# US 11,030,018 B2 · On-demand multi-tiered hang buster for SMT microprocessor > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,030,018 B2 - Document ID: US11030018B2 - Status: Granted - Kind: B2 - Application number: 15/831,658 - Filed: 2017-12-05 - Published: 2019-06-06 - Granted: 2021-06-08 - Earliest claimed: 2017-12-05 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Steven J. Battle; Dung Q. Nguyen; Susan E. Eisen; Kenneth L. Ward; Eula Faye Abalos Tolentino; Cliff Kucharski; Glenn O. Kincaid; David S. Walder - Web page: https://patentgrep.com/patent/US11030018B2 ## CPC - G06F 9/4881: {Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues} - G06F 9/524: {Deadlock detection or avoidance} - G06F 9/38: Concurrent instruction execution, e.g. pipeline or look ahead - G06F 9/3854: {Instruction completion, e.g. retiring, committing or graduating} - G06F 12/0855: Overlapped cache accessing, e.g. pipeline (G06F12/0846 takes precedence) - G06F 12/0877: Cache access modes - G06F 12/0895: of parts of caches, e.g. directory or tag array - G06F 12/0897: with two or more cache hierarchy levels (with multilevel cache hierarchies G06F12/0811) - G06F 2212/40: Specific encoding of data in memory or cache - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Embodiments include systems, methods, and computer program products for using a multi-tier hang buster for detecting and breaking out of hang conditions in a processor. One method includes determining a plurality of actions available at each of a plurality of tiers used for breaking out of the hang condition in the processor. The method also includes, after detecting the hang condition on a first thread of the processor, performing one or more actions available at a first tier of the plurality of tiers to break out of the hang condition. The method further includes, after performing the one or more actions at the first tier and determining that the hang condition is still present, performing one or more actions available at one or more second tiers of the plurality of tiers to break out of the hang condition. ## Claims ### Claim 1 A computer-implemented method for breaking out of a hang condition in a processor, comprising: determining a first plurality of actions available at each of a plurality of tiers used for breaking out of the hang condition in the processor; after detecting the hang condition on a first thread of the processor, performing one or more actions available at a first tier of the plurality of tiers to break out of the hang condition; after performing the one or more actions at the first tier and determining that the hang condition is still present, performing one or more actions available at one or more second tiers of the plurality of tiers to break out of the hang condition; and upon reaching a last tier of the plurality of tiers after performing the one or more actions at the first tier and performing the one or more actions at the one or more second tiers: determining that the hang condition is still present; and performing a flush operation to break out of the hang condition, wherein the flush operation is one of the first plurality of actions and is performed at an initial time at the last tier of the plurality of tiers in order to break out of the hang condition. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein detecting the hang condition comprises: monitoring an amount of time that has elapsed since a previous completion on the first thread; and determining that the amount of time that has elapsed since the previous completion on the first thread is above a threshold. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein detecting the hang condition comprises: monitoring a number of completions on at least a second thread in the processor; and determining that the number of completions on the at least a second thread in the processor is above a threshold. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the one or more actions available at the one or more second tiers have a greater impact on operation of the processor than the one or more actions available at the first tier, wherein the impact on the operation of the processor comprises at least one of impact to memory, load, and speed of the processor. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising: detecting that at least one of the first plurality of actions performed by the first thread impacts at least a second thread in the processor; determining, based on one or more conditions, at least one of whether to allow subsequent actions from the at least second thread to impact the first thread and whether to allow subsequent actions from the first thread to impact the at least second thread; and managing the subsequent actions from the first thread at the at least second thread based on the determination of whether to allow the subsequent actions. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein the one or more conditions comprise at least one of whether the first thread was a last thread to request a flush, a current tier on each of the first thread and at least second thread, and whether there was a completion on the first thread or at least second thread after a flush operation. ### Claim 7 A system, comprising: at least one processor; and a memory storing a program, which, when executed on the at least one processor, perform an operation for breaking out of a hang condition in the processor, the operation comprising: determining a first plurality of actions available at each of a plurality of tiers used for breaking out of the hang condition in the processor; after detecting the hang condition on a first thread of the processor, performing one or more actions available at a first tier of the plurality of tiers to break out of the hang condition; after performing the one or more actions at the first tier and determining that the hang condition is still present, performing one or more actions available at one or more second