Thermal prediction and regulation during integrated circuit testing

Abstract

One embodiment of the invention sets forth an integrated circuit testing technique. The technique includes generating one or more temperature predictions for a region of interest within an integrated circuit based on sensor data associated with the integrated circuit. The technique also includes determining one or more control parameters associated with testing of the integrated circuit based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit. The technique further includes causing testing of the integrated circuit to be performed based on the one or more control parameters.

Description

[0001] This application claims priority to U.S. Patent Application No. 63 / 755,975, filed February 7, 2025, and U.S. Patent Application No. 19 / 340,752, filed September 25, 2025, the full text of which is incorporated herein by reference. Technical Field

[0002] This invention relates generally to electronic and integrated circuit testing, and more specifically to thermal prediction and conditioning during integrated circuit testing. Background Technology

[0003] Integrated circuit (IC) testing typically involves a series of tests to verify the functionality, performance, and reliability of the IC and / or semiconductor device. These tests are generally used to detect various defects and / or problems associated with different stages of IC and / or IC manufacturing.

[0004] A typical integrated circuit testing process may include sorting tests to verify the conductivity, leakage current, and / or other electrical characteristics of the chips on the wafer. Chips that pass the sorting tests are separated and packaged, and each packaged chip undergoes aging tests, subjecting it to stress at its rated operating temperature or higher to identify and reject devices that fail early. Each packaged chip may also undergo final testing to evaluate its power consumption, frequency, input / output (I / O) functionality, and / or other characteristics, and to verify that it meets design specifications for integrity and reliability. Multiple chips can also be assembled into integrated systems (e.g., System-on-a-Chip (SoC), System-in-Package (SiP), Multi-Chip Module (MCM), Printed Circuit Board (PCB), etc.) and undergo system-level testing to verify the correct operation of the integrated system in its intended end-use application.

[0005] During integrated circuit (IC) testing, ICs may be exposed to a variety of environmental conditions, causing their internal temperatures to rise and / or fluctuate. To prevent temperatures from reaching unsafe or potentially damaging levels, thermal management systems are typically used to monitor and / or regulate temperatures during IC testing. For example, a conventional thermal management system compares temperature measurements from various temperature sensors, average temperature values ​​(e.g., the average temperature from multiple temperature sensors), power consumption, and / or other values ​​measured or determined during a particular IC test to one or more preset limits. When a corresponding temperature measurement or value reaches or exceeds a preset limit, the thermal management system sends a control signal to heat or cool the IC.

[0006] However, with the continuous improvement of the performance and integration of modern integrated circuits (ICs), testing requirements are becoming increasingly complex, and the difficulty of controlling IC temperature is also increasing. More specifically, the increase in IC power consumption leads to a corresponding increase in the risk of thermal damage. At the same time, the actual temperature of certain areas in complex ICs, packages, and / or IC integration systems may differ significantly from the temperature measured by temperature sensors. Furthermore, during testing and / or rapid switching of operating modes, the thermal management system may fail to respond promptly to sudden changes in heat generation. This inability to monitor the temperature of the entire IC and / or compensate for sudden changes in heat generation during IC testing may cause temperatures to reach or even exceed relevant limits, ultimately resulting in IC damage or scrapping.

[0007] As mentioned above, what is needed in the field is a more effective technology to determine and control the temperature of ICs during IC testing. Summary of the Invention

[0008] One embodiment of the present invention describes an integrated circuit testing technique. This technique includes: generating one or more temperature predictions for a region of interest within the integrated circuit based on sensor data associated with the integrated circuit. The technique further includes: determining one or more control parameters related to integrated circuit testing based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit. The technique further includes: causing testing of the integrated circuit based on the one or more control parameters.

[0009] Compared to existing technologies, one advantage of the disclosed technology is its ability to accurately and timely monitor and control the temperature of each region of interest (AOI) within an integrated circuit (IC). Therefore, the disclosed technology can avoid and / or mitigate thermal damage and / or failure caused by the limited temperature monitoring capabilities of temperature sensors and / or their inability to quickly respond to sudden temperature changes during IC testing. These advantages provide one or more technological improvements superior to existing technologies. Attached Figure Description

[0010] To gain a more detailed understanding of the features of the present invention described above, a more specific description of the invention can be obtained with reference to the accompanying drawings (which have been briefly outlined above). It should be noted that the drawings only illustrate typical embodiments of the invention and should not be considered as limiting the scope of the invention, as the invention may have other equally effective embodiments.

[0011] Figure 1 is a block diagram of one or more oriented computing systems for implementing various embodiments.

[0012] Figure 2 is a more detailed illustration of the training engine and execution engine in Figure 1 according to various embodiments.

[0013] Figure 3 shows exemplary training data for the machine learning model in Figure 2 according to various embodiments.

[0014] Figure 4 is an example system for performing integrated circuit (IC) tests according to various embodiments.

[0015] Figure 5 is a flowchart of the method steps for performing IC testing according to various embodiments. Detailed Implementation

[0016] The following description lists many specific details to provide a more comprehensive understanding of embodiments of the invention. However, those skilled in the art will understand that embodiments of the invention can be practiced even without considering these specific details.

[0017] Overview

[0018] As described above, a conventional thermal management system can regulate the temperature inside an integrated circuit by comparing the temperature values ​​measured by individual temperature sensors, the average temperature value (e.g., the average temperature value from multiple temperature sensors), power consumption, and / or other values ​​measured or determined during a specific integrated circuit test with one or more preset limits. When the preset limits are reached or exceeded during testing, the thermal management system can generate control signals to heat or cool the corresponding integrated circuit.

[0019] However, the temperature in certain areas of complex integrated circuits, packages, and / or integrated circuit integration systems may differ significantly from the temperature sensor measurements. Furthermore, thermal management systems may fail to respond promptly to thermal fluctuations arising from events such as (but not limited to) rapid switching during testing phases and / or operating modes. This inadequate ability to monitor the temperature of the entire integrated circuit and / or compensate for thermal fluctuations during integrated circuit testing can lead to temperatures reaching and / or exceeding appropriate limits, resulting in damage or destruction of the integrated circuit.

[0020] To overcome the aforementioned limitations, the disclosed technology is able to predict and regulate the temperature of individual regions of interest (AOIs) within an integrated circuit during testing. Each AOI may include (but is not limited to) two-dimensional (2D) and / or three-dimensional (3D) mesh cells, functional modules, regions of arbitrary shapes, and / or other types of regions or zones within the integrated circuit.

[0021] More specifically, the disclosed technology utilizes machine learning models, physics-based models, simulations, and / or other techniques to generate temperature predictions for each region of interest (AOI) within an integrated circuit, based on sensor data, test parameters, and / or other inputs collected during integrated circuit testing. Each temperature prediction can represent the “true silicon temperature” at the corresponding AOI, rather than the temperature measured by one or more temperature sensors located outside the AOI and / or at different distances from the AOI.

[0022] Temperature prediction is also used in conjunction with machine learning models, physics-based models, simulations, and / or other techniques to generate control parameters for regulating the temperature within the region of interest (AOI) during integrated circuit testing. These control parameters may include, but are not limited to, target temperature, thermal control signals for the thermal system, control signals for adjusting test pacing, and / or control signals for initiating or stopping test operations. The determination of control parameters may be based on temperature prediction and / or test-related control characteristics, such as, but not limited to, temperature limits, temperature slopes, power slopes, safety offsets, and / or other numerical values ​​characterizing test progress and / or conditions.

