qemu.run_serial returns 1 on success and 0 on failure while the ssh qemu.run command returns 0 and -N on failure In tests, both are used the same way: status, output = qemu.run(cmd) status, output = qemu.run_serial(cmd) This is confusing, maybe both functions should have a similar return. For information, the ssh runner uses the python module subprocess https://docs.python.org/3/library/subprocess.html run_serial should be changed to match the subprocess usage: - return 0 in case of success - return N in case of failure - add an option check. if true raise an error Note that the ssh command allow much more setting (stdin, stderr, stdout...)
Louis can you send a patch?
It is possible for me to work on the subject. Perhaps I was waiting from an external point of view. Changing the behaviour of the method run_serial and will break actual usages. Then, should we create a distinct method (serial_run for example) and deprecate this one ? Or should we change run_serial and patch its usage in oe-core ?
Louis, The best approach is to make a change and send the patch to the email list for review or if you really want feedback on the approach, just send an email describing the problem, pointing to this bug and suggest the approach you plan to take. Thanks, ../Randy
Moved en masse from 4.2 to 4.3 by Randy.
Patch review https://lore.kernel.org/all/20230411150503.2105880-1-lrannou@baylibre.com/
There is a patch but it needs some additional work. Louis if you can continue, please assign this bug to back to yourself. Thanks.
Bulk move of 5.0 medium importance issues to 5.99. Move to 5.1 if you want to actively work on an issue.
Move to 5.99 since no one is actively working on this issue. Move to 5.x if you start on a task, please.