> I assumed there is a limit on the amount of buffered and not yet committed data, and when that is cross the call would block until more data is flushed to di

> I assumed there is a limit on the amount of buffered and not yet committed data, and when that is cross the call would block until more data is flushed to disk. There is. It's tunable through /proc/sys/vm/dirty_ratio. When there is that much write cache, application writes will start to writeback synchronously. There is also dirty_background_ratio, which is the threshold at which writeback starts happening in the background (that is, in a kernel thread).