**** BEGIN LOGGING AT Sat May 02 02:59:59 2020 May 02 03:00:50 He wrote all kinds of books for trash shack May 02 03:00:58 Oh. May 02 03:01:02 Who is trash shack? May 02 03:01:06 Oh. Radio Shack? May 02 03:01:58 I remember starting in TTL chips reading his book and burnt the chips to a crisp May 02 03:02:05 Ha. May 02 03:02:17 "You should have known." May 02 03:02:57 While I'm in a typing mode any other changes? No this was late 60's May 02 03:03:31 Oh. May 02 03:03:41 The beggining of the times! May 02 03:04:09 Typing mode = KenUnix = NO! May 02 03:04:26 KenUnix: Yes! May 02 03:04:33 There are a ton of capes! May 02 03:04:38 First there was DTL then ECL then TTL then CMOS May 02 03:04:44 Look at GHI or were you told otherwise? May 02 03:05:11 I never once heard of DTL. My historical references still suffer. May 02 03:05:29 Let me show you. May 02 03:05:57 https://beagleboard.org/capes May 02 03:05:59 Diode Transistor Logic, ECL Emitter Coupled Logic May 02 03:06:05 There are the official capes from what I understood. May 02 03:06:09 Oh. May 02 03:06:12 Thank you, sir. May 02 03:06:49 One of the bagging companies was Fairchild May 02 03:07:00 One of the begging companies was Fairchild May 02 03:07:22 Things change but those Capes "work" mostly. I got the MotorCape, "RelayCape," and ServoCape. The ServoCape is my arch nemisis right now. May 02 03:07:46 So, I am engulfing time in building my own circuit of servo infrastructure. May 02 03:08:16 KenUnix: Fairchild was big biz. I think they still are these days. May 02 03:08:44 bagging/begging = beginning? May 02 03:11:14 They had one of the early video sames, used cartridges, had a Fairchild CPU May 02 03:12:31 I thought I did capes??? Where are there on elinux? May 02 03:13:13 They had one of the early video sames, used cartridges, had a Fairchild CPU . The F8 May 02 03:15:27 Oh. May 02 03:15:42 elinux.org did not get the GHI cape. May 02 03:15:46 capes. May 02 03:16:05 You just had one on the elinux.org page, i.e. the RelayCape. May 02 03:16:14 i will go and check again. May 02 03:17:53 Too many pages. Lost my train of thought May 02 03:17:58 They just keep spinnin'. May 02 03:18:00 Okay. May 02 03:18:20 I tried to click on the RelayCape link but it just kept spinnin'. May 02 03:18:30 Got the elinux url? May 02 03:18:35 Yes. May 02 03:18:39 Do you want it? May 02 03:19:12 https://elinux.org/BeagleBone_Community#Capes is you need it oh-so-badly. Ha. May 02 03:19:16 yes May 02 03:19:45 Oh. You mean the bbb.io/capes link? May 02 03:20:04 this one? https://beagleboard.org/capes May 02 03:20:06 I got lots of capes there May 02 03:20:23 Yes but why not put them on elinux.org too? May 02 03:21:22 I thought that was what this whole fiasco was over. May 02 03:21:34 "Putting things on elinux.org!" May 02 03:21:53 There are 18 capes May 02 03:22:17 Fiasco ? May 02 03:24:16 Yes. It is not easy to grab content and place it in other locations. Even C & P gets mundane. May 02 03:24:24 I am just worried that I cannot take you the entire way. May 02 03:24:50 There are ideas relative to PRUs and .dts files that I cannot handle yet. May 02 03:25:26 Only 18? May 02 03:25:47 I think there are more. Let me check. What about the Cape documentation on elinux.org? May 02 03:26:29 https://elinux.org/Beagleboard:BeagleBone_Capes <<< Yikes! Talk about some work. May 02 03:27:18 Wow, thanks May 02 03:31:45 YOu are welcome. May 02 03:32:19 Yep. I noticed that years ago but never really took time to look over the capes in it. It was like I was scared to get one. May 02 03:32:54 BBB Capatible <<< that made me more aware of what was offered then. May 02 03:34:42 Tomorrow . Booked those pages.. Eyes tired. Gotta go. May 02 03:35:27 Tomorrow . Bookmarked those pages.. Eyes tired. Gotta go.=( =( May 02 03:36:00 Okay. May 02 03:36:02 Have fun. May 02 03:36:38 Check bwbasic May 02 03:50:47 kenunix I didn't do much but you are welcome (bwbasic assist) May 02 03:52:16 GenTooMan! May 02 04:12:35 There was someone learning or trying to use qmake w/ qt creator, I think. Anyway, I found a book but it is four years old. It used qt5 and C++14. May 02 04:13:01 If that person is still having an issue, you can try to use me if needed. May 02 04:13:17 set_ yes I am alive for now. May 02 04:13:27 GenTooMan: Hello! May 02 04:13:43 Are you using qt5 and C++14? May 02 04:14:01 w qmake instead of cmake? May 02 04:15:24 anyway...if that person or you are having issues w/ compilation, try to add