Jun 29 04:42:19 <[g2]> send the code
Jun 29 04:42:43 <[g2]> exec 0x1000034 or whatever the first instruction address is
Jun 29 04:42:54 <[g2]> or go 0x1000034
Jun 29 04:44:38 <ep1220> g2: How do I get into redboot. I see a '+' and a little later the startup messages
Jun 29 04:44:52 <[g2]> just hit the ctrl-c
Jun 29 04:45:03 <[g2]> after the '+'
Jun 29 04:56:14 <[g2]> ep1220, luck with Redboot ?
Jun 29 05:00:00 <ep1220> g2: No, I see what it sends, but it ignores what I type
Jun 29 05:01:33 <[g2]> Have you used serail for input before ?
Jun 29 05:02:16 <ep1220> Yes. I also have already been in redboot (just didnot remember the key)
Jun 29 05:03:12 <[g2]> you can also get into Redboot via Telnet
Jun 29 05:03:40 <[g2]> it's a tighter window, but it sounds like the receive side of your serial isn't working
Jun 29 05:03:55 <[g2]> or your terminal emulator isn't properly configured
Jun 29 05:04:12 <[g2]> like it's setup for flow-control 
Jun 29 05:12:43 <ep1220> I get me a cup of tea and then try again :-)
Jun 29 05:18:03 <NAiL> heh, I think I figured out which cable is rx... If I touch the blue wire, I get lots of garbage :-P
Jun 29 06:00:31 <VoodooZ_Work> Good work guys. I'm keeping my fingers crossed we can figure out a way to get a "free" speed boost.
Jun 29 06:01:48 <NAiL> I wouldn't mind a free speed boost either. But I think it's unlikely ;)
Jun 29 06:03:21 <VoodooZ_Work> yeah, I'm just a dreamer :)
Jun 29 06:04:53 <NAiL> Nah, A Visionary. Sounds better :D
Jun 29 06:05:24 <VoodooZ_Work> ok. if you insist. :)
Jun 29 07:20:06 <ep1220> g2: ping
Jun 29 07:24:01 <VoodooZ_Work> ep1220: Found anything more?
Jun 29 07:24:14 <ep1220> I had a broken wire
Jun 29 07:24:20 <ep1220> now I am in redboot
Jun 29 07:24:24 <ep1220> loaded the code
Jun 29 07:24:37 <ep1220> when i exec it does not wha it should :-(
Jun 29 07:24:53 <ep1220> i use -b 10000000
Jun 29 07:24:55 <VoodooZ_Work> odd.
Jun 29 07:25:02 <[g2]> ep1220, poing
Jun 29 07:25:12 <VoodooZ_Work> [g2] to the rescue!
Jun 29 07:25:12 <ep1220> but it says it loads to 989680
Jun 29 07:25:28 <ep1220> when i dump: x -b 10000000
Jun 29 07:25:41 <ep1220> the output starts at 989680
Jun 29 07:25:45 <ep1220> is this normal ?
Jun 29 07:26:09 <[g2]> I think you have too many zeros :)
Jun 29 07:26:31 <[g2]> reset
Jun 29 07:26:44 <ep1220> done; seesm I have 1 too much
Jun 29 07:27:00 <VoodooZ_Work> been there!
Jun 29 07:27:08 <[g2]> yes
Jun 29 07:27:29 <[g2]> it's 10 5-zeros
Jun 29 07:27:36 <[g2]> which is 16 MB
Jun 29 07:27:52 <[g2]> 5-zeros being 2**20
Jun 29 07:31:34 <[g2]> ep1220, is that helping ?
Jun 29 07:32:18 <ep1220> Well, now the addresses match, but still no blinking LED
Jun 29 07:32:39 <ep1220> When I dump, I see the LSB first, is this OK ?
Jun 29 07:33:23 <[g2]> I think you should be seeing MSB
Jun 29 07:33:42 <[g2]> are you go -b 0x1000000 ?
Jun 29 07:33:54 <[g2]> it should probably be -b 0x100034
Jun 29 07:34:04 <[g2]> objdump is your friend
Jun 29 07:34:17 <ep1220> no I do exec 0x100000
Jun 29 07:34:30 <[g2]> that's why it's not working
Jun 29 07:34:56 <ep1220> the first codeword is at 0x100000
Jun 29 07:35:15 <[g2]> then it should work
Jun 29 07:35:37 <[g2]> you should see the same instructions in memory as the hex
Jun 29 07:37:50 <ep1220> Is identical
Jun 29 07:37:55 <ep1220> now I use go
Jun 29 07:38:30 <ep1220> this gives: T050f:00100008;0d:01ff2fec;#e3
Jun 29 07:39:37 <[g2]> I think that's the debugger
Jun 29 07:40:03 <[g2]> your not executing stuff right or you compiled it in LE mode
Jun 29 07:46:45 <ep1220> yes it is LE
Jun 29 07:47:12 <ep1220> I have to make a phone-call now and will try again afterwards
Jun 29 07:48:12 <[g2]> nod.
Jun 29 07:51:31 <jacques> did you guys see what jbowler-zzz said? 
Jun 29 07:51:49 <jacques> <jbowler-away> Am I missing something?  That code seems to be writing to the physical address of the GPIO within the loop, but writing the GPIO is very slow - it won't show the core clock speed at all.
Jun 29 07:51:49 <jacques> <jbowler-away> mov r0, #0
Jun 29 07:51:49 <jacques> <jbowler-away> 0: subs r0, r0, #1
Jun 29 07:51:49 <jacques> <jbowler-away> bne 0b
Jun 29 07:51:49 <jacques> <jbowler-away> Should take about 32s with a 266MHz core clock (2^33 instructions)
Jun 29 07:53:07 <[g2]> jacques, I saw it and I didn't see any reference in the Developers Manual about the timing for writing to that register
Jun 29 07:55:23 <jacques> `k just wanted to make sure you saw it
Jun 29 07:55:36 <ep1220> g2: big endian works better :-)
Jun 29 07:56:31 <[g2]> ep1220, :)
Jun 29 07:56:32 <ep1220> is still slower, here I see 4.12MHz
Jun 29 07:57:58 <[g2]> is cache on ?
Jun 29 07:58:34 <ep1220> How can I check ?
Jun 29 07:59:08 <ep1220> made a mistake: it is 6.58Mhz
Jun 29 08:00:03 <[g2]> cache on
Jun 29 08:01:40 <[g2]> 6.58 is better than 4.12 :)
Jun 29 08:03:50 <ep1220> cache on gives "illegal command jhddcache"
Jun 29 08:06:42 <ep1220> Using "cache ON": I get 7.25MHz
Jun 29 08:13:06 <ep1220> I now added a move r3,r2 to the loop
Jun 29 08:13:15 <ep1220> gives 5,45MHz
Jun 29 08:13:58 <ep1220> if i calc the delta and asume this is one cycle i get a coreclock of approx- 44Mhz
Jun 29 08:14:00 <ep1220> As jacques said above: I start to believe that writing to GPIO is not a valid way to find out the internal clockrate
Jun 29 08:15:12 <ep1220> Maybe a larger count-down inside the loop is better
Jun 29 08:15:57 <[g2]> we have a loop inside
Jun 29 08:16:18 <[g2]> a million 0x100000
Jun 29 08:17:11 <[g2]> that'll give an approximate freq in MIPS (if we use 0x80000
Jun 29 08:26:59 <ep1220> Added a loop
Jun 29 08:27:09 <ep1220> count down 4096 to 0
Jun 29 08:27:28 <ep1220> I assume the sub and branch take 1 cycle each
Jun 29 08:27:58 <ep1220> Now the GPIO is low/high for 62microseconds (each phase)
Jun 29 08:28:20 <ep1220> This gives a clock rate of 132,12MHz
Jun 29 08:35:51 <VoodooZ_Work> THe question is: how can we modify this? :)
Jun 29 08:38:13 <ep1220> Is the assumptin: branch takes one cycle correct ?
Jun 29 08:38:24 <ep1220> It it were 3 cycles, we were at 266MHz
Jun 29 08:39:55 <VoodooZ_Work> [g2] would be able to answer that.
Jun 29 08:40:14 <VoodooZ_Work> I'm more of a 8 bit micro guy.
Jun 29 08:41:58 <ep1220> There is a register named EXP_CNFG0
Jun 29 08:42:31 <ep1220> I ahve an app-note which says that bits 23:21 in there can be used to under-clock a part.
Jun 29 08:43:04 <VoodooZ_Work> interesting!
Jun 29 08:43:16 <ep1220> It also allows to select the PCI clock
Jun 29 08:43:20 <ep1220> 33 or 66
Jun 29 08:43:21 <VoodooZ_Work> We'd have to figure out what it's being set at in the slug.
Jun 29 08:43:26 <VoodooZ_Work> nice.
Jun 29 08:44:12 <ep1220> It is at 0xC4000020
Jun 29 08:45:19 <VoodooZ_Work> I'd have to browse the different data sheet as I'm a total ARM noob
Jun 29 08:46:50 <ep1220> tried to dump it in Redboot -> gives a strange line, maybe an exception (?)
Jun 29 08:48:07 <ep1220> I'll be back later
Jun 29 08:48:25 <VoodooZ_Work> ok, we'll have to share your findings with the others.
Jun 29 08:51:07 <[g2]> APEX can dump it :)
Jun 29 08:51:51 <VoodooZ_Work> hehe
Jun 29 08:51:57 * [g2] wonders which app note :)
Jun 29 08:52:19 <VoodooZ_Work> A link on the performance wiki page would be useful
Jun 29 08:52:41 <VoodooZ_Work> Did you see ep1220's assumption above? Can you confirm?
Jun 29 08:53:02 <VoodooZ_Work> (branches take 1 cycle or more?)
Jun 29 08:53:27 <[g2]> I'd have to check
**** BEGIN LOGGING AT Wed Jun 29 11:10:53 2005
Jun 29 11:40:56 <ep1220> [g2]-away: the appNote is IXP42X Boot-up Options; Intel doc number from Filename: 25406701; printed inside: 254067-002
Jun 29 11:52:41 <VoodooZ_Lunch> interesting read. 
