**** BEGIN LOGGING AT Mon Feb 27 10:59:57 2006
Feb 27 12:57:34 <prpplague> ~seen beewoolie-afk
Feb 27 12:57:40 <purl> beewoolie-afk <n=beewooli@206.124.142.26> was last seen on IRC in channel #openjtag, 5d 7h 27m 11s ago, saying: 'On one device I've looked at, the largest packet is 128 bytes.'.
Feb 27 12:58:33 * prpplague wonders if beewoolie got a job
Feb 27 17:15:09 <beewoolie-afk> lennert: hello
Feb 27 17:15:45 <vmaster> hey beewoolie-afk 
Feb 27 17:15:54 <beewoolie-afk> Hey
Feb 27 17:16:26 <beewoolie-afk> I'm really close to having something releasable.
Feb 27 17:16:48 <beewoolie-afk> Another sad discovery is that some of the chips I'm using don't have boundary scan chains.
Feb 27 17:17:01 <beewoolie-afk> Only Armv4 stuff such as the embedded ice chain.
Feb 27 17:17:02 <vmaster> mostly -s cores, i guess
Feb 27 17:17:20 <beewoolie-afk> I don't think so,  but I'm not sure.
Feb 27 17:17:27 <beewoolie-afk> Isn't S the synthesizable core?
Feb 27 17:17:30 <vmaster> yes
Feb 27 17:17:44 <beewoolie-afk> We talked about this before and I thought you said that these weren't -s cores.
Feb 27 17:17:49 <beewoolie-afk> The Sharp LH parts.
Feb 27 17:17:51 <vmaster> ah, ok
Feb 27 17:17:53 <vmaster> no, they aren't -s
Feb 27 17:18:21 <beewoolie-afk> I think that one of the problems is that they goofed on their first implementation and didn't setup the external chain properly.
Feb 27 17:18:36 <beewoolie-afk> It latches outputs on the rising edge instead of the falling clock edge.
Feb 27 17:18:43 <vmaster> ah, yeah
Feb 27 17:18:59 <beewoolie-afk> Perhaps they dropped it altogether in the second implementation of the ARM7.
Feb 27 17:19:07 <beewoolie-afk> The ARM9's have external scan chains that work.
Feb 27 17:20:47 <vmaster> ok, seems boundary scan is left out randomly
Feb 27 17:21:05 <vmaster> the sam9261 (9-ejs) has a boundary scan interface
Feb 27 17:21:06 <prpplague> beewoolie-afk: hey hey
Feb 27 17:21:10 <prpplague> beewoolie-afk: whats cookin
Feb 27 17:21:24 <beewoolie-afk> prpplague: Hey.
Feb 27 17:21:36 <beewoolie-afk> prpplague:  I send mail about those drivers.
Feb 27 17:21:42 * prpplague looks
Feb 27 17:21:46 <beewoolie-afk> vmaster: I'm not following you.
Feb 27 17:22:24 <beewoolie-afk> Are you saying that other -S parts don't have boundary scan chains, but the samsung part does?
Feb 27 17:23:22 <vmaster> beewoolie-afk: no. i originally thought that boundary scan was missing at -s parts, because all i've looked at missed it - but as the at91sam9261 (-s core) has one, and the samsungs (not -s) don't, my assumptions proved to be wrong
Feb 27 17:24:13 <ka6sox> morning
Feb 27 17:24:56 <beewoolie-afk> ka6sox: 
Feb 27 17:25:01 <beewoolie-afk> hi
Feb 27 17:25:14 <beewoolie-afk> vmaster: I've gotten most of the pieces working.
Feb 27 17:25:34 <beewoolie-afk> The trouble is that there are so many holes in the hardware, it's been hard to make sure that everything is good.
Feb 27 17:25:50 <beewoolie-afk> I've been looking for a part that has both boundary scan and embeddedice.
Feb 27 17:26:04 <vmaster> i have one, the at91rm9200
Feb 27 17:26:11 <beewoolie-afk> *and* that works with the ftdi2232.
Feb 27 17:26:16 <vmaster> the at91rm9200
Feb 27 17:26:28 <beewoolie-afk> vmaster: :-)  That will help, eventually.
Feb 27 17:26:36 <prpplague> beewoolie-afk: did you get your sd/mmc driver working in apex?
