**** BEGIN LOGGING AT Sat Feb 05 19:33:03 2005 **** ENDING LOGGING AT Sat Feb 05 20:14:00 2005 **** BEGIN LOGGING AT Sat Feb 05 20:15:04 2005 Feb 05 21:05:18 welcome Randy. Feb 05 21:05:25 hi, thanks Feb 05 21:05:38 this is the one I mostly hang out in. Feb 05 21:05:48 we talk about Bootloaders and JTAG things. Feb 05 21:06:04 you're some kind of admin type here ka6sox-away? Feb 05 21:06:34 uh...I guess so...I'm the crazy who started this project. Feb 05 21:08:18 we have finally settled on the hardware/software combo for this project. Feb 05 21:08:50 this isn't the remote radio is it? Feb 05 21:11:31 no it's not ;) Feb 05 21:11:35 we are building a FAST JTAG programmer... Feb 05 21:11:49 and Debugger for embedded linux. Feb 05 21:12:19 fun Feb 05 21:15:17 it is...been some spirited discussions and its a very good group. Feb 05 21:40:15 dyoung: you still around? Feb 05 21:40:54 dyoung: I think that we need to be on the same distro. One of us is going to have to change. Feb 05 21:41:26 beewoolie, what distro are you on? Feb 05 21:42:03 I've never upgraded. Still stock. Feb 05 21:42:17 I figured that it was best to *not* change too many variables at one time. Feb 05 21:44:20 probably a good idea Feb 05 21:44:31 he should have a stock one there. Feb 05 21:44:33 :) Feb 05 21:44:42 he's gonna love that. Feb 05 21:44:55 There is an alternative....which he's also gonna love. Feb 05 21:45:07 uh oh... Feb 05 21:45:13 I'm thinking it would be helpful to figure out how far the loader is getting. Feb 05 21:45:32 I'd put led markers in the code and see what markers are coming on. Feb 05 21:46:02 Once APEX comes to a prompt, I can do the rest of the debug without a jatgg crutch. Feb 05 21:46:26 What's dy's TZ? Feb 05 21:50:54 HST Feb 05 21:51:38 not that they ever change. Feb 05 23:24:34 ka6sox: Did prplague's inclination to use the MMC controller chip have any allure? Feb 05 23:26:01 for me? Feb 05 23:26:12 yes...I would like to use MMC in a project. Feb 05 23:26:13 or at all? Feb 05 23:26:34 I mean that he found it possible to use one of these USB->MMC controllers to handle JTAG. Feb 05 23:26:50 I know we need some IO pins to control the voltages and muxes. Feb 05 23:27:04 Could it be sufficient? Feb 05 23:27:12 is it lower cost and as available? Feb 05 23:27:39 Dunno. He gave me a link. Let me check. Feb 05 23:27:39 MMC is really SPI taken to a faster level. Feb 05 23:27:54 and I have done SPI for many years. Feb 05 23:30:42 Neither digikey nor mouser is coming up with the part. It's an Alcor9331. http://www.alcormicro.com/product_au9331.htm Feb 05 23:31:32 Seems like it should be a cheap part. Feb 05 23:33:56 Get this. Best Buy has a SD reader, USB interface, for $20. If that isn't cheap, I don't know what is. Feb 05 23:35:14 And it has a USB 2.0 interface. Feb 05 23:42:42 SD has IP restrictions..and the other problem is no Open Feb 05 23:42:48 source drivers. Feb 05 23:43:01 that is *primo important to me. Feb 05 23:44:11 This wouldn't be for SD. It would be for JTAG. He believe these devices can be used to run the JTAG scan. Feb 05 23:44:31 We can write drivers as long as there is documentation on the USB protocol. Feb 05 23:45:15 I only asked because I don't yet have a full grasp on the scope of the hardware needed. Feb 05 23:45:17 let me look and see if the device datasheet has all the instructions. Feb 05 23:45:31 The 2232 will certainly suffice. Feb 05 23:46:59 They don't appear to have a datasheet, just an technical overview. There are not details about the interface. Feb 05 23:48:14 ugh...that scares me. Feb 05 23:48:27 but the MMC technology would be nice Feb 05 23:48:39 http://www.alcormicro.com/product_au6330.htm Feb 05 23:48:55 I'm not sure I follow you. What about MMC is interesting to you? Feb 05 23:49:50 Perhaps I'm just dumb about USB. There really aren't any details about how the USB attaches to the SD. I suppose it could be such a simple conduit that data presented at the USB side is piped directly to the SD. interface and vice versa. Feb 05 23:50:18 It says there is a Linux driver... Feb 05 23:51:31 I see that. Feb 05 23:51:39 but havent' found the driver yet...still looking. Feb 05 23:51:51 Me too. No searches are coming up. Feb 05 23:52:09 the 2232 still seems like the way to go. Feb 05 23:54:32 I don't have a problem with that. with the alcor chip, it isn't clear how we'd handle configurability. Feb 05 23:54:53 yeah. Feb 05 23:55:07 that is the advantage of the 2232 Feb 05 23:55:16 having more than one LUN helps. Feb 05 23:56:13 * beewoolie nods Feb 05 23:57:10 It seems like we'd have to do some shenanigans like this. There are four data pins on SD. We'd use one for TDI and another for TDO. this means that the data rate would be 1/4th of the USB transfer rate. Feb 05 23:57:50 probably Feb 05 23:58:27 I don't know. I suppose it could be done. I think I'd rather hook my pony to a train that I can recognize. We may have more parts, but I'm sure we can make it work. Feb 05 23:58:58 Perhaps later, we can adapt the JTAG software to handle other dongles. Feb 05 23:59:17 thats fine with me... Feb 06 00:00:39 I feel I need to get my hands on the JTAG spec. I'll check to see if the UW library has it. Feb 06 00:01:07 good place to find it. Feb 06 00:02:22 Night. **** ENDING LOGGING AT Sun Feb 06 05:39:56 2005