[0023] Temperature prediction and / or control parameters can be generated by a separate computing unit (e.g., a modular computer (COM), a single-board computer, etc.). This computing unit operates independently of and concurrently with the thermal and test systems (e.g., automated test equipment (ATE), test procedures, etc.). The thermal system uses these control parameters to thermally control the integrated circuit (IC), while the test system tests the IC. This separate computing unit acts as an intermediate control point, decoupling the test and thermal systems while allowing the operation of each system to influence each other. Furthermore, this separate computing unit allows the thermal system, test system, and / or other components to access the IC-related temperature on demand (e.g., there is no "blind spot" where temperature sensors are unavailable).

[0024] Furthermore, temperature prediction and control can be performed at different stages of integrated circuit testing, such as (but not limited to) wafer sorting testing, chip-level testing, aging testing, final testing, and / or system-level testing. Data collected in early testing phases can be tracked and used for thermal prediction and control of the same integrated circuit in subsequent phases. This data can also be used to design, evaluate, and / or modify the chip layout and / or packaging involved in this integrated circuit, test configuration, and / or other parameters related to the testing and / or use of this integrated circuit.

[0025] Compared to existing technologies, one of the advantages of the disclosed technology is its ability to accurately and timely monitor and control the temperature of each area of ​​interest (AOI) within an integrated circuit (IC). Therefore, the disclosed technology can avoid and / or mitigate thermal damage and / or destruction caused by the limited temperature monitoring capabilities of temperature sensors and / or their inability to quickly respond to sudden changes in heat generated during IC testing. Furthermore, by incorporating thermal data collected in earlier testing phases into thermal prediction and control in subsequent testing phases (involving the same IC), the disclosed technology can improve the design and / or thermal management of complex ICs, packages, and / or multi-component integrated systems with different thermal responses. These advantages provide one or more technological improvements superior to existing technologies.

[0026] System Overview

[0027] Figure 1 is a block diagram of one or more orientations of a computing system 100 for implementing various embodiments. In at least one embodiment, the computing system 100 may include any type of computing device, including but not limited to servers, server platforms, desktop computers, laptop computers, handheld / mobile devices, digital kiosks, in-vehicle infotainment systems, smart speakers or displays, televisions, and / or wearable devices. In at least one embodiment, the computing system 100 is a server running in a data center or cloud computing environment that provides scalable computing resources as a service over a network.

[0028] In various embodiments, computing system 100 includes, but is not limited to, one or more processors 102 and one or more memories 104, which are connected to parallel processing subsystem 112 via memory bridge 105 and communication path 113. Memory bridge 105 is also connected to I / O (input / output) bridge 107 via communication path 106, and I / O bridge 107 is in turn connected to switch 116.

[0029] In one embodiment, I / O bridge 107 is configured to receive user input from optional input device 108, such as (but not limited to) a keyboard, mouse, touchscreen, sensor data analytics (e.g., evaluating gestures, voice, or other data in one or more sensor fields of view or light field for one or more purposes), VR / MR / AR headset, gesture recognition system, steering wheel, mechanical, digital, or touch buttons or input components, and / or microphone, and forward the input data to processor 102 for processing. In at least one embodiment, computing system 100 may be a server in a cloud computing environment. In these embodiments, computing system 100 may omit input device 108 and receive equivalent input in the form of commands (e.g., in response to one or more inputs from a remote computing device) and / or information, which are transmitted over a network and received via network adapter 118. In at least one embodiment, switch 116 is configured to provide connectivity between I / O bridge 107 and other components of computing system 100, such as network adapter 118 and various add-on cards 120 and 121.

[0030] In at least one embodiment, I / O bridge 107 is connected to system disk 114, which is configured to store content, applications, and data used by processor 102 and parallel processing subsystem 112. In one embodiment, system disk 114 provides non-volatile storage for applications and data and may include fixed or removable hard drives, flash drives, and CD-ROMs (CD-ROMs), DVD-ROMs (Digital Multifunction Optical Drives), Blu-ray discs, HD-DVDs (High Definition DVDs), or other magnetic, optical, or solid-state storage devices. In various embodiments, other components, such as universal serial buses or other port connections, optical disc drives, digital multifunction optical disc drives, film recording devices, etc., may also be connected to I / O bridge 107.

[0031] In various embodiments, memory bridge 105 may be a northbridge chip, and I / O bridge 107 may be a southbridge chip. Furthermore, communication paths 106 and 113, as well as other communication paths within computing system 100, may be implemented using any technically suitable protocol, including but not limited to AGP (Accelerated Graphics Port), HyperTransport, or any other bus or point-to-point communication protocol known in the art.

[0032] In at least one embodiment, the parallel processing subsystem 112 includes a graphics subsystem that transmits pixels to an optional display device 110, which can be any conventional cathode ray tube display, liquid crystal display, light-emitting diode display, and / or similar device. In such embodiments, the parallel processing subsystem 112 may include circuitry optimized for graphics and video processing, such as video output circuitry. Such circuitry may be integrated into one or more parallel processing units (PPUs) within the parallel processing subsystem 112, which are also referred to herein as parallel processors.

[0033] In at least one embodiment, the parallel processing subsystem 112 includes circuitry optimized for general and / or computational processing (e.g., optimized circuitry). Furthermore, such circuitry may be integrated into one or more parallel processing units (PPUs) within the parallel processing subsystem 112, which are configured to perform such general and / or computational tasks. In other embodiments, one or more PPUs within the parallel processing subsystem 112 may be configured to perform graphics processing, general processing, and / or computational processing operations. Memory 104 includes at least one device driver configured to manage processing tasks of one or more PPUs within the parallel processing subsystem 112. Additionally, memory 104 also includes instructions for implementing the training engine 122 and the execution engine 124, which can be executed by the processor and / or the parallel processing subsystem 112.

[0034] In various embodiments, the parallel processing subsystem 112 may be integrated with one or more other elements in FIG. 1 to form a single system. For example, the parallel processing subsystem 112 may be integrated with the processor 102 and other connectivity circuitry on a single chip to form a system-on-a-chip (SoC).

[0035] Processor 102 may include any suitable processor, such as a central processing unit (CPU), graphics processing unit (GPU), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), artificial intelligence (AI) accelerator, deep learning accelerator (DLA), parallel processing unit (PPU), data processing unit (DPU), or vision processing accelerator (DLA), parallel processing unit (PPU), data processing unit (DPU), vector or vision processor (VPU), programmable vision accelerator (PVA) (which may include one or more VPUs, pixel processing engine (PPE) and / or direct memory access (DMA) system), any other type of processing unit, or a combination of different processing units, such as a CPU configured to work in conjunction with a GPU. Generally, processor 102 may include any technically feasible hardware unit capable of processing data and / or executing software applications. Furthermore, in the context of this disclosure, the computing elements shown in computing system 100 may correspond to physical computing systems (e.g., systems or machines in a data center), and / or may correspond to virtual computing instances running in a computing cloud.

[0036] In at least one embodiment, processor 102 issues commands to control the operation of the PPU. In at least one embodiment, communication path 113 is a Peripheral Component Interconnect High-Speed ​​(PCIe) connection, wherein each PPU is allocated a dedicated channel. Other communication paths may also be used. The PPU preferably implements a highly parallel processing architecture and may be equipped with any amount of local parallel processing memory (PP memory).

