Age | Commit message (Collapse) | Author |
|
Users may use "sumo" or previous release branches to build for this
machine.
Signed-off-by: Arsalan H. Awan <Arsalan_Awan@mentor.com>
|
|
Signed-off-by: Sanjay R Mehta <sanju.mehta@amd.com>
|
|
This enables the ability to get max. and used UVD handles from
userspace.
JIRA Ticket: INTAMDDET-1890
Signed-off-by: Ahsan Hussain <ahsan_hussain@mentor.com>
|
|
These configurations conflict with KGDB debugging in
some cases and generate stack dumps in various other
situations.
The stack dumps related to mwaitx are properly fixed
through
https://patchwork.kernel.org/patch/8549611/
but the mentioned patch is only available for Yocto
kernel 4.8 and we don't use these so it is much
better to avoid the problem with this for the time
being rather than backporting the patch.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Fixes warnings printed out during kernel_configcheck
phase regarding CONFIG_BT_6LOWPAN and
CONFIG_DRM_I915_PRELIMINARY_HW_SUPPORT.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This commit does a few things which cannot be separated easily
so it cannot be split in to separate commits.
* Drops all 4.1 kernel bits.
* Moves all common patches from meta-amdfalconx86 to common.
* Moves SE/BE builds to 4.4 kernel.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
|
|
|
|
Some components that are required for the I2S audio
to work correctly are part of the AMDGPU driver like
the amd_gnb_bus. The AMDGPU driver is not RT safe yet
so it is disabled which makes the RT builds fail.
We now disable these failing components on the RT
kernel builds to cover up the issue.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Sanjay R Mehta <sanju.mehta@amd.com>
|
|
Signed-off-by: Drew Moseley <drew_moseley@mentor.com>
|
|
Signed-off-by: Sanjay Mehta <sanju.mehta@amd.com>
Signed-off-by: Arindam Nath <arindam.nath@amd.com>
|
|
These configurations are already provided through the common
layer.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This was only being applied to the amdfalconx86 BSP earlier.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The issue was first noted and fixed for the amdfalconx86 BSP. We
now move it to common so it applies across all.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
- amdfalconx86 / SteppeEagle: IMC and AMD SPI SDK
- BaldEagle: Enable IMC
Signed-off-by: Arindam Nath <arindam.nath@amd.com>
Signed-off-by: Michael Powell <michael_powell@mentor.com>
|
|
|
|
Some configurations were duplicated among different
configuration fragments or enabled in one and then
disabled in the other vice versa. This has been
consolidated and corrected now.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The closest kernel understandable machine for the amdfalconx86
platform is the common-pc-64. We now use it as KMACHINE per
the yocto standards and manipulate configurations
accordingly.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
linux-yocto/rt: split out standard only configurations
|
|
linux-yocto: fix kgdb hangs and warnings
|
|
Some of the configuration being passed through user config fragment
are explicitly disabled in the kernel for RT type. This produces
build warnings that the configuration does not apply cleanly.
We split out the mentioned configuration to a separate config
fragment and use that in case of standard type kernel only.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
KGDB does not play nice if DEBUG_PREEMPT and DEBUG_RODATA
are enabled in the kernel. The connecting GDB gives
warning that further execution would not be possible
once it hits a breakpoint. Additionally, continuing
or stepping while debugging runs into hangs sometimes
and freezes the target.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This enables support for touchscreens with multitouch
support. The hid-multitouch driver supports a number
of devices so it would be a good idea keeping it in
as a module.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The AMD Falcon family uses an R8168 NIC for ethernet
which is a Realtek based card. The in-kernel R8169
driver shows flaky behavior on some of the Falcon
family variants so we integrate and leverage through
the official R8168 driver provided by Realtek.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The handling of AMD CZ SMBus device isn't yet available
in the mainline 4.1 kernel so we patch the support in
to get the on-board eMMC working for the Falcon family
of platforms.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
In order to use MICROCODE_EARLY which is an intention for the
latest AMD GPU to work correctly the MICRODE configuration
needs to be built-in (y) rather than configured as a
module.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
We disable the R8168 driver which is not needed anymore as
we've dropped it while upgrading for 4.1 kernel version
and the in-tree R8169 driver handles compatibility
issues with the DASH controller which were seen earlier
with the 3.14 kernel version.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
With the upgrade to 4.1.15 a separate directory was created
for patches and configurations which applied on the 4.1 kernel
separately although some of the old configuration fragments were
still used. With the removal of version 3.14 this is no longer
necessary, creates confusion and would be harder to maintain
at the end of the day.