Jun 29 11:53:42 <VoodooZ_Lunch> Makes me wonder if configuring the 266 part as if it was a 533 part but configured for 266 operation would give us 133Mhz?
Jun 29 11:54:59 <VoodooZ_Lunch> [g2]-away: if you get a chance can you dump the content of that config reg in Apex?
Jun 29 12:00:36 <jacques>  IXP42X Boot-up Options - ftp://download.intel.com/design/network/applnots/25406701.pdf
Jun 29 12:00:45 <jacques> interesting indeed
Jun 29 12:02:25 <VoodooZ_Lunch> yeah, I doubt it's that simple but I'm curious to see what our beloved slug is configured.
Jun 29 12:02:53 <jacques> I just want to get to the bottom of this one way or another
Jun 29 12:02:58 <VoodooZ_Lunch> yeah, me too.
Jun 29 12:03:11 <jacques> and if we end up being able to increase our performance in the process, I will be very heppy
Jun 29 12:03:14 <jacques> happy
Jun 29 12:03:17 <VoodooZ_Lunch> the DS says these config bits can be configured with weak pullups.
Jun 29 12:03:23 <VoodooZ_Lunch> damn right!
Jun 29 12:03:34 <VoodooZ_Lunch> that would help my computer vision thread quite a bit!
Jun 29 12:03:42 <jacques> I bet it would
Jun 29 12:05:26 <jacques> hmm, they still don't show a way to actually underclock a 266MHz part
Jun 29 12:07:21 <VoodooZ_Lunch> yeah, but sometimes they don't show all "possible" combinations in those tables. I'd love to try a few other values.
Jun 29 12:07:50 <jacques> VoodooZ_Lunch, agreed
Jun 29 12:08:07 <jacques> first we ned to know what it's set to
Jun 29 12:08:12 <VoodooZ_Work> oops. forgot to change nick.
Jun 29 12:08:17 <VoodooZ_Work> yep.
Jun 29 12:08:28 <jacques> and if it's not 1,0,0 I will be very curious
Jun 29 12:08:33 <VoodooZ_Work> [g2] said he could read that value using Apex.
Jun 29 12:08:37 <VoodooZ_Work> indeed!
Jun 29 12:08:40 <jacques> if it's 0,1,1 I will think we've found the answer
Jun 29 12:08:41 <VoodooZ_Work> I'm all excited!
Jun 29 12:09:02 <VoodooZ_Work> Nothing like a good overclock to make me smile!
Jun 29 12:10:15 <VoodooZ_Work> unless it's possible to read it using redboot? ep1220 said he had errors.
Jun 29 12:12:06 <VoodooZ_Work> a simple 10k pull-down should do it then.
Jun 29 12:12:20 <VoodooZ_Work> Unless we don't have easy access to those pins? 
Jun 29 12:12:31 <NAiL> That'd be a drag
Jun 29 12:12:37 <VoodooZ_Work> damn right! 
Jun 29 12:18:42 <NAiL> But I'd imagine it would generate a lot of heat though
Jun 29 12:18:42 <NAiL> hmm... ping me someone?
Jun 29 12:19:59 <VoodooZ_Work> yeah, but I would still like to do it as my robot doesn't run for that long anyways.
Jun 29 12:20:26 <VoodooZ_Work> I tried doing a dump of that memory area from redboot but it keeps giving me weird stuff.
Jun 29 12:20:38 <VoodooZ_Work> someone know how to stop the dump command?
Jun 29 12:21:18 <NAiL> BE/LE?
Jun 29 12:23:31 <VoodooZ_Work> well, the default slug redboot mode.
Jun 29 13:27:32 <[g2]> dump 0xc4000020
Jun 29 13:27:32 <[g2]> c4000020: 00 41 ff ee 00 00 00 00  00 00 09 fc 00 00 00 00  .A??.... ...?....
Jun 29 13:27:32 <[g2]> c4000030: 00 00 29 46 08 04 40 c4  00 00 00 00 00 00 00 00  ..)F..@? ........
Jun 29 13:27:32 <[g2]> c4000040: 10 00 00 00 96 54 22 21  ff ff ff ec ff ff ff ff  .....T"! ????????
Jun 29 13:35:25 <[g2]> jacques, I think that 4 makes it 0, 1 0
Jun 29 13:41:12 <ep1220> unfortunately the bits setting the clockfrequency can not be written to
Jun 29 13:43:19 <[g2]> it's the pull ups right ?
Jun 29 13:43:27 <ep1220> nod
Jun 29 13:43:54 <ep1220> I already tried 2 values, no change in execution speed.
Jun 29 13:44:54 <[g2]> tried writing two values ?
Jun 29 13:45:16 <ep1220> yes.  1,0,0 and 0,1,1
Jun 29 13:45:26 <[g2]> does the write work ?
Jun 29 13:45:52 <ep1220> it does not crash.
Jun 29 13:46:06 <ep1220> But i did a "blind" write.
Jun 29 13:46:08 <[g2]> I meant did you read back the value ?
Jun 29 13:46:14 <ep1220> no
Jun 29 13:46:47 <ep1220> just added code to set these bits before the toggling
Jun 29 13:47:24 <ep1220> does the dump command also work in redboot ?
Jun 29 13:47:40 <ep1220> I tried x -b 0xc4000020 -l 4
Jun 29 13:47:46 <ep1220> gave strange repsonse
Jun 29 13:48:58 <[g2]> beewoolie-afk, in da house !
Jun 29 13:49:36 <ep1220> g2: did You verify how this reg is set on your faster machine (avila ?)
Jun 29 13:50:10 <[g2]> ep1220, no, not yet
Jun 29 13:50:21 <[g2]> I forget which boot loader I've got there (probably APEX)
Jun 29 13:51:12 <[g2]> I was going to try changing the value on the non-fat slug with APEX before the boot
Jun 29 13:53:39 <[g2]> beewoolie-afk, is there a write other than fill ?
Jun 29 13:56:23 <beewoolie-afk> [g2]: yo man.  did you follow the thread with ep1220?
Jun 29 13:56:47 <beewoolie-afk> [g2]: What are you looking to do?
Jun 29 13:56:55 <beewoolie-afk> [g2]: re fill tha is
Jun 29 13:56:57 <[g2]> beewoolie-afk, I think so, I didn't do the math on the instructions
Jun 29 13:57:08 <[g2]> can you fill more than a byte ?
Jun 29 13:57:45 <[g2]> fill 0x0080ffee 0xC4000020+4
Jun 29 13:58:02 <[g2]> fill 0x0081ffee 0xC4000020+4
Jun 29 14:00:06 <beewoolie-afk> [g2]: That should work.
Jun 29 14:00:30 <beewoolie-afk> Oh, you are wondering if the source can be more than one byte.  I don't recall.  I think that I never made that work.
Jun 29 14:00:56 <[g2]> nod. it didn't :)
Jun 29 14:01:51 <[g2]> I'd think jacques would have mmap'ed that reg by now
Jun 29 14:02:34 <dyoung-zzzz> We've talked about this appnote and register before.
Jun 29 14:03:51 <dyoung-zzzz> what we really need to do is see if CFG1 and/or CFG0 are pulled up.
Jun 29 14:03:54 <dyoung-zzzz> externally.
Jun 29 14:04:22 <[g2]> I've been wanting to dump those initial values
Jun 29 14:04:34 <[g2]> actuall all the values from the pull ups
Jun 29 14:04:44 <[g2]> it's in CPSR15 iirc
Jun 29 14:04:57 <[g2]> that's one of the first values read
Jun 29 14:07:13 <[g2]> hey
Jun 29 14:07:25 <[g2]> dump 0xc4000020
Jun 29 14:07:25 <[g2]> c4000020: 00 81 ff ee 
Jun 29 14:07:27 <[g2]> BINGO
Jun 29 14:07:45 <[g2]> ep1220, did you see that ?
Jun 29 14:07:52 <ep1220> yes
Jun 29 14:08:39 <ep1220> now you can run your benchmark
Jun 29 14:08:56 <[g2]> ep1220, that's the *avila* board :)
Jun 29 14:09:21 <ep1220> :-(  I thought You could modify them on the slug
Jun 29 14:09:56 <ep1220> looks like we are back on searching traces
Jun 29 14:10:16 <ep1220> the Flash includes only A21
Jun 29 14:10:49 <[g2]> it's on the expansion address lines
Jun 29 14:13:49 <ep1220> right, just found A23 in the wiki, So r64 pulls-down A23
Jun 29 14:16:33 <[g2]> ep1220, which wiki page ?
Jun 29 14:16:43 <ep1220> http://www.nslu2-linux.org/wiki/Info/FlashExpansion
Jun 29 14:16:49 <[g2]> thx
Jun 29 14:18:05 <[g2]> ep1220, I thought it just read the lower 16 for the ixp420
Jun 29 14:19:41 <ep1220> as I understand this appnote: at reset it reads A0:30 into the reg at c4000020
Jun 29 14:21:26 <ep1220> a21:23 set the clock frequency
Jun 29 14:21:54 <[g2]> I'm probably remembering wrong
Jun 29 14:21:55 <dyoung-zzzz> Okay I remember now.
Jun 29 14:22:10 <dyoung-zzzz> page 326 of 25248005.pdf
Jun 29 14:23:27 <[g2]-dinner> so addr 23 and 22 need tweaked 
Jun 29 14:23:44 <[g2]-dinner> and the 010 is undoco
Jun 29 14:23:54 <dyoung-zzzz> Well, I dunno, according to that table the 266Mhz can only be 266, 
Jun 29 14:24:00 <ep1220> i verified R64 is a pull-down, but it might be connected to the FLASH only
Jun 29 14:24:26 <dyoung-zzzz> but 101 110 and 111 are not documented.
Jun 29 14:24:41 <dyoung-zzzz> plus what [g2] said/
Jun 29 14:25:09 <beewoolie-afk> My read of those registers is that these are just the settings for memory timing.  Moreover, this register is just for the flash rom.
