**** BEGIN LOGGING AT Mon Jun 13 23:59:56 2005
Jun 14 15:33:07 <[g2]> beewoolie, DUDE you rock!
Jun 14 15:33:19 <beewoolie> [g2]: it doesn't work for you yet.
Jun 14 15:33:22 <beewoolie> Or does it?
Jun 14 15:33:34 <[g2]> well it mostly works
Jun 14 15:33:40 <[g2]> it doesn't fully work
Jun 14 15:33:52 <[g2]> but you still ROCK
Jun 14 15:33:56 <beewoolie> I know that you were having problems with listing the /boot directory.
Jun 14 15:34:04 <[g2]> nod.
Jun 14 15:34:05 <beewoolie> And then reading from an offset didn't work.
Jun 14 15:34:13 <beewoolie> Can you load a kernel anyway?
Jun 14 15:34:14 <[g2]> nod.
Jun 14 15:34:24 <[g2]> haven't tried loading the kernel
Jun 14 15:34:42 <[g2]> I was trying to load a file first and then I was going to dump memory and verify the contents
Jun 14 15:35:24 <beewoolie> I haven't tested the code on a slug yet.  I've been working with one of my dev boards instead since it is hooked up and going.
Jun 14 15:36:03 <[g2]> this was running on the fatslug
Jun 14 15:36:13 <beewoolie> That ought not matter.
Jun 14 15:36:29 <[g2]> nod... that's why we try it right ? :)
Jun 14 15:36:38 <beewoolie> Indeed.
Jun 14 15:36:52 <[g2]> is there anything I can try for you ?
Jun 14 15:37:17 <beewoolie> I'm thinking on it right now.
Jun 14 15:37:24 <[g2]> OK
Jun 14 15:37:27 <beewoolie> ...actually, you could try one thing tho.
Jun 14 15:37:32 <[g2]> sure name it
Jun 14 15:38:10 <beewoolie> If you build the loader with the #define TALK defined in src/drivers/drv-jffs2.c and then try info on the /boot directory, it will write a bunch of stuff to the console.
Jun 14 15:38:17 <beewoolie> Capture it and send it by email.
Jun 14 15:38:41 <beewoolie> Also, try the copy command and send that output.
Jun 14 15:40:47 <[g2]> you mean the last one from the e-mail I sent you ?
Jun 14 15:40:52 <[g2]> I tried that with foo
Jun 14 15:40:58 <beewoolie> yeah, that's it.
Jun 14 15:41:06 <[g2]> that was in the last e-mail
Jun 14 15:41:12 <beewoolie> The TALK output will tell me what it's doing.
Jun 14 15:41:19 <[g2]> ah
Jun 14 15:41:26 <beewoolie> You sent me the console output with TALK enabled?
Jun 14 15:41:39 <[g2]> it's built and scp'd to the host, I"m verifying the MD5 then I'll reflash
Jun 14 15:42:01 <beewoolie> You can run it without reflashing.
Jun 14 15:42:05 <beewoolie> That's what I do.
Jun 14 15:42:15 <beewoolie> Just load it to 0x0 and then g 0x0
Jun 14 15:42:22 <beewoolie> Safer than reflashing.
Jun 14 15:42:43 <[g2]> you load over serial ?
Jun 14 15:42:49 <beewoolie> Usually.
Jun 14 15:42:52 <beewoolie> It doesn't take long.
Jun 14 15:43:03 <beewoolie> apex> xr 0
Jun 14 15:44:13 <[g2]> or so just xr 0 and then xmodem the file ?
Jun 14 15:44:18 <beewoolie> Yep.
Jun 14 15:44:24 <[g2]> I've been wanting to do this
Jun 14 15:44:33 <beewoolie> it takes a couple of seconds for the two ends to sync.
Jun 14 15:47:02 <[g2]> I'm trying the xmodem send from minicom which isn't clear
Jun 14 15:47:15 <beewoolie> Send-> Xmodem
Jun 14 15:47:28 <beewoolie> The interface for browsing file sux.
Jun 14 15:47:52 <[g2]> I figured that much ... I didn't see the tabs at the bottom :)
Jun 14 15:50:37 <[g2]> minicom still isn't playing well :*
Jun 14 15:50:53 <beewoolie> Do you have rzsz installed?
Jun 14 15:51:05 <[g2]> dunno
Jun 14 15:51:16 <beewoolie> try 'which sx"
Jun 14 15:51:29 <[g2]> no sz
Jun 14 15:51:31 <[g2]> no sx
Jun 14 15:51:37 <beewoolie> There's your problem.
Jun 14 15:51:47 <beewoolie> The package on debian is lrzsz.
Jun 14 15:53:16 <[g2]> ok it's emerged
Jun 14 15:53:53 <[g2]> Sending apex.bin, 369 blocks: Give your local XMODEM receive ?
Jun 14 15:53:53 <[g2]> stopped k?command now. 
Jun 14 15:54:15 <beewoolie> Hmm/
Jun 14 15:54:17 <beewoolie>  ?
Jun 14 15:54:26 <[g2]> it might be sending
Jun 14 15:54:42 <[g2]> it's certainly better after the emerge :)
Jun 14 15:55:03 <[g2]> Ok I think it worked
Jun 14 15:55:10 <[g2]> I'm back at the APEX prompt
Jun 14 15:55:18 <beewoolie> Nice, eh?
Jun 14 15:55:26 <[g2]> yeah very nice
Jun 14 15:55:32 <beewoolie> I think there is a bug in it, tho.  it doesn't want to write directly to flash at the moment.
Jun 14 15:55:38 <beewoolie> It should be possible to do it.
Jun 14 15:55:46 <[g2]> so go 0
Jun 14 15:56:05 <beewoolie> Yep.
