Master broken?

Master broken?

[Gary Thomas]

I just updated to org.openembedded.dev 556819235f501bf762c8d8cc9f41a1a8244b106f

Now I get these errors during the initial parse:
   $ bitbake xfce-image
   NOTE: Handling BitBake files: | (0046/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing /local/Angstrom_BeagleBoard/openembedded 
/recipes/images/x11-gpe-image.bb
   NOTE: Handling BitBake files: \ (0053/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing 
/local/Angstrom_BeagleBoard/openembedded/recipes/images/angstrom-gnome-image.bb
   NOTE: Handling BitBake files: \ (0113/7335) [ 1 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing 
/local/Angstrom_BeagleBoard/openembedded/recipes/images/helloworld-image.bb
    ... etc

What am I missing?

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

Re: Master broken?

[Gary Thomas]

On 04/04/2011 08:30 AM, Gary Thomas wrote:
> I just updated to org.openembedded.dev 556819235f501bf762c8d8cc9f41a1a8244b106f

To be clear, local.conf includes these settings:
   DISTRO = "angstrom-2008.1"
   MACHINE = "beagleboard"
   IMAGE_PKGTYPE ?= "ipk"

> Now I get these errors during the initial parse:
> $ bitbake xfce-image
> NOTE: Handling BitBake files: | (0046/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing /local/Angstrom_BeagleBoard/openembedded
> /recipes/images/x11-gpe-image.bb
> NOTE: Handling BitBake files: \ (0053/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
> /local/Angstrom_BeagleBoard/openembedded/recipes/images/angstrom-gnome-image.bb
> NOTE: Handling BitBake files: \ (0113/7335) [ 1 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
> /local/Angstrom_BeagleBoard/openembedded/recipes/images/helloworld-image.bb
> ... etc
>
> What am I missing?
>

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

Re: Master broken?

[Sachin Kamboj]

The problem is "angstrom-2008.1" has been retired and is no longer available in the master branch. Use "angstrom-2010.x" instead as your distro. Alternatively, use the maintenance branch:

http://www.angstrom-distribution.org/angstrom-20081-moves-maintenance-branch
 
/Sachin.

On Apr 4, 2011, at 10:32 AM, Gary Thomas wrote:

> On 04/04/2011 08:30 AM, Gary Thomas wrote:
>> I just updated to org.openembedded.dev 556819235f501bf762c8d8cc9f41a1a8244b106f
> 
> To be clear, local.conf includes these settings:
>  DISTRO = "angstrom-2008.1"
>  MACHINE = "beagleboard"
>  IMAGE_PKGTYPE ?= "ipk"
> 
>> Now I get these errors during the initial parse:
>> $ bitbake xfce-image
>> NOTE: Handling BitBake files: | (0046/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing /local/Angstrom_BeagleBoard/openembedded
>> /recipes/images/x11-gpe-image.bb
>> NOTE: Handling BitBake files: \ (0053/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
>> /local/Angstrom_BeagleBoard/openembedded/recipes/images/angstrom-gnome-image.bb
>> NOTE: Handling BitBake files: \ (0113/7335) [ 1 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
>> /local/Angstrom_BeagleBoard/openembedded/recipes/images/helloworld-image.bb
>> ... etc
>> 
>> What am I missing?
>> 
> 
> -- 
> ------------------------------------------------------------
> Gary Thomas                 |  Consulting for the
> MLB Associates              |    Embedded world
> ------------------------------------------------------------
> 
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

Re: Master broken?

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1263 bytes --]

On Mon, Apr 04, 2011 at 08:32:41AM -0600, Gary Thomas wrote:
> On 04/04/2011 08:30 AM, Gary Thomas wrote:
> > I just updated to org.openembedded.dev 556819235f501bf762c8d8cc9f41a1a8244b106f
> 
> To be clear, local.conf includes these settings:
>    DISTRO = "angstrom-2008.1"
>    MACHINE = "beagleboard"
>    IMAGE_PKGTYPE ?= "ipk"
> 
> > Now I get these errors during the initial parse:
> > $ bitbake xfce-image
> > NOTE: Handling BitBake files: | (0046/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing /local/Angstrom_BeagleBoard/openembedded
> > /recipes/images/x11-gpe-image.bb
> > NOTE: Handling BitBake files: \ (0053/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
> > /local/Angstrom_BeagleBoard/openembedded/recipes/images/angstrom-gnome-image.bb
> > NOTE: Handling BitBake files: \ (0113/7335) [ 1 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
> > /local/Angstrom_BeagleBoard/openembedded/recipes/images/helloworld-image.bb
> > ... etc
> >
> > What am I missing?

