Werner changed the topic of #armbian to: armbian - Linux for ARM development boards | www.armbian.com | Github: github.com/armbian | Commits: #armbian-commits | Help our crowdfunding: https://bit.ly/3kcuCvj | Type 'help' for help | Logs: -> irc.armbian.com
<TRS-80> I didn't realize it was a derogitory term.
<archetech> no nerve just dont do drive by put downs
<TRS-80> I am i3 and rofi user for fucssakes
<TRS-80> my intent was never a put down
<TRS-80> I finally got consistent themes (mostly) everywhere, good enough for me, but not good enough I would call "rice." Some of those look amazing, IMO.
<archetech> Tony laughed at it it was a put down dont be coy with me
<c0rnelius> ha Ricer: from the latin word Ricarius meaning to suck at everything you attempt
<archetech> own it
<archetech> lol c0rnelius no
<c0rnelius> its funny and it goes on
<archetech> its a slanf term coined in the local drag scene
<archetech> slang
<c0rnelius> A person who makes unecessary modifications to their most often import car (hence the term "rice") to make it (mostly make it look) faster. The most common modifications are (but not limited to
<archetech> yes
<archetech> probably started here in detroit where cars are still objects of ego
<c0rnelius> archetech: I don't think ur a ricer and nor have I ever used the term. :D But I do think its funny.
<c0rnelius> probably
<TRS-80> Are you a car or desktop ricer? I guess I mostly refer to the latter.
<TRS-80> Sorry, "entusiast." Ahem.
<archetech> funny yes but doesnt apply to me why I didnt take much offense
<archetech> I test dev drivers
<archetech> for n2 gpu
<c0rnelius> archetech: stop picking on TRS-80, he be a nice fella.
<archetech> that has a ton of real world performance ramifications
<archetech> check out the thread at odroid if ya like its very interesting tech
<TRS-80> of course it does
<archetech> those panfrost devs are neck deep in C++ on it
<archetech> TRS-80, did you get that sbc to work yet?
<TRS-80> No. And now I'm exhausted from "arguing with people who are wrong on the Internet." :/
<archetech> LOL
<archetech> so whats it need a new uboot?
<TRS-80> I suspect so
<archetech> lmgtfy
<archetech> not
<archetech> you'll find it
<TRS-80> I am not sure it's that easy (but maybe it is).
<TRS-80> I will. Maybe or maybe not tonight.
<archetech> boot.ini boot.scr gah my enemy
<archetech> so tricky
<archetech> then dtb;s then kernels
<archetech> then drivers
<archetech> when I came here I knew nothing
<archetech> and had a crap board a rock64 v2
<archetech> now i have a corvette and it works pretty well
<TRS-80> You know how to surgically replace u-boot? Tony said just do dd from x location of y length but the idea of that scares the crap out of me. lol
<archetech> yes the old way is at sunxi
<TRS-80> Yes, the page I linked that resulted in flamefest.
<archetech> new way apt pacman gimme a new uboot
<TRS-80> Is that relevant to Armbian though?
<archetech> tried using armbian-config on that board?
<TRS-80> That is how it got broke in first place, actually.
<TRS-80> Now I can't boot.
<archetech> ah yeah been there
<nekomancer[m]> Tonymac32: how can I test your https://github.com/armbian/build/pull/2264 ? Should it work?
<archetech> TRS-80, what the board model?
<TRS-80> cubietruck
<archetech> see if arch has the steps
<TRS-80> gotta run, help bring in groceries, be back (hopefully)
<archetech> who makes the "its a TRUCK"!
<c0rnelius> don't own the board but I can't imagine the uboot is super hard.
<archetech> c0rnelius, true
<c0rnelius> whats the SoC?
<archetech> allwinner
<c0rnelius> armv7 or v8?
<archetech> dd if=u-boot-sunxi-with-spl.bin of=/dev/sdX bs=1024 seek=8
<archetech> done next!
<c0rnelius> oh v7 even easier... dont even need ATF.
<archetech> correct
<archetech> TRS-80, Ill just leave that there for u
<c0rnelius> skip boot scr and just use extlinux
<archetech> dont confuse him
<c0rnelius> just sayin, its way easier
<archetech> that isnt hard
<c0rnelius> no commands needed for extlinux
<c0rnelius> write said file, place file in correct place... boot.
<c0rnelius> its like syslinux or grub for sbc's
<TRS-80> Thanks guise. I can get the files from Armbian though (I found them, I think).
<TRS-80> And I think boot.scr is already in place and correct. Just need u-boot I'm pretty sure.
<archetech> its the dd ya needed
<TRS-80> right
<c0rnelius> which is the same on all allwinner socs I've used anyway
<archetech> its a TRUCK!
<c0rnelius> I've said it before and I'll say it again allwinner has the easiest uboot there is.
<c0rnelius> And unlike amlogic and rockchip you don't need to offset partitions to use it.
<archetech> ok ok
<archetech> now fix it and we can get ice cream
<c0rnelius> I'm lactose intolerant
<c0rnelius> How about we just get some shots of whiskey and a few beers?
<archetech> chianti for me dont need to bring the house down
<TRS-80> OK, so info at https://linux-sunxi.org/Bootable_SD_card#SD_Card_Layout is still correct? It essentially aligns with dd command that archetech posted (start writing at 8K) for u-boot. I really appreciate all your feedback (since I never did this before).
* TRS-80 will be drinking soon enough, one way or the other
<c0rnelius> Gay
<TRS-80> Remember, this all started with me saying to lanefu "Here, hold my beer" right before attempting a kernel upgrade... :D
<TRS-80> c0rnelius: Remember, he took offense at "ricer"...
<archetech> how'd the old sicillian get so old chianti and humor
<c0rnelius> Doesn't Armbian for some reason like to upgrade your u-boot whilst upgrading your kernel?
<archetech> sometimes
<TRS-80> c0rnelius: I would really like to know the answer to that as well. If so, I suspect that's where it all may have gone wrong.
* TRS-80 goes to flash u-boot, finally and after much trepidation
<c0rnelius> TRS-80: he'll be fine.
<archetech> when its unbootable what ya got lose?
<archetech> to
<TRS-80> yes, same logic I followed already
<TRS-80> dd to arbitrary location makes me a bit naervous is all
<c0rnelius> I'm pretty sure they push updates to uboot during kernel upgrades. You would have to Igor, but I can swear I've seen it.
<c0rnelius> have to ask*
<archetech> well doing that is a double edged sword
<c0rnelius> yep
<TRS-80> I will probably add that in my thread as a folluw up question when this is all over
<archetech> why arch does it after its first booted
<archetech> their kernels often work with old and new uboots
<archetech> but I dont count on that newkernel means new uboot when I can
<c0rnelius> u-boot is way more personal and I feel like that should be something ur asking for and not forced to do. the binaries are kept under /usr/lib
<c0rnelius> on armbian
<archetech> I just went through this yesterday on my n2 on arch
<archetech> oh look new kernel bam
<lanefu> TRS-80: are you saying I didnt hold your beer
<TRS-80> lanefu: Apparently not correctly, anyway!
<TRS-80> No, I take all the blame.
<c0rnelius> arch is shady and always has been. jacked all the good ideas from CRUX.
<TRS-80> lanefu: Also, Happy Friday! :)
<c0rnelius> the 13th!
<archetech> lanefu, beer dropper!
<lanefu> TRS-80:that dd is what i did the other day
<lanefu> to sector 8 or whatever
<TRS-80> a .deb is just an archive, right? Just get the nib out of there I guess (/me goes searching internet)?
<lanefu> lol
<TRS-80> ORLY?
<lanefu> yeah i mean i installed the uboot because you see the version being crazy new
<TRS-80> Your vim fingers of fury were going too fast for my Emacs brain apparently to follow along
<archetech> its a uboot why are you so stuck on armbian way?
<lanefu> also i havent been following irc today
<lanefu> so keep on doin what ur doin
<c0rnelius> yes it contains binaries and pre and post scripts which do certain things during install
* archetech face palms
<archetech> make it boot per my link then break again ;p
<archetech> c0rnelius, arch isnt shady original dev used LFS
<c0rnelius> bullshit
<c0rnelius> ha
<archetech> crus was just a iea
<archetech> crux
<archetech> idea he followed
<c0rnelius> Arch Linux was inspired by CRUX, another minimalist Linux distribution.
<archetech> bs eh its all at arch the history
<archetech> yeah but inspire isnt build
<c0rnelius> which in turn was inspired by the simplicity of BSD
<archetech> what got him the base crux? no
<c0rnelius> who got what who?
<archetech> else it would be just a crux fork
<c0rnelius> It is, which a diff package manager
stavrakis has quit [Quit: Leaving]
<c0rnelius> with*
<archetech> read some more
<c0rnelius> and then turn into bloat garbage
<archetech> nothing beats Arch
<c0rnelius> I have never run an Arch based system for more than a year without it breaking
<TRS-80> lanefu: Oh of course: U-Boot SPL 2020.10-armbian (Oct 18 2020 - 23:16:40 +0200) <- the current boot looping u-boot
<c0rnelius> own personal experience sure, but... my experience nonetheless.