Feb 27 17:26:44 <beewoolie-afk> prpplague: Nope.
Feb 27 17:26:49 <vmaster> i can run the ft2232 at the full 6mhz with the rm9200
Feb 27 17:26:51 <beewoolie-afk> prpplague: I haven't looked at it in a while.
Feb 27 17:27:01 <beewoolie-afk> vmaster: I haven't had speed problems. 
Feb 27 17:27:11 <prpplague> beewoolie-afk: ok, i wasn't sure if you still wanted my patch, but i'll get that ready to submit as well
Feb 27 17:27:36 <beewoolie-afk> vmaster: ATM, I have a part that isn't responding to a IDCODE scan with the ftdi2232.  It works with a parallel port dongle and I can use the same code on a different board.
Feb 27 17:28:11 <vmaster> beewoolie-afk: what chip?
Feb 27 17:28:18 <beewoolie-afk> prpplague: I do.  I've been working on the JTAG code for some time now.  I'm eager to get it to a working state before I start a new job.
Feb 27 17:28:42 <prpplague> beewoolie-afk: understood
Feb 27 17:28:49 <prpplague> beewoolie-afk: i hate leaving a project
Feb 27 17:28:58 <prpplague> beewoolie-afk: takes twice as long to get back into the game
Feb 27 17:28:58 <beewoolie-afk> vmaster: I've got two, the LH7A400 and LH7A40X.  I thought that these used to work, but I suspect that I broke something when I was working on the IXP42x code.
Feb 27 17:29:05 <beewoolie-afk> prpplague: Yeah.
Feb 27 17:29:31 <beewoolie-afk> Once this JTAG code is really useful, I'll release it and see if it works for other folks.
Feb 27 17:30:08 <prpplague> beewoolie-afk: cool beans
Feb 27 17:30:39 <beewoolie-afk> vmaster: I know that there is room for wrong code producing OK results on some targets.
Feb 27 17:31:32 <vmaster> beewoolie-afk: heh, yeah, had enough of this
Feb 27 17:31:33 <beewoolie-afk> vmaster: in my parallel port driver, I forgot to assert the TDI data on the low clock pulse.  It would assert on the rising pulse with the TMS signal and TCK.  One parallel cable worked fine with this. Another did not.
Feb 27 17:31:57 <beewoolie-afk> vmaster: Huh?
Feb 27 17:32:17 <vmaster> beewoolie-afk: my code worked with one device, but not with the other - mostly due to bugs in the code
Feb 27 17:32:37 <beewoolie-afk> That is our job, after all.
Feb 27 17:32:54 <lennert> beewoolie-afk: hello
Feb 27 17:32:59 <beewoolie-afk> lennert: Hey!
Feb 27 17:33:12 <vmaster> hey lennert 
Feb 27 17:33:40 <lennert> vmaster: hi :)
Feb 27 17:37:54 <prpplague> hehe, this channels just comes alive when the bee shows up
Feb 27 17:38:02 <lennert> hehe
Feb 27 17:38:05 <lennert> the Big Bee
Feb 27 17:38:12 <lennert> vmaster: you had a nice talk, btw
Feb 27 17:38:26 <lennert> vmaster: you're quite knowledgable about arm cores, for one
Feb 27 17:39:23 <vmaster> lennert: heh, thx. well, i hope it was interesting for some
Feb 27 17:39:35 <lennert> definitely interesting
Feb 27 17:39:47 <prpplague> vmaster: the openocd speech?
Feb 27 17:39:52 <vmaster> prpplague: yeah
Feb 27 17:40:03 <prpplague> vmaster: cool, sounds like it went over well then
Feb 27 17:40:47 <beewoolie-afk> vmaster: did you give a presentation?
Feb 27 17:41:14 <vmaster> beewoolie-afk: yeah, i presented the openocd and arm-debugging in general at the fosdem embedded softwares room
Feb 27 17:41:53 <lennert> the embedded room was disappointing except for one or two talks
Feb 27 17:42:06 <vmaster> heh, yeah, it was
Feb 27 17:42:09 <lennert> there were people from montavista teaching the audience how to best circumvent the gpl, etc.
Feb 27 17:42:19 <vmaster> i left after he said that
Feb 27 17:42:21 <beewoolie-afk> OUCH!
