Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Martin Jansa]

On Tue, Dec 14, 2010 at 06:40:36PM +0000, git version control wrote:
> Module: openembedded.git
> Branch: org.openembedded.dev
> Commit: 6f0ed3e2cebc76cc992ddc102ade8bfa13e17acb
> URL:    http://gitweb.openembedded.net/?p=openembedded.git&a=commit;h=6f0ed3e2cebc76cc992ddc102ade8bfa13e17acb
> 
> Author: Steve Sakoman <steve@sakoman.com>
> Date:   Sat Dec 11 07:16:26 2010 -0800
> 
> coreutils: experimental coreutils fix

what was wrong?

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Philip Balister]

On 12/14/2010 03:59 PM, Martin Jansa wrote:
> On Tue, Dec 14, 2010 at 06:40:36PM +0000, git version control wrote:
>> Module: openembedded.git
>> Branch: org.openembedded.dev
>> Commit: 6f0ed3e2cebc76cc992ddc102ade8bfa13e17acb
>> URL:    http://gitweb.openembedded.net/?p=openembedded.git&a=commit;h=6f0ed3e2cebc76cc992ddc102ade8bfa13e17acb
>>
>> Author: Steve Sakoman<steve@sakoman.com>
>> Date:   Sat Dec 11 07:16:26 2010 -0800
>>
>> coreutils: experimental coreutils fix
>
> what was wrong?
>

Sorry, I committed a fix I pulled from Steve's repo an dhe had a quick 
message on the commit.

Basically, the coreutils post install script that linked the coreutils 
version to the correct did not work. We can't figure out why, but 
removing the test solved the problem. As a side note, Poky also dropped 
the test.

Since this changes creates output that is more correct than the old 
behavior, I pushed it.

Philip

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Martin Jansa]

On Tue, Dec 14, 2010 at 04:19:07PM -0500, Philip Balister wrote:
> On 12/14/2010 03:59 PM, Martin Jansa wrote:
> > On Tue, Dec 14, 2010 at 06:40:36PM +0000, git version control wrote:
> >> Module: openembedded.git
> >> Branch: org.openembedded.dev
> >> Commit: 6f0ed3e2cebc76cc992ddc102ade8bfa13e17acb
> >> URL:    http://gitweb.openembedded.net/?p=openembedded.git&a=commit;h=6f0ed3e2cebc76cc992ddc102ade8bfa13e17acb
> >>
> >> Author: Steve Sakoman<steve@sakoman.com>
> >> Date:   Sat Dec 11 07:16:26 2010 -0800
> >>
> >> coreutils: experimental coreutils fix
> >
> > what was wrong?
> >
> 
> Sorry, I committed a fix I pulled from Steve's repo an dhe had a quick 
> message on the commit.

OK, no problem

> Basically, the coreutils post install script that linked the coreutils 
> version to the correct did not work. We can't figure out why, but 
> removing the test solved the problem.

Ah :/ never noticed it here.

> As a side note, Poky also dropped the test.

cannot find coreutils-target.inc or cherry-pick of 
http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=05794e79b4af2d2f3ded69a02c0275335b75a47c
in poky.. :)

> Since this changes creates output that is more correct than the old 
> behavior, I pushed it.

Then please remove all older coreutils versions which doesn't provide
complete set of bindir_progs, base_bindir_progs, sbindir_progs, because
ie coreutils-6.0 now breaks ie timeout:

SHR root@gjama ~ $ ll /usr/bin/timeout
lrwxrwxrwx    1 root     root            17 Dec 14 22:34 /usr/bin/timeout -> ../../bin/busybox

SHR root@gjama ~ $ opkg install coreutils
Installing coreutils (6.0-r6.0.6) to root...
update-alternatives: Linking //usr/bin/basename to basename.coreutils
...
update-alternatives: Linking //usr/bin/timeout to timeout.coreutils
...
update-alternatives: Linking //usr/bin/[ to lbracket.coreutils

SHR root@gjama ~ $ timeout
-bash: timeout: command not found
SHR root@gjama ~ $ which timeout
SHR root@gjama ~ $ ls -lah /usr/bin/timeout*
lrwxrwxrwx 1 root root 17 Dec 14 22:41 /usr/bin/timeout -> timeout.coreutils

Regards,

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Philip Balister]

On 12/14/2010 04:50 PM, Martin Jansa wrote:
>
> Then please remove all older coreutils versions which doesn't provide
> complete set of bindir_progs, base_bindir_progs, sbindir_progs, because
> ie coreutils-6.0 now breaks ie timeout:

So it looks like we can remove the coreutils-7.2 directory since there 
is no recipe for it. Does anyone object to removing coreutils-6.0?

Philip

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Frans Meulenbroeks]

2010/12/16 Philip Balister <philip@balister.org>:
> On 12/14/2010 04:50 PM, Martin Jansa wrote:
>>
>> Then please remove all older coreutils versions which doesn't provide
>> complete set of bindir_progs, base_bindir_progs, sbindir_progs, because
>> ie coreutils-6.0 now breaks ie timeout:
>
> So it looks like we can remove the coreutils-7.2 directory since there is no
> recipe for it. Does anyone object to removing coreutils-6.0?
>
> Philip

You have my ack to remove both.
Frans

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Martin Jansa]

On Thu, Dec 16, 2010 at 09:58:37AM +0100, Frans Meulenbroeks wrote:
> 2010/12/16 Philip Balister <philip@balister.org>:
> > On 12/14/2010 04:50 PM, Martin Jansa wrote:
> >>
> >> Then please remove all older coreutils versions which doesn't provide
> >> complete set of bindir_progs, base_bindir_progs, sbindir_progs, because
> >> ie coreutils-6.0 now breaks ie timeout:
> >
> > So it looks like we can remove the coreutils-7.2 directory since there is no
> > recipe for it. Does anyone object to removing coreutils-6.0?
> >
> > Philip
> 
> You have my ack to remove both.
> Frans

My too
(I've built coreutils-6.0 only to show example why that test was needed)

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

Re: [oe-commits] Steve Sakoman : coreutils: experimental coreutils fix

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 665 bytes --]

On Thu, Dec 16, 2010 at 09:58:37AM +0100, Frans Meulenbroeks wrote:
> 2010/12/16 Philip Balister <philip@balister.org>:
> > On 12/14/2010 04:50 PM, Martin Jansa wrote:
> >>
> >> Then please remove all older coreutils versions which doesn't provide
> >> complete set of bindir_progs, base_bindir_progs, sbindir_progs, because
> >> ie coreutils-6.0 now breaks ie timeout:
> >
> > So it looks like we can remove the coreutils-7.2 directory since there is no
> > recipe for it. Does anyone object to removing coreutils-6.0?
> >
> > Philip
> 
> You have my ack to remove both.
> Frans

ping?
-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Type: application/pgp-signature, Size: 205 bytes --]