Jun 29 14:25:49 <beewoolie-afk> Ooops.  Wrong page.
Jun 29 14:25:57 <dyoung-zzzz> beewoolie-afk, the table is titled, "Intel XScale Core Speed Expansion Bus Configuration Strappings"
Jun 29 14:26:46 <beewoolie-afk> But there isnt anything in the manual about 133MHz.
Jun 29 14:27:04 <dyoung-zzzz> exactly my point.
Jun 29 14:27:21 <ep1220> i think the table does not necessarily imply that other settings do not work
Jun 29 14:27:21 <beewoolie-afk> Yet, that doesn't mean that they don't detune parts.
Jun 29 14:27:27 <dyoung-zzzz> I looked at this (in the manual only) before and thought tht 133Mhz was not supported.
Jun 29 14:27:47 <dyoung-zzzz> ep1220, Right.  and I'm happy that someone is looking at it physically!
Jun 29 14:27:54 <beewoolie-afk> Well, we could try psuing the value to 0b100 to see if it makes the CPU run faster.
Jun 29 14:28:24 <ep1220> the APPnote says these bits can not be written to
Jun 29 14:28:47 <ep1220> (i already tried, was not faster, but not 100% sure my prog was correct)
Jun 29 14:28:52 <dyoung-zzzz> if we compare against the 533Mhz part, it says 011 is the half speed; but 011 for the 266 is still 266.
Jun 29 14:28:55 <beewoolie-afk> Hmm.  And we're already using 0b100
Jun 29 14:29:20 <ep1220> g2: read out 0b010
Jun 29 14:29:36 <ep1220> this is nowhere in the table
Jun 29 14:30:03 <beewoolie-afk> Didn't you read out 0x0081ffee?  That looks like 0b100 to me.
Jun 29 14:30:19 <ep1220> this was on the avila
Jun 29 14:30:25 <beewoolie-afk> Ah.
Jun 29 14:30:39 <ep1220> it was 0x0041ffee on the slug
Jun 29 14:30:39 <beewoolie-afk> so, perhaps we could change one of the resistors?
Jun 29 14:30:56 <beewoolie-afk> Very interesting.
Jun 29 14:30:57 <ep1220> when we find them: yes :-)
Jun 29 14:31:06 <ep1220> but need change 2 to get 100
Jun 29 14:32:40 <beewoolie-afk> Or make it zero.
Jun 29 14:33:45 <beewoolie-afk> Making it zero would probably be easier.
Jun 29 14:33:57 <ep1220> zero is undocumented as well
Jun 29 14:34:25 <ep1220> based on the table I'd say unless a23 is 1 we might not get full speed
Jun 29 14:34:57 <beewoolie-afk> Perhaps.  All zeros is full speed in every other case.
Jun 29 14:34:59 <dyoung-zzzz> ep1220, look at that page 326, it documents 100 000 001 011 for being 266Mhz.
Jun 29 14:34:59 <ep1220> in the appnote a23 is named cfg_en_n
Jun 29 14:35:44 <dyoung-zzzz> (thats 23:21)
Jun 29 14:37:24 <ep1220> Funny, the Names of these bits are different than in the appNote
Jun 29 14:38:03 <ep1220> (on page 325)
Jun 29 14:38:59 <ep1220> the devmanual is the more recent document
Jun 29 14:39:09 <beewoolie-afk> According to the pin schedule, these are (23-21) J24 G26 F26
Jun 29 14:44:06 <beewoolie-afk> I'm not so good at the topological twistings here.
Jun 29 14:46:09 * dyoung-zzzz me toos
Jun 29 14:46:22 <dyoung-zzzz> tracking traces is painful
Jun 29 14:46:33 <ep1220> It is close to midnight here, so i too am not in the mood for tracking traces :-)
Jun 29 14:46:51 <ep1220> I give it a try tomorrow.
Jun 29 14:46:57 <VoodooZ> thanks ep1220 
Jun 29 14:47:04 <ep1220> g'night all
Jun 29 14:47:05 <dyoung-zzzz> if I'm feeling brave, I'll look at it late
Jun 29 14:47:06 <dyoung-zzzz> r
Jun 29 14:47:18 <dyoung-zzzz> printit out gigantic.
Jun 29 14:47:39 <VoodooZ> cool!
Jun 29 14:47:53 <beewoolie-afk> It looks to me to be a via to the other side of the board.
Jun 29 14:49:20 <beewoolie-afk> It would be a cinch if I had one of those bar boards.
Jun 29 14:49:24 <beewoolie-afk> s/bar/bare/
Jun 29 14:49:32 <beewoolie-afk> A DVM would make quick work of the search.
Jun 29 14:50:45 <beewoolie-afk> VoodooZ: if the zeros are pull downs and the 1's are floating, then it should be simple to bring up the high bit.
Jun 29 14:51:26 <VoodooZ> yeah.
Jun 29 14:52:17 <VoodooZ> the DS says to use 10K weak pull-downs to set the bits.
Jun 29 14:52:33 <VoodooZ> as the pins are internall pulled-up IIRC
Jun 29 14:56:44 <dyoung-zzzz> ok I printed it it super gigantic.
Jun 29 14:56:54 <beewoolie-afk> ginormous?
Jun 29 14:57:02 <dyoung-zzzz> Gimangous!
Jun 29 14:57:25 <beewoolie-afk> BTW, if this works, it will make the slug a *rockin* platform.
Jun 29 14:57:46 <dyoung-zzzz> I guess I need to either take a slug with me, or print out the populated board too
Jun 29 14:57:49 <beewoolie-afk> I'll have to buy a few extras in case it makes linksys mad.
Jun 29 14:58:21 <NAiL> dyoung-zzzz: Going to read till you fall asleep? ;)
Jun 29 15:00:53 <[g2]> hey can you guys join me in #nslu2 ?
Jun 29 17:03:35 <jbowler-zzz> The Developers manual defines these settings: 100, 000, 001, 011.  [g2] saw 010 (undefined) on NSLU2 and 100 (defined) on avila
Jun 29 17:04:47 <jbowler-zzz> So my guess is that LinkSys have screwed up and not managed to pull down A22 (to give 000), or maybe accidentally pulled down A23 instead of A22
Jun 29 17:05:49 <jbowler-zzz> <-> A22 instead of A23...  Since these pins are just configuring a clock multiplier it's entirely possible that an undefined setting gives 133MHz...
Jun 29 17:07:01 <jbowler-zzz> The weird thing is that I have seen BogoMIPS reported by the kernel as 266 on one occasion - maybe there is some power on race condition?
Jun 29 17:23:11 <NAiL> That might explain why my slug is sometimes *very* hot
Jun 29 17:23:57 <jbowler-zzz> ... if you assume LinkSys got some undocumented way of declocking the slug from Intel to fix an overheating problem - possible.
Jun 29 17:25:19 <VoodooZ> ok, so what's next? Can we try to de/solder a resistor or something?
Jun 29 17:25:27 <jbowler-zzz> Anyway, those three bits cannot be written (at least not from RedBoot - the user config bits can, but not the clock speed ones.)
Jun 29 17:25:58 <VoodooZ> the DS says to use a 10K weak pull-down
Jun 29 17:26:27 <VoodooZ> I guess we need to trace those pins first
Jun 29 17:28:02 <jbowler-zzz> The PCI clock is set to 33MHz too.  I wonder what happens if I change that...
Jun 29 17:28:18 <VoodooZ> hehehehe. Evil!!!
Jun 29 17:28:27 <beewoolie-afk> Messing with the PCI clock is unlikely to be a pleasing experience.
Jun 29 17:28:33 <NAiL> jbowler-zzz: Will that change anything? Isn't that because the ext clock is 33Mhz?
Jun 29 17:29:25 <jbowler-zzz> You are probably correct...
Jun 29 17:29:32 <beewoolie-afk> NAiL: The CPU shouldn't be hotter at one clock speed or the other because most of the time it is idle.
Jun 29 17:29:46 <NAiL> beewoolie-afk: you haven't seen my slug :-P
Jun 29 17:29:49 <beewoolie-afk> You'd only see excessive heat if you were cranking it to capacity.
Jun 29 17:29:59 <beewoolie-afk> NAiL: I have not.  What did you do to it?
Jun 29 17:30:08 <NAiL> I'm using it to compile natively
Jun 29 17:30:21 <NAiL> (and I regularly hit ~4 in loadavg with samba)
Jun 29 17:30:34 <beewoolie-afk> Ah yes.  The other thing to look at is the daya sheets.  There could be a couple of reasons to detune the CPU.
Jun 29 17:30:37 <beewoolie-afk> 1) Heat.
Jun 29 17:30:42 <beewoolie-afk> 2) Power consumption
Jun 29 17:30:47 <beewoolie-afk> 3) Marginal parts
Jun 29 17:30:57 <beewoolie-afk> aka cheaper parts.
Jun 29 17:31:15 <beewoolie-afk> BTW, IMHO, we are better off cross compiling than native compiling.
Jun 29 17:32:41 <NAiL> Yes. A package like mysql is so fun to compile natively. Especially when you forget something, and have to recompile...
Jun 29 17:32:56 <NAiL> It works just fine cross, why compile it natively?
Jun 29 17:33:17 <NAiL> (lots of stuff works nice cross, not everything though)
Jun 29 17:34:20 <beewoolie-afk> AFAICT, everything can be cross compiled.  I have a BSP builder that does the whole thing.  OE also builds everything cross-wise.
Jun 29 17:35:09 <NAiL> Some packages are worse than others. Getting samba to compile with lfs support is one. :-P
Jun 29 17:35:16 <[g2]-away> perl is not built properly
Jun 29 17:35:28 <NAiL> no, neither is mysql at the moment. 
Jun 29 17:35:38 <NAiL> I'm working on mysql though ;)
Jun 29 17:35:41 <beewoolie-afk> I am using a cross-build perl.
Jun 29 17:35:55 <[g2]-away> for xscale ?
Jun 29 17:36:02 <beewoolie-afk> The only reason for a cross-built package to fail is if there is something wrong in your setup.
