Buffer display data in a chiplet architecture
View Patent ↗An apparatus and method for efficiently managing power consumption among multiple, replicated functional blocks of an integrated circuit. An integrated circuit includes multiple, replicated functional blocks that use separate power domains. Data of a given type is stored in an interleaved manner among at least two of the multiple functional blocks. In one implementation, a prior static allocation determines that only a subset of the functional blocks store the data of the given type. In another implementation, each of the functional blocks stores the data of the given type, and when an idle state has occurred, data of the given type is moved between the multiple functional blocks until one or more functional blocks no longer store data of the given type. When a transition to the idle state has occurred, the functional blocks that do not store the data of the given type are transitioned to a sleep state.
1. An integrated circuit comprising:
a plurality of functional blocks, each comprising circuitry configured to execute tasks and a cache configured to store data; and
power management circuitry configured to:
assign operating parameters to the plurality of functional blocks; and
in response to an idle state of the plurality of functional blocks:
assign first operating parameters to one or more functional blocks of the plurality of functional blocks that do not store data of a given type; and
assign second operating parameters to one or more functional blocks of the plurality of functional blocks that store data of the given type, wherein the first operating parameters provide lower power consumption than the second operating parameters.
2. The integrated circuit as recited in claim 1 , wherein the data of the given type is video frame data.
3. The integrated circuit as recited in claim 1 , wherein the power management circuitry is further configured to:
identify a first subset of the plurality of functional blocks that do not utilize a corresponding cache above a threshold for storing data of a given type; and
transfer data of the given type between functional blocks of the first subset until at least one functional block of the first subset that previously stored data of the given type no longer stores data of the given type.
4. The integrated circuit as recited in claim 1 , wherein the plurality of functional blocks are configured to store data of the given type in an interleaved manner among only a second subset of the plurality of functional blocks.
5. The integrated circuit as recited in claim 3 , wherein the power management circuitry is further configured to transfer data of the given type between functional blocks of the first subset until at least one functional block of the first subset utilizes the cache above the threshold for storing data of the given type.
6. The integrated circuit as recited in claim 3 , wherein one or more of the plurality of functional blocks are configured to:
perform a writeback operation to store a copy of data in a memory when storing the data in the cache; and
skip performing the writeback operation when writing data to the cache based on transferring data of the given type between functional blocks of the first subset.
7. The integrated circuit as recited in claim 3 , wherein the plurality of functional blocks are configured to store data of the given type in an interleaved manner among each of the plurality of functional blocks.
8. A method comprising:
assigning, by power management circuitry, operating parameters to a plurality of functional blocks, each comprising circuitry configured to execute tasks and a cache configured to store data; and
in response to an idle state of the plurality of functional blocks:
assigning, by the power management circuitry, first operating parameters to one or more functional blocks of the plurality of functional blocks that do not store data of a given type; and
assigning, by the power management circuitry, second operating parameters to one or more functional blocks of the plurality of functional blocks that store data of the given type, wherein the first operating parameters provide lower power consumption than the second operating parameters.
9. The method as recited in claim 8 , wherein data of the given type is video frame data.
10. The method as recited in claim 8 , further comprising:
identifying, by the power management circuitry, a first subset of the plurality of functional blocks that do not utilize a corresponding cache above a threshold for storing data of a given type; and
transferring data of the given type, by the power management circuitry, between functional blocks of the first subset until at least one functional block of the first subset that previously stored data of the given type no longer stores data of the given type.
11. The method as recited in claim 8 , further comprising storing data of the given type, by the plurality of functional blocks, in an interleaved manner among only a second subset of the plurality of functional blocks.
12. The method as recited in claim 10 , further comprising transferring data of the given type, by the power management circuitry, between functional blocks of the first subset until at least one functional block of the first subset utilizes the cache above the threshold for storing data of the given type.
13. The method as recited in claim 10 , further comprising:
performing a writeback operation, by one or more of the plurality of functional blocks, to store a copy of data in a memory when storing the data in the cache; and
skipping performing the writeback operation, by one or more of the plurality of functional blocks, when writing data to the cache based on transferring data of the given type between functional blocks of the first subset.
14. The method as recited in claim 10 , further comprising storing data of the given type, by the plurality of functional blocks, in an interleaved manner among each of the plurality of functional blocks.
15. A computing system comprising:
a memory configured to store instructions of one or more tasks and source data to be processed by the one or more tasks;
an integrated circuit configured to execute the instructions using the source data, wherein the integrated circuit comprises:
a plurality of functional blocks, each comprising circuitry configured to execute tasks and a cache configured to store data; and
power management circuitry configured to:
assign operating parameters to the plurality of functional blocks; and
in response to an idle state of the plurality of functional blocks:
assign first operating parameters to one or more functional blocks of the plurality of functional blocks that do not store data of a given type; and
assign second operating parameters to one or more functional blocks of the plurality of functional blocks that store data of the given type, wherein the first operating parameters provide lower power consumption than the second operating parameters.
16. The computing system as recited in claim 15 , wherein the power management circuitry is further configured to transition the one or more functional blocks that do not store data of the given type to a sleep state.
17. The computing system as recited in claim 15 , wherein the power management circuitry is further configured to:
identify a first subset of the plurality of functional blocks that do not utilize a corresponding cache above a threshold for storing data of a given type; and
transfer data of the given type between functional blocks of the first subset until at least one functional block of the first subset that previously stored data of the given type no longer stores data of the given type.
18. The computing system as recited in claim 17 , wherein the power management circuitry is further configured to transfer data of the given type between functional blocks of the first subset until at least one functional block of the first subset utilizes the cache above the threshold for storing data of the given type.
19. The computing system as recited in claim 17 , wherein one or more of the plurality of functional blocks are configured to:
perform a writeback operation to store a copy of data in a memory when storing the data in the cache; and
skip performing the writeback operation when writing data to the cache based on transferring data of the given type between functional blocks of the first subset.
20. The computing system as recited in claim 17 , wherein the plurality of functional blocks are configured to store data of the given type in an interleaved manner among each of the plurality of functional blocks.