The dev kernel can frequently fail, and is not anything that is used in
production. Allow failure to prevent CI issues but still notify that
there are potential issues.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Mirrors of meta-arm may focus their development on a small subset of
MACHINEs so provide the option to restrict the boards that are built on
CI using the variable BUILD_ENABLE_REGEX. If set, it conditionally
enables builds; if unset there is no change in behavior.
This variable could be overridden in a scheduled build, to e.g. build
all the MACHINEs weekly.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Installing SMM Gateway SP on the N1SDP may stop the platform from booting
for on-device testing in CI.
In n1sdp.yml, remove `ts-smm-gateway` if it has been added
Keep `ts-smm-gateway` in default SP set so it can still be tested with
`qemu-secureboot`
Signed-off-by: Adam Johnston <adam.johnston@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Apply the patch from scp-firmware to the third copy of the buggy
Makefiles which fail randomly under parallel builds.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The UEFI capsule generated is in the incorrect build directory.
This patch copies it to IMGDEPLOYDIR.
Signed-off-by: Gowtham Suresh Kumar <gowtham.sureshkumar@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
This patch uses the json config file for UEFI capsule generation
as this is efficient and easily scalable to generate multiple
capsules.
Signed-off-by: Gowtham Suresh Kumar <gowtham.sureshkumar@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The BBCLASSEXTEND configuration can generate native sdk and target
recipes as well. The cp command used in do_install will
create host contamination issues for these recipes, so this patch
makes the recipe native only.
Signed-off-by: Gowtham Suresh Kumar <gowtham.sureshkumar@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Inherits the UEFI capsule generation class and configures the capsule
variables for the wic.nopt image
Signed-off-by: Gowtham Suresh Kumar <gowtham.sureshkumar@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
This class currently supports only a single firmware binary. The
required capsule parameters needs to be set, if not the build fails.
Signed-off-by: Gowtham Suresh Kumar <gowtham.sureshkumar@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The native recipe installs the UEFI capsule generation tool
along with the other base tools to native sysroot.
Signed-off-by: Gowtham Suresh Kumar <gowtham.sureshkumar@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
TFM repos do not use release branches. TFM-test repo does not use master
branch and use detached head for the 1.6 release. This commits makes necessary
changes to be able to build tfm and tfm-test again for langdale.
Signed-off-by: Emekcan Aras <emekcan.aras@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
For some reason the kas 3.2.1 container fails:
No such file or directory: '/builds/engineering/yocto/meta-arm/ci/ci/base.yml'
Note the repeated /ci/, which is wrong.
Pin the kas container to 3.2 for now until this is resolved.
Signed-off-by: Ross Burton <ross.burton@arm.com>
A recent commit compressed the kernel image (to Image.gz) and
by default enabled an initramfs image. In the case for when
such that (initramfs) is not desirable, the deploy step of the
Juno firmware will still try to install the Image file, (not
Image.gz), so this fails:
ERROR: firmware-image-juno-1.0-r0 do_deploy: ExecutionError('/oe/build/tmp-glibc/work/juno-oe-linux/firmware-image-juno/1.0-r0/temp/run.do_deploy.360477', 1, None, None)
ERROR: Logfile of failure stored in: /oe/build/tmp-glibc/work/juno-oe-linux/firmware-image-juno/1.0-r0/temp/log.do_deploy.360477
Log data follows:
| DEBUG: Executing python function sstate_task_prefunc
| DEBUG: Python function sstate_task_prefunc finished
| DEBUG: Executing shell function do_deploy
| cp: cannot stat '/oe/build/tmp-glibc/deploy/images/juno/Image': No such file or directory
| WARNING: /oe/build/tmp-glibc/work/juno-oe-linux/firmware-image-juno/1.0-r0/temp/run.do_deploy.360477:152 exit 1 from 'cp -L -f /oe/build/tmp-glibc/deploy/images/juno/Image /oe/build/tmp-glibc/work/juno-oe-linux/firmware-image-juno/1.0-r0/image/juno-firmware-19.06/SOFTWARE/'
| WARNING: Backtrace (BB generated script):
| #1: do_deploy, /oe/build/tmp-glibc/work/juno-oe-linux/firmware-image-juno/1.0-r0/temp/run.do_deploy.360477, line 152
| #2: main, /oe/build/tmp-glibc/work/juno-oe-linux/firmware-image-juno/1.0-r0/temp/run.do_deploy.360477, line 163
NOTE: recipe firmware-image-juno-1.0-r0: task do_deploy: Failed
ERROR: Task (../meta-arm/meta-arm-bsp/recipes-bsp/images/firmware-image-juno.bb:do_deploy) failed with exit code '1'
This updates the else case for when an initramfs image is not
in use so that the right kernel image is deployed, by using
the KERNEL_IMAGETYPE variable, to use either version of the
kernel image.
