Due to the special nature of the init process, direct use of wait outside the wait4 loop is racy. This change copies the wstatus from wait4 loop state instead.
Signed-off-by: Ophestra <cat@gensokyo.uk>
Most tests already had no global state, however parallel was never enabled. This change enables it for all applicable tests.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This makes it possible to optionally omit the identifying verbose message, for when the Op implementation can provide a much more useful message in its case, using information not yet available to the String method.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This synchronises the wait4 return after the toplevel signal call in lowlastcap_signaled_cancel_forward_error.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This gets inlined and does not cause problems usually but turns out -coverpkg uninlines it and breaks the recovery.
Signed-off-by: Ophestra <cat@gensokyo.uk>
These are very useful in many packages containing relatively large amount of code making calls to difficult or impossible to stub functions.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This used to be entirely done via integration tests, with almost no hope of error injection and coverage profile. These tests significantly increase confidence of future work in this area.
Signed-off-by: Ophestra <cat@gensokyo.uk>
Scheduler nondeterminism cannot be accounted for, so do this instead.
There should not be any performance penalty as these calls are optimised out for direct.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This allows tests to stub all kernel behaviour, enabling measurement of all function call arguments and error injection.
Signed-off-by: Ophestra <cat@gensokyo.uk>
