Andy, >> Second, I'd like to keep separate the idea of using random data ring allocation from the idea of completing commands out of order. Although certainly the second one depends on the first. (And it's not random, it's probably better to say heap-like data ring allocation, in comparison to the stack-like allocation tcmu currently does.) This patch changes the circular data ring into a FIFO stack of pages. This is not what we want for async. > > It is not a FIFO stack. data_pages_used is actually a free-list so data pages can be allocated and freed in random order. I meant 'free_data_pages' here, not 'data_pages_used'. Max -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html