* update-alternatives automatic rename happens too late
@ 2013-03-13 13:35 Martin Jansa
2013-03-13 15:51 ` Mark Hatle
0 siblings, 1 reply; 2+ messages in thread
From: Martin Jansa @ 2013-03-13 13:35 UTC (permalink / raw)
To: Mark Hatle; +Cc: openembedded-core
[-- Attachment #1: Type: text/plain, Size: 610 bytes --]
When using update-alternatives feature to rename targets to
target.${BPN} it happens in perform_packagecopy, but for recipes where
that target is also staged to sysroot it's too late, both recipes will
try to install the same target file to sysroot.
One way to resolve this is to rename target to target.${BPN} directly in
recipe's do_install, but can we move this automatic renaming to
do_install? More importantly do we have some examples where we need to
run u-a for files in sysroot (e.g. when unpacking archive from
sstate-cache)?
--
Martin 'JaMa' Jansa jabber: Martin.Jansa@gmail.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: update-alternatives automatic rename happens too late
2013-03-13 13:35 update-alternatives automatic rename happens too late Martin Jansa
@ 2013-03-13 15:51 ` Mark Hatle
0 siblings, 0 replies; 2+ messages in thread
From: Mark Hatle @ 2013-03-13 15:51 UTC (permalink / raw)
To: Martin Jansa; +Cc: openembedded-core
On 3/13/13 8:35 AM, Martin Jansa wrote:
> When using update-alternatives feature to rename targets to
> target.${BPN} it happens in perform_packagecopy, but for recipes where
> that target is also staged to sysroot it's too late, both recipes will
> try to install the same target file to sysroot.
>
> One way to resolve this is to rename target to target.${BPN} directly in
> recipe's do_install, but can we move this automatic renaming to
> do_install? More importantly do we have some examples where we need to
> run u-a for files in sysroot (e.g. when unpacking archive from
> sstate-cache)?
>
It was specifically NOT done via a do_install(_append) because other things
could have been appending and doing odd behaviors that may or may not be supported.
Also as far as the sysroot went, all of the files that collided were
executables, man pages, etc... (not libraries or headers!) so they generally
are not used during the compilation process, but I can see with the additional
sysroot checking that this can add problems.
Perhaps the right fix is to add an additional task before the do_package (where
the sysroot is setup) but after the do_install that is specific to the
update-alternatives configuration? Change from doing it in the ${D} directory
as well...
--Mark
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2013-03-13 16:08 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-03-13 13:35 update-alternatives automatic rename happens too late Martin Jansa
2013-03-13 15:51 ` Mark Hatle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox