TWI483098B - Computer system and power sharing method thereof - Google Patents

Computer system and power sharing method thereof Download PDF

Info

Publication number
TWI483098B
TWI483098B TW101106064A TW101106064A TWI483098B TW I483098 B TWI483098 B TW I483098B TW 101106064 A TW101106064 A TW 101106064A TW 101106064 A TW101106064 A TW 101106064A TW I483098 B TWI483098 B TW I483098B
Authority
TW
Taiwan
Prior art keywords
power
processor
pulse width
width modulation
modulation signal
Prior art date
Application number
TW101106064A
Other languages
Chinese (zh)
Other versions
TW201335735A (en
Inventor
Che Wei Lin
Yih Chiou
Original Assignee
Acer Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Acer Inc filed Critical Acer Inc
Priority to TW101106064A priority Critical patent/TWI483098B/en
Publication of TW201335735A publication Critical patent/TW201335735A/en
Application granted granted Critical
Publication of TWI483098B publication Critical patent/TWI483098B/en

Links

Landscapes

  • Dc-Dc Converters (AREA)

Description

電腦系統及其功率分享方法 Computer system and its power sharing method

本發明是有關於一種功率分享方法,且特別是有關於一種用於電腦系統的功率分享方法與使用此方法的電腦系統。 The present invention relates to a power sharing method, and more particularly to a power sharing method for a computer system and a computer system using the same.

在電腦系統架構中,中央處理器(Central Processing Unit,CPU)與圖形處理器(Graphics Processing Unit,GPU)等處理器的操作頻率高低對系統效能的影響甚鉅。處理器超頻技術便是讓中央處理器或圖形處理器在一定時間內進行超頻,以提供處理器晶片所能承受的最大超頻幅度,進而提升電腦系統的效能。 In the computer system architecture, the operating frequency of a processor such as a central processing unit (CPU) and a graphics processing unit (GPU) has a great influence on system performance. Processor overclocking technology allows the central processing unit or graphics processor to overclock in a certain period of time to provide the maximum overclocking range that the processor chip can withstand, thereby improving the performance of the computer system.

然而,提升操作頻率使得電腦系統在運作時散發出來的熱能也愈來愈多。以筆記型電腦為例,由於其機殼散熱空間有限,因此在利用超頻技術來提升系統效能之餘,則十分難以對系統溫度進行控制。為了確保重要元件能正常工作而不至於因高溫而燒毀,電腦系統多半具有溫度監控機制以確保系統溫度在上升至均溫後不再提高。然而,根據測試及使用結果顯示,當電腦系統的溫度到達均溫後,由於要對溫度進行控管,因而會導致電腦系統的效能被大幅壓低。換句話說,目前的溫度監控機制較難在電腦系統的溫度及效能表現上取得平衡,而容易對操作感受造成負面影響。 However, increasing the operating frequency has caused more and more thermal energy from the computer system during operation. Taking a notebook computer as an example, because the heat dissipation space of the casing is limited, it is very difficult to control the system temperature while using the overclocking technology to improve the system performance. In order to ensure that important components can work properly without burning due to high temperatures, most computer systems have temperature monitoring mechanisms to ensure that the system temperature does not increase after rising to ambient temperature. However, according to the test and usage results, when the temperature of the computer system reaches the average temperature, the performance of the computer system is greatly reduced due to the temperature control. In other words, the current temperature monitoring mechanism is more difficult to balance the temperature and performance of the computer system, and it is easy to have a negative impact on the operating experience.

有鑑於此,本發明提供一種電腦系統及其功率分享方法,避免在電腦系統上升到均溫後便難以提升其效能的情況。 In view of this, the present invention provides a computer system and a power sharing method thereof, which are difficult to improve the performance of a computer system after it has risen to a uniform temperature.

本發明提出一種功率分享方法,用於包括第一處理器與第二處理器的電腦系統,此方法包括下列步驟:接收第一脈寬調變信號,第一脈寬調變信號具有一責任週期。將系統散熱功率定義為第一處理器與第二處理器個別預設之溫度設計功率(Thermal Design Power,TDP)的總和。判斷系統散熱功率是否大於或等於預設上限值。若是,則利用第一脈寬調變信號調整第一處理器與第二處理器的功率,其中利用升壓提高第一處理器之功率的時間與利用升壓提高第二處理器之功率的時間互不重疊。判斷電腦系統的系統溫度是否上升。若是,減少責任週期或系統散熱功率以縮短利用升壓以提高功率的時間。若否,增加責任週期或系統散熱功率以增長利用升壓以提高功率的時間。本方法重複判斷系統散熱功率是否大於或等於預設上限值、利用第一脈寬調變信號調整第一與第二處理器的功率,以及根據系統溫度增減責任週期或系統散熱功率等步驟,直到系統散熱功率小於預設上限值。 The present invention provides a power sharing method for a computer system including a first processor and a second processor, the method comprising the steps of: receiving a first pulse width modulation signal, the first pulse width modulation signal having a duty cycle . The system heat dissipation power is defined as the sum of the first preset temperature design power (TDP) of the first processor and the second processor. Determine whether the system cooling power is greater than or equal to the preset upper limit. If yes, adjusting the power of the first processor and the second processor by using the first pulse width modulation signal, wherein increasing the power of the first processor by using the boost and increasing the power of the second processor by using the boosting Do not overlap each other. Determine if the system temperature of the computer system is rising. If so, reduce the duty cycle or system cooling power to reduce the time to use boost to increase power. If not, increase the duty cycle or system cooling power to increase the time to use boost to increase power. The method repeatedly determines whether the system heat dissipation power is greater than or equal to a preset upper limit value, adjusts the power of the first and second processors by using the first pulse width modulation signal, and increases or decreases the duty cycle or the system heat dissipation power according to the system temperature. Until the system cooling power is less than the preset upper limit.