<archetech> and I run arch testing and it rarely breaks
<archetech> so its the user not the distro design
<c0rnelius> yeah sure... update? ok! reboot? fuck
<c0rnelius> thats my experience
<archetech> and its easy to fix winthin 24 hrs theres a fix
<c0rnelius> why would I want to wait?
<c0rnelius> thats silly
<archetech> cuz its sleek and fast
<c0rnelius> how about not break my shit
<c0rnelius> don't push an update without venting it first
<archetech> then your whiney butt needs debian stable ;p
<archetech> next!
<c0rnelius> sorry if I prefer to get shit done and not debug some lazy fools mess
<c0rnelius> bad on me
<archetech> arch isnt for everyone
<c0rnelius> hey man I tried...
<archetech> only the technically astute
<archetech> gotta love their irc eh
<archetech> um my archwont boot after upgrade FLAME HIM!
<c0rnelius> I was CRUX user for years and really gave Arch the old college try, but at the end of the day. Its not up to par when it comes to stable. Not even remotely.
<archetech> yeah you got that treatment and got butthurt
<archetech> you not the only one
<c0rnelius> I don't get butthurt, I move on. ;)
<archetech> right I got butthurt and stayed the course only difference
<c0rnelius> i think that's called stubborn
<archetech> some jackass kids dont get my goat
<archetech> same for Igor lol
<archetech> he tries
<c0rnelius> As a matter of fact when I first started using linux some 20 years ago I hated debian.
<archetech> bullies just stand up to them
<c0rnelius> But I will say this about it, it fucking works.
<archetech> give em a flick in the forehead they go running
<c0rnelius> who?
<c0rnelius> arch user?
<archetech> arch peeps etc
<c0rnelius> yeah they like to complain
<c0rnelius> lil bitches them kids
<c0rnelius> to bad about slackware
<archetech> just dont flick a truepsychopath which there many in FOSS
<c0rnelius> that was my jams for a long time
<archetech> so be wise
* lanefu not making any reiserfs jokes
<archetech> lol
<c0rnelius> BSD also is awesome, but hardware support there is just such trash.
<stipa> arch is ok
<c0rnelius> lanefu, don't like wife killers? :D
<archetech> c0rnelius, cuz they be the high priest pureists
ChriChri_ has joined #armbian
<archetech> got gey beards to prove their superior
<archetech> gray
<c0rnelius> gay
<archetech> those guys ive had little deakings with dont like em at al
<archetech> all
<lanefu> i've been working on a pointy beard for more street cred
<c0rnelius> my beard just keeps growing baby
<archetech> yes and when the enemy shaves their beards right before battle who ya gonna shoot? lol
<c0rnelius> haha
<archetech> your buddy thats who
<c0rnelius> what enemy ?
<c0rnelius> You need to hurry up and get the fuck out of Detroit
<archetech> thers hard times a comin all im sayin
ChriChri has quit [Ping timeout: 265 seconds]
ChriChri_ is now known as ChriChri
<c0rnelius> bs
<stipa> i heard china busniessmen are buying factories?
<stipa> actually watched
<c0rnelius> hard times have been here... its the lack of people taking responsibility that is creating all these problems.
<rneese> CHina aint buying crap in the US right now and if things go right they wont for the next4 to 8 years
<stipa> like Americans work for chinses in America
torv has quit [Remote host closed the connection]
<c0rnelius> blame him, blame them... ur color is diff than mine and therefore ur the problem. these are the idiots we need to deal with.
torv has joined #armbian
<rneese> well its time to take politics to another channel please. this is a channel for armbian discussion
<rneese> and it should have been done earlier
<stipa> found it in this documentary https://www.youtube.com/watch?v=m36QeKOJ2Fc
* lanefu wouldnt mind gettin back to linux a but it strue
<c0rnelius> K
<archetech> there is no correcting whats happening
<c0rnelius> There is only remedies
<ArmbianTwitter> @ameridroidSBC (ameriDroid.com): Lubuntu 20.04 for ODROID-C4 with Panfrost Hardware Acceleration https://t.co/Zk0r6EHuhN #ODROID #ODROIDC4 #C4 #SBC #SingleBoardComputer #Lubuntu #Linux #Armbian (15s ago)
<c0rnelius> And basic education
<TRS-80> archetech, you are slowly growing on me (not unlike a fungus)
<archetech> no remedies either
<TRS-80> you are right "there is no correcting..."
<c0rnelius> The problem with this world is in my opinion that people suffer from something called amnesia they don't know where they came from they don't know how they got there and they have no interest in knowing.
phipli has joined #armbian
<stipa> sad but true
<archetech> dont want to preach ill get booted theres my hint
phipli_ has quit [Ping timeout: 256 seconds]
<TRS-80> c0rnelius: I hope you are not ignoring the decades (centuries?) long campaigns of TPTB to deliberately make people forget about those sort of things? To make us all feel afraid, alone, and not connected to anything (history, land, culture, etc)?
<TRS-80> archetech: hint: you aren't the only one that knows ;)
<archetech> <rneese> well its time to take politics to another channel please. this is a channel for armbian discussion
<archetech> <rneese> and it should have been done earlier
<archetech> if you prefer near silence on a fri night that can be arranged im sure
<TRS-80> I am not aware of any specific offtopic policy other than "Armbian takes precedence" but I could have also just made that up. :)
<c0rnelius> I don't ignore history and as a matter of fact I feel as if I may be a steward of such. I believe it's one of the most important things that we have.
<TRS-80> c0rnelius: I knew I liked you
<archetech> TRS-80, lol
<archetech> sounds right
<c0rnelius> :)
<lanefu> yeah once its 11PM Tony_mac32 and i talk about car shit :P
<lanefu> once the europeans goto bed
<c0rnelius> ha
<TRS-80> I mean, I didn't just pull out from arse. Myself and other mods ^ nvm lane beat me to it
<archetech> I enjoy hostory of all kinds to bad about who gets published and who doesnt
<TRS-80> It's Miller Time now, gentlemen
<archetech> lanefu, word
<TRS-80> Hope we are not bothering you too much, rneese ?
<lanefu> speaking of Miller, and cars... https://www.youtube.com/watch?v=LRsralnOrcU
<lanefu> archetech: prmpting you in advance to chime in
<lanefu> rneese: so in X11 land.. there's a different X drivers...... fbdev, fbturbo, armsoc, and (kernel)modesetting
<lanefu> fbturbos are usualyl bastardization
<lanefu> fbdev is the simplest to working
<archetech> Founded in 1907, the Organization of American Historians (OAH) is the largest professional society dedicated to the teaching and study of American history.
<lanefu> armsoc is soemwhat specialized, and honestly i thought it was deprecated.. not sur ewhento euse it exactly
<archetech> that is a bad organization
<TRS-80> archetech: as they say, the winners write it... I am also a student of history, for long time now.
<lanefu> and "modesetting" is "kernel mode setting" or kms.. which is typically what pairs nicely with the drm video driver for a soc
<archetech> lanefu, lol
<archetech> when the goin gets tough the tough cracka miller
<lanefu> its soooo goood!
<lanefu> "that bolt was torqued in detroit"
<archetech> and salt added
<archetech> lanefu, nice sumation
<archetech> on arm vdrv
<lanefu> thanks did i miss anything
<archetech> idk TRS-80 wont tell uss if his toy booted
* TRS-80 just came back from eating dinner and making a freshie
<TRS-80> Going to try now
<lanefu> okay armbian u-boot pop quiz... where the F does ${fdtfile} get set https://github.com/armbian/build/blob/master/config/bootscripts/boot-sunxi.cmd#L55
<TRS-80> this is relevant to my interests lol
<archetech> trs is fueled and now he's adding nitro ot that
<archetech> to
<archetech> car talk
<archetech> which this is the ONLY irc chan that conflates car talk and arm talk in the world!
<lanefu> man that's a promise lol
<archetech> thks to tony lane and others
<stipa> some cars have arms
<lanefu> speaking of which I think i've conceeded to just ordering a carb instead of fuel injection for jeep
<archetech> some arms are in cars
* lanefu wants armbian in his jeep
<archetech> maybe we can get whats his name with the lec car to donate one for us to slap armbian in
<stipa> i think the first wall would be mine
<archetech> firewall fight
<lanefu> man i tel ya there's a non-zero chance there might be some armbian in a tesla lol
<archetech> we can change that
<stipa> supercar from Croatia has linux in
<TRS-80> Just started poring through it, but here is result of `rg fdtfile`, if anyone wants to help: https://paste.debian.net/hidden/5fe83fad/
<stipa> electric
<archetech> stipa, are you near Igor?
<stipa> around 150 km apart
<stipa> it's near for US standards
<archetech> hmm so like 80 miles
<stipa> :D
<stipa> probably archetech
* archetech has a flashback to speeding in Canada due to the km/hr
<archetech> man they sure let ya go fast here lol
<stipa> Canda yeah
<stipa> who would tell
<stipa> makes no sense
<archetech> Can and Australia would be the best countries in the world but that bust body Queen
<archetech> busy
stipa has quit [Ping timeout: 240 seconds]
stipa has joined #armbian
<lanefu> /buffer 4
<lanefu> thats right i said it.. /buffer 4
<TRS-80> lanefu: ^ why did that not work? :?