Jun 14 15:56:40 <[g2]> it's unhappy
Jun 14 15:56:50 <[g2]> x-mas tree lights all lit up
Jun 14 15:56:57 <beewoolie> Hmm.
Jun 14 15:57:02 <[g2]> go 0
Jun 14 15:57:02 <[g2]> Jumping to 0x00000000...
Jun 14 15:57:09 <beewoolie> Crash?
Jun 14 15:57:17 <beewoolie> Hmm.
Jun 14 15:57:23 <[g2]> that'd be my guess
Jun 14 15:57:26 <[g2]> or lockup
Jun 14 15:57:44 <[g2]> power-cycle fixes it
Jun 14 15:57:56 <beewoolie> Well, you can try writing it to flash to see if it's happier.  I'll hook it up in a little while to see if I can determine the cause of the failure.
Jun 14 15:58:04 <beewoolie> All on means something, too.
Jun 14 15:58:21 <[g2]> I'd imagine TALK is pretty safe right ?
Jun 14 15:58:47 <beewoolie> It will make it more verbose, that's all.
Jun 14 15:59:02 <beewoolie> As you have JTAG, I'm not too worried.  :-)
Jun 14 15:59:22 <[g2]> well JTAG hasn't been verified on this unit
Jun 14 15:59:33 <beewoolie> Hmm.
Jun 14 15:59:37 <beewoolie> Is this the avila?
Jun 14 15:59:47 <[g2]> no this is on the fatsly
Jun 14 15:59:50 <[g2]> fatslug
Jun 14 16:00:04 <beewoolie> is there a way to try the latest APEX on the avila?
Jun 14 16:00:17 <[g2]> Although I got the UBER-DOOBER fat slug on loan today
Jun 14 16:00:42 <[g2]> I'll have it for 3-4 weeks
Jun 14 16:00:43 <beewoolie> What does that mean?
Jun 14 16:00:49 <beewoolie> RAM and flash upgrades?
Jun 14 16:01:07 <[g2]> IXP425 Intel Dev board 256MB 425 @ 533 I think
Jun 14 16:01:15 <[g2]> I know its got 256MB
Jun 14 16:01:40 <beewoolie> Nice.
Jun 14 16:01:49 <[g2]> and one of my most favorite pieces of hw
Jun 14 16:01:51 <beewoolie> Seems like testing JTAG would be a good idea.
Jun 14 16:02:08 <[g2]> on the fat slug or the 425 ?
Jun 14 16:03:00 <[g2]> it's got the little plastic coffin that you can switch the flash parts in and out
Jun 14 16:03:11 <beewoolie> I like those.
Jun 14 16:03:21 <beewoolie> I think those sockets are $40 or so.
Jun 14 16:03:26 <[g2]> that's "one of my favorite hw parts"
Jun 14 16:03:47 <[g2]> no if I only had a flash programmer I'd really be set :)
Jun 14 16:04:28 <[g2]> come to think of it those would be super handy to have on the slug
Jun 14 16:04:34 <[g2]> and a programmer of course :)
Jun 14 16:05:15 <beewoolie> It would be slick for development, yeah.
Jun 14 16:06:31 <[g2]> the IXP425 kit or the little coffin ?
Jun 14 16:07:06 <beewoolie> Well, both, but a coffin'd slug could mean easier development.
Jun 14 16:07:27 <[g2]> if we had a flash programmer
Jun 14 16:07:46 <[g2]> I've got some 16MB samples it would be nice to just plug them in
Jun 14 16:09:53 <[g2]> those would be *way* cool for memory chips :)
Jun 14 16:15:22 <[g2]> beewoolie, you've got mail :)
Jun 14 16:15:25 <beewoolie> KO.
Jun 14 16:17:12 <[g2]> beewoolie, BTW I hope I'm not bugging you too much with this stuff!
Jun 14 16:17:17 <beewoolie> Nope.
Jun 14 16:17:29 <beewoolie> I'm doing several other things at the same time, so you'll have to be patient.
Jun 14 16:17:30 <[g2]> I hope I'm the helpful tester
Jun 14 16:18:19 <[g2]> NP, me too :)
Jun 14 16:18:25 <beewoolie> Yeah.  This output is helpfil.  The i -1 in the read is important.
Jun 14 16:18:55 <[g2]> diagnostic's are good.  It's great to be able to just turn some on
Jun 14 16:19:09 <[g2]> or even diagnostics
Jun 14 16:27:03 <beewoolie> Hmm.  I think I might have found the problem.
Jun 14 16:27:13 <[g2]> excellent
Jun 14 16:27:19 <[g2]> you're fast
Jun 14 16:27:25 <[g2]> what do you think it is ?
Jun 14 16:27:35 * [g2] hopes it's not the tester :)
Jun 14 16:27:56 <beewoolie> I changed one of the routines.  I optimized it.  And I think that that optimization doesn't work as I thought.
Jun 14 16:28:19 <[g2]> with that cache ?
Jun 14 16:28:19 <beewoolie> It seems to be that the jffs2 filesystem you sent has important differences from the one I have in flash.
Jun 14 16:28:42 <[g2]> how's that ?
Jun 14 16:28:44 <beewoolie> It's the find_cached_inode() function.  I changed it to do a complete binary search instead of a bsearch just on the inode number.
Jun 14 16:29:05 <beewoolie> I don't know what is different about your data, but it certainly helps with finding bugs.
Jun 14 16:29:55 <beewoolie> Oh yeah.  I think I found it.
Jun 14 16:30:01 <[g2]> COOL..
