gitlint: check for bad authors

Similar to how we expect Signed-off-by to be legal names matching
something from a CLA, it is helpful if the Author is formatted
similarly. Often, a new user especial

gitlint: check for bad authors

Similar to how we expect Signed-off-by to be legal names matching
something from a CLA, it is helpful if the Author is formatted
similarly. Often, a new user especially, we fix their git-config and
the Signed-off-by but leave the author as a username-like. This leads
to the SRCREV bump script creating a commit message referencing a
[likely non-public] username instead of the proper name of the author.

Tested:
```
Running commit_gitlint (0.19.1)
-: UC4 Author user has too few words; likely user id instead of legal name?: "joe-bob"
```

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I6a3da5b27cd57de5758ba9277ae5d8b2e0570a5e

show more ...

Prepare for building meta machines on jenkins

We are preparing to build anacapa, bletchley15, santabarbara, ventura2,
and yosemite5 on jenkins. This ensures that when those new jobs start
running on

Prepare for building meta machines on jenkins

We are preparing to build anacapa, bletchley15, santabarbara, ventura2,
and yosemite5 on jenkins. This ensures that when those new jobs start
running on the jenkins workers, they will be fast since the sstate
will have data for each of the new machines.

Signed-off-by: James Athappilly <jamesatha@gmail.com>
Depends-on: Iae149f1da8822f48783199b51858a07fdaa476e2
Change-Id: Id6bb685014ac7d4e4978f11717462e216b10c7f2

show more ...

Invalidate docker cache to fix clang upgrade error

This should address build errors from running
```
/bin/sh -c apt-get update && apt-get dist-upgrade -yy
```

that resulted in the following error:

Invalidate docker cache to fix clang upgrade error

This should address build errors from running
```
/bin/sh -c apt-get update && apt-get dist-upgrade -yy
```

that resulted in the following error:
```
dpkg: error processing archive /tmp/...60_arm64.deb (--unpack):
trying to overwrite '/usr/lib/llvm-21/bin/clang-scan-deps'
```

This forces docker to actually do an apt-get update/upgrade instead of
using the cached image on the host

Change-Id: I99aa148a9af51a3375ba813e93621b7dbd302680
Signed-off-by: James Athappilly <jamesatha@gmail.com>

show more ...

scripts: unit-test: Reduce NOFILE rlimit for valgrind

Avoid triggering its file descriptor limit test and failure exit:

```
--27:0:libcfile Valgrind: FATAL: Private file creation failed.
The cur

scripts: unit-test: Reduce NOFILE rlimit for valgrind

Avoid triggering its file descriptor limit test and failure exit:

