run-ptest ran each case as "./test.sh $test > /dev/null 2>&1", throwing
away the only thing that explains a failure: test.sh prints which check
failed and which words were misrecognised, e.g.
Fail in base.good. Good words recognised as wrong:
<words>
Without it a failing hunspell ptest reports a bare "FAIL: <name>" and
gives no way to tell a packaging problem from a real defect when the
suite runs on target.
Capture the output and print it, indented, under the FAIL line, and
take test.sh's exit status directly rather than reading $? inside the
else branch of the if that consumed it.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
All 25 ptest failures are missing glibc character-set converters. With
run-ptest no longer discarding test.sh output, the cause is explicit:
error - iconv_open: ISO8859-1 -> UTF-8
error - iconv_open: UTF-8 -> ISO8859-1
=============================================
Fail in allcaps3.good. Good words recognised as wrong:
Aggregated over the run: 382 ISO8859-1 and 10 ISO8859-15 iconv_open
failures, and every one of the 25 failing tests reports them.
Only glibc-gconv-iso8859-2 was pulled in, but the test corpus needs more
than that. Dictionaries with no SET line in their .aff fall back to
ISO8859-1 in both consumers of the default:
csutil.hxx:99 #define SPELL_ENCODING "ISO8859-1"
affixmgr.cxx:3502 if (encoding.empty()) encoding = SPELL_ENCODING;
hashmgr.cxx:97 if (!csconv) csconv = get_current_cs(SPELL_ENCODING);
and 92 of the 144 installed .aff files have no SET, which is why tests
with pure-ASCII data and no SET line still needed a converter. The
encodings actually referenced across the corpus are ISO8859-1,
ISO8859-15, ISO-8859-15, ISO8859-2 and UTF-8; UTF-8 is built into glibc,
so add the two missing ISO8859 modules.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
tests/test-proxy.c setup() spawns the reference bus implementation
directly:
g_subprocess_launcher_spawn (launcher, &error, "dbus-daemon",
"--session", "--print-address=1",
"--nofork", NULL);
so the test aborts on distros that select a different
VIRTUAL-RUNTIME_dbus:
ERROR:../tests/test-proxy.c:75:setup: assertion failed (error == NULL):
Failed to execute child process "dbus-daemon" (No such file or directory)
FAIL: xdg-dbus-proxy/test-proxy.test (Child process killed by signal 6)
dbus-daemon is packaged only in dbus, and dbus-broker carries
"RCONFLICTS:dbus-broker: dbus", so the two cannot be co-installed;
dbus-broker-launch is not CLI-compatible and cannot stand in. The
existing RDEPENDS on ${VIRTUAL-RUNTIME_dbus} already pulls in dbus (and
hence dbus-daemon) on distros that use it, so report the test as skipped
rather than failed when the daemon genuinely is not installable.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
TestCommandLineToCommaSeparatedValues fails because its JSON test cases
are never installed.
With no arguments the test runs runJsonTests(), which globs
TEST_SOURCE_DIR/cmdline-tests/*.json via FindFirstFileA and returns -1
when the directory cannot be opened. do_configure:prepend already
rewrites TEST_SOURCE_DIR for this CMakeLists to ${PTEST_PATH}/test, and
the built driver confirms it:
$ strings Testing/TestWinPRUtils | grep -e cmdline-tests -e ptest/test
cmdline-tests
cmdline-tests%c*.json
/usr/lib/freerdp3/ptest/test
but do_install_ptest only copies *bmp out of
winpr/libwinpr/utils/test, so /usr/lib/freerdp3/ptest/test/cmdline-tests
does not exist on target and the glob always fails. Copy the directory
alongside the other test data.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
All 6 ptest failures (t/31methcache, t/35thrclone, t/40profile,
t/41prof_dump, t/42prof_data, t/43prof_env) are missing runtime perl
modules, in two groups:
Module::Load, required by DBI/Profile.pm line 682:
Can't locate Module/Load.pm in @INC ... at
/usr/lib/perl5/vendor_perl/5.44.0/aarch64-linux/DBI/Profile.pm line 682.
This breaks t/40profile and t/43prof_env directly, and cascades into
t/41prof_dump and t/42prof_data: once DBI/Profile.pm fails to compile,
%INC is poisoned, so those two report "Attempt to reload DBI/Profile.pm
aborted" and a missing DBI::Profile->flush_to_disk method rather than
the underlying cause. Provided by perl-module-load.
Test2::Util::Sig, required by Test2/IPC/Driver/Files.pm line 18, which
Test::Builder pulls in when a test enables Test2 IPC (threads/forking):
Can't locate Test2/Util/Sig.pm in @INC ... at
/usr/lib/perl5/5.44.0/Test2/IPC/Driver/Files.pm line 18.
