| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Introduce a new embeddedsw-source recipe that is a single point for unpack,
and patch of the sources. This will allow someone to universally patch
(via a bbappend or similar) the esw sources.
As part of this work, we transfered the various patches for fsbl, pmu, plm,
and psm to the universe source recipe.
Most of this is transparent to any recipe using xlnx-embeddedsw. The class
automaticaly detects if the recipe is a user our the special embeddedsw-source
recipe and enables the correct behavior.
Since the esw sources expect the S and B to be the same, the cmake objects
get written into S, we are not able to follow the gcc-source example.
Instead recipes that use the xlnx-embeddedsw class will hard link or copy
the source files to their working directory. This avoid the penalty for
fetch, unpack and patch. Note these recipes may also add their own
SRC_URI which WILL be processed as normal. This is often used to
introduce special tcl scripts. See the do_copy_shared_src task injected
by the xlnx-embeddedsw.bbclass for technical details.
Additionally, correct some of the existing patch Upstream-Status entries
to conform to current YP guidelines.
For 2024.2 introduce a special Xil_Assert fix for the current toolchain
which is more strict on type conversion.. See the patch itself for
more details.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
| |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
|
|
|
| |
Remove most direct references to WORKDIR.
S = "${WORKDIR}" is no longer allowed, adjust this to refer to UNPACKDIR
${WORKDIR}/package is also incorrect, adjust to ${PKGD}.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Rework the image recover recipe to match image selector. imgrcry is now
the top level recipe which will pull in the required version of
image-recovery depending on build flow.
This change also updates the output file name in a number of places and
fixes an issue in image selector.
Signed-off-by: John Toomey <john.toomey@amd.com>
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
|
| |
Also add imgsel wrapper allowing for the correct image-selector recipe
to be built depending if the XSCT or SDT flow is in use
Signed-off-by: John Toomey <john.toomey@amd.com>
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
SOC_VARIANT has been remove, we are now only using the YP standard SOC_FAMILY
configuration. The defined families are: zynq, zynqmp, versal and versal-net.
Our decision of breaking up versal-net from versal, is based on the SoC CPU
changes from cortexa72/r5 to cortexa78/r52, thus we're treating it as a
different SoC family.
In order to capture the individual capabilities that we used to handle via
SOC_VARIANT, we have defined the following features (some may have been
previously defined):
- mali400 (zynqmp eg and ev)
- vcu (zynqmp ev)
- rfsoc (zynqmp dr RF capabiltiies)
- aie - (versal ai & premium)
- vdu - (versal ai)
SOC_VARIANT_ARCH and SOC_FAMILY_ARCH are now obsolete and replaced by
MACHINE_ARCH. This is based on the guideline that any recipes that use
MACHINE_FEATURES should be MACHINE_ARCH specific.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
| |
Clenaup and rework the code so that the 'generic' implementation no longers
affects xsct, or SDT (or future) variations.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
| |
Ensure that each bbappend only applies when XILINX_WITH_ESW or a related
DISTRO is enabled.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
|
|
|
| |
Adjust the way the PREFERRED_VERSION happens to allow us to use the same
login in layers that depend on this one.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
| |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|\
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|\|
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|\|
| |
| |
| |
| |
| |
| | |
Resolved conflicts
Preserved 2022* versions
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| |
| | |
Signed-off-by: Sandeep Gundlupet Raju <sandeep.gundlupet-raju@amd.com>
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Remove workaround for:
xpm_domain_iso.c:724:42: error: potential null pointer dereference [-Werror=null-dereference]
Also the other workaround can be removed, this has apparently been fixed
in the upstream code.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Update the integration to match 2023 changes.
Update the Makefile and versal pathes for 2023 specific version
Add a workaround for -Werror -Wnull-dereference failing during some compiles
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Set COMPATIBLE_HOST more loosely to allow for baremetal builds.
Update processing to use OBJCOPY and add additional deploy names.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
| |
| |
| |
| |
| | |
Signed-off-by: Sandeep Gundlupet Raju <sandeep.gundlupet-raju@amd.com>
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|/
|
|
|
|
| |
Update the integration to match latest changes.
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|
|
|
| |
Signed-off-by: Mark Hatle <mark.hatle@amd.com>
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
Update FSBL_IMAGE_NAME to align with expected file name to be
fsbl-${MACHINE}.
Signed-off-by: Sandeep Gundlupet Raju <sandeep.gundlupet-raju@xilinx.com>
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
device-tree, fsbl, plmfw, pmufw, psmfw do basic validation checks, when in
an eSDK skip these checks. Otherwise a partially configured system can't
generate an eSDK for the end user.
Sync Core and Standalone usages to be consistent.
