These layers all contain machine configuration that depends on having
the meta-intel common layer enabled as well, so use LAYERDEPENDS in the
layer configuration for each one to make this explicit.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
We want a failure to occur if the files specified in these statements
can't be found (e.g. when there is a misconfiguration), so use require
instead of include.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Move the interfaces file to a MACHINE appropriate subdirectory of
files so that:
a) base.bbclass appropriately marks the resultant package MACHINE
specific.
b) builds for alternative MACHINES whilst this layer is enabled
don't get the sys940x interfaces file
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
These machines have no MAC in the hardware and require that it
be set manually. Specify RANDOM_MAC for the eth0 interface which
the genmac init script will replace with RANDOM_MAC on first boot.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
For machines that do not have a MAC in hardware and with drivers that don't
generate a random one in the kernel, this init script will replace the string
RANDOM_MAC in the network/interfaces file with one generated with "ranpwd -m".
Care is taken to ensure multiple interfaces can use RANDOM_MAC and receive
unique addresses. ranpwd generates MACs with the locally administered bit set
and the multicast bit disabled.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Signed-off-by: Joshua Lock <josh@linux.intel.com>
ranpwd is used to generate random strings of various types, including
passwords, UUIDs, and MAC addresses.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
The Inforce SYS940x-ECX Developer-Ready Reference Platform features:
o Intel Atom E6xx (0.6-1.6 GHz)
o Up to 1GB on-board DDR2
o Intel Platform Controller Hub EG20T
o VGA,LVDS
o HD Audio
o SD Card
o Dual SATA
o Mini-PCIe
http://www.inforcecomputing.com/SYS940X_ECX.html
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
The new emgd-driver-bin_1.10 recipe no longer requires manually
extracting and installing emgd binaries, so remove the section that
deals with that.
It does require a new LICENSE_FLAGS_WHITELIST entry in local.conf, so
add instructions detailing that.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
The new emgd-driver-bin_1.10 recipe no longer requires manually
extracting and installing emgd binaries, so remove the section that
deals with that.
It does require a new LICENSE_FLAGS_WHITELIST entry in local.conf, so
add instructions detailing that.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
crownbay specifies a preferred version of 1.8 for emgd, but there's
really no reason to do that at this point - it should be able to use
the new 1.10 version.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
fri2 specifies a preferred version of 1.8 for emgd, but there's really
no reason to do that at this point - it should be able to use the new
1.10 version.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Add a custom asound.state for the n450 so that the alsa-state init
script sets (and restores) sound configuration appropriately.
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
A couple of things that had previously been warnings are now errors,
so they need to be fixed up.
The first problem is a comparison between the address of a static
struct and NULL, which can never be valid. A different fix for this
is upstream, which includes an API usage change; we don't need that to
fix this problem.
The second problem is a cast from pointer to integer in fbdevhw.c.
This also is fixed upstream by removing the whole section of code
which is bogus anyway, which is also done here.
This also adds a missing PR to the xserver-xorg recipe.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
A couple of things that had previously been warnings are now errors,
so they need to be fixed up.
The first problem is the same problem encountered by xserver-kdrive,
and has the same fix, namely create a new version of REGION_INIT and
have all calls that pass in a pointer to a static struct use it
instead. A different fix for this is upstream, which includes an API
change; we don't need all that to fix this problem.
The second problem is a cast from pointer to integer in fbdevhw.c.
This also is fixed upstream by removing the whole section of code
which is bogus anyway, which is also done here.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
This layer provides the initial version of the BSP for Cedar Trail
platform. The Cedar Trail platform is based on the Cedarview processor
and Tiger Point Chipset.
Signed-off-by: Kishore Bodke <kishore.k.bodke@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
XSERVER_IA32_EMGD should depend on libva-x11, which in turn depends on
libva, instead of just libva.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Acked-by: Darren Hart <dvhart@linux.intel.com>
libva provides some extra packages such as libva-glx. Add PROVIDES
for them so the rest of the build system knows about it.