<TRS-80> you can also just do C-n, vim peasant
<lanefu> cuz i accidently escape it with //buffer 4
<TRS-80> aah, should have figured
<TRS-80> probably you can also do C-downinvimkey whats that like k or l?
<lanefu> j
<lanefu> problem is im bouncing between buffer 2, 4 and 29
<TRS-80> M-j NUM
<lanefu> and i need to like remap some meta key
<TRS-80> you might call it Alt
<lanefu> yeah alt==command equals mac steals it
<TRS-80> no just Alt-j then let go then enter num
<TRS-80> buffer number
<lanefu> ∆∆2
<TRS-80> we talking weechat, right?
<lanefu> yeah bro but im in iterm2 on a mac
<TRS-80> what the actual fuck.jpg
<TRS-80> it's grabbing it?
<lanefu> alt is typically the command key
<archetech> dude how longs it take to do a dd
<TRS-80> oh there is only on on mac, right?
<lanefu> so i can lke turn it off specifically in iterm
<lanefu> but i use it a lot
<TRS-80> there is no Super (Win) and Meta (Alt), amirite?
<lanefu> control, option, command
<lanefu> typically option and command translate to win and alt
<TRS-80> OK
<TRS-80> so command-j then NUMBER, no?
<lanefu> nope
<TRS-80> chek keybindings on your platform, maybe they are different because mac?
<TRS-80> anyway, I better go flash that, archetech getting impatient for the exciting conclusion :D
<TRS-80> sory "copy"
<archetech> heh i htought ya want ed it fixed pardon me
<TRS-80> no I am fucking off as usual
<TRS-80> but I will now
<lanefu> okay ive assinged the left option key to meta
<TRS-80> and?
<lanefu> ironically
<lanefu> less
<TRS-80> ?
<TRS-80> the pager?
<lanefu> give me another alt command
<lanefu> less as it less happened
<lanefu> as in
<TRS-80> Alt + arrow up down should also change buffers
<lanefu> okay that works
<TRS-80> so should C + n/p
<lanefu> alt-j doesnt
<TRS-80> to be clear, you have to do (M-j NUM <RET>) where NUM is an integer
<lanefu> correct
<TRS-80> ;5Ano return
<TRS-80> wait a sec, now It's not working here either
drobo_00 has quit [Quit: drobo_00]
<lanefu> rofl
<lanefu> alright copy your files
<TRS-80> maybe something changed?
<TRS-80> I just did that like yesterday though
<TRS-80> I mean, I do it all the time, esp with double digit buffers
<TRS-80> OK, leaving and I mean it this time
<nekomancer[m]> <c0rnelius "lanefu, don't like wife killers?"> but reiser4 not contain any wife killer's code
<archetech> no I just run meson --prefix /usr ..
<lanefu> i stopped using pv after i learned about dd option `status=progress`
<stipa> nice
<archetech> whats pv
<archetech> yes status works
<lanefu> pretty IO status for piping file streams
<lanefu> and if it knows source file it does eta
<archetech> ah
<lanefu> so you could like pv armbianimage|dd of=/dev/disk bs=1M conv=fsync
<archetech> will it help me stay out of FB jail? then i dont need it
<lanefu> oh man can i show you guys to nerd tools i found
<lanefu> no more less on log files https://github.com/tstack/lnav
<Tony_mac32> lol oh?
<lanefu> its written in C too
* archetech uses |more
<Tony_mac32> lol
<Tony_mac32> oh, other note, WD blue NVMe's are like $55 for 500 gig
<Tony_mac32> I didn't realize they had gotten cheap
<lanefu> archetech:and then pipe to lpr on your dot matrix
<lanefu> wow yeah that is cheap
<stipa> cool
<Tony_mac32> lanefu I had an epson LX810 until about 3 years ago
<Tony_mac32> don't knock the dot matrix
<lanefu> yeah... they're pretty great
<lanefu> sadly my first printer was a 9 pin
<lanefu> so it was slooow
<lanefu> tandy dmp-107
<Tony_mac32> it would do "graphics"
<Tony_mac32> rofl
<Tony_mac32> and loud enough to wake the neighbors
<lanefu> yeah like slate tablet style
<Tony_mac32> I think that was a 9-pin setup
<Tony_mac32> anyway I snagged 2 of them, one for my nanopi M4V2
<lanefu> ohh nice
* lanefu pondering replacing all his NAS 2TB enterprise spinning rust with 2TB SSDs slowly over time
cubie has quit [Ping timeout: 258 seconds]
DaRock has joined #armbian
<c0rnelius> nekomancer[m]: didn't notice there was one? anyone using this?
* nekomancer[m] knows no one. but it still developed, I see news article not long ago
SebastianLee2 has joined #armbian
<archetech> nekomancer[m], got those hdd's yet?
<c0rnelius> back in the day I actually used reiserfs and didn't stop doing so because of his dirty deeds. Development stopped, so I dropped it.
<nekomancer[m]> <archetech "nekomancer, got those hdd's ye"> emm... sorry, I am out of context... ?
<archetech> helios
<c0rnelius> clang++: error: unable to execute command: Killed
<c0rnelius> clang++: error: clang frontend command failed due to signal (use -v to see invocation)
<c0rnelius> clang version 9.0.1-15
<c0rnelius> Target: aarch64-unknown-linux-gnu
<c0rnelius> Thread model: posix
<c0rnelius> InstalledDir: /usr/bin
<c0rnelius> clang++: note: diagnostic msg: PLEASE submit a bug report to https://bugs.llvm.org/ and include the crash backtrace, preprocessed source, and associated run script.
<c0rnelius> clang++: note: diagnostic msg:
<c0rnelius> ********************
<c0rnelius> PLEASE ATTACH THE FOLLOWING FILES TO THE BUG REPORT:
<c0rnelius> Preprocessed source(s) and associated run script(s) are located at:
<c0rnelius> clang++: note: diagnostic msg: ../../tools/clang/crashreports/target_auto_attacher-c7b28a.cpp
<c0rnelius> clang++: note: diagnostic msg: ../../tools/clang/crashreports/target_auto_attacher-c7b28a.sh
<c0rnelius> clang++: note: diagnostic msg:
<c0rnelius> ********************
<c0rnelius> damn chromium
<nekomancer[m]> A! no. covid, quarantine, job shrink, I am scared possible end of job, enter money-saving mode.
<stipa> right
<stipa> it's no good
<stipa> people without savings are doomed
<archetech> nekomancer[m], ok hang in there
<nekomancer[m]> a lot of collegues was faired already :(
<nekomancer[m]> I am not yet, but it is possible.
SebastianLee has quit [Quit: The Lounge - https://thelounge.chat]
SebastianLee2 is now known as SebastianLee
<nekomancer[m]> I can live 2-3 month for a price of 2 14 TB HDD I plan to buy.
<archetech> got any old hdd's layin around ya can use?
<stipa> yeah, maybe it's better not to buy
<nekomancer[m]> hope I will not try...
<archetech> really only need 2 for a raid 0
<nekomancer[m]> hm... maybe I have 1-2 3.5" 1-2 TB old hdd with bad blocks. Just not think in this direction
<archetech> I have 2 ebay $10 250GB drives I still use
<stipa> i haven't had success in repairing disks with bad blocks
<stipa> for some reason
<c0rnelius> If you think the virus is bad now then just give it a few more months.
<c0rnelius> If we are lucky it won't mutate into something worse
<archetech> no bad blocks is a bad sign dont use that
<nekomancer[m]> need to dig into old boxes...
<stipa> c0rnelius: right
* lanefu likes the cheap hitachi / hgst enterprise refurb drives
<archetech> used is a crap shoot but if ya know a good reliable supplier its not too risky
<archetech> I got mine fr some guy who replaces drives per age not usage
archetech has quit [Quit: Leaving]
<lanefu> nekomancer[m]:look for 2-4TB HGST drives
archetech has joined #armbian
<nekomancer[m]> we have no local trusty second-hand for hdd.
<nekomancer[m]> orders from EU have expensive pastage. From outside EU (USA, China) — expensive post, 35+ days in transit, blocker prices to returns.
<nekomancer[m]> that's why nobody buys not new HDD in Latvia :)
<nekomancer[m]> only new and with local shop warranty/
<nekomancer[m]> all others ways too risky.
<TRS-80> Tony_mac32: Is that a sale? Or normal price these days?
<TRS-80> also, ahem, IT VEEEEERRRRKS
<TRS-80> :D
<lanefu> Yay
<TRS-80> lanefu: Now, I would like to talk to you about pets and cattle. But maybe that's the last thing you like to discuss on a Friday night? lol
<TRS-80> I mean, that's what you do all week, right?
* TRS-80 is OK with just drinking my freshie, too
* Tony_mac32 raises glass
<TRS-80> Cheers, mate!