Also cleanup the BB_DONT_CACHE to be consistent in all usages.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
fsbl, plmfw, pufw, and psmfw are reworked to provide earlier error messages,
and better handle multiconfig builds.
Also adjust MACHINE specific changes to use .* for compatibility based on the
SOC_FAMILY override. The system doesn't always set the SOC_FAMILY as
COMPATIBLE_MACHINE, so this avoids an error of incompatible machine.
Align all of the expected firmware filenames to be <firmware>-${MACHINE}.
The user can override as necessary.
The license moves to CLOSED as we don't know the license of the software being
provided by the user or other dependencies. It'll be up to the user to
reconcile it.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| |
| | |
PACKAGE_ARCH should not be MACHINE_ARCH, unless a specific override is
being used for compatibility.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
(cherry picked from commit eef9f49887e2471768b649450a0769859fd63ba5)
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Rename recipes from 2021.2 to 2022.1 as necessary.
Move internal references to 2022.1 (layer.conf and local.conf.sample)
Move kernel to 2022.1 branch, as well as various kernel modules.
Signed-off-by: Mark Hatle <mark.hatle@kernel.crashing.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Update FSBL_IMAGE_NAME to align with expected file name to be
fsbl-${MACHINE}.
Signed-off-by: Sandeep Gundlupet Raju <sandeep.gundlupet-raju@xilinx.com>
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| | |
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
device-tree, fsbl, plmfw, pmufw, psmfw do basic validation checks, when in
an eSDK skip these checks. Otherwise a partially configured system can't
generate an eSDK for the end user.
Sync Core and Standalone usages to be consistent.
Also cleanup the BB_DONT_CACHE to be consistent in all usages.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
fsbl, plmfw, pufw, and psmfw are reworked to provide earlier error messages,
and better handle multiconfig builds.
Also adjust MACHINE specific changes to use .* for compatibility based on the
SOC_FAMILY override. The system doesn't always set the SOC_FAMILY as
COMPATIBLE_MACHINE, so this avoids an error of incompatible machine.
Align all of the expected firmware filenames to be <firmware>-${MACHINE}.
The user can override as necessary.
The license moves to CLOSED as we don't know the license of the software being
provided by the user or other dependencies. It'll be up to the user to
reconcile it.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|/
|
|
|
|
|
| |
PACKAGE_ARCH should not be MACHINE_ARCH, unless a specific override is
being used for compatibility.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
|
|
|
| |
This is the result of automated script (0.9.0) conversion:
oe-core/scripts/contrib/convert-overrides.py .
converting the metadata to use ":" as the override character instead of "_".
Signed-off-by: Sai Hari Chandana Kalluri <chandana.kalluri@xilinx.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The latest embeddedsw components no longer need integration workarounds,
other then the couple of patches being applied. This simplifies eliminates
the custom do_configure, and simplifies the do_compile.
EXTRA_COMPILER_FLAGS is now different then the default version.
The versal defaults were also adjusted to match the expected output.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
|
| |
Different boards may configure their firmwares differently. One such
example is ZU67DR RFSoC which requires libmetal libraries in some
configurtions.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
| |
The fsbl has been renamed to fsbl-firmware, and some adjustments to the
build method are required to match how standalone and xsct version work.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
| |
Signed-off-by: Sai Hari Chandana Kalluri <chandana.kalluri@xilinx.com>
|
|
|
|
| |
Signed-off-by: Sai Hari Chandana Kalluri <chandana.kalluri@xilinx.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Issue: CR-1087560
git version no longer requires most patches, so clear these.
bin generation was not working consistently for standalone (not xsct) build,
correct this behavior.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
| |
WHen processing for cmake, PARLLEL_MAKE is processed by a python chunk
and if empty returns 'None', instead of 1.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
|
| |
Issue: ER-6946
Different workflows may require a slightly different name.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The various *fw.bb Linux recipes need to exclude the PLM_DEPLOY_DIR from hash
calculations. This location can change, but the actual value does not
matter.
Move plm-firmware, psm-firmware, pmu-firmware to use IMAGE_VERSION_SUFFIX
(and the associate image-artifact-names class) by default. This matches
how modern deployed components should work.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Following the example of the psm, pmu and plm, we split the FSBL into both
a packaging component (fsbl) and the build of the firmware (fsbl-firmware).
This also adds an additional multilib, fsbl-fw, that is used to handle the
multiconfig version, as the Linux and baremetal builds are done differently.
Typical build approach is:
Add to local.conf:
BBMULTICONFIG += "fsbl-fw"
then build using:
MACHINE=zynqmp-generic bitbake fsbl
Note, while 'zynq' is implemented, it does not currently function. To
build the fsbl for zynq, you must use the meta-xilinx-tools version.
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|
|
|
|
| |
Signed-off-by: Mark Hatle <mark.hatle@xilinx.com>
|