[0037] It should be understood that the system shown in this document is merely an example, and actual systems may have various variations and modifications. The connection topology (including the number and arrangement of bridges, the number of processors 102, and the number of parallel processing subsystems 112) can be modified as needed. For example, in at least one embodiment, memory 104 may be directly connected to processor 102 instead of through memory bridge 105; other devices may communicate with memory 104 through memory bridge 105 and processor 102. In other embodiments, parallel processing subsystems 112 may be connected to I / O bridge 107 or directly to processor 102 instead of to memory bridge 105. In other embodiments, I / O bridge 107 and memory bridge 105 may be integrated into a single chip rather than existing as one or more separate devices. In some embodiments, one or more elements shown in FIG. 1 may be absent. For example, switch 116 can be omitted, and network adapter 118 and add-on cards 120 and 121 can be directly connected to I / O bridge 107. Furthermore, in some embodiments, one or more of the components shown in FIG1 can be implemented as virtualized resources in a virtual computing environment (e.g., a cloud computing environment).

[0038] In one or more embodiments, computing system 100 is configured to execute training engine 122, execution engine 124, and management engine 126 residing in system memory 104. Training engine 122, execution engine 124, and management engine 126 may be stored in system disk 114 and / or other storage devices and loaded into system memory 104 at execution time.

[0039] In some embodiments, training engine 122, execution engine 124, and management engine 126 include functions for performing thermal prediction and regulation during integrated circuit (IC) testing. More specifically, training engine 122 and execution engine 124 may use machine learning models, physics-based models, simulations, experiments, and / or other techniques to generate temperature predictions for various regions of interest (AOIs) within a specific IC. Training engine 122 and execution engine 124 may also use the same and / or different techniques to generate control parameters and / or control signals used to regulate the temperature within the AOI during IC testing. Temperature predictions, control parameters, and / or control signals may be generated and / or updated in a control loop to adjust the temperature of the AOI before it reaches a level that could damage the IC.

[0040] Management engine 126 can generate, store, and / or update temperature predictions, control parameters, control signals, test results, chip layout, test configurations, and / or other data records used at various stages of integrated circuit testing, including (but not limited to) wafer sorting, chip-level processing, aging tests, final tests, and / or system-level tests. Management engine 126 can also use data collected in earlier testing stages to perform thermal prediction and / or temperature control on the same integrated circuit in subsequent testing stages. The operation of training engine 122, execution engine 124, and management engine 126 will be described in detail below.

[0041] Thermal prediction and regulation during integrated circuit testing

[0042] Figure 2 is a more detailed illustration of the training engine 122, execution engine 124, and management engine 126 in Figure 1 according to various embodiments. As described above, the training engine 122, execution engine 124, and management engine 126 all include functions for performing thermal prediction and regulation during IC testing. Each component will be described in more detail below.

[0043] In one or more embodiments, thermal conditioning and control are performed using one or more machine learning models 208 during testing of a particular integrated circuit (IC). More specifically, the one or more machine learning models 208 may generate temperature predictions 242 for various regions of interest (AOIs) on the IC based on inputs 240 related to the testing and / or IC. Each temperature prediction represents the “true silicon temperature” at the corresponding AOI, rather than the temperature measured by one or more temperature sensors located outside the AOI and / or at different distances from the AOI. Therefore, the temperature predictions 242 can be used to evaluate the thermal response and / or characteristics of locations within the IC that cannot be directly measured using temperature sensors.

[0044] The same machine learning model 208 and / or one or more additional machine learning models 208 can also generate control parameters 246, which are used to adjust the temperature within the region of interest (AOI) during testing based on temperature prediction 242 and / or other inputs 240. For example, one or more machine learning models 208 can be used to generate target temperatures, temperature limits, thermal control signals, signals for adjusting test pacing and / or stopping testing, and / or other values ​​that can be used to control and / or adjust the temperature within the IC and / or within each AOI of the IC.

[0045] In one or more embodiments, an AOI corresponds to a region within an IC. For example, the IC may be divided into two-dimensional (2D) and / or three-dimensional (3D) mesh cells, each mesh cell corresponding to a different AOI. In another example, each AOI may be defined as a region within the IC having an arbitrary shape and / or size. In a third example, a particular AOI may correspond to a functional module, layer, and / or other discrete part of the IC.

[0046] Machine learning model 208 may include (but is not limited to) multilayer perceptrons, recurrent neural networks (RNNs), convolutional neural networks (CNNs), Transformer neural networks, regression models, tree-based models, time series models, support vector regression (SVR) models, and / or other models capable of processing numerical and / or time series data. Predictions generated by a particular machine learning model may be supplemented and / or replaced by predictions generated through one or more other techniques, such as (but not limited to) physics-based models, control loops, simulations, and / or experiments. For example, for a specific region of interest (AOI), multiple temperature predictions generated by machine learning model 208, physics-based models, control loops, simulations, and / or experiments may be combined into an overall temperature prediction for that AOI through averaging, weighted combination, additional machine learning models, and / or other aggregation techniques.

[0047] Training engine 122 trains machine learning model 208 to generate temperature predictions 242 and / or control parameters 246 related to a region of interest (AOI). As shown in Figure 2, training engine 122 uses training data 200 containing training inputs 202 and training baseline ground truth values ​​204 to update the model parameters 206 of machine learning model 208. For example, training engine 122 can generate and / or retrieve training data 200 through one or more simulations, experiments, physics-based models, measurements, historical testing of integrated circuits (ICs), and / or other techniques. Training engine 122 can feed training inputs 202 related to one or more AOIs into one or more machine learning models 208. Training engine 122 can process the training inputs 202 using the model parameters 206 of the machine learning models and obtain training outputs 210 containing temperature predictions 242 and / or control parameters 246 of the AOI from one or more layers, modules, and / or components of the machine learning models. The training engine can compute one or more losses 254 (e.g., mean squared error, mean absolute error, Huber loss, etc.) between the training output 210 and the corresponding true training baseline 204 in the training data. Then, the training engine 122 can iteratively update the model parameters 206 of the machine learning model using training techniques (e.g., gradient descent and backpropagation) to reduce the losses 254.

[0048] Figure 3 shows exemplary training data for the machine learning model 208 in Figure 2 according to various embodiments. As shown in Figure 3, the training data 200 may include (but is not limited to) test configurations 234, sensor data 302, control characteristics 304, reference true temperature 306, and / or reference true control parameters 308 related to a specific test and / or IC.

[0049] Each test configuration 234 contains data related to certain types of tests for a specific integrated circuit (IC). For example, a test configuration may include (but is not limited to) one or more target temperatures, one or more temperature limits (e.g., test-based temperature limits, damage limits, damage "presets" that cause the test procedure to pause until a safe temperature is reached, etc.), AOIs associated with the test, measurements to be performed during the test, test steps to be performed during the test, equipment used in the test, and / or other numerical values ​​characterizing the test.

[0050] Sensor data 302 includes real and / or analog measurements of attributes associated with a particular IC. These attributes may include (but are not limited to) temperature (e.g., the location of one or more sensors on and / or around the IC), power (e.g., via a power monitor), aging, current, voltage, voltage drop, and / or other values ​​characterizing the IC's condition.

