Werner changed the topic of #armbian to: armbian - Linux for ARM development boards | Upcoming Release: 20.05 "Kagu" | www.armbian.com | Github: github.com/armbian | Commits: #armbian-commits | Forums Feed: #armbian-rss | This channel is logged -> irc.armbian.com
xec has joined #armbian
xecuter has quit [Ping timeout: 260 seconds]
_whitelogger has joined #armbian
ChriChri_ has joined #armbian
ChriChri has quit [Ping timeout: 256 seconds]
ChriChri_ is now known as ChriChri
Tenkawa has quit [Quit: Leaving.]
Tenkawa has joined #armbian
c0rnelius_ has joined #armbian
c0rnelius_ has quit [Client Quit]
c0rnelius has quit [*.net *.split]
solderfumes[m] has quit [Ping timeout: 240 seconds]
psydruid has quit [Ping timeout: 252 seconds]
nekomancer[m] has quit [Ping timeout: 244 seconds]
hyperreal has quit [Ping timeout: 244 seconds]
JuniorJPDJ has quit [Ping timeout: 256 seconds]
solderfumes[m] has joined #armbian
psydruid has joined #armbian
hyperreal has joined #armbian
nekomancer[m] has joined #armbian
Tenkawa has quit [Quit: Leaving.]
willmore has quit [Ping timeout: 240 seconds]
JuniorJPDJ has joined #armbian
willmore has joined #armbian
dddddd has quit [Ping timeout: 265 seconds]
torv has quit [Remote host closed the connection]
archetech has quit [Quit: Konversation terminated!]
torv has joined #armbian
<Werner> Good morning
<IgorPec> morning
Werner changed the topic of #armbian to: armbian - Linux for ARM development boards | Current Release: 20.05.1 "Kagu" | www.armbian.com | Github: github.com/armbian | Commits: #armbian-commits | Forums Feed: #armbian-rss | This channel is logged -> irc.armbian.com
<Werner> Woohoo \o/
IgorPec has quit [Remote host closed the connection]
macc24 has joined #armbian
Elpaulo has quit [Quit: Elpaulo]
archetech has joined #armbian
<indy> hi all, how much is different building kernel for some board compared to building kernel using https://kernel-team.pages.debian.net/kernel-handbook/ch-common-tasks.html ?
<indy> i my case i need rockpis
<Werner> I guess the main difference is that everything is done in cross compile. Users dont see much about all the commands involved in the building process since the build script does the job for them.
<indy> problem is that with 4.4 legacy kernel (which run on rockpi s 100%) comes with uncomplete linux-headers package because i'm not able to build out-of-tree modules
<indy> and 5.6 kernel sometime runs. sometime not
<Werner> The header package for 4.4 is incomplete when building with the build script?
<indy> it is installable but module building fails with some vdso incomplete something
<indy> ...message
<Werner> Are you sure that it relys on the headers and not on the kernel source itself maybe?
<indy> it does rely on source tree
<Werner> Try installing the linux-source package as well
<indy> tried yesterday image with 5.6.15 and it works
<indy> weird
<indy> i put sd card - stuck after loading kernel
<indy> reset buttong worked
<Werner> Newer kernels are unfortunately not known for stability...
<indy> looks like
hellotesttttt has joined #armbian
mpmc has quit [Quit: ZNC - https://znc.in]
mpmc has joined #armbian
macc24 has quit [Ping timeout: 240 seconds]
macc24 has joined #armbian
dddddd has joined #armbian
hellotesttttt has quit [Remote host closed the connection]
Tenkawa has joined #armbian
macc24 has quit [Ping timeout: 258 seconds]
macc24 has joined #armbian
IgorPec has joined #armbian
IgorPec has quit [Changing host]
IgorPec has joined #armbian
<Werner> IgorPec, I did use the build script. Once on Bionic and once on Focal. Did not make any difference at booting beforehand
macc24 has quit [Quit: WeeChat 2.8]
macc24 has joined #armbian
macc24 has quit [Ping timeout: 256 seconds]
<IgorPec> so its fauiling on focal build host too?
