CN112650128A - Automatic protection method and device for chip overheating - Google Patents
Automatic protection method and device for chip overheating Download PDFInfo
- Publication number
- CN112650128A CN112650128A CN202011589421.4A CN202011589421A CN112650128A CN 112650128 A CN112650128 A CN 112650128A CN 202011589421 A CN202011589421 A CN 202011589421A CN 112650128 A CN112650128 A CN 112650128A
- Authority
- CN
- China
- Prior art keywords
- control
- chip
- temperature
- automatic
- decision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0428—Safety, monitoring
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24024—Safety, surveillance
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Control Of Temperature (AREA)
Abstract
The invention provides a chip overheating automatic protection method and a device, which comprises the steps of receiving output values of a plurality of temperature sensors, wherein the temperature sensors are respectively positioned at different positions in a chip, and each temperature sensor has a corresponding control threshold; generating an interrupt signal when the output value of at least one temperature sensor is higher than the corresponding control threshold value; forming a decision for the interrupt signal to generate a decision control result; and triggering a corresponding automatic control state machine according to the decision control result to automatically adjust the temperature of the chip. The invention can quickly and effectively execute hardware response and control, so that the chip can be recovered to a normal temperature range in a short time, and a plurality of control strategies are provided to reduce the load of the chip.
Description
Technical Field
The invention belongs to the field of chip protection, and particularly relates to a method and a device for automatically protecting a chip from overheating.
Background
In the operation process of computer and other electronic equipment, various chips belong to heat-generating devices. The excessive temperature generated by the chip can bring great influence on the work of the chip, which can cause the abnormal work of the chip, shorten the normal service life of the chip, and burn the chip under severe conditions, so the overheating protection strategy and mechanism of the chip are the problems that the chip must be considered when working.
At present, there are mainly external means and internal means for reducing the temperature of the chip. External means, namely, introducing cooling means into the board-level system, for example, adding a heat dissipation patch, a fan, a liquid cooling system and the like into the board-level system; the internal means reduces the working load of the chip to reduce or reduce the heat productivity of the chip and make the chip return to the normal working temperature range. For the method for intervening the working load of the chip, conventionally, the method mainly relies on a temperature sensor to initiate overheating early warning to a main control CPU in the chip, and the main control CPU enters an interrupt response service program to process and reduce the load of the chip after receiving corresponding overheating early warning interrupt, so that the following defects cannot be avoided: firstly, the response time is long, and because the CPU needs to enter the interrupt service program for processing, the response to the interrupt response priority is delayed and the software processing is delayed; secondly, the CPU load is high, and due to the natural defects of software response and CPU intervention, the interrupted response raises the CPU load in a short time, so that the situation of sudden CPU load rise has to be faced when the load needs to be reduced.
Disclosure of Invention
The invention aims to provide a chip overheating protection method for actively discovering and interfering in the working load of a chip, which properly and actively and automatically reduces the chip load on the premise of not influencing the normal work of the chip, so that the normal work of the chip is not interfered, and the problem of reducing the temperature in a chip under the condition of reducing the load is solved.
The invention provides a chip overheating automatic protection method in a first aspect, which comprises the following steps:
receiving output values of a plurality of temperature sensors, wherein the plurality of temperature sensors are respectively positioned at different positions in the chip, and each temperature sensor has a corresponding control threshold value;
generating an interrupt signal when the output value of at least one temperature sensor is higher than the corresponding control threshold value;
forming a decision for the interrupt signal to generate a decision control result;
and triggering a corresponding automatic control state machine according to the decision control result to automatically adjust the temperature of the chip.
Preferably, each said temperature sensor has a respective level m +1 control threshold; wherein m is a positive integer greater than 1;
when the output value of at least one temperature sensor is higher than the corresponding control threshold value, an interrupt signal is generated, further comprising:
generating a first interrupt signal when an output value of the temperature sensor is higher than a first control threshold and lower than a second control threshold;
generating an ith interrupt signal when an output value of the temperature sensor is higher than the ith control threshold and lower than the (i +1) th control threshold; wherein i is a positive integer and i belongs to [1, m +1 ];
wherein the decision control results of the first to m +1 th interrupt signals are used to trigger automatic control state machines of different levels.
Preferably, when the automatic control state machine receives a plurality of control requests at the same time, the automatic control state machine performs priority ranking according to the levels of the plurality of control requests, and executes control responses according to the ranked control requests.
