qschulz has quit [Remote host closed the connection]
qschulz has joined #yocto
Vonter has quit [Ping timeout: 260 seconds]
belsirk is now known as rfuentess
dmoseley has quit [Excess Flood]
dmoseley has joined #yocto
Vonter has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
olani_ has joined #yocto
dvergatal has quit [Quit: leaving]
<Guest6>
Hy i never helped an open source project and since i really like the yocto project is there a part where i could contribute? I can write C, C++, Java and a little bit of python.
<rburton>
Guest6: if you want to do some recipe work then there's a huge patchbomb of upgrades for meta-oe that the tooling just sent. the successful ones need verification, and the failing ones need a human to fix. https://lore.kernel.org/openembedded-devel/?q=AUH
<Guest6>
this sounds like something to get into it
<Guest6>
@rburton thanks i willt take a look on it
<rburton>
feel free to ask here for any more specific advice or if you have questions
<LetoThe2nd>
rburton: heh, spreading auhbombs!
<LetoThe2nd>
as a general heads up, who will be at plumbers?
BobPungartnik has joined #yocto
starblue has quit [Ping timeout: 245 seconds]
starblue has joined #yocto
olani- has quit [Remote host closed the connection]
<kanavin>
RP: thanks for those, the plan for me is to continue with oe-replicate-build and get it to a functional prototype asap, but I can do something else if you'd like me to
<kanavin>
I certainly would like to re-enable sstate mirror test as well
otavio has quit [Remote host closed the connection]
<RP>
kanavin: may as well finish that off, then get back to this. I have kind of worked out why the test fails but I still don't understand exactly why
<rburton>
LetoThe2nd: jonmason will be
danlor has quit [Ping timeout: 248 seconds]
flom84 has joined #yocto
otavio has joined #yocto
xmn has joined #yocto
rob_w has quit [Remote host closed the connection]
ptsneves1 has joined #yocto
<wmills>
LetoThe2nd: I will be there. nick is pretty transparent but this is Bill Mills
<LetoThe2nd>
wmills: ya i know the nick. cool!
<wmills>
rburton: over the weekend I used poky nanbield HEAD and meta-arm master HEAD for qemuarm64 and it blew up tring to patch linux-yocto. Was that supposed to work ATM? I can isolate it if you want. I just remove meta-arm from that build as I only use it for generic-arm64 and that build was using a core MACHINE
<rburton>
wmills: it worked last night in our CI
<rburton>
be sure you pulled meta-arm as there were changes fairly recently to fix kernel argh
<rburton>
i need to rebase my generic-arm64-to-poky branch
<wmills>
rburton: that was another question I had does the autobuilder test those combos while we are in this in-between state. If it tested nanbield / master or master / master or both?
<rburton>
right now it tests master of meta-arm for everything as there is no nanbield branch
<mcfrisk>
meta-arm issues caused by the upstreamed kernel patch, I presume... :/ I wish upstreaming would be smoother
<rburton>
when we release i'll update the AB to track the branches as appropriate
<wmills>
Let me investigate further. Both were fetched and were using the latest on the tip as of Nov 4 (of the respective branches)
<rburton>
that worked in CI last night
flom84 has quit [Ping timeout: 264 seconds]
ykrons has joined #yocto
silbe has quit [Remote host closed the connection]
zpfvo has quit [Ping timeout: 240 seconds]
zpfvo has joined #yocto
yocton has quit [Server closed connection]
yocton has joined #yocto
amitk_ has joined #yocto
zpfvo has quit [Ping timeout: 264 seconds]
zpfvo has joined #yocto
amitk has quit [Ping timeout: 264 seconds]
Chaser has quit [Quit: Chaser]
Chaser has joined #yocto
<wmills>
rburton: I don't know why it worked in CI I will look at that next. I did reproduce the issue. It is this line: SRC_URI:append:aarch64 = " \
<wmills>
Merge branch 'v6.5/standard/base' into v6.5/standard/qemuarm64
<wmills>
Author: Bruce Ashfield <bruce.ashfield@gmail.com>
<wmills>
That seems pretty old.
<RP>
kanavin: I've tried to explain why I think the test is failing, let me know if it doesn't make sense. Not quite sure what the best fix is yet
<rburton>
wmills: the fun with generic-arm64 is going to be adding all the configs missing in in linux-yocto's fragments. i don't suppose there's a list of the options needed to bring up the hardware...