This makes Test::More unusable in t/31methcache and t/35thrclone, which
is why only those two of the many Test::More-based tests fail. Provided
by perl-module-test2-util-sig.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
14 of the 257 pegtl tests failed - every test that reads a data file
(input_file_input, input_mmap_input, internal_file_mapper,
internal_read_file_stdio, stream_*_input, example_json, ...) while the
243 tests that parse in-memory strings passed.
Upstream runs its tests from the source root:
src/test/CMakeLists.txt:299
add_test(NAME ${exename} WORKING_DIRECTORY ${CMAKE_SOURCE_DIR} ...)
so the tests open their data with CWD-relative literals - grepping the
test sources shows src/test/data/duseltronik.txt (7 references),
src/test/data/test_data.txt (4) and src/test/data/{blns,pass1-3,
fail1-39}.json. run-ptest executes the binaries from ${PTEST_PATH}, so
the data has to be at ${PTEST_PATH}/src/test/data/.
The recipe instead installed it to src/test/pegtl/data - one directory
too deep - and copied only *.json, so test_data.txt and
duseltronik.txt were missing entirely. Install all of src/test/data
to the path the tests actually use. The binaries stay in
src/test/pegtl, which is where run-ptest globs for them.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
run-ptest has a #!/bin/sh shebang but used two constructs that are not
available in the target /bin/sh (dash), so the suite died immediately
after starting the server and reported no test results at all:
run-ptest: line 54: stdbuf: not found
run-ptest: line 58: syntax error: bad substitution
ERROR: Exit status is 2
${PIPESTATUS[0]} is a bash array reference; dash has no PIPESTATUS and
fails with "bad substitution". stdbuf is coreutils-only and is not
guaranteed to be installed on the target.
Capture pg_regress output to a temporary file and use $? directly
instead of piping live into sed, which removes the need for both
PIPESTATUS and stdbuf while keeping the same PASS:/FAIL: translation.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Same problem as xdg-dbus-proxy: the hardcoded "dbus" RDEPENDS makes
meta-oe-image-ptest-fwupd fail do_rootfs on a distro whose
VIRTUAL-RUNTIME_dbus is dbus-broker, since dbus-broker RCONFLICTS with
the dbus provided by dbus-1 while systemd requires dbus-broker.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Hardcoding "dbus" makes the ptest image unbuildable on any distro that
selects a different system bus implementation. On a systemd distro
setting VIRTUAL-RUNTIME_dbus = "dbus-broker", do_rootfs fails to
solve:
package dbus-broker-37 conflicts with dbus provided by dbus-1-1.16.2
package systemd requires dbus-broker, but none of the providers can
be installed
package xdg-dbus-proxy-ptest requires dbus, but none of the providers
can be installed
Depend on ${VIRTUAL-RUNTIME_dbus} instead so the ptest package pulls
in whichever bus the distro actually uses.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
ntpdate has been deprecated. meta-openembedded@6315006aadd in theory
dropped it, but in practice the binary was still being built and
shipped, but now as part of ntp-utils package.
To avoid this, let's patch it out of the build.
Signed-off-by: Michal Sieron <michal.sieron@nokia.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
The product-only "pymongo" value emits a wildcard-vendor identity and
omits the second authoritative name for the packaged MongoDB Python
driver.
Use "mongodb:python_driver" for its NVD dictionary CPE family and
"mongodb:pymongo" for its NVD dictionary CPE and configuration identity.
With sbom-cve-check 1.3.3 and the pinned database snapshots, the
generated product identity changes; the current CVE report is unchanged.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
The current product-only "m2crypto" mapping generates a wildcard-vendor
identity instead of the two vendors used for the packaged source.
Use "heikkitoivonen:m2crypto" and "m2crypto_project:m2crypto" for their
exact NVD dictionary CPE and NVD configuration identities. This changes
the generated CPE set, but sbom-cve-check 1.3.3 with the pinned NVD
snapshot has no current CVE report delta.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
The product-only "py" mapping generates a wildcard-vendor identity
instead of the exact NVD identity for the packaged pytest-dev py
source. Use "pytest:py" for its NVD dictionary CPE and configuration
matches.
This changes the generated product identity. With sbom-cve-check 1.3.3,
the current CVE report is unchanged using the pinned database snapshots;
both mappings report CVE-2020-29651 and CVE-2022-42969.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
The current inherited "python:simpleeval" mapping is wrong for the
packaged danthedeckie SimpleEval source and misses its vulnerability
record.
