> But what about ordering between first set & next set of 10 slave writes ? > Whether first set & next set can get mixed ? > It is the same process that runs the first 10, as well as the next 10 PCI writes. Consider the scenario: 1. The program/process starts executing. 2. After first 10 iterations of the loop, scheduler decides to migrate the process to the second processor. 3. The next 10 iterations will be done in the second processor. The for loop of 20 iterations is a process undergoing sequential execution. Unless you fork another process in the beginning, and ask it to do the latter half -10 iterations of the for loop, there is no parallelism of execution, and hence there is no way that the second 10 iterations can happen before the first 10 executions. The logic which made you say that the first 10 will be ordered among themselves, and so also for the second 10, holds for both the sets, and will not be hampered, where-so-ever the process runs. -- With regards, Jagadeesh Bhaskar P -------------------- Bugs are by far the largest and most successful class of entity, with nearly a million known species. In this respect they outnumber all the other known creatures about four to one. —Professor Snopes' Encyclopedia of Animal Life ------- -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