<rburton>
or do we get to play diff-the-configs
tgamblin has joined #yocto
<wmills>
That is the right commit for nanbield. 6.5.7
<mcfrisk>
wmills: poky and meta-arm need to be in sync for this, after poky update to 6.5.7 meta-arm needs to be new enough to drop that patch which is now fixed in upstream. same story with 6.1 point releases in stable branch as the patch was applied to stable
<wmills>
mcfrisk, good point let me double check meta-arm commit is correct
<wmills>
NOPE! thanks
<wmills>
rburton: WRT defconfig that file I pointed you to has a lot of =y. Sometime over the next week I will make a new one that removed things we don't need and changes as many =y to =m as possible
<wmills>
We have already talked to Arnd Bergman about getting upstream defconfig to just work for zynqmp and TI SOC on beagleplay. (AM62 I think but I don't live this everyday anymore)
<rburton>
wmills: to be in meta-yocto-bsp the machine needs to use the yocto-kernel-cache pieces, which means splitting it up into fragments
<mcfrisk>
wmills: FWIW, if kernel.org releases change a config from tristate to bool and a config snippet has =m, then the options will be silently =n. we just hit that with some drm stuff..
<wmills>
of course that will be 6.8 or beyond. We still need someting until then
<rburton>
(meta-arm is a big fan of defconfig because its easier and _should_ work, but i don't expect RP to be happy with that)
<wmills>
mcfrisk: I don't understand that. Are you saying for configs not in the kernel your using or all =m's For generic-arm64 we need a LOT of =m by default.
<wmills>
Regardless of what happens in linux-yocto we want the defconfig in upstream kernel to work for more embedded SOCs out of the box.
<mcfrisk>
wmills: for long term maintenance, it's important to notice if upstream kernel config options change, get renamed etc. change from tristate to bool isn't captured by yocto tooling currently, but a complete rename or removal is.
<wmills>
mcfrisk OK I understand. thanks
<rburton>
wmills: dare i ask how many kernel patches are needed for the play still, if using 6.5?
<wmills>
I run zynqmp with 6.5.8 with zero patches. (well 1 patch for OpenAMP stuff but you don't need that)
Daanct12 has quit [Quit: WeeChat 4.1.1]
<wmills>
I even noticed that this kernel is enumeration mali 400 with the lima driver. :) I don't know if it works but it is a start
<wmills>
OHHH you said "for the play". I read "in order to play". I don't know we can ask denix when he is online or I can ask TI
<wmills>
denix: do you know if beagleplay can boot to UART and ethernet from 6.5 with zero patches?
<wmills>
Sooo, OpenAMP is trying to make its October release this week. (Can you relate?) Should I wait for the official YP 4.3 tag or just go with the lock commit of nanbield from Sunday?
yannd has joined #yocto
<zeddii>
good old kitchen sink configs!
Vonter has quit [Ping timeout: 245 seconds]
Vonter has joined #yocto
<wmills>
rburton: (( my issue was in layer setup I changed the comment of meta-arm to say "use the latest as of Nov 4" but did not actually change the commit hash. sorry for the noise ))
danlor has joined #yocto
Minvera has joined #yocto
<denix>
wmills: not sure, but I can find out
<fray>
wmillsI think Nanbield is close enough to use it for the OA release
<wmills>
denix: thanks for the pointer. Where is that running? U-boot 2021.01 is pretty old but I guess for this purpose it is a good thing. Nothing in latest u-boot needed for 6.6 kernel.
<wmills>
I don't see anything checking the ethernet in the kernel. It looks like it get enum'ed and there are two interrupts from enet. It must not be doing dhco or there would be more interrupts
<wmills>
dhcp
<denix>
wmills: unfortunately, 2021 still uses custom k3-image-gen to manage SYSFW, DM-FM, etc. with 2023 that all moved to binman and I'm still seeing issues with Play...
<denix>
[ 6.327019] IP-Config: Got DHCP answer from 192.168.0.1, my address is 192.168.0.73
<wmills>
denix: I am not particular about how it does that part but we do want EFI boot for SystemReady IR conformance
<wmills>
denix: OK I missed it thanks. I am surprised it only took 2 interrupts. I guess 2 TX and 2 RX?
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
<RP>
wmills: basically 4.3 will ship as QA tested in the QA report but we have docs commits that will still come in
Xagen has joined #yocto
Vonter has quit [Ping timeout: 258 seconds]
Vonter has joined #yocto
<wmills>
denix: I am not familiar with kernelci practices. Is that using the defconfig or can you give a link to the config that it does use?
<wmills>
RP: thanks. It sounds like I can go with the commits from Sunday. Any outlook for the tag w/ docs and everything?? This week? Next?