<Werner> IgorPec, yes
<IgorPec> ok, i want to know if you issued chroot on your x86 desktop to run update
<IgorPec> without arm emultor
<Werner> No, I used arm64, OPi1+ to say
<IgorPec> ahaa, on arm64 builds this seems fine
<IgorPec> i tried a few arm64 focal builds, no problem
<Werner> No sure if you get me correctly. The image was for OPi1 which is armhf and I used arm64 to fix this image
<IgorPec> yes - correct
<Werner> Okay
<IgorPec> it seems that armhf update is broken
<IgorPec> by using arm64 binary there
<Werner> Interesting
<IgorPec> it juse a guess, not sure if this will solve, but worth trying
<Werner> So if this works we have an upstream bug from Ubuntu's armhf initramfs-tools I guess
<IgorPec> they are dropping armhf
<IgorPec> yes, this is upstream ... if this is the problem
<Werner> Did not know they were going to drop 32bit arm but would make sense since they dropped x86 too
<IgorPec> well, not 100% sure, but possible
sassinak-work has quit [Ping timeout: 272 seconds]
sassinak-work has joined #armbian
<IgorPec> werner: hack works
<Werner> Nice
<IgorPec> but to be sure, we need few more tests
<IgorPec> it could be some random issue
<Werner> Which boards did you test?
<IgorPec> nanopi m1
<Werner> Send me the line, I'll test the OPi One
<Werner> With Bionic build host
<IgorPec> add QEMU_BINARY="qemu-aarch64-static"
<IgorPec> i will add [[ if release == focal ]] later
<IgorPec> actually you can also send a PR and i go for a beer ;) i need one
<IgorPec> ... if it will also work for you of course
epsilonKNOT has quit [Quit: ZNC 1.7.5 - https://znc.in]
<Werner> Just added it and started build
heroapoorva has quit [Quit: WeeChat 2.8]
<IgorPec> next thing to try is https://patchwork.kernel.org/patch/7435061/
drobo_00 has joined #armbian
<Werner> Funny. I stumbled across CONFIG_BLK_DEV_RAM_SIZE today as well as I was searching for a solution to this U-Boot issue
<IgorPec> i will only change that for start
drobo_00 has quit [Client Quit]
epsilonKNOT has joined #armbian
macc24 has joined #armbian
<IgorPec> works as well :)
<IgorPec> +CONFIG_BLK_DEV_RAM_COUNT=16
<IgorPec> +CONFIG_BLK_DEV_RAM_SIZE=8192
<Werner> OPi1 does not boot. Let me double check that I did not made a mistake somewhere
<Werner> if [[ release == focal ]]; then
<Werner> QEMU_BINARY="qemu-aarch64-static"
epsilonKNOT has quit [Quit: ZNC 1.7.5 - https://znc.in]
epsilonKNOT has joined #armbian
DaRock has quit [Ping timeout: 260 seconds]
Wernertest has joined #armbian
<Werner> _recent
<Wernertest> AR-294: sunxi current on Focal boot broken, reported by Werner at 2020-06-01. Status: To Do.
<Werner> _getissue AR-123
<Wernertest> AR-123: OpenHAB2 installation is failing, reported by Igor Pecovnik at 2019-12-27. Status: Done.
<Werner> _part
Wernertest has left #armbian ["Limnoria 2020.04.11"]
<Werner> Working on a new toy in the meantime :)
<IgorPec> if [[ release == focal ]]; then
<IgorPec> $RELEASE == focal
<ArmbianTwitter> IgorPec: Error: "RELEASE" is not a valid command.
<IgorPec> "$RELEASE == focal "
<Werner> Ah yes, of course... head vs table
<Werner> I do too much Python lately
<IgorPec> you don't need to add if for testing, just force it
<IgorPec> i will remake opione, zero, duo on a build host with kernel changes
<Werner> Still got two builds running, both OPi1, one with bionic as host, the other with focal as host
<Werner> Just for the heck of it to see if there is a difference
<IgorPec> my build host for releases is bionic
<Werner> Again, OPi one does not boot
<Werner> Trying the focal host image next
<Werner> Nope. Either I did a mistake somewhere and cannot find it or the hack does not work for the OPi One at least
<IgorPec> in a few minutes i will have them online
<Werner> Okay
<IgorPec> 10-15
Strykar has quit [Quit: /quit]
Strykar has joined #armbian
<Werner> Anything new?
<IgorPec> yes
<IgorPec> uploaded ... moment
<Werner> Writing to sd...
<Werner> Lets hope I made just a mistake when creating mine...
