<kanavin>
viric, this happens when you change your configuration so that there are no existing sstate objects that can be reused. In this case, bitbake has to clean up ${WORKDIR} so that there are no copies of sstate objects left over from previous builds.
florian_kc has joined #yocto
<kanavin>
bitbake never removes anything from the actual sstate cache, unless you ask with -c cleansstate
<kanavin>
I agree that the message is somewhat ambiguous
aduskett has quit [Ping timeout: 252 seconds]
DixitP has joined #yocto
<RP>
viric: It removes from the sysroots, not the sstate cache. I wrote the code ;-)
ahussain has joined #yocto
ehussain has quit [Ping timeout: 252 seconds]
ahussain is now known as ehussain
sakoman has quit [Ping timeout: 252 seconds]
gspbirel56 has quit [Ping timeout: 255 seconds]
sakoman has joined #yocto
<viric>
RP ah!
<viric>
The problem I have is with 'systemd'. If I modify systemd, it always rebuilds it and its dependencies. It doesn't check sstate for systemd.
<viric>
I have tried 'bitbake -DD systemd', and there should be a line with 'SState' (note two capital S) that should say if systemd sstate files are found, and the lines don't exist
gspbirel56 has joined #yocto
<viric>
So I have systemd with a bbappend and with out. Let's call it systemd A and systemd B. If I build A, then I change code to B, and I build B, then I change code to A, then I build A again.... I'd expect systemd A is in the sstate already from the first build. But it is not checked.
florian has quit [Ping timeout: 252 seconds]
florian_kc has quit [Ping timeout: 248 seconds]
sakoman has quit [Ping timeout: 252 seconds]
<viric>
I can see SState checked for many other packages but never for systemd
<viric>
kanavin: I definitely have sstate pieces that can be reused. I'm building conf A, then conf B, then conf A, then conf B ... and it build systemd & mozjs & etc. every time
<kanavin>
viric, this is not supposed to happen. but it's hard to say what is going on without seeing specifics.
florian has joined #yocto
<viric>
I imagine it can be easily reproducible by adding some line to a systemd bb and removing it mh
sakoman has joined #yocto
florian_kc has joined #yocto
<kanavin>
viric, not really. I'm certain in my setup it will reuse sstate.
<kanavin>
it's a standard scenario and it works for everyone who's not doing strange things.
<viric>
hm I wonder what I am overlooking then
<kanavin>
viric, you might want to start with basic poky, and build systemd only
<viric>
Yes, I understand. Now suddenly I stopped being able to reproduce what I thought was evident. I'm clearly missing some detail
<viric>
thank you
florian_kc has quit [Ping timeout: 245 seconds]
florian_kc has joined #yocto
thomas_34 has quit [Quit: Client closed]
<viric>
how can I tell bitbake to build INITRAMFS_IMAGE ?
<viric>
so, the package the variable points to
angeloc__ has quit [Ping timeout: 252 seconds]
aduskett has joined #yocto
aduskett has quit [Read error: Connection reset by peer]
aduskett has joined #yocto
aduskett has quit [Client Quit]
DixitP has quit [Quit: Client closed]
sgw has quit [Ping timeout: 245 seconds]
sgw has joined #yocto
thomas_34 has joined #yocto
sakoman has quit [Ping timeout: 265 seconds]
sakoman has joined #yocto
angeloc__ has joined #yocto
thomas_34 has quit [Quit: Client closed]
debbiemartin has joined #yocto
ehussain has quit [Remote host closed the connection]
rob_w has quit [Remote host closed the connection]
Xagen has quit [Ping timeout: 260 seconds]
<jonmason>
Is there a plan to remove linux-yocto-rt now that the real time patches are fully upstreamed in the latest kernel?
<RP>
jonmason: that kernel recipe enables the rt features so probably not. zeddii would know more
cyxae has joined #yocto
<jonmason>
Seems like that should be a yocto-kernel-cache feature, but probably something that should be discussed publicly
<RP>
jonmason: I think it is a KTYPE iirc
<RP>
so slightly higher level
<jonmason>
gotcha. I haven't looked into it, just thinking about simplifying our CI and the recent update on that came to mind
<RP>
jonmason: ironically we'll probably test more on the YP side now it is more standardised
<jonmason>
the combinatorial explosion that comes from 3 kernels, 3 compilers, uboot/edk2. glibc/musl...for all of our platforms. You know better than I with the autobuilder, but it's a lot
<RP>
jonmason: we can't make -rt an always on or always off option, it will always need a separate test
<RP>
whether it is a separate recipe of not doesn't really matter
<jonmason>
of course, but not sure it's something that needs to be tested on an architecture level (e.g., arm/arm64) for every possible board
<jonmason>
it should "just work"
<RP>
jonmason: er, I'm not convinced on that
<jonmason>
lol, fair
<RP>
you know how invasive -rt is right?