<TRS-80> stressful couple days
<TRS-80> Internet was out right before my cubietruck blew up
<TRS-80> for a couple days
<archetech> Now, I would like to talk to you about pets and cattle. thats hilarious line
<archetech> for an arm chan
<archetech> animal husbandry is down the hall
<mike64> where would I go to figure out what this error means while compiling? make[1]: *** [scripts/Makefile.build:117: arch/arm/kernel/asm-offsets.s] Error 1
<mike64> actually if I changed the value that's listed in here would that fix my error?./include/linux/seqlock.h:248:15: error: implicit declaration of function ‘__seqcount_sequence’; did you mean ‘__seqcount_assert’? [-Werror=implicit-function-declaration]
<TRS-80> archetech: You see I got it working? :)
<archetech> yes enjoy
<TRS-80> Also, "pets and cattle" is a DevOps idea, where instead of carefully hand curating the config on a particular box/service (like my current situation) you just spawn services in containers, etc. So if one dies/fails or whatever, you just spawn another one, ideally automatically. Pets vs cattle.
<TRS-80> Thanks for your help earlier.
<archetech> 0ah ok np
<TRS-80> mike64: Go to line 117 in the Makefile and start looking what it does? Maybe that's what led you to your second question? Beyond that I'm afraid I have no idea (maybe someone else will come along who does, but it's kinda late).
<mike64> where exactly is that makefile in the armbian build directory I'm having a heck of a time finding it.
<TRS-80> should be right in root but hang on and I'll look
<TRS-80> where do you call make?
<TRS-80> I just searched whole build project, it's not in there. In fact I think it might be sub-project.
<TRS-80> build repository, rather
<mike64> so I'm not crazy not finding it
<TRS-80> I suspect it's in parent. Do you have the whole project?
<mike64> i followed the instructions on armbian website and cloned a depth of 1 from git. if i change that to a hight number would that show up to me? (this is my first time takling something to this degree)
<TRS-80> Cloned which? build?
<TRS-80> armbian/build in other words?
<mike64> this is what I used git clone --depth 1 https://github.com/armbian/build
<TRS-80> Oh it must get called in that shell script you invoke
<mike64> so dig in the compile.sh?
<TRS-80> I already did, it seems to lead to lib/build-all-ng.sh but still looking (also I could be wrong)
<TRS-80> I dunno man, lots of stuff going on there
<TRS-80> But I never built from scratch, either
<TRS-80> mike64: --depth <depth> Create a shallow clone with a history truncated to the specified number of commits
<mike64> ok so that would not be helpful
<TRS-80> So that just makes ut so you don't get the whole history of the project
<mike64> yeah it's tricky where it is hiding
<TRS-80> Gound that in `help git clone`, FYI.
<TRS-80> s/Gound/Found/
<ArmbianHelper> TRS-80 meant to say: Found that in `help git clone`, FYI.
<mike64> thanks I've been lost in readings.
<TRS-80> The path to true wizardry lie through many scrolls of ancient wisdom (i.e., source code, man pages, etc...). ;)
<mike64> I try to. the biggest trick is understanding something before you need enough to remember where it is when you do.
<TRS-80> The more you do it, you start to become familiar.
<TRS-80> Some things I also make notes. Especially like the second or third time I need to look them up. lol
<TRS-80> I have extensive notes in Orgmode. I also try to keep good project logs, those can really come in handy a year or two down the line some times.
<TRS-80> Which reminds me...
* TRS-80 goes and restarts lisp machine now that network drives are mounted
<TRS-80> yaay my comfy env
<TRS-80> mike64: It should "just work" though. Are you using supported environment? What changes did you make (if any). Make can be a complicated beast; have you used it much before (I want to help but not insult by presuming what you may or may not know)?
<mike64> I'm triyng to patch the rt preempt patch I can compile without adding that
<TRS-80> aah ok
<mike64> I'm using the supported ubuntu environment in virtual machine
<TRS-80> Well, that is where I would start looking then. :)
<TRS-80> Or perhaps consult with whoever is working on that?
<TRS-80> Or is that you? lol
<mike64> i'm on fourth try between current and legacy kernels i'm getting different errors. I just have never gotten this deep into so I'm leaning new errors and fixes
<nekomancer[m]> lisp machine — something from times near Turing machine?
<TRS-80> nekomancer[m]: You probably know it as Emacs. ;)
<nekomancer[m]> :)
<TRS-80> mike64: Has that patch ever applied cleanly?
<mike64> I have not been able to
<mike64> truth be told I have never applied a patch before so I might not even be doing it right
<mike64> but it seems to just add it to the correct folder in armbian build.
<TRS-80> here is what I mean by comfy env: highlight 'rt preempt' in terminal, s-B, first dialog is already on default search provider, <RET>, second dialog for query already filled with x-primary selection, <RET> again and new browser tab opens that query at that search provider. Not bad if I say so myself... :D
<TRS-80> you apply with git? It's pretty straightforward if you follow instructions you can finds everywhere in internet.
<TRS-80> Although, where did you get the patch from? If not formatted exactly right (i.e., email done wrong) it can mess up the formatting.
<TRS-80> You download it you should be fine though.
<mike64> i downloaded it from the rtwiki site that builds it
<TRS-80> Did the patch ever touch your Windows system?
<mike64> I don't use windows
<TRS-80> oh, sorry
<TRS-80> good man :)
<TRS-80> mixed you up I think with someone else
<TRS-80> mike64: What arch?
<mike64> whar arch for what?
<TRS-80> You trying to copile for (I am looking at rt preempt wiki)?
<mike64> i'm runnig an orangepi2 and have been trynig between the sunxi-current which is 5.9.9 and sunxi-legacy which is 5.4
<mike64> I might have to go back to a distro that works with a 4.x kernel because I have see stuff about it working then.
<mike64> but I feel like it's related to the fact of using the Ubunutu 20.04 environment and building for ARM I must have something not set right for the crosscompile but I don't know where
<TRS-80> You have heard of this working on OrangePi 2? Isn't that ARM7?
<TRS-80> (I am trying to look up now)
<mike64> I have definately seen success on 3.x and 4.x. I might have to go back in time and build up from there.
<TRS-80> Oh OK then
<TRS-80> I just didn't see ARM7 in that list of known working arch on the wiki (doesn't mean it doesn't or can't work)
<TRS-80> yeah Orange Pi 2 -> H3 -> Cortex-A7 -> ARMv7-A
<TRS-80> I see a lot of ARM9, but no ARM7 here: https://rt.wiki.kernel.org/index.php/CONFIG_PREEMPT_RT_Patch
<mike64> I've seen posts such as this "Placed unpacked 'patch-4.6.4-rt8.patch.xz' in 'userpatches/kernel/sun8i-dev'
<mike64> Compilation successful." so I might have to go further back. So clonig deeper into the git history would that allow me to grab an older armbian image from that kernel time?
<TRS-80> How well do you know git?
<mike64> I'm learning. (this is as great way to learn more.)
<TRS-80> But basically, yes. Not sure how cloning deeper works, once you already cloned? Just try I guess.
<TRS-80> Worst case, just delete the whole repo and clone all over, just without the depth arg.
<mike64> or maybe use an older raspbian image from the 3.x time and build up from there. I'm not sure which would be more work to try to go from.
<mike64> I do have one of those
<mike64> so for a statement like this. where are those chunks saved to? I can't find that. 1 out of 7 hunks FAILED -- saving rejects to file arch/powerpc/platforms/pseries/iommu.c.rej
<TRS-80> Raspbian make images for OrangePi?
<TRS-80> I guess not at the path specified? I wouldn't know where else to look.
<mike64> I have an old rasbian image from the time that works that was from the orange pi website. but it's old
<mike64> i think it's based of Jessie
<TRS-80> Sources? Or a compiled image?
<mike64> image for the sdcard
<TRS-80> Well, you need sources to compile.
<TRS-80> Infact, now that you mention it. I don't know how Armbian build script really works tbh. Does it pull in an already compiled kernel? Or does it compile it? Because I think these patches need to be applied to the kernel, right? Sorry if that's obvious (I'm learning here).
<TRS-80> And depending how Armbian build script works, you might need to apply kernel patched first? But surely someone helping you earlier would have told you such things.
<mike64> I don't really know how armbian build works either...that's why I'm here.
<TRS-80> OK, so we are on somewhat equal footing then. lol
<TRS-80> You picking up what I am putting down then?
<archetech> it pulls sources in and does w/e the vars you set tell it todo
<TRS-80> Thanks archetech
<TRS-80> I try to help, but...
<archetech> best way to see is look at the armbian repos it pulls its configs from
<archetech> it just does what those say
<TRS-80> I poked around in there few times, but there is a lot going on. lol
<TRS-80> Lane did get me to clone the build repo today to my machine for the first time though, sneaky bastard. lol
<archetech> plus you can add patches etc
<archetech> rea config.sh and where it loops
<archetech> read
<archetech> goes all over the place its not linear
<TRS-80> yeah I was only scanning quickly through there, not studying, but there is a lot going on
<TRS-80> muh recursive algorithms
<mike64> that nonlinear is what is throwing me off because I don't know where my makefile or anything is
<mike64> but I I getting good at looking at logs
<TRS-80> mike64: How well do you know GNU Make? Or even bash for that matter?
