Reducing translation latency within a memory management unit using external caching structures
Reducing translation latency within a memory management unit (MMU) using external caching structures including requesting, by the MMU on a node, page table entry (PTE) data and coherent ownership of the PTE data from a page table in memory; receiving, by the MMU, the PTE data, a source flag, and an indication that the MMU has coherent ownership of the PTE data, wherein the source flag identifies a source location of the PTE data; performing a lateral cast out to a local high-level cache on the node in response to determining that the source flag indicates that the source location of the PTE data is external to the node; and directing at least one subsequent request for the PTE data to the local high-level cache.
1. A computer processor for reducing translation latency within a memory management unit (MMU) using external caching structures, the computer processor configured to carry out the steps of:
requesting, by the MMU on a node, page table entry (PTE) data and coherent ownership of the PTE data from a page table in memory;
receiving, by the MMU, the PTE data, a source flag, and an indication that the MMU has coherent ownership of the PTE data, wherein the source flag identifies a source location of the PTE data;
performing, by the MMU, a lateral cast out of the PTE data to a local high-level cache on the node in response to determining that the source flag indicates that the source location of the PTE data is external to the node; and
directing at least one subsequent request for the PTE data to the local high-level cache.
2. The computer processor of claim 1 , further configured to carry out the steps of:
in response to determining that the source flag indicates that the source location of the PTE data is on the node, accessing the PTE data without performing the lateral cast out to the local high-level cache.
3. The computer processor of claim 1 , wherein the PTE data comprises a change status indicating changes made to a page referenced by the PTE data, and wherein performing the lateral cast out to the local high-level cache on the node is further in response to evaluating the change status.
4. The computer processor of claim 1 , wherein performing the lateral cast out to the local high-level cache on the node comprises storing the PTE data as a cacheline entry in the high-level cache.
5. The computer processor of claim 1 , wherein the MMU is one unit of a distributed MMU on a group of nodes.
6. The computer processor of claim 1 , wherein the local high-level cache is shared by at least one processing core on the node.
7. The computer processor of claim 1 , wherein the memory is external to the node.
8. A computing system, the computing system including a computer processor for reducing translation latency within a memory management unit (MMU) using external caching structures, the computer processor configured to carry out the steps of:
requesting, by the MMU on a node, page table entry (PTE) data and coherent ownership of the PTE data from a page table in memory;
receiving, by the MMU, the PTE data, a source flag, and an indication that the MMU has coherent ownership of the PTE data, wherein the source flag identifies a source location of the PTE data;
performing, by the MMU, a lateral cast out of the PTE data to a local high-level cache on the node in response to determining that the source flag indicates that the source location of the PTE data is external to the node; and
directing at least one subsequent request for the PTE data to the local high-level cache.
9. The computing system of claim 8 , the computer processor further configured to carry out the steps of:
in response to determining that the source flag indicates that the source location of the PTE data is on the node, accessing the PTE data without performing the lateral cast out to the local high-level cache.
10. The computing system of claim 8 , wherein the PTE data comprises a change status indicating changes made to a page referenced by the PTE data, and wherein performing the lateral cast out to the local high-level cache on the node is further in response to evaluating the change status.
11. The computing system of claim 8 , wherein performing the lateral cast out to the local high-level cache on the node comprises storing the PTE data as a cacheline entry in the high-level cache.
12. The computing system of claim 8 , wherein the MMU is one unit of a distributed MMU on a group of nodes.
13. The computing system of claim 8 , wherein the local high-level cache is shared by at least one processing core on the node.