Jun 29 17:36:20 <NAiL> *cough*
Jun 29 17:36:22 <beewoolie-afk> What do you think is causing the problem with cross-built perl.
Jun 29 17:36:55 <beewoolie-afk> I've spent a lot of time getting the build process right.  A lot of packages are dumb about how they build.
Jun 29 17:37:16 <beewoolie-afk> so far I've found solutions to every one...painful tho it may be.
Jun 29 17:37:43 <NAiL> I bet I compiled samba 40 times before getting it right :P
Jun 29 17:38:09 <beewoolie-afk> That's why I like cross-building on an athlon64.
Jun 29 17:38:25 <NAiL> I'm not that lucky :P
Jun 29 17:38:38 <beewoolie-afk> The BSP build takes about 6 hours on my athlon64.  I cannot imagine there are enough hours in the year to do the same natively.
Jun 29 17:38:58 <beewoolie-afk> Athlon 64's are cheap.
Jun 29 17:39:05 <VoodooZ> AMD RULES!!!
Jun 29 17:39:20 <NAiL> If you have a budget at all, they're somewhat cheap yes. 
Jun 29 17:39:37 <beewoolie-afk> If your budget is zero...well, ahem.
Jun 29 17:39:39 <NAiL> At the moment I have to squeeze my budget for the slug I just ordered :P
Jun 29 17:39:55 <VoodooZ> another good reason to overclock the hell out of it! :)
Jun 29 17:39:59 <beewoolie-afk> Save up.  My system was probably only a couple of hundred.
Jun 29 17:40:28 <jbowler-zzz> Well setting the PCI clock bit to 1 (66MHz) from RedBoot doesn't seem to stop OpenSlug booting - of course it may not have stuck...
Jun 29 17:40:42 <beewoolie-afk> If you consider the time it takes to do work on the slower machine, you could get a second job and pay for it quite quickly.
Jun 29 17:41:08 <[g2]-away> the 420/422 doesn't do pci 66 iirc
Jun 29 17:41:08 <beewoolie-afk> I think that the PCI clock setting is for the benefit of the USB controller.  does it still work?
Jun 29 17:41:29 <jbowler-zzz> It's booted off a USB disk.
Jun 29 17:41:46 <beewoolie-afk> As [g2]-away says, it may not be supported anyway.
Jun 29 17:41:58 <jbowler-zzz> As NAiL pointed out - it might not have any effect on the frequency...
Jun 29 17:42:00 <NAiL> beewoolie-afk: Uh.. cheapest AMD64 I can get my hands on is >$200 itself. Then another >$100 for a mobo, closer to $200 for memory. 
Jun 29 17:42:22 <beewoolie-afk> Hmm.  Let me look.  I am using one of the slower models...
Jun 29 17:42:27 <[g2]-away> tomorrow I'm gonna see if the big native compiles take 1/2 the wall clock time on the Avila board
Jun 29 17:43:02 <[g2]-away> it's not fully a straight comparison
Jun 29 17:43:31 <[g2]-away> slug versus 266 avila with GigeE NFS mount
Jun 29 17:43:46 <[g2]-away> the GigE ran 200Mbs plus which is about 16MBs
Jun 29 17:43:53 <[g2]-away> hmmm 24MBs
Jun 29 17:44:03 <NAiL> The slug would be brilliant with GigE :D
Jun 29 17:44:11 <[g2]-away> but the swap will have to be out there
Jun 29 17:44:14 <NAiL> (if the usb could keep up)
Jun 29 17:44:24 <[g2]-away> that'll be on NFS
Jun 29 17:44:57 <[g2]-away> the Maxtor Storage Server was clocking hdparm around 23-24MBs
Jun 29 17:45:09 <[g2]-away> from what the guy in #openmss said
Jun 29 17:45:57 <[g2]-away> :)
Jun 29 17:46:06 * dyoung-overclk bashes his head repeatedly on desk
Jun 29 17:46:13 <VoodooZ> hehe
Jun 29 17:46:26 <[g2]-away> let's see that BogoMips number 
Jun 29 17:46:44 <beewoolie-afk> NAiL: MB $65, CPU $180, RAM $60.  I had the number a little wrong, but not too far.
Jun 29 17:46:49 <dyoung-overclk> Do you guys watch tv?
Jun 29 17:47:03 <[g2]-away> a show or two on Tivo
Jun 29 17:47:03 <dyoung-overclk> specifically the geico commercials.
Jun 29 17:47:12 <NAiL> I will as soon as I get my pvr250 working :-P
Jun 29 17:47:16 <[g2]-away> skip almost all commercials
Jun 29 17:47:25 <[g2]-away> s-p-s-30-s
Jun 29 17:47:33 <dyoung-overclk> ok
Jun 29 17:47:37 <[g2]-away> Select-Play-Select-30-Select
Jun 29 17:47:46 <dyoung-overclk> I wont say it then.
Jun 29 17:47:57 <[g2]-away> it's the cheat code for the 30 second advance on the TIVO S1
Jun 29 17:48:38 <[g2]-away> it's funny my *kids* skip through most of the commericals
Jun 29 17:48:51 <beewoolie-afk> NAiL: BTW, I'm using one of the 739 socket models.  Sux for upgrade, but the 939's were way to expensive when I bought this one.
Jun 29 17:48:55 <[g2]-away> that balances out with the ones they rewind 100 times
Jun 29 17:49:07 <[g2]-away> I've got the 939
Jun 29 17:49:17 <VoodooZ> nice
Jun 29 17:49:25 <[g2]-away> It was $405 for mobo, cpu and ram
Jun 29 17:49:34 <NAiL> beewoolie-afk: Yeah, I'm gonna go for a 939 when I can afford it.
Jun 29 17:49:42 <VoodooZ> dyoung-overclk, did you overclock anything yet? :)
Jun 29 17:49:48 <NAiL> Problem is, I'm gonna upgrade my workstation first. Then my dev/server box
Jun 29 17:49:53 <dyoung-overclk> Okay I'll say it.
Jun 29 17:49:58 <dyoung-overclk> "No, but I do have good news"
Jun 29 17:50:01 <[g2]-away> I can serve up nearly 400Mbs from HD
Jun 29 17:50:11 <VoodooZ> uh?
Jun 29 17:50:17 <[g2]-away> 600 from memory
Jun 29 17:50:24 <dyoung-overclk> "I just saved a bunch of money on my car insurace by switching to geico!"
Jun 29 17:50:29 <dyoung-overclk> actually thats not it.
Jun 29 17:50:36 <VoodooZ> hehehehe
Jun 29 17:50:50 <dyoung-overclk> "But I do have good news!  I just picked up my 2 PCR1000's unblocked.
Jun 29 17:50:52 <dyoung-overclk> "
Jun 29 17:51:06 <VoodooZ> ?
Jun 29 17:51:50 <dyoung-overclk> http://www.icomamerica.com/products/receivers/pcr1000/
Jun 29 17:52:07 <[g2]-away> dyoung-overclk, so what's up with the OC ?
Jun 29 17:52:15 <[g2]-away> do you have an OC'd slug
Jun 29 17:52:19 <VoodooZ> ya
Jun 29 17:52:26 <dyoung-overclk> Not quite that far yet.
Jun 29 17:52:31 <dyoung-overclk> ask me again in 10 hours.
Jun 29 17:52:35 <dyoung-overclk> I should have it by then.
Jun 29 17:52:46 <dyoung-overclk> I gotta finish something righ tnow.  back in a bit.
Jun 29 17:52:53 <[g2]-away> GL
Jun 29 17:53:10 <beewoolie-afk> What does unblocked mean with this pcr1000?
Jun 29 17:53:28 <beewoolie-afk> Oh.  I see now.
Jun 29 17:53:45 <VoodooZ> cool. thanks dyoung. I have to leave later so I'll check back with you tommorow.
Jun 29 17:53:57 <dyoung-overclk> you need to have FCC permisison to legally buy a radio that can receive certain frequency bands.
Jun 29 17:54:20 <dyoung-overclk> type acceptance, blahblahblah
Jun 29 17:54:24 <dyoung-overclk> fill this, stamp that
Jun 29 17:54:33 <dyoung-overclk> yadda yadda yadda.
Jun 29 17:54:44 <beewoolie-afk> So, did you buy from Japan?
Jun 29 17:54:55 <[g2]-away> so ppl can monitor the cell bands
Jun 29 17:55:00 <[g2]-away> can't
Jun 29 17:55:18 <[g2]-away> we leave that for the g-MAN
Jun 29 17:55:37 <beewoolie-afk> I thought mobiles were going digital anyway.
Jun 29 17:56:00 <[g2]-away> like it's that hard decoding GSM ?
Jun 29 17:56:25 <[g2]-away> somebody deployed those systems
Jun 29 17:56:28 <[g2]-away> FOA
Jun 29 17:57:13 <beewoolie-afk> The difference is that decoding the signal requires more SW.  Listening in on the analog signals is cake with a good receiver.
Jun 29 17:58:17 <[g2]> sure AMPS is much easier
Jun 29 17:58:49 <[g2]> just tune the fwd and rev channels
Jun 29 17:59:27 <beewoolie-afk> I'm not hack in this area.  I ain't got the time to listen to other people's banal conversations.
Jun 29 18:00:13 <[g2]> I spent 5 years building/testing/deploying a big ($5B) cell system
Jun 29 18:00:25 <[g2]> it was a yob-man
Jun 29 18:00:39 <[g2]> nice consulting gig
Jun 29 18:00:53 <[g2]> ah... the 90's
Jun 29 18:01:02 <beewoolie-afk> they were gogogogogogogogogogogog
Jun 29 18:01:05 <beewoolie-afk> ...o
Jun 29 18:01:19 <[g2]> :
Jun 29 18:01:20 <[g2]> :)
Jun 29 18:01:39 <beewoolie-afk> Do you think it is right that someone should be able to be ticketed for driving a zamboni drunk?
Jun 29 18:02:01 <[g2]> during a game sure
Jun 29 18:02:16 <beewoolie-afk> when no one is on the ice?  
