WO2016197152A1 - Dynamic frequency scaling based on supply current monitoring - Google Patents
Dynamic frequency scaling based on supply current monitoring Download PDFInfo
- Publication number
- WO2016197152A1 WO2016197152A1 PCT/US2016/036106 US2016036106W WO2016197152A1 WO 2016197152 A1 WO2016197152 A1 WO 2016197152A1 US 2016036106 W US2016036106 W US 2016036106W WO 2016197152 A1 WO2016197152 A1 WO 2016197152A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- load current
- current
- frequency
- predefined
- operating frequency
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/324—Power saving characterised by the action undertaken by lowering clock frequency
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/3296—Power saving characterised by the action undertaken by lowering the supply or operating voltage
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/04—Generating or distributing clock signals or signals derived directly therefrom
- G06F1/08—Clock generators with changeable or programmable clock frequency
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3206—Monitoring of events, devices or parameters that trigger a change in power modality
- G06F1/3212—Monitoring battery levels, e.g. power saving mode being initiated when battery voltage goes below a certain level
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
A system for performing dynamic frequency scaling may include a voltage regulator and a decision block. The decision block may be configured to determine a target frequency of operations for a system on chip.
Description
DYNAMIC FREQUENCY SCALING BASED ON SUPPLY CURRENT MONITORING
BACKGROUND OF THE INVENTION
[0001] The present invention relates generally to power regulation of integrated circuits, and more particularly to load current based dynamic frequency scaling for integrated circuits.
[0002] Integrated circuits generally require provision of power within particular parameters during operation. The provision of such power may face many complexities. For example, some devices may be powered by batteries having relatively small capacities, while the devices themselves, at least at various times, may require large amounts of power. Operating such devices to avoid excessive power usage may be desirable.
[0003] One way to reduce power usage by an integrated circuit is to operate the integrated circuit at a lower clock speed. Unfortunately, high level determinations of expected power usage by an integrated circuit, for example by estimating a processing workload by considering types of operations expected to be performed, may not accurately gauge expected power requirements. BRIEF SUMMARY OF THE INVENTION
[0004] Some embodiments in accordance with aspects of the invention provide for dynamic frequency scaling based on an indication of load current.
[0005] Some embodiments in accordance with aspects of the invention provide a system useful for controlling power consumption by integrated circuitry, comprising: a voltage regulator configured to provide regulated power to a load comprising integrated circuitry of a system-on- chip (SOC), the voltage regulator including circuitry for determining a digital signal indicative of load current; phase-locked loop (PLL) circuitry configured to provide a plurality of clock signals, each of the plurality of clock signals at a different frequency signal indicative of current
operating frequency to be used by a system-on-chip (SOC); and a decision block configured to generate a selection signal for selecting one of the plurality of clock signals, for use by the integrated circuitry of the SOC, based on a currently selected clock signal and the signal indicative of the load current.
[0006] Some embodiments in accordance with aspects of the invention provide a method useful for determining a target operating frequency to be used by a system-on-chip (SOC), comprising: selecting a first target operating frequency; determining whether load current exceeds a first predefined operational current; if the load current exceeds the first predefined operational current, selecting a second target operating frequency; determining whether the load current exceeds a second predefined operational current; if the load current exceeds the second predefined operational current, selecting a third target operating frequency; determining whether the load current exceeds a third predefined operational current; and if the load current exceeds the third predefined operational current, selecting a fourth target operating frequency.
[0007] These and other aspects of the invention are more fully comprehended upon review of this disclosure.
BRIEF DESCRIPTION OF THE FIGURES
[0008] FIG. 1 is a block diagram of a system for performing dynamic frequency scaling (DFS) operations in accordance with aspects of the invention.
[0009] FIG. 2 is a flow diagram of a process for determining a target operating frequency based on load current in accordance with aspects of the invention.
[0010] FIG. 3 provides a chart showing simulated results of various levels of load current with respect to system on chip (SoC) activity, for a particular process such as the process of FIG. 2.
[0011] FIG. 4 is a semi-schematic, semi-block diagram of a voltage regulator in accordance with aspects of the invention.
[0012] FIG. 5 provides a chart showing simulated averages of the CMPBP and CMPADJ signals with respect to load current, for a particular operating topology such as that of the embodiment of FIG. 4.