C++xx to the qmake file. I will show you. May 02 04:19:29 https://pastebin.com/KNtR8RBu will show you an easy set up for a window in qt5 and C++14. May 02 04:19:46 I will get the source. Please hold. May 02 04:23:20 https://pastebin.com/k01DqYUt is the short source for making a simple window. May 02 04:24:11 no I am not use qt5 nor C++14. May 02 04:24:17 Okay. May 02 04:24:24 I guess it was someone else. May 02 04:24:27 have not used... uh... brain is slowing down. May 02 04:24:32 Yea! May 02 04:24:45 Coffee GenTooMan, coffee! May 02 04:28:54 I've been up 19 hours... May 02 04:29:24 damn. May 02 04:29:29 Godspeed. May 02 04:29:38 maybe? I need to get regular sleep not what I've been doing :D May 02 04:29:50 Oh. Smart. Get sleep, my bretheren. May 02 04:30:11 All who sit shall sleep well. May 02 06:06:01 Sleep... May 02 07:20:58 set_: I'm not sure what you mean exactly, but in my case I did not upgrade (in the sense of apt upgrade), but I just reflashed a new image from SD. May 02 07:22:52 zmatt: Thanks for the tip about bb-customizations. I'm a bit hesitant to upgrade things like that, since these are embedded device that I do not have easy physical access to, and some of them only have a 3G connection, so that makes them a bit fragile. May 02 07:24:00 blathijs: I mean, if you "reflashed a new image from SD" then you'll have a sufficiently recent bb-customizations May 02 07:24:15 assuming you mean the 2020-04-06 image May 02 07:25:36 and for devices "out in the field" I'd definitely keep a local mirror device for testing updates/modifications before deploying them to the hard-to-access production device May 02 07:27:09 oh you do mean the 2020 image since that's why you noticed that the spidev numbering changed from wrong to correct May 02 07:28:45 of course if you need to modify software to use the /dev/spi/ symlinks then you may as well just change it to use the new (correct) numbering, unless you also need to support older kernels that have the off-by-one error in spi bus numbering May 02 07:32:30 I'm also a bit puzzled about this situation... you've reflashed a device to a new image and then put it in a hard to access location without first testing that your software works on this new image? May 02 07:34:58 because... that does not sound like a smart thing to do May 02 07:40:17 zmatt: Thanks for clarifying. I do have a local mirror device, and there I reflashed and noticed the SPI numbering change. But I indeed have older devices to support, ideally with the same configuration, so I've now updated my configuration to use /dev/spi/0.0 if available, and fall back to /dev/spidev1.0 otherwise (which assumes that /dev/spi and the numbering change happened at the same time, which might May 02 07:40:19 not be true, but close enough for the image versions I'm running). May 02 07:40:50 zmatt: As for bb-customizations, I though you suggested updating just that one package on existing devices, which I'm not too fond of because they are hard to reach. May 02 07:41:23 presence of /dev/spi/* depends on udev rules the are installed as part of the bb-customizations packages, the spidev numbering depends on kernel version (or more specifically the device tree) May 02 07:41:38 Yeah, I gathered. May 02 07:41:38 you could just manually install the specific udev rules file May 02 07:42:23 zmatt: But you'd also need devicetree customizations, not just udev, right? Or is the udev file responsible for the spidev name change? May 02 07:43:07 oh actually I thought it was purely based on an udev rule but now that I look at it seems to simply use 10-of-symlink.rules I think May 02 07:43:15 i.e. the actual symlink name comes from DT May 02 07:43:23 yeah exactly May 02 07:43:26 zmatt: But you'd also need devicetree customizations, not just udev, right? Or is the udev file responsible for the spidev name change? May 02 07:43:36 w00ps, double post May 02 07:43:40 zmatt: You say "the new (correct) numbering", does that mean that /dev/spidevx.x can be considered stable (and now correct), but was just bugged before? If so, what is the point of the /dev/spi symlinks? Did those predate the fixed spidev names? May 02 07:44:14 yeah they predated them May 02 07:46:19 In any case, I think I'm good now, I'm not just using the symlinks when they exist, and falling back to the old spidev numbering otherwise, which seems to work across all my devices. May 02 07:46:34 actually it's not the main DT that creates the symlinks, it's the