Jun 29 18:02:27 <beewoolie-afk> I thought that drunk driving only pertains to being on the road.
Jun 29 18:03:19 <[g2]> depends on the level
Jun 29 18:03:37 <[g2]> what .04-.06 that might be iffy
Jun 29 18:03:46 <[g2]> .10+ 
Jun 29 18:03:51 <[g2]> bad enws probably
Jun 29 18:04:06 <beewoolie-afk> The point is that the DUI is based on the hazard.  There's no one else on the ice.
Jun 29 18:04:38 <beewoolie-afk> I know it is possible to get a DUI on a bicycle, which IMHO, is ridiculous since there is no license involved.
Jun 29 18:04:43 <[g2]> it's a hazard to the life of the guy/gal driving
Jun 29 18:04:44 <beewoolie-afk> But in a rink?
Jun 29 18:04:50 <[g2]> not to mention property right ?
Jun 29 18:04:54 <beewoolie-afk> Ah, you're of that camp.
Jun 29 18:05:07 <[g2]> what responsible behavior ?
Jun 29 18:05:21 <beewoolie-afk> I don't believe in such legislation.  If a guy wants to kill himself, he can be my guest.
Jun 29 18:06:17 <[g2]> well I don't know how the legislation is written or what article/law you are talking about but it probably says
Jun 29 18:06:29 <[g2]> "Driving a vehicle"
Jun 29 18:06:38 <beewoolie-afk> http://www.thesmokinggun.com/archive/0628052zamboni1.html
Jun 29 18:08:02 <beewoolie-afk> Depends on the wording, too.  Is a bicycle a motor vehicle?
Jun 29 18:08:23 <[g2]> ok .12 with heavy machinery and ppl near by
Jun 29 18:08:54 <[g2]> that's like the law in TX which you must be at least 10 feet from the road to discharge a firearm
Jun 29 18:09:17 <[g2]> because ppl were shooting jackrabbit while driving and shooting across the lane
Jun 29 18:12:05 <[g2]> hey I *need* one of these
Jun 29 18:12:22 <[g2]> referring article says $50 bucks
Jun 29 18:12:26 <[g2]> http://www.thegreenhead.com/cool-stuff/2005/06/10-million-candlepower-spotlight.php
Jun 29 18:12:50 <[g2]> then I can run a POE device for a long time with the rechargeable battery
Jun 29 18:18:05 <NAiL> POE is nice
Jun 29 18:18:22 <[g2]> A POE Slug would be cool :)
Jun 29 18:18:38 <NAiL> Yeah :)
Jun 29 18:18:52 <NAiL> Dunno how POE actually works though
Jun 29 18:19:05 <NAiL> I've seen a POE "injector"-thing
Jun 29 18:19:44 <beewoolie-afk> NAiL: It think it puts DC on the line and super-imposes the Ethernet signals on top.
Jun 29 18:19:55 <Tiersten> It uses the unused pairs
Jun 29 18:19:56 <[g2]> no
Jun 29 18:20:02 <[g2]> nod to tman
Jun 29 18:20:07 <[g2]> the extra pairs
Jun 29 18:20:17 <[g2]> 8 wires 4 unused
Jun 29 18:20:18 <beewoolie-afk> Those won't be unused forever.
Jun 29 18:20:35 <Tiersten> They're used in certain GigE implementations
Jun 29 18:20:49 <beewoolie-afk> The old gige did, iirc.
Jun 29 18:21:14 <[g2]> now there's jumbo frames and pause frames and all kinda stuff VLAN tagging
Jun 29 18:21:26 <NAiL> aha
Jun 29 18:21:33 <NAiL> Alternative 2: The data pairs are used. Since Ethernet pairs are transformer coupled at each end, it is possible to apply DC power to the center tap of the isolation transformer without upsetting the data transfer. In this mode of operation the pair on pins 3 and 6 and the pair on pins 1 and 2 can be of either polarity.
Jun 29 18:22:27 <[g2]> Tiersten, what's the POE spec ?
Jun 29 18:22:33 <Tiersten> 802.3af
Jun 29 18:22:38 <NAiL> So basically, I can hack the cable I use for my slug to also carry power over the unused pairs, and pull them off before they reach the board?
Jun 29 18:22:45 <[g2]> that's what I thought 
Jun 29 18:22:48 <[g2]> IEEE
Jun 29 18:22:51 <Tiersten> There are two modes.  One uses the data pairs.  The other uses unused ones in 10/100baseT
Jun 29 18:22:56 <NAiL> http://www.poweroverethernet.com/articles.php?article_id=52
Jun 29 18:23:06 <NAiL> ^-- seems like ok ref
Jun 29 18:23:12 <Tiersten> NAiL: In theory yes but it's not quite as simple as that unless you buy premade PoE adapters
Jun 29 18:23:38 <NAiL> Tiersten: Why?
Jun 29 18:23:40 <Tiersten> It puts 48V down the cable as that was the telecom spec and it also should probe to see if the client device actually wants PoE and what pairs it wants it on
Jun 29 18:23:51 <Tiersten> This is if you want to do it properly that is
Jun 29 18:23:59 <NAiL> I don't care about proper! :P
Jun 29 18:24:05 * [g2] cares
Jun 29 18:24:06 <Tiersten> Nothing to stop you wiring the two extra pairs together and pumping 5V @ 2A down it :)
Jun 29 18:24:15 <NAiL> Exactly :D
Jun 29 18:24:22 <[g2]> and it's limited to 1A right ?
Jun 29 18:24:38 <Tiersten> 802.3af is 48V@350mA
Jun 29 18:25:08 <Tiersten> You're okay with putting 5V@2A down it with two pairs connected together
Jun 29 18:25:15 <NAiL> An obvious requirement of the spec is to prevent damage to existing Ethernet equipment. A "discovery process", run from the Power Sourcing Equipment (PSE), examines the Ethernet cables, looking for devices that comply with the specification. It does this by applying a small current-limited voltage to the cable and checks for the presence of a 25k ohm resistor in the remote device. Only if the resistor is present is the full 48V is applied, but this i
Jun 29 18:26:38 <Tiersten> Cisco do their own thing as well regarding PoE which isn't compatible.  But their new stuff is 802.3af compliant now
Jun 29 18:26:41 <NAiL> Are the unused pairs traced out on the board on the slug? Do I have to yank the pairs *before* the plug?
Jun 29 18:26:53 <Tiersten> Don't know.  Never looked
Jun 29 18:27:03 <NAiL> On the backside, only four are traced out
Jun 29 18:27:04 <Tiersten> You can probably see from the PCB pics on the wiki
Jun 29 18:27:10 <NAiL> ah, yes
Jun 29 18:28:23 <NAiL> Unless they're connected to a layer in the middle of the pcb, they're not traced at all
Jun 29 18:28:24 <Tiersten> Most devices are reasonably safe with having PoE applied even when it doesn't support it
Jun 29 18:29:08 <NAiL> Which (hopefully) means that I could actually just wire from the pins on the rj45 to the power input.
Jun 29 18:29:30 <Tiersten> Easy enough to test anyway.  Get the multimeter out on continuity setting
Jun 29 18:30:03 <NAiL> Then I'd have to disconnect my slug. At the moment I'm using it to store the music I'm listening to :P
Jun 29 18:30:45 <Tiersten> Sacrifices must be made :)
Jun 29 18:31:33 <NAiL> btw, anyone know if it's possible to light up the hdd lights on the slug now? 
Jun 29 18:31:55 <beewoolie-afk> What do you mean, light up the lights
Jun 29 18:32:27 <NAiL> the hdd1 and hdd2 lights never come on. I'd like to use them to show when a disk is being fsck'ed when booting
Jun 29 18:33:18 <beewoolie-afk> They light when the drive is plugged in, but they aren't activity lights.
Jun 29 18:33:24 <beewoolie-afk> You could patch the kernel to do it.
Jun 29 18:33:41 <NAiL> beewoolie-afk: They don't light up on OpenSlug. 
Jun 29 18:34:05 <NAiL> Someone did something so the status led now works. But I don't know if the hdd lights also works now.
Jun 29 18:34:08 <beewoolie-afk> Hmm.  that would be a user-space issue.
Jun 29 18:34:22 <NAiL> Tiersten: That made me none the wiser
Jun 29 18:34:46 <jbowler-zzz> leds +1 
Jun 29 18:34:48 <jbowler-zzz> leds +2
Jun 29 18:35:02 <NAiL> jbowler-zzz: Neat :)
Jun 29 18:35:04 <beewoolie-afk> If you just want to know if a drive is plugged in before user-space, you'd have to put it in the linuxrc script.
Jun 29 18:35:27 <jbowler-zzz> hotplug could do it
Jun 29 18:35:39 <jbowler-zzz> Indeed, in this case hotplug should do it.
Jun 29 18:35:46 <jbowler-zzz> (but it doesn't of course)
Jun 29 18:36:34 <jbowler-zzz> Is there any way of telling which usb socket a drive is plugged into?
Jun 29 18:36:46 <beewoolie-afk> I think so.  check lsusb.
Jun 29 18:36:47 <NAiL> jbowler-zzz: I *think* so. 
Jun 29 18:37:03 <NAiL> jbowler-zzz: with udevinfo you get that info, IIRC.
Jun 29 18:37:55 <beewoolie-afk> There had to be a way.  I moved the mouse from one port to the other on my notebook and I cannot tell which jack it is using.
Jun 29 18:38:37 <NAiL> Heh, windows keeps installing drivers each time I switch USB-ports on anything. So there's got to be a way :-P
Jun 29 18:39:01 <jbowler-zzz> cat /proc/bus/usb/devices
Jun 29 18:39:19 <NAiL> (My slug is on the table right now, so ...)
Jun 29 18:39:22 <jbowler-zzz> For my OnetTouch II I get: T:  Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  2 Spd=480 MxCh= 0
Jun 29 18:39:44 <NAiL> "Port" sounds promising ;)
Jun 29 18:39:46 <jbowler-zzz> It's in the 'disk 1' port
Jun 29 18:40:23 <beewoolie-afk> That works for me.