DETAILED DESCRIPTION
[0013] FIG. 1 is a block diagram of a system for performing dynamic frequency scaling (DFS) operations in accordance with aspects of the invention. In the system of FIG. 1 , dynamic frequency scaling is performed based on an indication of load current. In FIG. 1 , an embedded voltage regulator (EVR) 1 13 receives an input voltage VIN and a clock signal from a phase- locked loop (PLL) 1 15, and supplies a regulated power to a load, which is shown as a system on chip (SOC) 1 1 1. In some embodiments, the EVR and the SOC are all part of one device, for example a cellular or smart telephone device or other portable computer device. Moreover, in various embodiments the EVR and the SOC are within the same package for semiconductors, and in some embodiments are in the same multichip module, or even on the same chip.
[0014] Power consumption by the SOC is to an extent controlled by a decision block 1 19. Power consumed by the SOC, when the SOC is performing operations, is generally a function of frequency of operations performed by the SOC. Reduction of frequency of clock signals provided to the SOC generally reduces power consumed by the SOC. The decision block 1 19 generally determines a target frequency of operations, which may be considered a frequency operating point, or frequency set or operating point. The decision block, for example, may generate a decision signal for deciding a frequency operating point in response to an indication of current operating frequency of the SOC and an indication of load current provided by the EVR
to the SOC. As such, the decision signal is a function of the frequency of the clock signal and the indication of load current. In some embodiments, and as shown in FIG. 1 , the EVR may forward the clock signal provided by the PLL directly to the decision block. In various embodiments, the EVR is a switching converter, for example, operating high side and low side switches so as to regulate voltage applied to a load. In some embodiments, the EVR is a switching converter as discussed in more detail in U.S. Patent Application No. 62/171,983, entitled Voltage Regulator Current Load Sensing, filed of even date herewith, the disclosure of which is incorporated herein by reference for all purposes.
[0015] As illustrated in FIG. 1 , a multiplexer 1 17 receives multiple clock signals, operating at various frequencies from a highest frequency FCL to a lowest frequency, shown as FCL /4 in FIG. 1. The multiplexer receives the clock signals from the PLL. As one skilled in the art would understand, a plurality of separate PLLs may instead be used to provide the multiple clock signal, with for example each of the separate PLLs providing one of the clock signals, and the PLL or the plurality of separate PLLs may alone or together be considered PLL circuitry. In some embodiments, the PLL may include divisors in dividing a system clock signal, which may be provided by a voltage-controlled oscillator (VCO). In some embodiments, and as illustrated in FIG. 1 , the clock signals include a first clock signal at a frequency FCLK, a second clock signal at a frequency 3FCLK/4, a third clock signal at a frequency FCLK/2, and a fourth clock signal at a frequency FCLK/4. The multiplexer selects one of the clock signals for use based on the decision signal of the decision block 1 19, and provides the selected clock signal to the SOC as a frequency of operation. In some embodiments, the SOC may be a processor or a microcontroller. In various embodiments, the SOC may contain digital, analog, mixed-signal, and/or radio-frequency functions.
[0016] FIG. 2 is a flow diagram of a process for determining a target operating frequency based on load current in accordance with aspects of the invention. In some embodiments, the process is performed by the decision block 1 19 of FIG. 1. In some embodiments, the process is performed by digital circuitry. In some embodiments, the process is performed by processor circuitry configured by program instructions.
[0017] In block 211, the process selects a target operating frequency of one-fourth of a frequency of a system clock signal (FCLK/4). If the process determines that the load current exceeds one-sixteenth of a maximum operational current (IMAX/16), then the process takes transition 213 to block 215.
[0018] At block 215, the process selects a target operating frequency of one-half of the frequency of the system clock signal (FCLK/2). If the process determines that the load current is less than one-eighth of the maximum operational current (IMAX/8), then the process takes transition 216 back to block 21 1. If, however, the process determines that the load current exceeds one-fourth of the maximum operational current (IMAX/4), then the process takes transition 217 to block 219.
[0019] At block 219, the process selects a target operating frequency of three-fourths of the frequency of the system clock signal (3FCL /4). If the process determines that the load cun-ent is less than three-eighths of the maximum operational current (3 IMAX/8), then the process takes transition 220 back to block 215. If, however, the process determines that the load current exceeds nine-sixteenth of the maximum operational current (9IMAX/16), then the process takes transition 221 to block 223.
