**** BEGIN LOGGING AT Fri Jan 27 03:00:04 2006 Jan 27 07:31:56 hmm... how difficult is it to get a jtag header onto a nslu2? Jan 27 07:35:49 vmaster: http://nslu2-linux.thestuffguy.com/gallery/jtag Jan 27 07:36:03 thanks Jan 27 09:08:40 velinp, any word on your fattened Slug? Jan 27 09:10:20 ka6sox: not fattened; friend hesitating about risk/benefit; maybe later (found a local supplier of 256Mbit chips, price ~12 EUR; need some soldering hardware update, etc.) Jan 27 09:11:13 velinp, its not easy with those small traces. Jan 27 09:12:52 yes, i know :) friend has experience; i trust him when he says 'it;s risky'; how are bob and wendy doing? velnas is stuck with mingw32 for >60 hours; Jan 27 09:13:18 they are both chugging away... Jan 27 09:14:32 I'm still working on getting fatter hardware to reduce the swapping Jan 27 11:04:43 http://www.goepel.com/content/html_en/index.php Jan 27 11:04:48 anyone looked at these guys Jan 27 11:04:56 they are covered in the latest circuit cellar Jan 27 12:35:59 beewoolie: hey hey Jan 27 12:36:05 prpplague: Jan 27 12:36:08 prpplague: hi Jan 27 12:36:48 The good news is that I understand what I have to do to perform accelerated debugging on the Xscale. Jan 27 12:37:07 The bad news is that I don't have a jtag emulator with the necessary signals. Jan 27 12:38:23 beewoolie, bummer Jan 27 12:38:29 *soon* (tm) Jan 27 12:38:32 A minor setback. Jan 27 12:38:53 The dongle from GW may have the nSRST signal. Jan 27 12:39:08 It turns out that there isn't really a substitute on this CPU. Jan 27 12:40:54 [g2]: Any word on the GW docs? Jan 27 12:43:09 <[g2]> beewoolie I got them last night. However, all the pages said "..... Confidential" across them.... Jan 27 12:43:10 * [g2] was not pleased Jan 27 12:43:30 OK. Let me know if/when they are willing to share. Jan 27 12:48:00 <[g2]> beewoolie I'm sure I could send you a copy as I don't think you release the docs to the world Jan 27 12:48:17 <[g2]> However, I wound not want the software you write to be restricted Jan 27 12:48:25 [g2]: How would they feel about an OSS driver... yeah. Jan 27 12:48:52 As long as it only works with their device, is there really an issue? Jan 27 12:48:55 <[g2]> We may be better off putting our energies into a different platform Jan 27 12:49:03 Fair enough. Jan 27 12:49:17 I was hoping for something in the short term that supports sRST. Jan 27 12:49:32 I need it for Xscale. Jan 27 12:49:52 I may have to start with the ep1220 board instead. Jan 27 12:50:02 <[g2]> it may well be that for the short-term (next few months) we use this Jan 27 12:50:31 <[g2]> I'm not sure Jan 27 12:50:45 Does their SW work from Linux? Jan 27 12:50:46 <[g2]> I wanted to cool off a little before I speak with them about it Jan 27 12:50:56 Are you mad? Jan 27 12:51:20 <[g2]> I'm not mad, I'm a tiny bit disappointed Jan 27 12:51:20 Or has the weather been unseasonably toasty? Jan 27 12:51:24 Hmm. Jan 27 12:51:58 I'm glad to return it if this doesn't work out. Jan 27 12:54:42 <[g2]> nod Jan 27 12:54:44 <[g2]> thx Jan 27 16:46:36 [g2]: still there? Jan 27 17:02:12 beewoolie: if you're looking for something cheap that's known to work, have a look at the olimex arm-jtag (sparkfun distributes to the u.s., iirc) - it's a simple parport wiggler clone, with working nSRST Jan 27 17:02:29 vmaster: Cool. Jan 27 17:02:34 * beewoolie browses... Jan 27 17:10:22 vmaster: $20. Not a lot of corn. Jan 27 17:11:33 i doubt any commercial offering could be a lot less Jan 27 17:11:51 Indeed. Jan 27 17:12:22 i have one on my desk, and i know that i'm able to control both nTRST and nSRST separately Jan 27 17:13:06 which is what you need for xscale debugging - you hold the core in reset while loading the mini i-cache Jan 27 17:13:16 That's exactly the issue. Jan 27 17:15:52 Oddly, they don't have a schematic for the arm-jtag. There is one for the MSP hip Jan 27 17:17:11 there's a posting in the sparkfun forum from olimex documenting the arm-jtag connections