High-throughput out-of-order cipher text stealing
A method and device for out-of-order cryptographic processing to eliminate or ameliorate data dependency stalls, in the cryptographic pipeline of a data storage device, resulting from the application of cipher text stealing. A cryptographic processing pipeline performs the steps of: a) cryptographically process a penultimate data block, to produce a penultimate processed data block, and store a portion of the penultimate processed data block in the memory store; b) cryptographically process at least one leading data block, to produce at least one processed leading data block; and c) cryptographically process a concatenation of a last data block and the portion of the penultimate processed data block, to produce a last processed data block, wherein the cryptographic processing pipeline starts to perform step b) before completing step a).
1. A data storage device comprising:
a non-transitory storage medium; and
a cryptographic processing unit, integrated with the data storage device, and configured to:
receive a first portion of a data object;
receive a second portion of the data object, wherein:
the first portion includes a penultimate data block; and
the second portion includes at least one leading data block of one or more leading data blocks; and
the first portion is received before the second portion; and
perform cryptographic processing of the data object to produce a processed data object, wherein:
the data object comprises a sequence of data blocks, including:
the one or more leading data blocks;
the penultimate data block; and
a last data block;
the cryptographic processing unit comprises:
a memory store;
a cryptographic processing pipeline; and
a pipeline controller;
the pipeline controller is configured to control the cryptographic processing pipeline to perform the steps of:
a) cryptographically process the penultimate data block, to produce a penultimate processed data block, and store a portion of the penultimate processed data block in the memory store;
b) cryptographically process at least one of the one or more leading data blocks, to produce one or more processed leading data blocks; and
c) cryptographically process a concatenation of the last data block and the portion of the penultimate processed data block, to produce a last processed data block; and
the pipeline controller is further configured to control the cryptographic processing pipeline to start to perform step b) before the cryptographic processing pipeline completes step a).
2. The data storage device of claim 1 , wherein performing cryptographic processing comprises performing encryption.
3. The data storage device of claim 1 , wherein performing cryptographic processing comprises performing decryption.
4. The data storage device of claim 1 , wherein:
each leading data block has a size equal to a set block size;
the penultimate data block has a size equal to a set block size; and
the last data block has a size less than the set block size.
5. The data storage device of claim 1 , wherein the cryptographic processing pipeline comprises a plurality of pipeline stages comprising, at least, a first stage and a second stage.
6. The data storage device of claim 5 , wherein the pipeline controller is further configured to control the first stage of the cryptographic processing pipeline to start to perform step b) while the second stage of the cryptographic processing pipeline continues to perform step a).
7. The data storage device of claim 1 , wherein the cryptographic processing pipeline is clocked by a clock signal.
8. The data storage device of claim 7 , wherein the cryptographic processing pipeline stores the portion of the penultimate processed data block in the memory store at least one clock cycle of the clock signal before an input stage of the cryptographic processing pipeline starts to perform step c).
9. The data storage device of claim 1 , wherein the cryptographic processing pipeline processes all of the one or more leading data blocks in step b).
10. The data storage device of claim 1 , further comprising:
an output buffer configured to store the penultimate processed data block; and
an output port configured to:
output processed data blocks from the cryptographic processing unit, and
output the last processed data block before outputting the penultimate processed data block, stored in the output buffer.
11. The data storage device of claim 1 , further comprising an input port in communication with a source of data blocks of the data object, wherein the cryptographic processing unit is configured to request and receive data blocks of the data object from the source via the input port.
12. The data storage device of claim 11 , wherein the cryptographic processing unit is further configured to:
request the first portion of the data object from the source via the input port; and
then request the second portion of the data object from the source via the input port.
13. The data storage device of claim 1 , wherein step a) comprises determining a tweak value associated with the penultimate data block.
14. The data storage device of claim 13 , wherein the cryptographic processing unit is configured to determine the tweak value associated with the penultimate data block before performing step a).
15. The data storage device of claim 1 , wherein the cryptographic processing unit is configured to calculate a tweak value for each data block in the sequence of data blocks.
16. The data storage device of claim 1 , wherein the cryptographic processing unit is configured to perform encryption in accordance with an XTS-AES encryption algorithm.
17. The data storage device of claim 1 , wherein the cryptographic processing unit is configured to perform decryption in accordance with an XTS-AES decryption algorithm.
18. A method for performing cryptographic processing of a data object to produce a processed data object, wherein:
the data object comprises a sequence of data blocks, including,
one or more leading data blocks;
a penultimate data block; and
a last data blocks;
the method comprises ordered steps of:
a) receiving a first portion of the data object, the first portion including the penultimate data block;
b) receiving a second portion of the data object, the second portion including at least one leading data block of the one or more leading data blocks;
performing cryptographic processing of penultimate data block to produce a penultimate processed data block and store a portion of the penultimate processed data block in a memory store;
d) performing cryptographic processing of at least one of the one or more leading data blocks to produce one or more processed leading data blocks; and
e) performing cryptographic processing of a concatenation of the last data block and the portion of the penultimate processed data block stored in the memory store to produce a last processed data block; and
step d) is started before step c) is completed.
19. The method of claim 18 , wherein the method is performed by a cryptographic unit comprising:
a memory store;
a hardware circuitry pipeline configured to encrypt a plaintext data block to produce a cipher text block; and
a pipeline controller configured to control the pipeline.
20. A data storage device for performing cryptographic processing of a data object to produce a processed data object, wherein:
the data object comprises a sequence of data blocks, including,
one or more leading data blocks;
a penultimate data block; and
a last data block;
the data storage device comprises means for:
a) receiving a first portion of the data object, the first portion including the penultimate data block;
b) receiving a second portion of the data object, the second portion including at least one leading data block of the one or more leading data blocks;
c) cryptographically processing the penultimate data block, to produce a penultimate processed data block, and storing a portion of the penultimate processed data block in a memory store;
d) cryptographically processing at least one of the one or more leading data blocks, to produce one or more processed leading data blocks; and
e) cryptographically processing a concatenation of the last data block and the portion of the penultimate processed data block, to produce a last processed data block,
the first portion is received before the second portion; and
step d) is started before step c) is completed.