How can caches be defeated?

Question

I have this question on my assignment this week, and I don't understand how the caches can be defeated, or how I can show it with an assembly program.. Can someone point me in the right direction?

Show, with assembly program examples, how the two different caches (Associative and Direct Mapping) can be defeated. Explain why this occurs and how it can be fixed. Are the same programs used to defeat the caches the same?

Note: This is homework. Don't just answer the question for me, it won't help me to understand the material.

Thank you for correctly identifying as homework :) retagged for you — Jesus Ramos, Jul 10 '11 at 15:04
You get an upvote for not only admitting home-work, but also for being clear that you wish to understand the material. — Moo-Juice, Jul 10 '11 at 15:11

score 6 · Accepted Answer · answered Jul 10 '11 at 15:06

6

A cache is there to increase performance. So defeating a cache means finding a pattern of memory accesses that decreases performance (in the presence of the cache) rather than increases it.

Bear in mind that the cache is limited in size (smaller than main memory, for instance) so typically defeating the cache involves filling it up so that it throws away the data you're just about to access, just before you access it.

answered Jul 10 '11 at 15:06

Paul Butcher

10,722
3
40
44

So, am I correct in thinking that the best way to show this would be to just to create a loop that is at least one cache line greater than the size of the cache, or a combination of a loop and reading/storing data that would have to be split across two cache lines? – John Jul 12 '11 at 00:24
You're thinking along the right lines, yes. Although bear in mind that you're trying to find a pattern that is less efficient with the cache than without. Depending upon the exact details of the cache, simply reading a large amount of data sequentially probably won't defeat it (you might not be getting much benefit from it, but neither will it be hurting you). The trick is to find a pattern where the cost of the cache exceeds the benefit that it gives you. – Paul Butcher Jul 12 '11 at 07:40

score 3 · Answer 2 · answered Jul 10 '11 at 15:58

If you're looking for a hint, think about splitting a data word across 2 cache lines.

(In case you're also looking for the answer, a similar problem was encountered by the x264 developers -- more information available here and here. The links are highly informative, and I really suggest you read them even after you've found your answer.)

score 0 · Answer 3 · answered Jul 11 '11 at 17:10

Another thing to keep in mind is whether the caches you deal with are virtually or physically indexed / tagged. In some variants, cache aliasing forces line replacements even if the cache as such isn't completely filled. In other variants, cache/page coloring collisions might cause evictions. Finally, in multiprocessor systems under certain workloads, cacheline migrations (between the caches of different CPUs) may limit the usefulness of CPU caches.

How can caches be defeated?

3 Answers3