<PhoenixMage>
After some advice. I have created a class that will create a grub-efi fitImage that can be chainloaded by u-boot. The logic is because some of the boards I am using have more recent dtb's in the kernel source then u-boot (rock 5b). My question is how I should integrate this into bitbake? At this point I have created a custom KERNEL_CLASS though I am not sure this is best way to go. Perhaps a better
<PhoenixMage>
idea would be to bbappend the grub-efi or u-boot recipes?
<fennec>
i'm trying to drive ws2812b leds through spi on a verdin-imx8mp (using spidev).
<fennec>
it works just fine on a raspberry pi 3, but on the imx8mp i'm facing inconsistant behavior
<fennec>
using a logic analyser doesn't really work, as the spi keeps sending signals after it has been used once.
<fennec>
i tried disabling DMA, which gives a bit more consistant results but is still actually wrong
<fennec>
what can i try next?
<fennec>
i have been checking a lot of things, including the logs of this channel
<fennec>
there was a person modifying spi drivers especially for ws2812b, unfortunately the pastebin of their code is now gone
Thorn has joined #yocto
alessioigor has joined #yocto
rob_w has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
Guest98 has joined #yocto
olani has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
mckoan|away is now known as mckoan
<mckoan>
good morning
<mckoan>
fennec: do you see signals on the SPI bus ?
<fennec>
mckoan: i do yes
<fennec>
the issue is that they are almost garbage
<fennec>
and as mentioned i can't really use the logic analyser
<fennec>
since there is always noise on the signal
<fennec>
i have no way to have it start recording when i'm sending data
<fennec>
the sampling starts right away
Schlumpf has joined #yocto
<fennec>
and this is not an issue with the analyser afaict
<fennec>
since it works perfectly on the pi
xmn has joined #yocto
<barath>
morning
<Guest98>
morning
<barath>
anyone know of a "canonical" way to get python modules into yocto? specifically I'm trying to write a custom fetcher. I see that meta-rust does this adding an anonymous function via a bbclass, but this function will be called for every file. not sure if there's a better way https://github.com/meta-rust/meta-rust/blob/master/classes/crate-fetch.bbclass#L12
goliath has joined #yocto
Kubu_work has joined #yocto
rfuentess has joined #yocto
LocutusOfBorg has quit [Quit: ZNC 1.8.2+deb3 - https://znc.in]
manuel_ has joined #yocto
LocutusOfBorg has joined #yocto
gsalazar has joined #yocto
<LetoThe2nd>
yo dudX
<mckoan>
LetoThe2nd: welcome back!
<jclsn>
Good morning
<jclsn>
mckoan: Was he ever gone?
<kanavin_>
barath, pypi class perhaps?
wooosaiiii has quit [Quit: wooosaiiii]
wooosaiiii has joined #yocto
<barath>
mhm, but was thinking of the best way of making the module available to yocto's fetch2 class / fetcher system
<LetoThe2nd>
jclsn: last week was at Prague, for EOSS-
<Guest98>
if you recorded and shared the event in prague, where can i find it?
<jclsn>
LetoThe2nd: Ah yeah, how was it? Asked my manager if I could attend, but it was to late to get the costs covered.
<LetoThe2nd>
jclsn: I liked it very much and I think generally it was very positive, but it sure depends on your personal preference.
<LetoThe2nd>
Guest98: expect videos on YT in about 6 weeks, give or take.
<Guest98>
LetoThe2nd thanks.
SheatNoisette has joined #yocto
Perflosopher has quit [Ping timeout: 260 seconds]
xmn has quit [Ping timeout: 246 seconds]
<jclsn>
LetoThe2nd: Did they serve beer?
leon-anavi has joined #yocto
<LetoThe2nd>
jclsn: yes.
xmn has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
<kanavin_>
LetoThe2nd served beer too
<kanavin_>
but only if you would enter yocto BoF
<kanavin_>
and threw chocolate into the audience
<sudip>
and Whiskey too
<LocutusOfBorg>
hello, I have a doubt
<LocutusOfBorg>
if I want my service to be not restarted during upgrade
<LocutusOfBorg>
e.g. I want to restart the target after the installation is fully done
<LocutusOfBorg>
should I just systemd_postinst() override, or do we have something more clever?
