<Guest98>
i want to use the "systemd" recipes in poky(dunfell) but in poky(zeus). because, i need the version in dunfell. if i take the dunfell systemd folder and put it in mymeta-layer. then how do i append or override the poky(zeus) version? i know it's not a logical move, but i need it.
wooosaiiii has quit [Quit: wooosaiiii]
wooosaiiii has joined #yocto
<marex>
Guest98: xilinx ?
<Guest98>
im not using meta-xilinx
<marex>
they were the only ones using such devastatingly obsolete software (zeus) iirc
<marex>
Guest98: probably create systemd-dunfell_<version>.bb , and then set PREFERRED_PROVIDER/PREFERRED_VERSION to point to it
<marex>
then you avoid overwriting recipes
<marex>
you might need to add PROVIDES/RPROVIDES into systemd-dunfell_%.bbappend , keep custom changes to that recipe in bbappend so you can sync the recipe with upstream easily
davidinux2 has joined #yocto
<Guest98>
there are multiple .bb recipes in the systemd folder. will systemd-dunfell_<version>.bb be enough?
<marex>
you might need to pick whatever other dependencies you need too, indeed
<marex>
but seriously, you might want to consider moving on from zeus instead of building this franken-oe
<Guest98>
if i increase the priority of my layer? i mean put it above poky. will that be enough?
<marex>
you dont have to, if you have separate recipes and you set PREFERRED_PROVIDER/PREFERRED_VERSION , then the separate recipes would be used
<Guest98>
i think, i understand. it will be enough to set PREFERRED_PROVIDER/PREFERRED_VERSION for each recipe i will use in the systemd folder.
<Guest98>
marex thank you for help btw.
<marex>
yeah, that should work
<marex>
btw. its called directory
<Guest98>
yes, my bad :D
<marex>
Guest98: btw when in doubt, use 'bitbake -e'
<marex>
that will dump the entire order of variable expansion in bitbake and how it parsed the recipes when it traversed them
<marex>
that should give you an idea why some recipe was not picked, or was picked in the wrong order
nemik has quit [Ping timeout: 268 seconds]
nemik has joined #yocto
<Guest98>
marex thanks! from rookie with love.
nemik has quit [Ping timeout: 240 seconds]
nemik has joined #yocto
astlep550474 has quit [Ping timeout: 240 seconds]
astlep5504744 has joined #yocto
astlep5504886 has joined #yocto
astlep550488 has quit [Ping timeout: 268 seconds]
xmn has quit [Ping timeout: 240 seconds]
amitk has quit [Ping timeout: 268 seconds]
rfuentess has joined #yocto
mckoan|away is now known as mckoan
<mckoan>
good morning
<Guest98>
morning
amitk has joined #yocto
Guest98 has quit [Quit: Client closed]
Guest98 has joined #yocto
zpfvo has joined #yocto
<kanavin>
rburton, the actual problem is how to submit this upstream: we need to demonstrate breakage. if we ourselves don't have a scenario where it reliably happens, I'd rather drop the patch.
<RP>
kanavin: I think rburton's point was the issue wasn't happening any more so we are good to drop the patch?
<kanavin>
RP: either way, I'd like to drop it, yes :) if the issue reappears, we can quickly revert that, and then we'll have some kind of lead into where the problem is, but with the patch in place that can't happen
<RP>
kanavin: I'm not in a good frame of mind for more issues, but yes :/
* RP
wonders why we're seeing a lot of mips giving "AssertionError: 1 != 0 : Test log string not found in /var/log/messages or logread. Output: logread: can't find syslogd buffer: No such file or directory"
<RP>
kanavin: I'm going to be very happy to have that lockfile fix in as it hit the master builds that didn't have it again last night!
<RP>
I've merged it now
<kanavin>
RP: I only hope it does fix the issue for good
<kanavin>
I don't want to start more a-fulls, as it's rather crowded on AB lately, and you're still sorting out that kernel fail I think
gsalazar has joined #yocto
perdmann has quit [Server closed connection]
perdmann has joined #yocto
florian_kc has joined #yocto
<RP>
kanavin: I've been running a lot of builds as it is the only way to reproduce the issue. I've been trying to drip feed other patches through at the same time but it is tricky
<RP>
kanavin: the good news is I think we identified the kernel issue
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
rsalveti has quit [Quit: Connection closed for inactivity]
* sudip
looks at abelloni's mail and realises ptests for libssh2 patch were only tested on oe-core and not on poky :(
bps has joined #yocto
bps has quit [Changing host]
bps has joined #yocto
bps2 has joined #yocto
<kayterina[m]>
Hello, the repo I am building has a "configure" script inside $S/deep/deeper/configure (S = "${WORKDIR}/git"), so I have to 'cd' there in do_compile/install. Is there a way to move to $B or is it ok to just leave it as is?