Jun 29 18:40:48 <NAiL> hmm
Jun 29 18:41:20 <NAiL> According to my multimeter, all the pins in the rj45 goes to a big lump'o'metal, then goes to separate pins again.
Jun 29 18:41:35 <VoodooZ> good night guys...
Jun 29 18:41:51 <NAiL> There's no resistance between any combination of pins, and when I buzz I get buzz on *all* pins.
Jun 29 18:42:02 <NAiL> I don't really get that...
Jun 29 18:42:06 <beewoolie-afk> Nite, VoodooZ_Log 
Jun 29 18:42:23 <beewoolie-afk> NAiL: Depends on the circuit.
Jun 29 18:43:05 <beewoolie-afk> IIRC, the interface for those pins often use pull-ups or pull-downs.  Depends on the manufacturers spec.
Jun 29 18:43:09 <NAiL> Well.. anyways it sounds like it's not exactly safe to run the power through the rj45 plug. I'll just have to take it from before the plug.
Jun 29 18:44:42 <NAiL> I'll wait with that until I get another slug next week
Jun 29 18:44:47 <jbowler-zzz> Ok, with some USB flash device in 'disk 2' I get: T:  Bus=01 Lev=01 Prnt=01 Port=01 Cnt=02 Dev#=  4 Spd=480 MxCh= 0
Jun 29 18:45:06 <NAiL> Port and Count changed
Jun 29 18:45:50 <NAiL> Hopefully, if you swap 'em, the flash disk ends up at pport 1, and the maxtor ends up at port 2
Jun 29 18:45:51 <jbowler-zzz> and Dev#
Jun 29 18:46:00 <NAiL> yeah
Jun 29 18:46:03 <NAiL> Didn't notice that one
Jun 29 18:46:19 <jbowler-zzz> Right, but I'm not going to try hotswapping /, I suspect it won't work.
Jun 29 18:46:34 <NAiL> haha, no
Jun 29 18:47:30 <Tiersten> If you read the sercomm sources for the NSLU2 you'll find that's how they're doing the port 1 = /dev/sda and port 2 = /dev/sdb thing
Jun 29 18:47:44 <NAiL> aha
Jun 29 18:48:30 <NAiL> That's possible to fake with udev
Jun 29 18:48:50 <beewoolie-afk> udev is cool.
Jun 29 18:49:05 <NAiL> Udev isn't working as it should atm :P
Jun 29 18:49:06 <jbowler-zzz> So it's possible to do in user space - take the device being checked (/dev/sd??), look up that Dev# (probably) in /proc and do leds !1 or leds !2 as appropriate, then do leds +1 on the mount (mountall.sh).
Jun 29 18:49:32 <Tiersten> Yeah but sercomm are a cowboy operation and just hack it so it just works
Jun 29 18:49:42 <jbowler-zzz> Ye ha!
Jun 29 18:49:55 <Tiersten> heh
Jun 29 18:50:28 <NAiL> jbowler-zzz: plug in the device, udev finds which port and makes a symlink. In the hotplug script you can see which device is being plugged in, and thus light the correct light.
Jun 29 18:50:47 <NAiL> I'm working on a hotplug script for udev/openslug ;-)
Jun 29 18:50:52 <jbowler-zzz> Having udev do /dev/usb1* and /dev/usb2* would be good ;-)
Jun 29 18:51:27 <NAiL> No problem, as long as it doesn't already exist.
Jun 29 18:51:35 <jbowler-zzz> I.e. I don't want to know that sda? was inserted before sdb?, I want to know which damned port it's been plugged into (as a user).
Jun 29 18:52:09 <NAiL> no problem whatsoever :D
Jun 29 18:53:43 <jbowler-zzz> Hum.  How many layers on the NSLU2 board?  Just the two we see?
Jun 29 18:56:03 * NAiL makes a wild guess at 3 ;)
Jun 29 18:56:12 * [g2] guesses 6
Jun 29 18:56:26 <[g2]> maybe 8
Jun 29 18:56:36 <[g2]> most likely 6
Jun 29 18:56:39 <NAiL> Why that many?
Jun 29 18:56:50 <[g2]> 498 pins in a BGA
Jun 29 18:57:01 <[g2]> or however many there are
Jun 29 18:57:34 <[g2]> plus I guess you need some ground planes
Jun 29 19:00:22 <jbowler-zzz> So finding where those (that) 10k pull down resistor is could be quite difficult...  (There are a lot of 10k resistors on the board).
Jun 29 19:01:43 <[g2]> allan's our man
Jun 29 19:04:01 <[g2]> dyoung, does the unslung native make for perl / php run the tests too ?
Jun 29 19:07:02 <NAiL> jbowler-zzz: leds ![12] makes 'em flash?
Jun 29 19:07:10 <NAiL> duh
Jun 29 19:07:14 * NAiL tries instead
Jun 29 19:07:20 <jbowler-zzz> ;-)
Jun 29 19:07:56 <NAiL> haha
Jun 29 19:08:07 <NAiL> using ! as a parameter is not always bright
Jun 29 19:08:08 <NAiL> leds vi ipkg.conf
Jun 29 19:08:38 <NAiL> leds \!1 worked as it should though ;-)
Jun 29 19:09:07 <jbowler-zzz> ha!  not a problem with ash...
Jun 29 19:09:13 <NAiL> no, but I've got bash
Jun 29 19:09:31 <jbowler-zzz> I thought about that, then figured ash would also be used in the boot scripts, but maybe not.
Jun 29 19:09:35 <NAiL> Interesting trivia: bash is pronounced like poo in norwegian.
Jun 29 19:11:08 <NAiL> hmm.. I've gotta take a long hard look at the bootscripts and see if I can make them act like I want
Jun 29 19:11:35 <jbowler-zzz> I started that but didn't want to deviate too much from the OE boot.
Jun 29 19:11:46 <NAiL> signaling fsck & lighting the correct led on mount would be very nice
Jun 29 19:11:53 <jbowler-zzz> That said, I think the OE boot sequence is damaged in several areas.
Jun 29 19:12:18 <jbowler-zzz> Right.
Jun 29 19:13:05 <jbowler-zzz> It's very tempting just to give up on the OE packages/initscripts - the main maintenance work seems to be to add machine specific stuff...
Jun 29 19:15:32 <NAiL> Haven't looked that hard on the startup stuff yet. As soon as I get serial up I'm gonna do so.
Jun 29 19:15:45 <jbowler-zzz> bash is safe with ! in scripts (it doesn't treat it as a history escape.)
Jun 29 19:16:00 <NAiL> Good :)
Jun 29 19:16:33 <NAiL> neat. I can make led1 and led2 flash alternatingly
Jun 29 19:16:45 <jbowler-zzz> That's by timing though...
Jun 29 19:16:54 <NAiL> (means: It's my period. I'm not gonna boot until you give me some chocholate!)
Jun 29 19:17:29 <NAiL> Yeah, but shouldn't be hard to do in a script?
Jun 29 19:17:39 <NAiL> I don't know how fast they flash
Jun 29 19:17:47 <jbowler-zzz> If a big ethernet packet comes in between the two commands...
Jun 29 19:18:42 <jbowler-zzz> One of the things I've thought about doing is changing the kernel nslu2-io.c so that it uses just one timer, not four (one per LED) and allows precise setting of the on/off time of each LED.
Jun 29 19:19:31 <jbowler-zzz> Using one timer must be good for performance, and different flash rates is useful for UI.
Jun 29 19:19:43 <NAiL> yeah
Jun 29 19:19:58 <NAiL> well.. Actually I can't think of more usages of the leds right now ;-)
Jun 29 19:20:15 <NAiL> But then again, 640k should be enough for anybody
Jun 29 19:20:16 <jbowler-zzz> ;-) but it's the only output device!
Jun 29 19:20:40 <NAiL> ... --- ...
Jun 29 19:21:00 <[g2]> there is the speaker :)
Jun 29 19:21:47 <jbowler-zzz> Oh yes, the squeaker - and, for that matter, the power LED (but it can only be switched off.)
Jun 29 19:21:56 <NAiL> -.- . .-. -. . .-.. / .--. .- -. .. -.-. / ..--.. / 
Jun 29 19:22:17 <NAiL> A user would understand that that is "Kernel Panic", right? ;)
Jun 29 19:22:21 <[g2]> ~decode_morse
Jun 29 19:22:25 <jbowler-zzz> Not my strong point, morse code.
Jun 29 19:22:49 <jbowler-zzz> In fact not my strong point language!=english
Jun 29 19:23:02 <dyoung-overclk> I need a quick and dirty way to strip the 8th bit off of some defective ascii screams.
Jun 29 19:23:04 <dyoung-overclk> streams
Jun 29 19:23:34 <dyoung-overclk> anyone have any UBE suggestions for me/
Jun 29 19:23:41 <NAiL> tr?
Jun 29 19:23:50 <NAiL> urgh, my inet line is failing again
Jun 29 19:23:52 <[g2]> 8th as in high order ?
Jun 29 19:23:53 <dyoung-overclk> man tr
Jun 29 19:23:57 <dyoung-overclk> [g2] yes.
Jun 29 19:24:07 <[g2]> %128 ?
Jun 29 19:24:46 <NAiL> tr -d \128-\255 <-- dunno, might work
Jun 29 19:24:55 <dyoung-overclk> [g2] in C?
Jun 29 19:25:04 <[g2]> yeah
Jun 29 19:25:14 <dyoung-overclk> yeah, I was trying to avoid writing a c program. :-)
Jun 29 19:25:15 <[g2]> isn't % the modulo operator
Jun 29 19:25:22 <[g2]> oh then perl
Jun 29 19:25:42 <[g2]> just open a binary file
Jun 29 19:26:08 <jbowler-zzz>  #include <stdio.h> int main() { int ch; while ((ch=getchar())!=-1) putchar(ch&0x7f); return 0; }
Jun 29 19:26:09 <[g2]> but sed or tr would probably work
Jun 29 19:26:39 <[g2]> LOL -- jbowler-zzz  ROCKS with the 1-liner
Jun 29 19:27:16 <[g2]> foo.. | strip | ...