angstrom-2008.1

http://lists.linuxtogo.org/pipermail/angstrom-distro-devel/2011-March/004383.html

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

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

Re: Master broken?

[Gary Thomas]

On 04/04/2011 08:37 AM, Sachin Kamboj wrote:
> The problem is "angstrom-2008.1" has been retired and is no longer available in the master branch. Use "angstrom-2010.x" instead as your distro. Alternatively, use the maintenance branch:
>
> http://www.angstrom-distribution.org/angstrom-20081-moves-maintenance-branch

Thanks, that got me going.

Why didn't I get an error when I selected DISTRO="angstrom-2008.1"?  There is no
longer a .../conf/distro/angstrom-2008.1.conf file on the master branch.

> On Apr 4, 2011, at 10:32 AM, Gary Thomas wrote:
>
>> On 04/04/2011 08:30 AM, Gary Thomas wrote:
>>> I just updated to org.openembedded.dev 556819235f501bf762c8d8cc9f41a1a8244b106f
>>
>> To be clear, local.conf includes these settings:
>>   DISTRO = "angstrom-2008.1"
>>   MACHINE = "beagleboard"
>>   IMAGE_PKGTYPE ?= "ipk"
>>
>>> Now I get these errors during the initial parse:
>>> $ bitbake xfce-image
>>> NOTE: Handling BitBake files: | (0046/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing /local/Angstrom_BeagleBoard/openembedded
>>> /recipes/images/x11-gpe-image.bb
>>> NOTE: Handling BitBake files: \ (0053/7335) [ 0 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
>>> /local/Angstrom_BeagleBoard/openembedded/recipes/images/angstrom-gnome-image.bb
>>> NOTE: Handling BitBake files: \ (0113/7335) [ 1 %]ERROR: global name 'runtime_mapping_rename' is not defined while parsing
>>> /local/Angstrom_BeagleBoard/openembedded/recipes/images/helloworld-image.bb
>>> ... etc
>>>
>>> What am I missing?

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

Re: Master broken?

[Dr. Michael Lauer]

Am 04.04.2011 um 16:43 schrieb Gary Thomas:

> On 04/04/2011 08:37 AM, Sachin Kamboj wrote:
>> The problem is "angstrom-2008.1" has been retired and is no longer available in the master branch. Use "angstrom-2010.x" instead as your distro. Alternatively, use the maintenance branch:
>> 
>> http://www.angstrom-distribution.org/angstrom-20081-moves-maintenance-branch
> 
> Thanks, that got me going.
> 
> Why didn't I get an error when I selected DISTRO="angstrom-2008.1"?  There is no
> longer a .../conf/distro/angstrom-2008.1.conf file on the master branch.

Take a look at bitbake.conf The DISTRO configuration is included via the
'include' directive, which is a permissive include – in contrast to 'require',
which is a mandatory include.

I don't recall our reasons for keeping it like that, but I remember that 'require'
wasn't present at the time we wrote bitbake.conf. Perhaps we should reconsider
that 'include'.

Cheers,

:M:

Re: Master broken?

[Phil Blundell]

On Tue, 2011-04-05 at 11:22 +0200, Dr. Michael Lauer wrote:
> Take a look at bitbake.conf The DISTRO configuration is included via the
> 'include' directive, which is a permissive include – in contrast to 'require',
> which is a mandatory include.
> 
> I don't recall our reasons for keeping it like that, but I remember that 'require'
> wasn't present at the time we wrote bitbake.conf. Perhaps we should reconsider
> that 'include'.