bps has quit [Ping timeout: 240 seconds]
<mcfrisk>
kayterina[m]: You can change S or even B, but note that applying patches to S via SRC_URI may be more difficult in a sub directory as paths differ
<mcfrisk>
kayterina[m]: also note that depending on build tooling, you may be able to adjust the default build script to a correct path. for autotools.bbclass this is done via AUTOTOOLS_SCRIPT_PATH and/or CONFIGURE_SCRIPT variables
nemik has quit [Ping timeout: 240 seconds]
nemik has joined #yocto
ptsneves has joined #yocto
nemik has quit [Ping timeout: 265 seconds]
gsalazar has quit [Ping timeout: 260 seconds]
nemik has joined #yocto
vladest has quit [Ping timeout: 240 seconds]
prabhakarlad has joined #yocto
luneff has quit [Read error: Connection reset by peer]
luneff has joined #yocto
<rburton>
kanavin: i have a rewrite of the ghostscript recipe now :)
<kanavin>
one of the oldest, ugliest hacks in oe-core
<rburton>
yay
<rburton>
does 3.12 actually improve things there?
<kanavin>
yes, by removing distutils altogether
<kanavin>
as expected over a two year deprecation period
<kanavin>
of course there will be components that don't pay attention to deprecation warnings, or any warnings at all
vladest has joined #yocto
pidge has joined #yocto
<rburton>
oh ghostscript really is terrible
<rburton>
it has copies of so much code in its source tree and uses those over system libraries
<rburton>
to the point where you need to _delete them_ to get it to use the system
<landgraf1>
it's even worse...
<kanavin>
rburton, I think that the legacy of being an apple-driven project?
<rburton>
possibly. i think its more 'building stuff is hard'.
<rburton>
like it doesn't use pkgconfig macros but calls pkgconfig directly
<kanavin>
I still don't understand why smartphone apps like, oh, linkedin, take up 300 Mb of space, and do it casually
<kanavin>
and you still can't sort by date there
d-s-e has joined #yocto
kpo_ has quit [Ping timeout: 246 seconds]
kpo has joined #yocto
Guest98 has quit [Quit: Client closed]
gsalazar has joined #yocto
goliath has joined #yocto
<rburton>
kanavin: can you mail me a log.do_configure from coreutils 9.3 that works for you?
<kanavin>
rburton, building that for arm64 now, hold
<mcfrisk>
kanavin: apps, because of a custom embedded chromium to bypass any tracking, privacy etc settings users may have otherwise set... you are the product :/
<kanavin>
mcfrisk, I still don't understand why closed source software is not illegal
<kanavin>
it bloody should be
<rburton>
kanavin: ha i made it work by changing what locales i have available
<kanavin>
rburton, so log.do_configure not needed?
<rburton>
would be interesting to check my working case is doing the same as yours, i don't care what the target is
<kanavin>
ok I'll let the build finish
<rburton>
my theory is if you remove your fr_FR.utf8 locale then the build breaks :)
<mcfrisk>
kanavin: well, Android is open source and Google still gets all the data... /rants
<kanavin>
mcfrisk, open source per gpl3 definition
<kanavin>
can rebuild and reinstall
<rburton>
no, _add_ the locale
<kanavin>
rburton, locale -a on the host shows it is installed
<rburton>
hm
beroset has quit [Quit: Client closed]
prabhakarlad has quit [Quit: Client closed]
Guest98 has joined #yocto
<kanavin>
rburton, sent
<rburton>
cheers
<kanavin>
rburton, note that none of the a-full distros had issues with it either :)
<rburton>
there's an AC_TRY_RUN with a missing ,
<rburton>
note that the top says there's no fallback
<rburton>
it's guarded with a check on the fr.utf8 locale being available
<rburton>
kanavin: eg fedora38-ty-1 only has en_* locales present. my theory is adding fr_FR.UTF-8 would break it.
Guest98 has quit [Quit: Client closed]
nemik has quit [Ping timeout: 246 seconds]
nemik has joined #yocto
beroset has joined #yocto
nemik has quit [Ping timeout: 268 seconds]
nemik has joined #yocto
d-s-e has quit [Ping timeout: 260 seconds]
<kanavin>
rburton, but log.do_configure I sent is from the host that has
<kanavin>
has it
<rburton>
checking for a french Unicode locale... none
<rburton>
you have fr_fr but not the utf8 version
<mcfrisk>
Q: should USERADD_GID_TABLES build reprodocubility be set in layer or distro config, or image recipe, or special local.conf for reproduction tests only?