Jan 27 17:18:09 do you have a link? Jan 27 17:18:31 http://www.sparkfun.com/cgi-bin/phpbb/viewtopic.php?t=2068&highlight=armjtag Jan 27 17:19:14 there are typos, DB is always DB25, not DB26, but the signals are correct Jan 27 17:19:42 and NTRST is really nSRST, TRST is nTRST Jan 27 17:20:20 That can be deduced from the JTAG pin numbers. Looks good. Jan 27 17:20:42 Someone just pointed me to a device called jlink. Jan 27 17:20:47 I'm going to check that out, too. Jan 27 17:20:47 segger Jan 27 17:20:56 it's proprietary withouting documentation Jan 27 17:21:03 Not good. Jan 27 17:21:46 Is it the ftdi2322? Jan 27 17:22:20 I don't think so, as it's faster than the FT2232 ever could be Jan 27 17:22:47 Looks like it's windows centric? Jan 27 17:22:55 yes Jan 27 17:23:02 Non starter Jan 27 17:23:43 Probably not $20. Jan 27 17:24:26 it's reliable debugging hardware, if you're willing to pay the price Jan 27 17:25:01 I bought a bdi2000 for that reason. It's reliable, but not flexible. Can't use it for xscale unless I want to pony up another US$1000 Jan 27 17:25:18 And since I don't get paid for the Xscale work, that's not happening. Jan 27 17:27:44 i'm looking to get an xscale development board, as it's always been on my wish-list for the openocd Jan 27 17:28:33 but xscale debugging is a whole new world - most of what i've done one the host for arm7/9 targets has to happen on the xscale Jan 27 17:38:05 I'm off - night everyone Jan 27 17:38:36 vmaster: nite Jan 27 17:47:25 <[g2]> vmaster nite Jan 27 17:47:35 <[g2]> I'll catch you tomorrow Jan 27 17:48:14 <[g2]> lennert beewoolie can I ask you a quick question ? Jan 27 17:48:27 as long as I can ask you one. :-) Jan 27 17:49:37 <[g2]> sure I've got to run in a couple minutes but can ask now Jan 27 17:51:24 <[g2-lap]> http://pastebin.ca/38858 Jan 27 17:51:46 <[g2-lap]> The madwifi driver is oopsing on an address that seems valid Jan 27 17:52:08 <[g2-lap]> The PCI space is mapped as a 64K region at 0x48000000 Jan 27 17:52:24 <[g2-lap]> and the address being referenced is 0x48004004 Jan 27 17:54:26 <[g2]> beewoolie what's your Q ? Jan 27 17:54:38 hang on, I'm reading your Q Jan 27 17:55:55 It could be that those addresses are PCI mapped and that it only goes for 0x4000 bytes Jan 27 17:56:08 An lspci would tell you. Jan 27 17:56:45 That would be my guess. Jan 27 17:56:49 The error is bonafide Jan 27 17:56:58 It is saying that the descriptor at that address is zero. Jan 27 17:57:39 Oh. I see. Jan 27 17:57:41 You have that. Jan 27 17:58:33 Also remember that the memory reported by PCI is physical. Jan 27 17:58:47 The read is a virtual address read. Jan 27 17:58:57 There isn't a guarantee that they'll be identity mapped. Jan 27 17:59:07 Usually, they aren't, but I'm not so familiar with PCI. Jan 27 18:00:09 My question is easier. Jan 27 18:00:25 I haven't been able to find Harvold's router distro. Jan 27 18:00:31 Do you have a keyword or a link? Jan 27 18:08:06 <[g2-lap]> https://svn.gnumonks.org/trunk/grouter/ Jan 27 18:08:14 <[g2]> beewoolie how's that ? :) Jan 27 18:08:24 sweet Jan 27 18:08:51 googling grouter got a lot of tile stuff Jan 27 18:09:07 <[g2-lap]> http://gnumonks.org/~laforge/weblog/2005/ Jan 27 18:09:16 <[g2-lap]> search that page for grouter / linwrap Jan 27 18:09:26 <[g2-lap]> bbl Jan 27 18:12:42 thanks Jan 27 18:12:59 <[g2]> np Jan 27 18:13:09 He says that it is x86 only. Jan 27 19:34:36 <[g2-lap]> re Jan 27 19:35:44 <[g2-lap]> beewoolie I think we'd be able to take those apps and the configs and run them pretty easily Jan 28 01:05:11 vmaster_: I'm looking at the LH 79520 datasheet again to see how I missed the fact that it's an r3 part. Jan 28 01:05:18 The file I'm looking at: http://document.sharpsma.com/files/tec_datasheet_lh79520.pdf Jan 28 01:05:31 If it says r3, it's well hidden. Is there something else you were reading? Jan 28 01:06:52 Ah, now I see. It's burried in the description. **** ENDING LOGGING AT Sat Jan 28 02:59:57 2006