<paulg>
what is the easiest way to get one of those "dateless" symlinks in an image dir for an added file? As a made up example, say I wanted core-image-minimal.ext4.md5sum linked to core-image-minimal-20230504201932.ext4.md5sum
<paulg>
my reading so far tends to make me think it is triggered off IMAGE_TYPES and friends but I haven't quite figured it all out yet.
<wooosaiiii>
here it is: "When working with varflags, the same syntax, with the exception of overrides, applies. In other words, you can set, append, and prepend varflags just like variables. See the “Variable Flag Syntax” section for details."
rfuentess has quit [Remote host closed the connection]
thomasd13 has quit [Remote host closed the connection]
thomasd13 has joined #yocto
Blok has joined #yocto
<Blok>
I am trying to add a build instance id (job id). This would typically be a nono as it breaks determinism. What would be the recommended way to deal with such meta-data?
<rburton>
you might want to expand on what you want to do
<Blok>
The job id would match a build and be a compact way of describing the version of each included layer. I have a long manifest with all the SHAs from all commits, but that is less human friendly. I am looking for a human friendly way to describe the version.
alessioigor has joined #yocto
seninha has joined #yocto
seninha has quit [Remote host closed the connection]
seninha has joined #yocto
kscherer has joined #yocto
rob_w has quit [Quit: Leaving]
d-s-e has quit [Quit: Konversation terminated!]
Blok has quit [Ping timeout: 245 seconds]
seninha has quit [Ping timeout: 256 seconds]
alessioigor has quit [Quit: alessioigor]
zpfvo has quit [Ping timeout: 260 seconds]
rich1234 has quit [Quit: Client closed]
Minvera has joined #yocto
seninha has joined #yocto
zpfvo has joined #yocto
nemik has quit [Ping timeout: 240 seconds]
nemik has joined #yocto
amitk__ has quit [Ping timeout: 264 seconds]
nemik has quit [Ping timeout: 268 seconds]
nemik has joined #yocto
<olani>
Blok: You could look at image-buildinfo.bbclass
<crazy_imp>
inside the devshell - is there a way to call the do_compile function?
<crazy_imp>
is calling ../../temp/run.do_compile the best way to do this?
florian has quit [Quit: Ex-Chat]
<crazy_imp>
(amount of ../ may vary depending on ${S})
<rburton>
crazy_imp: its the way to do it
<rburton>
(running that script is what 'task execution' is)
seninha_ has joined #yocto
seninha has quit [Remote host closed the connection]
<fullstop>
Hi all, I have a platform in which I want to generate both an initramfs as well as a root file system. I want the initramfs to use busybox-mdev but the other image to use systemd. I added DISTRO_FEATURES_remove = " systemd" to the image recipe, but systemd components are still present in the generated image.
<fullstop>
Is it best to have two different trees for this?
<rburton>
you can't change distro features in an image, but you can pick a different init manager in an image recipe with the INIT_MANAGER variable
<fullstop>
rburton: thanks! So the systemd components will be present no matter what I do?
<fullstop>
The image is small enough with those pieces in, but I don't know how it affects things.
<rburton>
if you set INIT_MANAGER and VIRTUAL-RUNTIME_dev_manager pull in busybox parts, you won't get systemd
<fullstop>
I'm missing INIT_MANAGER
<fullstop>
let me try that
<fullstop>
this is back on the latest dunfell, for what it's worth.
<rburton>
actually if you set INIT_MANAGER then it sets the dev-manager for you
<rburton>
your distro can set INIT_MANAGER=systemd in the default config and then the initramfs set INIT_MANAGER=mdev-busybox to switch for that image
<rburton>
that should work, anyway
<fullstop>
Perhaps I should take the time to look more into distro things. I have most of these settings in local.conf
<fullstop>
setting INIT_MANAGER to busybox (in the image recipe) still retained systemd components.
<fullstop>
Also, as an aside, thanks for all of the work that you (and everyone else) does with Yocto. It's hugely beneficial in my life and I greatly appreciate it.