在本發明之一實施例中,其中減少責任週期或系統散熱功率以縮短利用升壓來提高功率之時間的步驟包括判斷責任週期是否到達週期下限。若否,則減少責任週期。若是,則減少系統散熱功率。 In one embodiment of the invention, the step of reducing the duty cycle or system heat dissipation power to reduce the time to increase power with boosting includes determining whether the duty cycle has reached a lower cycle limit. If not, reduce the duty cycle. If yes, reduce the system cooling power.

在本發明之一實施例中,其中增加責任週期或系統散熱功率以增長利用升壓以提高功率的時間的步驟包括判斷責任週期是否到達週期上限。若否,則增加責任週期。若是,則增加系統散熱功率。 In an embodiment of the invention, the step of increasing the duty cycle or system heat dissipation power to increase the time to utilize boost to increase power includes determining whether the duty cycle has reached the upper cycle limit. If not, increase the duty cycle. If yes, increase the system cooling power.

在本發明之一實施例中,其中在利用第一脈寬調變信號調整第一處理器與第二處理器的功率與判斷電腦系統的系統溫度是否上升之間,此功率分享方法更包括在需要判斷第一處理器的使用率時,取得第一處理器的目前使用功率,並根據第一處理器的目前使用功率與溫度設計功率判斷第一處理器是處於高使用率狀態或低使用率狀態。若第一處理器處於低使用率狀態,則降低第一處理器的溫度設計功率,並將第二處理器的溫度設計功率設定為系統散熱功率與第一處理器目前之溫度設計功率的差值。若第一處理器處於高使用率狀態,則增加第一處理器的溫度設計功率,並將第二處理器的溫度設計功率設定為系統散熱功率與第一處理器目前之溫度設計功率的差值。 In an embodiment of the present invention, wherein the power sharing method is further included between adjusting the power of the first processor and the second processor by using the first pulse width modulation signal and determining whether the system temperature of the computer system is rising. When the usage rate of the first processor needs to be determined, the current usage power of the first processor is obtained, and the first processor is determined to be in a high usage state or a low usage rate according to the current usage power and temperature design power of the first processor. status. If the first processor is in the low usage state, the temperature design power of the first processor is lowered, and the temperature design power of the second processor is set as the difference between the system cooling power and the current temperature design power of the first processor. . If the first processor is in the high usage state, increasing the temperature design power of the first processor, and setting the temperature design power of the second processor to a difference between the system cooling power and the current temperature design power of the first processor .

在本發明之一實施例中,其中利用第一脈寬調變信號調整第一處理器與第二處理器的功率的步驟包括在第一脈寬調變信號處於第一邏輯準位時以升壓提高第一處理器之功率,並在第一脈寬調變信號處於第二邏輯準位時以升壓提高第二處理器之功率。 In an embodiment of the invention, the step of adjusting the power of the first processor and the second processor by using the first pulse width modulation signal comprises increasing the first pulse width modulation signal when the first logic level is at the first logic level The voltage increases the power of the first processor and boosts the power of the second processor with the boost when the first pulse width modulation signal is at the second logic level.

在本發明之一實施例中,其中利用第一脈寬調變信號調整第一處理器與第二處理器的功率的步驟包括提供第二脈寬調變信號,此第二脈寬調變信號與第一脈寬調變信號 互為反相。在第一脈寬調變信號處於第一邏輯準位時以升壓提高第一處理器之功率,並且在第二脈寬調變信號處於第一邏輯準位時以升壓提高第二處理器之功率。 In an embodiment of the invention, the step of adjusting the power of the first processor and the second processor by using the first pulse width modulation signal comprises providing a second pulse width modulation signal, the second pulse width modulation signal First pulse width modulation signal Mutual inversion. Raising the power of the first processor with the boost when the first pulse width modulation signal is at the first logic level, and boosting the second processor with the boost when the second pulse width modulation signal is at the first logic level Power.