Jun 14 16:30:10 <[g2]> I told you beewoolie ROCKS
Jun 14 16:30:46 <[g2]> BTW did you see that article about the virtual ARM simulator ? it's supposed to run at 217Mhz on a 2-3G pentium
Jun 14 16:30:57 <beewoolie> That could be handy.
Jun 14 16:31:02 <beewoolie> Where was the article?
Jun 14 16:31:19 <[g2]> it was about a week or so ago... I can probably dig it up
Jun 14 16:31:48 <jacques> linked from linuxdevices
Jun 14 16:31:56 <jacques> vir-someting
Jun 14 16:32:15 <[g2]> http://www.linuxdevices.com/news/NS4665550218.html
Jun 14 16:32:25 <[g2]> that's where I thought I saw it
Jun 14 16:32:33 <jacques> http://www.virtera.com/
Jun 14 16:32:36 <[g2]> they were running OPIE in a simulator
Jun 14 16:33:06 <jacques> no BE tho
Jun 14 16:33:17 <[g2]> yeah 217Mhz on a 2.6 Pentium
Jun 14 16:33:26 <[g2]> well we can test in LE right :)
Jun 14 16:33:36 <[g2]> it'll work for jacmet 
Jun 14 16:33:44 <beewoolie> [g2]: I can now read the /boot directory.
Jun 14 16:33:57 <[g2]> go beewoolie go beewoolie 
Jun 14 16:34:01 <beewoolie> Don't quite know why I didn't catch that one myself.
Jun 14 16:34:09 <beewoolie> I need to do some testing here before I can call it OK.
Jun 14 16:34:27 <[g2]> OK...
Jun 14 16:34:43 <[g2]> I'm gonna have to try building my 425 kernel sometime
Jun 14 16:38:47 <beewoolie> bugs bugs bugs
Jun 14 16:39:29 <[g2]> I'm guessing you're squashing bugs bugs bugs
Jun 14 16:39:56 <beewoolie> ...finding...and then SMASHING!  Beewoolie SMASH!
Jun 14 16:40:19 <jacques> lol
Jun 14 17:03:35 <beewoolie> [g2]: OK man.  Try 1.2.15.  I fixed a whole mess of bugs in the way it handles extended regions. e.g. jffs2:/boot/zImage@10 and such.
Jun 14 17:03:41 <beewoolie> I'm not sure there aren't more problems.
Jun 14 17:10:06 <[g2]> beewoolie, EXCELLENT
Jun 14 17:10:48 <[g2]> so the copy will be copy jffs2:/boot/bzImage+0x.... 0x1d00000 ?
Jun 14 17:11:01 <beewoolie> Wait a second.
Jun 14 17:12:10 <beewoolie> apex> copy j:/boot/bzImage 0x8000
Jun 14 17:12:17 <beewoolie> That will put the kernel in the right place.
Jun 14 17:12:38 <beewoolie> You can put an offset on it if there is a 16 byte header.
Jun 14 17:12:39 <[g2]> OK, I *love* that syntax
Jun 14 17:12:56 <beewoolie> apex> copy j:/boot/zImage@16 0x8000
Jun 14 17:13:10 <[g2]> ok so the @ is +16 decimal
Jun 14 17:13:23 <beewoolie> yeah.  The form is @start+length
Jun 14 17:13:27 <[g2]> and it pulls the lenght from the fs as one would hope 
Jun 14 17:13:54 <[g2]> I'l pull and build .15 now
Jun 14 17:14:31 <beewoolie> There are still some errors in the handling of the ranges.  I'm revising as we speak.  It shouldn't break, tho.
Jun 14 17:15:36 <[g2]> ok
Jun 14 17:19:35 <[g2]> HEH.. there we go
Jun 14 17:20:06 <[g2]> I knew I wrote down the recipie somewhere lately
Jun 14 17:23:21 <beewoolie> [g2]: Did it work?
Jun 14 17:23:34 <[g2]> I'm rebuilding now
Jun 14 17:23:47 <[g2]> I had my recipe written down...
Jun 14 17:23:51 <[g2]> I was looking for it 
Jun 14 17:24:25 <[g2]> today when I went to meet the Intel Rep, I wrote his cell # down and tore off the paper and put it in my pocket :)
Jun 14 17:24:58 <[g2]> it took me a couple minutes to realize that's the piece of paper that has all the APEX notes on it :)
Jun 14 17:25:03 <beewoolie> One of my collegues at Intel asked if I knew anyone interested in working in Shanghai for a couple of years.
Jun 14 17:25:27 <beewoolie> ...that was sort of a nonsequitir...
Jun 14 17:25:53 <[g2]> there's a bunch of development going international
Jun 14 17:29:56 <[g2]> ok... let's try this agin
Jun 14 17:29:59 <[g2]> ok... let's try this again
Jun 14 17:30:03 <[g2]> I'm in 2.14 :)
Jun 14 17:30:58 <beewoolie> It isn't exactly the development going international.  Well, maybe.  They've got fabs in China and they want a Linux dude to make it all go smoothly.  Probably do some kernel specific things for them.
Jun 14 17:31:20 <beewoolie> [g2]: BTW, you unearthed a small warren of buglets.
Jun 14 17:32:56 <[g2]> cool
Jun 14 17:33:05 <[g2]> and you probably fixed them all :)
Jun 14 17:33:17 <beewoolie> Still discovering them...
Jun 14 17:34:03 <beewoolie> After I discovered that the bounded region IO for files was porked in jffs2, I looked in ext2 and found similar problems.
Jun 14 17:34:26 <[g2]> APEX just gets stronger
Jun 14 17:35:32 <[g2]> beewoolie, I should probably turn on TALK just in case eh ?
Jun 14 17:35:57 <beewoolie> Were you successful in running from RAM?