Jun 29 19:27:44 <jbowler-zzz> [g2]: one liner requires \n after the #include, alas.
Jun 29 19:28:03 <dyoung-overclk> holy shit
Jun 29 19:28:06 <[g2]> ;
Jun 29 19:28:11 <dyoung-overclk> now THATS what I'm talking about.
Jun 29 19:28:33 <dyoung-overclk> hug jbowler
Jun 29 19:29:29 <dyoung-overclk> okay gotta get the info back to our tech in mauritius.
Jun 29 19:29:32 <dyoung-overclk> be back in a sec
Jun 29 19:30:04 <jbowler-zzz> devio '$(1; wb @127&1;$)1'
Jun 29 19:30:12 <NAiL> haha
Jun 29 19:30:15 <jbowler-zzz> It terminates with a (read) error.
Jun 29 19:30:42 * NAiL goes and cuts off his internet cable
Jun 29 19:30:44 <NAiL> brb
Jun 29 19:39:24 <NAiL> there
Jun 29 19:40:43 <dyoung-overclk> Done, thanks, that saved me a lot of time.
Jun 29 19:41:51 <NAiL> More time for overclocking! :-P
Jun 29 19:43:47 <dyoung-overclk> Right.
Jun 29 19:44:14 * NAiL is just being annoying
Jun 29 19:44:45 <dyoung-overclk> No, that was taking away from my valuable tracking down the trace to overclock time.
Jun 29 19:56:30 <jbowler-zzz> overclocking time?  That's, like, all our watches slow down so it looks like we have 266MHz slugs, despite LinkSys?
Jun 29 19:59:19 <[g2]> must resist.... no overclock ....
Jun 29 19:59:48 <NAiL> Imagine a BeoWulf-cluster of OC'ed slugs! :-P
Jun 29 20:00:07 <[g2]> can you say fire hazard ? :)
Jun 29 20:00:08 <dyoung-overclk> strange thing happened just now.
Jun 29 20:00:21 <[g2]> black smoke was let out ?
Jun 29 20:00:21 * NAiL listens intently
Jun 29 20:00:28 * [g2] too
Jun 29 20:00:29 <dyoung-overclk> for some reason, everything is moving twice as fast.
Jun 29 20:00:45 <dyoung-overclk> suddenly my T42 is 3.4Ghz.
Jun 29 20:00:46 <NAiL> [g2]: In my experience, the magic smoke tends to be blueish gray ;)
Jun 29 20:01:03 <dyoung-overclk> and my cellphone minutes get used twice as fast.
Jun 29 20:01:08 <NAiL> haha
Jun 29 20:01:12 <[g2]> Ahhh... the black smoke went it
Jun 29 20:01:14 <[g2]> Ahhh... the black smoke went in
Jun 29 20:01:30 * NAiL wonders what dyoung is smoking ;)
Jun 29 20:01:57 <dyoung-overclk> Ganja Pot Budz.
Jun 29 20:02:02 <dyoung-overclk> Woot Dawg!
Jun 29 20:02:05 <[g2]> I think dyoung is running slugtime* 1/2
Jun 29 20:02:14 <[g2]> or * 2
Jun 29 20:06:12 <NAiL> woohooo
Jun 29 20:06:14 <NAiL> neat
Jun 29 20:06:15 <NAiL> disco
Jun 29 20:07:37 <NAiL> leds \!A;sleep 1;leds \!g;leds \!1 ;sleep 1;leds \!2;
Jun 29 20:08:24 <NAiL> Neat. Status cycles through three colors and blinks, while leds 1&2 flashes alternatingly :-D
Jun 29 20:09:21 * NAiL adds HowTo/SlugOnAcid page
Jun 29 20:11:10 * ByronT removes NAiL
Jun 29 20:11:29 <NAiL> *sigh*
Jun 29 20:11:43 <ByronT> :)
Jun 29 20:12:39 <[g2]> nite all
Jun 29 20:12:56 <[g2]> dyoung, gl with the OC
Jun 29 20:13:05 <[g2]> NAiL, GL with the serial
Jun 29 20:16:56 <dyoung-web> R81/R82
Jun 29 20:17:20 <dyoung-web> are A22 and A21.
Jun 29 20:20:54 <dyoung-web> R81 not populated.
Jun 29 20:21:06 <dyoung-web> R82 is 10k
Jun 29 20:21:37 <dyoung-web> I dunno what to yet, I need to find a physical slug for that info.
Jun 29 20:31:43 <jbowler-zzz> A23 should have 10k on it too.
Jun 29 20:34:02 <jbowler-zzz> probably R83, also populated with 10k
Jun 29 20:35:14 <dyoung-web> Actually I think A23 ep1220 traced out to R64.
Jun 29 20:35:19 <dyoung-web> I didnt double verify yet.
Jun 29 20:36:01 <dyoung-web> that ones a pull down.
Jun 29 20:36:09 <jbowler-zzz> populate R81 to get one of the 533MHz settings (000) - that should reveal what the actual core (internal) clock setting is
Jun 29 20:36:47 * NAiL still hasn't found R81/82/64
Jun 29 20:37:09 <NAiL> ah, 64.
Jun 29 20:37:13 <jbowler-zzz> top left on the picture on the web site
Jun 29 20:38:45 <dyoung-web> I agree.
Jun 29 20:38:58 <dyoung-web> if someone wants to put a 10k on R81, that would be helpful.
Jun 29 20:39:39 <dyoung-web> Nail, R81/82 are nedxt to the power button on the component side
Jun 29 20:40:16 <beewoolie-afk> g
Jun 29 20:40:24 <dyoung-web> Yo G Dawg.
Jun 29 20:40:31 <NAiL> ah, yeah, found them
Jun 29 20:40:40 <beewoolie-afk> dyoung-web: Nice work.  I thought, tho, that A22 is the one pulled high.
Jun 29 20:40:48 <NAiL> Now, if I was leet with SMD, I'd do it
Jun 29 20:40:59 <beewoolie-afk> Or, is it that the value is 1 unless there's a pull-down.
Jun 29 20:41:02 <dyoung-web> I dunno, I need a slug to see what R82 is pulled to.
Jun 29 20:41:10 <dyoung-web> I dunno if thats a ground or vcc bus
Jun 29 20:41:47 <dyoung-web> I can check later; but I have an appt with About To Be Angry Woman.
Jun 29 20:42:04 <NAiL> gl, dyoung
Jun 29 20:42:26 <beewoolie-afk> Good work.
Jun 29 20:42:50 <beewoolie-afk> I've got a slug here, but I only see R84 on the solder mask.
Jun 29 20:43:18 <NAiL> beewoolie-afk: the ones below are R80, 82, 83, 81
Jun 29 20:43:30 <NAiL> If you see just above the flash, you'll see the numbers there
Jun 29 20:43:35 <NAiL> (beneath the serial header)
Jun 29 20:43:56 <NAiL> With a white line going from the box with the numbers to the box with the R's
Jun 29 20:44:00 <beewoolie-afk> I see it now.
Jun 29 20:44:13 <beewoolie-afk> So, which resistor is on which line?
Jun 29 20:45:20 <dyoung-web> The legend is extended.
Jun 29 20:45:42 <dyoung-web> The bottom one is R81 (the only one with nothing on the pads)
Jun 29 20:46:30 <dyoung-web> so from top to bottom, R84 80 82 83 81
Jun 29 20:46:53 <beewoolie-afk> I got that.  Which resistor is on which address line?
Jun 29 20:47:05 <beewoolie-afk> ~praise dyoung-web 
Jun 29 20:47:06 <purl> All hail dyoung-web!
Jun 29 20:47:06 <NAiL> 05:16 < dyoung-web> R81/R82
Jun 29 20:47:06 <NAiL> 05:17 < dyoung-web> are A22 and A21.
Jun 29 20:47:30 <dyoung-web> oh, my bad.
Jun 29 20:48:21 <beewoolie-afk> So, Moving the resistor from R64 ro R81would put us in the 0b100 mode, right?
Jun 29 20:48:39 <dyoung-web> A22 = R81, A23 = R82
Jun 29 20:48:52 <beewoolie-afk> Oh.  That's different.
Jun 29 20:49:03 <dyoung-web> AIG
Jun 29 20:49:04 <dyoung-web> sorry
Jun 29 20:49:11 <beewoolie-afk> So, moving the R82 resistor to R81 would be the right thing.
Jun 29 20:49:12 <dyoung-web> cant type because my ass is about to be kicked.
Jun 29 20:49:18 <dyoung-web> sorry lemme get this straight.
Jun 29 20:49:21 <beewoolie-afk> Dude!  We need you.
Jun 29 20:49:27 <beewoolie-afk> don't get all ass-kicked.
Jun 29 20:49:46 <dyoung-web> A22 = R81 A21 = R82 A23 = R64
Jun 29 20:49:59 <beewoolie-afk> If you think that's what we need, I'll see if I have time to try it tonight.
Jun 29 20:50:06 * NAiL watches a towering shadow grow behind dyoung
Jun 29 20:50:36 <beewoolie-afk> Right.  So I was right the first time.  Moving R64 to R81 should do the trick.
Jun 29 20:50:45 <dyoung-web> Right.
Jun 29 20:50:50 <dyoung-web> sorry for my ignorance.
Jun 29 20:50:51 <beewoolie-afk> Okey dokey.
Jun 29 20:51:03 <NAiL> have fun :)
Jun 29 20:51:05 <beewoolie-afk> ~praise dyoung-web 
Jun 29 20:51:05 <purl> All hail dyoung-web!
Jun 29 20:51:08 <dyoung-web> okay I'll check bakc after my ass gets kicked.
Jun 29 20:51:14 <beewoolie-afk> doh!