從另一觀點來看,本發明提出一種電腦系統,包括第一處理器、第二處理器、晶片組、溫度感測器,以及控制器。其中,晶片組耦接第一處理器與第二處理器。溫度感測器用以感測電腦系統的系統溫度。控制器耦接晶片組與溫度感測器。控制器接收第一脈寬調變信號,第一脈寬調變信號具有一責任週期,以及將系統散熱功率定義為第一處理器與第二處理器個別預設之溫度設計功率的總和。控制器在判斷系統散熱功率大於或等於預設上限值時,利用第一脈寬調變信號調整第一處理器與第二處理器的功率,其中利用升壓提高第一處理器之功率的時間與利用升壓提高第二處理器之功率的時間互不重疊,並判斷電腦系統的系統溫度是否上升,在判定系統溫度上升時減少責任週期或系統散熱功率以縮短利用升壓來提高功率的時間,以及在判定系統溫度下降時增加責任週期或系統散熱功率以增長利用升壓來提高功率的時間。控制器重複前述操作直到系統散熱功率小於預設上限值。 From another point of view, the present invention provides a computer system including a first processor, a second processor, a chipset, a temperature sensor, and a controller. The chipset is coupled to the first processor and the second processor. A temperature sensor is used to sense the system temperature of the computer system. The controller is coupled to the chip set and the temperature sensor. The controller receives the first pulse width modulation signal, the first pulse width modulation signal has a duty cycle, and defines the system heat dissipation power as a sum of the temperature preset design powers preset by the first processor and the second processor. The controller adjusts the power of the first processor and the second processor by using the first pulse width modulation signal when determining that the system heat dissipation power is greater than or equal to the preset upper limit value, wherein boosting the power of the first processor by using the boosting The time does not overlap with the time of boosting the power of the second processor by using the boost, and determines whether the system temperature of the computer system rises, and reduces the duty cycle or the system heat dissipation power when determining the temperature rise of the system to shorten the use of the boost to increase the power. Time, and increase the duty cycle or system cooling power when determining the system temperature drops to increase the time to boost power with boost. The controller repeats the foregoing operation until the system heat dissipation power is less than the preset upper limit value.

在本發明之一實施例中,其中若控制器判斷責任週期未到達週期下限,則減少責任週期以縮短利用升壓來提高功率的時間,而若控制器判斷責任週期到達週期下限,則減少系統散熱功率以縮短利用升壓來提高功率的時間。 In an embodiment of the present invention, if the controller determines that the duty cycle does not reach the lower limit of the cycle, the duty cycle is reduced to shorten the time for boosting the power by using the boost, and if the controller determines that the duty cycle reaches the lower limit of the cycle, the system is reduced. Heat dissipation power to reduce the time to increase power with boost.

在本發明之一實施例中,其中若控制器判斷責任週期 未到達週期上限,則增加責任週期以增長利用升壓來提高功率的時間,而若控制器判斷責任週期到達週期上限,則增加系統散熱功率以增長利用升壓來提高功率的時間。 In an embodiment of the invention, wherein the controller determines the duty cycle If the upper limit of the cycle is not reached, the duty cycle is increased to increase the time to increase the power by using the boost, and if the controller determines that the duty cycle reaches the upper limit of the cycle, the system heat dissipation power is increased to increase the time for boosting the power.

在本發明之一實施例中,其中控制器在利用第一脈寬調變信號調整第一處理器與第二處理器的功率以及判斷電腦系統的系統溫度是否上升之間,更包括在需要判斷第一處理器的使用率時取得第一處理器的目前使用功率,根據第一處理器的目前使用功率與溫度設計功率判斷第一處理器是處於高使用率狀態或低使用率狀態。若第一處理器處於低使用率狀態,則降低第一處理器的溫度設計功率,並將第二處理器的溫度設計功率設定為系統散熱功率與第一處理器目前之溫度設計功率的差值。若第一處理器處於高使用率狀態,則增加第一處理器的溫度設計功率,並將第二處理器的溫度設計功率設定為系統散熱功率與第一處理器目前之溫度設計功率的差值。 In an embodiment of the present invention, wherein the controller adjusts the power of the first processor and the second processor by using the first pulse width modulation signal and determines whether the system temperature of the computer system rises, and further includes determining The usage rate of the first processor obtains the current usage power of the first processor, and determines whether the first processor is in a high usage state or a low usage state according to the current usage power and temperature design power of the first processor. If the first processor is in the low usage state, the temperature design power of the first processor is lowered, and the temperature design power of the second processor is set as the difference between the system cooling power and the current temperature design power of the first processor. . If the first processor is in the high usage state, increasing the temperature design power of the first processor, and setting the temperature design power of the second processor to a difference between the system cooling power and the current temperature design power of the first processor .

在本發明之一實施例中,其中控制器在第一脈寬調變信號處於第一邏輯準位時以升壓提高第一處理器之功率,並在第一脈寬調變信號處於第二邏輯準位時以升壓提高第二處理器之功率。 In an embodiment of the invention, the controller increases the power of the first processor by boosting when the first pulse width modulation signal is at the first logic level, and is at the second of the first pulse width modulation signal. The logic level increases the power of the second processor with a boost.

在本發明之一實施例中,其中控制器接收第二脈寬調變信號,此第二脈寬調變信號與第一脈寬調變信號互為反相。控制器在第一脈寬調變信號處於第一邏輯準位時以升壓提高第一處理器之功率,並在第二脈寬調變信號處於第一邏輯準位時以升壓提高第二處理器之功率。 In an embodiment of the invention, the controller receives the second pulse width modulation signal, and the second pulse width modulation signal and the first pulse width modulation signal are mutually inverted. The controller increases the power of the first processor by boosting when the first pulse width modulation signal is at the first logic level, and increases the boost by the boost when the second pulse width modulation signal is at the first logic level The power of the processor.