tiers of the plurality of tiers to break out of the hang condition; and upon reaching a last tier of the plurality of tiers after performing the one or more actions at the first tier and performing the one or more actions at the one or more second tiers: determining that the hang condition is still present; and performing a flush operation to break out of the hang condition, wherein the flush operation is one of the first plurality of actions and is performed at an initial time at the last tier of the plurality of tiers in order to break out of the hang condition. ### Claim 8 (depends on 7) The system of claim 7, wherein detecting the hang condition comprises: monitoring an amount of time that has elapsed since a previous completion on the first thread; and determining that the amount of time that has elapsed since the previous completion on the first thread is above a threshold. ### Claim 9 (depends on 7) The system of claim 7, wherein detecting the hang condition comprises: monitoring a number of completions on at least a second thread in the processor; and determining that the number of completions on the at least a second thread in the processor is above a threshold. ### Claim 10 (depends on 7) The system of claim 7, wherein the one or more actions available at the one or more second tiers have a greater impact on operation of the processor than the one or more actions available at the first tier, wherein the impact on the operation of the processor comprises at least one of impact to memory, load, and speed of the processor. ### Claim 11 (depends on 7) The system of claim 7, the operation further comprising: detecting that at least one of the first plurality of actions performed by the first thread impacts at least a second thread in the processor; determining, based on one or more conditions, at least one of whether to allow subsequent actions from the at least second thread to impact the first thread and whether to allow subsequent actions from the first thread to impact the at least second thread; and managing the subsequent actions from the first thread at the at least second thread based on the determination of whether to allow the subsequent actions. ### Claim 12 (depends on 11) The system of claim 11, wherein the one or more conditions comprise at least one of whether the first thread was a last thread to request a flush, a current tier on each of the first thread and at least second thread, and whether there was a completion on the first thread or at least second thread after a flush operation. ### Claim 13 A computer program product, the computer program product comprising: a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform an operation for breaking out of a hang condition in a processing unit, the operation comprising: determining a first plurality of actions available at each of a plurality of tiers used for breaking out of the hang condition in the processing unit; after detecting the hang condition on a first thread of the processor, performing one or more actions available at a first tier of the plurality of tiers to break out of the hang condition; after performing the one or more actions at the first tier and determining that the hang condition is still present, performing one or more actions available at one or more second tiers of the plurality of tiers to break out of the hang condition; and upon reaching a last tier of the plurality of tiers after performing the one or more actions at the first tier and performing the one or more actions at the one or more second tiers: determining that the hang condition is still present; and performing a flush operation to break out of the hang condition, wherein the flush operation is one of the first plurality of actions and is performed at an initial time at the last tier of the plurality of tiers in order to break out of the hang condition. ### Claim 14 (depends on 13) The computer program product of claim 13, wherein detecting the hang condition comprises: monitoring an amount of time that has elapsed since a previous completion on the first thread; and determining that the amount of time that has elapsed since the previous completion on the first thread is above a threshold. ### Claim 15 (depends on 13) The computer program product of claim 13, wherein detecting the hang condition comprises: monitoring a number of completions on at least a second thread in the processing unit; and determining that the number of completions on the at least a second thread in the processing unit is above a threshold. ### Claim 16 (depends on 13) The computer program product of claim 13, wherein the one or more actions available at the one or more second tiers have a greater impact on operation of the processing unit than the one or more actions available at the first tier, wherein the impact on the operation of the processing unit comprises at least one of impact to memory, load, and speed of the processing unit. ### Claim 17 (depends on 13) The computer program product of claim 13, the operation further comprising: detecting that at least one of the first plurality of actions performed by the first thread impacts at least a second thread in the processing unit; determining, based on one or more conditions, at least one of whether to allow subsequent actions from the at least second thread to impact the first thread and whether to allow subsequent actions from the first thread to impact the at least second thread; and managing the subsequent actions from the first thread at the at least second thread based on the determination of whether to allow the subsequent actions. ### Claim 18 (depends on 17) The computer program product of claim 17, wherein the one or more conditions comprise at least one of whether the first thread was a last thread to request a flush, a current tier on each of the first thread and at least second thread, and whether there was a completion on the first thread or at least second thread after a flush operation.