<archetech> make is done from the repos configs
<mike64> I've used them to build. But I don't know alot. Trying to get better.
<archetech> igor wont explain it to me lol
<archetech> that would cost too much $$
<TRS-80> lol
<TRS-80> Hey, give the guy some slack, he has had this project largely on his shoulders for some years now.
<TRS-80> It would be nice to flesh out the wiki/docs though maybe to give a little more help / background on building, as that is one of main focus of project.
<TRS-80> Maybe we can figure some things out together.
<TRS-80> Some thing that "new" people often don't realize is that their perspective is actually valuable to (any) project. Because the devs already know a lot, they may not write the best docs. But new people, it's all fresh in the mind what was confusing at first. And so sometimes can write better docs (IMO) because of that perspective.
<mike64> Yeah learning and helping with the docs is something that I could help with.
<TRS-80> mike64: That would be an extremely valuable contribution. Start doing something like that, and Igor will take more of his time to help you, I can just about guarantee.
<mike64> though I think for tonight it is time to rest. Thanks to all for their insights.
<TRS-80> Getting pretty tired myself. Cheers, mate.
<archetech> yes it needs better docs
<TRS-80> "it" is "us" archetech, ya dig? ;)
<TRS-80> If by "it" you meant the project.
<archetech> the builder
<Werner> Good morning
archetech has quit [Quit: Leaving]
MrFixIt has quit [Read error: Connection reset by peer]
MrFixIt has joined #armbian
archetech has joined #armbian
s_frit has quit [Ping timeout: 240 seconds]
TRS-80 has quit [Quit: WeeChat 2.9]
MrFixIt has quit [Remote host closed the connection]
MrFixIt has joined #armbian
<IgorPec> good morning
margoaron has quit [K-Lined]
<Werner> Morning
<buZz> wow a apology
<buZz> 02:45:00 <TRS-80> Hey, sorry if I over-reacted earlier.
emOne has joined #armbian
emOne has joined #armbian
emOne has quit [Changing host]
lgranie has joined #armbian
<lgranie> hi
<Werner> Hi
<lgranie> looking forward to update a x905x2 kernel
<lgranie> did linux-image-current-meson64 will do the job?
<Werner> Depends on the actual board
<lgranie> x96max
<Werner> If it is a random tv box then you should consider stating your question here because in this channel nobody tinkers/cares about those :) https://forum.armbian.com/clubs/1-tv-boxes/
<lgranie> ok thx
<lgranie> have a nice day
<Werner> likewise
lgranie has quit [Remote host closed the connection]
Redfoxmoon has quit [Ping timeout: 260 seconds]
<stipa> buZz: maybe you should do the same :D
Unit193 has quit [Read error: Connection reset by peer]
Redfoxmoon has joined #armbian
Redfoxmoon has quit [Changing host]
Redfoxmoon has joined #armbian
Unit193 has joined #armbian
<buZz> for him misinterpreting? lol
<rneese> lide is a isinterpetation
<rneese> lol
<rneese> morning
<rneese> whats going on this am
<buZz> not a lot :)
nkuttler_ has joined #armbian
Net147_ has joined #armbian
Net147 has quit [*.net *.split]
mpmc has quit [*.net *.split]
nkuttler has quit [*.net *.split]
silver_hook has quit [*.net *.split]
[TheBug] has quit [*.net *.split]
creemj has quit [*.net *.split]
silver_hook_ has joined #armbian
[TheBug] has joined #armbian
mpmc has joined #armbian
nkuttler_ is now known as nkuttler
nekomancer[m] has quit [*.net *.split]
creemj has joined #armbian
sassinak-work has quit [*.net *.split]
TheNumb has quit [*.net *.split]
sassinak-work has joined #armbian
nekomancer[m] has joined #armbian
TheNumb has joined #armbian
psydruid has quit [Ping timeout: 240 seconds]
forest[m] has quit [Ping timeout: 240 seconds]
initbasti[m] has quit [Ping timeout: 244 seconds]
TiD91 has quit [Ping timeout: 244 seconds]
solderfumes[m] has quit [Ping timeout: 246 seconds]
nekomancer[m] has quit [Ping timeout: 246 seconds]
<lanefu> mike64: which opi do you have
<rneese> mornig lanefu
<rneese> so I by chance do you know what boards use the meson64
<lanefu> howdy
<lanefu> rneese: meson64 == le potato, odroid c2, odroidn2
<lanefu> meson64 == amlogic64
<rneese> II have been working on meson64-common.inc ok
<lanefu> now all 3 of those are different subtypes to mak things fun
<rneese> we need a lepotato to test the xorg.conf
<rneese> so the n2/n2+ WORK
<rneese> THE C4 IS NOT AMLOGIC
<rneese> the c4 is amlogic
<rneese> x905
<rneese> S905x3
<rneese> sorry
<rneese> do we have one for testing
nekomancer[m] has joined #armbian
<rneese> I can put the first change you can review
solderfumes[m] has joined #armbian
psydruid has joined #armbian
<rneese> i put 2 files in the package/bsp/odroid
TiD91 has joined #armbian
<rneese> and i rm the 02-driver from the meson64-common.inc
<rneese> then I just have to coonfigure the the boards under the desktops to the right driver
<lanefu> i have tater
<rneese> ?
<lanefu> i have le potato
<rneese> ok can you test and see if it works
<lanefu> S905x
<rneese> with the change
<lanefu> yeah build an image
<rneese> for somereason the new builder is failing me
<rneese> its not merging in the desktop into the root fs
<rneese> I should be bale to build with out apps
forest[m] has joined #armbian
<rneese> building you a deepin img
initbasti[m] has joined #armbian
archetech has quit [Quit: Leaving]
archetech has joined #armbian
jock has joined #armbian
<jock> hello all!
<Werner> Hi
<jock> hi Werner, got a bit busy last weeks... may I ask if armbian 2020.11 already went out?
<Werner> No, might even be delayed a bit to early december
<jock> Ahhh ok, thanks
<lanefu> mike64: See this for inspiration.. a bit of a brute force attempt, but seems to build https://github.com/lanefu/build/tree/RTAttempt
<lanefu> mike64: also search for mike64 here https://armbian.lane-fu.com/debs-beta/
TRS-80 has joined #armbian
DaRock has quit [Ping timeout: 260 seconds]
<IgorPec> hej
<TRS-80> Good morining Igor
<TRS-80> I was just writing you in my thread, but maybe better ask here (if you have a moment).
<IgorPec> its afternoon :) been busy with family stuff
<IgorPec> shoot
rneese has quit [Remote host closed the connection]
<TRS-80> I plan on working on some minor doc updates, based on my recent experience. I already cloned some repos (with plain git) to my local machine. But how to get changes back to the project? Do I have to do the whole GitHub song and dance where I clone the project to my account there and bla bla or can I just push directly up from my dev machine with plain git somehow?
rneese has joined #armbian
emOne has quit [Ping timeout: 246 seconds]
<TRS-80> Or do you accept patches via email, etc.? I know this isn't really a mailing list driven project.
<IgorPec> you clone to your account, clone that to your comp, update, update, ... when you are happy, create a pull reqest
jock has quit [Remote host closed the connection]
emOne has joined #armbian
emOne has quit [Changing host]
emOne has joined #armbian
<IgorPec> mailing lists are odd in 2020 :)
<IgorPec> but people do use it ofc
jock has joined #armbian
<TRS-80> I just hate needing to insert fucking GitHub into the mix, just a useless middle man, ya know?
<IgorPec> if you create a patch, it should be created with Git that it contains your data etc
<rneese> language young man
<IgorPec> well, until we host stuff there
<TRS-80> IgorPec: I think most of projects I contribute to are on Mailing lists, including some pretty big ones.
<IgorPec> agree, but this require manual intervention
<TRS-80> That's why I ask. I know your busy and have some workflow probably you prefer.
<IgorPec> for docs i can give you edit rights and its done
jock has quit [Remote host closed the connection]
<TRS-80> What do you mean, on GitHub?
<IgorPec> yes
<IgorPec> >_)
<lanefu> TRS-80: you could find some github cli tools to create github PRs
<TRS-80> lanefu: Good point. Infact I think there are some in Emacs.
<lanefu> but yeah on docs yeah edit writes on master
<lanefu> is fien
<IgorPec> i also mainly do CLI git
<IgorPec> once we will perhaps host our own git, but until then its as is
<TRS-80> Don't be surpries if I bugging you guys in here to make sure of things, I still don't know a lot and I don't want to be putting wrong informations. So I appreciate the write priveleges, but still want review probably to make sure I'm not spreading BS as tkaiser might say. :D
<IgorPec> then making a pull request is the way to go
<IgorPec> you can do everything from CLI
<TRS-80> You guys require any assignment? Real names? FSF for instance require the former, but not the latter. :)
<IgorPec> no, we don't have that strict policy
<TRS-80> cool
<TRS-80> I habe more problem with the real name part than the assignment, tbh.