Feb 27 17:42:21 <lennert> btw, in case you didn't know yet, montavista wrote the linux kernel
Feb 27 17:42:30 <beewoolie-afk> OUCH!
Feb 27 17:42:35 <lennert> there's some guy in finland who contributed some bits, but montavista did most of the work
Feb 27 17:42:37 <beewoolie-afk> They are not nice people.
Feb 27 17:42:39 <lennert> oh, and by the way
Feb 27 17:42:46 <beewoolie-afk> I met them at a show here in Seattle.
Feb 27 17:42:58 <lennert> there are 25 million linux phones and 5 million linux PCs, so it's only fair that montavista get to decide what goes into the linux kernel
Feb 27 17:43:18 <lennert> so the next time daniel walker gets flamed on lkml, remember taht
Feb 27 17:43:49 <beewoolie-afk> Huh?
Feb 27 17:44:07 <lennert> daniel walker is some montavista dude that wants to push all kinds of rt crap into the kernel
Feb 27 17:44:38 <beewoolie-afk> OK.  I'm assuming that your tongue has poked a hole in your cheek.
Feb 27 17:44:46 <lennert> yeah
Feb 27 17:44:46 * p2-mate missed the MV realtime talk @fosdem
Feb 27 17:45:02 <lennert> p2-mate: you didn't miss anything
Feb 27 17:45:13 <p2-mate> otherwise I missed food :)
Feb 27 17:45:16 <lennert> hehe
Feb 27 17:45:19 <lennert> and nice beer
Feb 27 17:45:26 <p2-mate> that as well
Feb 27 17:45:29 <beewoolie-afk> vmaster: are there slides of your talk? 
Feb 27 17:45:33 <p2-mate> although I can always drink nice beer :)
Feb 27 17:45:41 <lennert> i ate about six tuna sandwiches on sunday and got really stuffed
Feb 27 17:45:58 <vmaster> beewoolie-afk: yeah, they're going to be at the fosdem site, but i can put them on a server, too
Feb 27 17:46:24 <lennert> the montavista dude's arguments were totally flawed
Feb 27 17:46:28 <beewoolie-afk> No hurry.
Feb 27 17:46:41 <lennert> like, linus has been wrong in the past, ergo, we can better leave the decisions to montavista
Feb 27 17:47:03 <p2-mate> lennert: this was on the rt talk ? not on the boot time thing ?
Feb 27 17:47:08 <beewoolie-afk> Fortunately, his confidence doesn't affect our ability to use Linux.
Feb 27 17:47:22 <lennert> he described some really crazy setups where the cpu had to service some peripheral ten gazillion times per second, where i figured that you might as well add a $5 cpld and avoid all the complexity
Feb 27 17:47:35 <lennert> p2-mate: i missed the boot time thing (by vitaly wool, right?)
Feb 27 17:48:30 <p2-mate> lennert: yes
Feb 27 17:48:32 <lennert> oh
Feb 27 17:48:57 <lennert> and he also made a snide remark that if you mess up the licensing, harald welte is gonna get you
Feb 27 17:49:08 <p2-mate> lennert: he isn't a very good speaker, but it was a reasonably nice talk
Feb 27 17:49:28 <lennert> i dropped in at the end of vitaly's talk, he was mostly staring at his laptop it seemed
Feb 27 17:50:23 <p2-mate> I liked the lock free data exchange talk
Feb 27 17:50:29 <lennert> yeah
Feb 27 17:50:39 <lennert> although it was kind of disappointing that it applied to UP only
Feb 27 17:51:03 <lennert> still, he mentioned some pretty nice properties (on UP) that i wasn't aware of
Feb 27 17:51:31 <p2-mate> yes. but we still have some time before embedded SMP becomes really mainstream to fix that :)
Feb 27 17:51:38 <lennert> hehe
Feb 27 17:52:02 <lennert> actually, at least one VM kernel patch didn't go into the arm tree because it didn't work on smp
Feb 27 17:52:11 <prpplague> yea MV is struggling to find a place for themselves these days
Feb 27 17:52:26 <p2-mate> arm11 SMP does seem to exist :)
Feb 27 17:52:35 <lennert> rmk has a bunch of arm smp boxes, i think
Feb 27 17:52:52 <lennert> whether those are fpga-based or not i don't know
Feb 27 17:52:57 <p2-mate> is there real arm11 SMP silicon ?