Jun 14 17:36:07 <[g2]> not last time
Jun 14 17:36:24 <beewoolie> I'd try it first without TALK.
Jun 14 17:41:40 <[g2]> DOH!
Jun 14 17:41:51 <[g2]> copy j:/foo 0x1000000
Jun 14 17:41:51 <[g2]> 0 bytes transferred
Jun 14 17:42:12 <[g2]> not a really a big DOH!
Jun 14 17:42:26 <[g2]> info j:/
Jun 14 17:42:26 <[g2]> caching jffs2 filesystem: 838 directory nodes 2972 inodes nodes
Jun 14 17:42:26 <[g2]> bin/
Jun 14 17:42:58 <beewoolie> That's it?
Jun 14 17:43:08 <beewoolie> It only shows bin/ in the root?
Jun 14 17:43:14 <[g2]> for the directory it show all
Jun 14 17:43:21 <beewoolie> What about info j:/foo
Jun 14 17:43:40 <[g2]> info j:/foo
Jun 14 17:43:40 <[g2]> bin/
Jun 14 17:43:40 <[g2]> boot/
Jun 14 17:43:40 <[g2]> dev/
Jun 14 17:43:40 <[g2]> etc/
Jun 14 17:43:42 <[g2]> home/
Jun 14 17:43:44 <[g2]> initrd/
Jun 14 17:43:46 <[g2]> lib/
Jun 14 17:43:48 <[g2]> media/
Jun 14 17:43:50 <[g2]> mnt/
Jun 14 17:43:52 <[g2]> proc/
Jun 14 17:43:54 <[g2]> sbin/
Jun 14 17:43:56 <[g2]> sys/
Jun 14 17:43:58 <[g2]> usr/
Jun 14 17:44:00 <[g2]> var/
Jun 14 17:44:02 <[g2]> tmp@
Jun 14 17:44:04 <[g2]> linuxrc.sav
Jun 14 17:44:08 <[g2]> linuxrc
Jun 14 17:44:10 <[g2]> foo
Jun 14 17:44:12 <[g2]> APEX Boot Loader 1.2.15 -- Copyright (c) 2004,2005 Marc Singer
Jun 14 17:44:14 <beewoolie> OK.  It's definitely not finding things in the inode cache.
Jun 14 17:44:14 * [g2] turns on TALK
Jun 14 17:48:47 <[g2]> info j:/foo
Jun 14 17:48:47 <[g2]> jffs2_identify
Jun 14 17:48:47 <[g2]> jffs2_path_to_inode
Jun 14 17:48:47 <[g2]> jffs2_path_to_inode: starting with inode 1
Jun 14 17:48:47 <[g2]> find_cached_parent_inode
Jun 14 17:48:48 <[g2]> bin/
Jun 14 17:51:40 <[g2]> beewoolie, I just sent you an e-mail with TALK enabled for 1.2.15 and the commands
Jun 14 17:51:57 <[g2]> it doesn't appear to generate the -1 any more.
Jun 14 17:52:06 <[g2]> It looks like it's just picking the first node
Jun 14 17:52:08 <[g2]> inode 1
Jun 14 17:53:57 <beewoolie> Yeah.
Jun 14 17:57:27 <[g2]> are you going to be around in an hour ?
Jun 14 17:58:10 <beewoolie> Perhaps.  I've got a cold and am feeling like I'll be lying down soon.
Jun 14 17:58:26 <beewoolie> kas11: hey there.
Jun 14 17:58:40 <kas11> hiya beewoolie
Jun 14 17:59:20 <rwhitby-away> hey kas11!
Jun 14 17:59:37 <kas11> hey rwhitby-away
Jun 14 18:00:54 <rwhitby-away> kas11: what are you up to these days, slug-wise?
Jun 14 18:02:05 <kas11> rwhitby-away nada slugwise...mostly learning rpm packaging these days
Jun 14 18:05:48 <[g2]> hey kas11 !
Jun 14 18:06:26 <kas11> how goes it, [g2] ?
Jun 14 18:06:34 <beewoolie> [g2]: I've got an intuition...
Jun 14 18:06:48 <[g2]> kas11, pretty good thx
Jun 14 18:06:54 <[g2]> beewoolie, What's that ?
Jun 14 18:06:59 <beewoolie> ...
Jun 14 18:07:09 <[g2]> drum roll....
Jun 14 18:07:22 <beewoolie> let me prognosticate.  Your trouble is due to a config issue.
Jun 14 18:07:28 <beewoolie> The envelope please...
Jun 14 18:07:47 <[g2]> nor:0x180000+0x660000 ?
Jun 14 18:07:58 <beewoolie> ...hang on...
Jun 14 18:08:14 * [g2] waits with baited breath
Jun 14 18:08:31 <beewoolie> Are either of the filesystem types enabled: FAT or EXT2?
Jun 14 18:08:42 <[g2]> nope
Jun 14 18:09:03 <beewoolie> That's the problem.  Just for grins, enable EXT2 and try it.  I'll fix the source so this isn't necessary.
Jun 14 18:09:14 <[g2]> that's easy
Jun 14 18:09:17 <[g2]> will do
Jun 14 18:09:48 <beewoolie> What's happening is that the region parser doesn't include the filename parsing code because it is hard coded to either fat or ext2.  I'm going to make it be included more reliably.
Jun 14 18:09:53 <[g2]> do I need any underlying region ?
Jun 14 18:09:58 <beewoolie> No.  
Jun 14 18:10:09 <[g2]> ok
Jun 14 18:10:16 <beewoolie> It's just to make sure the parser includes the pathname parsing code.
Jun 14 18:11:32 <[g2]> if this work you'll *definitely* have to prognosticate more often!
Jun 14 18:12:04 <beewoolie> The clue was in the  TALK output.  There just wasn't enough of it.