<IgorPec> i hope we nailed the problem
<Werner> We did not...Shi..
<IgorPec> the same? :(
<Werner> Yeah, Bad CRC
<Werner> Lets see if I can "fix" it again with the other SBC
<IgorPec> ok, now adding ugly hack QEMU_BINARY="qemu-aarch64-static"
archetech has quit [Quit: Leaving]
<IgorPec> this will be faster. Then beer ... :)
<IgorPec> working or not working
<Werner> Fixed it with the arm64 board again
<Werner> [ 6.042188] thermal thermal_zone0: binding zone cpu_thermal with cdev thermal-cpufreq-0 failed:-22
<Werner> That is new...
<IgorPec> perhaps we are running low on space? huh
<Werner> low space?
<IgorPec> image size is determined with some overhead ... perhaps this doesn't work good enough for focal
<Werner> Then aarch64 images should fail as well?
<IgorPec> yeah, but i am not 100% sure that either
<IgorPec> redownload
<Werner> Burning...
<Werner> The Jira bot feature will mostly be written from scratch. The best addon for the bot I could find that handles Jira was six years old and was relying on a library that seem not to even exist in this form anymore...I'm excited about that
<IgorPec> what can we extract from that bot?
<Werner> Theoretically everything can be seen or edited via the bot. Though that is a bit too much for me. I will only cover things that are publicly available and do not need any authentication.
<IgorPec> doesn't help, my m1 is not booting
<Werner> Some for the One
<Werner> *Same
<IgorPec> ok ... changing size
<Werner> Is this even longer necessary? There is no "default" branch anymore
<IgorPec> can be removed
<IgorPec> perhaps $INITRD_ARCH is empty in chroot
<Werner> Yes and no. It is empty until the 99-uboot call calls the armbian-release file
<IgorPec> ahaa
<IgorPec> ok, images are rebuild
<Werner> Downloading
<Werner> Writing
<IgorPec> no need, the same
<IgorPec> fuck
<HerculeP> umm, Armbian_20.05.1_Odroidc2_focal_current_5.6.15_desktop.img doesn't boot either :(
<Werner> I have another OPi One running as my PiHole here...let me check if I can fix a One image with another One
<IgorPec> herculeP: yeah, the same problem
macc24 has quit [Quit: WeeChat 2.8]
<IgorPec> and we have no clue yet
<Werner> Worked...Now lets try the same on a intel machine
<Werner> What we know so far is that the bad CRC can be fixed with chrooting into the sd card from another armhf/arm64 and simply executing update-initramfs -u
<IgorPec> we can try running it once again in the sctipz
<Werner> chroot into the sd card on x86 does not work which is to be expected. Exec format error
<Werner> IgorPec, Makes no sense but well if it works....
<Werner> Fixing the ramdisk via QEMU works as well. Takes longer than native and spams unsupported syscalls 412 und 403 but works.
<IgorPec> so we didn't get anywhere :(
<Werner> That pretty much covers the situation, yes.
<Werner> update-initramfs -uv -k ${VER}-${LINUXFAMILY}
<Werner> What I did was just update-initramfs -u
macc24 has joined #armbian
<IgorPec> -v = verbose -k version Specify kernel version or 'all'
<Werner> Yeah I know, I am just looking for things that are different from what I just did with qemu by hand
<IgorPec> perhaps some utility is missing?
<Werner> Then something should yell ERROR
<IgorPec> "/usr/sbin/update-initramfs"
<IgorPec> if scripts are made well, yes, but not always are
<Werner> True
<IgorPec> do we have /usr/bin/basename on focal cli
<Werner> Yes
<IgorPec> dpkg-trigger
nekomancer[m] has quit [Ping timeout: 256 seconds]
<Werner> Yes
JuniorJPDJ has quit [Ping timeout: 244 seconds]
solderfumes[m] has quit [Ping timeout: 246 seconds]
hyperreal has quit [Ping timeout: 260 seconds]
psydruid has quit [Ping timeout: 260 seconds]
<IgorPec> chroot: failed to run command ‘/bin/bash’: No such file or directory
<Werner> Hm I did use "chroot target/ qemu-arm-static /bin/bash"
<Werner> What happens this way? chroot $chroot_target $QEMU_BINARY /bin/bash -c "$update_initramfs_cmd" >> $DEST/debug/install.log 2>&1
<IgorPec> yes, this is the command
<Werner> Besides the $QEMU_BINARY
<IgorPec> qemu: Unsupported syscall: 412 is after each line when creating initird on broken image
<Werner> Yup, spams the console
<Werner> Not with Focal as build host though
<IgorPec> and not on Mint
<Werner> Adding $QEMU_BINARY did not work either
<IgorPec> yeah, i know.