[0051] Control characteristic 304 contains data related to the progress of a specific test and / or the thermal behavior of the integrated circuit (IC) during the test. For example, control characteristic 304 may include (but is not limited to): the sum of temperature values ​​above and below target temperatures and / or limits, temperature slope, power slope, increase or decrease in temperature or power slope, minimum and / or maximum deviation of temperature (or other attribute) from the corresponding target and / or limit, warnings and / or alarms related to damage limits (or other types of events), test continuity signals, one or more safety offsets related to the target temperature, power value distribution for each AOI, and / or anomalous / outlier values ​​of temperature, power, and / or other attributes.

[0052] The reference true temperature 306 includes temperature values ​​for each region of interest (AOI) within the integrated circuit (IC). For example, each reference true temperature can be generated through simulation, experimentation, physics-based modeling, measurement, and / or other techniques. Each reference true temperature can also be associated with a corresponding test configuration, sensor data set 302, and / or control characteristics 304, so that the test configuration, sensor data 302, and / or control characteristics 304 can serve as features for machine learning models to predict the reference true temperature.

[0053] The benchmark true control parameter 308 includes values ​​related to temperature control and / or regulation within the integrated circuit (IC) and / or various areas of interest (AOIs) within the IC. For example, the benchmark true control parameter 308 may include (but is not limited to) target temperatures associated with temperature sensors inside and / or around the IC, target temperatures associated with specific AOIs within the IC, another thermal control signal for controlling and regulating the temperature of the IC and / or one or more portions of the IC, and / or any other type of input that can be used to control the thermal conditions of the IC. The benchmark true control parameter 308 may also include (or instead of) control inputs related to test execution, such as (but not limited to) signals related to adjusting test pacing (e.g., enabling cooling and / or temperature stabilization) and / or stopping the test (e.g., preventing IC damage). The benchmark true control parameter 308 may also include quality improvement data, statistics, root cause analysis of failures, and / or other data and / or test results that can be used to evaluate the quality of the integrated circuit (IC) and / or recorded for subsequent use. A specific set of baseline true control parameters 308 can be associated with a test configuration, a set of sensor data 302, a control characteristic 304, and / or a baseline true temperature 306, thereby enabling the test configuration, sensor data 302, control characteristic 304, and / or baseline true temperature 306 to serve as characterizations for machine learning models to predict these baseline true control parameters 308.

[0054] In some embodiments, test configuration 234, sensor data 302, and / or control characteristics 304 are used as training inputs 202, processed by a specific machine learning model to produce a corresponding training output 210. These training inputs 202 may also be associated with training baseline true values ​​204, including baseline true temperatures 306, baseline true control parameters 308, and / or other values ​​to be predicted from the training inputs 202. For example, a specific test configuration, set of sensor data 302, and / or set of control characteristics 304 may be input into a machine learning model and / or used to select a machine learning model trained to predict one or more corresponding baseline true temperatures 306. In another example, a specific test configuration, set of sensor data 302, set of control characteristics 304, and / or set of baseline true temperatures 306 may be input into a machine learning model and / or used to select a machine learning model trained to predict one or more corresponding baseline true control parameters 308.

[0055] In one or more embodiments, training engine 122 trains machine learning model 208 in one or more training phases to optimize machine learning model 208 based on various losses 254, training output 210 types, performance characteristics, and / or other objectives or factors. For example, training engine 122 may train a first set of one or more machine learning models 208 to predict a baseline true temperature 306 based on a corresponding test configuration 234, sensor data 302, and / or control characteristics 304. Training engine 122 may train a second set of one or more machine learning models 208 separately to predict a baseline true control parameter 308 based on a corresponding test configuration 234, sensor data 302, control characteristics 304, and / or baseline true temperature 306. Training engine 122 may also (or instead) train two sets of machine learning models 208 in an end-to-end manner to optimize control parameters 246 generated by the second set of machine learning models 208 based on temperature predictions 242 generated by the first set of machine learning models 208. The training engine 122 can also (or instead) train individual machine learning models 208 to generate temperature predictions 242 and control parameters 246 for different ICs and / or AOIs within the IC.

[0056] Returning to the discussion in Figure 2, after the training of the machine learning model 208 is completed, the training engine 122 can store the parameters and / or other representations of the trained machine learning model 208 in a database, cloud storage, distributed archive system and / or other types of data storage 212 for subsequent retrieval and use.

[0057] Execution engine 124 uses a trained machine learning model 208, physics-based models, simulations, experiments, control loops, and / or other techniques from training engine 122 and / or data storage 212 to generate temperature predictions 242 and control parameters 246 for controlling the temperature and / or other parameters of various IC tests. More specifically, execution engine 124 may collect inputs 240, which include (but are not limited to) temperature, power, aging, current, voltage, voltage drop, and / or other sensor data characterizing the IC's state. Inputs 240 may also include target temperature, temperature limits, AOI (region of interest), measured values, test steps, equipment used, and / or other test configuration attributes characterizing IC-specific tests. Input 240 may also include the sum of temperature values ​​above and below the target temperature and / or limit, temperature slope, power slope, changes in temperature or power slope, minimum and / or maximum deviation of temperature from the corresponding target and / or limit, warnings and / or alarms related to damage limits, test continuity signals, one or more safety offsets related to the target temperature, power value distribution for each AOI, outlier / abnormal attribute values ​​and / or other control characteristics characterizing test progress.

[0058] Execution engine 124 may use machine learning model 208 and / or other techniques to generate a temperature prediction 242 for the AOI based on input 240. Execution engine 124 may also use machine learning model 208 and / or other techniques to generate control parameters 246 for adjusting the temperature of the AOI and / or IC based on input 240 and / or temperature prediction 242. Execution engine 124 may also generate test outputs 248, which include: temperature prediction 242; control parameters 246; sensor data collected before, during, and / or after applying a specific set of control parameters 246 to the IC; control characteristics; test steps; and / or other data related to temperature prediction 242 and / or control parameters 246. These test outputs 248 may be stored in data storage 212 and / or provided as additional input 240 for use by machine learning model 208 and / or other techniques to generate new temperature predictions 242 and / or control parameters 246 for subsequent time steps in the test.

[0059] In one or more embodiments, the execution engine 124 runs on a separate computing unit independent of the test system used to perform IC testing and the thermal system used to regulate the temperature of the integrated circuit (IC) during IC testing. As further described below with reference to Figure 4, this separate computing unit can serve as an intermediate control point, decoupling the IC testing by the test system from the thermal control of the IC by the thermal system, while allowing the test system and the thermal system to influence each other during testing.

[0060] Figure 4 illustrates an example system 400 for performing integrated circuit (IC) testing according to various embodiments. Figure 4 As shown, system 400 includes thermal system 402, test system 404, and execution engine 124.

[0061] Test system 404 includes an automated test equipment (ATE) that tests the device under test (DUT) 406 (e.g., an integrated circuit) according to a series of test steps defined in test procedure 408. For example, test procedure 408 may include commands, control signals, and / or instructions executed by test system 404 to complete the test steps. While executing the test steps, sensors integrated into test system 404 can acquire sensor data 412, which includes (but is not limited to) temperature measurements, power consumption values, voltage readings, current measurements, and / or other attributes characterizing the operating state of the integrated circuit.

