| agneli | boot_scripts=boot.scr.uimg boot.scr | 06:47 |
|---|---|---|
| agneli | this variable in your u-boot conf you need to change to get gnartest.scr to be recognized | 06:48 |
| agneli | for example: setenv boot_scripts gnartest.scr boot.scr yyhaaa.ext etc0.etc | 06:50 |
| agneli | then saveenv and next reboot u-boot will be looking for the listed files exactly in that order | 06:51 |
| agneli | do you have ls comand compiled in u-boot? | 06:56 |
| agneli | ls mmc 0:1 / | 06:56 |
| agneli | should list your current boot command | 06:56 |
| agneli | ls mmc 0:2 / | 06:56 |
| agneli | should list your current / partition | 06:56 |
| agneli | *should list your current /boot partition :) | 06:56 |
| agneli | also this command could come useful https://u-boot.readthedocs.io/en/latest/usage/cmd/part.html | 07:00 |
| agneli | you can really really check what partitions are visible there by their UUIDs | 07:00 |
| agneli | this prints my pt on my emmc: part list mmc 0 | 07:01 |
| agneli | part list mmc 1 | 07:02 |
| agneli | MMC: no card present | 07:02 |
| agneli | this says my sdcard is not preset :) | 07:02 |
| agneli | there are very nice examples at the end of the part.html document shared above | 07:03 |
| agneli | so you can really really list and read all partitions in your *.scr boot script to check everything | 07:04 |
| agneli | there must be a logical reason why the card behaves like it behaves | 07:04 |
| agneli | it is not windows to have some non-deterministic behavior | 07:05 |
| agneli | and from what you are saying - you are able to mount that very same card / partition on that very same port/device (pb laptop) while booted from emmc | 07:06 |
| agneli | the card is OK and nothing is alergic | 07:06 |
| agneli | there is some very simple configuration bug to be resolved | 07:06 |
| agneli | just toy for the 15 minutes with the above commands and that is all it is needed. and I am still saying root=UUID is the fast and easy solution | 07:07 |
| agneli | all the above commands I tested on my device's u-boot shell | 07:08 |
| agneli | my uboot is a mainile from earlier this year with some tegra2 specific patches I was testing, that were supposed to be mainlined | 07:10 |
| agneli | gnarface: why do not you get latest official mainline u-boot? some patches? | 07:11 |
| agneli | also I think it is time you kindly share some outputs, words are just words, can have different meaning for different ppl, chances are I do not understand what you are saying in the way you want me to understand... | 07:30 |
| agneli | :) | 07:30 |
| agneli | did I miss anything interesting? :) | 20:01 |
| onefang | No one said anything since you left. | 20:03 |
| agneli | ;) | 20:03 |
| agneli | thx | 20:03 |
| gnarface | agneli: i left badblocks running on it through last night. no errors. i'm copying the install to it again right now. | 20:03 |
| agneli | I am telling you our card is OK | 20:04 |
| agneli | and if you want to really check use f3 nothing else | 20:04 |
| gnarface | well, i had to be sure | 20:04 |
| agneli | it will give you everything icnluding coctail at the end of the check | 20:04 |
| gnarface | but i think now the only explanation has to be that the card is too slow to work | 20:05 |
| agneli | nah, it would not mount | 20:05 |
| gnarface | i think with this hardware it might be different | 20:06 |
| agneli | I encourage you to use commands I provided in the morning | 20:06 |
| gnarface | due to poor controller design or something | 20:06 |
| agneli | I tested them all on my device | 20:06 |
| gnarface | i did try all those, nothing unexpected returned | 20:06 |
| gnarface | thank you for the references by the way | 20:06 |
| agneli | so from uboot shell you are able to list files on the toor partition of the card in question | 20:06 |