universal overlay May 02 07:46:50 i.e. it depends on the bb.org-overlays package May 02 07:48:07 added more than a year ago: https://github.com/beagleboard/bb.org-overlays/commit/d21b116ebee9 May 02 07:49:28 zmatt: Ok. Thanks for clarifying, I'm off now! May 02 08:10:31 Hi! After updating the distribution I can not configure pins. May 02 08:11:32 ? May 02 08:11:49 I'm pretty sure nothing has changed about how pin configuration works May 02 08:12:02 sorry, not all message was sent May 02 08:12:35 "ERROR: write() to /sys/devices/platform/ocp/ocp:P9_11_pinmux/state failed, No such device" May 02 08:12:56 Debian 10.3 Buster IoT Image 2020-04-27 May 02 08:13:07 What could be the problem? May 02 08:13:26 dunno, I have that exact image freshly installed on a beaglebone here and I can configure P9_11 just fine May 02 08:13:54 did you reflash eMMC or are you booting from SD card? May 02 08:14:10 SD card May 02 08:14:35 installed yesterday May 02 08:14:52 in that case I recommend either reflashing eMMC or, if for some reason you want to continue running from SD card instead, wiping eMMC using "sudo blkdiscard /dev/mmcblk1" May 02 08:15:34 since an old bootloader on eMMC in combination with a new image on SD card can cause all sorts of problems (due to the old bootloader not being able to understand modern directives in /boot/uEnv.txt) May 02 08:15:55 if eMMC (or at least the bootloader on it) is wiped then it'll use the bootloader on SD card instead May 02 08:16:07 (and reflashing eMMC will also replace the bootloader) May 02 08:17:01 thanks, I'll try May 02 09:30:12 I transferred the image to eMMC, but the problem has not changed May 02 09:34:33 bone:~$ config-pin P9_11 pruout ERROR: write() to /sys/devices/platform/ocp/ocp:P9_11_pinmux/state failed, No such device May 02 10:02:33 Kiba25: uhh, and you didn't change anything in /boot/uEnv.txt (since you downloaded the image and installed it to SD card) May 02 10:03:39 can you run sudo /opt/scripts/tools/version.sh and share its output (via some paste service like pastebin.com, never paste multiline output directly into chat) May 02 10:14:30 Thanks for the tip, https://pastebin.com/wLYsLMcG May 02 10:16:42 I just disabled hdmi May 02 10:23:42 oh wait I misread the you originally got... ENODEV (No such device) means the pinmux state you tried to set doesn't exist for that pin May 02 10:24:09 which is true, P9_11 does not have a pruout mode May 02 10:24:29 so this isn't a new error (like you implies), it cannot have worked in any image May 02 10:24:35 *implied May 02 10:31:55 I tried just "out" and tried other pins May 02 10:32:01 however, now I tried "config-pin P9_30 pruout" and it seems to work May 02 10:32:19 "out" is not a pinmux mode and will also never have worked May 02 10:32:32 "Current mode for P9_30 is: pruout" is this normal? May 02 10:32:49 yes I don't know why it prints that either, seems rather spammy to me May 02 10:32:59 but I assume that hasn't changed with this new image either May 02 10:41:54 In general, I updated the image due to the fact that the PRU program did not have access to the port. I ran the command "config-pin P9_30 PRUOUT" and did not receive a response, and still could not get access (change high to low on any pin). Now I got this answer, but have not tested the PRU program yet May 02 10:43:48 pinmux states are case-sensitive, so it's "pruout", not "PRUOUT" May 02 10:44:14 use e.g. config-pin -l P9_30 to list valid modes for a pin May 02 10:47:28 Thanks a lot, it helped) May 02 10:59:17 Kiba25: and just in case you're not sure which pin is what for pru: https://pastebin.com/zK85mXex May 02 10:59:36 (extracted from my detailed pins spreadsheet, https://goo.gl/Jkcg0w ) May 03 00:36:44 hi May 03 00:37:39 does anyone know how to resolve this issue, or is windows 7 just not supported on beagleboard black w/ latest image: https://github.com/RobertCNelson/bb-kernel/issues/49 May 03 00:38:27 usb on windows is just endless misery May 03 00:39:12 microsoft can't be fucked to implement any standard, and it's even a giant chore to convince windows to use its own fucking driver for microsoft's own proprietary usb networking protocol May 03 00:39:49 if at all possible just connect via ethernet instead of usb-networking May 03 00:41:36 oh actually, this is about the serial gadget May 03 00:41:39 which you don't need at all May 03 00:42:59 as long as you can connect via networking (either usb-networking or ethernet) you're fine May 03 00:43:47 really the only reason I can think of to ever use the usb serial gadget is if you need to debug networking issues and can't reach the bbb via networking May 03 00:55:15 dizzy0ny: you should really try to migrate away from Windows 7 anyway since microsoft has ceased to support it and no longer even provides security updates May 03 01:07:04 win 10 ... a throw back to the bad old days of computing when you didn't own anything and were completely at the mercy of the company who made the computer. May 03 01:29:08 Anybody home ??