Preferably, after triggering the corresponding automatic control state machine according to the decision control result and automatically adjusting the temperature of the chip, the method further includes:
and detecting whether the temperature of the chip is normal within a predefined time, if so, exiting the automatic control of the automatic control state machine, otherwise, re-triggering the automatic control of the automatic control state machine, and continuously adjusting the temperature of the chip.
Preferably, after the automatically adjusting the temperature of the chip, the method further comprises:
and executing interrupt clearing control, and executing delay waiting operation based on an interrupt clearing time window preset in a timer so that the interrupt signal is cleared.
The present invention provides in a second aspect an automatic protection device for chip overheating, comprising:
the temperature sensors are respectively positioned at different positions in the chip and used for sensing and outputting temperature values of the positions, and each temperature sensor has a corresponding control threshold;
a threshold detector for generating an interrupt signal when an output value of at least one temperature sensor is higher than a corresponding control threshold;
the decision controller is used for forming a decision for the interrupt signal and generating a decision control result;
and the automatic control state machine is used for responding to the decision control result and triggering automatic adjustment of the temperature of the chip.
Preferably, each said temperature sensor has a respective level m +1 control threshold; wherein m is a positive integer greater than 1;
the threshold detector is further configured to:
each temperature sensor has a corresponding m +1 level control threshold; wherein m is a positive integer greater than 1;
when the output value of at least one temperature sensor is higher than the corresponding control threshold value, an interrupt signal is generated, further comprising:
generating a first interrupt signal when an output value of the temperature sensor is higher than a first control threshold and lower than a second control threshold;
generating an ith interrupt signal when an output value of the temperature sensor is higher than the ith control threshold and lower than the (i +1) th control threshold; wherein i is a positive integer and i belongs to [1, m +1 ];
wherein the decision control results of the first to m +1 th interrupt signals are used to trigger automatic control state machines of different levels.
The decision controller is further configured to:
and forming a decision for the first to the (m +1) th interrupt signals, wherein the generated decision control result is used for triggering automatic control state machines of different levels.
Preferably, the apparatus further comprises a priority controller for, when the automatic control state machine receives a plurality of control requests simultaneously, performing priority sorting according to levels of the plurality of control requests and executing control responses according to the sorted control requests.
Preferably, the automatic control state machine is further configured to:
after the temperature of the chip is automatically adjusted, whether the temperature of the chip is normal is detected within a predefined time, if the temperature of the chip is normal, the automatic control of the automatic control state machine is quitted, otherwise, the automatic control of the automatic control state machine is triggered again, and the temperature of the chip is continuously adjusted.
Preferably, the automatic control state machine is further configured to:
after the temperature of the chip is automatically adjusted, interrupt clear control is performed, and a delay waiting operation is performed through an interrupt clear time window preset in a timer, so that the interrupt signal is cleared.
Compared with the prior art, the invention has the following advantages:
the chip overheating automatic protection method and device provided by the invention directly intervenes the chip overheating condition through sensor data, quickly and effectively executes hardware response and control, shortens the temperature control response time, and simultaneously provides a plurality of control strategies to reduce the chip working load, so that the chip can be recovered to the normal working temperature range in a short time.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 shows a block diagram of a chip overheat protection device according to the present invention.
Fig. 2 shows a schematic diagram of an implementation process of the automatic control state machine according to the invention.
Fig. 3 shows a flow chart of a chip overheating protection method according to the present invention.
Fig. 4 shows a schematic diagram of an implementation of an automatic state machine without interrupt clearing according to the invention.
Fig. 5 shows a schematic diagram of an implementation of an automatic state machine for interrupt-free clearing and control of time according to the invention.
Fig. 6 shows a schematic diagram of an implementation process of the downconversion control automatic state machine according to the present invention.
Fig. 7 shows a schematic diagram of temperature sensor threshold setting according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The chip overheating automatic protection scheme of the invention utilizes a temperature sensor network which is pre-arranged in a chip to monitor the place which is easy to heat in the chip in real time, obtains the temperature dynamic change conditions of different positions in the chip at the same time, forms the combined control of the temperature monitoring network after the register configuration, then forms the control decision process, finally generates the corresponding temperature overheating early warnings of different levels, and triggers the automatic control of different levels based on the early warnings. For example, when the temperature is too high, the chip load is changed in real time according to different results to reduce the temperature of the chip, so that the normal work of the chip is ensured.