<LocutusOfBorg>
I know Debian did something for this, because of e.g. ssh connections on daemon restart
SheatNoisette has quit [Quit: Client closed]
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
amsobr has joined #yocto
fennec has quit [Quit: Client closed]
Perflosopher has joined #yocto
<kanavin_>
LocutusOfBorg, generally there's little interest in updating systems via classical package system. not robust enough for embedded.
<LocutusOfBorg>
so my solution might be the easiest one
<LocutusOfBorg>
sadly this is what customer wants :D
<kanavin_>
all those rpm/deb/ipk transactions just don't seem resistant to power cuts. you may end up with booting a half-updated system when the power comes back.
Maxxed has quit [Server closed connection]
Maxxed has joined #yocto
JakubKunert[m] has joined #yocto
starblue3 has quit [Ping timeout: 246 seconds]
starblue3 has joined #yocto
xmn has quit [Quit: ZZZzzz…]
amitk_ has joined #yocto
amsobr has quit [Quit: Konversation terminated!]
amsobr has joined #yocto
adrianf has joined #yocto
jonmason has quit [Server closed connection]
jonmason has joined #yocto
florian_kc has joined #yocto
<Guest98>
there is a bbappend recipe and i added module.patch and fragment.cfg with "SRC_URI_append_mymachine". but, while patch is approved, fragment.cfg is not. why could it be?
BobPungartnik has quit [Quit: Leaving]
mkazantsev has joined #yocto
florian_kc has quit [Ping timeout: 246 seconds]
<amsobr>
check out the recipe and see if there's any specific variable in use to list configuration fragments. It may happen that you need to add that fragment to some variable
<mcfrisk>
Guest98: syntax should be SRC_URI:append:machine on all maintained branches. Then config fragments aren't applied if you have a full static defconfig. Check your kernel.bbclass and bitbake -e output from kernel recipe for details.
<amsobr>
mcfrisk: I assume the syntax is right because adding the patch worked.
<amsobr>
Guest98: on which branch are you?
<Guest98>
amsobr zeus. syntax is correct.
mkazantsev has quit [Quit: Leaving]
mkazantsev has joined #yocto
<amsobr>
OK, but it's a good point about defconfig, might be worth looking.
rfuentess has quit [Remote host closed the connection]
<amsobr>
If possible, you should consider upgrading to a newer release :)
wooosaiiii has quit [Quit: wooosaiiii]
wooosaiiii has joined #yocto
rfuentess has joined #yocto
<Guest98>
amsobr thanks.
mkazantsev has quit [Quit: Leaving]
fennec has joined #yocto
<fennec>
getting back to my spi question if i may
<fennec>
what can cause the spi to be always sending signal?
<fennec>
and what can i try to avoid it, i'm running out of ideas (?)
<JerryM>
hey folks a little while ago I send some patches to the oe-devel mailing list for meta-java, unfortunately I haven't had any response nor have I been able to contact the listed maintainers, I was wondering what to do in this situation
<LetoThe2nd>
JerryM: in a perfect world: ditch java ;-)
<LetoThe2nd>
JerryM: in a more realistic world, it is indeed unfortunate - try again, and if nothing else works - fork.
<rburton>
the list is "the list of git repositories on this server"
<JerryM>
so how would something be added to said server
<rburton>
first, you don't _have_ to keep it on that server, there are major layers (meta-clang, for example) that live on github. we offer a git server for convenience.