=D =D May 03 01:31:27 kenunix I am at home May 03 01:33:44 I loaded Ubuntu 20.04 on my laptop. Three strange problems (1 gcc -Os don't work) (2 shell sleep fails cannot read realtime clock) and (3 sudo cat abc >> def fails. Root owns abc and def. Appending to the end of def. Used to work. Says Permission denied. Used to work} =( =( May 03 01:35:00 GenTooMan I know off topic. I'm frustrated X) May 03 01:35:58 GenToo - Wonder if Ubuntu has an IRC?X-P May 03 01:40:01 KenUnix: what on earth is going on on your system May 03 01:40:30 kenunix not likely it would be swamped with whinging people... May 03 01:41:05 KenUnix: "sudo cat abc >>def" does not work as you want it to, and never will have May 03 01:41:29 you're redirecting the output of "sudo cat abc", which means root privileges only apply to reading abc, not to appending to def May 03 01:42:14 so you're wrong that it "used to work"... it only could have worked if sudo were unnecessary in the first place May 03 01:42:19 That's how I used to append to the end of a root owned file May 03 01:42:25 no, you didn't May 03 01:43:00 that would not have worked 2 decades ago and it will not work today May 03 01:43:40 ^ agreed May 03 01:43:46 OK then how should I append to a root owned file? May 03 01:44:32 piping data into sudo tee -a FILE is one way May 03 01:45:00 or just applying sudo to a small utility that performs the task May 03 01:45:16 pipes and redirection are quite precise arts :) May 03 01:45:46 and easy to get wrong, and pseudo-right by mistake May 03 01:46:55 didn't work, cat abc | sudo tee -a def May 03 01:47:11 define "didn't work" May 03 01:47:46 because that most definitely will append the contents of the file "abc" to the file "def" with root privileges May 03 01:48:04 (it's also an example of "useless use of cat" since you can just write "sudo tee -a def @zmatt i understand...i have a new laptop ordered, so i'll be off of windows 7...but they don't have a delivery date of when they will get in stock! May 03 01:48:56 dizzy0ny: ok. anyway, like I said whether the usb serial gadget works or not isn't that important anyway May 03 01:49:56 in the mean time, i was hoping to get a jump start using this beaglebone black that has been sitting around collecting dust...that's why i ask if there is a way to connect it via usb on windows 7 (after flashing with new image) May 03 01:50:17 OK It works but I also see the scouce file May 03 01:50:23 dizzy0ny: so, does the rndis driver also fail to work? May 03 01:50:37 KenUnix: ??? May 03 01:50:51 rndis ? May 03 01:50:56 usb networking May 03 01:51:02 yes May 03 01:51:08 well that's the only thing you need from usb May 03 01:51:37 After doing this cat aaa | sudo tee -a bbb I see file contents while it's working May 03 01:52:11 KenUnix: yeah, just redirect the output of tee to /dev/null to suppress that May 03 01:52:29 wait May 03 01:52:57 @zmatt, if i remove the SD card w/ latest image and connect it with the old firmware, usb connects fine. Im not sure how i will push code over if i don't ave usb? there is a wifi adapter plugged into it, but don't see it showing up on the network either May 03 01:53:04 tee writes its stdin to stdout *and* writes (or, with the -a option, appends) to file May 03 01:53:28 dizzy0ny: so you're saying usb networking *isn't* working? May 03 01:53:33 That worked cat aaa | sudo tee -a bbb >/dev/null May 03 01:54:00 zmatt, that's right...it apparently is known issue with windows 7 and later firmware May 03 01:54:07 dizzy0ny: ah my bad, I misread what you said earlier May 03 01:54:24 https://github.com/RobertCNelson/bb-kernel/issues/49 May 03 01:54:49 that github issue isn't about usb networking though but about usb serial, which is not relevant May 03 01:57:02 like I said earlier, if you can connect it via Ethernet or wifi use that. I think if the wifi adapter is recognized by the new image it *might* start an access point on it automatically, but