# US 11,443,402 B2 · Synchronized data chaining using on-chip cache > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,443,402 B2 - Document ID: US11443402B2 - Status: Granted - Kind: B2 - Application number: 16/609,084 - Filed: 2018-12-04 - Published: 2020-06-18 - Granted: 2022-09-13 - Earliest claimed: 2017-12-04 - Assignees: Google Inc. - Applicants: Google LLC - Inventors: Benjamin Dodge; Jason Rupert Redgrave; Xiaoyu Ma - Web page: https://patentgrep.com/patent/US11443402B2 ## CPC - G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining - G06F 13/28: using burst mode transfer, e.g. direct memory access {DMA}, cycle steal (G06F13/32 takes precedence) - G06F 15/7825: {Globally asynchronous, locally synchronous, e.g. network on chip} - G06T 1/60: Memory management - Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE ## Abstract Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating, by an image sensor of a computing device, frame data comprising sub-frames of image pixel data. A first resource of the system-on-chip provides the frame data to a second resource of the system-on-chip. The frame data is provided to the second resource using a first data path included in the system-on-chip. The first resource provides a token to the second resource using a second data path included in the system-on-chip. A processor of the system-on-chip, uses the token to synchronize production of sub-frames of image pixel data provided by the first resource to the second resource and to synchronize consumption of the sub-frames of image pixel data received by the second resource from the elastic memory buffer. ## Claims ### Claim 1 A computer-implemented method performed using a system-on-chip, the method comprising: generating, by an image sensor of a computing device, frame data comprising sub-frames of image pixel data; providing, by a first resource of the system-on-chip, the frame data to at least one second resource of the system-on-chip, wherein the frame data is provided to the at least one second resource using an elastic memory buffer and a first data path included in the system-on-chip; providing, by the first resource of the system-on-chip, a token to the at least one second resource of the system-on-chip, wherein the token is provided to the at least one second resource using a second data path included in the system-on-chip; and using, by a processor of the system-on-chip, at least the token provided using the second data path to: i) synchronize production of sub-frames of image pixel data provided by the first resource to the at least one second resource; and ii) synchronize consumption of the sub-frames of image pixel data received by the at least one second resource from the elastic memory buffer. ### Claim 2 (depends on 1) The method of claim 1, further comprising: providing, by the at least one second resource of the system-on-chip and using the second data path, a return token to the first resource, wherein the return token corresponds to the token provided by the first resource, and the return token is used along with the token to synchronize production of the sub-frames of image pixel data. ### Claim 3 (depends on 2) The method of claim 2, wherein the first resource is a producer resource that uses a producer direct memory access unit to: i) provide the frame data using the first data path; ii) provide the token using the second data path; and iii) receive the return token using the second data path. ### Claim 4 (depends on 3) The method of claim 3, wherein the at least one second resource is a consumer resource that uses a consumer direct memory access unit to: i) consume the frame data provided using the first path; ii) receive the token provided using the second path; and iii) provide the return token using the second data path. ### Claim 5 (depends on 4) The method of claim 4, wherein: i) the processor allocates a variable amount of memory included in the elastic memory buffer, the memory being used for transfers of sub-frame data between multiple respective producer-consumer pairs; and ii) the processor adjusts an initial amount of allocated memory based on a quantity of producer-consumer pairs that actively transfer sub-frame data, and wherein the sub-frame data has a data size that is less than a threshold data size. ### Claim 6 (depends on 2, 3, 4, 5) The method of any one of claims 2 to 5, wherein the first resource provides the frame data to the at least one second resource via the first data path based on a sub-frame data consumption rate of the at least one second resource, wherein the sub-frame data consumption rate is computed by the processor at least by using the return token. ### Claim 7 (depends on 2, 3, 4, 5, 6) The method of any one of claims 2 to 6, further comprising: i) determining, by the processor and using the return token, a production rate by which the first resource is required to produce frame data to synchronize production of the sub-frames of image pixel data; and ii) providing, by the first resource, the frame data based on the determined production rate. ### Claim 8 (depends on 7) The method of claim 7, further comprising: i) determining, by the processor and using the token, a consumption rate by which the at least one second resource is required to consume sub-frames of data to synchronize consumption of the sub-frames of image pixel data; and ii) consuming, by the at least one second resource, the sub-frames of data based on the determined consumption rate. ### Claim 9 The method of any one of the preceding claims, wherein the first path and the second path are part of the same data communications network included in the system-on-chip. ### Claim 10 The method of any one of the preceding claims, wherein the second path is