**** BEGIN LOGGING AT Wed Jul 31 02:59:56 2019 Jul 31 03:18:35 hello everybody, I'm triyng to disconnect and reconnect an usb_f_mass_storeage driver to reload a file from a boneblack and notice the host the changes, someone knows how to achieve that?? Jul 31 03:58:44 I'm worried by that question Jul 31 04:03:02 lol Jul 31 07:46:56 Hi Jul 31 07:47:26 Could you please provide the HTS and ECCN for BeagleBoard-X15 Jul 31 07:50:52 Arun33: ask your distributor, we're not a sales channel Jul 31 07:51:39 okay Jul 31 07:52:30 and/or just google... I found an HTS for the BeagleBoard-X15 with ease (on Arrow's product page for it) Jul 31 07:58:32 but your best bet is to ask your distributor Jul 31 12:22:18 m Jul 31 19:37:26 https://www.cnx-software.com/2019/07/31/beaglebone-green-gateway-octavo-osd3358-sip-ethernet-dc-jack/ Jul 31 19:56:44 schamtics? Jul 31 20:18:50 jkridner: how expensive is the eMMC on the BBB roughly? the OSD3358-512M-BCB (which includes 4GB eMMC) is $11.76/unit more expensive than the OSD3358-512M-BSM (on digikey, prices for 100 units, which is the highest quantity for which they listed prices)... which seems like a lot to me Jul 31 20:19:58 hmmmm..... less than $7. Jul 31 20:20:33 zmatt: don't think I can share schematics yet. Seeed should publish when ready. Jul 31 20:20:52 I hope they're better than their previous work... *cough* :P Jul 31 20:21:11 zmatt: I think the value for the integrated eMMC is simplicity. also eMMC prices in low volume are still rather high. Jul 31 20:21:32 it seems unlikely octavo buys them in low volume? Jul 31 20:21:32 zmatt: you picking on the cape header stuff or another issue? Jul 31 20:21:45 no, typing a reply to a comment in that article, just doing some fact-checking Jul 31 20:22:08 zmatt: true.... I'm just saying that the extra price for users might be justifiable. Jul 31 20:23:46 zmatt: I'll say that one negative is that the Ethernet is USB-based. They made a trade-off to make sure they didn't use any cape-header resources for WiFi. So, the headers should now be fully BeagleBone Black compatible. Jul 31 20:23:48 Here's what I ended up typing in reply to the comment wondering why they didn't use the C-SIP (my comment is still pending moderation): https://pastebin.com/UaVXT3wX Jul 31 20:23:59 oof Jul 31 20:24:07 I had assumed/hoped the wifi would be usb based Jul 31 20:24:09 I made the same comment re #1 Jul 31 20:24:33 yeah, that'd been better, but they stuck with the WL1837MOD. Jul 31 20:24:54 has good quality, features, but lacks 802.11ac. Jul 31 20:25:08 sacrificing real ethernet sounds like a big oof to me Jul 31 20:25:18 but I guess it depends on one's priorities Jul 31 20:26:42 I think your facts are right, but I do like that new SIP still.... just not for a Beagle. I'd love to see a PocketBone built with it. Jul 31 20:27:10 Hello all, anyone available to help troubleshoot the SPI1 channel? Jul 31 20:27:25 jkridner: just deadbug one ;) Jul 31 20:27:46 bRitch022: the easy way to find out is by asking your question / explaining your problem Jul 31 20:28:28 (and have patience. there are plenty of knowledgable people here, but most of them only glance at chat occasionally so it can take a while to get a response) Jul 31 20:28:38 I am having trouble getting the SPI1 to work properly. I am using the test-spi.py file (though I edited it to poll the spi channel again after a 1 second delay) and have verified that pins P9_29 & P_30 are jumped together for the loopback test, which is unsuccessful. Jul 31 20:28:38 clock signal at all. Jul 31 20:28:53 @zmatt thank you Jul 31 20:29:27 it looks to melike some of your question got lost: https://pastebin.com/raw/RSY2yFx3 Jul 31 20:30:08 I have also attached an oscilloscope to P9_31, and I am not seeing an oscillating clock signal at all Jul 31 20:30:29 Detailed explanation and attempted fixes: Jul 31 20:30:29 https://forums.adafruit.com/viewtopic.php?f=49&t=154691 Jul 31 20:30:39 first thing I'd double-check is the pin config Jul 31 20:31:01 Triple-checked, promise! Jul 31 20:31:02 e.g. using my show-pins utility: https://github.com/mvduin/bbb-pin-utils#show-pins Jul 31 20:31:37 Ok, giving that a go Jul 31 20:31:41 just for the quadruple check, can you install my utility and share the output of "sudo show-pins | grep spi" e.g. via pastebin.com (or some other paste service) ? Jul 31 20:32:29 if it's any comfort: I know spi1 works fine, I have a board sitting here which uses it intensively Jul 31 20:33:30 That's certainly a light at the end of the tunnel here! Thanks zmatt Jul 31 