[0020] At block 223, the process selects a target operating frequency equivalent to the frequency of the system clock signal (FCL ). If the process determines that the load current is
less than three-fourths of the maximum operational current (3IMAX/4), then the process takes transition 224 back to block 219.
[0021] In various embodiments, the process selects the target operating frequency by commanding, e.g. the decision block 119 of FIG. 1, to send a signal to, e.g. the multiplexer 1 17 of FIG. 1 , to select the target operating frequency, which in some embodiments, may be provided by a phase-locked loop, e.g. the PLL 1 15 of FIG. 1. In some embodiments, the load current may be an output load current provided by a voltage regulator, e.g. the EVR 113 of FIG. 1.
[0022] FIG. 3 provides a chart showing simulated results of various levels of load current with respect to system on chip (SoC) activity, for a particular process such as the process of FIG. 2.
[0023] As shown in FIG. 3, the simulated current load generally increases linearly as SoC activity increases. The SoC generally operates in different operating regions depending on a selected target operating frequency. Moreover, levels of current jump may vary when switching from one target operating frequency to another.
[0024] More specifically, when a target operating frequency of FCL /4 is selected, where FCLK may be a frequency of a system clock, current jump is equivalent to IMAX/4, where IMAX may be a maximum operational current. With the target operating frequency of FCLK/4, as shown in FIG. 3, the SoC operates in low power idle region 311. When a target operating frequency of FCLK/2 is selected, current jump is equivalent to 3IMAX/8. With the target operating frequency of FCLK/2, the SoC operates in low power regular region 313. When a target operating frequency of 3FCLK/4 is selected, current jump is also 3I AX/8. With the target operating frequency of 3 FCLK/4, the SoC operates in high performance regular region 315. And when a target operating frequency of FCLK is selected, current jump is also IMAX/4, as with the target operating frequency
of FCLK /4. With respect to this target operating frequency, the SoC operates in high performance busy region 317.
[0025] FIG. 4 is a semi-schematic, semi-block diagram of a voltage regulator in accordance with aspects of the invention. The voltage regulator includes circuitry, digital circuitry in various embodiments, for determining an indication of load current supplied to a load. The indication of load current may be useful in many respects, including in operation of the voltage regulator, in determining if an over current situation exists, and in allowing for improved thermal management. In various embodiments the circuitry includes circuitry for determining whether an output voltage of the voltage regulator is above and/or below a predefined range of voltages, for determining an average of such occurrences, and for determining an indication of load current based on the average of such occurences. In some embodiments, the voltage regulator of FIG. 4 may be implemented circuitry of the embedded voltage regulator (EVR) 113 of FIG. 1.
[0026] As illustrated in FIG. 4, the converter includes a high side switch 413a, a low side switch 413b, a bypass switch 420, an output inductor 415, an output capacitor 417, a logic circuitry 421 for controlling the high side, low side, and bypass switches, a first comparator 423, a second comparator 424, a third comparator 422, a first pulse-width modulation (PWM) generator 425a, a second PWM generator 425b, a first digital average block 441a, a second digital average block 441b, a first digital function block 443a, and a second digital function block 443b.
[0027] The voltage regulator of FIG. 4 operates the high side, low side, and bypass switches so as to regulate voltage applied to a load 419. In doing so, the converter operates the high side and low side switches either in accordance with PWM signals. In the embodiment of FIG. 4, multiple PWM generators are shown, but in many embodiments only a single PWM generator
may be used. In embodiments with multiple PWM generators, PWM signals generated by either the first PWM generator 425a or the second PWM generator 425b may be selected for example based on whether an output voltage of the converter is below a predetermined magnitude, for example, as indicated by the first comparator 423. The PWM generators, and other aspects of the voltage regulator of FIG. 4, are discussed in more detail in U.S. Patent Application No. 14/732,505, entitled DC-DC Converter Having Digital Control and Reference PWM Generators, filed of even date herewith, the disclosure of which is incorporated herein by reference for all purposes. The converter also operates the bypass switch, also optimal in many embodiments, in accordance with an output provided by the second comparator 424 indicating whether the output voltage of the converter is above a predetermined magnitude.