Use "danthedeckie:simpleeval" for the source-aligned NVD dictionary
CPE, NVD configuration identity, and CNA affected-data identity.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
pcsc-lite 2.5.0 defaults to installing systemd user units, which leaves the
package without a system-level pcscd.service while SYSTEMD_SERVICE still
enables pcscd.socket. On target systems this makes systemctl status pcscd
fail, and systemctl --user start pcscd can fail because the user unit
requires polkit.service from the user manager.
Select upstream Meson systemdunit=system so pcscd.service and pcscd.socket
are installed as system units. Keep upstream sysusers.d packaging so
systemd-sysusers can create the pcscd system user required by the
generated service and socket units.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Linux commit 8f9616500c59 ("atm: remove orphaned uAPI for deleted drivers,
protocols and SVCs") deleted atm uAPI headers and ioctls in
linux/atmdev.h, legacy ATM drivers had been dropped from the kernel.
linux-atm is the userspace side of exactly those interfaces, so it no
longer builds once linux-libc-headers is 7.2:
| ispl_l.l:15:10: fatal error: 'linux/atmsvc.h' file not found
| ispl_y.y:15:10: fatal error: 'linux/atmsvc.h' file not found
| src/test/isp.c:19:10: fatal error: 'linux/atmsvc.h' file not found
| make[3]: *** [Makefile:589: ispl_y.o] Error 1
Remove the package
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Now that KERN_MAJ reflects the target kernel instead of the build host's
uname -r, vboxvideo is actually built for kernels older than 7.x -- and it
does not compile against 6.18.44:
vbox_fb.c:336:16: error: implicit declaration of function 'drm_fb_helper_alloc_info'; did you mean 'drm_fb_helper_fill_info'? [-Wimplicit-function-declaration]
vbox_fb.c:336:14: error: assignment to 'struct fb_info *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
Linux commit 63c971af4036 ("drm/fb-helper: Allocate and release fb_info in a
single place") moved the struct fb_info allocation out of the drivers and into
the DRM core, and made drm_fb_helper_alloc_info() static; drivers now find the
instance ready to use in fb_helper->info by the time their .fbdev_probe
callback runs. vbox_fb.c guards its call with RTLNX_VER_MIN(6,19,0), which is
where the change landed in mainline.
Bumping that version check is not a fix. The commit was pulled into stable as
a Stable-dep-of and therefore appears in the *middle* of several series --
v6.6.151, v6.12.103 and v6.18.44 have all dropped the declaration that
v6.6.150, v6.12.102 and v6.18.43 still carry. No LINUX_VERSION_CODE
comparison can express that, and guessing wrong the other way is worse than a
build failure: on a 6.18.0..6.18.43 kernel fb_helper->info is still NULL when
.fbdev_probe runs, so the driver would dereference NULL at runtime.
So add an escape hatch to vbox_fb.c and let the build system decide. do_compile
greps the kernel headers we are actually compiling against and passes
-DVBOX_NO_DRM_FB_HELPER_ALLOC_INFO through KCFLAGS when the declaration is
gone. KCFLAGS rather than EXTRA_CFLAGS/VBOXMOD_CFLAGS because a command-line
assignment of the latter two would clobber the include paths they carry, while
KCFLAGS is untouched by VirtualBox's makefiles and propagates cleanly into the
nested $(MAKE) -C $(KERN_DIR). Leaving the macro undefined keeps the previous
behaviour, so nothing changes for kernels that still export the function, and
7.x is unaffected -- VBOX_VIDEO_MODULE is empty there and the in-tree
drivers/gpu/drm/vboxvideo driver is used instead.
Verified on qemux86-64 against linux-yocto 6.18.44, from cleansstate:
vboxguest.ko, vboxsf.ko and vboxvideo.ko all build with no modpost warnings,
the -D reaches the vbox_fb.o compile line, kernel-module-vboxvideo ships the
module, and `nm -u vboxvideo.ko` no longer references
drm_fb_helper_alloc_info -- all 155 remaining undefined symbols are exported
by the kernel's Module.symvers.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The inherited python:filelock mapping does not identify the tox-dev source packaged by this recipe, so filelock CVEs are missed.
Use tox-dev:filelock to match the source identity used by NVD and CNA.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Fast serialization and validation library for JSON, MessagePack, YAML and
TOML. Added as a runtime dependency of the forthcoming python3-max recipe,
but useful on its own.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
SuiteSparse adds -Wl,-rpath=$(INSTALL_LIB) while linking shared
libraries on Linux. In the OpenEmbedded build this value can resolve to
a build or install path under TMPDIR, so installed ELF files can keep an
absolute build path and fail buildpaths QA.