[0062] Execution engine 124 receives sensor data 412 from the device under test (DUT) 406, detailed information about the current test step from the test procedure 408, and / or other inputs 240 related to the DUT 406 and / or the test. Execution engine 124 processes these inputs 240 using a trained machine learning model 208, physics-based models, simulations, and / or other techniques, and generates temperature predictions 242 for various regions of interest (AOIs) within the DUT 406.

[0063] The execution engine 124 also generates control parameters 246 based on temperature prediction 242, test-related control characteristics, and / or other test-related data. Control parameters 246 may include target temperature for thermal conditioning, thermal control signals, adjustments to test steps or test operations, and / or other parameters that affect the thermal conditions of the device under test (DUT) 406.

[0064] The thermal system 402 receives control parameters 246 from the execution engine 124 and uses these control parameters 246 to control the temperature of the DUT 406. For example, the thermal system 402 may include heating elements, cooling systems, temperature controllers and / or other thermal controls 410, which can be used to regulate the thermal environment of the DUT 406.

[0065] As shown in Figure 4, the thermal system 402 can directly receive sensor data 412 from the test system 404 and use the sensor data 412 to adjust the thermal control 410 without using the control parameters 246 received from the execution engine 124. Similarly, the test system 404 can receive temperature prediction 242 and / or control parameters 246 from the execution engine 124, and / or receive thermal control 410 from the thermal system 402. The test system 404 can use the received values ​​to adjust the execution of specific test steps in the test procedure 408 (e.g., by adjusting the test pace, stopping the test, executing and / or omitting certain test steps, etc.).

[0066] As described above, the execution engine 124 can operate on a separate computing unit (e.g., a modular computer (COM), a single-board computer, etc.) independent of the test system 404 and the thermal system 402. This separate computing unit acts as an intermediate control point, decoupling the testing of the DUT 406 by the test system 404 from the thermal regulation of the DUT 406 by the thermal system 402. Simultaneously, the temperature prediction 242 and / or control parameters 246 generated by the execution engine 124 are based on test content related to: (i) the test program 408 and / or the test system 404; and (ii) thermal sensor data 412 reflecting the effect of thermal control 410 on the temperature of the DUT 406. This allows the test system 404 and the thermal system 402 to influence each other during the testing process. Furthermore, since the rate at which the execution engine 124 generates temperature prediction 242 and / or control parameters 246 is independent of the rate at which sensor data 412 is acquired, the thermal system 402, the test system 404, and / or other components can access thermal data (e.g., temperature prediction 242, control parameters 246, etc.) related to the DUT 406 on demand (e.g., avoiding "blind spots" where temperature and / or other sensors cannot generate sensor data 412).

[0067] Returning to the discussion in Figure 2, management engine 126 generates and updates records 214, which are used to track test outputs 248 and / or other types of data or entities related to a specific IC test. These records 214 may be stored in data storage 212 (e.g., one or more database tables, archives, etc.) and / or shared with training engine 122, execution engine 124, and / or other components involved in thermal prediction and conditioning during IC testing.

[0068] As shown in Figure 2, record 214 includes sorting data 220, chip-level data 222, package-level data 224, aging test data 226, final test data 228, and / or system-level data 230 collected at the corresponding stages of IC testing. Sorting data 220 may include conductivity, leakage current, voltage, frequency, thermal data, and / or data collected during wafer sorting testing. Therefore, sorting data 220 can reflect the thermal behavior of individual chips on the wafer before packaging.

[0069] Chip-level data 222 may include data generated during chip-level testing after the chip is separated from the wafer and before packaging. For example, chip-level data 222 may include temperature profiles, power consumption patterns, thermal resistance values, and other chip-specific thermal characteristics. Chip-level data 222 may also include data related to chip placement, orientation, physical characteristics, and / or other chip properties that affect thermal behavior during testing.

[0070] Package-level data 224 may include data collected after the chips are assembled into a package. For example, package-level data 224 may include package materials, thermal interfaces, package geometry, and / or other properties that affect the thermal response of the integrated circuit. Package-level data 224 may also include (or instead) case temperature, thermal resistance between the interface and the case, heat dissipation mode, thermal coupling between multiple chips in a multi-chip package, and / or other package-specific thermal characteristics.

[0071] Aging test data 226 may include data collected during aging tests of the packaged chip. For example, aging test data 226 may include temperature drift, power consumption changes, thermal stability metrics, and / or other thermal characteristics over time and / or under different stress conditions. Aging test data 226 may also include data related to the thermal cycling effects and temperature-related failure modes of the packaged chip.

[0072] Final test data 228 may include data collected during the final testing of the packaged chip. For example, final test data 228 may include temperature profiles related to various operating modes, power states, and / or test conditions.

[0073] System-level data 230 may include data collected during system-level testing of multiple integrated circuits integrated into a large system. For example, system-level data 230 may capture thermal interactions between different components, system-level thermal management requirements, and / or thermal performance under various operating conditions and / or in end-use applications. System-level data 230 may also include data related to thermal coupling effects, thermal diffusion, and system-level thermal control strategies.

[0074] In some embodiments, sorting data 220, chip-level data 222, package-level data 224, aging test data 226, final test data 228, and / or system-level data 230 include identification codes associated with the corresponding test stage (e.g., test stage identification codes, IC identification codes, etc.), inputs 240, temperature predictions 242, test parameters 246, and / or test outputs 248. Furthermore, for a specific IC, sorting data 220, chip-level data 222, package-level data 224, aging test data 226, final test data 228, and / or system-level data 230 can be stored in data storage 212 and used by management engine 126 to improve the testing or thermal conditioning of the IC at various test stages.

[0075] More specifically, management engine 126 can provide data collected in early testing phases as input 240 to machine learning model 208 and / or other components to generate temperature predictions 242, control parameters 246, and / or test outputs 248 in subsequent testing phases involving the same integrated circuits (ICs) and / or components. For example, management engine 126 can use sorting data 220 and / or chip-level data 222 for a specific chip as a set of reference thermal characteristics to generate temperature predictions 242, control parameters 246, and / or thermal responses for the same chip in subsequent testing phases. As another example, management engine 126 can use package-level data 224, aging test data 226, and / or final test data 228 for a specific package to provide data for system-level testing and / or thermal conditioning of the integrated system containing that package.

[0076] The management engine 126 may also use (or instead use) sorting data 220, chip-level data 222, package-level data 224, aging test data 226, final test data 228, and / or system-level data 230 in record 214 to generate and / or modify chip layout 232, test configuration 234, and / or test results 236 associated with a specific IC. As shown in Figure 2, the management engine 126 may provide a user interface 216 for viewing, searching, modifying, and / or otherwise accessing record 214. The user interface 216 may also be used (or instead of otherwise) to generate, view, and / or modify chip layout 232 in a packaged and / or integrated system based on the thermal response of the corresponding chip (e.g., represented by temperature prediction 242, control parameters 246, and / or other thermal data in chip-related sorting data 220 and / or chip-level data 222). User interface 216 can also be used (or alternatively) to set and / or adjust the thermal limits and / or other properties of test configuration 234 for one or more test phases based on the thermal response determined in previous test phases involving the same IC. User interface 216 can also be used (or alternatively) to view and / or compare test results 236 for the same IC and / or different ICs in test phases. Therefore, management engine 126 and / or user interface 216 can allow users and / or other entities involved in the design and / or testing of wafers, chips, packages, integrated systems and / or other IC-based systems or components to optimize the various phases of the design and testing of IC-based systems or components.