Jun 14 18:12:16 <beewoolie> And I still found a whole mess of bugs!
Jun 14 18:15:11 <[g2]> this is look'n better :)
Jun 14 18:16:12 <[g2]> Hey! I just loaded my first JFFS2 file into memory with APEX
Jun 14 18:16:17 <[g2]> congrats dude !
Jun 14 18:16:20 <beewoolie> Yipee!
Jun 14 18:16:30 <[g2]>  dump 0x1000000
Jun 14 18:16:30 <[g2]> 01000000: 01 02 03 04 05 06 07 08  09 0a 0b 0c 0d 0e 0f 10  ........ ........
Jun 14 18:16:30 <[g2]> 01000010: 11 12 13 14 15 16 17 18  19 1a 1b 1c 1d 1e 1f 20  ........ .......
Jun 14 18:16:30 <[g2]> 01000020: 21 22 23 0a 27 b8 1b 17  27 ba 1b 17 27 b8 1b 17  !"#.'... '
Jun 14 18:16:39 <beewoolie> Is that right?
Jun 14 18:17:02 <beewoolie> you should be able to do this, too
Jun 14 18:17:06 <[g2]> copy j:/foo 0x1000000
Jun 14 18:17:08 <beewoolie> apex> du j:/foo
Jun 14 18:18:07 <[g2]> beewoolie, NOD : same stuff
Jun 14 18:18:11 <dyoung-web> Beewoolie-Smash!
Jun 14 18:18:17 <[g2]> I'm just checking the lenght
Jun 14 18:18:18 <dyoung-web> Attack  400
Jun 14 18:18:25 <dyoung-web> Plus 400.
Jun 14 18:18:57 * beewoolie slices the gelatinous cube into  bite-sized pieces.
Jun 14 18:19:36 <[g2]>  ls -l foo
Jun 14 18:19:36 <[g2]> -rw-r--r--  1 tom users 36 Jun 14 07:14 foo
Jun 14 18:19:36 <[g2]> bash-2.05b$ hexdump foo
Jun 14 18:19:36 <[g2]> 0000000 0201 0403 0605 0807 0a09 0c0b 0e0d 100f
Jun 14 18:19:38 <[g2]> 0000010 1211 1413 1615 1817 1a19 1c1b 1e1d 201f
Jun 14 18:19:42 <[g2]> 0000020 2221 0a23
Jun 14 18:19:44 <[g2]> 0000024
Jun 14 18:19:46 <[g2]> that's from the host side
Jun 14 18:20:25 <[g2]> hold on one minute
Jun 14 18:22:02 <[g2]>  od -t x1 foo
Jun 14 18:22:02 <[g2]> 0000000 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f 10
Jun 14 18:22:02 <[g2]> 0000020 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20
Jun 14 18:22:02 <[g2]> 0000040 21 22 23 0a
Jun 14 18:22:02 <[g2]> 0000044
Jun 14 18:22:13 <[g2]> that's better
Jun 14 18:22:29 <[g2]> no endianness filtering going on
Jun 14 18:23:02 <beewoolie> The length calc stuff seems to be correct now, too.
Jun 14 18:23:08 <[g2]> nod
Jun 14 18:23:15 <beewoolie> I just did some testing to make sure we don't read past the end of a file.
Jun 14 18:23:16 <dyoung-web> Beewoolie, I'm halfway interested in finding more information about your Intel Shanghai bit.  It must be at ICSL; lots of stuff going on there.
Jun 14 18:23:19 <[g2]> do you think I can load the kernel ?
Jun 14 18:23:30 <[g2]> or is the TALK stuff too much for a 1MB file ?
Jun 14 18:23:32 <beewoolie> [g2]: try it.
Jun 14 18:24:05 <beewoolie> dyoung-web: I sent some questions to the dude yesterday.  Do you want his email?  I'm sure he wouldn't mind as long as you're genuinely interested.
Jun 14 18:24:28 <beewoolie> dyoung-web: I suppose I should ask him first, tho.
Jun 14 18:24:30 <[g2]> hey I can see in /boot now :)
Jun 14 18:24:44 <beewoolie> Duhn, daaa!
Jun 14 18:25:03 <dyoung-web> beewoolie: I'm more curious at this point.
Jun 14 18:25:15 <beewoolie> What details would interest you to know.
Jun 14 18:25:20 <beewoolie>  ?
Jun 14 18:25:30 <[g2]> copy j:/boot/zImage 0x8000
Jun 14 18:25:34 <[g2]> that's the one ?
Jun 14 18:25:49 <beewoolie> Should be.
Jun 14 18:25:56 <[g2]> then boot ?
Jun 14 18:25:58 <beewoolie> Depends on the name of the source file.
Jun 14 18:26:07 <[g2]> that's the name of the symlink
Jun 14 18:26:17 <beewoolie> The destination address can be verified because it is in the environment.
Jun 14 18:26:20 <beewoolie> apex> print
Jun 14 18:26:25 <beewoolie> Look for bootaddr.
Jun 14 18:26:30 <[g2]> nod
Jun 14 18:26:33 <[g2]> verified
Jun 14 18:27:30 <dyoung-web> Contractor or Employee?  Scope of work?  Location in Shanghai?  (ALthough its gotta be Pudong) that sort of thing.
Jun 14 18:27:51 <dyoung-web> Unfortunately, but mandarin skills suck.
Jun 14 18:27:55 <[g2]> ffs2_read: d->index 880124  d->length 880124
Jun 14 18:27:55 <[g2]> 880124 bytes transferred
Jun 14 18:27:55 <[g2]> jffs2_close
Jun 14 18:30:18 <beewoolie> dyoung-web: I asked about the status of the position.  He said Shanghai.
