[OE-core] [oe-core][PATCHv2 04/30] xf86-video-omapfb: bump SRCREV and add patch from meta-oe

[Martin Jansa]

On Mon, Oct 03, 2011 at 05:00:35PM +0100, Phil Blundell wrote:
> On Mon, 2011-10-03 at 15:28 +0200, Martin Jansa wrote:
> > From: Martin Jansa <martin.jansa at gmail.com>
> > 
> > Signed-off-by: Martin Jansa <Martin.Jansa at gmail.com>
> > Signed-off-by: Richard Purdie <richard.purdie at linuxfoundation.org>
> > Signed-off-by: Martin Jansa <Martin.Jansa at gmail.com>
> > ---

Sorry about 3 SOB lines... it's because I had it from
oe-core/master-next, but I belive RP will merge master-next first where
it's only with his and mine SOB.

> This is a bit of a lame checkin message.  "Add patch from meta-oe"
> doesn't really say anything useful about what you're changing, and there
> is no long comment at all.  Also ...

Rotation didn't work when using VFRB, this patch is fixing that and is from 
koen upstream repo as patch header says and I cannot probably describe it 
better then original author.

> >@@ -1,25 +1,25 @@
> >-require xorg-driver-driver.inc
> >+require xorg-driver-video.inc
> > 
> > SUMMARY = "X.Org X server -- Texas Instruments OMAP framebuffer driver"
> >
> > DESCRIPTION = "omapfb driver supports the basic Texas Instruments OMAP \
> > framebuffer."
> >
> >-LICENSE = "MIT-X"
> >-LIC_FILES_CHKSUM = "file://src/omapfb-driver.c;beginline=1;endline=30;md5=a44c2a37e04d1c2c5f0313afb493f833"
> >+LICENSE = "MIT-X & GPLv2+"
> >+LIC_FILES_CHKSUM = "file://COPYING;md5=63e2cbac53863f60e2f43343fb34367f"
> > DEPENDS += "virtual/libx11"
> 
> ... these changes don't seem to be covered by the description above.

because COPYING file wasn't available in SRCREV used in old oe-core
version.. so this change is just simplification and improving license
metadata, because original MIT-X src/omapfb-driver.c doesn't cover what
was explained in COPYING file:
The src/omapfb.h header is under the GPL license.

> And, finally, if this is an OMAP-specific driver, why is it in oe-core
> in the first place?  I would have thought it would be better placed in
> meta-ti or some such layer.

We've discussed it on oe-devel ML already
http://lists.linuxtogo.org/pipermail/openembedded-devel/2011-September/034781.html

now we have very similar recipes in oe-core and meta-oe which is imho
worst case.. I'm fine with dropping it from oe-core and keeping it in
meta-oe, but I would prefer to keep it in meta-oe ie because
nokia900.conf is maintained in meta-smartphone layer and I don't want to
add meta-ti as dependency just because this one recipe.

Regards,
-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa at gmail.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.openembedded.org/pipermail/openembedded-core/attachments/20111004/f61e075e/attachment-0002.sig>