**** BEGIN LOGGING AT Wed Jun 27 03:00:02 2018 Jun 27 03:00:17 How dare you.. Jun 27 03:00:38 Oops. Jun 27 03:00:57 CoffeeBreakfast: What book are you using? Jun 27 03:02:46 Minix, by Tanenbaum Jun 27 03:02:57 Oh. Jun 27 03:03:03 Sorry. My fault. Jun 27 03:44:48 (Actually, Minix is not the name of the book) Jun 27 03:44:57 Oh? Jun 27 05:13:42 tdy Jun 27 07:24:47 Hi Jun 27 07:29:07 Hi Jun 27 09:15:45 Hi, I have a question regarding sd card Jun 27 09:16:07 the uboot loads kernel from sd card and put it into ram Jun 27 09:16:33 the ram address we refer as entry point and loadaddr. Jun 27 09:16:46 but what about kernel storage on sd card. Jun 27 09:17:06 How uboot gets to know where the kernel is stored? which address on sd card? Jun 27 12:48:14 jkridner, I am attempting to boot archlinux on my pocketbeagle and it is a no go currently. I am going t assume it comes down to uboot. I saw that there was a patch in march but not sure where to look to see if that is the current stable version for me to pull and compile Jun 27 12:48:32 also can this be crosscompiled from a amd64 machine Jun 27 13:25:32 mistawright: I'm sure it can be cross compiled. Do you have output from the debug UART? Jun 27 13:28:53 unfortunately i dont Jun 27 13:30:15 tbr, i also noticed that compiling uboot for 2018.05 and using dd to write the uboot it booted but only the two bottom leds came on and they blink so I don't know if thats archlinux-arm's way of booting said device. I get some rndis devices in dmesg but nothing apart from that Jun 27 13:35:36 mistawright: got serial logs? Jun 27 13:35:52 mistawright: stop everything you are doing and get a usb-serial device, NOW. Jun 27 13:36:07 hello i have query Jun 27 13:36:07 pocketbeagle works great with USB UART Click. Jun 27 13:36:25 mistawright: doing any sort of boot problem debugging and such without UART access is madness Jun 27 13:36:31 help me anyone Jun 27 13:36:41 Dixith: you have not stated your problem Jun 27 13:37:16 tbr, i need to find my usb serial dongle thing but its at home. what pins would I end up using for this as this isnt a beagle bone black with the inner serial pins Jun 27 13:37:44 We are facing problem when booting the BBB Jun 27 13:38:07 the leds are not flashing after holding the s2 buttons for severaal minutes Jun 27 13:39:01 FYI: we have disconnected the BBB several times without power off.. Does it corrut the board? Jun 27 13:39:17 corrupt* Jun 27 13:39:26 Dixith: what happens if you don't hold down S2 and connect power? Jun 27 13:40:04 the led 0 is blinking continously Jun 27 13:40:29 ok, then you didn't prepare your SD-card correctly. What did you do? Jun 27 13:40:36 mistawright: the ROM code will send RNDIS. Jun 27 13:41:03 mistawright: you can try using my node-beagle-boot netconsole hack to get some u-boot console output. Jun 27 13:41:34 mistawright: you'll miss out on the kernel output though for now. Jun 27 13:41:57 we have changed the .xz debain file to .img the nwith win32imager we have wrote on the sd card Jun 27 13:42:18 also, we did format before doing this Jun 27 13:42:22 Dixith: you just renamed the file or what did you do? Jun 27 13:42:35 mistawright: https://www.youtube.com/watch?v=9CxfKfjykz4 Jun 27 13:42:59 nope I used the 7zip Jun 27 13:43:02 and changed it Jun 27 13:43:12 ok Jun 27 13:43:19 so you unpacked it? Jun 27 13:43:26 yes Jun 27 13:43:30 https://elciotna18.sched.com/event/DXmz/bring-up-with-netconsole-and-usb-without-a-serial-debug-net-jason-kridner-texas-instruments-and-beagleboardorg-foundation Jun 27 13:43:37 for 2-3 times I was able to see the splash screen but all of suddent Jun 27 13:43:43 sudden* Jun 27 13:43:44 wish they'd link the slides with the video. Jun 27 13:44:04 Dixith: so the SD-card worked before? Jun 27 13:44:16 yes Jun 27 13:44:32 it stoped working Jun 27 13:45:15 I'd recommend to rewrite the SD-card using win32imager Jun 27 13:46:10 done atleast 4-5 times.. Jun 27 13:46:16 but nothing changed# Jun 27 13:47:09 try a different card Jun 27 13:47:34 ohh.. are you sure? Jun 27 13:48:04 I'm not saying the card is broken. I'm saying it might be worth to try with a second card. Jun 27 13:48:21 anyway, I have to leave now, but I'm sure jkridner and the others can help you further Jun 27 13:49:02 ökay thank you Jun 27 13:49:41 @jkrindner are you there? Jun 27 13:54:14 "after holding the s2 buttons for several minutes" --- don't do that! in the rare situation that the S2 button is needed/useful, it is used by: 1. press down the S2 button while the board is unpowered 2. connect power 3. release the S2 button Jun 27 13:55:07 there's no point in holding the S2 button after the power led has turned on, and I don't think the hw appreciates it if you press it once the video output has been enabled Jun 27 13:57:09 oh, you left Jun 27 14:12:13 why the hell does the S2 button have a mere 100 ohm series resistor anyway... if the video output is enabled and lcd_data2 is driven high, that results in 33 mA of current when you press the S2 button, which I think could definitely damage the hardware Jun 27 14:14:49 s/definitely/possibly/ .. I'm not actually how easily that happens, but it definitely strains the hardware Jun 27 16:01:39 Dixith: I am, but you re not. Jun 27 16:02:45 zmatt: good question.... agree that it should be higher. Jun 27 16:03:32 jkridner: 680 ohm suffices to keep the current below 6 mA Jun 27 16:06:05 hmm, and I should really pick up my little project again of generating V/I curves from the AM335x IBIS data. step 1 probably being posting on E2E to complain about some of the obvious garbage that's in the IBIS file :-/ (missing data, values that blatantly contradict both the datasheet and reality, etc) Jun 27 16:10:31 or, well, step 0 being to check if there's updated IBIS data that fixes the issue I guess Jun 27 21:37:25 anyone knows how to configure a regular user on a beagleboard x15 to be able to use the opencl DSP processor? I can use it as root, but don't know how to setup the permissions of a regular user to access it. Jun 27 22:00:32 usually such things can be accomplished, but I haven't played with the DSP on the x15 yet and don't know how interaction with it is typically handled Jun 27 22:04:00 keep in mind code on an auxiliary core typically has access to all of RAM (and indeed most of the SoC), hence a user able to load code onto auxiliary cores is effectively root from a security perspective Jun 27 22:05:45 (but I appreciate that using a non-root user may still be desirable for various reasons, and it may be possible to provide adequate isolation using the L3 and L4 firewalls) **** ENDING LOGGING AT Thu Jun 28 03:00:02 2018