[0028] Outputs of each of the comparators are also provided to the first digital average block 441a and the second digital average block 441b, respectively. The digital averages are provided to the first digital function block 443a and the second digital function block 443b, respectively, which determine an indication of load current. In some embodiments, the indication of load current may be provided to a decision block, e.g. the decision block 119 of FIG. 1.
[0029] Referring to FIG. 4, the high side switch 413a and the low side switch 413b are coupled in series between a first voltage source and a second voltage source. The first voltage source is at a higher voltage than the second voltage source, with the high side switch coupling the first voltage source to the low side switch, and the low side switch coupling the second voltage source to the high side switch. The high side and low side switches may be formed, for example, with metal-oxide-semiconductor field-effect transistor (MOSFET) transistors, with a p- channel MOS transistor forming the high side switch and an n-channel MOS transistor forming the low side switch. In operation, either the high side switch is active, the low side switch is
active, or neither switch is active. For illustrative purposes, the high side and low side switches also show a resistance (RDSON) provided by the switches.
[0030] The output inductor 415 has one end coupled to a node between the high side switch 413a and the low side switch 413b, and also to a first end of the bypass switch 420. Another end of the output inductor is coupled to the output capacitor 417, a second end of the bypass switch 420, and the load 419, with the load current Load passing through the load. A node coupling the other end of the output inductor, the output capacitor, and the load generally may be considered the output of the voltage regulator. For illustrative puiposes, the other end of the output inductor 415 also shows a resistance (RDCR) provided by the output inductor and associated circuit paths, e.g., a parasitic effect.
[0031] The first comparator 423, the second comparator 424, and the third comparator 422 generally have a first input coupled to the output node, their second inputs coupled to reference voltages, and the comparators configured to determine which input is greater. With respect to the first comparator 423, the reference voltage, for example, may be a desired output voltage of the voltage regulator minus a tolerance voltage. The first comparator therefore determines whether the output voltage of the voltage regulator is less than or greater than a desired output voltage minus a tolerance voltage. With respect to the second comparator 424, the reference voltage may be the desired output voltage of the voltage regulator plus a tolerance voltage. The second comparator therefore determines whether the output voltage of the voltage regulator is greater than or less than the desired output voltage plus the tolerance voltage. With respect to the third comparator 422, the reference voltage may be a minimum operational voltage for the voltage regulator. The third comparator therefore determines whether the output voltage of the voltage regulator drops below the minimum operational voltage. Operations below the minimum
operational voltage generally indicates a short circuit, and an output of the third comparator is may be provided to a short-circuit alarm to prevent a device from operating under conditions indicating a short circuit situation.
[0032] The logic circuitry 421 may receive the output signals from the first and second comparators, and the signals generated by the first and second PWM generators to control states of the high side, low side, and bypass switches. The logic circuitry 421 generally controls the states of the high side, low side, and bypass switches by way of producing control signals for controlling those switches. Referring to FIG. 4, the logic circuitry 421 includes a multiplexer 427 which receives the PWM and PWMadj signals, and selects one of them for use based on an output CMPADJ of the first comparator 423.
[0033] As shown in FIG. 4, a latch 429 stores the signal produced by the second comparator 424. The latch stores the signal when an output of the multiplexer, indicating an end of the duty cycle of the converter switches, transitions to a high state. An output of the latch (which may be referred to as CMPBP) is provided to a gate of the bypass switch, an OR gate 431 and, after passing through an inverter 435, to an AND gate 433. The OR gate also receives the output of the multiplexer, and provides an output to a gate of the high side switch. The high side switch, active when its gate input is low, is therefore active when both the output of the multiplexer and the output of the latch are low. The AND gate also receives the output of the multiplexer, and provides an output to a gate of the low side switch. The low side switch, active when its gate input is high, is therefore active when the inverted latch output is high and the output of the multiplexer is high.
[0034] As further shown in FIG. 4, the first digital average block 441 a receives the output (CMPBP) of the latch 429. In various embodiments, the first digital average block monitors the
output CMPBP by way of recording the output CMPBP over a period of time, and generates a digital average (which may be referred to as <CMPBP>) of the output CMPBP based on recorded values of the output CMPBP. The first digital logic block 443a, in various embodiments, receives the digital average <CMPBP> from the first digital average block, and determines and outputs a first digital load current based on the digital average <CMPBP>. The digital average <CMPBP> may be considered a function of load current, a converter input voltage, the desired output voltage of the voltage regulator, an inductance value of the output inductor, and a period of a switching frequency of the voltage regulator. In some embodiments a correlation of <CMPBP> and load current is determined based on voltage regulator topology. In some embodiments the correlation of <CMPBP> and load current is determined based on simulation and/or measurement of voltage regulator operation under various load conditions. In some embodiments, and as later discussed with respect to FIG. 5, <CMPBP> may be considered to be related to load current by a first order equation, for example of the form <CMPBP> = mlLoad + b.
