# US 12,007,820 B2 · Systems, devices, and methods for controller devices handling fault events > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,007,820 B2 - Document ID: US12007820B2 - Status: Granted - Kind: B2 - Application number: 17/337,650 - Filed: 2021-06-03 - Published: 2021-12-09 - Granted: 2024-06-11 - Earliest claimed: 2020-06-04 - Assignees: Infineon Technologies AG - Applicants: Infineon Technologies AG - Inventors: Konstantin Ivanchenko; Petru Bacinski; Bejoy Mathews - Web page: https://patentgrep.com/patent/US12007820B2 ## CPC - G06F 1/30: Means for acting in the event of power-supply failure or interruption, e.g. power-supply fluctuations (for resetting only G06F1/24) - G06F 11/3058: {Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations (thermal management in cooling arrangements of a computing system G06F1/206)} - G06F 11/0721: {within a central processing unit [CPU]} - G06F 11/079: {Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36)} - G06F 11/2015: {Redundant power supplies (power supply failure G06F1/30)} - G06F 11/2041: {with more than one idle spare processing component} - G06F 11/2043: {where the redundant components share a common memory address space} - G06F 11/2046: {where the redundant components share persistent storage (G06F11/2043 takes precedence)} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A controller chip includes a first cluster including one or more first controller units, a first power supply grid, a first clock tree structure to supply one or more clock signals, and at least a first power supply input. A second cluster includes one or more second controller units, a second power supply grid, a second clock tree structure to supply one or more clock signals, and at least a second power supply input. A monitoring cluster includes a monitoring circuit configured to: monitor the power supply and the clock signal supply of each of the first cluster and second cluster, and in the event of determining at least one of a power supply failure or a clock signal supply failure in one cluster of the first cluster or the second cluster, indicate the failure to the other cluster to take one or more actions. ## Claims ### Claim 1 A controller chip, comprising: a first cluster comprising one or more first controller units configured to provide a first computational effort for an application, a first power supply grid configured to power the one or more first controller units, a first clock tree structure configured to supply one or more clock signals to the one or more first controller units, a first reset tree configured to reset one or more components of the first cluster, and a first power supply input configured to provide power to the first cluster; a second cluster comprising one or more second controller units configured to provide a second computational effort for the application, a second power supply grid configured to power the one or more second controller units, a second clock tree structure configured to supply one or more clock signals to the one or more second controller units, a second reset tree configured to reset one or more components of the second cluster, and a second power supply input configured to provide power to the second cluster; and a monitoring cluster different from the first cluster and the second cluster, and comprising a third clock tree structure and a monitoring circuit, wherein the first, second, and third clock tree structures are configured to operate independently of each other, and the monitoring circuit is configured to: monitor the first power supply input and the first clock tree structure of the first cluster, and the second power supply input and the second clock tree structure of the second cluster; and in response to determining a power supply failure and/or a clock signal supply failure in the first cluster, indicate the determined power supply failure and/or the determined clock signal supply failure to the second cluster to take one or more actions; wherein the second cluster is configured, in response to receiving an indication from the monitoring circuit of the power supply failure or the clock signal supply failure in the first cluster, to re-assign the first computational effort from the first cluster to the second cluster according to the indication. ### Claim 2 (depends on 1) The controller chip of claim 1, wherein the monitoring circuit is further configured to, in response to determining the power supply failure and/or the clock signal supply failure in the first cluster, initiate at least a partial restart of the first cluster. ### Claim 3 (depends on 1) The controller chip of claim 1, wherein the first cluster, the second cluster, and the monitoring cluster are formed on a single monolithic semiconductor substrate. ### Claim 4 (depends on 1) The controller chip of claim 1, wherein the first power supply input configured to provide power to the first cluster is independent from the second power supply input configured to provide power to the second cluster. ### Claim 5 (depends on 4) The controller chip of claim 4, further comprising: a third power supply input configured to power the monitoring circuit, the third power supply input being independent from the first power supply input and the second power supply input. ### Claim 6 (depends on 1) The controller chip of claim 1, wherein the monitoring cluster comprises a power management circuit coupled to the first cluster and to the second cluster, wherein in response to the monitoring circuit determining the power supply failure in the first cluster, the power management circuit is configured to cause power from a secondary source to be provided to the first cluster having the power supply failure. ### Claim 7 (depends on 1) The controller chip of claim 1, wherein