The specific implementation architecture of the preferred embodiment of the present invention is shown in fig. 1, and an entire array of temperature sensors is used to perform real-time temperature monitoring on the interior of a chip, compare different temperature thresholds, and generate interrupts corresponding to the different temperature thresholds according to the comparison result, so that decision inputs are formed by using the interrupts, processing mechanisms of different levels are generated by using a decision control algorithm, and finally, a corresponding control state machine is triggered by an enable controller to generate automatic adjustment of the chip load, and the corresponding interrupts are cleared within a preset time window after the automatic adjustment. The internal structure of the chip overheat protection device shown in fig. 1 is described in detail below.
Register C0 generates specific configuration information and control information for other components. The invention is mainly configured by a corresponding configuration bus, and configuration information and control information are generated by pre-configuring the register by a CPU;
the temperature sensor array arranged at different positions in the chip comprises a temperature sensor 0, a temperature sensor 1, … and a temperature sensor n which are respectively denoted as C1.0, C1.1, … and C1.n and are used for monitoring the temperature condition in the chip in real time. The specific number n of the temperature sensors is determined according to the conditions of the chip, including the conditions of the chip such as the position of a hot spot, the process characteristics, the size and the like, and the balance is comprehensively achieved. Preferably, the temperature sensor is a digital temperature sensor, that is, the output result is a digital vector signal after analog-to-digital conversion, the output result has a specific corresponding relationship with the temperature inside the chip, and the temperature condition of the position inside the chip corresponding to the sensor can be obtained by using the output result;
a plurality of threshold detectors C2.0, C2.1, …, C2.n threshold the output of the respective temperature sensors and generate interrupts. Preferably, the same temperature sensor can set a plurality of thresholds and generate a plurality of interrupts corresponding to different thresholds for distinguishing different temperature intervals, so that different chip overheating degrees can be obtained by most directly utilizing the different thresholds. Storing a plurality of threshold items C3 in the threshold value detector, wherein the threshold items are obtained by setting values of registers; for each threshold value item, a comparator C4 and an interrupt generator C5 are arranged in the threshold value detector corresponding to the threshold value item, and are used for judging whether the output result of the corresponding temperature sensor in the chip exceeds a given threshold value, and if the output result exceeds the given threshold value, the temperature of the corresponding position of the corresponding temperature sensor in the chip exceeds the temperature corresponding to the threshold value.
When the comparator C4 outputs the comparison result of the threshold, the interrupt generator C5 generates the interrupt by sampling, and provides the necessary interfaces for interrupt control, such as interrupt setting, interrupt masking, interrupt status, interrupt clearing, etc., so that the software can perform the necessary management and control, and the interrupt can be automatically cleared by the hardware within the set time after the hardware generates the automatic control.
The decision controller C6 provides an internal over-temperature protection algorithm for the chip and forms a decision control for making an algorithm-specific decision on the interrupts generated by the threshold detection at different temperature thresholds associated with different locations, and finally generates control results at different response levels. The specific decision control algorithm can be realized by adopting simple decision control or complex decision control, and is determined according to the requirement. For example, a simple decision control may simply classify, combine and output the interrupt results corresponding to the threshold values. For example, if the response level is defined as 4 levels, and the result of each level corresponds to a different hardware automatic processing mode, the threshold of each temperature sensor is also set to 4 levels, each level also corresponds to a different temperature threshold, and finally, the decision control only needs to simply execute or output the interrupt of each level, which indicates that the corresponding control is adopted only if any one temperature sensor generates an interrupt exceeding the specified temperature threshold.
Preferably, the decision controller C6 further includes a multi-level decision node for performing complex decision control, performing multi-level decision after combining the input interrupts with threshold classification, and finally generating response outputs of different levels. As shown in fig. 1, each stage of decision may take the previous output result into consideration, and make decisions using different parameters.
Referring to fig. 1, the multi-stage decision nodes specifically include m + 10 th-stage decision nodes C7.0, C7.1, …, C7.m, and are configured to classify input (n +1) × (m +1) interrupts according to a predefined algorithm and form a primary decision according to a threshold level, where n +1 is the number of temperature sensors, and m +1 is the number of thresholds; k +1 level 1 decision nodes C8.0, C8.1, …, C8.k, for forming a decision on the input (m +1) × (k +1) results according to a predefined algorithm, k +1 being the number of previous level decision outputs. Similarly, the decision controller C6 sequentially includes i +1 stages of decision nodes, i > 1, where each stage of decision node forms a decision on the output result of the previous stage. As a final stage, the decision controller may include n +1 i-th stage decision nodes C9.0, C9.1, …, C9.n, form a decision on the output result of the i-th stage according to a predefined algorithm, and finally output n +1 decision results.