Feb 27 17:52:58 <vmaster> heh, yeah, and if you offer them your soul in return you might actually get an arm11 with docs to play with
Feb 27 17:53:05 <p2-mate> I thought it was still all FPGAs
Feb 27 17:53:14 <lennert> probably fpgas
Feb 27 17:53:20 <lennert> vmaster: yeah :)
Feb 27 17:53:53 <p2-mate> there has been  MIPS SMP SoC  silicon for a few years now :)
Feb 27 17:53:58 <lennert> we're having lots of fun with the ixp2350 since intel decided to use the armv5 isa and the armv6 mmu
Feb 27 17:54:17 <vmaster> i talked to someone from arm at the e-world in nuremberg, and there might actually be an off the shelf cortex available sometime this year
Feb 27 17:54:18 <lennert> and the armv6 mmu only allows supersection (16M granularity) mappings of physical addresses  > 4 GB
Feb 27 17:54:30 <lennert> (the ixp2350 has 33 bit physical address space)
Feb 27 17:54:49 <p2-mate> right. they should use MIPS :)
Feb 27 17:54:56 <lennert> p2-mate: i know of the bcm1250, where you generally just run linux on one core and packet processing on the other one
Feb 27 17:55:06 <p2-mate> lennert: it can do SMP linux too
Feb 27 17:55:07 <lennert> vmaster: a $10000 development kit? :)
Feb 27 17:55:30 <vmaster> mhh, no, like a real core in quantities
Feb 27 17:55:59 <lennert> p2-mate: i have some dual ixp2000 boards but the external bus isn't cache coherent so you have to run two kernels... :)
Feb 27 17:57:51 <p2-mate> right :)
Feb 27 17:58:03 <p2-mate> the 1250 is cache coherent
Feb 27 17:58:16 <p2-mate> and 64bit :)
Feb 27 18:04:06 <p2-mate> would be cool to have openocd support for. no open docs though :(
Feb 27 18:06:01 <lennert> i asked like 5 times what a bcm1250 board would cost.. never even got a reply
Feb 27 18:06:12 <lennert> so i guess the pricing structure isn't open either ;)
Feb 27 18:06:24 <p2-mate> no :(
Feb 27 18:06:33 <p2-mate> debian got some
Feb 27 18:06:40 <lennert> yeah
Feb 27 18:07:03 <lennert> tbm was looking for a fast arm board.. he thought 600mhz was too slow since he has a dual 800mhz mips and he's used to that :)
Feb 27 18:07:13 <p2-mate> :)
Feb 27 18:08:26 <p2-mate> I suggested IOP boards to several debian people already, but they somehow do not seem to get it
Feb 27 18:08:44 <p2-mate> (for fast arm boards)
Feb 27 18:08:51 <lennert> you mean like the allnet nas thingy?  i recommended that one to tbm, but he thought it was too slow.
Feb 27 18:08:56 <lennert> i don't know what other iop boards are out there.
Feb 27 18:09:02 <p2-mate> lennert: intel devboards
Feb 27 18:09:10 <lennert> hmmmmm, what do those cost?
Feb 27 18:09:14 <vmaster> hehehhe
Feb 27 18:09:20 <lennert> we have a dual ixdp2850 dev board which cost like $25000 or so
Feb 27 18:09:24 <p2-mate> right :)
Feb 27 18:09:40 <p2-mate> there is one for 500USD iirc
Feb 27 18:09:45 <lennert> last time vince asked intel for a donation they were willing to donate a 200 mhz pxa board
Feb 27 18:09:46 <p2-mate> or was it 1000 USD ?