基於上述,本發明是以脈寬調變信號控制兩處理器以高速交替切換至其功率上限,並且根據處理器的使用率動態分配電腦系統能提供的系統散熱功率。據此能確保系統溫度不致過熱同時提高系統效能,而達到兼顧電腦系統之效能及溫度表現的目的。 Based on the above, the present invention controls the two processors to switch to their upper power limit at high speed with a pulse width modulation signal, and dynamically allocates the system heat dissipation power that the computer system can provide according to the usage rate of the processor. According to this, it can ensure that the system temperature does not overheat and improve the system performance, and achieve the purpose of both the performance and temperature performance of the computer system.

為讓本發明之上述特徵和優點能更明顯易懂,下文特舉實施例,並配合所附圖式作詳細說明如下。 The above described features and advantages of the present invention will be more apparent from the following description.

在使用電腦系統的過程中,處理器之功率對系統效能及溫度有相當直接的影響。由於系統會經常監測處理器的功率(例如,每一百毫秒進行一次監測),因此在對處理器進行超頻而提高系統效能之餘,若能在監測到功率上升時隨即做出對應處理,而並非等待系統溫度已過度升高時才進行控制,則可使電腦系統在效能與溫度表現上取得平衡。本發明便是基於上述觀點進而發展出的一種電腦系統及其功率分享方法,為了使本發明之內容更為明瞭,以下特舉實施例作為本發明確實能夠據以實施的範例。 In the process of using a computer system, the power of the processor has a fairly direct impact on system performance and temperature. Since the system constantly monitors the power of the processor (for example, monitoring every hundred milliseconds), if the processor is overclocked to improve system performance, if the power is detected, the corresponding processing will be performed immediately. Controlling the system without excessive temperature rise allows the computer system to balance performance and temperature performance. The present invention is a computer system and a power sharing method thereof which have been developed based on the above-mentioned viewpoints. In order to clarify the content of the present invention, the following specific embodiments are exemplified by the present invention.

圖1是依照本發明之一實施例所繪示之電腦系統的方塊圖。請參閱圖1,電腦系統100包括第一處理器110、第二處理器120、晶片組130、溫度感測器140,以及控制器150。本實施例之電腦系統100例如是筆記型電腦系統,然而本發明並不以此為限。在其他實施例中,電腦系統100也可以是桌上型電腦系統或平板電腦等等。 1 is a block diagram of a computer system in accordance with an embodiment of the present invention. Referring to FIG. 1, the computer system 100 includes a first processor 110, a second processor 120, a chipset 130, a temperature sensor 140, and a controller 150. The computer system 100 of this embodiment is, for example, a notebook computer system, but the invention is not limited thereto. In other embodiments, computer system 100 can also be a desktop computer system or tablet computer or the like.

第一處理器110例如是中央處理器(Central Processing Unit,CPU),用以控管電腦系統100的整體運作。第二處理器120例如是圖形處理器(Graphics Processing Unit,GPU),用以產生呈現於螢幕或其他顯示裝置(未繪示)的影像訊號。 The first processor 110 is, for example, a central processing unit (Central Processing) Unit, CPU), used to control the overall operation of the computer system 100. The second processor 120 is, for example, a graphics processing unit (GPU) for generating an image signal presented on a screen or other display device (not shown).

晶片組130耦接第一處理器110、第二處理器120以及控制器150。在一實施例中,晶片組130包括南橋晶片與北橋晶片,其中南橋晶片用以連接控制器150、基本輸入輸出系統(未繪示),以及速度較慢的週邊設備,而北橋晶片則連接如第一處理器110、第二處理器120,以及主記憶體(未繪示)等速度較快的元件。 The chipset 130 is coupled to the first processor 110, the second processor 120, and the controller 150. In one embodiment, the wafer set 130 includes a south bridge wafer and a north bridge wafer, wherein the south bridge wafer is used to connect the controller 150, a basic input/output system (not shown), and a slower peripheral device, and the north bridge wafer is connected as The first processor 110, the second processor 120, and a faster memory component such as a main memory (not shown).

溫度感測器140用以偵測電腦系統100的系統溫度,其可包括中央處理器的溫度、圖形處理器的溫度、主記憶體的溫度,以及電腦系統100的表面溫度等。舉例來說,溫度感測器140可包括數個感測器,並分別置於中央處理器、圖形處理器,以及主記憶體等元件的四周以偵測其溫度。 The temperature sensor 140 is configured to detect the system temperature of the computer system 100, which may include the temperature of the central processing unit, the temperature of the graphics processor, the temperature of the main memory, and the surface temperature of the computer system 100. For example, the temperature sensor 140 can include a plurality of sensors disposed around the components of the central processing unit, the graphics processor, and the main memory to detect the temperature thereof.

