Age | Commit message (Collapse) | Author |
|
This renames the snowyowl machine and its all references to e3000.
Signed-off-by: Arsalan H. Awan <Arsalan_Awan@mentor.com>
|
|
This fixes the driver to report link status correctly otherwise
we run into situations where misleading timeout errors are
generated.
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
This removes the leftovers from 08e1e59 in b7d8bab which are not
required anymore as it was fixed upstream -> oe-core -> 148fd7a
Signed-off-by: Arsalan H. Awan <Arsalan_Awan@mentor.com>
|
|
Just like we've bisected for the XGBE fragment, this
now allows the eMMC patches to be overriden from other
BSPs which want to modify/drop the fragment.
We're dropping the fragment name from the snowyowl layer
SRC_URI because the common append now provides it already.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Separate out XGBE patches from the GPU patches fragment. We
apply this on all the BSPs and any BSP that wants to override
the fragment can simply provide the same named fragment from
its own layer.
We're dropping the fragment name from the snowyowl layer
SRC_URI because the common append now provides it already.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Just like the AMD XGBE patches we're now bisecting the
eMMC patches so different BSPs can provide a different
patchset while the most common piece will be kept in
the common layer.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
We're trying to consolidate the common layer so it can be
used for multiple BSPs. This separates out the AMD XGBE
fragment so it can be overriden from elsewhere.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
We're now supporting the 4.14 kernel.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
|
|
This reverts commit 97734e5e31f0436ca1b966ef81f2a3813ef82190.
Network instabilities with 1 Gbps & 10 Gbps Copper SFPs were introduced
into amd-xgbe with this commit.
Fixes issue: AMDDET-503
Signed-off-by: Arsalan H. Awan <Arsalan_Awan@mentor.com>
|
|
The upgraded version of lttng-modules from upstream
does not require the instrumentation backports so
drop them and update the version checking workaround
in order to fix the build.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
The change being applied to the recipe is generic and
breaks whenever the recipe version is updated so use
a wildcard for versioning to keep it maintainable.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The backports for Spectre/Meltdown changes include things
that break the builds for lttng-modules otherwise these
changes. We get build failures such as
git/probes/../probes/lttng-tracepoint-event-impl.h:142:6: error: conflicting types for 'trace_kvm_mmio'
| void trace_##_name(_proto);
| ^
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This moves the VIRTIO configs to the KVM configuration
fragment and enables CONFIG_VIRTIO_INPUT.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This moves the parts that can be utilized through the
common layer for the RT kernel and generalizes the
support.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The RT kernel patches modify the interrupt handling mechanisms
in a way that using do_softirq directly is prohibited. Rather
the support forces the user to use a thread based version for
the same sort of funtionality by calling thread_do_softirq.
This fixes the build issue by calling the appropriate function
depending on the kernel configuration.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This will allow for an easier maintenance procedure such that
the common fragments can be handled more seamlessly.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This drops the unnecessary MCE patches which conflict
with the upcoming RT kernel support and everything
functions as expected without these as well.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
These generate unnecessary boot logs and are not needed
for the platform to work correctly.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
AMD Spectre/Meltdown Upstream backports
|
|
This moves further common settings from the BSP
specific appends to the common fragment.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
The kernel 4.9 recipe is being used for v1000 as well
as snowyowl. Move the base recipe to common and then
use appends for BSP overrides.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
This enables the userspace I/O drivers support
which is required by DPDK to function properly.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
snowyowl-user-config.cfg: set CONFIG_MMC_BLOCK=y
|
|
Snowyowl
|
|
The Snowy Owl board features an eMMC device that can
be used as a boot media. If the block device isn't
made part of the kernel the installation mechanism
which runs through initrd without kernel modules
fails to find the device which can be selected as
an installation candidate.
Signed-off-by: Awais Belal <awais_belal@mentor.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|
|
Signed-off-by: Sudheesh Mavila <sudheesh.mavila@amd.com>
|