Row hammer mitigation using a victim cache
Row hammer attacks takes advantage of unintended and undesirable side effects of memory devices in which memory cells interact electrically between themselves by leaking their charges and possibly changing the contents of nearby memory rows that were not addressed in an original memory access. Row hammer attacks are mitigated by using a victim cache. Data is written to cache lines of a cache. A least recently used cache line of the cache is written to the victim cache.
1. A method, comprising:
writing data to cache lines of a first cache; and
performing a row hammer mitigation operation by writing a least recently used cache line from the first cache to a cache line of a second cache; and
writing a least recently used cache line from the second cache back to the second cache.
2. The method of claim 1 , wherein:
the first cache is a set associative cache, and
the second cache is a fully associative cache.
3. The method of claim 1 , wherein the first cache and the second cache each operate according to a least recently used (LRU) caching policy.
4. The method of claim 1 , further comprising:
determining that a least recently used cache line from the second cache is not to be written to a memory device coupled to the first cache or the second cache, or both.
5. The method of claim 1 , further comprising writing the data to the cache lines of the first cache according to a compute express link (CXL) protocol.
6. The method of claim 1 , wherein the first cache comprises a greater quantity of cache lines than the second cache.
7. An apparatus, comprising:
a controller;
a first cache coupled to the controller, the first cache configured to operate according to a least recently used (LRU) scheme;
a second cache coupled to the controller, the second cache configured to operate according to a LRU scheme; and
a memory device coupled to the first cache or the second cache, or both via at least one memory channel, wherein the controller is configured to:
control writing of data to cache lines of the first cache; and
control writing of a least recently used cache line of the first cache to the second cache to perform a row hammer mitigation operation; and
control writing of a least recently used cache line from the second cache back to the second cache.
8. The apparatus of claim 7 , wherein the controller comprises a central controller portion of a memory controller that is configured to operate according to a Compute Express Link protocol.
9. The apparatus of claim 8 , wherein the first cache and the second cache are resident on the central controller portion of the memory controller.
10. The apparatus of claim 7 , wherein the second cache comprises at least 256 cache lines, and wherein the first cache contains at least twice as many cache lines as the second cache.
11. The apparatus of claim 7 , wherein:
the first cache is a set associative cache, and
the second cache is a fully associative cache.
12. The apparatus of claim 7 , wherein the controller is further configured to:
determine that a least recently used cache line from the second cache is not to be written to the memory device.
13. The apparatus of claim 7 , wherein the first cache is physically distinct from the second cache.
14. A system, comprising:
a memory controller comprising:
a central controller;
a first cache comprises a first quantity of cache lines resident on the central controller;
a second cache comprising a second quantity of cache lines resident on the central controller; and
a memory device coupled to the memory controller, wherein the memory controller is configured to:
cause data to be written to the first cache according to a Compute Express Link (CXL) protocol; and
cause a performance of a row hammer mitigation operation that includes writing a least recently used cache line of the first cache to the second cache;
cause a least recently used cache line from the second cache to be written back to the second cache.
15. The system of claim 14 , wherein:
the first cache is a set associative cache, and
the second cache is a fully associative cache.
16. The system of claim 14 , wherein the first cache or the second cache, or both are configured to operate according to a least recently used caching policy.
17. The system of claim 14 , wherein the second cache comprises a lesser quantity of cache lines than the first cache.
18. The system of claim 14 , wherein the memory controller is configured to utilize an instruction level parallelism scheme to cause the data to be written to the first cache.
19. The system of claim 14 , wherein the memory controller is further configured to:
determine that a least recently used cache line from the second cache is not to be written to the memory device.
20. The system of claim 14 , wherein the memory controller is configured to cause the data to be written to the first cache at a rate of at least 32 gigatransfers per second.