Hello Richard, On Tue, Dec 13, 2022 at 11:52:08PM +0000, Richard Purdie wrote: > On Mon, 2022-12-12 at 16:18 -0600, Patrick Williams wrote: > > In order to build a fitimage, we don't need the image to be "complete" > > but we need the cpio portion of it built since the cpio is what ends > > up inside the FIT. By reducing the dependency to `do_image_cpio` we > > are able to include an image's rootfs as the ramdisk for a FIT and > > then bundle that FIT into a larger SPI flash layout. > > > > Signed-off-by: Patrick Williams <[email protected]> > > ---
> > - d.appendVarFlag('do_assemble_fitimage_initramfs', 'depends', '
> > ${INITRAMFS_IMAGE}:do_image_complete')
> > + d.appendVarFlag('do_assemble_fitimage_initramfs', 'depends', '
> > ${INITRAMFS_IMAGE}:do_image_cpio')
>
> I merged this but then realised you can't do this.
...
> As such I reverted it.
Apologies for that. It seemed to be working on my end but I was not
going to be surprised if someone told me I had very bad ideas as was
completely overlooking something.
> The copy to
> DEPLOY_DIR_IMAGE happens during do_image_complete so you can't rely on
> this to be present until that task.
I didn't notice this issue, but I very well could have had stale data in
my DEPLOY_DIR_IMAGE. I'll investigate further. Is there any reason
this couldn't be moved?
> We need the copy and sstate code so
> we have a manifest of what is "installed" so we can clean up the task
> and replace it with code re-running.
I'll look into this as well. Is it fine to make do_image_cpio sstated?
--
Patrick Williams
signature.asc
Description: PGP signature
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#174548): https://lists.openembedded.org/g/openembedded-core/message/174548 Mute This Topic: https://lists.openembedded.org/mt/95632880/21656 Group Owner: [email protected] Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