It should be noted that the connection relationship between each decision node in each stage and the decision nodes in the preceding and following stages is determined only by the decision result, and theoretically, any corresponding relationship may be presented. That is, the numbers m, k, …, n, etc. are all independent positive integers, and the numbers thereof may be all the same, partially the same, or all different, and may be arbitrarily set by those skilled in the art according to system requirements.
The complex decision control has the advantages that the algorithm can be adjusted according to different conditions, and the control precision and effect are better. To further reduce complexity, in a preferred embodiment of the present invention, each stage is implemented with simple combinational logic between decision nodes. Thus, a good regulating effect is still ensured under the condition of greatly reducing the complexity.
Preferably, a plurality of enabling controllers C10 correspond to a plurality of last-stage decision nodes of the decision controller C6, respectively, for enabling corresponding automatic control state machines. Only in the case of enabling, the output result of the decision controller will trigger the corresponding automatic control state machine. The enabling controller can enable the chip system to selectively adopt corresponding control only on output results of certain decision control, so that the flexibility is improved;
the automatic control state machines C11.0, C11.1, …, C11.n correspond to the plurality of enable controllers C10, respectively, and the automatic control state machines generate response control to the chips for the corresponding events. The response control generated between different automatic control state machines may be the same or different. Wherein, when the response control generated between different automatic control state machines is the same, it indicates that the control action taken on the chip can be of the same type in control parameter. For example, the most direct method for reducing the chip load is to perform down-conversion processing on functional modules inside the chip, and although different automatic control state machines may perform down-conversion processing on the chip, the frequency division ratio of the down-conversion may be different between them.
The purpose of the automatic control state machine is to produce an automatic protection control of the chip in case of overheating, so that once the corresponding event is responded, the corresponding interrupt needs to be cleared after a certain time delay. In view of the timeliness of the automatic control, after the corresponding control is given and a certain time passes, the automatic control state machine needs to automatically cancel the control, so that the chip enters the normal state again to work.
The priority controller C12 is connected to the automatic control state machines C11.0, C11.1, …, C11.n, and is configured to receive multiple control requests of different levels simultaneously generated by the automatic control state machines, and perform priority ordering on the multiple control requests through a preset arbitration mechanism, so as to ensure that the control requests of high levels obtain the fastest response preferentially.
Alternatively, the implementation architecture shown in FIG. 1 may not include registers. In this case, all configuration and control parameters need not be set by registers, for example, may be set manually by a user or stored in a non-volatile memory of the electronic device, or may be downloaded from a cloud server, or the like. Alternatively, the implementation architecture shown in fig. 1 may not include a decision controller. That is, after the threshold detection result is sent to the interrupt generator, the corresponding automatic control state machine can be directly triggered without any decision process.
In one embodiment, the implementation of the automatic control state machine is illustrated in FIG. 2. The state machine mainly comprises the following states:
state S0 is an idle state, and after the reset is released, the automatic control state machine enters the idle state;
the state S1 is a control start state, and in the state S0, when the corresponding input event is 1, the automatic control state machine starts control, and then the automatic control state machine enters a state S1, in which the control mainly performed includes:
s1_ C0: and starting control timing so as to cancel the control after the timing is reached. Since the control state generated by the automatic control state machine is temporary control of an abnormal state, the automatic control state machine finally needs to cancel the control;
s1_ C1: this secondary control is initiated.
State S2 is an interrupt clear state, i.e., the interrupt generated by the corresponding threshold detector is cleared. Specifically, the control mainly performed in this state includes:
s2_ C0: starting interrupt clearing;
s2_ C1: starting the interrupt clear delay count in order to ensure sufficient interrupt clear time;
state S3 is an interrupt clear delay complete state, and enters state S4 after the interrupt clear delay is reached;
the state S4 is a state where the interrupt clear is canceled, and waits for the control timing to end, and when the control timing ends, the state S5 is entered;
the state S5 is a control timing end state, and when the control timing ends, it is detected whether there is still a current input event, if the temperature in the chip is still too high, the input event of the automatic control state machine will not be cancelled due to the interrupt clearing generated by the automatic control state machine, so that the corresponding input event still generated at this time is 1, and the automatic control state machine enters the state S1 to start the next control; otherwise, generating a corresponding input event of 0, and entering a state S6;
state S6 is a deactivated control state, and after the control is deactivated this time, the automatic control state machine goes to state S0 to remain idle until the next event occurs, and a new state transition process is started.