I'm not sure about that May 03 01:57:21 hmm..ok i may be confused here.. May 03 01:57:51 What WiF adapter is he using? May 03 01:58:07 Ok i do see the rndis driver installed May 03 01:58:18 and i see it showing up as my 'E:" drive May 03 01:58:36 the drive showing up is usb mass storage, that part should always work May 03 01:59:24 He should be able usinf putty to connect to virtual com port May 03 01:59:38 KenUnix: no, that part won't work May 03 01:59:55 @zmatt that doesn't work when i have the SD card in w/ new firmware (i hold user/boot button, and power it us w/ the SD card inserted). May 03 02:00:07 dizzy0ny: what doesn't? May 03 02:00:17 there is no E drive w/ new firmware May 03 02:00:32 oh did they finally remove it? I assumed it would still be there May 03 02:00:38 what i see in Device Manager instead are... May 03 02:00:43 but maybe they dropped it since the drivers are long irrelevant May 03 02:01:23 hmm no I think it should still show up as mass storage May 03 02:01:24 'CDC ECM', 'CDC Serial', 'Gadget Serial'..in all cases it is unable to install/find drivers May 03 02:01:58 just to verify, which image are you using? May 03 02:02:32 okay I guess the "CDC Serial" might actually be RNDIS? maybe windows is just really confused May 03 02:02:40 there's definitely only a single serial interface May 03 02:02:58 AM3358 Debian 10.3 2020-04-06 4GB SD IoT May 03 02:03:06 ECM is the usb standard for networking, which microsoft doesn't support because fuck standards May 03 02:03:55 @zmatt RNDIS is there and is installed. perhaps with the latest version, the usb is not considered an RNDIS device May 03 02:04:17 I mean, they try to do everything to tell Windows that it is in fact RNDIS May 03 02:04:51 but Windows is too stupid to manage loading its own driver for microsoft's own networking protocol May 03 02:05:03 not sure..but this forced me to end up just buying a Raspberry PI 4. I guess whenever that laptop with windows 10 gets here (if ever), i will try this again May 03 02:05:09 at least it used to be, I *think* they got it working on more recent windows versions May 03 02:05:40 Thanks for you help..and trying May 03 02:06:19 I think the problem is in part due to exposing a composite device... but not doing so means not being able support Mac, which properly implements the USB standard but obviously not microsoft's proprietary RNDIS May 03 02:07:11 How does a raspberry or other micro controllers solve this problem? or do they have the same issue? May 03 02:08:15 there's nothing beaglebone-specific about this problem, we've had exactly the same problems with trying to get our microcontroller device understood as usb device by windows May 03 02:09:07 I guess the rpi foundation just has more resources available to get this fixed one way or the other, possibly by just paying microsoft regularly to get "signed" drivers (note: said "driver" is just a text file telling windows to use its own driver for your device) May 03 02:09:30 what port though if the device is not being recognized? May 03 02:09:37 ? May 03 02:09:44 oops..wrong window May 03 02:10:12 note it's not about the device as a whole being recognized, normally anyway, it's about each individual interface May 03 02:10:28 then again, it's Windows with its broken usb stack, so who knows May 03 02:10:39 understood May 03 02:12:44 FYI ..looks like RobertCNelson responded on that issue at the GitHub link...guess i will just come back to this when i have windows 10 May 03 02:13:00 yes but again, that issue isn't your actual issue May 03 02:13:14 acm never worked on windows 7 May 03 02:13:18 but you don't need it May 03 02:14:24 i just had an idea...if i were to say run ubuntu vai VirtualBox..i wonder if it would be able to connect then? or will still have to proxy through the windows driver (which isn't installed0 May 03 02:15:26 if it shows up as usb device at all (i.e. if virtualbox has usb pass-through) then it should work yes May 03 02:15:26 We'll I'm using Windows 10 and it works May 03 02:15:41 KenUnix: yes but this is windows 7 May 03 02:15:58 hmm..i may try that May 03 02:17:41 The latest verison of virtual box supports usb pass-through with the free extension. May 03 02:19:06 Sorry I only have Winbloes on a VM May 03 02:19:30 Sorry I only have Winbloes 7 on a VM May 03 02:35:10 OK I fired up Win 7 installed putty and can ssh to bbb May 03 02:40:34 OK I installed winscp and it works **** ENDING LOGGING AT Sun May 03 02:59:57 2020