[0077] Figure 5 is a flowchart of method steps for performing integrated circuit testing according to various embodiments. Although the method steps are described in conjunction with the system shown in Figures 1-2 and 4, those skilled in the art will understand that any system configured to perform some or all of the method steps in any order is within the scope of this disclosure.

[0078] As shown in the figure, in step 502, training engine 122 collects training data, including sensor data, temperature, test parameters, control characteristics, and / or control parameters related to the testing of one or more integrated circuits (ICs). For example, training engine 122 can generate and / or retrieve training data through one or more simulation, experimentation, physics-based modeling, measurement, historical testing of ICs, and / or other techniques.

[0079] In step 504, training engine 122 uses training data to train one or more machine learning models to predict the temperature and / or control parameters of one or more regions of interest (AOIs) within an integrated circuit (IC). For example, training engine 122 may train a first machine learning model to predict the temperature of a specific AOI within a particular IC based on inputs such as sensor data collected during IC testing. Training engine 122 may also train a second machine learning model to predict target temperatures, thermal control signals, and / or other types of thermal control parameters associated with that AOI and / or IC, based on the predicted temperature output by the first machine learning model (or other techniques) and / or control characteristics calculated during IC testing.

[0080] In step 506, execution engine 124 generates one or more temperature predictions for each region of interest (AOI) in the integrated circuit (IC) based on sensor data collected during a test instance corresponding to the test. For example, execution engine 124 may run on a separate computing unit during test instance execution. While collecting sensor data, execution engine 124 may periodically and / or continuously generate temperature predictions for each AOI using one or more machine learning models trained through steps 502 and 504 and / or other techniques. When it is necessary to generate temperature predictions for multiple AOIs and / or multiple ICs on the IC simultaneously, execution engine 124 may generate these temperature predictions in parallel (e.g., on multiple processors, processor cores, etc.). Each temperature prediction may represent the “true silicon temperature” at the corresponding AOI, rather than the temperature measured by a sensor that may be located outside the AOI.

[0081] In one embodiment, execution engine 124 generates temperature predictions using additional data related to the integrated circuit (IC). For example, execution engine 124 may use additional machine learning models, physics-based models, simulations, and / or other techniques trained in steps 502 and 504 to predict the temperature of each region of interest (AOI) within the IC based on parameters including: sensor data, known and / or simulated physical characteristics of the integrated circuit (IC), statistics related to the integrated circuit (IC) and / or IC testing, existing test results related to the integrated circuit (IC) (e.g., from historical testing and / or early testing phases of the integrated circuit (IC)), test parameters related to testing, manufacturing data related to the integrated circuit (IC), and / or other applicable data related to the integrated circuit (IC) and / or testing.

[0082] In step 508, execution engine 124 determines one or more control parameters based on temperature predictions and one or more control characteristics related to the test instance. For example, execution engine 124 may use one or more machine learning models trained through steps 502 and 504 and / or other techniques to generate target temperatures, thermal control signals, commands, and / or other types of numerical values ​​that can be used to control the operation of the thermal system controlling the IC temperature during the test instance and / or the operation of the test system executing the test instance.

[0083] In step 510, execution engine 124 executes the test instance according to control parameters. For example, execution engine 124 may transmit control parameters to a thermal system and / or a test system, and / or otherwise generate outputs that propagate control parameters to the thermal system and / or the test system. The thermal system may use the control parameters to control the temperature of the integrated circuit (IC) and / or various parts of the IC (e.g., to bring the temperature to and / or maintain a target temperature, to avoid exceeding temperature limits, etc.). The test system may use the control parameters to change the execution pace of the test instance, pause the test instance, modify the test steps executed in the test instance, and / or otherwise adjust the execution of the test instance. Execution engine 124 and / or management engine 126 may also (or instead) generate and / or record test results and / or other data related to the test instance based on temperature predictions, control parameters, and / or other data collected and / or generated during the test instance.

[0084] In step 512, execution engine 124 determines whether the test instance should continue execution. For example, execution engine 124 may determine that the test instance should continue execution until all steps in the test have been completed, the test result has been determined, a certain amount of time has elapsed, and / or other conditions are met.

[0085] When execution engine 124 determines that the test instance needs to continue execution, it repeats steps 506, 508, and 510 to generate temperature predictions, control parameters, and / or other outputs that can be used to predict and / or regulate the temperature of the integrated circuit (IC). Additionally, execution engine 124 also periodically and / or continuously executes step 512 to determine whether to continue executing the test instance.

[0086] Once execution engine 124 determines in step 512 that the execution of the test instance should not continue, management engine 126 executes step 514, storing the temperature predictions, control parameters, and / or other thermal data generated during the test instance into one or more records associated with that test instance. For example, management engine 126 may store in the records a mapping between thermal data and one or more identifiers of the test instance and / or IC. These records may then be stored in a data store; linked to additional data collected in other testing phases involving the IC; used to adjust subsequent thermal control and / or testing of the IC; used to generate and / or modify the design of the package and / or integrated system containing the IC; and / or used to perform other tasks related to thermal conditioning, testing, and / or IC design.

[0087] In summary, the disclosed technology enables the prediction and adjustment of the temperature of individual regions of interest (AOIs) within an integrated circuit during integrated circuit testing. Each AOI may include (but is not limited to) two-dimensional (2D) and / or three-dimensional (3D) mesh cells, functional modules, regions of arbitrary shapes, and / or other types of regions or zones within the integrated circuit.

[0088] More specifically, the disclosed technology utilizes machine learning models, physics-based models, simulations, and / or other techniques to generate temperature predictions for each region of interest (AOI) within an integrated circuit, based on sensor data, test parameters, and / or other inputs collected during integrated circuit testing. Each temperature prediction can represent the “true silicon temperature” at the corresponding AOI, rather than the temperature measured by one or more temperature sensors located outside the AOI and / or at different distances from the AOI.

[0089] Temperature prediction is also used in conjunction with machine learning models, physics-based models, simulations, and / or other techniques to generate control parameters for regulating the temperature within the region of interest (AOI) during integrated circuit testing. These control parameters may include, but are not limited to, target temperature, thermal control signals for the thermal system, control signals for adjusting test pacing, and / or control signals for initiating or stopping test operations. The determination of control parameters may be based on temperature prediction and / or test-related control characteristics, such as, but not limited to, temperature limits, temperature slopes, power slopes, safety offsets, and / or other numerical values ​​characterizing test progress and / or conditions.

[0090] Temperature prediction and / or control parameters can be generated by a separate computing unit (e.g., a modular computer (COM), a single-board computer, etc.). This computing unit operates independently of and concurrently with the thermal and test systems (e.g., automated test equipment (ATE), test procedures, etc.). The thermal system uses these control parameters to thermally control the integrated circuit (IC), while the test system tests the IC. This separate computing unit acts as an intermediate control point, decoupling the test and thermal systems while allowing the operation of each system to influence each other. Furthermore, this separate computing unit allows the thermal system, test system, and / or other components to access the IC-related temperature on demand (e.g., there is no "blind spot" where temperature sensors are unavailable).