Partly because, as you say, bitbake didn't support "require" at the
time; and also partly because, when that part of bitbake.conf was
originally written, both DISTRO and MACHINE were much more optional than
they are today.  (In fact, the sample local.conf still implies that it's
legitimate to not select a MACHINE or DISTRO, though in practice I'm not
sure that it's feasible to get any useful results if you don't.)

I agree, it seems reasonable to change that include to a require.
Anybody who genuinely doesn't want a distro configuration is presumably
going to be savvy enough to create a dummy file in order to keep bitbake
happy.

p.

master broken?

[Leif Lindholm]

Hi,

Preparing a new prerequisite set for the ARM* EFI changes, I find
myself to test build any of the x86 targets after

21e4a6fa0 multiboot: disentangle multiboot and multiboot2

Build fails with
loader/multiboot_mbi2.c:25:29: fatal error: grub/multiboot2.h: No such file or directory  #include <grub/multiboot2.h>

For now, I will submit the set rebased to the commit before that.

/
    Leif

Re: master broken?

[Vladimir 'phcoder' Serbinenko]

[-- Attachment #1: Type: text/plain, Size: 510 bytes --]

Repairing

On Tue, Sep 5, 2017, 21:11 Leif Lindholm <leif.lindholm@linaro.org> wrote:

> Hi,
>
> Preparing a new prerequisite set for the ARM* EFI changes, I find
> myself to test build any of the x86 targets after
>
> 21e4a6fa0 multiboot: disentangle multiboot and multiboot2
>
> Build fails with
> loader/multiboot_mbi2.c:25:29: fatal error: grub/multiboot2.h: No such
> file or directory  #include <grub/multiboot2.h>
>
> For now, I will submit the set rebased to the commit before that.
>
> /
>     Leif
>

[-- Attachment #2: Type: text/html, Size: 791 bytes --]

Re: master broken?

[Leif Lindholm]

This seems to have fixed the "bios" build.
But the efi variant (--with-efi) still breaks with

loader/multiboot.c: In function ‘efi_boot’:
loader/multiboot.c:142:47: error: ‘MULTIBOOT2_EFI_INITIAL_STATE’ undeclared (first use in this function)
    struct grub_relocator_efi_state state_efi = MULTIBOOT2_EFI_INITIAL_STATE;
                                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
loader/multiboot.c:142:47: note: each undeclared identifier is reported only once for each function it appears in
loader/multiboot.c:146:44: error: ‘grub_multiboot_payload_eip’ undeclared (first use in this function)
    state_efi.MULTIBOOT_EFI_ENTRY_REGISTER = grub_multiboot_payload_eip;                                                                                                                                  ^~~~~~~~~~~~~~~~~~~~~~~~~~

/
    Leif

On Tue, Sep 05, 2017 at 07:11:32PM +0000, Vladimir 'phcoder' Serbinenko wrote:
> Repairing
> 
> On Tue, Sep 5, 2017, 21:11 Leif Lindholm <leif.lindholm@linaro.org> wrote:
> 
> > Hi,
> >
> > Preparing a new prerequisite set for the ARM* EFI changes, I find
> > myself to test build any of the x86 targets after
> >
> > 21e4a6fa0 multiboot: disentangle multiboot and multiboot2
> >
> > Build fails with
> > loader/multiboot_mbi2.c:25:29: fatal error: grub/multiboot2.h: No such
> > file or directory  #include <grub/multiboot2.h>
> >
> > For now, I will submit the set rebased to the commit before that.
> >
> > /
> >     Leif
> >

Re: master broken?

[Vladimir 'phcoder' Serbinenko]

[-- Attachment #1: Type: text/plain, Size: 1663 bytes --]

On Wed, Sep 6, 2017, 15:43 Leif Lindholm <leif.lindholm@linaro.org> wrote:

> This seems to have fixed the "bios" build.
> But the efi variant (--with-efi) still breaks with
>
I tried x86_64-efi and it seems to work. And --with-efi doesn't seem to be
one of our options. Which exact platform fails?