[0035] Similarly, the second digital average block 441b receives the output (CMPADJ) of the first comparator 423. In various embodiments, the second digital average block monitors the output CMPADJ by way of recording the output CMPADJ over a period of time, and generates a digital average (which may be referred to as <CMPADJ>) of the output CMPADJ based on recorded values of the output CMPADJ. The second digital logic block 443b, in various embodiments, receives the digital average <CMPAD.I> from the second digital average block, and determines and outputs a second digital load current based on the digital average <CMPADJ>. The digital average <CMPADJ> may be considered a function of the second digital load current, the bias voltage, the voltage offset, and a parasitic resistance (for example of the switches and output inductor) of the voltage regulator.
[0036] FIG. 5 provides a chart showing simulated averages of the CMPBP and CMPADJ signals with respect to load current, for a particular operating topology such as that of the embodiment of FIG. 4. For the topology of FIG. 4, the PWM generator may be considered to determine PWM duty cycles using an input voltage of a desired converter output voltage plus a bias voltage vbias, and the PWMadj generator may be considered to determine PWMadj duty cycles using an input voltage of the desired converter output voltage plus the bias voltage vbias plus an adjustment voltage vac. As shown in FIG. 5, the simulated average of the CMPBP signal 513 generally decreases linearly in a first region rl of lower load current, and thereafter remains at a minimum value in a second region r2 of higher load current. The simulated average of the CMPADJ signal 511, to the contrary, generally increases linearly in the second region r2 of higher load current, after being at a minimum value in the first region rl of lower load current.
[0037] The first region rl may be considered to overlap and/or correspond to time periods in which the converter is operating in PFM mode, as the CMPBP periodically goes high, as indicated by the non-minimum values of the simulated average of CMPBP. Similarly, the second region r2 may be considered to overlap and/or correspond to time periods in which the converter is operating in PWM mode, as the CMPBP signal generally does not go high.
[0038] From FIG. 5, it may be seen that a first line 517 described by one first order equation largely corresponds to the simulated digital average of CMPBP with respect to load current, and a second line 515 described by another first order equation largely correspond to the simulated digital average of CMPADJ with respect to load current. Accordingly, it is believed the digital average of CMPBP may be used to determine an indication of load current when the converter is operating in PFM mode, and the digital average of CMPADJ may be used to determine an indication of load current when the converter is operating in PWM mode.
[0039] Although the invention has been discussed with respect to various embodiments, it should be recognized that the invention comprises the novel and non-obvious claims supported by this disclosure.
Claims
1. A system useful for controlling power consumption by integrated circuitry, comprising: a voltage regulator configured to provide regulated power to a load comprising integrated circuitry of a system-on-chip (SOC), the voltage regulator including circuitry for determining a digital signal indicative of load current;
phase-locked loop (PLL) circuitry configured to provide a plurality of clock signals, each of the plurality of clock signals at a different frequency signal indicative of current operating frequency to be used by a system-on-chip (SOC); and
a decision block configured to generate a selection signal for selecting one of the plurality of clock signals, for use by the integrated circuitry of the SOC, based on a currently selected clock signal and the signal indicative of the load current.
2. The system of claim 1 further comprising a multiplexer configured to receive the plurality of clock signals from the PLL circuitry, the multiplexer configured to provide one of the plurality of clock signals to the SOC based on the selection signal generated by the decision block.
3. The system of claim 1, wherein the decision block is configured to generate the selection signal so as to increase a target frequency of operations to be used by the integrated circuitry of the SOC when the load current exceeds a first predefined magnitude.
4. The system of claim 1 , wherein the decision block is configured to generate the selection signal so as to reduce the target frequency of operations to be used by the SOC when the load current is below a second predefined magnitude.
