[OE-core] [RESEND PATCH v3] util-linux: create util-linux-runuser package

Ioan-Adrian Ratiu adrian.ratiu at ni.com
Tue Jan 12 09:03:01 UTC 2016


On Tue, 12 Jan 2016 10:26:02 +0200
Ioan-Adrian Ratiu <adrian.ratiu at ni.com> wrote:

> On Tue, 12 Jan 2016 15:52:50 +1300
> Paul Eggleton <paul.eggleton at linux.intel.com> wrote:
> 
> > On Wed, 23 Dec 2015 12:03:03 Ioan-Adrian Ratiu wrote:  
> > > On Tue, 22 Dec 2015 17:34:12 +0000
> > > 
> > > "Burton, Ross" <ross.burton at intel.com> wrote:    
> > > > On 15 December 2015 at 09:49, Ioan-Adrian Ratiu <adrian.ratiu at ni.com>     
> > wrote:  
> > > > > Split runuser into its own package (previously provided by util-linux).
> > > > > Since runuser is compiled only when DISTRO_FEATURES includes pam,
> > > > > the creation of util-linux-runuser is also constrained by this.    
> > > > 
> > > > What's the rationale for pulling this out into a separate package?    
> > > 
> > > I need runuser on a busybox image and can't install the entire util-linux
> > > package.    
> > > > >  PACKAGES += "${@bb.utils.contains('PACKAGECONFIG', 'pylibmount',
> > > > > 
> > > > > 'util-linux-pylibmount', '', d)}"
> > > > > +PACKAGES =+ "${@bb.utils.contains('DISTRO_FEATURES', 'pam',
> > > > > 'util-linux-runuser', '', d)}"    
> > > > 
> > > > No need for conditionals: empty packages are not created.    
> > > 
> > > Thank you for this tip, I've modified and sent v4.    
> > 
> > Unfortunately this wasn't right - the package being excluded from PACKAGES was 
> > serving to avoid the libpam dependency when pam was not in DISTRO_FEATURES; as 
> > of this patch we are now building it unconditionally leading to a warning.
> > 
> > Ioan-Adrian, could you please send a patch that changes it back to how you had 
> > it in v3?  
> 
> Yes, I'll send v5 which will revert the change from v4.
> 

Uh oh, I just rebased on origin/master and it looks like v4 was commited there a few hours ago.

So I'll just send a new patch to introduce the pam conditional on top of the already integrated change.

> > 
> > Thanks,
> > Paul
> >   
> 




More information about the Openembedded-core mailing list