Jun 14 18:30:27 <beewoolie> [g2]: does it boot.
Jun 14 18:30:36 <[g2]> uname -a
Jun 14 18:30:36 <[g2]> Linux LKG7CC52D 2.6.11.2 #1 Mon Jun 6 21:39:19 EDT 2005 armv5teb unknown unknown GNU/Linux
Jun 14 18:30:46 <beewoolie> That looks like a yes.
Jun 14 18:30:54 <[g2]> I'd like to try one more test
Jun 14 18:31:03 <beewoolie> Did you disable TALK?
Jun 14 18:31:12 <[g2]> no.. that's with TALK enabled
Jun 14 18:31:30 <beewoolie> I'm going to release 1.2.16...
Jun 14 18:31:49 <beewoolie> OK.  1.2.16 is up.
Jun 14 18:32:06 <beewoolie> You don't need to enable the ext2 driver anymore.
Jun 14 18:32:36 <[g2]> doh.. I'm missed the timeout gotta reboot again
Jun 14 18:32:57 <[g2]> THIS is JUST SO DAMN COOL! :)
Jun 14 18:33:09 <beewoolie> You're telling me?
Jun 14 18:33:26 <beewoolie> I'm adding aliases at the moment.
Jun 14 18:34:35 <[g2]> so fill 0x8000+0x100000 0 is the cmd I'm looking for right ?
Jun 14 18:34:47 <beewoolie> Close.
Jun 14 18:34:53 <beewoolie> Try apex> help fill
Jun 14 18:35:12 <[g2]> ok that works :)
Jun 14 18:35:19 <beewoolie> apex> fi 0 0x8000+2m
Jun 14 18:36:08 <[g2]> ok... it's cleared
Jun 14 18:36:14 <[g2]> I did 1m
Jun 14 18:36:21 <dyoung-web> I feel like I'm missing out.
Jun 14 18:36:32 <beewoolie> :-)
Jun 14 18:36:53 <[g2]> copy j:/boot/bzImage 0x8000
Jun 14 18:36:54 <beewoolie> Sounds like dyoung as FMS.
Jun 14 18:37:02 <[g2]> FMS ?
Jun 14 18:37:04 <beewoolie> s/as/has/
Jun 14 18:37:10 <beewoolie> Want to try to guess?
Jun 14 18:37:16 <dyoung-web> Full Metal ...
Jun 14 18:37:18 <[g2]> Firmware missing syndrome
Jun 14 18:37:25 <dyoung-web> some kine of syndrome.
Jun 14 18:37:28 <beewoolie> Getting warmer.
Jun 14 18:37:46 <dyoung-web> Mental...
Jun 14 18:37:53 <beewoolie> lol
Jun 14 18:38:18 <dyoung-web> Firmware Mental something....
Jun 14 18:38:24 <[g2]> ok it's zImage not bzImage
Jun 14 18:39:26 <dyoung-web> fading memory syndrome?
Jun 14 18:39:46 <beewoolie> hint: it isn't specific to what we're working on.  Lots of people have it.  It can be recognized by hearing people say things like: "What was that?  What's going on?"
Jun 14 18:39:49 <[g2]> OK.. Clearly... APEX is either a GREAT story teller or booting well from the kernel inside the jffs2 *with* the symlink I might add
Jun 14 18:39:50 <beewoolie> lol
Jun 14 18:40:07 <beewoolie> All in less than 64k
Jun 14 18:40:21 <dyoung-web> sounds like selective hearing.
Jun 14 18:40:25 <[g2]> I first filled the boot area with 0's do make sure it wasn't just lying around in memory
Jun 14 18:40:31 <beewoolie> BTW, I'm adding aliases on the way to supporting chained drivers.
Jun 14 18:40:39 <[g2]> then loaded and booted from with the jffs2 partition
Jun 14 18:41:13 <dyoung-web> Fatigued m... syndrome?
Jun 14 18:41:21 <beewoolie> Let me know when you give up.
Jun 14 18:41:53 <dyoung-web> grr, okay give!
Jun 14 18:42:15 <beewoolie> Fear of Missing Something
Jun 14 18:42:32 <dyoung-web> Oh.  Yeah, I have that every now and then.
Jun 14 18:42:38 <beewoolie> lol
Jun 14 18:42:44 <dyoung-web> hehe and you saw me have it 3 times today.
Jun 14 18:42:48 <beewoolie> Most of us do.
Jun 14 18:43:20 <dyoung-web> I want to be onboard the Apex Train.
Jun 14 18:43:29 <dyoung-web> But then I have itchy feet.
Jun 14 18:43:35 <dyoung-web> but guessing acronyms is fun.
Jun 14 18:43:50 <beewoolie> There's no fee to board.
Jun 14 18:44:09 <dyoung-web> But there might be.
Jun 14 18:44:10 <beewoolie> It's just that [g2] is motivated to try out the new features.
Jun 14 18:44:26 <dyoung-web> The JTAG Toll may need to be paid at some point.
Jun 14 18:44:32 <beewoolie> And he's a darn good tester.  He breaks something almost every time.
Jun 14 18:44:36 <dyoung-web> lucky i prepaid it.
Jun 14 18:45:03 <[g2]> it's really fun working together with such ASS-KICKERs
Jun 14 18:45:26 <dyoung-web> so does the new jffs2 driver read only?
Jun 14 18:45:36 <beewoolie> I'm adding aliases so that it's easier to write apex to flash.
Jun 14 18:45:44 <beewoolie> apex> copy $apex nor:0
Jun 14 18:46:04 <[g2]> this is a big mileston
Jun 14 18:46:05 <[g2]> this is a big milestone
Jun 14 18:46:15 <beewoolie> I had a client corrupt flash because he didn't read the instructions.  He just pasted in my instructions.