<RP>
its like me saying armv5 doesn't need testing as it is mostly just arm :)
<jonmason>
I mean, it shouldn't.... :D
<RP>
totally :)
<jonmason>
well see if rburton hard nack's my patch to remove it from CI
<RP>
jonmason: If rburton naks a removal patch I'd suspect alien abduction
<jonmason>
We need to get shirts made. "Burn it all down" - rburton
<jonmason>
I suspect internal people will still want coverage for it, something to discuss tomorrow
<avocadosand>
I’ll take this into consideration for next years swag ;)
<RP>
jonmason: it does have a lot of users for better or worse
<jonmason>
avocadosand: extra credit if you can AI his pic into the little girl buring down the house meme
<jonmason>
or the "this is fine" one
<RP>
avocadosand: I wonder if the different project members all have a particular catchphrase...
<avocadosand>
I gave out the This is Fine pins at OSS in Vienna. I have inspirational knitted creatures for ATO
<jonmason>
RP: the ones from zeddii cannot be put on a t-shirt ;-)
<RP>
avocadosand: I'm sad to be missing out on these! I'm quite proud to own a "Mentally Contaminated" pin badge...
<RP>
jonmason: there are ways and means...
<avocadosand>
You have one waiting for you when I see you next!
<RP>
avocadosand: awesome, thanks :)
<jonmason>
avocadosand: bring them to Cambridge in January
<rburton>
I pop to the tip to get rid of some crap and look what I come back to
<zeddii>
jonmason, no, it definitely stays.
<zeddii>
the biggest misconception is going to be "rt just works"
<zeddii>
there will a) still be patches b) be a dedicated configuration.
<zeddii>
I'm already testing those changes on linux-yoccto-dev
<rburton>
somewhere on my laptop is design for an OpenEmbedded Certified Tool tshirt
<jonmason>
zeddii: the kernel guys say it just works, and they'd never lie or exagerate
<rburton>
zeddii: did you see debbiemartin's question yesterday about the kernel and clang? is that a combo you've tried, or does the kernel explicitly force gcc right now?
<zeddii>
it doesn't force it per-se, but I've never tried it.
<jonmason>
I thought they got clang working with the kernel ages ago
<rburton>
yeah upstream, but not our recipes
<jonmason>
gotcha
<zeddii>
it might just be our variable definitions that have gcc assumptions for kernel_cc, etc.
wojci has quit [Ping timeout: 245 seconds]
Daanct12 has quit [Quit: WeeChat 4.4.2]
olani- has joined #yocto
olani_ has joined #yocto
RP has quit [Ping timeout: 248 seconds]
Xagen has joined #yocto
goliath has quit [Quit: SIGSEGV]
RP has joined #yocto
xmn has joined #yocto
Xagen has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Xagen has joined #yocto
aduskett has joined #yocto
aduskett has quit [Client Quit]
rfuentess has quit [Remote host closed the connection]
jmiehe has joined #yocto
jmiehe has quit [Client Quit]
Guest81 has joined #yocto
mbulut has quit [Ping timeout: 244 seconds]
aduskett has joined #yocto
mbulut has joined #yocto
khem has quit [Quit: WeeChat 4.4.2]
khem has joined #yocto
<JPEW>
RP: The code has changed forms slightly over time, but the idea of setting a variable to indicate that a package has an incompatible license has been the mechanism since the feature to check package licenses independently was added in 2012
<RP>
JPEW: I'm happy to change this. That commit explains a few things!
<JPEW>
Cool, I'll do the tests about var deps you suggested and send the patches
Vonter_ has joined #yocto
Vonter has quit [Ping timeout: 265 seconds]
<RP>
thanks
<RP>
JPEW: I'm much happier when I know we've done the due diligence like this!
gspbirel56 has quit [Ping timeout: 252 seconds]
Vonter_ has quit [Read error: Connection reset by peer]
Vonter has joined #yocto
Vonter has quit [Client Quit]
Vonter has joined #yocto
jmd has joined #yocto
florian_kc has quit [Ping timeout: 248 seconds]
florian has quit [Quit: Ex-Chat]
zpfvo has quit [Remote host closed the connection]
PhoenixMage has quit [Ping timeout: 252 seconds]
PhoenixMage has joined #yocto
PhoenixMage has quit [Ping timeout: 255 seconds]
PhoenixMage has joined #yocto
mjm has joined #yocto
<vvn>
do you guys run downloads and sstate mirrors locally, if so, do you serve them from a single server instance, or two servers?
frieder has quit [Remote host closed the connection]
jmd has left #yocto [ERC 5.4 (IRC client for GNU Emacs 28.2)]
PhoenixMage has quit [Ping timeout: 252 seconds]
<rburton>
personally, single nfs server
PhoenixMage has joined #yocto
<rburton>
the AB is a single nfs mount, or http for public
leon-anavi has quit [Quit: Leaving]
florian_kc has joined #yocto
<Guest81>
Hi, in branch scarthgap, commit 711c93422918 ("linux-firmware: upgrade 20240312 -> 20240909") introduced a regression were all firmware files are now installed (1.2G). Is this something that has been fixed, maybe in another branch?
qschulz has quit [Remote host closed the connection]