控制器150耦接溫度感測器140。在本實施例中,控制器150例如是嵌入式控制器(Embedded Controller,EC),用以控管電腦系統100的電源、輸入裝置(例如鍵盤或觸控墊等),並可接收由溫度感測器140所感測到的溫度資訊。特別是,本實施例之控制器150會利用所接收的脈寬調變信號PWM1控制第一處理器110與第二處理器120在短時間交互進行超頻,並確保第一處理器110與第二處理器120不會發生同時超頻的情況,以在同樣的溫度 表現下令電腦系統100提供更高的效能。 The controller 150 is coupled to the temperature sensor 140. In this embodiment, the controller 150 is, for example, an embedded controller (EC) for controlling the power of the computer system 100, an input device (such as a keyboard or a touch pad, etc.), and can receive a sense of temperature. The temperature information sensed by the detector 140. In particular, the present embodiment the controller 150 of the embodiment will utilize pulse width modulation signal PWM received by control processor 110 of the first and the second processor 120 in a short time overclocking interaction, and to ensure that the first and the second processor 110 The second processor 120 does not experience simultaneous overclocking to provide higher performance to the computer system 100 at the same temperature performance.

以下將以另一實施例配合圖1來說明在電腦系統100實現功率分享方法的詳細流程。圖2是依照本發明之一實施例所繪示之功率分享方法的流程圖。請同時參閱圖1與圖2。 A detailed flow of implementing the power sharing method in the computer system 100 will be described below with another embodiment in conjunction with FIG. 2 is a flow chart of a power sharing method according to an embodiment of the invention. Please also refer to Figure 1 and Figure 2.

首先如步驟S210所示,控制器150接收脈寬調變信號PWM1,其具有一責任週期(duty cycle)。以圖3A為例,脈寬調變信號PWM1的責任週期為20%,亦即在每一時脈週期T內,脈寬調變信號PWM1有20%的時間會處於高邏輯準位H,而有80%的時間處於低邏輯準位L。而在圖3B所示之實施例中,脈寬調變信號PWM1的責任週期為60%,亦即在每一時脈週期T內,脈寬調變信號PWM1有60%的時間會處於高邏輯準位H,而有40%的時間處於低邏輯準位L。 First, as shown in step S210, the controller 150 receives the pulse width modulation signal PWM 1 having a duty cycle. Taking FIG. 3A as an example, the duty cycle of the PWM signal PWM 1 is 20%, that is, within each clock cycle T, the pulse width modulation signal PWM 1 is at a high logic level H for 20% of the time. And 80% of the time is at the low logic level L. In the embodiment shown in FIG. 3B, the duty cycle of the PWM signal PWM 1 is 60%, that is, the pulse width modulation signal PWM1 is 60% of the time in each clock cycle T. Level H, and 40% of the time is at low logic level L.

接著在步驟S220中,控制器150將系統散熱功率定義為第一處理器110與第二處理器120個別預設之溫度設計功率(Thermal Design Power,TDP)的總和。在本實施例中,第一處理器110與第二處理器120的溫度設計功率是可被調整的,然而其初始預設值會明確定義在處理器的規格中,控制器150可透過晶片組130讀取第一處理器110與第二處理器120個別之溫度設計功率的預設值。 Next, in step S220, the controller 150 defines the system heat dissipation power as the sum of the temperature design power (TDP) preset by the first processor 110 and the second processor 120. In this embodiment, the temperature design power of the first processor 110 and the second processor 120 can be adjusted, however, the initial preset value is clearly defined in the specification of the processor, and the controller 150 can pass through the chipset. The preset value of the temperature design power of the first processor 110 and the second processor 120 is read.

在步驟S230中,控制器150判斷系統散熱功率是否大於或等於預設上限值。若系統散熱功率小於預設上限值,表示電腦系統100在目前尚不需執行功率分享的機 制,因而結束本實施例之功率分享方法的流程。 In step S230, the controller 150 determines whether the system heat dissipation power is greater than or equal to a preset upper limit value. If the system cooling power is less than the preset upper limit, it means that the computer system 100 does not need to perform power sharing at present. Thus, the flow of the power sharing method of the present embodiment is ended.

然而,倘若系統散熱功率大於或等於預設上限值,接著如步驟S240所示,控制器150利用脈寬調變信號PWM1調整第一處理器110與第二處理器120的功率。其中,利用升壓(boost)來提高第一處理器110之功率的時間與利用升壓提高第二處理器120之功率的時間互不重疊。 However, if the system heat dissipation power is greater than or equal to the preset upper limit value, then the controller 150 adjusts the power of the first processor 110 and the second processor 120 by using the pulse width modulation signal PWM 1 as shown in step S240. The time for boosting the power of the first processor 110 by boosting does not overlap with the time for boosting the power of the second processor 120 by boosting.

