DOMAINS0001 runs into a assert in sel4test-tests/src/main.c@:250

[axel-h]

I've just seen these assertions during test runs on all architectures. Seems this happen reqularly. Is this just a QEMU issue, or could DOMAINS0001 be improved to avoid this message, or it these a real problem?

[...]
Running test DOMAINS0001 (Change domain successfully())
Test DOMAINS0001 passed
Assertion failed: !"unreachable" (/github/workspace/projects/sel4test/apps/sel4test-tests/src/main.c: main: 250)
    </testcase>
    <testcase classname="sel4test" name="DOMAINS0002">
Running test DOMAINS0002 (Try non-existant domain())
<<seL4(CPU 0) [decodeDomainInvocation/1584 T0xfc411400 "32" @80654eb]: Domain Configure: invalid domain (14 >= 4).>>
Test DOMAINS0002 passed
[...]
All is well in the universe

int main(int argc, char **argv)
{
   [...]
   printf("Test %s %s\n", init_data->name, result == SUCCESS ? "passed" : "failed");
    /* send our result back */
    seL4_MessageInfo_t info = seL4_MessageInfo_new(seL4_Fault_NullFault, 0, 0, 1);
    seL4_SetMR(0, result);
    seL4_Send(endpoint, info);

    /* It is expected that we are torn down by the test driver before we are
     * scheduled to run again after signalling them with the above send.
     */
    assert(!"unreachable");
    return 0;
}

[lsf37]

Interesting, I have not seen that one yet. Do you have a pointer to a run?

[axel-h]

For example ARMVIRT_debug_DOM_clang_64 at https://github.com/seL4/seL4/actions/runs/6785175754/job/18442892398?pr=824

[lsf37]

Oh right, of course, they would be in successful runs. That explains why I haven't seen them. Yes, that would definitely be good to address.

[axel-h]

Yes, the test pass. I saw this just by chance in one run and then found it in more places also. I've clarified the log except above to say that the test run passes.