Re: [PATCH] image: Run copy_boot_files after rootfs postprocessing

[Harald Seiler <hws@denx.de>]

Hi Jan,

On Tue, 2020-10-13 at 12:19 +0200, Jan Kiszka wrote:
> On 25.06.20 17:33, Harald Seiler wrote:
> > In some cases, postprocessing might trigger an update of the initramfs
> > which would not appear in the deployed version.  Fix this by running
> > copy_boot_files only after all postprocessing completed.
> > 
> > Signed-off-by: Harald Seiler <hws@denx.de>
> > ---
> >  meta/classes/image.bbclass | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/meta/classes/image.bbclass b/meta/classes/image.bbclass
> > index 0150f2718573..21d634a8f34f 100644
> > --- a/meta/classes/image.bbclass
> > +++ b/meta/classes/image.bbclass
> > @@ -153,7 +153,7 @@ do_copy_boot_files() {
> >          cp -f "$dtb" "${DEPLOY_DIR_IMAGE}/"
> >      done
> >  }
> > -addtask copy_boot_files before do_rootfs_postprocess after do_rootfs_install
> > +addtask copy_boot_files before do_rootfs after do_rootfs_postprocess
> >  
> >  python do_image_tools() {
> >      """Virtual task"""
> > 
> 
> There was a lengthy discussion after this, but I don't think the
> conclusion was to hold back this patch, right? We are using it at least
> in one case internally, so this should likely go in now.

At least from my side we can drop this.  In the project where I needed
this it is now gone, because I switched to the alternate initramfs
mechanism over there.

Also, the arguments where this patch might break downstream use-cases do
hold some truth so I'd rather err on the side of caution and not go this
route.

-- 
Harald

DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-62  Fax: +49-8142-66989-80   Email: hws@denx.de