> Memory ordering issue, and a barrier of some sort is required?
> Does the lockup happen on the same chip as the original
> usleep-less Android implementation?
Seems I misremembered. It does not lock up, it just sends
incomplete messages. I added an udelay(100) in commit
de839b8f06bc5dd3f5037c4409a720cbb9bf21c3 [1] which seems to
prevent that.