<RP>
kanavin: dependencies are hard ;-)
<RP>
(app size on phones)
d-s-e has joined #yocto
<RP>
mcfrisk: yes ;-)
<RP>
mcfrisk: it depends what you're doing, there is no right answer to that
OnkelUlla has quit [Ping timeout: 246 seconds]
<qschulz_>
it's not because I don't participate much anymore in the chat that I don't see you blaming it on the French rburton
<rburton>
to be fair the bug only happened if you didn't have french, and unsurprisingly it was fixed by a french person :)
<mcfrisk>
RP: gah, USERADD/GROUPADD fun it is then, trying figure out what would be best.. I think static config file for all builds
Minvera has joined #yocto
Guest98 has joined #yocto
old_boy has quit [Quit: Client closed]
<Guest98>
hi everyone.
<Guest98>
let's say there is a "linux_4.14.98.bb" recipe' in "main-layer". this recipe got append from "other-layer"'s "linux_4.14.%.bbappend" . i also created "linux_4.14.%.bbappend" from "my-layer" and i want to append it. my purpose, i want to change the src_uri in "linux_4.14.%.bbappend" to be machine specific src_uri from my-layer.
<LetoThe2nd>
Guest98: sounds legit, what is the question?
<Guest98>
The value "src_uri" machine spesific is not returning right value.
kscherer has joined #yocto
<Guest98>
LetoThe2nd oh, my bad. rookie doing rookie things :D
<LetoThe2nd>
Guest98: np, have fun.
<Guest98>
i forgot to put the recipe in the directory.
shoragan has quit [Server closed connection]
shoragan has joined #yocto
<RP>
mcfrisk: we created a file which works for OE-Core but obviously we can't cover everything in every other layer
xmn has joined #yocto
<Guest98>
if i have "other-machine.conf" and it contains "MACHINEOVERRIDES =. "mx8:mx8m:mx8mq:". Also, i have "my-machine.conf" in my own layer and it contains: https://pastebin.com/MCY0txgY
<Guest98>
a function that works as "do_install_other-machine" also affects "my-machine", right? if i use "do_install_my-machine" it only affects my-machine.
<mcfrisk>
RP: yes, I'm copying that and adding custom files to my layer (it seems file contents are not merged so can't amend files in poky) and now need to decide if the files are to be used in all builds or just the reproducibility test builds, and where the config should enable those. image recipe wasn't the correct place even if it felt right, layer config file feels very much wrong. distro config doesn't work
<mcfrisk>
for all uses. machine config is too low level. so to local.conf then maybe
sakoman has joined #yocto
<RP>
mcfrisk: I'm open to ways to improve it but at the time I wanted to at least get the tests working for core
Colin_Finck has quit [Server closed connection]
Colin_Finck has joined #yocto
prabhakarlad has joined #yocto
<kayterina[m]>
In my autotools recipe, by setting AUTOTOOLS_SCRIPT_PATH and creating the Makefiles and .o in $B, it fails in do_compile with "No rule to make target". When simply 'cd' in $S and build there, it compiles ok. Is it logical? My attempt is here: https://pastebin.com/SpqKppMH
<kayterina[m]>
What else changes when I use the autotools variables to split the dirs?
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
<rburton>
kayterina[m]: autoconf will write the Makefiles to ${B}
<rburton>
you can delete your do_compile entirely
alessioigor has quit [Quit: alessioigor]
<rburton>
oh sorry the snippet was too small
<rburton>
i'm guessing the autotools files are broken and don't like out-of-tree builds
alessioigor has joined #yocto
<rburton>
inherit autotools-brokensep and file a bug with them because they're not doing automake right
ptsneves has quit [Ping timeout: 265 seconds]
kscherer has quit [Ping timeout: 258 seconds]
cambrian_invader has joined #yocto
ptsneves has joined #yocto
AKN has joined #yocto
tnovotny has quit [Quit: Leaving]
d-s-e has quit [Quit: Konversation terminated!]
Guest98 has quit [Quit: Client closed]
prabhakarlad has quit [Quit: Client closed]
AKN has quit [Read error: Connection reset by peer]
zpfvo has quit [Ping timeout: 265 seconds]
zpfvo has joined #yocto
amitk has quit [Ping timeout: 240 seconds]
BobPungartnik has joined #yocto
Guest98 has joined #yocto
BobPungartnik has quit [Client Quit]
kpo has quit [Ping timeout: 258 seconds]
florian_kc has quit [Ping timeout: 268 seconds]
mckoan is now known as mckoan|away
vladest has quit [Ping timeout: 265 seconds]
alessioigor has quit [Quit: alessioigor]
florian has quit [Quit: Ex-Chat]
nemik has quit [Ping timeout: 240 seconds]
nemik has joined #yocto
rfuentess has quit [Remote host closed the connection]
Guest98 has quit [Quit: Client closed]
nemik has quit [Ping timeout: 240 seconds]
nemik has joined #yocto
NishanthMenon[m] has quit [Remote host closed the connection]
leon-anavi has quit [Quit: Leaving]
zpfvo has quit [Remote host closed the connection]
bps2 has quit [Ping timeout: 240 seconds]
<khem>
yeah sometimes automake files assume relative paths for build dir
<khem>
Use ${abs_top_builddir} and ${abs_top_srcddir} in Makefile.am
<khem>
instead of ${builddir} and ${srcdir}
ptsneves has quit [Ping timeout: 258 seconds]
OnkelUlla has joined #yocto
Guest98 has joined #yocto
<Guest98>
hi everyone,
<Guest98>
if there are 2 bsp directories and 1 meta-common. does it make sense to put images in meta-common or should they be in bsp?