[0091] Furthermore, temperature prediction and control can be performed at different stages of integrated circuit testing, such as (but not limited to) wafer sorting testing, chip-level testing, aging testing, final testing, and / or system-level testing. Data collected in early testing phases can be tracked and used for thermal prediction and control of the same integrated circuit in subsequent phases. This data can also be used to design, evaluate, and / or modify the chip layout and / or packaging involved in this integrated circuit, test configuration, and / or other parameters related to the testing and / or use of this integrated circuit.

[0092] Compared to existing technologies, one of the advantages of the disclosed technology is its ability to accurately and timely monitor and control the temperature of each area of ​​interest (AOI) within an integrated circuit (IC). Therefore, the disclosed technology can avoid and / or mitigate thermal damage and / or failure caused by the limited temperature monitoring capabilities of temperature sensors and / or their inability to quickly respond to sudden changes in heat generated during IC testing. Furthermore, by incorporating thermal data collected in earlier testing phases into thermal prediction and control in subsequent testing phases (involving the same IC), the disclosed technology can improve the design and / or thermal management of complex ICs, packages, and / or multi-component integrated systems with different thermal responses. These advantages provide one or more technological improvements superior to existing technologies.

[0093] For ease of explanation, the description of various embodiments herein is not intended to be exhaustive, nor is it limited to the disclosed embodiments. Many modifications and variations can be made to the invention by those skilled in the art without departing from the scope and spirit of the described embodiments.

[0094] Clause 1. In various embodiments, a computer-implemented method for performing integrated circuit testing includes: generating one or more temperature predictions for a region of interest within the integrated circuit based on sensor data associated with the integrated circuit; determining one or more control parameters related to testing the integrated circuit based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit; and causing the testing to be performed on the integrated circuit based on the one or more control parameters.

[0095] Clause 2. The computer-implemented method as described in Clause 1 includes: generating one or more additional temperature predictions and one or more additional control parameters based on the results of the test; and causing additional tests to be performed on the integrated circuit based on the one or more additional control parameters.

[0096] Clause 3. The computer implementation method as described in Clause 1 or 2, wherein the test is related to the wafer or chip associated with the integrated circuit, and the additional test is related to the package associated with the integrated circuit.

[0097] Clause 4. A computer implementation method as described in any one of Clauses 1-3, wherein causing the test to be performed on the integrated circuit based on the one or more control parameters includes controlling the temperature associated with the integrated circuit using the one or more control parameters.

[0098] Article 5. The computer implementation method as described in any one of Articles 1-4, wherein the one or more temperature predictions are generated by a first machine learning model, and the one or more control parameters are generated by a second machine learning model.

[0099] Article 6. The computer implementation method as described in any one of Articles 1-5, wherein the one or more temperature predictions and the one or more control parameters are generated based on at least one of a machine learning model, a physics-based model, a control loop, a simulation, or an experiment.

[0100] Article 7. The computer implementation method as described in any one of Articles 1-6, wherein the one or more control parameters include at least one of a target temperature, a control signal for a thermal system related to the test of the integrated circuit, a control signal related to the pacing of the test, or a control signal related to starting or stopping the test.

[0101] Article 8. The computer implementation method as described in any one of Articles 1-7, wherein the one or more control characteristics include at least one of temperature limit, maximum deviation from temperature limit, minimum deviation from temperature limit, temperature slope, temperature integral, temperature deviation sum, power slope, safety offset, power value distribution, or one or more outliers.

[0102] Article 9. The computer implementation method as described in any one of Articles 1-8, wherein the sensor data includes at least one of temperature, power, age, current, voltage, or voltage drop.

[0103] Article 10. The computer implementation method as described in any one of Articles 1-9, wherein the one or more temperature predictions are generated based on at least one of the following: physical characteristics of the integrated circuit, statistical data related to the integrated circuit, test results related to the integrated circuit, test parameters related to the test, or manufacturing data related to the integrated circuit.

[0104] Article 11. In one or more embodiments, one or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause one or more processors to perform the following steps: generating one or more temperature predictions for a region of interest within an integrated circuit based on sensor data associated with the integrated circuit; determining one or more control parameters related to testing the integrated circuit based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit; and adjusting one or more thermal controls related to the testing based on the one or more control parameters.

[0105] Article 12. One or more non-transitory computer-readable media as described in Article 11, wherein the instructions also cause the one or more processors to perform the following steps: determining (i) a set of reference true temperatures associated with the region of interest and (ii) a set of reference true control parameters associated with the integrated circuit; and training one or more machine learning models using the set of reference true temperatures and the set of reference true control parameters, wherein the one or more machine learning models are subsequently used to generate the one or more temperature predictions or the one or more control parameters.

[0106] Article 13. One or more non-transitory computer-readable media as described in Article 11 or 12, wherein training the one or more machine learning models comprises: generating a predicted temperature of the region of interest based on input including training sensor data and training test procedure data associated with the integrated circuit by executing a machine learning model included in the one or more machine learning models; and training the machine learning model based on one or more losses calculated using the predicted temperature and corresponding reference true temperatures included in the reference true temperature set.

[0107] Article 14. One or more non-transitory computer-readable media as described in any one of Articles 11-13, wherein training the one or more machine learning models comprises: generating predictive control parameters based on an input including a predicted temperature of the region of interest by executing a machine learning model included in the one or more machine learning models; and training the machine learning model based on one or more losses calculated using the predictive control parameters and corresponding benchmark true control parameters included in the benchmark true control parameter set.

[0108] Article 15. One or more non-transitory computer-readable media as described in any one of Articles 11-14, wherein the instructions also cause the one or more processors to perform the following steps: generating one or more additional temperature predictions and the one or more additional control parameters based on the one or more temperature predictions and the one or more control parameters; and adjusting the one or more thermal controls based on the one or more additional control parameters.

[0109] Clause 16. One or more non-transitory computer-readable media as described in any one of Clauses 11-15, wherein the one or more thermal controls are adjusted during the test or subsequent test of the integrated circuit.

[0110] Article 17. One or more non-transitory computer-readable media as described in any one of Articles 11-16, wherein the one or more temperature predictions and the one or more control parameters are generated based on at least one of a machine learning model, a physics-based model, a control loop, a simulation, or an experiment.

[0111] Article 18. One or more non-transitory computer-readable media as described in any one of Articles 11-17, wherein the sensor data is collected by one or more sensors located outside the region of interest.

[0112] Article 19. One or more non-transitory computer-readable media as described in any one of Articles 11-18, wherein the region of interest includes a two-dimensional or three-dimensional region within the integrated circuit.

[0113] Article 20. In one or more embodiments, a system includes: one or more memories storing instructions, and one or more processors coupled to the one or more memories, and configured, when executing the instructions, to perform the following steps: generating one or more temperature predictions for the region of interest within the integrated circuit based on sensor data associated with one or more locations located outside the region of interest in the integrated circuit; determining one or more control parameters related to testing the integrated circuit based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit; and causing the testing to be performed on the integrated circuit based on the one or more control parameters.

[0114] Any claim element and / or any combination of any elements described in any claim of this application, in whatever manner, is within the scope of this embodiment and protection.

[0115] For ease of explanation, the description of various embodiments herein is not intended to be exhaustive, nor is it limited to the disclosed embodiments. Many modifications and variations can be made to the invention by those skilled in the art without departing from the scope and spirit of the described embodiments.