5. The system of claim 1, wherein the plurality of clock signals includes a first clock signal at a frequency F, a second clock signal at a frequency 3F/4, a third clock signal at a frequency F/2, and a fourth clock signal at a frequency F/4.
6. The system of claim 5, wherein the decision block is configured to generate the selection signal to select the second clock signal when the first clock is the currently selected clock frequency and the signal indicative of the load current indicates a load current greater than a first predefined fraction of a predefined maximum load current.
7. The system of claim 6, wherein the decision block is configured to generate the selection signal to select the third clock signal when the second clock is the currently selected clock frequency and the signal indicative of the load current indicates a load current greater than a second predefined fraction of a predefined maximum load current.
8. The system of claim 7, wherein the decision block is configured to generate the selection signal to select the fourth clock signal when the third clock is the currently selected clock frequency and the signal indicative of the load current indicates a load current greater than a third predefined fraction of a predefined maximum load current.
9. The system of claim 8, wherein the decision block is configured to generate the selection signal to select the third clock signal when the fourth clock is the currently selected clock frequency and the signal indicative of the load current indicates a load current greater than a fourth predefined fraction of a predefined maximum load current.
10. A method useful for determining a target operating frequency to be used by a system-on- chip (SOC), comprising:
selecting a first target operating frequency;
determining whether load current exceeds a first predefined operational current;
if the load current exceeds the first predefined operational current, selecting a second target operating frequency;
determining whether the load current exceeds a second predefined operational current; if the load current exceeds the second predefined operational current, selecting a third target operating frequency;
determining whether the load current exceeds a third predefined operational current; and if the load current exceeds the third predefined operational current, selecting a fourth target operating frequency.
11. The method of claim 10, wherein the first target operating frequency is one-fourth of a frequency of a system clock signal, the second target operating frequency is one-half of the frequency of the system clock signal, the third target operating frequency is three-fourths of the frequency of the system clock signal, and the fourth target operating frequency is the frequency of the system clock signal.
12. The method of claim 10, wherein the first predefined operational current is one-sixteenth of a maximum operational current, the second predefined operational current is one-fourth of the maximum operational current, and the third predefined operational current is nine-sixteenth of the maximum operational current.
13. The method of claim 10, further comprising:
determining whether the load current is less than a fourth predefined operational current; if the load current is less than the fourth predefined operational current, selecting the first target operating frequency;
determining whether the load current is less than a fifth predefined operational current;
if the load current is less than the fifth predefined operational current, selecting the second target operating frequency;
determining whether the load current is less than a sixth predefined operational current; and
if the load current is less than the sixth predefined operational current, selecting the third target operating frequency.
14. The method of claim 13, wherein the fourth predefined operational current is one-eighth of a maximum operational current, the fifth predefined operational current is three-eighths of the maximum operational current, and the sixth operational current is three-fourths of the maximum operational current.
15. The method of claim 10, wherein the SOC operates in low power idle region when the first target operating frequency is selected.
16. The method of claim 10, wherein the SOC operates in low power regular region when the second target operating frequency is selected.
17. The method of claim 10, wherein the SOC operates in high performance regular region when the third target operating frequency is selected.