Signed-off-by: Daniel Díaz <daniel.diaz@linaro.org>
Signed-off-by: Jon Mason <jon.mason@arm.com>
For models that require a license, ARMLMD_LICENSE_FILE is used to define
the location of a license file or server. If the variable is not set in
Bitbake it will not be set in the model environment.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
FVP_ENV_PASSTHROUGH may contain variables that have not been set.
d.getVar returns None in this case. Detect this and skip setting the
variable in the model environment.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Define FVP_ENV_PASSTHROUGH's vardeps to equal itself, so that the
fvpconf is regenerated if any of the defined variables change.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
aa89fe3f ensured environment variables necessary for GUI applications
are passed through to the model despite runfvp env var restrictions. Add
XAUTHORITY to this list. This is useful when doing X-forwarding with
Kas, which creates its own home directory.
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Change u-boot and machine config to default to booting a compressed
initramfs. This allows for easier testing. A compressed image is
needed as the image is too big for the storage, and the error notifying
of such is vague.
Signed-off-by: Jon Mason <jon.mason@arm.com>
The libts recipe assumes generated cmake file will be suffixed with
'-noconfig'. This is only true when building with the default type
i.e. "".
Check which target cmake file has been generated before trying to
patch it. This fixes 'no such file' error when building with an
explicit type (Debug, Release, etc).
Signed-off-by: Adam Johnston <adam.johnston@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The SCP-firmware CMake compile step automatically attempts to execute
cppcheck if both:
* cppcheck can be located using find_program
* DISABLE_CPPCHECK is not defined
cppcheck is not readily available in OE-core and is not an essential
part of the compilation process for end-users, so explicitly disable the
cppcheck step by passing DISABLE_CPPCHECK to CMake.
Additionally, because the OE-core CMake toolchain file cannot be used,
find_program may locate cppcheck on the host machine, which will cause
the build to fail if it is not the recommended version (as it is in
recent Linux distros).
Issue-Id: SCM-5864
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: Ia87a0cbbb67ac1d6f3b26cfb5747a85b46131f81
Signed-off-by: Jon Mason <jon.mason@arm.com>
If CMAKE_BUILD_TYPE=debug (lowercase), SCP-firmware builds with debug
compiler flags but BUILD_MODE_DEBUG is not defined in C code so features
that are conditionally enabled/disabled in debug mode are not active.
Pass capitalized "Debug" and "Release" strings to CMAKE_BUILD_TYPE to
ensure Debug mode is fully enabled when SCP_BUILD_RELEASE = "0".
SCP_BUILD_RELEASE = "1" (the default) for all machines in meta-arm-bsp
so they are unaffected.
Issue-Id: SCM-5864
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I93220420eedd2e3e6c169679efcaf4642dd5bc51
Signed-off-by: Jon Mason <jon.mason@arm.com>
apply_local_src_patches.bbclass was added in a previous patch to handle
the application of patch files located inside the fetched source code.
find is used to collect the patch files which does not guarantee the
order of its output. Pipe the output of find into sort to ensure patch
files are applied in the correct order.
Issue-Id: SCM-5864
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I1082fb7a726a7745289a5aa8bb6447bef57a94b0
Signed-off-by: Jon Mason <jon.mason@arm.com>
meta-clang has a langdale branch now, so use that instead of master.
linux-yocto needs to use non-clang objcopy, apply the change locally
until the commit has been merged into meta-clang's langdale branch.
perf needs some patches backported to langdale, until that has been done
use gcc to build perf.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Timing Annotation is a feature of the model that enables high-level
performance estimations to be made [1]. It is not needed to demonstrate
a functioning software stack so set FASTSIM_DISABLE_TA to 1 in the model
environment to disable this feature. This also improves model
performance.
[1] https://developer.arm.com/documentation/100965/1119/Timing-Annotation
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
DEFAULT_TAG and CPU_REQUEST are being used to help with internal Gitlab
pipeline setups know which type of machines to run on, but has no value
outside of Arm Corp. Gitlab CI allows for variables to be overridden
by default. So, we can give it a default value of NULL/empty and have
everything work internally and externally by default.
Signed-off-by: Jon Mason <jon.mason@arm.com>
Tag all jobs with the DEFAULT_TAG variable so each instance can control
what tags the jobs have, whilst still explicitly tagging the jobs which
need specific tags (such as x86_64 for jobs which need to run x86-only
binaries)
Signed-off-by: Ross Burton <ross.burton@arm.com>
The Kas container needs to use the entrypoint as that is where the user
changes from root to a normal user.
Also set the KUBERNETES_CPU_REQUEST to the variable CPU_REQUEST as this
needs to be tuned per-deployment.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Update version in documentation.
Issue-Id: SCM-4874
Signed-off-by: Robbie Cao <robbie.cao@arm.com>
Change-Id: Ic66bdcdc5c6309331f80faab6eaf2e3e936a5da4
Signed-off-by: Jon Mason <jon.mason@arm.com>
Ensure meta-atp recipes are only performed if a compatible machine is
selected.