<RP>
wmills: I've seen patches so hoping the next few days
varjag has quit [Quit: ERC (IRC client for Emacs 27.1)]
<dvergatal>
hi guys I've recently encountered a problem with curl-native and that by default it is configured with the "--with-ca-bundle=${sysconfdir}/ssl/certs/ca-certificates.crt" which is supposed to use the path from native system but somehow it uses the path in which it's installed and for that reson i'm getting this error curl: (77) error setting certificate file:
<halstead>
We are rushing. As soon as the migration is complete and the new website is live we will put the connectivity.html file back in place.
khem has joined #yocto
<jonmason>
halstead: don't stress. hopefully if anyone else hits the issue, they can see my workaround above
frieder has quit [Remote host closed the connection]
wmills_ has joined #yocto
wmills_ has quit [Client Quit]
flom84 has quit [Ping timeout: 255 seconds]
wmills has quit [Ping timeout: 255 seconds]
wmills has joined #yocto
<marex>
jonmason: that, or they just wait, the world wont fall apart because of it
<wmills>
zeddii: is there a way in yocto kernel tooling to say "build all dtbs that the ARCH knows". Basically the equivalent of make dtbs dtbs_install
Marmottus1 has joined #yocto
smurray_ has joined #yocto
raghavgururajan_ has joined #yocto
Net147_ has joined #yocto
Haxxa has quit [Quit: Haxxa flies away.]
Piraty_ has joined #yocto
Chaser has quit [Quit: Chaser]
Ch^W_ has joined #yocto
Haxxa has joined #yocto
Net147 has quit [Ping timeout: 246 seconds]
smurray has quit [Ping timeout: 246 seconds]
smurray_ is now known as smurray
Saur has quit [Ping timeout: 246 seconds]
raghavgururajan has quit [Ping timeout: 246 seconds]
Piraty has quit [Ping timeout: 246 seconds]
Ch^W has quit [Ping timeout: 246 seconds]
Marmottus has quit [Ping timeout: 246 seconds]
Marmottus1 is now known as Marmottus
raghavgururajan_ has quit [Read error: Connection reset by peer]
tleb has quit [Write error: Connection reset by peer]
jonesv has quit [Write error: Connection reset by peer]
raghavgururajan has joined #yocto
tleb has joined #yocto
ptsneves has quit [Read error: Connection reset by peer]
ptsneves has joined #yocto
jonesv has joined #yocto
florian_kc has joined #yocto
Saur has joined #yocto
<ablu>
wmills: You can call dtbs_install yourself from do_install:
<halstead>
dl9pf: I'm not sure what the cause was but if it's resolved I'll let it go.
<dl9pf>
halstead: yeah, it works now
prabhakarlad has quit [Ping timeout: 248 seconds]
mvlad has quit [Remote host closed the connection]
justache is now known as justThanks
alessioigor has quit [Quit: alessioigor]
kevinrowland has joined #yocto
Chocky has quit [Quit: Leaving.]
Chocky has joined #yocto
<yocton>
Is there a time limit for the logs in the autobuilder? Is it safe to keep a build URL and go back to it in a few month or should I save the logs I want to keep?
dl9pf has quit [Server closed connection]
dl9pf has joined #yocto
<RP>
yocton: we haven't cleaned logs in a while, we tend to do it only when we need to clear space or do database migrations from buildbot upgrades
BobPungartnik has quit [Quit: Leaving]
<yocton>
That should be safe enough to leave in a todolist :) Thanks!
praneeth_ has quit [Server closed connection]
praneeth_ has joined #yocto
ldts has quit [Server closed connection]
ldts has joined #yocto
rber|res has quit [Ping timeout: 240 seconds]
speeder has quit [Remote host closed the connection]
GParker_ has joined #yocto
speeder has joined #yocto
geoffhp has quit [Ping timeout: 240 seconds]
rber|res has joined #yocto
GParker_ has quit [Quit: Leaving]
GParker_ has joined #yocto
goliath has quit [Quit: SIGSEGV]
GParker_ has quit [Quit: Leaving]
<kevinrowland>
If I DEPEND on a `python3-*-native` package, I see that it lands in my recipe's sysroot. But what's the right way to pass that sysroot to a python script in `do_configure` so that it can find the package? Just append to PYTHONPATH?
sgw has joined #yocto
armpit has quit [Server closed connection]
<fray>
it's been a while, but usually you use the python helpers to give you the path to the regular bits and then add your module if it's in a custom path
armpit has joined #yocto
<fray>
(otherwise try to keep it in a standard path and that won't be necessary0.. if a configure script actually goes in an interogates it, then you should be able to use the same process to get the path to pass to configure
<RP>
kevinrowland: the python3native and python3targetconfig classes can sometimes help