<IgorPec> running low on ideas :(
<Werner> Stash focal images for now and put bionic back...
<IgorPec> that's the last option to go with
<Werner> Best to do to minimize the impact for now
<IgorPec> there is one nasty workaround
<IgorPec> prepare ramdisks and copy binaries at the end :) but thats lame
<Werner> Let me try one last thing before I give up on it for today
<IgorPec> yes, i will also do a few more
<IgorPec> then done
<IgorPec> i compiled a working one now, but i haven't change anything
phantomse[m] has joined #armbian
<Werner> W: Couldn't identify type of root file system for fsck hook
<IgorPec> where did you get that?
<Werner> From reading through the install.log
<IgorPec> could be something
<Werner> What if the initramfs update happens before everything is moved into the blank image?
<IgorPec> no, but fstab could not be read / be present for some reason?
<IgorPec> stop there and check if file is present at that time and if content is correct
<IgorPec> i build an image right now
<IgorPec> it has those errors in the log but its booting
<IgorPec> now :)
<Werner> I just copied the update initramfs command right after umounting the image just to see what happens. The output was the same. So the update could also be applied without being inside the mounted image
<IgorPec> you mean without qemu?
<Werner> LOL
<Werner> Mine is booting to?
<IgorPec> don't understand
<Werner> Me neither
<IgorPec> lets link forum to this conversation and stop
<Werner> I copied that line between 558 and 559
<Werner> With the $QEMU part
<IgorPec> yes, it should probably work as well
<Werner> Still makes no sense to my why but for now at least something
<Werner> *me
<IgorPec> ok, i will add some warning on the download pages
psydruid has joined #armbian
andrevallestero[ has joined #armbian
hyperreal has joined #armbian
JuniorJPDJ has joined #armbian
solderfumes[m] has joined #armbian
xlogik has joined #armbian
nekomancer[m] has joined #armbian
<IgorPec> will be this way ok=
<Werner> yeah
<IgorPec> changed to 1.6.2020 Some Focal images are. Please wait until we fix them!
<Werner> missing "broken"
<IgorPec> added
<Werner> Well that was an interesting Monday I guess...
<IgorPec> yeah, lets call it off for today ... users are warned
<Werner> Right. I am out for today. Bye
<IgorPec> tnx cya
<mrueg> shot in the blue: spaces in $chroot_target?
<mrueg> would try to quote that one everywhere to make sure it's not splitting on space
<IgorPec> we don't use windows :)
<IgorPec> no, we don't have spaces ... and that would not explain why works everywhere else ... just not focal
<HerculeP> hehe
* HerculeP ducks
<mrueg> IgorPec: it could depend on $MOUNT and thus on $RELEASE
<mrueg> i don't see anything obvious there though
<IgorPec> yeah, i am totally low on ideas and have to stop for today. brain is not working
<IgorPec> i hope someone else will sove this one :)
<IgorPec> one can see what we tried, not to repeat
<Miouyouyou> So, what's the issue actually ? OPi image is b0rked ?
<HerculeP> and others
<Miouyouyou> Ah, the ramdisk breaks out
<IgorPec> yeah, mainly armhf builds
<IgorPec> not sure if aarch64 as well
<IgorPec> i go get some beer, otherwise my head will go off :)
<Miouyouyou> Alright :3
Tenkawa has quit [Quit: Leaving.]
Tenkawa has joined #armbian
<mrueg> Cheers!
macc24 has quit [Ping timeout: 265 seconds]
<ArmbianTwitter> @armbian (armbian): RT @guidol70: Its alive 😉 #imsaisim from #z80pack does now work on my NanoPi A64 with #armbian desktop buster #altairsim does now also wor… (2s ago)
macc24 has joined #armbian
macc24 has quit [Quit: WeeChat 2.8]
sassinak-work has quit [Ping timeout: 240 seconds]
sassinak-work has joined #armbian