linux/poky.git/meta/lib/oeqa, branch styhead-5.1.1

selftest/sstatetests: run CDN mirror check only once

2024-11-30T13:41:59+00:00

The first no-fail check was an attempt to work around the old CDN's instability (and it didn't really help); it should not be necessary with the new CDN, and only delays a-full completion. (From OE-Core rev: 6ffde2643e6ee8a2a0eddca173a254dbdc4a79bb) Signed-off-by: Alexander Kanavin Signed-off-by: Richard Purdie (cherry picked from commit 7f75c42b7fcf60a9ca58d3ded9047df675d76dc2) Signed-off-by: Steve Sakoman

qemurunner: Clean up serial_lock handling

2024-11-26T13:37:09+00:00

Avoid "RuntimeError: release unlocked lock" since the lock shouldn't be locked even in the error path. Add a try/finally path to ensure this. (From OE-Core rev: 0d08aab2e1c1ff1a8cb41b62b2e763bb88ca8b5a) Signed-off-by: Richard Purdie (cherry picked from commit b0732ee009ca47580d1d2ad75334f4aa50e6efd5) Signed-off-by: Steve Sakoman

oeqa/runtime/ssh: Fix incorrect timeout fix

2024-11-26T13:37:09+00:00

I'd meant to change the command timeout in the previous change, fix the correct one. (From OE-Core rev: 41ab062f0b711e30720d0e52aab79ef387ec4f28) Signed-off-by: Richard Purdie (cherry picked from commit bb991988cb23be2c8947171726ada321f27e6eed) Signed-off-by: Steve Sakoman

oeqa/runtime/ssh: Rework ssh timeout

2024-11-26T13:37:09+00:00

After the changes to improve this test, we keep seeing image testing ssh failures, particularly on mips. It looks like part of the problem is that on a loaded system, 5s is too short for mips to reliably establish an ssh connection. I've seen logs where it keeps timing out and fails to work, then the debug code successfully uses ssh later after everything else fails. Change the timings/retries to give slow platforms enough time to respond. (From OE-Core rev: 5158ceb4179ec53e396a57068714aea7d81e3f59) Signed-off-by: Richard Purdie (cherry picked from commit ba64ccf3ad6e40461219b72d60eb0fe5cb38fddd) Signed-off-by: Steve Sakoman

oeqa/selftest: Update the BB_HASHSERVE_UPSTREAM

2024-11-02T13:15:08+00:00

(From OE-Core rev: d5ba3f78f340d4627cf33eca14969d61b00f5766) Signed-off-by: Jose Quaresma Signed-off-by: Richard Purdie Signed-off-by: Steve Sakoman

oeqa/selftest/gcc: Fix kex exchange identification error

2024-10-29T12:51:03+00:00

while runnig oe-selftest for gcc, some of the testcases that need to be run on qemu are not running due to below failures: - kex_exchange_identification: read: Connection reset by peer^M Connection reset by 192.168.7.2 port 22^M ERROR: Couldn't create remote directory /tmp/runtest.3549814 on ssh To resolve kex exchange identification error increased the MaxStartups. (From OE-Core rev: 971d63d5b4982ce9ef4c4be742ba83a9708267da) Signed-off-by: Harish Sadineni Signed-off-by: Richard Purdie (cherry picked from commit df64d5ab6eb37dcdc2046f449ec539a3f4b985c8) Signed-off-by: Steve Sakoman

oeqa/postactions: Fix archive retrieval from target

2024-09-20T15:00:27+00:00

A previous patch broke archive retrieval because of an undeclared variable. Declare the archive_name variable as expected. (From OE-Core rev: 0d0f3541552b073f3536176e2caf847bec24548d) Signed-off-by: Mathieu Dubois-Briand Reported-by: Richard Purdie CC: Alexis Lothoré CC: Alexandre Belloni Signed-off-by: Richard Purdie

rust: Rust Oe-Selftest Reduce the testcases in exclude list

2024-09-12T15:15:09+00:00

We had previously excluded certain test cases due to failures in earlier versions. However, with the latest version of Rust, many of these test cases are passing. As a result, we have removed them from the exclude list and added them back into the test suite for execution. The rust-1.79 has been successfully tested with the current test inclusions. (From OE-Core rev: e8dae0ee5ea958bb84af33d9b9a29ab357d96e31) Signed-off-by: Deepesh Varatharajan Signed-off-by: Richard Purdie

oeqa/selftest: Only rewrite envvars paths that absolutely point to builddir

2024-09-12T15:15:09+00:00

When building the new selftest builddir, paths in environment variables are rewritten to point to the new buildir, but users can have environment variables that point outside of the build dir using relative paths from builddir. We must not rewrite those. Check this by verifying that the absolute path still contains the builddir. Fixes [YOCTO #15241] (From OE-Core rev: c5e70500caffcd0518899cc6eba23a38bc3be108) Signed-off-by: Yoann Congal Signed-off-by: Richard Purdie

meta-ide-support: Mark recipe as MACHINE-specific

2024-09-12T15:15:09+00:00

meta-ide-support:do_write_test_data dumps the bitbake data dictionary to a file using export2json(). As this obviously includes the value of MACHINE, and other MACHINE-specific variables, the recipe needs to be marked as MACHINE-specific. RP: Note that this patch does change the name of the environment script since it is no longer package arch specific but machine arch specific. [RP: Fix selftest to reference new environment file] (From OE-Core rev: 3be2bc8a9b0c9d6a178329c8b451a6bedf255d6c) Signed-off-by: Paul Barker Signed-off-by: Richard Purdie