It should be noted that the state machine shown in fig. 2 is only used for illustrating and not limiting the technical solution of the present invention. Those skilled in the art should understand that any conceivable modifications of the conditions, states, actions and the like of the state machine can be made on the basis of the examples of the present invention according to actual needs, and the present invention should not be limited to the specific conditions, states and actions of the above examples.
Based on the above-described implementation architecture and state machine description of automatic protection against chip overheating, the implementation method for automatic protection against chip overheating according to the present invention is shown in fig. 3, and includes the following steps:
step S1: and starting.
Step S2: and performing initial configuration on the register. The configuration of the register comprises setting of different thresholds of different temperature sensors, setting of control node parameters of different levels of a decision control algorithm, enabling of a dynamic control state machine, setting of interrupt clearing delay time, setting of duration of a control timer, setting of different control parameters, setting of control strategies and the like.
Step S3: the threshold detector generates an interrupt to the temperature sensor.
When a certain temperature sensor senses that the temperature exceeds a predefined threshold, the threshold detector generates a corresponding interrupt for the temperature sensor and inputs the corresponding interrupt to the decision controller.
Step S4: the decision controller forms a decision based on the interrupt.
Preferably, the decision controller forms a decision step by step according to the multi-stage decision nodes, and outputs an event corresponding to the decision to the automatic control state machine.
Step S5: the automatic control state machine performs automatic control.
Step S6: the automatic control state machine performs interrupt clearing control.
Since the automatic control of step S5 is a temporary control, it is necessary to control the corresponding threshold detector that generated the interrupt in step S3 to clear the interrupt previously generated by the threshold detector.
Step S7: an interrupt clear end operation is performed.
Specifically, the delay waiting operation may be performed by setting a first time window in advance in the timer, ensuring a sufficient interrupt clearing time, and ensuring that the interrupt is cleared.
Step S8: a control time ending operation is performed.
Specifically, the delay waiting operation can be executed by presetting a second time window in the timer, so that sufficient automatic control time is ensured, and the control process is completely finished.
Step S9: and judging whether a corresponding decision control event exists or not. If the corresponding decision control event still exists, returning to the step S5, executing the next automatic control process, otherwise, entering S10.
Under normal conditions, if the interruption of the threshold detector is cleared, the corresponding decision control event disappears, and then step S10 is entered to end the automatic control. On the contrary, if the chip overheating condition is not improved or the improvement effect is not obvious after the corresponding control is generated, the decision control event still exists, and it is necessary to return to the step S5 to re-execute the automatic control.
Step S10: the automatic control is ended.
Step S11: the chip returns to normal state and the method ends.
At this time, the system judges that there is no overheat event, so the chip returns to the normal working state before the automatic control through the state transition operation.
In an alternative embodiment, if the interrupt generation in the threshold detector is not subjected to sampling processing, but directly changes with the result of the threshold comparator, the interrupt clearing control is not required to be generated in the control process, in this case, the automatic control state machine only needs to generate the automatic control enough time or cancel the control after the automatic control is generated and the chip overheat condition disappears, and the state machine schematic diagrams at this time are respectively shown in fig. 4 and fig. 5. The interruption automatically disappears when the chip load decreases and the temperature also decreases below the threshold.
With specific implementation, fig. 6 shows a schematic diagram of a method for effectively reducing the load of a chip according to an embodiment of the present invention. The embodiment specifically adopts a frequency reduction control mode, and realizes the frequency reduction control process in the form of an automatic control state machine. When the corresponding automatic control state machine generates a control instruction, all clock trees of the chip reduce the running frequency according to the preset proportion until the overheating condition disappears. The implementation of the down-conversion control in fig. 6 may add a frequency division enable controller to the clock tree, and when the down-conversion control is effective, the frequency division enable controller starts to operate, so that the clock tree frequency is reduced, and the chip load is reduced.