<TRS-80> Many people gripe about FSF assignment, but you ever see them lose a lawsuit? They don't even get into them very often. And when they do, they win. ;)
<lanefu> whats assignment
<TRS-80> copyright assignment
<lanefu> oh
<TRS-80> You sign your work over to FSF, that's how it works on FSF projects.
<TRS-80> Then they "own" it, but it means they can defend it, etc. legally.
<lanefu> oh makes sense
<TRS-80> Final question: I was thinking armbian-config should be recommended way to flash u-boot? But is that something that even needs to be done often (should it even be in docs)? Or you only need to do it in weird edge cases?
<IgorPec> no, its usually optional
<IgorPec> rarely you need to upgrade it
<IgorPec> idk what happened in your case, but back in the old days we had auto updating
<TRS-80> u-boot?
<IgorPec> but this was disabled 3+ y ago i think
<IgorPec> yes
<IgorPec> u-boot is ... don't touch it if it boots
<TRS-80> at one point, I did not upgrade for long time, I think I missed some upgrade or change in the meantime and that's what caused the problems
<TRS-80> remember this is cubietruck that I been running for years now
<IgorPec> also we don't change boot scripts .. and perhaps your "v1.0" is not compatible with current boot loader
<TRS-80> Like you Igor, my first non RPi SPC :)
<TRS-80> yes my boot script was VERY VERY different
<IgorPec> then problems are expected
<TRS-80> right, hence my conclusion
<IgorPec> most people don't touch it
<TRS-80> which is why I question this should even be i ndocs or not, it's like a weird edge case
<IgorPec> updating u-boot manually is not corner case
<TRS-80> ok good to know
<IgorPec> if you somehow erase it
<IgorPec> you should find info how to install new perhaps
<TRS-80> new from? Armbian? Or upstream?
<IgorPec> ours
<TRS-80> I ended up getting mine out of the deb, but mayber there is better way lol
<IgorPec> yes, that's it
<TRS-80> I can write the guide, jjust need advice on which way is best
<TRS-80> OK, can do
<TRS-80> perhaps even extending armbian-config to fetch it?
<TRS-80> fetch and extract?
<IgorPec> if you have a working machine, this is not a problem
<IgorPec> updating via armbianm-config will work
<TRS-80> Oh right
<TRS-80> this is manual recovery
<IgorPec> yes
<IgorPec> that part perhaps needs some improvement
<TRS-80> Yes, I'll put some time into it today
<IgorPec> if you have time to rework armbian-config ... also nice :)
<TRS-80> for the "I can still boot" case, you mean?
<IgorPec> no, in general
<TRS-80> ok lol
<IgorPec> :)
<TRS-80> Man I dunno, I might ruffle too many feathers. I would want to reformat everything to fit in 80 columns. LOL
<TRS-80> I follow those old school FSF type guidelines
<IgorPec> armbian-config is kind of old school, but not that old :)
<TRS-80> Emacs Lisp is 65 recommended
<TRS-80> But it's a much more elegant language
<TRS-80> OK I guess I get to work
<TRS-80> Thanks for tips. :)
<IgorPec> no
<IgorPec> p
<TRS-80> no to what? lol
<[TheBug]> I think he meant 'no problem' but maybe he ran outa steam halfway after he started typing... that reminds me, time for some coffee!
* IgorPec indeed needs some coffee.
* TRS-80 is halfway into a quite large cup, and it's a new day, hence the motivation
<Werner> Somebody said coffee
* lanefu low on coffee
* Werner ate two pizza and now needs to take it easy
Liptibilly has joined #armbian
Elpaulo has joined #armbian
<Liptibilly> Hi ~, could someone help? I have a OrangePi PC+ (H3) and I would like to compile v4l2-request on it, but it seems that some defines are missing on linux/videodev2.h header file. Looks like this file hasn't been uptated for a while.
<Werner> This file might come from upstream, either Ubuntu/Debian repository (whatever libxy-dev package or so) or kernel which would be megi's sources: https://github.com/megous/linux/tree/orange-pi-5.9
<Liptibilly> Yes, it come from kernel. The required patch has been introduced in mainline in 2018, but I take a new armbian image today, and it is still the old version
<Werner> Try having a look at the sources I linked above. From those all sunxi/sunxi64 images are built (with a couple of Armbian patches on top)
<Liptibilly> Yes, I just checked, and this repo have the right version of my file, but I don't know how to use it :/
<Liptibilly> Should I rebuild the kernel from that sources?
<Werner> So when you use an up-to-date Armbian image this particular file is older than in the sources from which the image is made from?
<Liptibilly> Yes.
<Werner> Odd...
<Werner> The file is part of the kernel headers I assume?
<Liptibilly> yes
<Liptibilly> I checked a couple of armban images for several boards (orangepi ones) and they all have the old version.
<IgorPec> all images uses the same kernel
<IgorPec> they are divided into families, allwinner, rockchip, imx
<IgorPec> ..
<Werner> Um...did you find this particular file in megis sources?
<Liptibilly> yes, and this is the right version
<Werner> Can you paste the path to it real quick? Cannot find it directly
<IgorPec> but we use that sources ... perhaps our patch overwrites it?
<IgorPec> for some reason
<Werner> if megis sources are up to date that must be the reason
<Werner> thx
<Liptibilly> but these files have de defines I nees
<Werner> Ah. That file is slightly bigger ^^
<Liptibilly> need*
<Liptibilly> But on the image I downloaded today, the /include/uapi folder doesn't event exists
<Werner> At least from the number of lines the files are identical in our header package and megis sources
<TRS-80> Trying to set up my GitHub like /armbian/repo by creating project /armbian and then cloning {documentation, build, etc.} into there as submodules; is that going to mess things up when I try and do PRs?
<lanefu> we don't use submodules
<lanefu> and i'd say you're making it too complicated if you are
<Liptibilly> Weird... I just reinstall the package you mention Werner, and still no uapi directory on my tree...
<Werner> Liptibilly, Indeed. Try steamroller tactics and extract the deb using ar x
<TRS-80> I just don't want all the different armbian repos loose in my GitHub flat at top level. Which is how they will be if cloned one by one. Unless I am missing something (this is why I asked)?
<lanefu> TRS-80: you'll still have to initiate a PR from the submodule's repo
<TRS-80> yeah fine, just trying to keep things organized
<lanefu> yeah fortunately github is kind of flat from that perspective
<lanefu> just make your checksouts orangized
<TRS-80> ?
<lanefu> so your github account will have the 2 or 3 forked repos, just flat in the top of your namespace
<lanefu> i'm saying when you check this out locally
<lanefu> just be organized there... ex: `~/GIT/armbian/build ~/GIT/armbian/documentation`
<TRS-80> yes of course
<TRS-80> that is the easy part
<TRS-80> in fact I had to create armbian project on GitHub, clone it local, now I clone each git submodule into there and push back up? Because I don't see how else to do this in stupid UI.
<Tony_mac32> I have never attempted submodules. Teach me senpai
<TRS-80> basically ^
<TRS-80> let me see if this works, then you can also look at my github as an example
<lanefu> a submodule is just a reference to anotehr git repo
<lanefu> so if you have a submodule that is armbian/documentation
<lanefu> you can only push to armbian/docuemntation
<lanefu> it's not going to make a seperate repo in your github
<TRS-80> I noticed it's no different for you guys (I looked at lanefu GitHub) even though you are members of the org. How retarded. Unless I am missing something?
<lanefu> TRS-80: right 90% of the time I just push a branch to my own repo fork..and initiate a PR
<lanefu> now because i'm a member of hte org. i can make and manage branches on armbian/build
<lanefu> but I prefer not to
<TRS-80> But I mean, even being member of Org, you cannot (it seems) organize all the Armbian repos under one common folder (in GitHub). Amirite?
<lanefu> TRS-80: correct.... repos are stored flat in an organization.... You can make teams and associate repos to teams
<lanefu> but ultimately its just like an alias
<lanefu> gitlab lets you nest nicely
<lanefu> has a concept of groups
<TRS-80> I dunno, not as clean as I had hoped; we'll see how it goes I guess: https://github.com/TRSx80/armbian
<lanefu> TRS-80: yeah but that submodule is just pointing to the armbian repo
<lanefu> so other than for your own conviencence of checking 3 thigns out at once locally
<lanefu> there's no aspet of that which you'll use to interact with PRs back to armbian
<TRS-80> so it's like one way thing?
<TRS-80> n oway to do a pr from there?
epsilonKNOT has left #armbian [#armbian]
<lanefu> i believe hte answer is no
<TRS-80> Do you understand how much I love you guys to tolerate this GitHub retardation? :D
<TRS-80> reeee
<Liptibilly> Thanks Werner, but nothing work... I'll wait until this is fixed in armbian images.
Liptibilly has quit [Remote host closed the connection]
<lanefu> TRS-80: there ya go
Chevremw has joined #armbian
Chevremw has quit [Remote host closed the connection]
<Werner> Not sure what to fix since the file seems correctly in place....
<Werner> Maybe he was using legacy or even older legacy
<TRS-80> I think Recovery should be a new child topic under USER GUIDE parent, and move recovery stuff into there (including flashing u-boot stuff I am about to add) and then leave rest of "Advanced Features" stuff (wireless, freeze, docker, etc.) there where it is.