Jun 14 18:46:16 <[g2]> actually
Jun 14 18:46:21 <[g2]> this is a *big* milestone
Jun 14 18:46:24 <beewoolie> I agree.
Jun 14 18:47:03 <[g2]> it means we can go to dual full rootfs's and free up a 1MB partition
Jun 14 18:47:08 <beewoolie> I think that this means we can use it.
Jun 14 18:47:17 <dyoung-web> YAY!
Jun 14 18:47:22 <dyoung-web> ~praise beewoolie 
Jun 14 18:47:23 <purl> All hail beewoolie!
Jun 14 18:47:26 <dyoung-web> ~praise [g2]
Jun 14 18:47:28 <purl> All hail [g2]!
Jun 14 18:47:29 * beewoolie celebrates
Jun 14 18:47:33 <[g2]> ~praise beewoolie 
Jun 14 18:47:34 <purl> All hail beewoolie!
Jun 14 18:47:51 * beewoolie throws flowers on the crowd
Jun 14 18:47:57 <dyoung-web> beewoolie: best I can guess is your job is here: 
Jun 14 18:47:59 <[g2]> LOL
Jun 14 18:48:00 <dyoung-web> http://www.intel.com/jobs/china/sites/shanghai.htm
Jun 14 18:48:58 <beewoolie> I wouldn't mind seeing the wall.
Jun 14 18:49:10 <beewoolie> Dollface thinks it would be an fine adventure
Jun 14 18:49:14 <dyoung-web> I climbed quite a ways up it.
Jun 14 18:49:36 <dyoung-web> near beijing.  Basically until it ended. :-)
Jun 14 18:49:50 <[g2]> it'd probably be fun to bike ride on
Jun 14 18:50:05 <dyoung-web> Right up until those guys with guns start chasing you.
Jun 14 18:50:22 <[g2]> ok bike ride fast :)
Jun 14 18:50:23 <beewoolie> Really?
Jun 14 18:50:28 <beewoolie> Guns?
Jun 14 18:50:40 <beewoolie> Is it sacred?
Jun 14 18:50:49 <dyoung-web> The military presence is fairly immense.
Jun 14 18:50:56 <dyoung-web> all over the place.
Jun 14 18:51:30 <dyoung-web> [g2] actually it would be pretty technical ride because the height of the stairs isnt uniform.  They did that to make it difficult for horses to ascend.
Jun 14 18:52:08 <[g2]> clearly not tough enough for the mongols right ?
Jun 14 18:52:21 <[g2]> or was it the Huns ?
Jun 14 18:53:18 <dyoung-web> Thats Mr Khan to the rest of is.
Jun 14 18:53:31 <[g2]> it's interesting because I think we are attacking the modern day wall
Jun 14 18:53:54 <[g2]> the last two are the bootloader/JTAG  and building HW walls
Jun 14 18:54:20 <dyoung-web> So which of you is Kublai and which is Genghis?
Jun 14 18:54:34 * [g2] is the water boy
Jun 14 18:54:37 <dyoung-web> whats the 3rd brothers name?
Jun 14 18:54:55 <beewoolie> Shemp
Jun 14 18:55:15 <beewoolie> Oh.  He's the fourth.
Jun 14 18:55:23 <beewoolie> Curley.
Jun 14 18:55:38 <dyoung-web> Hey, isht that a differnt brother??
Jun 14 18:55:42 <dyoung-web> yeah, there you go.
Jun 14 18:56:47 <dyoung-web> Togril?
Jun 14 18:56:55 <dyoung-web> something like that...
Jun 14 18:56:58 <dyoung-web> I forget.
Jun 14 18:57:22 <[g2]> beewoolie, I'll be taking a look at the PCI and USB interface
Jun 14 18:57:24 <dyoung-web> he was the lesser brpother anways.
Jun 14 18:57:59 <[g2]> it'll be on simmer on the back burner
Jun 14 18:58:09 * kas11 waves @ ka6sox-away ... y'all be good
Jun 14 18:58:26 <beewoolie> [g2]: I thought you might.
Jun 14 18:58:42 <[g2]> I'd like to have a usb-serial console and booting to EXT2 off disk
Jun 14 18:58:42 <beewoolie> He only killed in his spare time?
Jun 14 18:59:04 <beewoolie> Will the madness ever stop?!
Jun 14 18:59:35 <[g2]> what madness ?
Jun 14 19:00:24 <[g2]> maybe... serial is good for now
Jun 14 19:00:58 <[g2]> it's a cheap mod and I think we can get a $12.95 part form digilent that does the level shifting
Jun 14 19:02:48 <dyoung-web> for the usb serial?
Jun 14 19:03:00 <[g2]> no for actual serial
Jun 14 19:03:17 <[g2]> they've got a 6 pin adapter for the cerebot board
Jun 14 19:03:25 <beewoolie> Can't you use a cheaper part for level shifting?
Jun 14 19:04:41 <dyoung-web> I think [g2] is looking at already made stuff.
Jun 14 19:04:55 <dyoung-web> the part is dirt cheap.
Jun 14 19:05:19 <[g2]> I can't make everyting to start with :)
Jun 14 19:05:47 <[g2]> what do you think that $12 part can be made for ?
Jun 14 19:05:59 <dyoung-web> I think the last batch of MAX3232's I bought were $3/ea/
Jun 14 19:06:15 <dyoung-web> Plus the 4 0.1mf caps.
Jun 14 19:06:50 <dyoung-web> the circuit is easy.  but I dunno, I'm sure theres a product that could work out of the box.