20:33:52 have you disabled audio in /boot/uEnv.txt ? since hdmi audio occupies some pins you need Jul 31 20:34:07 Yes Jul 31 20:36:21 this is what it looks like for me: https://pastebin.com/raw/HvwbGWha Jul 31 20:36:30 zmatt: I've been contributing to https://github.com/RobertCNelson/boot-scripts/blob/master/device/bone/show-pins.pl Jul 31 20:37:04 zmatt: maybe rcn-ee[m] should just package your upstream if you accept the patches? Jul 31 20:37:18 ie, turn it into a debian package, rather than live in /opt/scripts Jul 31 20:37:31 jkridner: I didn't know it lived there Jul 31 20:37:46 doh Jul 31 20:37:49 https://pastebin.com/DMHkttc5 Jul 31 20:38:08 yeah, sorry I didn't reach out to you earlier myself. kinda thought I'd mentioned it, but I probably forgot. Jul 31 20:38:32 * jkridner goes nuts typing /opt/scripts/device/bone/show-pins.pl rather than just show-pins! Jul 31 20:38:40 jkridner: you mentioned something about patches earlier, but didn't say where, and I didn't see any pull requests or whatever to my repo Jul 31 20:39:20 jkridner: he added a .pl suffix? why? :P Jul 31 20:39:30 the language it's written in is an implementation detail Jul 31 20:39:46 +1 Jul 31 20:40:00 check my patches. I've been using it extensively for BBAI Jul 31 20:40:34 oof, looks like his version is so old that I didn't have a github repo yet Jul 31 20:40:46 oh! Jul 31 20:40:57 maybe the merge will be tough then? Jul 31 20:41:25 he added it november 2015, my initial commit was august 2016 Jul 31 20:41:44 https://github.com/mvduin/bbb-pin-utils/commits/master Jul 31 20:42:14 https://github.com/mvduin/bbb-pin-utils/commits/master/show-pins ... more useful list of commits Jul 31 20:42:17 so, not too many commits Jul 31 20:42:28 but I also have branches for bbblue, bbb-wireless, and bbx15 Jul 31 20:42:49 oh and bbw apparently, forgot about that Jul 31 20:43:26 zmatt: https://pastebin.com/ELeTjhJx Jul 31 20:43:28 I should probably add autodetection of the device model and unify them Jul 31 20:43:29 Looks pretty similar to yours Jul 31 20:43:57 bRitch022: woops, forgot to reply to you. yeah that looks fine to me Jul 31 20:45:15 Nice utility! Jul 31 20:45:48 when you said "I am having trouble getting the SPI1 to work properly" ... did you mean to imply you do have SPI0 working, or is this just your first attempt at using SPI? in the latter case, the probability it's just a software issue obviously greatly increases Jul 31 20:45:57 thanks :) Jul 31 20:46:10 jkridner: it probably will, but I'll take a look at your patches later Jul 31 20:47:01 zmatt: why branches and not if-statements? Jul 31 20:47:08 zmatt: I have not tried SPI0 yet Jul 31 20:49:30 zmatt: this is all I'm trying to run Jul 31 20:49:30 https://pastebin.com/D9G8kB1L Jul 31 20:50:07 jkridner: apart from the bbx15 version, they all just differ in the data table. like I said, I probably should have made it auto-detect model and add a description column per model supported to the data table, but this was less effort and as long as I have people download and install the script into /usr/local/sbin anyway, it's not really a problem to just use the version correct for their actual ... Jul 31 20:50:13 ...board Jul 31 20:50:38 I had to add a module for multiple inline files. Jul 31 20:50:57 * jkridner misses using perl on a regular basis. Jul 31 20:51:22 perl is a tough language to remember, but an easy language to program when doing it all the time. Jul 31 20:52:03 bRitch022: at home I have a bbb with an sd card containing a relatively clean version of the latest official image, so there I can test stuff more easily... the beaglebones I have here are running custom images (and most of them are embedded into stuff) Jul 31 20:52:13 zmatt: I included your data table in https://docs.google.com/spreadsheets/d/1fE-AsDZvJ-bBwzNBj1_sPDrutvEvsmARqFwvbw_HkrE/edit?usp=sharing Jul 31 20:52:58 jkridner: I used to be a pretty deep perl guru, but I sort of drifted away. I still use it every now and then though, I'm pretty productive in perl (maybe in part because I'm not particularly motivated to write "nice" code ;-) Jul 31 20:53:04 for the BBAI SRM, we are looking at documenting with pins in the column.... that way the vertical high-visibility range goes across all the mux modes. Jul 31 20:53:30 switched SRM to asciidoc with the hope of importing .csv Jul 31 20:53:48 bRitch022: so if you're not in a hurry, it's probably easier to me to help you when I get home (probably in an hour or so) Jul 31 20:53:53 colors inspired by PocketBeagle documentation. Jul 