>
> loader/multiboot.c: In function ‘efi_boot’:
> loader/multiboot.c:142:47: error: ‘MULTIBOOT2_EFI_INITIAL_STATE’
> undeclared (first use in this function)
>     struct grub_relocator_efi_state state_efi =
> MULTIBOOT2_EFI_INITIAL_STATE;
>
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> loader/multiboot.c:142:47: note: each undeclared identifier is reported
> only once for each function it appears in
> loader/multiboot.c:146:44: error: ‘grub_multiboot_payload_eip’ undeclared
> (first use in this function)
>     state_efi.MULTIBOOT_EFI_ENTRY_REGISTER = grub_multiboot_payload_eip;
>
>
> ^~~~~~~~~~~~~~~~~~~~~~~~~~
>
> /
>     Leif
>
> On Tue, Sep 05, 2017 at 07:11:32PM +0000, Vladimir 'phcoder' Serbinenko
> wrote:
> > Repairing
> >
> > On Tue, Sep 5, 2017, 21:11 Leif Lindholm <leif.lindholm@linaro.org>
> wrote:
> >
> > > Hi,
> > >
> > > Preparing a new prerequisite set for the ARM* EFI changes, I find
> > > myself to test build any of the x86 targets after
> > >
> > > 21e4a6fa0 multiboot: disentangle multiboot and multiboot2
> > >
> > > Build fails with
> > > loader/multiboot_mbi2.c:25:29: fatal error: grub/multiboot2.h: No such
> > > file or directory  #include <grub/multiboot2.h>
> > >
> > > For now, I will submit the set rebased to the commit before that.
> > >
> > > /
> > >     Leif
> > >
>

[-- Attachment #2: Type: text/html, Size: 2561 bytes --]

Re: master broken?

[Leif Lindholm]

On Wed, Sep 06, 2017 at 03:13:36PM +0000, Vladimir 'phcoder' Serbinenko wrote:
> On Wed, Sep 6, 2017, 15:43 Leif Lindholm <leif.lindholm@linaro.org> wrote:
> 
> > This seems to have fixed the "bios" build.
> > But the efi variant (--with-efi) still breaks with
> >
> I tried x86_64-efi and it seems to work. And --with-efi doesn't seem to be
> one of our options. Which exact platform fails?

Sorry, typo - I meant --with-platform=efi

On my amd64 (Debian Stretch), running
git clean -fdx && ./autogen.sh && ./configure --with-platform=efi && make -j40

On commit 1b18d6b0d34872ced6b3dbfc3e7957c80efbcb7a.

There is no definition of MULTIBOOT2_EFI_INITIAL_STATE in my tree, but
there is MULTIBOOT_EFI_INITIAL_STATE.

/
    Leif

> >
> > loader/multiboot.c: In function ‘efi_boot’:
> > loader/multiboot.c:142:47: error: ‘MULTIBOOT2_EFI_INITIAL_STATE’
> > undeclared (first use in this function)
> >     struct grub_relocator_efi_state state_efi =
> > MULTIBOOT2_EFI_INITIAL_STATE;
> >
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > loader/multiboot.c:142:47: note: each undeclared identifier is reported
> > only once for each function it appears in
> > loader/multiboot.c:146:44: error: ‘grub_multiboot_payload_eip’ undeclared
> > (first use in this function)
> >     state_efi.MULTIBOOT_EFI_ENTRY_REGISTER = grub_multiboot_payload_eip;
> >
> >
> > ^~~~~~~~~~~~~~~~~~~~~~~~~~
> >
> > /
> >     Leif
> >
> > On Tue, Sep 05, 2017 at 07:11:32PM +0000, Vladimir 'phcoder' Serbinenko
> > wrote:
> > > Repairing
> > >
> > > On Tue, Sep 5, 2017, 21:11 Leif Lindholm <leif.lindholm@linaro.org>
> > wrote:
> > >
> > > > Hi,
> > > >
> > > > Preparing a new prerequisite set for the ARM* EFI changes, I find
> > > > myself to test build any of the x86 targets after
> > > >
> > > > 21e4a6fa0 multiboot: disentangle multiboot and multiboot2
> > > >
> > > > Build fails with
> > > > loader/multiboot_mbi2.c:25:29: fatal error: grub/multiboot2.h: No such
> > > > file or directory  #include <grub/multiboot2.h>
> > > >
> > > > For now, I will submit the set rebased to the commit before that.
> > > >
> > > > /
> > > >     Leif
> > > >
> >