| agneli | is that correct understanding? | 20:06 |
| gnarface | both cards, yes. and their boot.scr files were different in size so i knew they were actually different | 20:07 |
| agneli | ls mmc X:Y / ? | 20:07 |
| gnarface | yea | 20:07 |
| agneli | so card works 100% | 20:07 |
| agneli | root=UUID :) | 20:07 |
| gnarface | in answer to your question about why i haven't actually pasted output though, you should know that i have to disassemble the stupid thing to re-enable the serial console, and as that causes a lot of wear on the plastic housing i'm reluctant to do it except as an emergency measure | 20:08 |
| agneli | ach crap | 20:08 |
| agneli | excuse me | 20:08 |
| agneli | i keep forgettin | 20:08 |
| gnarface | see, it actually shares the audio jack with the serial connector, and i have to toggle an internal dip switch to change modes (this is one of the very early models) | 20:08 |
| agneli | I actually had a laptop serive shop to solder ma a connector | 20:09 |
| agneli | and I have all pulled out outside | 20:09 |
| agneli | so for me it is very easy | 20:09 |
| gnarface | i have considered drilling a hole in the case where the switch is, but i don't want it to be easier for dust to get in there | 20:09 |
| agneli | in fact I have serial connected as a sort of dockstation :) | 20:09 |
| agneli | i used kensington lock hole to pull the tinnest eth cable I could find | 20:10 |
| gnarface | heh, clever | 20:10 |
| gnarface | no kensington lock holes on these things though | 20:10 |
| agneli | I have ethernet connector sticking out with all singals provided by debug prot | 20:10 |
| agneli | serial inluded | 20:10 |
| agneli | so excuse me for that again | 20:11 |
| agneli | anway - whatever we are saying is just pointing at one direction - UUID, something strange happens | 20:11 |
| agneli | when kernel boots | 20:12 |
| gnarface | if this doesn't work i think i might try the same copy method to a 16GB microSD of the same brand, and if that works i'll just consider the 64GB one haunted | 20:12 |
| agneli | btw | 20:12 |
| agneli | when kernel boots | 20:12 |
| agneli | you have all possible logging enabled to the console, I mean lcd/panel | 20:12 |
| agneli | does it detect everything as it should? | 20:12 |
| gnarface | yea, i'm not passing "quiet" so a bunch of text goes by | 20:13 |
| gnarface | it detects the display fine | 20:13 |
| agneli | put the highest level possible | 20:13 |
| agneli | what I have, idr | 20:13 |
| agneli | 7? | 20:13 |
| gnarface | but it quickly kernel panics then starts a 10 second counter to reboot | 20:13 |
| agneli | I am talking about mmc devices | 20:13 |
| agneli | panics? | 20:14 |
| agneli | i was sure it was init not found | 20:14 |
| gnarface | yes, it says init not found then panics | 20:14 |
| gnarface | or maybe it panics then says init not found | 20:14 |
| agneli | ach | 20:14 |
| gnarface | i think the "init not found" part is an indirect result of the panic, not the cause | 20:15 |
| gnarface | there's a bunch of crap that looks like a dump from a read error or something too | 20:15 |
| agneli | f3 | 20:15 |
| agneli | please do run full test with f3 | 20:16 |
| gnarface | noted, but badblocks found no bad block even with a destructive read+write test | 20:16 |
| agneli | i am 100% sure you have some flaky cells | 20:16 |
| agneli | i doubt badblocks will find those | 20:16 |
| agneli | I _never_ has those issues | 20:16 |
| agneli | but my colleague actually just boght a pbp and aliexpress card | 20:16 |
| agneli | and that is exactly the case | 20:17 |
| agneli | 3 flaky cells | 20:17 |
| gnarface | hmm | 20:17 |
| agneli | somtimes ok sometimes not | 20:17 |
| agneli | f3 | 20:17 |
| agneli | idk why ppl are prefering f3 over badblocks | 20:17 |
| agneli | maybe it does sth differenlty | 20:17 |
| agneli | idnot care | 20:17 |