31 20:54:20 jkridner: the show-pins data table is directly based on my spreadsheet ( https://goo.gl/Jkcg0w ) Jul 31 20:54:33 probably lots of errors, certainly lots of gaps still in that table. Jul 31 20:54:45 btw, I can never get the greek characters to display properly in show-pins Jul 31 20:55:26 zmatt: I gotta jet in about a half hour. Will you be on on friday? Jul 31 20:55:34 that table's pretty. Jul 31 20:55:49 bRitch022: I'm around often enough, at random times. Jul 31 20:56:22 the trick is to just poke me here in chat and leave your client connected, I'll respond sooner or later ;) Jul 31 20:56:27 any thought of how to document iosets? Jul 31 20:56:34 the AM5279 iosets seem odd. Jul 31 20:57:01 Sounds good, thanks alot. Great meeting you, I'm glad I found this chat. It's interesting! Jul 31 20:57:30 jkridner: I don't think I attempted to document those yet... this is my current spreadsheet: https://goo.gl/jiazTL Jul 31 20:57:39 what are the xip-m1/2 columns? Jul 31 20:58:24 jkridner: on the first tab of my am335x pins spreadsheet? columns AB-AM are all about pins used by various boot modes Jul 31 20:58:38 k Jul 31 20:58:39 red if they're configured as output (at least some of the time) Jul 31 20:58:48 bRitch022: do not send private messages Jul 31 20:58:50 just use chat Jul 31 20:59:07 zmatt alright Jul 31 21:00:25 (private messages = an extra window in my irc client, which already has 18 windows so I don't particularly feel like having more of those. I'm also far less likely to see messages since I normally just have this channel as the active window) Jul 31 21:00:48 * jkridner finds the IO set tab Jul 31 21:01:01 would love to have that against the cape header pins. Jul 31 21:01:06 jkridner: note that "Pins" is generic for the AM335x, the "BBB" tab is the derivative for the BBB specifically Jul 31 21:01:16 the P9/P8 tabs are probably the ones I use the most Jul 31 21:01:34 Signals can also be useful, as can PRU Jul 31 21:02:06 yeah I did document the am335x iosets Jul 31 21:02:15 based on the json data I think Jul 31 21:02:50 I know I've done work on converting the json for the am3358 into something more easily digestable, but I don't remember where I got with that Jul 31 21:03:16 for the am572x I mean Jul 31 21:03:17 sorry Jul 31 21:03:45 in what sense do they "seem odd" to you? apart from the fact the am572x has iosets in the first place (you'd think that iodelay would make iosets superfluous) Jul 31 21:09:01 zmatt: agreed re: iodelay, but I think something still comes down to testing. Jul 31 21:09:51 * jkridner pulls up https://dev.ti.com/pinmux to try to find an example. Jul 31 21:13:11 what I recall is that for a peripheral without an IO set, selecting one of the pins still limited my ability to select some others. Jul 31 21:13:42 I'm wondering now if I was confusing myself with pins that were actually part of a different peripheral, since those show up in the pull-down. Jul 31 21:13:50 er, different peripheral instance. Jul 31 21:15:50 looking at mcasp, that's probably what it was.... I didn't have it locked on a particular peripheral instance. Jul 31 21:16:26 the pins without gpios completely baffle me Jul 31 21:17:52 pins without gpios? looks like that's just some very special-purpose ones Jul 31 21:18:13 never mind Jul 31 21:18:30 interesting, I don't think I ever actively noticed that Jul 31 21:19:13 lol, some have no GPIO mode, but do have PRU GPIO modes, wtf Jul 31 21:25:56 yeah weird those two seemingly random blocks of pins (189-194, 203-212) with no GPIO mode... those pins seem completely unremarkable otherwise. maybe they just didn't want to add a 9th gpio controller? Jul 31 21:29:41 looks like they left a few gpios unused though Jul 31 21:37:52 jkridner[m]: yeah I think they just didn't want to add yet another gpio controller... it's not like you're very likely to run out of gpios on it already :P Jul 31 21:53:38 jkridner[m]: lol... iosets for an uart... I think TI must be confused about the "A" in "UART" Jul 31 21:55:23 hah, iosets containing a single pin each... I guess a set of one element is still a set ;) Jul 31 23:09:47 jkridner[m]: doesn't seem there are that many things affected by iosets on the am572x tab though... I added an IOSets tab to my Vayu Pins spreadsheet ( https://goo.gl/jiazTL ) based the json, with the most obvious nonsense filtered out Jul 31 23:11:58 still playing a bit with the layout for readability Jul 31 23:12:58 eh, it'll have to do, time to head home.. bbl **** ENDING LOGGING AT Thu Aug 01 03:00:25 2019