It should be noted that the down-conversion control method shown in fig. 6 is only used for illustrating but not limiting the single method of chip load reduction of the present invention. It will be appreciated by those skilled in the art that any conventional method and combination of methods, including for example voltage frequency combination control, may be used in accordance with the actual requirements on the basis of the present invention.
For the case that a single temperature sensor is associated with a plurality of temperature thresholds, if the current automatic control cannot effectively reduce the temperature in the chip, the temperature in the chip continues to rise, which triggers the automatic control state machine of the next stage, so the thresholds of the temperature sensor must be set in a step-like manner, and the corresponding automatic control means generated by the corresponding control state machines of different levels is controlled in a step-like manner from weak to strong. Fig. 7 is a schematic diagram of threshold setting of the temperature sensor, which corresponds the output of the temperature sensor to 5 intervals: a normal temperature zone (for a chip only), a superheat zone 0, a superheat zone 1, a superheat zone 2 and a superheat zone 3. The 5 partitions need to correspond to 4 thresholds, the thresholds are set from low to high, and the number of the thresholds and the respective values are only examples. In this way, the warning level generated when the temperature reaches and exceeds the specified threshold value is also from low to high, wherein the normal temperature zone is considered as normal temperature, the hardware does not need to carry out any treatment, the overheating zone 0, the overheating zone 1, the overheating zone 2 and the overheating zone 3 are all the conditions that the temperature of the chip is overhigh, and the hardware needs to carry out different treatments. The overheating areas 0-3 can correspond to overheating levels so as to distinguish different emergency levels of chip overheating treatment, and corresponding decision control and a corresponding automatic control state machine are formed according to the information.
Therefore, the chip overheating automatic protection method can quickly and effectively execute hardware response and control, directly intervene in the chip overheating condition without software and CPU participation, can remarkably shorten the response time taken by the chip overheating, and simultaneously provides a plurality of effective control strategies, thereby effectively reducing the chip working load and enabling the chip to be recovered to the normal working temperature range in a very short time.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A chip overheating automatic protection method is characterized by comprising the following steps:
receiving output values of a plurality of temperature sensors, wherein the plurality of temperature sensors are respectively positioned at different positions in the chip, and each temperature sensor has a corresponding control threshold value;
generating an interrupt signal when the output value of at least one temperature sensor is higher than the corresponding control threshold value;
forming a decision for the interrupt signal to generate a decision control result;
and triggering a corresponding automatic control state machine according to the decision control result to automatically adjust the temperature of the chip.
2. The method according to claim 1, wherein each temperature sensor has a corresponding level m +1 control threshold, m being a positive integer greater than 2;
when the output value of at least one temperature sensor is higher than the corresponding control threshold value, an interrupt signal is generated, further comprising:
generating a first interrupt signal when an output value of the temperature sensor is higher than a first control threshold and lower than a second control threshold;
generating an ith interrupt signal when an output value of the temperature sensor is higher than the ith control threshold and lower than the (i +1) th control threshold; i is a positive integer, and i belongs to [1, m +1 ];
wherein the decision control results of the first to m +1 th interrupt signals are used to trigger automatic control state machines of different levels.
3. The method according to claim 1, wherein when the automatic control state machine receives a plurality of control requests at the same time, the automatic control state machine performs priority ordering according to the levels of the plurality of control requests, and executes control responses according to the ordered control requests.
4. The method according to claim 1, wherein the triggering a corresponding automatic control state machine according to the decision control result to automatically adjust the temperature of the chip further comprises:
and detecting whether the temperature of the chip is normal within a predefined time, if so, exiting the automatic control of the automatic control state machine, otherwise, re-triggering the automatic control of the automatic control state machine, and continuously adjusting the temperature of the chip.
5. The method for automatic protection against over-temperature of chip as claimed in claim 4, further comprising, after the automatic adjustment of the temperature of the chip:
and executing interrupt clearing control, and executing delay waiting operation based on an interrupt clearing time window preset in a timer so that the interrupt signal is cleared.
6. An automatic protection device for chip overheating, comprising:
the temperature sensors are respectively positioned at different positions in the chip and used for sensing and outputting temperature values of the positions, and each temperature sensor has a corresponding control threshold;
a threshold detector for generating an interrupt signal when an output value of at least one temperature sensor is higher than a corresponding control threshold;
the decision controller is used for forming a decision for the interrupt signal and generating a decision control result;
and the automatic control state machine is used for responding to the decision control result and triggering automatic adjustment of the temperature of the chip.
