11.1.3. Inclusive and exclusive caches

Consider a simple memory read, for example, LDR X0, [X1] in a single core processor.

This is a rather simplistic view. For multi-core and multi-cluster systems, before performing a load from external memory, the caches of L2 or L1 caches of cores within the cluster or of other clusters might also be checked. In addition, there is no consideration of either L3 or system caches at this point.

This is an inclusive cache model, where the same data can be present in both the L1 and L2 caches. In an exclusive cache, data can be present in only one cache and an address cannot be found in both the L1 and L2 caches at the same time.

Copyright © 2015 ARM. All rights reserved.ARM DEN0024A