18. The method of claim 10, wherein the SOC operates in high performance busy region when selecting the fourth target operating frequency.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP16804675.3A EP3304714A4 (en) | 2015-06-05 | 2016-06-06 | Dynamic frequency scaling based on supply current monitoring |
CN201680040268.5A CN108139786B (en) | 2015-06-05 | 2016-06-06 | Dynamic frequency scaling based on supply current monitoring |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201562172007P | 2015-06-05 | 2015-06-05 | |
US62/172,007 | 2015-06-05 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016197152A1 true WO2016197152A1 (en) | 2016-12-08 |
Family
ID=57442136
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2016/036106 WO2016197152A1 (en) | 2015-06-05 | 2016-06-06 | Dynamic frequency scaling based on supply current monitoring |
Country Status (4)
Country | Link |
---|---|
US (1) | US10139888B2 (en) |
EP (1) | EP3304714A4 (en) |
CN (1) | CN108139786B (en) |
WO (1) | WO2016197152A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110915116A (en) * | 2017-06-14 | 2020-03-24 | 朝阳半导体技术江阴有限公司 | Voltage regulator with time-aware current reporting |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108475985B (en) * | 2015-12-11 | 2021-01-08 | 朝阳半导体技术江阴有限公司 | Boost DC-DC converter with digitally controlled and reference PWM generator |
US10103630B1 (en) * | 2018-02-14 | 2018-10-16 | Nxp B.V. | SMPS filter optimization |
CN114721908B (en) * | 2022-04-11 | 2022-12-06 | 摩尔线程智能科技(北京)有限责任公司 | On-chip power consumption control circuit, chip and power consumption control method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1996028877A1 (en) * | 1995-03-09 | 1996-09-19 | Macronix International Co., Ltd. | Regulated reference voltage circuit for flash memory device and other integrated circuit applications |
JP2002312073A (en) * | 2001-04-12 | 2002-10-25 | Oki Electric Ind Co Ltd | Power saving integrated circuit and control method of power saving integrated circuit |
KR20080017641A (en) * | 2006-08-21 | 2008-02-27 | 삼성전자주식회사 | Clock multiplier and clock generator having the same |
US20100085031A1 (en) * | 2005-07-01 | 2010-04-08 | Dobberpuhl Daniel W | Operating an Integrated Circuit at a Minimum Supply Voltage |
US20140152278A1 (en) * | 2011-08-26 | 2014-06-05 | The Trustees Of Columbia University In The City Of New York | Systems and methods for switched-inductor integrated voltage regulators |
Family Cites Families (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4727318A (en) | 1984-10-04 | 1988-02-23 | Sony/Tektronix Corporation | Apparatus for measuring characteristics of electronic devices |
US20020087904A1 (en) * | 2000-12-28 | 2002-07-04 | Zhong-Ning (George) Cai | Method and apparatus for thermal sensitivity based dynamic power control |
US7279871B2 (en) * | 2004-01-12 | 2007-10-09 | Lionel Llc | Voltage regulator and method of regulating voltage |
CN1949128A (en) * | 2005-10-14 | 2007-04-18 | 鸿富锦精密工业(深圳)有限公司 | Central processor frequency regulating circuit |
WO2007045265A1 (en) * | 2005-10-21 | 2007-04-26 | Freescale Semiconductor, Inc. | Electronic device and method for controlling current |
TWI309766B (en) | 2006-05-11 | 2009-05-11 | Asustek Comp Inc | Method and apparatus for switching performance |
TW200923631A (en) * | 2007-11-22 | 2009-06-01 | Inventec Corp | Apparatus and method for adjusting working frequency of VRD by detecting current |
US8370663B2 (en) | 2008-02-11 | 2013-02-05 | Nvidia Corporation | Power management with dynamic frequency adjustments |
DE102008049714A1 (en) * | 2008-09-30 | 2010-04-01 | Siemens Enterprise Communications Gmbh & Co. Kg | Method and arrangement for operating an electronic system |
JP2010246287A (en) * | 2009-04-07 | 2010-10-28 | Renesas Electronics Corp | Current control circuit |
US8732495B2 (en) | 2010-08-31 | 2014-05-20 | Integrated Device Technology, Inc. | Systems, apparatuses and methods for dynamic voltage and frequency control of components used in a computer system |
KR101744356B1 (en) * | 2011-03-17 | 2017-06-08 | 삼성전자주식회사 | Device and method for reducing current consuming in wireless terminal |
US9563254B2 (en) * | 2011-12-22 | 2017-02-07 | Intel Corporation | System, method and apparatus for energy efficiency and energy conservation by configuring power management parameters during run time |
CN103246340A (en) | 2012-02-06 | 2013-08-14 | 索尼公司 | Device and method for dynamically adjusting frequency of central processing unit |
US9600024B2 (en) * | 2012-09-28 | 2017-03-21 | Mediatek Singapore Pte. Ltd. | Control method of clock gating for dithering in the clock signal to mitigate voltage transients |