```
--27:0:libcfile Valgrind: FATAL: Private file creation failed.
The current file descriptor limit is 1073741804.
If you are running in Docker please consider
lowering this limit with the shell built-in limit command.
--27:0:libcfile Exiting now.
```

Change-Id: I94b051c9843209fb0986575622735691f0320a13
Signed-off-by: Andrew Jeffery <andrew@codeconstruct.com.au>

show more ...

Fix valgrind safety check

Cleanup the code to not catch all errors, only errors when the valgrind
run is done. If the build, or anything else fails, that should be
promoted to a real exception. Pr

Fix valgrind safety check

Cleanup the code to not catch all errors, only errors when the valgrind
run is done. If the build, or anything else fails, that should be
promoted to a real exception. Previously, any error in the c file would
silently be treated as "valgrind not supported".

While we're here, clean up some other things. Use TemporaryDirectory as
a context rather than doing manual cleanup. Use the python DEVNULL
rather than constructing our own null pipe.

Change-Id: I0749b10cb3de579db007927fb0e3a58e815b5e22
Signed-off-by: Ed Tanous <etanous@nvidia.com>

show more ...

beautysh: fix to v6.2.1

The latest version of beautysh has a bug[1] that breaks formatting.
Fix to an older version until this is fixed.

[1]: https://github.com/lovesegfault/beautysh/issues/265

Si

beautysh: fix to v6.2.1

The latest version of beautysh has a bug[1] that breaks formatting.
Fix to an older version until this is fixed.

[1]: https://github.com/lovesegfault/beautysh/issues/265

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I6371098f68a2f61c0a901a1b527ac93d0e043044

show more ...

build-unit-test-docker: use ubuntu-toolchain-r for gcc

Ubuntu plucky uses GCC-15.0, which has a few known bugs. Switch
to picking up GCC from the ubuntu-toolchain-r repository, which
currently has

build-unit-test-docker: use ubuntu-toolchain-r for gcc

Ubuntu plucky uses GCC-15.0, which has a few known bugs. Switch
to picking up GCC from the ubuntu-toolchain-r repository, which
currently has GCC-15.2.

One specific issue the community has encountered included with
this fixes is reported by Google Test[1].

[1]: https://github.com/google/googletest/issues/4759

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: Id7f9cfd9c6292334b79c6c417c11210910876f06

show more ...

build-unit-test-docker: utilize https for llvm install

Some of our internal networks have started to completely block http.
This llvm.sh script utilizes a http address by default for some wget
comma

build-unit-test-docker: utilize https for llvm install

Some of our internal networks have started to completely block http.
This llvm.sh script utilizes a http address by default for some wget
commands. An issue[1] has been opened on the llvm project but for now
just utilize a workaround to tell the script to use https.

Tested:
- Confirmed script now correctly builds in environment without http
support

[1]: https://github.com/llvm/llvm-project/issues/163216

Change-Id: Iadcb05e94480c3a91dc65c33e41f274ac545fd99
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>

show more ...

qemu-build.sh: Allow WORKSPACE to be anywhere while building

When getting latest-qemu-aarch64 working, I encountered an error
saying the script which should have been mounted wasn't there [1].
Upon

qemu-build.sh: Allow WORKSPACE to be anywhere while building

When getting latest-qemu-aarch64 working, I encountered an error
saying the script which should have been mounted wasn't there [1].
Upon investigating, I realized that this works only if WORKSPACE
is a child of HOME which does get bind mounted. When building on
the meta machines, the main writable mount is not a users home
directory.

This change makes it so that we just bind WORKSPACE
directly and it simplifies things by remapping the docker root
user to the caller's UID/GID so permissions line up.

I was able to manually run this on the meta graviton host and
verified that it made the build artifacts with the correct
username and group.

[1] https://jenkins.openbmc.org/job/latest-qemu-aarch64/1/console

Change-Id: Ief4fb34c3f160767122c0129c447114a9e86692a
Signed-off-by: James Athappilly <jamesatha@gmail.com>

show more ...

userid-validation: add scaleway

CLA update has been processed and there is an active user in gerrit so
add to approve list

Change-Id: I5d05624bf6e623d30c6e6a52ce7c01051f9c5cd4
Signed-off-by: Andrew

userid-validation: add scaleway

CLA update has been processed and there is an active user in gerrit so
add to approve list

Change-Id: I5d05624bf6e623d30c6e6a52ce7c01051f9c5cd4
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>

show more ...

build-setup.sh: fix bitbake warnings

Resolved the warning messages encountered during the docker build

WARNING: local.conf:263 has a lack of whitespace around the
assignment: 'DL_DIR="/data0/jenkin

build-setup.sh: fix bitbake warnings

Resolved the warning messages encountered during the docker build

WARNING: local.conf:263 has a lack of whitespace around the
assignment: 'DL_DIR="/data0/jenkins/bitbake_downloads"'

WARNING: local.conf:264 has a lack of whitespace around the
assignment: 'SSTATE_DIR="/data0/jenkins/bitbake_sharedstatecache"'

WARNING: local.conf:267 has a lack of whitespace around
the assignment: 'TMPDIR="/data0/jenkins/workspace/latest-master/
label/docker-builder/target/bletchley/build"'

Change-Id: I1596bc0251f3d5b06c24069377313320c973eb7b
Signed-off-by: Hariharan Rangasamy <hariharanr@ami.com>

show more ...

build-unit-test-docker: update-alternatives for clang-apply-replacements

Meson expects a `clang-apply-replacements` implementation in order to
generate the `clang-tidy-fix` target[1,2]. With clang-

build-unit-test-docker: update-alternatives for clang-apply-replacements

Meson expects a `clang-apply-replacements` implementation in order to
generate the `clang-tidy-fix` target[1,2]. With clang-21, it seems
that this moved to a versioned implementation, so we need to include it
in the `update-alternatives` invocation.

[1]: https://github.com/mesonbuild/meson/blob/3708730dee5049eca7e497629d25629534ea4964/mesonbuild/backend/ninjabackend.py#L3907
[2]: https://github.com/mesonbuild/meson/blob/3708730dee5049eca7e497629d25629534ea4964/mesonbuild/environment.py#L320

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I7d3099ac7cd675755131a1385f4ce0fc15dc55f7

show more ...