在本實施例中,控制器150在脈寬調變信號PWM1處於第一邏輯準位(例如,高邏輯準位)時以升壓提高第一處理器110之功率,並在脈寬調變信號PWM1處於第二邏輯準位(例如,低邏輯準位)時以升壓提高第二處理器120之功率。進一步來說,控制器150在脈寬調變信號PWM1處於高邏輯準位時,利用升壓將第一處理器110之功率提高到功率上限(例如,溫度設計功率的最大值),而在脈寬調變信號PWM1處於低邏輯準位時,則將第一處理器110之功率調回溫度設計功率的原有設定值。相反地,第二處理器120之功率則會在脈寬調變信號PWM1處於低邏輯準位時被提升至最大值,並且在脈寬調變信號PWM1處於高邏輯準位時被調回原設定之功率。由於脈寬調變信號PWM1會不斷地在高邏輯準位與低邏輯準位之間切換,因此透過此機制可令第一處理器110與第二處理器120快速提高功率以增加電腦系統100的效能,並可在系統溫度尚未來得及升高之前就將功率拉低回原設定值,而能有效散熱來避免系統溫度過高。此外,由於脈寬調變信號PWM1不會同時處於高邏輯準位與低邏輯準位,故此機制亦可確 保不會同時將第一處理器110與第二處理器120的功率提升到上限,以確保不造成系統溫度激增。 In this embodiment, the controller 150 boosts the power of the first processor 110 by boosting when the pulse width modulation signal PWM 1 is at the first logic level (eg, a high logic level), and adjusts the pulse width. The power of the second processor 120 is boosted by the boost when the signal PWM 1 is at the second logic level (eg, low logic level). Further, the controller 150 boosts the power of the first processor 110 to the upper power limit (eg, the maximum value of the temperature design power) by using the boost when the pulse width modulation signal PWM 1 is at the high logic level. When the pulse width modulation signal PWM 1 is at the low logic level, the power of the first processor 110 is adjusted back to the original set value of the temperature design power. Conversely, the power of the second processor 120 is boosted to a maximum value when the pulse width modulation signal PWM 1 is at a low logic level, and is reset when the pulse width modulation signal PWM 1 is at a high logic level. The originally set power. Since the PWM signal PWM 1 is continuously switched between the high logic level and the low logic level, the first processor 110 and the second processor 120 can be quickly increased in power to increase the computer system 100. The efficiency and the power can be pulled back to the original set value before the system temperature has risen, and the heat can be effectively cooled to avoid the system temperature being too high. In addition, since the pulse width modulation signal PWM 1 is not at the high logic level and the low logic level at the same time, the mechanism can also ensure that the power of the first processor 110 and the second processor 120 are not raised to the upper limit at the same time. To ensure that no system temperature surges.

接著在步驟S250中,控制器150從溫度感測器140取得目前的系統溫度,並將其與之前的溫度相比以判斷電腦系統100的系統溫度是否上升。 Next, in step S250, the controller 150 obtains the current system temperature from the temperature sensor 140 and compares it with the previous temperature to determine whether the system temperature of the computer system 100 has risen.

若電腦系統100的系統溫度上升,如步驟S260所示,控制器150減少脈寬調變信號PWM1的責任週期或減少系統散熱功率,以縮短利用升壓來提高功率的時間。詳言之,若控制器150判斷責任週期未到達週期下限(例如10%,但本發明並不以此為限),則減少責任週期以縮短利用升壓來提高第一處理器110或第二處理器120之功率的時間。而若控制器150判斷責任週期已到達週期下限,則減少系統散熱功率來縮短利用升壓以提高第一處理器110或第二處理器120之功率的時間。 If the system temperature of the computer system 100 rises, as shown in step S260, the controller 150 reduces the duty cycle of the pulse width modulation signal PWM 1 or reduces the system heat dissipation power to shorten the time for boosting power by boosting. In detail, if the controller 150 determines that the duty cycle has not reached the lower limit of the period (for example, 10%, but the invention is not limited thereto), the duty cycle is reduced to shorten the use of the boost to improve the first processor 110 or the second. The time of the power of the processor 120. If the controller 150 determines that the duty cycle has reached the lower limit of the cycle, the system heat dissipation power is reduced to shorten the time for boosting the power of the first processor 110 or the second processor 120.

若電腦系統100的系統溫度下降,則如步驟S270所示,控制器150增加脈寬調變信號PWM1的責任週期或增加系統散熱功率,以增長利用升壓來提高功率的時間。具體來說,若控制器150判斷責任週期未到達週期上限(例如90%,但本發明並不以此為限),則增加責任週期以增長利用升壓來提高第一處理器110或第二處理器120之功率的時間,而若控制器150判斷責任週期已到達週期上限,則增加系統散熱功率以增長利用升壓以提高第一處理器110或第二處理器120之功率的時間。 If the system temperature of the computer system 100 drops, as shown in step S270, the controller 150 increases the duty cycle of the pulse width modulation signal PWM 1 or increases the system heat dissipation power to increase the time for boosting power by boosting. Specifically, if the controller 150 determines that the responsibility period does not reach the upper limit of the period (for example, 90%, but the invention is not limited thereto), the duty cycle is increased to increase the boosting to improve the first processor 110 or the second. The time of the power of the processor 120, and if the controller 150 determines that the duty cycle has reached the upper limit of the cycle, the system heat dissipation power is increased to increase the time during which the boost is utilized to increase the power of the first processor 110 or the second processor 120.

在經過上述步驟後,脈寬調變信號PWM1的責任週期 或系統散熱功率會有所增減,接著本實施例之功率分享方法將回到步驟S230,再次判斷目前的系統散熱功率是否大於或等於預設上限值。若是,則依據目前之脈寬調變信號PWM1的責任週期執行步驟S240之動作,並重新根據步驟S250的判斷結果決定執行步驟S260或步驟S270。本實施例之功率分享方法將重複步驟S230至步驟S270直到系統散熱功率小於預設上限值為止。 After the above steps, the duty cycle or the system heat dissipation power of the pulse width modulation signal PWM 1 may increase or decrease. Then, the power sharing method of the embodiment returns to step S230 to determine whether the current system heat dissipation power is greater than or Equal to the preset upper limit. If yes, the action of step S240 is performed according to the duty cycle of the current pulse width modulation signal PWM 1 , and step S260 or step S270 is determined to be performed according to the determination result of step S250. The power sharing method of this embodiment will repeat steps S230 to S270 until the system heat dissipation power is less than the preset upper limit value.