Jun 14 19:07:44 <[g2]> I mean if I can make it for $6 then it'd probably be worth it
Jun 14 19:08:41 <dyoung-web> that PMOD-RS232 looks nice.
Jun 14 19:08:52 <[g2]> that's the one
Jun 14 19:08:59 <[g2]> $12.95 right ?
Jun 14 19:09:33 <beewoolie> I see what you mean.  As long as the $13 doesn't require soldering, it's a good deal.
Jun 14 19:09:57 <dyoung-web> Well, the board is going to be around 1.5 square inches, so the board cost should be pretty low if we make 100 of them.
Jun 14 19:10:12 <dyoung-web> but I would rather spend $13.
Jun 14 19:10:23 <dyoung-web> or $20 for the compsys kit and get the hood. :-)
Jun 14 19:10:37 <dyoung-web> (a hood and nice solder pads)
Jun 14 19:10:48 <[g2]> well the hood is cheap right ?
Jun 14 19:10:54 <dyoung-web> Ah, but one could make a custom cable for this PMOD thing.
Jun 14 19:11:04 <dyoung-web> so I'll flip back to the PMOD.
Jun 14 19:11:25 <dyoung-web> The best deal is still cellphone cables when they go on closeout at radiocrap.
Jun 14 19:11:30 <[g2]> right the modded slug could have a 4 to 6 pin cable coming out
Jun 14 19:11:39 <beewoolie> I haven't been able to find any of those.
Jun 14 19:11:47 <beewoolie> cellphone cables that is.
Jun 14 19:11:56 <dyoung-web> Theres one more store I didnt hit yet.
Jun 14 19:12:08 <dyoung-web> I think I have 6 more of those cables if you want one.. speak up.
Jun 14 19:12:25 <beewoolie> I have one and that's all I need right now.
Jun 14 19:12:52 <beewoolie> I'm going to need some way to handle USB to USB.  Did we ever determine if that would work?
Jun 14 19:13:07 <[g2]> I've got a $19 cable
Jun 14 19:13:21 <[g2]> it does host-to-host and runs between the slug and pc
Jun 14 19:13:32 <[g2]> looks like an ETH2 port
Jun 14 19:13:38 <beewoolie> Hmm.
Jun 14 19:13:54 <[g2]> the mechanicals are shit... but it works for testing
Jun 14 19:13:54 <beewoolie> I'd like to be able to use it for the console.
Jun 14 19:14:24 <[g2]> for a console... I'd use a PIC with usb and drive a LCD display
Jun 14 19:14:41 <beewoolie> I mean, I'd like to use it like a serial console, silly.
Jun 14 19:14:53 <[g2]> you could with a usb keyboard :)
Jun 14 19:15:03 <beewoolie> Plug the USB port of the slug into the notebook computer.  Fire up minicom and go
Jun 14 19:15:20 <[g2]> Um.. I run one of my slugs like that now
Jun 14 19:15:27 <beewoolie> We talked about this and I don't remember if it worked.
Jun 14 19:15:38 <beewoolie> You do?  without async serial conversion?
Jun 14 19:16:03 <[g2]> ok... I've got PC-USB-Serial to Serial on the Slug
Jun 14 19:16:06 <beewoolie> I could imagine doing it with two async converters and a null modem.
Jun 14 19:16:30 <beewoolie> right.  I'm looking for USB to USB.  No serial involved.
Jun 14 19:16:42 <[g2]> I did that with the host-to-host
Jun 14 19:16:55 <[g2]> Slug-USB-USB-PC
Jun 14 19:16:59 <beewoolie> Right.
Jun 14 19:17:03 <[g2]> with the network connection setup
Jun 14 19:17:04 <beewoolie> special cable?
Jun 14 19:17:13 <[g2]> just the $19 host-to-host cable
Jun 14 19:17:41 <beewoolie> [g2]: Just checking the jtag schematic now.
Jun 14 19:17:45 <beewoolie> Hmm.
Jun 14 19:17:49 <beewoolie> I'll look for it.
Jun 14 19:18:25 <[g2]> the JTAG FTDI is a USB device
Jun 14 19:18:40 <[g2]> the host to host acts as 2 devices and bridges
Jun 14 19:18:56 <[g2]> but I don't really like the cable that much.
Jun 14 19:19:29 <[g2]> 50% of the time when I plugged it into the P4 it literally crashed the P4 from the USB port
Jun 14 19:19:36 <beewoolie> Hmm.
Jun 14 19:19:46 <beewoolie> Sounds like a SW problem.
Jun 14 19:19:52 <[g2]> I think HW
Jun 14 19:20:01 <[g2]> spike on the USB
Jun 14 19:20:22 <[g2]> I guess I could disable USB and see
Jun 14 19:20:53 <[g2]> but actually just touching the USB caused a little spark without plugging it in
Jun 14 19:21:13 <[g2]> it was winter but I didn't like that
Jun 14 19:21:16 <beewoolie> Oh.  That's bad.
Jun 14 19:21:22 <[g2]> nod.
Jun 14 19:21:40 <[g2]> but what do you want for $19 ?
Jun 14 19:21:44 <beewoolie> Gotta go.  Dollface made me some chicken soup.
Jun 14 19:21:52 <beewoolie> ttfn
Jun 14 19:21:55 <[g2]> be well my friend
Jun 14 19:21:57 <[g2]> THX
Jun 14 19:40:22 <dyoung-web> Hes the man.
Jun 14 19:41:22 <[g2]> the MAN!
Jun 14 19:43:20 <dyoung-web> I want to be like Beewoolie when I grow up.
Jun 14 19:43:38 <dyoung-web> Then I can earn the power of Beewoolie-SMASH for plus 400 attack. :-)
**** ENDING LOGGING AT Tue Jun 14 23:59:56 2005