each of the first cluster and the second cluster is configured to monitor for the power supply failure, the clock signal supply failure, or a reset failure of a cluster different from itself. ### Claim 8 (depends on 1) The controller chip of claim 1, wherein the first reset tree and the second reset tree are configured to operate independently of each other. ### Claim 9 (depends on 1) The controller chip of claim 1, wherein the monitoring cluster comprises a third power supply input configured to provide power to the monitoring cluster, wherein the third power supply input is independent of the first and second power supply inputs. ### Claim 10 (depends on 1) The controller chip of claim 1, wherein the monitoring circuit comprises a first voltage monitor coupled to the first power supply input of the first cluster and a second voltage monitor coupled to the second power supply input of the second cluster. ### Claim 11 (depends on 1) The controller chip of claim 1, wherein prior to re-assignment of the computational effort from the first cluster to the second cluster, the first cluster uses a first amount of computational effort and the second cluster use a second amount of computational effort, wherein the second amount of computational effort is less a maximum computational effort of the second cluster by a predetermined amount of computational effort, and wherein the first amount of computational effort is less than or equal to the predetermined amount such that after the re-assignment the second cluster has a resultant computational effort that is a sum of the first amount of computational effort and the second amount of computational effort. ### Claim 12 A method to be performed by a controller chip, wherein the controller chip includes: a first cluster comprising one or more first controller units, a first power supply grid to power the one or more first controller units, a first clock tree structure to supply one or more clock signals to the one or more first controller units, a first reset tree configured to reset one or more components of the first cluster, and a first power supply input to provide power to the first cluster; and a second cluster comprising one or more second controller units, a second power supply grid to power the one or more second controller units, a second clock tree structure to supply one or more clock signals to the one or more second controller units, a second reset tree configured to reset one or more components of the second cluster, and a second power supply input to provide power to the second cluster, the method comprising: monitoring, by a monitoring circuit of a monitoring cluster that includes a third clock tree, the first and second power supply inputs and the first and second clock tree structures, the monitoring cluster being different from the first cluster and the second cluster and the first, second, and third clock tree structures configured to operate independently of each other; identifying, by the monitoring circuit, at least one failure in the first power supply input and/or the first clock tree structure while the first cluster is performing one or more computational actions; indicating, by the monitoring circuit, the at least one failure to the second cluster; and performing, at least in part by the second cluster, the one or more computational actions previously performed by the first cluster. ### Claim 13 (depends on 12) The method of claim 12, wherein the first cluster, the second cluster, and the monitoring cluster are formed on a single semiconductor substrate. ### Claim 14 (depends on 12) The method of claim 12, wherein indicating the at least one failure to the second cluster comprises indicating the one or more computational actions previously performed by the first cluster to the second cluster. ### Claim 15 (depends on 12) The method of claim 12, wherein identifying the at least one failure from the first cluster comprises receiving one or more signals from the first cluster indicating the at least one failure. ### Claim 16 A controller chip, comprising: a plurality of clusters each including one or more controller units, a power supply grid configured to power the one or more controller units, a clock tree structure configured to supply one or more clock signals to the one or more controller units, a reset tree configured to reset one or more controller units of a first cluster, and a power supply input configured to provide power to a respective cluster of the plurality of clusters, wherein the clock tree structures of the plurality of clusters are configured to operate independently of each other; and at least one monitoring circuit configured to monitor a power supply input and a clock tree structure of at least one cluster of the plurality of clusters, and in response to determining a clock tree structure failure in the at least one cluster of the plurality of clusters, indicate the clock tree structure failure to at least one other non-failing cluster of the plurality of cluster. ### Claim 17 (depends on 16) The controller chip of claim 16, wherein the at least one monitoring circuit is configured to indicate the clock tree structure failure by transmitting a signal to the at least one other non-failing cluster of the plurality of clusters. ### Claim 18 (depends on 16) The controller chip of claim 16, wherein the at least one other non-failing cluster is configured, in response to receiving the indication from the monitoring circuit of the clock tree structure failure, to re-assign a computational effort from the at least one cluster to the at least one other non-failing cluster according to the indication, and subsequently re-assigning the computational effort back to the at least one cluster when the at least one cluster is repaired or reset.