[0116] The various aspects of this embodiment can be embodied as a system, method, or computer program product. Therefore, the various aspects of this disclosure can take the form of a completely hardware embodiment, a completely software embodiment (including firmware, resident software, microcode, etc.), or a combination of hardware and software embodiments, all of which are generally referred to herein as a "module," a "system," or a "computer." Furthermore, any hardware and / or software technology, process, function, component, engine, module, or system described in this disclosure can be implemented as a circuit or a set of circuits. Moreover, the various aspects of this disclosure can take the form of a computer program product embodied in one or more computer-readable media having computer-readable program code embodied thereon.

[0117] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, such as, but not limited to, any one or more of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer floppy disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or USB flash drive), optical fiber, USB flash drive read-only memory (CD-ROM), any of the above-mentioned memory or any of these devices. In this document, a computer-readable storage medium may be any tangible medium capable of containing or storing a program for use by or associated with an instruction execution system, device, or apparatus.

[0118] Various aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams illustrating methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each module and combination of modules in the flowchart illustrations and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to produce a machine. When these instructions are executed by the processor of the computer or other programmable data processing apparatus, the functions / operations specified in the flowchart illustration and / or block diagram modules are implemented. Such processors may be, but are not limited to, general-purpose processors, special-purpose processors, dedicated processors, or field-programmable gate arrays.

[0119] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, program code segment, or portion of program code, containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions shown in the blocks may differ from the order shown in the figures. For example, two blocks shown consecutively in a figure may actually execute substantially simultaneously, or, depending on the functions involved, these blocks may sometimes execute in reverse order. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of each block in the block diagrams and / or flowcharts, can be implemented via a dedicated hardware system that performs the specified function or operation, or a combination of dedicated hardware and computer instructions.

[0120] While the foregoing description pertains to embodiments of this disclosure, other and further embodiments of this disclosure may be devised without departing from the basic scope of this disclosure, the scope of which is defined by the following claims.

Claims

1. A computer-based method for performing integrated circuit testing, characterized in that, include: Based on sensor data associated with an integrated circuit, generate one or more temperature predictions for a region of interest within the integrated circuit; Based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit, determine one or more control parameters related to the testing of the integrated circuit; and Based on the one or more control parameters, the test is performed on the integrated circuit.

2. The computer implementation method as described in claim 1, characterized in that, include: Based on the results of the test, generate one or more additional temperature predictions and one or more additional control parameters; as well as Based on the one or more additional control parameters, additional tests are performed on the integrated circuit.

3. The computer implementation method as described in claim 2, characterized in that, The tests are related to the wafer or chip associated with the integrated circuit, and the additional tests are related to the package associated with the integrated circuit.

4. The computer implementation method as described in claim 1, characterized in that, Performing the test on the integrated circuit based on the one or more control parameters includes controlling the temperature associated with the integrated circuit using the one or more control parameters.

5. The computer implementation method as described in claim 1, characterized in that, The one or more temperature predictions are generated by a first machine learning model, and the one or more control parameters are generated by a second machine learning model.

6. The computer implementation method as described in claim 1, characterized in that, The one or more temperature predictions and the one or more control parameters are generated based on at least one of a machine learning model, a physics-based model, a control loop, a simulation, or an experiment.

7. The computer implementation method as described in claim 1, characterized in that, The one or more control parameters include at least one of the following: target temperature, control signal of the thermal system related to the test of the integrated circuit, control signal related to the pace of the test, or control signal related to starting or stopping the test.

8. The computer implementation method as described in claim 1, characterized in that, The one or more control characteristics include at least one of the following: temperature limit, maximum deviation from temperature limit, minimum deviation from temperature limit, temperature slope, temperature integral, temperature deviation sum, power slope, safety offset, power value distribution, or one or more outliers.

9. The computer implementation method as described in claim 1, characterized in that, The sensor data includes at least one of temperature, power, age, current, voltage, or voltage drop.

10. The computer implementation method as described in claim 1, characterized in that, The one or more temperature predictions are generated based on at least one of the following: physical characteristics of the integrated circuit, statistical data related to the integrated circuit, test results related to the integrated circuit, test parameters related to the test, or manufacturing data related to the integrated circuit.

11. One or more non-transitory computer-readable media, characterized in that, It stores instructions that, when executed by one or more processors, cause the processors to perform the following steps: Based on sensor data associated with an integrated circuit, generate one or more temperature predictions for a region of interest within the integrated circuit; Based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit, determine one or more control parameters related to the testing of the integrated circuit; as well as Based on the one or more control parameters, adjust one or more thermal controls related to the test.

12. The one or more non-transitory computer-readable media as claimed in claim 11, characterized in that, The instructions also cause the one or more processors to perform the following steps: Determine (i) a set of reference true temperatures associated with the region of interest and (ii) a set of reference true control parameters associated with the integrated circuit; and One or more machine learning models are trained using the benchmark true temperature set and the benchmark true control parameter set, wherein the one or more machine learning models are then used to generate the one or more temperature predictions or the one or more control parameters.

13. One or more non-transitory computer-readable media as claimed in claim 12, characterized in that, Training one or more machine learning models includes: By executing a machine learning model included in one or more of the machine learning models, a predicted temperature of the region of interest is generated based on input including training sensor data, and test program data related to the integrated circuit is trained; and The machine learning model is trained using the predicted temperature and one or more losses calculated using the corresponding benchmark true temperatures included in the benchmark true temperature set.

14. One or more non-transitory computer-readable media as claimed in claim 12, characterized in that, Training one or more machine learning models includes: By executing a machine learning model included in one or more of the machine learning models, predictive control parameters are generated based on inputs including the predicted temperature of the region of interest; and The machine learning model is trained using the predicted control parameters and one or more losses calculated using the corresponding benchmark true control parameters included in the benchmark true control parameter set.

15. One or more non-transitory computer-readable media as claimed in claim 11, characterized in that, The instructions also cause the one or more processors to perform the following steps: Based on the one or more temperature predictions and the one or more control parameters, generate one or more additional temperature predictions and the one or more additional control parameters; as well as The one or more thermal controls are adjusted based on the one or more additional control parameters.

16. The one or more non-transitory computer-readable media as claimed in claim 15, characterized in that, The one or more thermal controls are adjusted during the test or subsequent testing of the integrated circuit.

17. One or more non-transitory computer-readable media as claimed in claim 11, characterized in that, The one or more temperature predictions and the one or more control parameters are generated based on at least one of a machine learning model, a physics-based model, a control loop, a simulation, or an experiment.

18. One or more non-transitory computer-readable media as claimed in claim 11, characterized in that, The sensor data is collected by one or more sensors located outside the region of interest.

19. One or more non-transitory computer-readable media as claimed in claim 11, characterized in that, The region of interest includes a two-dimensional or three-dimensional region within the integrated circuit.

20. A system, characterized in that, include: One or more memory locations storing instructions, and One or more processors coupled to the one or more memory locations, and configured to perform the following steps when the instructions are executed: Based on sensor data related to one or more locations within the integrated circuit that are outside the region of interest, one or more temperature predictions are generated for the region of interest within the integrated circuit. Based on the one or more temperature predictions and one or more control characteristics associated with the integrated circuit, determine one or more control parameters related to the testing of the integrated circuit; as well as Based on the one or more control parameters, the test is performed on the integrated circuit.

Images