| agneli | but i trust they wisdom and they say f3 :) | 20:18 |
| agneli | https://packages.debian.org/en/stable/utils/f3 | 20:18 |
| agneli | unless it panics because init not found | 20:19 |
| agneli | then root=UUID :) | 20:20 |
| gnarface | agneli: heh, you're gonna love this one. i used a different brand microSD->SD adapter in the USB->SD adapter for this write, and it boots now | 21:29 |
| gnarface | so, the microSD is not defective, but the adapter that came with it is | 21:29 |
| agneli | you never checksummed the result? | 21:29 |
| gnarface | no, because i figured dd would throw errors! | 21:30 |
| agneli | ach it was not a direct copy | 21:30 |
| agneli | time to sleep | 21:30 |
| agneli | so it was one of those errors that well... are hard to spot :) | 21:30 |
| gnarface | seems that way | 21:31 |
| agneli | but yeah | 21:31 |
| agneli | you counquered it | 21:31 |
| agneli | congratulations | 21:31 |
| gnarface | i thought i'd already tried this, but on the previous attempt i must have switched back and forth interchangeably | 21:31 |
| gnarface | thanks for helping | 21:31 |
| agneli | it was my pleasure gnarface | 21:31 |
| agneli | it really was | 21:32 |
| gnarface | i would have given up on it if not for your persistence. i really appreciate it. | 21:32 |
| agneli | to quote Han Solo: "I know." :) | 21:33 |
| agneli | i am glad I helped | 21:33 |
| gnarface | the thing is, this is really stupid, i was told to throw away the pine64 microSD adapters, i was warned they're defective, but i tested them and kept the one that seemed to work... but it turns out that risk was not worth it. samsung adapter working fine. | 21:33 |
| gnarface | but i really didn't expect something like this could happen without dd complaining at all | 21:34 |
| gnarface | on the other ones, dd would complain, so i knew they were bad right away | 21:34 |
| gnarface | and to top that off, the error seemed like it must have happened in the same place every time... the possibility of this type of failure is completely baffling to me | 21:35 |
| agneli | i have my storeis with shitsung... | 21:35 |
| agneli | i have writen it on the wall today | 21:35 |
| agneli | never again _any_ shitsung devices | 21:36 |
| gnarface | well today it's what worked | 21:36 |
| agneli | unfrotunatelly i have a lot already | 21:36 |
| agneli | :) | 21:36 |
| agneli | I am joking a bit | 21:36 |
| gnarface | only other ones i have here besides the stack of apparently busted pine64 ones is 1 SanDisk and one Patriot | 21:36 |
| agneli | I am happy you have your solution | 21:36 |
| gnarface | any opinions on reliability for SanDisk or Patriot? | 21:37 |
| agneli | i am suing exclusively sandisk for sdcards | 21:37 |
| agneli | never any issues | 21:37 |
| gnarface | noted, i like them too, but their rep has tanked lately | 21:37 |
| agneli | mine all are like 2yrs old | 21:37 |
| agneli | what happend? | 21:37 |
| agneli | they screwed up under Western Digital? | 21:38 |
| gnarface | not sure, just starting about less than 2 years ago, lots of people over on irc.pine64.org started complaining that their SanDisk SDs and microSDs died very quickly. i suspect a rash of counterfeit devices may be to blame though. | 21:40 |
| agneli | i think so | 21:40 |
| agneli | i have 2 32 GB used bought like 4 years ago | 21:40 |
| agneli | on both is gentoo | 21:41 |
| agneli | without any binary packages | 21:41 |
| agneli | all compiled over and over again | 21:41 |
| agneli | 0 failures | 21:41 |
| agneli | i woudl bet on counterfeits... aliexpress is full of them | 21:41 |
| agneli | you can get them even free provided you buy android based iphone from them... :) | 21:42 |
| gnarface | agneli: you can bet i will certainly checksum the backup this time | 22:55 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!