如圖2所示,本實施例是根據電腦系統100之系統溫度的升降而控管利用升壓來提高處理器之功率的時間。如此一來可確保在提升電腦系統100之效能的同時,不對溫度表現造成負面影響。 As shown in FIG. 2, this embodiment controls the time for boosting the power of the processor by boosting according to the rise and fall of the system temperature of the computer system 100. This ensures that the performance of the computer system 100 is improved while not adversely affecting temperature performance.

從另一方面來看,由於在大部份的使用狀況下第一處理器110與第二處理器120並不會同時處於高使用率狀態,因此以下的實施例是說明如何在有限的散熱空間下,根據處理器的使用率而調控其功耗,以提升系統效能。 On the other hand, since the first processor 110 and the second processor 120 are not in a high usage state at the same time in most of the usage conditions, the following embodiments illustrate how to provide limited heat dissipation space. The power consumption is adjusted according to the usage rate of the processor to improve system performance.

圖4是依照本發明之另一實施例所繪示之功率分享方法的流程圖。請同時參閱圖1與圖4。由於圖4之步驟S405至步驟S420與圖2之步驟S210至步驟S240相同或相似,故在此不再贅述。 4 is a flow chart of a power sharing method according to another embodiment of the present invention. Please also refer to Figure 1 and Figure 4. Since steps S405 to S420 of FIG. 4 are the same as or similar to steps S210 to S240 of FIG. 2, they are not described herein again.

在本實施例中,控制器150在利用脈寬調變信號PWM1調整第一處理器110與第二處理器120的功率之後,接著如步驟S425所示,決定是否需要判斷第一處理器110的使用率。具體來說,若目前處於動態設定第一處理器110之功率上限的釋放週期(release cycle),則不需 判斷第一處理器110的使用率,因此本實施例之功率分享方法的流程將進入步驟S460。反之,若不處於上述釋放週期,則需要判斷第一處理器110的使用率,因此如步驟S430所示,控制器150取得第一處理器110的目前使用功率,並根據第一處理器110的目前使用功率與溫度設計功率來判斷第一處理器是處於高使用率狀態或低使用率狀態。在本實施例中,倘若目前使用功率與溫度設計功率的比值小於一第一預設值(例如,50%),則控制器150判定第一處理器110是處於低使用率狀態,而倘若目前使用功率與溫度設計功率的比值大於一第二預設值(例如,80%),則控制器150判定第一處理器110是處於高使用率狀態。 In this embodiment, after the controller 150 adjusts the power of the first processor 110 and the second processor 120 by using the pulse width modulation signal PWM 1 , then, as shown in step S425, it is determined whether the first processor 110 needs to be determined. Usage rate. Specifically, if the release cycle of the power upper limit of the first processor 110 is dynamically set, the usage rate of the first processor 110 is not required to be determined, so the flow of the power sharing method of the embodiment will enter. Step S460. On the other hand, if it is not in the above release period, it is necessary to determine the usage rate of the first processor 110. Therefore, as shown in step S430, the controller 150 obtains the current usage power of the first processor 110, and according to the first processor 110. Power and temperature design power is currently used to determine whether the first processor is in a high usage state or a low usage state. In this embodiment, if the ratio of the currently used power to the temperature design power is less than a first preset value (eg, 50%), the controller 150 determines that the first processor 110 is in a low usage state, and if currently If the ratio of power to temperature design power is greater than a second predetermined value (eg, 80%), the controller 150 determines that the first processor 110 is in a high usage state.

根據上述判斷機制,若步驟S435的判斷結果為是(即,第一處理器110處於低使用率狀態),接著在步驟S440中,控制器150降低第一處理器110的溫度設計功率,並將第二處理器120的溫度設計功率設定為系統散熱功率與第一處理器110目前之溫度設計功率的差值。也就是說,在此情況下第一處理器110的溫度設計功率會被調降,而多餘的功率則被分配轉移給第二處理器120。亦即,第二處理器120的溫度設計功率會被調升。 According to the above judgment mechanism, if the result of the determination in step S435 is YES (ie, the first processor 110 is in the low usage state), then in step S440, the controller 150 decreases the temperature design power of the first processor 110, and The temperature design power of the second processor 120 is set as the difference between the system heat dissipation power and the current temperature design power of the first processor 110. That is to say, in this case, the temperature design power of the first processor 110 is adjusted down, and the excess power is distributed to the second processor 120. That is, the temperature design power of the second processor 120 is boosted.

