Managing datapath validation on per-transaction basis
A technique for managing a datapath of a data storage system includes receiving a request to access target data and creating a transaction that includes multiple datapath elements in a cache, where the datapath elements are used for accessing the target data. In response to detecting that one of the datapath elements is invalid, the technique further includes processing the transaction in a rescue mode. The rescue mode attempts to replace each invalid datapath element of the transaction with a valid version thereof obtained from elsewhere in the data storage system. The technique further includes committing the transaction as processed in the rescue mode.
1. A method of managing a datapath of a data storage system, the method comprising:
receiving a request to access target data;
creating a transaction that includes multiple datapath elements in a cache, the datapath elements used for accessing the target data;
in response to detecting that one of the datapath elements is invalid, processing the transaction in a rescue mode, the rescue mode checking the datapath elements in the transaction for validity and attempting to replace each invalid datapath element of the transaction with a valid version thereof obtained from elsewhere in the data storage system; and
committing the transaction as processed in the rescue mode.
2. The method of claim 1 , wherein attempting to replace the invalid datapath element with a valid version includes checking for the valid version in a non-volatile store of the data storage system.
3. The method of claim 2 , wherein the non-volatile store is configured as a mirror having two sides, and wherein the method further comprises checking both sides of the mirror for the valid version.
4. The method of claim 3 , wherein the method further comprises, upon detecting that one side of the mirror stores the valid version but the other side of the mirror does not, updating the other side of the mirror so that it also contains the valid version.
5. The method of claim 2 wherein, upon replacing valid datapath element of the transaction with the valid version, the method further comprises marking the datapath element as dirty in the cache.
6. The method of claim 5 , wherein the cache is a first cache disposed in a first processor node, the data storage system also including a second cache disposed in a second processor node, and wherein the method further comprises, upon replacing the invalid datapath element with the valid version, evicting a corresponding datapath element from the second cache.
7. The method of claim 1 , wherein the cache is a first cache disposed in a first processor node, the data storage system also including a second cache disposed in a second processor node, and wherein attempting to replace the invalid datapath element with a valid version includes checking for the valid version in the second cache.
8. The method of claim 7 wherein, upon determining that the second cache contains the valid version of the datapath element, the method further comprises marking the datapath element in the second cache as dirty.
9. The method of claim 1 , wherein the cache is a first cache disposed in a first processor node, the data storage system also including a second cache disposed in a second processor node, and wherein the method further comprises, upon determining that a particular datapath element of the transaction is invalid, evicting the datapath element from the first cache and evicting a corresponding datapath element from the second cache.
10. The method of claim 1 , wherein processing the transaction in the rescue mode does not correct all invalid datapath elements of the transaction, and wherein the method further comprises executing an error handler to manage the uncorrected datapath elements.
11. A computerized apparatus, comprising control circuitry that includes a set of processing units coupled to memory, the control circuitry constructed and arranged to:
receive a request to access target data;
create a transaction that includes multiple datapath elements in a cache, the datapath elements used for accessing the target data;
in response to detecting that one of the datapath elements is invalid, process the transaction in a rescue mode, the rescue mode checking the datapath elements in the transaction for validity and attempting to replace each invalid datapath element of the transaction with a valid version thereof obtained from elsewhere in the data storage system; and
commit the transaction as processed in the rescue mode.
12. A computer program product including a set of non-transitory, computer-readable media having instructions which, when executed by control circuitry of a computerized apparatus, cause the computerized apparatus to perform a method of managing a datapath, the method comprising:
receiving a request to access target data;
creating a transaction that includes multiple datapath elements in a cache, the datapath elements used for accessing the target data;
in response to detecting that one of the datapath elements is invalid, processing the transaction in a rescue mode, the rescue mode checking the datapath elements in the transaction for validity and attempting to replace each invalid datapath element of the transaction with a valid version thereof obtained from elsewhere in the data storage system; and
committing the transaction as processed in the rescue mode.
13. The computer program product of claim 12 , wherein attempting to replace the invalid datapath element with a valid version includes checking for the valid version in a non-volatile store of the data storage system.
14. The computer program product of claim 13 , wherein the non-volatile store is configured as a mirror having two sides, and wherein the method further comprises checking both sides of the mirror for the valid version.
15. The computer program product of claim 14 , wherein the method further comprises, upon detecting that one side of the mirror stores the valid version but the other side of the mirror does not, updating the other side of the mirror so that it also contains the valid version.
16. The computer program product of claim 13 wherein, upon replacing valid datapath element of the transaction with the valid version, the method further comprises marking the datapath element as dirty in the cache.
17. The computer program product of claim 16 , wherein the cache is a first cache disposed in a first processor node, the data storage system also including a second cache disposed in a second processor node, and wherein the method further comprises, upon replacing the invalid datapath element with the valid version, evicting a corresponding datapath element from the second cache.
18. The computer program product of claim 12 , wherein the cache is a first cache disposed in a first processor node, the data storage system also including a second cache disposed in a second processor node, and wherein attempting to replace the invalid datapath element with a valid version includes checking for the valid version in the second cache.
19. The computer program product of claim 18 wherein, upon determining that the second cache contains the valid version of the datapath element, the method further comprises marking the datapath element in the second cache as dirty.
20. The computer program product of claim 12 , wherein the cache is a first cache disposed in a first processor node, the data storage system also including a second cache disposed in a second processor node, and wherein the method further comprises, upon determining that a particular datapath element of the transaction is invalid, evicting the datapath element from the first cache and evicting a corresponding datapath element from the second cache.