7. The automatic chip overheat protection device according to claim 6, wherein each of said temperature sensors has a corresponding level m +1 control threshold; wherein m is a positive integer greater than 1;
the threshold detector is further configured to:
each temperature sensor has a corresponding m +1 level control threshold; wherein m is a positive integer greater than 1;
when the output value of at least one temperature sensor is higher than the corresponding control threshold value, an interrupt signal is generated, further comprising:
generating a first interrupt signal when an output value of the temperature sensor is higher than a first control threshold and lower than a second control threshold;
generating an ith interrupt signal when an output value of the temperature sensor is higher than the ith control threshold and lower than the (i +1) th control threshold; wherein i is a positive integer and i belongs to [1, m +1 ];
wherein the decision control results of the first to the m +1 th interrupt signals are used for triggering automatic control state machines of different levels;
the decision controller is further configured to:
and forming a decision for the first to the (m +1) th interrupt signals, wherein the generated decision control result is used for triggering automatic control state machines of different levels.
8. The automatic protection device for chip overheating according to claim 6, further comprising a priority controller for performing a priority ranking according to the level of the plurality of control requests and performing a control response according to the ranked control requests when the plurality of control requests are received by the automatic control state machine at the same time.
9. The chip overheat automatic protection device according to claim 6, wherein the automatic control state machine is further configured to:
after the temperature of the chip is automatically adjusted, whether the temperature of the chip is normal is detected within a predefined time, if the temperature of the chip is normal, the automatic control of the automatic control state machine is quitted, otherwise, the automatic control of the automatic control state machine is triggered again, and the temperature of the chip is continuously adjusted.
10. The chip overheat automatic protection device according to claim 9, wherein the automatic control state machine is further configured to:
after the temperature of the chip is automatically adjusted, interrupt clear control is performed, and a delay waiting operation is performed through an interrupt clear time window preset in a timer, so that the interrupt signal is cleared.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011589421.4A CN112650128A (en) | 2020-12-29 | 2020-12-29 | Automatic protection method and device for chip overheating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011589421.4A CN112650128A (en) | 2020-12-29 | 2020-12-29 | Automatic protection method and device for chip overheating |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112650128A true CN112650128A (en) | 2021-04-13 |
Family
ID=75363605
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011589421.4A Pending CN112650128A (en) | 2020-12-29 | 2020-12-29 | Automatic protection method and device for chip overheating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112650128A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114489235A (en) * | 2022-01-18 | 2022-05-13 | 上海天数智芯半导体有限公司 | Chip frequency dynamic adjustment method and system for abnormal temperature processing |
| CN114979364A (en) * | 2022-05-25 | 2022-08-30 | 南昌黑鲨科技有限公司 | Intelligent heat dissipation system and method and computer readable storage medium |
| CN115454221A (en) * | 2022-09-02 | 2022-12-09 | 无锡众星微系统技术有限公司 | Temperature control method of SAS controller |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW567408B (en) * | 2002-03-29 | 2003-12-21 | Uniwill Comp Corp | Apparatus and method for controlling power and clock speed of electronic system |
| CN101431228A (en) * | 2008-12-18 | 2009-05-13 | 深圳华为通信技术有限公司 | Thermal protection control method and system |
| CN102025533A (en) * | 2010-11-04 | 2011-04-20 | 华为终端有限公司 | Main control unit, network processing unit as well as overheating protection system and method |
| CN103376859A (en) * | 2012-04-26 | 2013-10-30 | 华为技术有限公司 | Chip performance control method and device |
| CN103514011A (en) * | 2012-06-21 | 2014-01-15 | 苹果公司 | Thermal control apparatus and methodology |
| US20140191695A1 (en) * | 2013-01-08 | 2014-07-10 | Honeywell International, Inc., Patent Services M/S Ab/2B | Thermal protection method and system to maximize availability of electric drive system |
