On Thursday, November 26, 2020 at 2:09:56 PM UTC+5:30 Henning Schild wrote:

> Am Thu, 26 Nov 2020 13:33:28 +0530 
> schrieb Vijai Kumar K <Vijaikumar_...@mentor.com>: 
>
> > When we use --use-uuid along with bootimg plugin, it is observed 
> > that the uuid updated in /etc/fstab is not same as the one the 
> > disk created by bootimg has. 
> > 
> > The latest wic[1] creates a UUID for use for these plugins, our ISAR 
> > version of these plugins were not consuming the uuid generated by 
> > wic while generating the partition. Instead a default volume id is 
> > generated which is different from what is written by wic in fstab; 
> > resulting in boot failure. 
> > 
> > Fix the ISAR bootimg plugins to use the uuid from wic. 
> > Also, enable --use-uuid with bootimg-pcbios-isar plugin in 
> > common-isar.wks.inc for a testcase. 
> > 
> > [1] 
> > 
> https://github.com/openembedded/openembedded-core/commit/9256b8799495634ee8aee5d16ff71bd6e6e25ed4 
> > 
> > Reported-by: Henning Schild <henning...@siemens.com> 
> > Signed-off-by: Vijai Kumar K <Vijaikumar_...@mentor.com> 
> > --- 
> > meta-isar/scripts/lib/wic/canned-wks/common-isar.wks.inc | 2 +- 
> > meta/scripts/lib/wic/plugins/source/bootimg-efi-isar.py | 3 ++- 
> > meta/scripts/lib/wic/plugins/source/bootimg-pcbios-isar.py | 3 ++- 
> > 3 files changed, 5 insertions(+), 3 deletions(-) 
> > 
> > diff --git a/meta-isar/scripts/lib/wic/canned-wks/common-isar.wks.inc 
> > b/meta-isar/scripts/lib/wic/canned-wks/common-isar.wks.inc index 
> > cc38a01..4290edb 100644 --- 
> > a/meta-isar/scripts/lib/wic/canned-wks/common-isar.wks.inc +++ 
> > b/meta-isar/scripts/lib/wic/canned-wks/common-isar.wks.inc @@ -1,3 
> > +1,3 @@ # This file is included into 3 canned wks files from this 
> > directory -part /boot --source bootimg-pcbios-isar --ondisk sda 
> > --label boot --active --align 1024 +part /boot --use-uuid --source 
> > bootimg-pcbios-isar --ondisk sda --label boot --active --align 1024 
> > part / --source rootfs --ondisk sda --fstype=ext4 --label platform 
> > --align 1024 --exclude-path=boot diff --git 
>
> This is a deviation that you needed for testing, but it is not what 
> upstream wic does. I would suggest to drop that. 
>

Will drop that in v2.
 

>
> > a/meta/scripts/lib/wic/plugins/source/bootimg-efi-isar.py 
> > b/meta/scripts/lib/wic/plugins/source/bootimg-efi-isar.py index 
> > 2285d2e..6b2fbc2 100644 --- 
> > a/meta/scripts/lib/wic/plugins/source/bootimg-efi-isar.py +++ 
> > b/meta/scripts/lib/wic/plugins/source/bootimg-efi-isar.py @@ -276,7 
> > +276,8 @@ class BootimgEFIPlugin(SourcePlugin): # dosfs image, 
> > created by mkdosfs bootimg = "%s/boot.img" % cr_workdir 
> > - dosfs_cmd = "mkdosfs -n efi -C %s %d" % (bootimg, blocks) 
> > + dosfs_cmd = "mkdosfs -n efi -i %s -C %s %d" % \ 
> > + (part.fsuuid, bootimg, blocks) 
> > exec_cmd(dosfs_cmd) 
>
> This seems like stuff that was forgotten about when bumping wic, right? 
>
> Maybe worth diffing our fork plugins with their originals, in old and 
> new wic to see if anything else was forgotten. 
>

Yes. I will try to align our plugins close to their respective parent in 
upstream and will
send out a separate patchset if there are any changes. 

Thanks,
Vijai Kumar K
 

>
> Henning 
>
> > mcopy_cmd = "mcopy -i %s -s %s/* ::/" % (bootimg, hdddir) 
> > diff --git 
> > a/meta/scripts/lib/wic/plugins/source/bootimg-pcbios-isar.py 
> > b/meta/scripts/lib/wic/plugins/source/bootimg-pcbios-isar.py index 
> > 493615b..b3d71c5 100644 --- 
> > a/meta/scripts/lib/wic/plugins/source/bootimg-pcbios-isar.py +++ 
> > b/meta/scripts/lib/wic/plugins/source/bootimg-pcbios-isar.py @@ 
> > -209,7 +209,8 @@ class BootimgPcbiosIsarPlugin(SourcePlugin): # dosfs 
> > image, created by mkdosfs bootimg = "%s/boot.img" % cr_workdir 
> > - dosfs_cmd = "mkdosfs -n boot -S 512 -C %s %d" % (bootimg, 
> > blocks) 
> > + dosfs_cmd = "mkdosfs -n boot -i %s -S 512 -C %s %d" % \ 
> > + (part.fsuuid, bootimg, blocks) 
> > exec_native_cmd(dosfs_cmd, native_sysroot) 
> > 
> > mcopy_cmd = "mcopy -i %s -s %s/* ::/" % (bootimg, hdddir) 
>
>