Feb 27 18:10:02 <lennert> there's a dual ixp2350 pci-e board for $3000
Feb 27 18:10:04 <p2-mate> 200Mhz PXA I hope :)
Feb 27 18:10:08 <lennert> which is also too pricey
Feb 27 18:10:14 <p2-mate> but still :)
Feb 27 18:10:20 <lennert> yeah, not 200 millihertz :-D
Feb 27 18:10:46 <lennert> (200 millihertz.. that means a pipeline stall takes 1 sec :)
Feb 27 18:10:47 <p2-mate> but intel did pay for all of the wireless at dc5 including someone staying there for 1 week or so
Feb 27 18:11:13 <p2-mate> costs much more then a retail priced 1000USD board :)
Feb 27 18:11:21 <lennert> i think it depends on whether you find the right people within intel
Feb 27 18:11:26 <p2-mate> must be
Feb 27 18:11:32 <lennert> intel is a 100-headed monster
Feb 27 18:12:45 <vmaster> definitely
Feb 27 18:12:58 <p2-mate> devboard prices are also a bit arbitrary
Feb 27 18:13:36 <lennert> arbitrary, yes, very much so
Feb 27 18:15:40 <vmaster> beewoolie-afk: http://openocd.berlios.de/OpenOCD_Fosdem2006_talk.sxi
Feb 27 18:16:26 <p2-mate> vmaster: you are dominic ?
Feb 27 18:16:31 <vmaster> p2-mate: yeah
Feb 27 18:16:36 <p2-mate> ah ok :)
Feb 27 18:17:31 <beewoolie-afk> vmaster: thx
Feb 27 18:19:28 <beewoolie-afk> vmaster: What is sxi?
Feb 27 18:19:33 <lennert> staroffice ?
Feb 27 18:19:49 <beewoolie-afk> I have openoffice, but it doesn't use that extension by default...
Feb 27 18:20:00 <p2-mate> ooo yes
Feb 27 18:20:24 <vmaster> it's openoffice 1.1
Feb 27 18:20:29 <[g2]> HEY beewoolie-afk's in da house !!
Feb 27 18:20:31 <vmaster> but openoffice 2 opens that just as well
Feb 27 18:20:34 <beewoolie-afk> Must have gotten dumped when I updated.
Feb 27 18:20:40 <beewoolie-afk> [g2]: Hey
Feb 27 18:20:46 <[g2]> howdy :)
Feb 27 18:21:02 <vmaster> i had to use sxi as the laptop my university gave me didn't have ooo2
Feb 27 18:25:52 <vmaster> http://openocd.berlios.de/OpenOCD_Fosdem2006_talk.odp converted to open document presentation (OOo2)
Feb 27 18:28:35 <beewoolie-afk> vmaster: yeah, it was that OO had been uninstalled in the upgrade.
Feb 27 18:52:07 <prpplague> beewoolie-afk: hey, i sent a response email about the spi and such
Feb 27 18:52:14 <beewoolie-afk> OK.
Feb 27 18:52:16 <prpplague> beewoolie-afk: but i figured i'd tell you here anyway
Feb 27 18:52:30 <beewoolie-afk> :-)
Feb 27 18:52:32 <prpplague> beewoolie-afk: just curious about the spi drivers for apex
Feb 27 18:52:49 <prpplague> beewoolie-afk: till now, i've been putting them in the hardware specific type drivers
Feb 27 18:53:15 <prpplague> beewoolie-afk: do we want to do some generic spi framework?
Feb 27 18:53:49 <prpplague> beewoolie-afk: because i'm gonna have atleast 4 platforms, all with different spi drivers, all gonna be used for different things
Feb 27 18:54:02 <prpplague> beewoolie-afk: sd/mmc, enc28j60, and such
Feb 27 18:54:17 <beewoolie-afk> OK.
Feb 27 18:54:42 <beewoolie-afk> Since I've never been successful in using the MMC, I never worked on the spi driver.
Feb 27 18:56:15 <prpplague> beewoolie-afk: basically my question is, should i just have spi drivre for each platform, which is what i'm doing now, or start work on an api?
Feb 27 18:56:35 <prpplague> beewoolie-afk: i mean its basically like adding the other hardware specific drivers, like serial and timer
Feb 27 18:57:42 <prpplague> beewoolie-afk: the enc28j60 and sd/mmc drivers will be generic, but they will depend on the spi driver for the specific platform
Feb 27 18:58:09 <prpplague> beewoolie-afk: i.e. if the platform has no spi driver, then the enc28j60 and sd/mmc won't be available
Feb 27 18:58:29 <beewoolie-afk> That can be OK.
Feb 27 18:58:38 <beewoolie-afk> What I've been doing is defining a standard name.