| CN104104064A (en) * | 2013-04-08 | 2014-10-15 | 华为终端有限公司 | Thermal protection circuit |
| CN104156263A (en) * | 2013-05-14 | 2014-11-19 | 国际商业机器公司 | Interruption of chip component managing tasks, chip, and assembly comprising chip |
| CN109827666A (en) * | 2019-03-29 | 2019-05-31 | 上海剑桥科技股份有限公司 | Temperature observation circuit |
| CN111817762A (en) * | 2019-04-12 | 2020-10-23 | 三星电子株式会社 | Method and apparatus for thermal management in wireless communications |
-
2020
- 2020-12-29 CN CN202011589421.4A patent/CN112650128A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW567408B (en) * | 2002-03-29 | 2003-12-21 | Uniwill Comp Corp | Apparatus and method for controlling power and clock speed of electronic system |
| CN101431228A (en) * | 2008-12-18 | 2009-05-13 | 深圳华为通信技术有限公司 | Thermal protection control method and system |
| CN102025533A (en) * | 2010-11-04 | 2011-04-20 | 华为终端有限公司 | Main control unit, network processing unit as well as overheating protection system and method |
| CN103376859A (en) * | 2012-04-26 | 2013-10-30 | 华为技术有限公司 | Chip performance control method and device |
| CN103514011A (en) * | 2012-06-21 | 2014-01-15 | 苹果公司 | Thermal control apparatus and methodology |
| US20140191695A1 (en) * | 2013-01-08 | 2014-07-10 | Honeywell International, Inc., Patent Services M/S Ab/2B | Thermal protection method and system to maximize availability of electric drive system |
| CN104104064A (en) * | 2013-04-08 | 2014-10-15 | 华为终端有限公司 | Thermal protection circuit |
| CN104156263A (en) * | 2013-05-14 | 2014-11-19 | 国际商业机器公司 | Interruption of chip component managing tasks, chip, and assembly comprising chip |
| CN109827666A (en) * | 2019-03-29 | 2019-05-31 | 上海剑桥科技股份有限公司 | Temperature observation circuit |
| CN111817762A (en) * | 2019-04-12 | 2020-10-23 | 三星电子株式会社 | Method and apparatus for thermal management in wireless communications |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114489235A (en) * | 2022-01-18 | 2022-05-13 | 上海天数智芯半导体有限公司 | Chip frequency dynamic adjustment method and system for abnormal temperature processing |
| CN114979364A (en) * | 2022-05-25 | 2022-08-30 | 南昌黑鲨科技有限公司 | Intelligent heat dissipation system and method and computer readable storage medium |
| CN115454221A (en) * | 2022-09-02 | 2022-12-09 | 无锡众星微系统技术有限公司 | Temperature control method of SAS controller |
| CN115454221B (en) * | 2022-09-02 | 2023-05-30 | 无锡众星微系统技术有限公司 | Temperature control method of SAS controller |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112650128A (en) | Automatic protection method and device for chip overheating | |
| JP5362912B2 (en) | Temperature collection control device and method | |
| US4023143A (en) | Fixed priority interrupt control circuit | |
| CN109586952B (en) | Server capacity expansion method and device | |
| CN107566235B (en) | Message data transmission method and device and electric vehicle | |
| KR102247181B1 (en) | Method and device for generating anomalous behavior detection model using learning data generated based on xai | |
| US20140361885A1 (en) | Systems and Methods for Process Alarm Reduction | |
| CN108965066B (en) | Method and device for detecting transmission path | |
| US9612894B2 (en) | Detector for high frequency interrupts | |
| CN113515358B (en) | Task scheduling method and device, electronic equipment and storage medium | |
| JP7077815B2 (en) | Lighting control system | |
| CN107911229B (en) | Running state change reminding method and device, electronic equipment and storage medium | |
| CN110825023A (en) | Data interaction method and system between PLC remote modules | |
| EP3819799B1 (en) | Method of threat detection | |
| CN110851259B (en) | Interrupt control method, interrupt controller, computer device and storage medium | |
| CN110673937B (en) | Single chip microcomputer and synchronous and asynchronous control method thereof | |
| CN110890977B (en) | Host node monitoring method and device of cloud platform and computer equipment | |
| CN112636946A (en) | Edge main node election method and power industrial control terminal | |
| JP6347825B2 (en) | Episodic mining apparatus and method | |
| CN111967614B (en) | Artificial intelligence learning method and device | |
| US10521379B1 (en) | Systems and methods for I/O prioritization in nvme-compliant storage devices | |
| CN114489300B (en) | Method and device for resetting Expander chip | |
| JP3394437B2 (en) | Processor Decision Method by Load Balancing Multiprocessor Method | |
| CN117909332A (en) | Data processing system, method and computer storage medium | |
| CN116866270A (en) | Data processing method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210413 |