Guest98 has quit [Quit: Client closed]
<rburton>
are they specific to each bsp or common?
tgamblin has quit [Ping timeout: 246 seconds]
tgamblin has joined #yocto
nemik has quit [Ping timeout: 258 seconds]
nemik has joined #yocto
florian_kc has joined #yocto
nemik has quit [Ping timeout: 268 seconds]
nemik has joined #yocto
amitk has joined #yocto
rsalveti has joined #yocto
PhoenixM1ge has joined #yocto
PhoenixMage has quit [Ping timeout: 260 seconds]
gsalazar has quit [Ping timeout: 240 seconds]
seninha has quit [Remote host closed the connection]
schtobia has quit [Ping timeout: 265 seconds]
schtobia has joined #yocto
kscherer has joined #yocto
kscherer has quit [Ping timeout: 240 seconds]
kscherer has joined #yocto
testGuest93 has joined #yocto
testGuest93 has left #yocto [#yocto]
<rburton>
kanavin: ghostscript branch ready for squash and nice comment: 8 files changed, 20 insertions(+), 314 deletions(-)
ptsneves has joined #yocto
polprog has quit [Server closed connection]
polprog has joined #yocto
florian_kc has quit [Ping timeout: 258 seconds]
tgamblin has quit [Remote host closed the connection]
tgamblin has joined #yocto
ptsneves has quit [Ping timeout: 240 seconds]
amitk has quit [Ping timeout: 258 seconds]
Xagen has joined #yocto
amitk has joined #yocto
florian_kc has joined #yocto
amitk has quit [Remote host closed the connection]
<kanavin>
rburton, \0/ removing code is the best kind of commit
amitk has joined #yocto
<JaMa>
until finally no code, no problem
<kanavin>
the world needs less code for sure
<kanavin>
sometimes I think that exponential growth in ram and non-volatile storage capacity for code did everyone a disservice
Saur[m] has joined #yocto
<JaMa>
agreed, world needs less code and I need more RAM and non-volatile storage and instead of building the code I'll play all 125GB of Starfield :)
<Saur[m]>
Not sure I want to go back to the 64 kB I started with though (other than for nostalgia)...
<nerdboy>
but then 128k would be Woo hoo! bonus!
<nerdboy>
like the commode, er, commodore
<tgamblin>
kanavin: why learn to write efficient software when you can build out machine learning infrastructure and have it do that for you? :)
<Saur[m]>
Never had a C128, went from C64 to A500.
<nerdboy>
spell check proposed that ^^ so i went with it...
<nerdboy>
baseball on c128, now that's nostalgia
vladest has joined #yocto
amitk has quit [Ping timeout: 240 seconds]
<JaMa>
tgamblin: maybe because useful machine learning infrastructure would need exponentially more ram and non-volatile storage than what we have nowadays? :)
ecdhe has quit [Read error: Connection reset by peer]
ecdhe has joined #yocto
seninha has joined #yocto
old_boy has quit [Quit: Client closed]
<khem>
rburton: I had a patch for gn build with gcc13 for meta-arm its still pending
<khem>
IIRC jonmason mentioned that it was failing in your CI, do you have handy links to failure logs
old_boy has joined #yocto
<old_boy>
EXTRA_IMAGEDEPENDS has my test.bb file which installs a tool in my output folder whenever I am doing explicit 'bitbake test' but not when I am doing a complete-image-minimal. Is there any way to debug this?
florian_kc has quit [Ping timeout: 240 seconds]
nemik has quit [Ping timeout: 258 seconds]
nemik has joined #yocto
nemik has quit [Ping timeout: 240 seconds]
nemik has joined #yocto
<khem>
old_boy: are you adding EXTRA_IMAGEDEPENDS in your machine conf ?
<old_boy>
khem: yes
<khem>
try with core-image-base
mrpelotazo has quit [Ping timeout: 268 seconds]
mrpelotazo has joined #yocto
<old_boy>
let me try
<old_boy>
khem: weird issue, in my do_deploy which exists in test.bb file that works fine even with full build now whenever I am adding a log in that function like bbplain.
<old_boy>
with log it doesn't copy the artifacts to output folder..
<old_boy>
so can't even debug it if that function is called or not as adding logs is fixing the problem...