Signed-off-by: Adrián Herrera Arcila <adrian.herrera@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Previously, meta-atp extended the original gem5 recipe to add the ATP
Engine models; the .bbappend was complex, because it pulled from two
sources, and it was not possible to make machine-based overrides, as it
is a native recipe.
To solve this, we use the recent EXTRAS feature to add the gem5 models
from a different recipe.
Signed-off-by: Adrián Herrera Arcila <adrian.herrera@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
The m5 readfile recipe provides general utility for gem5 users to
run any script on OS boot. We hence move it to the meta-gem5 layer.
Signed-off-by: Adrian Herrera <adrian.herrera@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
By making m5readfile into its own recipe, we avoid modifications to the
m5ops recipe when using the meta-atp layer, which break the Yocto
compatibility of the layer.
Signed-off-by: Adrian Herrera <adrian.herrera@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
Test failed because there was no reference to maintainers in the
meta-atp README.
Following the common structure of other layers in the meta-arm
repository, the README in meta-atp now refers to the top-level README,
and a documentation directory contains the guidance that was present in
the original meta-atp README.
Signed-off-by: Adrián Herrera Arcila <adrian.herrera@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
N1SDP has 2 DRAM's. This change is to register 2nd DRAM which starts at
0x8080000000. Linux uses 1KB of this memory to share data with optee-os.
Signed-off-by: Jon Mason <jon.mason@arm.com>
721bec25 "arm/fvp: Join cli arguments in verbose logging" changed the
verbose output of FVPRunner to print the generated arguments using
shlex.join instead of as a list. However, this function is only
available in Python >= 3.8, whereas OE-core currently supports Python
3.6.
To fix this, backport its one-line implementation to a local function
shlex_join and update the call site.
Issue-Id: SCM-5314
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I56cab9dddcd0a91272464be15742a6ee726dad41
Signed-off-by: Jon Mason <jon.mason@arm.com>
As TF-M ships patches that it needs applied to mbedcrypto, we apply them as
part of do_patch by using a postfunc. There is an issue when do_patch is
executed after do_deploy_source_date_epoch_setscene and apply_local_patches
tries to apply the patches already applied.
To fix this, make usage of the apply_local_src_patches bbclass.
Change-Id: Ia115b540b37ad3a2cce30e1e0461abd1f5a6ccc1
Signed-off-by: Diego Sueiro <diego.sueiro@arm.com>
Signed-off-by: Jon Mason <jon.mason@arm.com>
This class is to be inherited by recipes where there are patches located inside
the fetched source code which need to be applied.
The following variables need to be set:
LOCAL_SRC_PATCHES_INPUT_DIR is the directory from where the patches are located
LOCAL_SRC_PATCHES_DEST_DIR is the directory where the patches will be applied
Signed-off-by: Diego Sueiro <diego.sueiro@arm.com>
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I8f9c16a5fbc9d5569cba60136560f1951408bd60
Signed-off-by: Jon Mason <jon.mason@arm.com>
Start a new thread to simultaneously log the output of FVP and the
telnet output if the --verbose flag is passed to runfvp. So that
ConsolePortParser can read the same stream, use itertools.tee to
temporarily duplicate the stream.
Use a custom log format string with an escape character to ensure that
log output always starts at the beginning of a line when interleaved
with console output.
Issue-Id: SCM-5314
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I3e815d9d899425e0d2af619524f09f2eda87562c
Signed-off-by: Jon Mason <jon.mason@arm.com>
It is often useful to inspect the FVP output after running the tests.
Refactor OEFVPSerialTarget._create_logfile into
OEFVPSSHTarget._create_log_filename, so that all FVP test controllers
are able to create secondary log files.
Pass a filehandle to the stdout argument of run_fvp so that the FVP
output is asynchronously logged to a file. Change the port parsing logic
to read back from the same log file instead of reading the stdout
directly.
Issue-Id: SCM-5314
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I2ddb423fa0d896c13d3e96884858c680c4d34555
Signed-off-by: Jon Mason <jon.mason@arm.com>
To simplify the FVPRunner class, create a separate ConsolePortParser
class to handle reading an iterator of lines and parsing port numbers
for FVP consoles. Use this in runfvp and the test targets.
This refactor also allows the stream being monitored to be changed more
easily, e.g. to a log file.
Issue-Id: SCM-5314
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: Iade3a4c803fb355b04af7afa298d0a41fe707d94
Signed-off-by: Jon Mason <jon.mason@arm.com>
FVPRunner relies heavily on asyncio, despite there being very little
concurrent work happening. Additionally, while the runfvp entry point
starts an asyncio runner, it is not practical to have a single asyncio
runtime during testimage, which is fully synchronous.
Refactor to use subprocess.Popen and related functionality. The process
object has a similar interface to its async equivalent.
Cascade the API changes to runfvp and the test target classes.
Issue-Id: SCM-5314
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Change-Id: I3e7517e8bcbb3b93c41405d43dbd8bd24a9e7eb8
Signed-off-by: Jon Mason <jon.mason@arm.com>
