Test fails due to rr hang

[neon12345]

Some test hang in "Task::read_bytes_fallible" for me.

This is due to the workaround running indefinitely:

   if (0 == nread && 0 == all_read && 0 == errno) {
   ...
      continue;
    }

With a condition to only check this one, the tests work fine.

I also applied #3436 which could contribute to the whole behavior.

But I don't know if the test coverage still works as expected with the changes.

[rocallahan]

When you get this loop, can you attach gdb to rr and get the addr value, then look up addr in /proc/<tid>/maps for the Task tid and tell us what's at that address?

[GitMensch]

@neon12345 the referenced issue was closed. Can you please recheck if this works for you now (using current master) and if it still hangs check for the requested Task tid?