Re: [yocto] [PATCH 5/9] task-core-boot: Allow DISTRO to disable keymaps

Re: [yocto] [PATCH 5/9] task-core-boot: Allow DISTRO to disable keymaps

[Koen Kooi]

[-- Attachment #1: Type: text/plain, Size: 1529 bytes --]


Op 21 dec. 2011, om 11:09 heeft Martin Jansa het volgende geschreven:

> On Wed, Dec 21, 2011 at 01:02:51AM -0800, Darren Hart wrote:
>> NOTE: I'm looking for a better way to do this.
>> 
>> keymaps pulls in gettext, which can't build without wide character support.
>> 
>> Rather than have to add a -tiny variant of every MACHINE to drop "keyboard" from
>> the MACHINE_FEATURES, just enable the DISTRO to define what gets added to
>> RDEPENDS_task-core-boot when MACHINE_FEATURES includes keyboard. By default,
>> keep "keymaps", but use DISTRO_KEYMAPS if it is set (to "" for example).
>> 
>> Signed-off-by: Darren Hart <dvhart@linux.intel.com>
>> ---
>> meta/recipes-core/tasks/task-core-boot.bb |    7 ++++++-
>> 1 files changed, 6 insertions(+), 1 deletions(-)
>> 
>> diff --git a/meta/recipes-core/tasks/task-core-boot.bb b/meta/recipes-core/tasks/task-core-boot.bb
>> index 05c280d..e097836 100644
>> --- a/meta/recipes-core/tasks/task-core-boot.bb
>> +++ b/meta/recipes-core/tasks/task-core-boot.bb
>> @@ -17,6 +17,11 @@ PR = "r9"
>> MACHINE_ESSENTIAL_EXTRA_RDEPENDS ?= ""
>> MACHINE_ESSENTIAL_EXTRA_RRECOMMENDS ?= ""
>> 
>> +#
>> +# Set by the distro configuration to empty if not supported
>> +#
>> +DISTRO_KEYMAPS ?= "keymaps"
> 
> Why not VIRTUAL-RUNTIME_keymaps like everywhere else?

Why not just create a proper task for -tiny instead of making the already horrible task-core-boot even more horrible?

Besided that: this patch needs to go to the oe-core list

regards,

Koen

[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 169 bytes --]

Re: [yocto] [PATCH 5/9] task-core-boot: Allow DISTRO to disable keymaps

[Darren Hart]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



On 12/21/2011 05:19 AM, Koen Kooi wrote:
> 
> Op 21 dec. 2011, om 11:09 heeft Martin Jansa het volgende 
> geschreven:
> 
>> On Wed, Dec 21, 2011 at 01:02:51AM -0800, Darren Hart wrote:
>>> NOTE: I'm looking for a better way to do this.
>>> 
>>> keymaps pulls in gettext, which can't build without wide 
>>> character support.
>>> 
>>> Rather than have to add a -tiny variant of every MACHINE to
>>> drop "keyboard" from the MACHINE_FEATURES, just enable the
>>> DISTRO to define what gets added to RDEPENDS_task-core-boot
>>> when MACHINE_FEATURES includes keyboard. By default, keep
>>> "keymaps", but use DISTRO_KEYMAPS if it is set (to "" for
>>> example).
>>> 
>>> Signed-off-by: Darren Hart <dvhart@linux.intel.com> --- 
>>> meta/recipes-core/tasks/task-core-boot.bb |    7 ++++++- 1
>>> files changed, 6 insertions(+), 1 deletions(-)
>>> 
>>> diff --git a/meta/recipes-core/tasks/task-core-boot.bb 
>>> b/meta/recipes-core/tasks/task-core-boot.bb index 
>>> 05c280d..e097836 100644 --- 
>>> a/meta/recipes-core/tasks/task-core-boot.bb +++ 
>>> b/meta/recipes-core/tasks/task-core-boot.bb @@ -17,6 +17,11 @@
>>> PR = "r9" MACHINE_ESSENTIAL_EXTRA_RDEPENDS ?= "" 
>>> MACHINE_ESSENTIAL_EXTRA_RRECOMMENDS ?= ""
>>> 
>>> +# +# Set by the distro configuration to empty if not supported
>>>  +# +DISTRO_KEYMAPS ?= "keymaps"
>> 
>> Why not VIRTUAL-RUNTIME_keymaps like everywhere else?
> 
> Why not just create a proper task for -tiny instead of making the 
> already horrible task-core-boot even more horrible?
> 
> Besided that: this patch needs to go to the oe-core list

I addressed both of these in 0/9.

task-core-boot really doesn't seem horrible to me. Recreating it
everytime somone wanted a slightly different image config seems like
it would explode into a ton of mostly duplicated
task-core-boot-VARIANT.bb files.

As for the list, I opted to keep everything together for the RFC as
without the distro definition patches the rest had no real context,
and that meant picking a list oe-core or yocto - since the ultimate
goal was a distro definition for yocto, I sent to yocto. I could have
sent to oe, then someone would have been upset that I didn't send to
yocto. As I said in 0/9, the non-RFC patches will go to their
respective lists.

- -- 
Darren Hart
Intel Open Source Technology Center
Yocto Project - Linux Kernel
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJO8gH7AAoJEKbMaAwKp3642RkH/iAXzR42uEF5MfCrcQQtuKd+
W5VieJRmMZ8Yfdd9DtNxBcBLuoLSECETFKQihBq95SaEgHacTsw7H3mAY2lPA/qq
ZUa4RczNC/21CHkKh5VZIYJYY0cIqMJou/7hvyuLyqNuMp9DYAAw4tNWKNNd78QI
WUvsB9BRW4fDw17lx3ZopyK1ge7U/tWhmCaxlcls0g9s++vWZPiUxQN9h+ZPPluj
qQ86sb8M3mO6kM6AvWafgQgMAc2fRaKlvvns9YR49kltaeAq/6yBRvowR5x3FLSq
xz4F2tbtlUuhz56q72y23wJN9/eF0ll1VnKwACTquYNpmWPQGcHe6KQICEluo3Y=
=JJu4
-----END PGP SIGNATURE-----