**** BEGIN LOGGING AT Mon Dec 22 02:59:56 2008
Dec 22 15:58:10 <prpplague> dirk2: hey bud
Dec 22 15:58:14 <prpplague> dirk2: how goes the battle
Dec 22 15:58:58 <dirk2> prpplague: You can read my progress at http://elinux.org/BeagleBoardOpenOCD :)
Dec 22 15:59:09 * prpplague looks
Dec 22 15:59:19 <dirk2> prpplague: I'm still at the learning & reproducing level
Dec 22 16:00:43 <dirk2> prpplague: I was successfully able to execute the *command* "jtag tapenable omap3530.cpu", but I'm not sure if it was successful
Dec 22 16:01:10 <dirk2> prpplague: And I still have to learn how to verify that ;)
Dec 22 16:01:14 <prpplague> dirk2: if you are able to re-scan the chain and it shows up, it is enabled
Dec 22 16:01:25 <prpplague> dirk2: right, which is why i was using the dm355 board
Dec 22 16:01:47 <prpplague> dirk2: the davinci already has openocd support, so i'd be able to do the tests on it
Dec 22 16:02:24 <dirk2> prpplague: There are we again at "learn level": How do I "re-scan" or/and what does this mean?
Dec 22 16:03:51 <prpplague> dirk2: the scan chain has a feature where you can get the ID's of everything on the chain, normally when the ICEPick is the only item enabled, you will only see the ID for the ICEPick on the chain
Dec 22 16:04:23 <prpplague> dirk2: once you successfully enable the arm tap, you should be able to scan again, this time getting two id's, one for the ICEPick and one for the arm
Dec 22 16:04:52 <dirk2> prpplague: You mean the command "scan_chain"?
Dec 22 16:05:04 <prpplague> dirk2: yea
Dec 22 16:05:19 <prpplague> dirk2: i was just looking at your output on the page
Dec 22 16:05:25 <prpplague> dirk2: looks like that is what you did
Dec 22 16:06:00 <prpplague> dirk2: something doesn't look right
Dec 22 16:06:06 <dirk2> prpplague: Yes, and i got 0x00000000 which means it doesn't work, right?
Dec 22 16:06:08 <prpplague> dirk2: there is a no ID information for the cpu
Dec 22 16:06:24 <prpplague> dirk2: yea, looks like something isn't working or misconfigured
Dec 22 16:07:05 <dirk2> prpplague: Can you have a look to the scripts in the middle of the page. there is a jtag configure $_CHIPNAME.cpu -event tap-enable
Dec 22 16:07:19 <prpplague> dirk2: one sec
Dec 22 16:07:46 <dirk2> prpplague: Is there a chance I can verify that each command is successful?
Dec 22 16:08:10 <dirk2> prpplague: What I understand is that we wirte some "JTAG registers" by this
Dec 22 16:08:11 <prpplague> dirk2: no way of knowing with out knowing what they do, and that is the black magic TI isn't giving up
Dec 22 16:08:16 <prpplague> dirk2: yea
Dec 22 16:08:39 <prpplague>  # force an error till we get a good number
Dec 22 16:08:39 <prpplague>   set _CPUTAPID 0xffffffff
Dec 22 16:08:40 <dirk2> prpplague: Am I able to read these "registers" back and look what are in them?
Dec 22 16:08:55 <prpplague> dirk2: as far as i know, no
Dec 22 16:09:35 <dirk2> prpplague: Hm :( I thought of it as something like I2C: Write registers, read registers
Dec 22 16:10:28 <prpplague> dirk2: right, i should qualify myself with that statement, i am sure there is a way to read back the registers, but since we have no idea what the registers are or the method for reading it back..........
Dec 22 16:11:16 <dirk2> prpplague: Then, I think an additional hint is in https://lists.berlios.de/pipermail/openocd-development/2008-December/004166.html
Dec 22 16:11:34 <dirk2> prpplague: I think we have to do a reset and then wait 10 TCK
Dec 22 16:12:53 <dirk2> prpplague: Rick mentioned "I think we can address this in the config fairly easily", but I have no idea how to do this. Main question is if there is already a command to wait 10 TCK, or if it has to be implemented (what most probably is 'easy')
Dec 22 16:13:19 <prpplague> dirk2: yea
Dec 22 16:14:28 <dirk2> prpplague: At the moment I'm thinking about something like
Dec 22 16:14:31 <dirk2> jtag_reset
Dec 22 16:14:36 <dirk2> wait 10 TCK
Dec 22 16:14:49 <dirk2> irscan omap3530.jrc 7
Dec 22 16:14:51 <dirk2> ...
Dec 22 16:14:55 <dirk2> to add to the script
Dec 22 16:17:11 <prpplague> sounds doable
Dec 22 16:17:33 <dirk2> yes, but the details ;)
Dec 22 16:19:01 <dirk2> prpplague: Conclusion is that I have success if I get something else than 0x00000000 IdCode, right?
Dec 22 16:19:20 <prpplague> dirk2: correct
Dec 22 16:20:07 <dirk2> prpplague: Do you know if Rick or Duane are sometimes here?
Dec 22 16:20:37 <prpplague> i've seen duane on the #beagle a few times
Dec 22 16:21:04 <dirk2> As "duane" ?
Dec 22 16:22:35 <prpplague> yea iirc
Dec 22 16:24:32 <dirk2> ok, will send a mail to mailing list if Rick would have some more details regarding "waiting 10 TCK".
Dec 22 16:28:49 <prpplague> good deal, sorry i can't spend more time on it right now, we have a product release scheduled for jan9, so not much time to finish up some of the details
Dec 22 16:31:10 <dirk2> np. You help already a lot with a short chat :)
Dec 22 16:31:42 <dirk2> And I learn a lot :)
Dec 22 16:32:44 <prpplague> dirk2: good deal then!
Dec 22 22:20:04 <flyback> m00
Dec 22 22:21:39 <L_Deluxe> o1nk
Dec 22 23:16:47 * flyback is feeling a little better about his life in some ways but still feels his body and mind are deteriating
Dec 23 00:30:22 * flyback headbangs to http://www.8bitpeoples.com/discography?page=6
Dec 23 00:37:13 <flyback> http://www.8bitpeoples.com/mp3/get/262/8bp038-05-bit_shifter-let_it_snow.mp3 <--- owns your "canuck"
**** ENDING LOGGING AT Tue Dec 23 02:59:57 2008