We also need to add the dependency relationships between libva,
libva-x11, and libva-glx.
Also remove the bogus RDEPENDS on the display components.
Also libva-glx depends on mesa-dri - if it's not there, libva-glx
won't be built.
Also, add missing -dev and -dbg packaging.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Acked-by: Darren Hart <dvhart@linux.intel.com>
The SRCREV update is required for the cfg/efi-ext as well as patches
to the standard/* branches to support EFI booting.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Fixes [YOCTO #1797]
Building with ROOTLESS_X results in X failing start with
the following error:
i915 0000:00:02.0: VGA-1: EDID block 0 invalid.
(EE) AIGLX error: dlopen of /usr/lib/dri/i915_dri.so failed
(/usr/lib/dri/i915_dri)
(EE) AIGLX: reverting to software rendering
(EE) AIGLX error: dlopen of /usr/lib/dri/swrast_dri.so failed
(/usr/lib/dri/swrast)
(EE) GLX: could not load software renderer
The n450 BSP is the only BSP in meta-intel using ROOTLESS_X. Dropping
it restores functionality and brings it inline with the rest of the BSPs.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Include the firmware package for the iwlwifi card for task-base based images
using the MACHINE_EXTRA_RRECOMMENDS variable. This will exclude the firmware
from being installed in core-image-minimal, but it doesn't have the tooling
required to connect over wifi anyway.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Add firmware required by the Intel Centrino Adnanced-N 6205 mini-pcie card.
The content is already available in the linux-firmware repository, this patch
just adds the meta-data to include the right files and define a package.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
The current PortOrder value assumes LVDS. Change the value to first try the
primary SDVO output, then the secondary SDVO, and fallback to LVDS if neither
are present.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
The FRI2 has an optional serial expansion board. Enable the use of this
as a console for both syslinux and the OS.
Note: while the settings all appear to be correct, the serial port is
not working on my test device in syslinux or the getty. The
kernel messages do make their way to the serial port after PCI
initialization.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Have crownbay use gst-va-intel so we can easily test and make use of
the video acceleration capabilities of this machine.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
We need some extra plugins for EMGD video acceleration use and
testing, so create a new 'gst-va-intel' package group for them.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
This reverts commit acf138682a.
This breaks non-compatible builds, and should probably be done per-BSP
anyway.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
We need some extra plugins for EMGD video acceleration use and
testing, so append them to gst-meta-base.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Adding support for video acceleration means some new components need
to be installed - update the instructions to reflect that.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
This adds the video binaries extracted from the following EMGD 1.8
binary rpms (emgd/1.8.2032/IEMGD_HEAD_Linux/common/video_plugin):
gst-plugins-mixvideo-0.10.30-1.i586.rpm
gst-plugins-va-0.10.7MFLD-1.i586.rpm
gst-vabuffer-0.10.5MFLD-1.i586.rpm
mixcommon-0.1.9-1.i586.rpm
mixvbp-0.1.24-1.i586.rpm
mixvideo-0.1.31-1.i586.rpm
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
EMGD 1.8 requires libva-1.0.12, so upgrade to that version, and we
don't actually need the 1.0.10 recipe, so remove it.
This also adds -dbg and -dev packaging fixes to fix QA problems,
removes some unnecessary settings, updates checksums, etc.
Also create and use a common libva.inc, as we'll soon be adding a more
updated version of libva to the current version.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
This adds the libva_1.0.10 recipe based on the poky-contrib recipe
by Yu Ke (commit msg copied below).
The original recipe was in meta/recipes-graphics; this patch moves it
to meta-intel/recipes-multimedia since it's multimedia- and
Intel-specific.
The original commit messages (indented):
libva: add new recpies libva
libva is a library implementation of "Video Acceleration (VA) API for Linux".
libva enables and provides access to graphics hardware (GPU) acceleration for
video processing on Linux and UNIX, includes video decoding, video encoding,
subpicture blending and rendering.
This patch add the libva recipes.
Signed-off-by: Yu Ke <ke.yu@intel.com>
libva: install some missing .so files
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