We now consolidate these directories to have a single one
which provides everything needed to go through a sane
build.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
We now use 4.1.15 for both RT and standard kernel so drop
the unneeded 3.14 kernel.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Fix kernel config warnings
|
|
Kernel modifications
|
|
The uprobe changes are available in the kernel with v3.14.24
which is in use currently so the patching fails. We also
need to adjust the crui.scc inclusion as was done earlier
for standard kernel as that is a mainstream bug resolved
in v3.14.29 and later.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Their are duplicate configs found in amdfalconx86.cfg and
amdfalconx86-user-config.cfg. Also, there are configurations
which are un-needed and do not make it to the final defconfig.
This generates warnings during the build process.
We fix this by removing the duplicate configs and removing
the unnecessary fragments.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Due to a faulty CRC generation issue the Module.symvers file
is updated incorrectly for some exported symbols which leads
to build time warnings and runtime failures for loadable
modules.
We now use an upstream kernel patch which fixes the problem
by correctly handling typeof() for generation of kernel
symbols.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The kernel hangs while trying to do a load_idt in the
early start up phase and there's nothing obvious to observe
on any of the consoles.
This happens because jmp targets for early_idt_handlers are
global which should always generate a long jmp which in
theory should work correctly but tends to fail with newer
assemblers.
Patch backported from
https://patchwork.kernel.org/patch/6469791/
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
No need to add the amdfalconx86-user-patches.scc to
the amdfalconx86.scc file since it is explicitly listed
in the SRC_URI.
Signed-off-by: Drew Moseley <drew_moseley@mentor.com>
|
|
Signed-off-by: Arindam Nath <arindam.nath@amd.com>
|
|
Signed-off-by: Arindam Nath <arindam.nath@amd.com>
|
|
The detection and enumeration of generic WiFi devices
is currently not available due to lack of the driver
support.
We now enable some generic drivers to cover for these
devices.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Arindam Nath <arindam.nath@amd.com>
|
|
We now enable the in-tree r8168 driver and disable
the externally built module. This gives us the ability
to boot over NFS.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The NICs on AMD Falcon family platforms utilize the
r8168 driver from realtek and fail to perform correctly
with the built-in r8169 driver.
We now merge the r8168 driver source in the kernel so
we are able to use it as built-in rather than an
external module.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
|
|
The amdgpu configs are only relevant for the standard
kernel as the driver is not yet RT-safe.
We now provide these configs through a separate
file so it could be handled differently for both
kernel types.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This commit enables support for the PREEMPT_RT kernel by
leveraging the linux-yocto-rt and providing required
configurations.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The dropped kernel configs are defined twice in the
same configuration fragment which produces a warning
in the configuration process.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The CONFIG_BACKLIGHT_LCD_SUPPORT is required for DRM driver,
this is added to avoid the below warning from kernel's compile
process
warning: (DRM_RADEON && DRM_AMDGPU && DRM_NOUVEAU && DRM_I915
&& DRM_GMA500 && DRM_SHMOBILE && DRM_TILCDC && FB_BACKLIGHT
&& USB_APPLEDISPLAY && FB_OLPC_DCON && ASUS_LAPTOP &&
SONY_LAPTOP && THINKPAD_ACPI && EEEPC_LAPTOP && ACPI_CMPC
&& SAMSUNG_Q10) selects BACKLIGHT_CLASS_DEVICE which has unmet
direct dependencies (HAS_IOMEM && BACKLIGHT_LCD_SUPPORT)
Please note that this warning can only be observed in the
log.do_compile for the kernel.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Drew Moseley <drew_moseley@mentor.com>
|