Liptibilly has joined #armbian
Liptibilly has quit [Remote host closed the connection]
tmaurice has joined #armbian
<Tony_mac32> my name is Tony, and I'm an ARMaholic
<Tony_mac32> that kind of recovery?
<Tony_mac32> :)
* Tony_mac32 hides
<Werner> hahahaha
<tmaurice> hey, weirdly enough the systemd + ansible issue I was bitching about 2 weeks ago (service in unknown state) disappeared in systemd 246
<tmaurice> \o/
<Tony_mac32> nice
<tmaurice> finally i can ansible my r2s <3
<tmaurice> is anyone here operating a helios64 NAS ?
<tmaurice> any catches I should be aware of ? I should recieve one soon
<Tony_mac32> mine is being shipped across the land at the moment
<lanefu> 246 is when systemd applied their fix. Ansible 2
<lanefu> 2.10 I think also has a workaround slated for next merge window
<tmaurice> i didnt update ansible, just systemd
<Tony_mac32> there is a forum thread, there are a couple odds and ends to take a look at.
<tmaurice> anyways 246 could be the default one in buster so we dont have to install it from backports ?
<tmaurice> cheers Tony_mac32 I'll have a look !
<tmaurice> when it arrives
<tmaurice> like in ages
<tmaurice> thanks royal mail
<lanefu> Itll be whatever debian has
<tmaurice> right
<tmaurice> i was just wondering
<tmaurice> cool cool thanks mate
<TRS-80> Tony_mac32: We expect a full forum writeup. With pictures! Or maybe a video.
<tmaurice> or both
<TRS-80> Or tmaurice Or whoever :)
<tmaurice> hahaha
<tmaurice> I already have an article to write about using a r2s to do ospf over vxlan over wireguard
<tmaurice> one thing at a time
<TRS-80> Tony_mac32: Your comment the other day made me question my thinking to wait about the ECC version.
<TRS-80> tmaurice: At your leisure. Keep it fun. No burnout. ;)
<tmaurice> ECC is for people that dont trust your ram, the same type of people that do unit testing
<TRS-80> do you even know who you are talking to REEEEE
<Tony_mac32> I just don't know how it will work on a non-ECC ram controller, but I'm also not that well versed in that part of this sort of thing
<tmaurice> @TRS-80: definitely, if I have the motivation I'll definitely do that tho!
<tmaurice> but tbh I expect that is just going to be assembling the thing, flashing an SD card and done
<tmaurice> and then backuping all my por^W debian isos
<TRS-80> I looked at the assembly doc (I was curious how the front panel connected[0]), assembly looks like a (minor) project
<TRS-80> [0] magnets (nicely)
<tmaurice> yeah there seem to be a lot of work required hahaha
<tmaurice> knowing myself I might just fuck it up
<tmaurice> I'm bad with HW
<TRS-80> docs looked very well done
<tmaurice> Hw being any kind of physical object
<tmaurice> the docs are amazing yes
<tmaurice> they really did it well
<tmaurice> and they update it quite often
<TRS-80> the whole project seems quite quality, I was actually surprised to read recent post complaining of fan noise (I can't imagine they used some cheap fans)
<tmaurice> Well I mean the fans canq be noisy, it's a NAS
<tmaurice> I would expect it to be noisy
<tmaurice> I'll tell you about the fan noise when I get mine but tbh that is something I would expect
<TRS-80> Yes I am very impressed with Kobol. And this from a generally cranky, cynical and disillusioned guy.
<tmaurice> cant have 80T of storage + 3.5Gbps network capacity without some noise :p
<TRS-80> Aren't the fans pretty big though? I would think so...
<TRS-80> (hope so)
<tmaurice> the fans are big yes
<tmaurice> and im not a big fan
<tmaurice> lel
<tmaurice> kill me now
<TRS-80> OK, well someone said they are still working on fine tuning, maybe stuck at 100% for now
<tmaurice> but yes kobol did a super good job, a bunch of things are missing in the official armbian image tho
<TRS-80> But still, large and good quality fans should not be noisy
<tmaurice> but overall 10/10
<tmaurice> worst case scenario if the fans are too noisy i would put it in the basement
<tmaurice> if i had a basement
<tmaurice> so like first i'll move
<tmaurice> thanks london housing
<tmaurice> I have really high expectations for this NAS
* TRS-80 lok over at multiple 140mm Corsair fans in a huge 750D case that I never think about (other than when the subject comes up) because I never hear them
<TRS-80> /s/lok/look/
<tmaurice> yeah worst case you can change the fans, but tbh given the chip it is, I'm not even sure you need 2 fans to cool it down
<tmaurice> that seems a bit overkill doesnt it ?
<TRS-80> That's what I am saying. More and bigger fans = less RPMs = less noise.
<tmaurice> fair enough yes
<TRS-80> And with all the care they seem to take, I would be very surprised they didn't pay any attention to that.
<TRS-80> Ergo, wait for firmware update. ;)
<tmaurice> also yes apparently they are PWM fans so basically adapting to the load
<tmaurice> fuck i want my toy now haha
<stipa> are the case on that NAS of aluminum?
<TRS-80> Not sure but they seem to be like bead blasted and powder coated or something; quite nice.
<stipa> i'm thinking, a bunch of alimunim like that would cool stuff nicely
<stipa> fans would help in very hot environments i guess
<TRS-80> lanefu: Vanilla Markdown (i.e., tabs for code blocks)? Or can we do GFM (code fences)?
<TRS-80> I am not even sure that is GFM. It says just "extended" https://www.markdownguide.org/extended-syntax/#fenced-code-blocks
<lanefu> You can use ```
<TRS-80> Sweet. In Emacs I can enter environment to edit in that language, but only inside ```
<TRS-80> Oh crap I should be making changes in a branch?
<TRS-80> lanefu: ^ I just remembered
<TRS-80> Hmm, well not according to https://github.com/armbian/documentation#quick-start (but I can update that, too)
Hokedli has joined #armbian
Redfoxmoon is now known as axp
axp is now known as Redfoxmoon
<TRS-80> Looks like we are hand editing mkdocs.yml now, instead of generating it with mkArmbianDocs.py? Otherwise I don't see any metadata to put things in the order we want in the menu. Amirite?
<Werner> Because generating a new mkdocs.yml screws everything up :)
tmaurice has quit [Ping timeout: 260 seconds]
<TRS-80> Yes there is no provision for ordering the items in the menu. I see lane trying to automate all the things though. :)
<Werner> Lots of things are automated but some things do not need that (yet). Since it does not happen everyday that new menu points are added I think it is fine to keep this task doing manually
<TRS-80> I agree
<TRS-80> So, I don't need to make my changes in a branch (before I get too far into this)?
Net147_ is now known as Net147
tmaurice has joined #armbian
<ArmbianTwitter> @Poddingue (Bruno Verachten 🍰): @marinusklasen Thanks a lot for pinging me about this subject. IMHO to this day, the easiest way without too much tinkering is to use an @OdroidH #XU4 or #MC1 Solo with a specific build of #ffmpeg : https://t.co/n3gVaxqMM7 (18s ago)
DarthGandalf has joined #armbian
Hokedli has quit [Quit: Konversation terminated!]
<DarthGandalf> Hello! I'm trying to use armbian kernel with with odroid-n2+ board but having problems. 5.9.6-meson64 (binary taken from the image) can't initialize network for me when doing it via "ip" command, but works with "ifconfig". The error is "Error: Nexthop has invalid gateway."
<DarthGandalf> But it starts working if I do "ifconfig eth0 up" after that
Hokedli has joined #armbian
Hokedli has quit [Client Quit]
<DarthGandalf> I also built 5.8.16+ from armbian sources, and this one works fine with network. But it cannot reboot: https://dpaste.org/DYfc
<DarthGandalf> 5.9.6-meson64 was rebooting fine
<lanefu> what image are you using?
<DarthGandalf> It started from the image from https://www.armbian.com/odroid-n2/ but I replaced everything except /boot and /lib/modules with /lib/firmware with gentoo arm64 stage3
<DarthGandalf> so, u-boot and one of the two kernels I mentioned is directly from that image
<IgorPec> n2 has currently some reboot issues, true
<lanefu> yeah some of the amlogic stuff reboots unreliably.. i've noticed mostly if you touch a HID device, it can impact it.... but headless usually okay
<DarthGandalf> I see. Reboot worked fine for me in 5.9.6-meson64 (so far)
<IgorPec> and see if it helps. this was recently added as from hardkernel
raver has quit [Read error: Connection reset by peer]
<IgorPec> its helping on C4 but possible this caused problem on N2 ... recompiling kernel witrhoiut this patch
tmaurice has quit [Ping timeout: 272 seconds]
<DarthGandalf> IgorPec: hm, I'll try to apply that patch to 5.8.16, maybe it'll fix reboot there
<IgorPec> no, remove it
<DarthGandalf> But you said "recently added". I assume 5.9.6 is newer than 5.8.16?