Feb 27 18:58:53 <beewoolie-afk> We can have a standard SPI device.
Feb 27 18:59:06 <beewoolie-afk> The region-spec will point to the correct device.
Feb 27 18:59:10 <beewoolie-afk> or chip select.
Feb 27 18:59:56 <prpplague> beewoolie-afk: ok
Feb 27 19:00:18 <prpplague> beewoolie-afk: i'll go with that then
Feb 27 19:14:20 <[g2]> vmaster nice talk!
Feb 27 19:14:24 <[g2]> congrats!
Feb 27 19:14:35 <prpplague> vmaster: yea congrats
Feb 27 19:14:47 <prpplague> its always nice to see someone who works hard get some credit
Feb 27 20:43:59 <prpplague> beewoolie-afk: hey, thoughts on adding a lock/unlock command for flash?
Feb 27 20:44:12 <beewoolie-afk> Right now, the driver does it automatically.
Feb 27 20:44:26 <beewoolie-afk> Or are you *not* talking about the block lock?
Feb 27 20:44:40 <prpplague> beewoolie-afk: more or less a write protect
Feb 27 20:44:51 <beewoolie-afk> There are two levels.
Feb 27 20:45:06 <beewoolie-afk> There is a hardware VPEN line that some boards control elswhere via GPIO.
Feb 27 20:45:15 <beewoolie-afk> The block unlock/lock I do automatically.
Feb 27 20:45:30 <prpplague> beewoolie-afk: right
Feb 27 20:45:39 <beewoolie-afk> I suppose I could add a requirement that users commit to a flash write.
Feb 27 20:46:03 <beewoolie-afk> I am not sure what more we need.
Feb 27 20:46:12 <beewoolie-afk> The user has to enter a nor: prefix to write to flash.
Feb 27 20:46:17 <prpplague> beewoolie-afk: just that i've fucked my kernel twice today by typing in the wrong copy command
Feb 27 20:46:39 <prpplague> beewoolie-afk: you sure of that?
Feb 27 20:46:48 <beewoolie-afk> Duude.  Don't fuk the kernel.  It's not worth the hassle.
Feb 27 20:46:58 <prpplague> beewoolie-afk: hehe
Feb 27 20:47:32 <beewoolie-afk> What do you imagine the command would be?
Feb 27 20:47:42 <beewoolie-afk> The drivers don't presently have an unlock/lock method
Feb 27 20:48:10 <prpplague> beewoolie-afk: lock nor:256k+640k
Feb 27 20:48:17 <beewoolie-afk> What you're asking for is a two-phase write.
Feb 27 20:48:39 <beewoolie-afk> I believe that the locks are not saved to flash.  They're in volatile storage.
Feb 27 20:49:04 <prpplague> beewoolie-afk: yea, i would expect this to be stored in the environ space for apex
Feb 27 20:49:14 <beewoolie-afk> That isn't what I mean.
Feb 27 20:49:29 <beewoolie-afk> That's up to the user.
Feb 27 20:49:43 <beewoolie-afk> Let me think about it.
Feb 27 20:49:51 <prpplague> beewoolie-afk: okie dokie
Feb 27 20:49:55 <beewoolie-afk> The downside is that I don't want to make the base image any larger.
Feb 27 20:50:02 <prpplague> beewoolie-afk: yea
Feb 27 20:50:04 <beewoolie-afk> Adding a command to unlock will make the smallest APEX build larger.
Feb 27 20:50:12 <prpplague> agreed
Feb 27 20:50:23 <beewoolie-afk> So, I may have to add an option to enable the command.  Otherwise, it would remain automatic.
Feb 27 20:50:24 <prpplague> it was just a thought
Feb 27 20:50:30 <beewoolie-afk> Thots are good.
Feb 27 20:53:15 <prpplague> i'm looking forward to having the s3c2410 part of apex
Feb 27 20:53:26 <prpplague> hopefully more ppl will be using it
Feb 27 20:53:41 <beewoolie-afk> cool.
Feb 27 21:03:32 <prpplague> beewoolie-afk: i'll make a post about it as soon as the patches are accepted
Feb 27 21:03:44 <prpplague> beewoolie-afk: i need to add support for the littlechips board next
Feb 27 21:03:48 <beewoolie-afk> OK.