build-unit-test-docker: clang: update to 21

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I795e88ada8e676c242b5a18888ce9c08afdedc93

boot-qemu: Use hostname -I to obtian container IP

The code was using `awk 'END{print $1}' /etc/hosts` to obtain the
container IP, it works OK for containers with IPv4-only address, but
does not work

boot-qemu: Use hostname -I to obtian container IP

The code was using `awk 'END{print $1}' /etc/hosts` to obtain the
container IP, it works OK for containers with IPv4-only address, but
does not work for containers with IPv6 address, because the last line of
`/etc/hosts` is IPv6 address in such containers. Then the script will
use `127.0.0.1` as the IP address to start the QEMU, which will not work
for QEMU CI because the host can not access the QEMU's ports.

Fix this by using `hostname -I` to obtain the IPv4 address.

Tested: Verify QEMU CI passes in containers with IPv6 address.

Change-Id: I3ce39b5633a913c303ccc8297cdf39a87f5be99a
Signed-off-by: Lei YU <yulei.sh@bytedance.com>

show more ...

build-qemu-robot-docker.sh: update to plucky

The previous version has gone EOL

Change-Id: Ic9522129e17a6359a49b7720f10a7b5229163c3e
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>

build-seed: add gb200nvl-obmc and remove palmetto

Nvidia has provided a build server so add a system per their request.
Palmetto was removed as it is no longer supported by IBM.

Change-Id: Ia03affe

build-seed: add gb200nvl-obmc and remove palmetto

Nvidia has provided a build server so add a system per their request.
Palmetto was removed as it is no longer supported by IBM.

Change-Id: Ia03affe3b6f879822552b1ce3c841e3c666a0f33
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>

show more ...

build-unit-test-docker: escape topic queries

Gerrit topics can contain spaces[1]. We need to escape those so
that the http request can be done successfully.

[1]: https://gerrit.openbmc.org/c/openb

build-unit-test-docker: escape topic queries

Gerrit topics can contain spaces[1]. We need to escape those so
that the http request can be done successfully.

[1]: https://gerrit.openbmc.org/c/openbmc/entity-manager/+/83520

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I934a012497eb4e33ee606f133f2f705b8e5fcd10

show more ...

build-unit-test-docker: sdbusplus: use pip for install

Currently sdbusplus uses a setup.py, which is a deprecated method
for installing python packages. If we use `pip` here, we can allow
it to det

build-unit-test-docker: sdbusplus: use pip for install

Currently sdbusplus uses a setup.py, which is a deprecated method
for installing python packages. If we use `pip` here, we can allow
it to detect how sdbusplus specifies its installation directives,
which will allow transition from the older setup.py to newer
pyproject.toml method.

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: Ie956fc32972f4e0ef225a1f8c04d98642098b6a2

show more ...

config-clang-tidy: support dropping config options

The latest docs .clang-tidy reference dropped some config options,
so add support for that to the tool.

Signed-off-by: Patrick Williams <patrick@s

config-clang-tidy: support dropping config options

The latest docs .clang-tidy reference dropped some config options,
so add support for that to the tool.

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I4b9c33f9d01697d06574f931aa2c8a782cafc262

show more ...

Upgrade to meson 1.9.0

Meson 1.9.0 includes a new feature to avoid pulling in all the headers
when running clang-tidy. While 1.9.0 isn't the version yocto uses, this
is an important change to decre

Upgrade to meson 1.9.0

Meson 1.9.0 includes a new feature to avoid pulling in all the headers
when running clang-tidy. While 1.9.0 isn't the version yocto uses, this
is an important change to decrease our CI times.

[1] https://mesonbuild.com/Release-notes-for-1-9-0.html#clangtidys-autogenerated-targets-correctly-select-source-files

Change-Id: I2cd844d79f751443f9400cb185840128c2d9f2a9
Signed-off-by: Ed Tanous <etanous@nvidia.com>

show more ...

build-seed: add evb-npcm845 and remove stale systems

Change-Id: Ic81415b670f37748696dda48a80a57425323a9fd
Signed-off-by: Andrew Geissler <geissonator@yahoo.com>

Add DMTF/libspdm v3.7.0 to docker unit test image

Add libspdm library to the docker build environment
to support SPDM.
The library is built with OpenSSL crypto backend
and configured for shared libr

Add DMTF/libspdm v3.7.0 to docker unit test image

Add libspdm library to the docker build environment
to support SPDM.
The library is built with OpenSSL crypto backend
and configured for shared library output.

Tested:
- Verified that local CI passes without errors.
- Confirmed all SPDM libraries and associated header
files are correctly built and included.

Change-Id: I04c97ae616ada2bc955ac1353ce743dd8ea3ba8a
Signed-off-by: Ratan Gupta <ratankgupta31@gmail.com>

show more ...

Make clang environment temporary

Pushing to the global environment has the unintended effect of running
sanitizers in clang. Moving to gcc-15 had some unintended consequences
where that broke in so

Make clang environment temporary

Pushing to the global environment has the unintended effect of running
sanitizers in clang. Moving to gcc-15 had some unintended consequences
where that broke in some cases with an error of:

ld.lld: error: undefined symbol: __asan_stack_malloc_1

Move the clang variables to a dict and pass it into the appropriate
calls so that it can be destroyed afterward.

Change-Id: I8d4e2b50156e285d7a569dfa1aa8831828c5f480
Signed-off-by: Ed Tanous <etanous@nvidia.com>

show more ...

Set CXXFLAGS for clang

CXXFLAGS is more arguably correct. Fix it.

Change-Id: I0cd9bacfd4e6453c9e36879581aadde9792e31bd
Signed-off-by: Ed Tanous <etanous@nvidia.com>

build-unit-test-docker: force libgpiod 1.x series

Latest ubuntu switched to libgpiod 2.x, which we do not support
yet in OpenBMC. Convert from using a deb to manually building the
1.x package.

Tes

build-unit-test-docker: force libgpiod 1.x series

Latest ubuntu switched to libgpiod 2.x, which we do not support
yet in OpenBMC. Convert from using a deb to manually building the
1.x package.

Tested:

Ran against entity-manager and confirmed that subproject build was
no longer selected.

```
Run-time dependency libgpiodcxx found: YES 1.6.5
```

Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Change-Id: I556574c3cf728932e90f98286871a365d92c91f9

show more ...