US9086708B2 (en) | 2012-12-31 | 2015-07-21 | Gazelle Semiconductor Inc. | High slew rate switching regulator circuits and methods |
US9268347B2 (en) | 2013-02-12 | 2016-02-23 | International Business Machines Corporation | Implementing dynamic regulator output current limiting |
KR102166935B1 (en) | 2013-11-11 | 2020-10-16 | 삼성전자주식회사 | Method of changing an operating frequency for performing a dynamic voltage and frequency scaling, system on-chip, and mobile device having the same |
US9760967B2 (en) | 2013-11-13 | 2017-09-12 | Qualcomm Incorporated | System and method of dynamically throttling CPU frequency for gaming workloads |
-
2016
- 2016-06-06 CN CN201680040268.5A patent/CN108139786B/en active Active
- 2016-06-06 WO PCT/US2016/036106 patent/WO2016197152A1/en active Application Filing
- 2016-06-06 EP EP16804675.3A patent/EP3304714A4/en not_active Withdrawn
- 2016-06-06 US US15/174,884 patent/US10139888B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1996028877A1 (en) * | 1995-03-09 | 1996-09-19 | Macronix International Co., Ltd. | Regulated reference voltage circuit for flash memory device and other integrated circuit applications |
JP2002312073A (en) * | 2001-04-12 | 2002-10-25 | Oki Electric Ind Co Ltd | Power saving integrated circuit and control method of power saving integrated circuit |
US20100085031A1 (en) * | 2005-07-01 | 2010-04-08 | Dobberpuhl Daniel W | Operating an Integrated Circuit at a Minimum Supply Voltage |
KR20080017641A (en) * | 2006-08-21 | 2008-02-27 | 삼성전자주식회사 | Clock multiplier and clock generator having the same |
US20140152278A1 (en) * | 2011-08-26 | 2014-06-05 | The Trustees Of Columbia University In The City Of New York | Systems and methods for switched-inductor integrated voltage regulators |
Non-Patent Citations (1)
Title |
---|
See also references of EP3304714A4 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110915116A (en) * | 2017-06-14 | 2020-03-24 | 朝阳半导体技术江阴有限公司 | Voltage regulator with time-aware current reporting |
CN110915116B (en) * | 2017-06-14 | 2023-11-10 | 超旸半导体(上海)有限公司 | Voltage Regulator with Time-Aware Current Reporting |
Also Published As
Publication number | Publication date |
---|---|
CN108139786A (en) | 2018-06-08 |
US10139888B2 (en) | 2018-11-27 |
US20160357242A1 (en) | 2016-12-08 |
CN108139786B (en) | 2021-07-23 |
EP3304714A1 (en) | 2018-04-11 |
EP3304714A4 (en) | 2018-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10139888B2 (en) | Dynamic frequency scaling based on supply current monitoring | |
US7498793B2 (en) | Current-mode DC-to-DC-converter | |
US10326364B2 (en) | Voltage regulator current load sensing | |
US10389237B1 (en) | Light-load efficiency improvement of hybrid switched capacitor converter | |
US9882472B2 (en) | Techniques for power supply topologies with capacitance management to reduce power loss associated with charging and discharging when cycling between power states | |
US9483065B2 (en) | Power regulation with load detection | |
JPH11289754A (en) | Power supply and control method for power circuit | |
US7786714B2 (en) | Voltage converter apparatus and method therefor | |
US20120153912A1 (en) | Controller for a Power Converter and Method of Operating the Same | |
US20070013351A1 (en) | Electric power unit operating in continuous and discontinuous conduction modes and control method therefor | |
US8829872B1 (en) | Systems and methods for dropping and/or adding phases in multiphase regulators | |
CN114556760A (en) | Method and system for buck converter current reuse | |
US20170364107A1 (en) | Switching regulator control circuit | |
US20170126124A1 (en) | Power converter load current control | |
US20200076305A1 (en) | Electronic converter and method of operating an electronic converter | |
US10164541B2 (en) | Power supply unit | |
JP6326967B2 (en) | Multi-phase power supply | |
CN113872416A (en) | DVFS power supply system and DVFS power supply control method | |
US20170085079A1 (en) | Power supply device and power supply device operating method | |
US10007286B2 (en) | Switching regulator overload detector | |
WO2019113584A1 (en) | Timer for power converter controller | |
Rahman et al. | Digital pulse-frequency/pulse-amplitude modulator for improving efficiency of SMPS operating under light loads | |
US9507356B2 (en) | Adaptive controller for control of a DC-DC power supply | |
JP2008517575A (en) | Converter circuit with improved efficiency | |
JP2023073225A (en) | Hardware Scheme for Dynamic Adjustment of DC-DC Converter Peak Current and Safe LDO Disable |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16804675 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2016804675 Country of ref document: EP |