Feb 27 21:14:04 <prpplague> beewoolie-afk: i really like apex it really picked up where blob left off
Feb 27 21:14:27 <beewoolie-afk> thanks.
Feb 27 21:15:03 <prpplague> beewoolie-afk: and the debug features are just lovely :)
Feb 27 21:16:18 <prpplague> beewoolie-afk: i've got to find time to dump the rom contents of the pixter-mm and see what they are using for bootloader
Feb 27 21:16:30 <beewoolie-afk> That would be interesting.
Feb 27 21:16:43 <beewoolie-afk> I'd guess that they use the same thing they used in the first rev.
Feb 27 21:28:57 <prpplague> beewoolie-afk: somehow if feel that they didn't as the the 3 rev was based on the LH75411
Feb 27 21:29:10 <prpplague> beewoolie-afk: the 4th (the pixter-mm) is on the lh79524
Feb 27 21:29:21 <prpplague> beewoolie-afk: just doesn't seem likely to me
Feb 27 21:29:28 <beewoolie-afk> let me know
Feb 27 21:29:52 <prpplague> beewoolie-afk: will do
Feb 27 21:30:12 <prpplague> beewoolie-afk: i've got a real nasty feeling that they might be using linux and violating GPL
Feb 27 21:30:32 <prpplague> beewoolie-afk: mattel and fisher-price have done so in the past when they use sub-contractors
Feb 27 21:30:45 <prpplague> beewoolie-afk: that way they can disavow knowledge
Feb 27 21:31:17 <beewoolie-afk> apparently, someone from montavista was giving advice on how to circumvent GPL
Feb 27 21:31:29 <prpplague> beewoolie-afk: yea, thats what i heard
Feb 27 21:47:47 <lennert> well, it was stuff like don't write kernel drivers cause then harald welte is going to get you
Feb 27 22:37:10 <z3ro___> hmm, one of the many reasons I don't like people form companies using the linux kernel (don't get me wrong, the ones who actually abide by the gpl are fine, but many do not)
Feb 27 22:37:32 <z3ro___> in fact, even the ones who do abide by the gpl usually have to be pushed to do so.
Feb 27 22:38:11 <z3ro___> which is a real shame, imho... seems that most of them don't realize that gpl is *NOT* public domain.
Feb 27 22:38:23 <lennert> they are starting to realise now
Feb 27 22:38:45 <z3ro___> some are... but it's still not as good as it should be.
Feb 27 22:38:55 <z3ro___> there are many companies violating the gpl.
Feb 27 22:41:44 <lennert> yeah
Feb 27 22:42:06 <lennert> claiming that you wrote the linux kernel is one way of circumventing the gpl :P
Feb 27 22:42:38 <z3ro___> wouldn't supprize me if someone at montavista tried that...
Feb 27 22:42:52 <z3ro___> "I wrote one kernel module, my kernel! mine!"
Feb 28 01:51:30 <AchiestDragon> :)
Feb 28 01:52:55 <ka6sox> is it mourning?
Feb 28 01:53:09 <AchiestDragon> 2am
Feb 28 01:55:00 <lennert> 3am here :)
Feb 28 02:14:59 <ka6sox> yikes!
Feb 28 02:22:14 <ka6sox> only 6pm here
Feb 28 02:47:54 <ka6sox> at 3:30 dwery goes to bed :)
Feb 28 03:16:04 <beewoolie-afk> I've posted a first version of my JTAG application.
Feb 28 03:16:18 <beewoolie-afk> It isn't *quite* useful except for armv4 cores.
Feb 28 03:16:37 <beewoolie-afk> And it may take  some coaxing at that.
Feb 28 03:16:57 <beewoolie-afk> I did this so that I could take a break and pretend that I've released.
Feb 28 03:17:09 <beewoolie-afk> ftp://ftp.buici.com/pub/taproot
Feb 28 03:17:34 <beewoolie-afk> If anyone wants to try it, let me know and I'll help you along.
Feb 28 03:17:58 <beewoolie-afk> Mainly, I want to see if it works with anyone else's hardware: adapters and targets alike.
Feb 28 03:18:17 <beewoolie-afk> If you want to add a target, make sure to get a BSDL file.
**** ENDING LOGGING AT Tue Feb 28 10:59:56 2006