<IgorPec> or you want to have 5.8.y ? but i didn't notice troubles before 5.9.y i think
<IgorPec> yes
<DarthGandalf> 5.9.6 reboots fine for me
<IgorPec> i have troubles :)
<DarthGandalf> Network in 5.9.6 is broken
<DarthGandalf> I tried 5.8.16 because of troubles with the network
<IgorPec> i have two n2 here ... one moment
<IgorPec> 1st works with 5.9.8 fine
<IgorPec> is N2+
tmaurice has joined #armbian
<IgorPec> second one running 5.9.6 (normal N2) also ok
<DarthGandalf> by ok you mean reboot?
<IgorPec> network works
<DarthGandalf> Do you use "ip" or "ifconfig" to configure it?
<IgorPec> network manager
<DarthGandalf> the patch didn't fix reboot for me in 5.8.16
<TRS-80> If NM is running, I think that can interfere with trying to manually set things (e.g., with ip, etc.). Amirite?
<DarthGandalf> TRS-80: that's what I would expect
emOne has quit [Ping timeout: 265 seconds]
emOne has joined #armbian
<DarthGandalf> Perhaps such interference is what *fixes* it...
<lanefu> i shouldn't ask.. but why gentoo
<DarthGandalf> because why not :) and because configurable
<lanefu> yeah sounds about right
<lanefu> lol
<TRS-80> Code block highlighting is really bothering me on my local machine (via mkdocs serve) but does not appear on the real site (https://docs.armbian.com).
<lanefu> TRS-80: might be a few breadcrumbs here http://ix.io/2Ebt
<TRS-80> OSX? pip for Python 3 on Debian is pip3
<TRS-80> I thought I would update that in Docs as well (or at least a note)
<TRS-80> Anyway, I already went down quite rabbit hole about syntax highlighting. But I appreciate it.
<lanefu> ubuntu
<TRS-80> really? That doesn't get you Python 2 pip?
<TRS-80> they must have alias it I guess
<c0rnelius> pip3 is python3-pip
<TRS-80> c0rnelius: in Ubuntu?
<c0rnelius> So regular pip should be obvious
<lanefu> root@stuff:/home/lane# which pip
<lanefu> /usr/local/bin/pip
<lanefu> root@stuff:/home/lane# pip --version
<lanefu> pip 20.2.3 from /usr/local/lib/python3.8/dist-packages/pip (python 3.8)
<c0rnelius> I don't see why you they would go through all this effort to change package naming. But then again Ubuntu does dumb shit.
<lanefu> will the pip(3) command is deprecated altogether
<lanefu> the most pythonic way is python -m pip blah
<TRS-80> yeah, the dist aliased it (assuming you didn't do that yourself)
<lanefu> anyway that's probably my doing
<lanefu> since /usr/local
<lanefu> with yeah by default there's no update-alternatieves setting to point "python" to python3
<c0rnelius> pip3 shouldn't be depreciated?
<c0rnelius> depending on the release its either python-pip or python3-pip... but Ubuntu might have adjusted this inb Focal.
* lanefu runs away from this conversation
<c0rnelius> pretty sure last time I was in ubuntu focal python3-pip installed fine, but complained when I used sudo to do pip installs.
<TRS-80> c0rnelius: I thought it wasn't good to do that, because it can interfere with the system Python packages?
<c0rnelius> it can, yes
<c0rnelius> but as I was only using it to install speedtest-cli I didn't really care :D
<TRS-80> On Debian at least, pip3 install foo now defaults to pip3 --user install foo (which ends up in ~/.local/lib/pythonX.Y/site-packages/foo) which I like a lot
<c0rnelius> pip also looks to see what is there before installing anyway, so its just precaution honestly. IF ur doing a large pip install but decide to install specific python packages system wide, pip will see this and skip over it.
<DarthGandalf> Huh. Adding "sleep 10" before the "ip" commands fixed it on 5.9 O_o
raver has joined #armbian
<TRS-80> DarthGandalf: should like some race condition
<TRS-80> s/should/sound/
<ArmbianHelper> TRS-80 meant to say: DarthGandalf: sound like some race condition
<TRS-80> In theory, I don't like systemd either, but it does take care of stuff like that with targets.
<c0rnelius> Armvuan? :)
<TRS-80> Like the interface is not up yet or something like that. I had some similar problem with mounting an NFS mount.
<TRS-80> c0rnelius: :D
lanefu has left #armbian ["WeeChat 2.8"]
Spads has quit [Remote host closed the connection]
<DarthGandalf> Found the error message (no idea why it wasn't shown before): https://dpaste.com/HXT6UZDTP
<IgorPec> with lates image?
<DarthGandalf> with 5.9.6
Spads has joined #armbian
Spads has joined #armbian
<IgorPec> well, we had this bug ... but it was fixed some time ago
tmaurice has quit [Ping timeout: 264 seconds]
<DarthGandalf> Fixed how?
<IgorPec> with a patch . i believe it was sent upstream
Spads has quit [Remote host closed the connection]
Spads has joined #armbian
Spads has quit [Remote host closed the connection]
Spads has joined #armbian
Spads has quit [Changing host]
Spads has joined #armbian
Redfoxmoon has quit [Read error: Connection reset by peer]
Redfoxmoon has joined #armbian
Redfoxmoon has quit [Changing host]
Redfoxmoon has joined #armbian
<DarthGandalf> IgorPec: I took 5.9.8 from kernel.org, applied armbian patches with meson64-current config, so the patch is definitely in now. It still fails though
<IgorPec> we are talking sources from kernel.org and add those patches
<DarthGandalf> (during application of patches, some wireless didn't cleanly apply - the patched files don't exist)
<IgorPec> that's not related to this
<IgorPec> well, hard to say ... probably your hw is out of spces a bit, try changing cables
<IgorPec> i have two working N2 here and see forum if there are any like reports recently
<DarthGandalf> I don't see how cables matter here. Since it's static IP, the cable doesn't even need to be inserted there
<IgorPec> no, this is below network stack
<DarthGandalf> IgorPec: heh. Removing that patch fixed the network for me
<DarthGandalf> as in, putting `reset-gpios = <&gpio GPIOZ_15 (GPIO_ACTIVE_LOW | GPIO_OPEN_DRAIN)>;` back
<IgorPec> yeah, possible
<IgorPec> welcome to our weird world :)
<DarthGandalf> Thanks for link!
raver has quit [Remote host closed the connection]
<DarthGandalf> Could you reopen this issue?
<IgorPec> we need more then one report
<DarthGandalf> I have only one N2, sorry
<TRS-80> lol
<IgorPec> its noted, but we have many N2 around and nobody have reported this
<IgorPec> if its a hw quality issue, we are not correct address
<TRS-80> Not laughing at you DarthGandalf, I just found that way more funny than I probably should have.
<IgorPec> what?
<DarthGandalf> TRS-80: I understand that
<TRS-80> IgorPec: Where he said "I have only one N2, sorry"
<DarthGandalf> I wonder how many of these N2 around use NM and how many use ip, and how many use ifconfig
<DarthGandalf> There was some difference between ip and ifconfig
<TRS-80> ifconfig I thought is supposed to be deprecated for long time now?
<DarthGandalf> It is
<DarthGandalf> But it worked better with this specific issue
<IgorPec> not sure if this has anything to do with
* DarthGandalf shrugs
<IgorPec> the problem is if i change this, we will have troubles
<IgorPec> majority
<IgorPec> and we will be running in circles
<stipa> and there wiould be only one N2 running Armbian in the world
<TRS-80> Tough break for one individual, but Igor has to keep in mind the greater good. Hopefully you can understand.
<DarthGandalf> TRS-80: I'm not asking you to revert the patch
<DarthGandalf> I was asking to reopen the issue on github
<DarthGandalf> Maybe there is some solution for the issue which will work for every hardware
<IgorPec> we spent a lot of time on this and we can't afford to spent more :(
<IgorPec> i would escalatge to manufacturer
<stipa> maybe they changed something in hardware
<IgorPec> we have seen that, possible, but unliekly with Hardkernel
<DarthGandalf> stipa: well, this one is N2+
<DarthGandalf> But I saw report in the same bug that N2+ was also fixed by the patch
<stipa> or the board has small fault in manufacturing somewhere
<stipa> oh
<stipa> N2+
<stipa> that could explain the problem
<DarthGandalf> it could? how?
<DarthGandalf> If that's the case, I'd propose adding a patch which puts that line to arch/arm64/boot/dts/amlogic/meson-g12b-odroid-n2-plus.dts
<DarthGandalf> But <DarthGandalf> But I saw report in the same bug that N2+ was also fixed by the patch
<DarthGandalf> so dunno
<IgorPec> exactly ... its a weird problem
<DarthGandalf> Anyway, I updated my comment to include more details. Maybe that will help others to google for the error message
<stipa> right, more poeple will buy N2+
<stipa> it's new as i see
<stipa> few months
<stipa> old
<stipa> it was promotodet to the market around 3 months ago
<stipa> promotoed*
<stipa> promoted**
<stipa> DarthGandalf: when did you get it?
<DarthGandalf> This Friday
<stipa> right
<stipa> soon there will be more than "you" people complaining
<DarthGandalf> We'll see
<stipa> yeah