The packaged libraries do not need this install-tree rpath. Runtime
resolution is handled through normal package dependencies and the target
library search path. Keep the librt link and drop only the rpath entry.
Signed-off-by: Himanshu Jadon <hjadon@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
SuiteSparse 5.10.1 still carries bundled CMake entry points with
2.x minimum versions. CMake 4 rejects projects which request
compatibility with versions older than 3.5, so configure fails before
SuiteSparse can build.
Backport the upstream SuiteSparse change which raises the bundled
Mongoose, METIS and GKlib minimum version to 3.13. This version also
reaches two bundled GraphBLAS CMake entry points with the same old
minimum, so update those in the backport as well.
Signed-off-by: Himanshu Jadon <hjadon@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
samba package also installs LDB command line tools and python LDB
module. libldb package installs the same tools and pyldb module from
standalone LDB source.
pkgdata shows the conflict clearly:
samba: /usr/bin/ldbadd
libldb: /usr/bin/ldbadd
samba-python3: site-packages/ldb.so
pyldb: site-packages/ldb.so
So when both samba and libldb are added in same image, rootfs can fail
because package manager sees same files from two different packages.
Both package sets are not expected to be installed together.
Add RCONFLICTS for samba/libldb and samba-python3/pyldb. This makes the
conflict explicit during dependency resolution instead of failing later
as file install conflict.
Older meta-networking had similar conflict handling for samba/libldb in
commit: https://git.openembedded.org/meta-openembedded/commit/?id=8e641e28f630dac61a03373ad70c72a2370ad11e
This change adds the same handling again with current python3 package names.
Signed-off-by: Himanshu Jadon <hjadon@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
The current product-only "httplib2" mapping generates a wildcard-vendor
product identity instead of the vendor assigned to the packaged project.
Use "httplib2_project:httplib2" for its exact NVD dictionary CPE and NVD
configuration identity. This changes the generated CPE, but sbom-cve-check
1.3.3 with the pinned NVD database snapshot has no current CVE report delta.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
The current inherited "python:web3" mapping does not match the web3.py identities used by NVD and CVE List V5, so its source-aligned CVE is missed.
Use "ethereum:web3.py" for the CNA affected-data identity and "apeworx:web3.py" for the NVD dictionary CPE and configuration identity.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
UTscapy writes its detailed report, including the traceback of every
failing test, to a file under ${PTEST_PATH}. Under testimage that path
only exists inside the ephemeral image, which is discarded as soon as the
run finishes, so a failure left nothing behind but a bare
FAIL: 000.01s Emulate interact()
line and diagnosing it required rebuilding the image by hand.
Pass -F so the report expands only the failed tests, and cat it to stdout
when there is one, where ptest-runner records it. A clean run is
unaffected.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The "Emulate interact()" UTScapy regression test failed:
IPython requested but not found !
Using the default Python shell: History is disabled.
Traceback (most recent call last):
File "/usr/lib/python3.14/site-packages/scapy/main.py", line 857, in interact
banner += "\n"
UnboundLocalError: cannot access local variable banner where it is
not associated with a value
scapy.main.interact() only binds "banner" in the branches handling the
ipython, ptipython, ptpython and bpython shells. The test asks for
ipython, which is not in the image, so scapy falls back to the plain
Python shell and leaves "banner" unbound while still running the
unconditional "banner += \n" for the caller-supplied mybanner.
The test is correct and the bug is in scapy, which still has it in
master, so carry a patch initializing banner from banner_text before the
shell-specific chain.
meta-networking-image-ptest-python3-scapy now reports PASSED=319
FAILED=0, up from PASSED=318 FAILED=1.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
ptest.bbclass pulls ptest-runner into the -ptest package with
RRECOMMENDS:${PN}-ptest += "ptest-runner"
but the recipe then used a hard assignment for its own recommendation,
which overwrote that and left openl2tp-ptest recommending only
kernel-module-l2tp-ppp. ptest-runner was therefore never installed in
the ptest image, and meta-networking-image-ptest-openl2tp skipped the
whole run with "Test requires ptest-runner to be installed" instead of
executing any test.
Append instead of assigning, so both recommendations survive. The
openl2tp ptest now runs and passes.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The product-only cbor2 mapping uses a wildcard vendor. Use
agronholm:cbor2, the NVD identity for the packaged source.
Signed-off-by: Devansh Patel <devanshp@cisco.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Khem Raj
Michal Sieron
Devansh Patel
Wang Mingyu
Khem Raj
Zheng Ruoqin
Ankur Tyagi
Jaipaul Cheernam
Himanshu Jadon