<RP>
rburton: should it? :/
<RP>
rburton: I'm not convinced it can be done per image
<rburton>
hm
<rburton>
maybe you need to write a custom initramfs recipe that pulls in the bits you want explicitly
vvmeson has joined #yocto
vmeson has quit [Ping timeout: 276 seconds]
<fullstop>
I could also delete the systemd pieces after it's built, but that seems wrong.
<rburton>
yes, that's right, you just need to write an initramfs recipe that doesn't use packagegroups and installs just the bits it wants
<rburton>
like core-image-minimal-initramfs does with PACKAGE_INSTALL
<fullstop>
all things to look at. Thanks everyone!
seninha_ has quit [Read error: Connection reset by peer]
seninha_ has joined #yocto
<fullstop>
Ideally I'd have the virtual/kernel recipe in one image reference the output of the initramfs image, but that sounds like a task for another day.
seninha_ has quit [Client Quit]
prabhakarlad has quit [Quit: Client closed]
prabhakar has quit [Quit: Connection closed]
prabhakar has joined #yocto
prabhakarlad has joined #yocto
thomasd13 has quit [Ping timeout: 268 seconds]
zpfvo has quit [Ping timeout: 256 seconds]
ndec_ is now known as ndec
alessioigor has joined #yocto
florian has joined #yocto
zpfvo has joined #yocto
zpfvo has quit [Ping timeout: 268 seconds]
zpfvo has joined #yocto
tomzy_0[m] has quit [Quit: You have been kicked for being idle]
alessioigor has quit [Ping timeout: 240 seconds]
zpfvo has quit [Quit: Leaving.]
Thorn_ has joined #yocto
Thorn has quit [Ping timeout: 264 seconds]
leonanavi has quit [Quit: Leaving]
leon-anavi has quit [Remote host closed the connection]
bps has joined #yocto
bps has joined #yocto
bps has quit [Changing host]
florian has quit [Ping timeout: 256 seconds]
DarkestDot has joined #yocto
Thorn_ has quit [Ping timeout: 256 seconds]
goliath has quit [Quit: SIGSEGV]
Thorn has joined #yocto
seninha has joined #yocto
<khem>
abelloni: You might want to drop * 7f003547f57 gcc: Upstream-Status patch updates for GCC. from your contrib master-next branch
rfs613 has quit [Ping timeout: 248 seconds]
<khem>
RP wants it to be reworked and its conflicting with my gcc-13 patches
<khem>
in order to rebase it will be helpful
amitk_ has joined #yocto
<kanavin>
I think INIT_MANAGER isn't tested or meant as a per image item, only as a distro level item. It may kinda work, but that's not guaranteed and people may run into weird fails.
goliath has joined #yocto
Guest7484 has joined #yocto
Guest7484 has quit [Client Quit]
rfs613 has joined #yocto
bps has quit [Remote host closed the connection]
bps has joined #yocto
bps has quit [Changing host]
bps has joined #yocto
<rburton>
yeag i was wrong, it won't
vvn has quit [Quit: WeeChat 3.8]
vvn has joined #yocto
prabhakarlad has quit [Quit: Client closed]
nerdboy has quit [Remote host closed the connection]
nerdboy has joined #yocto
nerdboy has quit [Changing host]
nerdboy has joined #yocto
pabigot has joined #yocto
nerdboy has quit [Ping timeout: 256 seconds]
seninha has quit [Ping timeout: 240 seconds]
goliath has quit [Quit: SIGSEGV]
nerdboy has joined #yocto
nerdboy has quit [Changing host]
nerdboy has joined #yocto
<fullstop>
I have the INIT_MANAGER working by creating two different distros. It works well, except the initramfs distro really wants to build the kernel and I don't need one.
<fullstop>
In fact, it's a problem because it's generating the cpio which is used in the other distro's kernel.
<smurray>
fullstop: you could try setting PREFERRED_PROVIDER_virtual/kernel = "linux-dummy" for that distro
<fullstop>
smurray: that's better than where I was going, which was going to be a zero-byte cpio file for that distro
mvlad has quit [Remote host closed the connection]
<fullstop>
smurray: maybe it does need to know about the kernel itself, because I'm using cryptsetup?
<smurray>
fullstop: it's possible. I've never tried using it for a initramfs, only for things like container images