<rburton>
just explain why your repo should be hosted
<barath>
someone know how I can add tools to the pseudo sysroot? I want write a custom fetcher, and it needs a binary which is installed on the host but cant be found because it seems to not exist in the sysroots available to tasks run by bb.process.run
<JerryM>
rburton I was just wondering about the process, thank you for the clarification (y)
<sudip>
rburton: fyi, link to meta-nim in the layer index is 404
Kubu_work1 has joined #yocto
belsirk has joined #yocto
rfuentess has quit [Ping timeout: 252 seconds]
kpo has joined #yocto
fennec20 has joined #yocto
Kubu_work1 has quit [Quit: Leaving.]
fennec has quit [Ping timeout: 246 seconds]
Kubu_work has joined #yocto
Omax has quit [Remote host closed the connection]
<barath>
answering my own question from above, it's the HOSTTOOLS variable
wooosaiiii has quit [Quit: wooosaiiii]
wooosaiiii has joined #yocto
fennec20 has quit [Quit: Client closed]
JerryM has quit [Quit: Client closed]
JerryM has joined #yocto
Schlumpf has joined #yocto
fennec has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
zhmylove has joined #yocto
Guest98 has quit [Quit: Client closed]
<rburton>
sudip: mailed the maintainer, can't find it elsewhere on github
prabhakarlad has quit [Quit: Client closed]
sakoman has joined #yocto
rob_w has quit [Remote host closed the connection]
belsirk is now known as rfuentess
Schlumpf8 has joined #yocto
Schlumpf has quit [Ping timeout: 246 seconds]
manuel_ has quit [Ping timeout: 246 seconds]
kpo has quit [Ping timeout: 246 seconds]
tprrt has quit [Server closed connection]
tprrt has joined #yocto
Crofton has quit [Server closed connection]
Crofton has joined #yocto
zhmylove has quit [Read error: Connection reset by peer]
goliath has quit [Quit: SIGSEGV]
olani has quit [Remote host closed the connection]
prabhakarlad has joined #yocto
<sudip>
thanks rburton
Omax has joined #yocto
JerryM has quit [Quit: Client closed]
<Wenqing>
there is nothing in yocto to prevent cache poisoning?
<rburton>
i presume you mean a malicious actor injecting bad information into sstate?
<Wenqing>
yeap
<rburton>
if your sstate is local and you worry about that then you have bigger problems
<Wenqing>
what is my sstate is in AWS s3?
<rburton>
if your sstate is remote then the remote can turn on signed sstate and you can verify the signers are good
<Wenqing>
what if*
alessioigor has quit [Quit: alessioigor]
<rburton>
by remote i mean "if someone else controls it"
alessioigor has joined #yocto
<rburton>
if its in s3 but its your repo then presumably you trust it
<rburton>
if you dont' trust that then you don't trust eg your layers to not be hijacked either
<rburton>
but either way, look up sstate signing
<Wenqing>
ok, thanks
Schlumpf8 has quit [Quit: Client closed]
xmn has joined #yocto
ptsneves has quit [Ping timeout: 246 seconds]
<Wenqing>
is there a similar feature for DL_DIR cache as well?
rfuentess has quit [Remote host closed the connection]
jetm- has joined #yocto
philmd- has joined #yocto
<rburton>
Wenqing: downloaded tarballs already have checksums in the recipe, and git clones have a SHA. the only attack is has collisions
<rburton>
hash collisions
<Wenqing>
ok, I forget the recipe checksum. Thanks rburton
otavio has quit [Remote host closed the connection]
otavio has joined #yocto
mckoan is now known as mckoan|away
Net147 has quit [Ping timeout: 240 seconds]
manuel_ has joined #yocto
goliath has joined #yocto
Net147 has joined #yocto
Net147 has joined #yocto
Net147 has quit [Changing host]
Kubu_work has quit [Quit: Leaving.]
nemik has quit [Ping timeout: 246 seconds]
nemik has joined #yocto
nemik has quit [Ping timeout: 246 seconds]
nemik has joined #yocto
leon-anavi has quit [Quit: Leaving]
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
geoff_ has joined #yocto
geoffhp has quit [Read error: Connection reset by peer]
florian has joined #yocto
gsalazar has quit [Remote host closed the connection]
<khem>
Would have been good if Alex would have sent a patch to meta-oe to remove it alongside when it was added to oe-core, nevertheless thanks for letting me know
<khem>
is there a tool to find duplicate recipes ?