formed at the system-on-chip using one of multiple virtual channels included in the system-on-chip. ### Claim 11 A system-on-chip comprising: one or more processing devices; and one or more non-transitory machine-readable storage devices storing instructions that are executable by the one or more processing devices to cause performance of operations comprising: generating, by an image sensor of a computing device, frame data comprising sub-frames of image pixel data; providing, by a first resource of the system-on-chip, the frame data to at least one second resource of the system-on-chip, wherein the frame data is provided to the at least one second resource using an elastic memory buffer and a first data path included in the system-on-chip; providing, by the first resource of the system-on-chip, a token to the at least one second resource of the system-on-chip, wherein the token is provided to the at least one second resource using a second data path included in the system-on-chip; and using, by a processor of the system-on-chip, at least the token provided using the second data path to: i) synchronize production of sub-frames of image pixel data provided by the first resource to the at least one second resource; and ii) synchronize consumption of the sub-frames of image pixel data received by the at least one second resource from the elastic memory buffer. ### Claim 12 (depends on 11) The system-on-chip of claim 11, further comprising: providing, by the at least one second resource of the system-on-chip and using the second data path, a return token to the first resource, wherein the return token corresponds to the token provided by the first resource, and the return token is used along with the token to synchronize production of the sub-frames of image pixel data. ### Claim 13 (depends on 12) The system-on-chip of claim 12, wherein the first resource is a producer resource that uses a producer direct memory access unit to: i) provide the frame data using the first data path; ii) provide the token using the second data path; and iii) receive the return token using the second data path. ### Claim 14 (depends on 13) The system-on-chip of claim 13, wherein the at least one second resource is a consumer resource that uses a consumer direct memory access unit to: i) consume the frame data provided using the first path; ii) receive the token provided using the second path; and iii) provide the return token using the second data path. ### Claim 15 (depends on 14) The system-on-chip of claim 14, wherein: i) the processor allocates a variable amount of memory included in the elastic memory buffer, the memory being used for transfers of sub-frame data between multiple respective producer-consumer pairs; and ii) the processor adjusts an initial amount of allocated memory based on a quantity of producer-consumer pairs that actively transfer sub-frame data, and wherein the sub-frame data has a data size that is less than a threshold data size. ### Claim 16 (depends on 12, 13, 14, 15) The system-on-chip of any one of claims 12 to 15, wherein the first resource provides the frame data to the at least one second resource via the first data path based on a sub-frame data consumption rate of the at least one second resource, wherein the sub-frame data consumption rate is computed by the processor at least by using the return token. ### Claim 17 (depends on 12, 13, 14, 15, 16) The system-on-chip of any one of claims 12 to 16, further comprising: i) determining, by the processor and using the return token, a production rate by which the first resource is required to produce frame data to synchronize production of the sub-frames of image pixel data; and ii) providing, by the first resource, the frame data based on the determined production rate. ### Claim 18 (depends on 17) The system-on-chip of claim 17, further comprising: i) determining, by the processor and using the token, a consumption rate by which the at least one second resource is required to consume sub-frames of data to synchronize consumption of the sub-frames of image pixel data; and ii) consuming, by the at least one second resource, the sub-frames of data based on the determined consumption rate. ### Claim 19 (depends on 11, 12, 13, 14, 15, 16, 17, 18) The system-on-chip of any one of claims 11 to 18, wherein: the first path and the second path are part of the same data communications network included in the system-on-chip; and the second path is formed at the system-on-chip using one of multiple virtual channels included in the system-on-chip. ### Claim 20 One or more non-transitory machine-readable storage devices storing instructions that are executable by one or more processing devices to cause performance of operations comprising: generating, by an image sensor of a computing device, frame data comprising sub-frames of image pixel data; providing, by a first resource of the system-on-chip, the frame data to at least one second resource of the system-on-chip, wherein the frame data is provided to the at least one second resource using an elastic memory buffer and a first data path included in the system-on-chip; providing, by the first resource of the system-on-chip, a token to the at least one second resource of the system-on-chip, wherein the token is provided to the at least one second resource using a second data path included in the system-on-chip; and using, by a processor of the system-on-chip, at least the token provided using the second data path to: i) synchronize production of sub-frames of image pixel data provided by the first resource to the at least one second resource; and ii) synchronize consumption of the sub-frames of image pixel data received by the at least one second resource from the elastic memory buffer.