Jun 29 20:51:18 <dyoung-web> :-)
Jun 29 20:51:21 <dyoung-web> Bye
Jun 29 22:19:15 <jbowler-zzz> If you have a steady hand and a 1/8W or similar 10k then holding it onto the unused pads (R81) will pull down A22 - currently floating high - and put b000 onto A21:23, which Intel documents as 533MHz (option 2 - I suspect 100 is actually a 650MHz setting).
Jun 29 22:19:51 <jbowler-zzz> It only needs to be pulled down during power on - after that the value is locked in.
Jun 29 22:20:25 <NAiL> jbowler-zzz: It works.
Jun 29 22:20:35 <NAiL> My slug has 266 bogomips
Jun 29 22:20:39 <jbowler-zzz> Yeah - reading the wrong channel ;-)
Jun 29 22:20:42 <NAiL> :D
Jun 29 22:21:01 <jbowler-zzz> Now you need a temp probe, just to see how hot it gets...
Jun 29 22:21:32 <NAiL> It's still not body temperature, so I'm not worried. If it get hot, I'll get worried.
Jun 29 22:21:33 <jbowler-zzz> I think my slug might generate two holes and a switch...
Jun 29 22:21:49 <NAiL> two holes and a switch?
Jun 29 22:23:00 <jbowler-zzz> 1/8W 10k resistor in series with a switch, poking out of the case...
Jun 29 22:23:13 <NAiL> haha, ok
Jun 29 22:23:16 <jbowler-zzz> (I don't have any surface mount components)
Jun 29 22:23:45 <jbowler-zzz> ~hail dyoung
Jun 29 22:23:46 * purl bows down to dyoung and chants, "I'M NOT WORTHY!!"
Jun 29 22:32:00 <beewoolie-afk> I'd be circumspect about putting a switch in this line.  It's an address line after all.
Jun 29 22:32:54 <NAiL> If he puts the resistor parallell to the switch, it'll work, won't it?
Jun 29 22:33:14 <NAiL> circuit broken = 10K, circuit closed = 0K, right?
Jun 29 22:33:37 <NAiL> The length shouldn't be too much either?
Jun 29 22:33:52 * NAiL is hopeful
Jun 29 22:34:14 * NAiL dreams of a 650Mhz slug :P
Jun 29 22:34:40 <NAiL> 400 or 533 would be extremely nice too ;-)
Jun 29 22:34:55 <NAiL> But I guess 400 would require a passive heatsink
Jun 29 22:35:06 <ByronT-Away> I have those....
Jun 29 22:35:31 <ByronT-Away> I had to mod a Linksys SD2008 (gigabit 8-port switch)
Jun 29 22:35:33 <NAiL> I can glue on a P4 heatsink! :)
Jun 29 22:35:42 <ByronT-Away> louder than a plane taking off...
Jun 29 22:35:50 <NAiL> haha
Jun 29 22:36:08 <NAiL> I think the heatsinks in my 16-port switch will do. It's semi-dead anyways
Jun 29 22:36:22 <beewoolie-afk> NAiL: it makes the trace significantly longer when the resistor is in the circuit.
Jun 29 22:36:39 <NAiL> yeah... 
Jun 29 22:36:47 <ByronT-Away> just google "SD2008 loud" to see what I mean
Jun 29 22:36:48 <NAiL> (don't discourage him, it might work!) :P
Jun 29 22:36:49 <beewoolie-afk> This part doesn't go that fast.  266 is the max.
Jun 29 22:37:10 <beewoolie-afk> I doubt anything I write will seriously inhibit anyone on this channel.
Jun 29 22:37:31 <NAiL> This switch is LOUD. VERY LOUD. Louder than all of my other computers and ...
Jun 29 22:38:51 <beewoolie-afk> ByronT-Away: did the heatsink mean you could remove the fan?
Jun 29 22:39:03 <beewoolie-afk> I have one of the smaller ones and it drives me nuts.
Jun 29 22:40:35 <ByronT-Away> no, I went with some RAM heatsinks and a slower fan
Jun 29 22:41:12 <ByronT-Away> here's where I got my inspiration for doing it... http://soltesz.net/sd2008/
Jun 29 22:42:29 <ByronT-Away> I have found that if something gets stacked on it, or it gets covered up, then it will lock up... but as long as it's room temp, and nothing is blocking air flow, it's happy
Jun 29 22:42:37 <ByronT-Away> (and now almost silent)
Jun 29 22:42:45 <ByronT-Away> in a Living room, no less
Jun 29 22:46:10 <beewoolie-afk> Nice.
Jun 29 22:47:29 <ByronT-Away> a slower fan with heatsinks definitely works, but IIRC finding a fan that was small enough (height wise) and slower was a PITA
Jun 29 22:49:03 <beewoolie-afk> Thanks for the tip.  I like the device, but the noice is definitely a problem.
Jun 29 22:51:00 <ByronT-Away> 3dcool.com had the fan I needed
Jun 29 22:51:28 <ByronT-Away> (expensive though)
Jun 29 22:51:52 <NAiL> Anyone got a temp probe I could use? ;)
Jun 29 22:52:25 <NAiL> Closest I get is the temp probe from a temp-controlled fan
Jun 29 22:52:34 <NAiL> But there's no way I can calibrate it :(
Jun 29 22:52:47 <beewoolie-afk> It is worthwhile getting one of the DVMs with a temp probe.
Jun 29 22:52:54 <ByronT-Away> just keep the finger there and see how long it takes to blister it
Jun 29 22:52:55 <beewoolie-afk> I bought the F179 Fluke.
Jun 29 22:53:04 <ByronT-Away> or what beewoolie-afk said
Jun 29 22:53:10 <NAiL> ByronT-Away: It won't blister at this temp ;)
Jun 29 22:53:35 <NAiL> It's like the forehead of someone with a fever. So I guess it's somewhere around 40
Jun 29 22:53:58 <NAiL> Touching it lowers the temp rather quickly though
Jun 29 22:56:29 * NAiL notes he is using Celcius, not Fahrenheit :-P
Jun 29 22:56:54 <ByronT-Away> yeah, that was easy to figure out
Jun 29 22:56:57 <beewoolie-afk> We may need to put a heatsink on it.
Jun 29 22:57:07 <beewoolie-afk> and cut holes in the case.
Jun 29 22:57:21 <NAiL> Don't need a huge heatsink tho
Jun 29 22:57:55 <ByronT-Away> beewoolie-afk, now you have another reason to buy some RAM heatsinks
Jun 29 22:58:02 <beewoolie-afk> Yipee!
Jun 29 22:58:05 <ByronT-Away> btw, the case already has holes in it
Jun 29 22:58:13 <NAiL> Yeah, but they're kinda small
Jun 29 22:58:32 <NAiL> The slug is noticeably hotter when its running inside the case
Jun 29 22:59:13 <ByronT-Away> I don't know that larger holes would help that much.... the problem is more of air flow
Jun 29 22:59:14 <NAiL> I can't try putting my slug inside its case, because the case is kinda... open ;)
Jun 29 22:59:34 <ByronT-Away> maybe holes in the top to allow the heat somewhere to escape
Jun 29 22:59:34 <beewoolie-afk> ByronT-Away: if the Broadcom datasheet says that the parts in the Linksys device don't need a fan or heatsink, why is one needed?
Jun 29 23:00:17 <ByronT-Away> I would think it's a cumalitive effect of the other components in the SD2008
Jun 29 23:00:55 <ByronT-Away> those times that the device locked on me, the case was hot enough that I would not call it comfortable to hold
Jun 29 23:01:48 <beewoolie-afk> I've never tried to run it that hard.
Jun 29 23:01:59 <beewoolie-afk> I don't think I have enough Gb devices to test it.
Jun 29 23:02:31 <beewoolie-afk> Let me get this straight.  You made your mods because it was failing, not just because of the noise?
Jun 29 23:02:46 <ByronT-Away> no, I made the mods *because* of the noise
Jun 29 23:03:24 <ByronT-Away> I only have two ethernet ports next to my Lazy Boy in the living room
Jun 29 23:03:56 <ByronT-Away> but I had two slugs and a laptop... so I needed a switch there... and I happened to find the SD2008
Jun 29 23:04:37 <ByronT-Away> I ignored the warnings about the noise - until I plugged it in... and the wife started staring at me... with *that* look
Jun 29 23:07:24 <beewoolie-afk> So, when were you finding that the device locked up?
Jun 29 23:07:48 <ByronT-Away> after I modded it - and left a bunch of books covering it
Jun 29 23:07:59 <beewoolie-afk> Ah, I see.
Jun 29 23:08:13 <beewoolie-afk> The metal case is probably part of the cooling design.
Jun 29 23:08:43 <ByronT-Away> when I pulled the book that was on top of another book on top of it, and it was warm, i knew nothing needs to be on it or around it
Jun 29 23:09:29 <beewoolie-afk> Mine is running my 'backbone' at the moment.  I'll disassemble it later when I can swap it for a hub.
Jun 29 23:09:40 <ByronT-Away> yes, the case is definitely part of the cooling design
Jun 29 23:10:14 <beewoolie-afk> the only thing I don't like is the way the lights work.  It is clear that there is an internal process handling the light blinking that doesn't do so smoothly.
Jun 29 23:11:08 <ByronT-Away> the blue leds from my bytecc ext. hd cases drown out the blinking from the SD2008
Jun 29 23:12:01 <beewoolie-afk> You are a fancy one..."the BLUE leds outshine the tawdry switch lights..."
Jun 29 23:13:12 <ByronT-Away> two slugs, two bytecc hd enclosures, one lacie big disk extreme, one gb switch later... and the wife thinks that the living room is a partial disco when the lights are off
Jun 29 23:13:32 <beewoolie-afk> At least it's quiet.
Jun 29 23:14:25 <ByronT-Away> for the time being... I've got a build machine clone that I have to setup somewhere
Jun 29 23:14:55 <ByronT-Away> I think the wife is going to relocate me and my devices...
Jun 29 23:17:31 <beewoolie-afk> Frankedslug is powered off.
Jun 29 23:17:40 <beewoolie-afk> Note
Jun 29 23:17:42 <beewoolie-afk> Nite
**** ENDING LOGGING AT Wed Jun 29 23:59:56 2005