若步驟S435的判斷結果為否,則在步驟S445的判斷結果為是時(即,第一處理器110處於高使用率狀態),如步驟S450所示,控制器150增加第一處理器110的溫度設計功率,並將第二處理器120的溫度設計功率設定為系統散熱功率與第一處理器110目前之溫度設計功率的差 值。亦即,當第一處理器110處於高使用率狀態時,本實施例之功率分享方法將減少第二處理器120的溫度設計功率,並將多餘之功率移轉到第一處理器110,以提升第一處理器110的溫度設計功率。 If the result of the determination in step S435 is negative, then if the result of the determination in step S445 is YES (ie, the first processor 110 is in the high usage state), the controller 150 increases the first processor 110 as shown in step S450. The temperature design power, and the temperature design power of the second processor 120 is set as the difference between the system heat dissipation power and the current temperature design power of the first processor 110. value. That is, when the first processor 110 is in the high usage state, the power sharing method of the embodiment reduces the temperature design power of the second processor 120, and transfers the excess power to the first processor 110 to The temperature design power of the first processor 110 is boosted.

簡言之,步驟S430至步驟S450是根據第一處理器110的負載來將系統散熱功率動態調配給第一處理器110與第二處理器120使用。接下來的步驟S460至步驟S480則是依據電腦系統100之系統溫度是否上升來決定如何控管提高處理器之功率的時間,由於步驟S460至步驟S480與圖2之步驟S250至步驟S270相同或相似,故在此不再贅述。圖4所示之功率分享方法將重複步驟S415至步驟S480直到系統散熱功率小於預設上限值為止。 In short, steps S430 to S450 dynamically allocate system cooling power to the first processor 110 and the second processor 120 according to the load of the first processor 110. The next step S460 to step S480 is to determine how to control the power of the processor according to whether the system temperature of the computer system 100 rises. Since steps S460 to S480 are the same as or similar to steps S250 to S270 of FIG. Therefore, it will not be repeated here. The power sharing method shown in FIG. 4 will repeat steps S415 to S480 until the system heat dissipation power is less than the preset upper limit value.

在上述實施例中,控制器150是藉由一組硬體線路以監測第一處理器110及第二處理器120的目前使用功率,並取得第一處理器110及第二處理器120之溫度設計功率的上限設定。在判斷其一處理器的負載較小時,將多餘之功率移轉給另一處理器,反之亦然,進而實現系統散熱功率的動態分配。 In the above embodiment, the controller 150 monitors the current usage power of the first processor 110 and the second processor 120 by using a set of hardware lines, and obtains the temperatures of the first processor 110 and the second processor 120. The upper limit setting of the design power. When it is judged that the load of one processor is small, the excess power is transferred to another processor, and vice versa, thereby realizing the dynamic allocation of the system heat dissipation power.

此外,即便在第一處理器110與第二處理器120均處於高功率需求之情況下,亦因採用快速且交替地將第一處理器110及第二處理器120之功率切換至上限,如此可以確保系統溫度不過度上升。而快速的切換功率上限則能使第一處理器110與第二處理器120在短時間內以高速運作,以達成更高的系統效能表現。 In addition, even if both the first processor 110 and the second processor 120 are in high power demand, the power of the first processor 110 and the second processor 120 is switched to the upper limit quickly and alternately. It can be ensured that the system temperature does not rise excessively. The fast switching power cap enables the first processor 110 and the second processor 120 to operate at high speed in a short time to achieve higher system performance.

雖然在上述實施例中控制器150是利用一個脈寬調變信號PWM1來提升兩處理器的功率上限,但在另一實施例中,控制器150亦可接收兩個脈寬調變信號PWM1與PWM2,並分別以這兩個脈寬調變信號調控兩個處理器。 Although in the above embodiment, the controller 150 uses a pulse width modulation signal PWM 1 to increase the power upper limit of the two processors, in another embodiment, the controller 150 can also receive two pulse width modulation signals PWM. 1 and PWM 2 , and control the two processors with the two pulse width modulation signals respectively.

詳言之,脈寬調變信號PWM2與脈寬調變信號PWM1互為反相。舉例來說,如圖5所示,脈寬調變信號PWM1與PWM2的責任週期均為50%,然而在每一時脈週期內當脈寬調變信號PWM1處於高邏輯準位時,脈寬調變信號PWM2便處於低邏輯準位,反之亦然。並且,控制器150調整脈寬調變信號PWM1之責任週期的長短會同時對脈寬調變信號PWM1與PWM2造成影響。具體而言,當脈寬調變信號PWM1的責任週期被控制器150調整,脈寬調變信號PWM2因要維持與脈寬調變信號PWM1互為反相,因此其責任週期也會隨之改變。 In detail, the pulse width modulation signal PWM 2 and the pulse width modulation signal PWM 1 are mutually inverted. For example, as shown in FIG. 5, the duty cycle of the PWM signals PWM 1 and PWM 2 are both 50%, but when the PWM signal PWM 1 is at a high logic level in each clock cycle, The pulse width modulation signal PWM 2 is at a low logic level and vice versa. Then, the controller 150 adjusts the PWM responsibility 1 cycle PWM signal will also affect the length of the pulse width modulation signals PWM 1 and PWM 2. Specifically, when the duty cycle of the pulse width modulation signal PWM 1 is adjusted by the controller 150, the pulse width modulation signal PWM 2 is maintained in phase with the pulse width modulation signal PWM 1 , so the duty cycle thereof is also Change with it.

在本實施例中,控制器150將在脈寬調變信號PWM1處於第一邏輯準位(例如,高邏輯準位)時以升壓提高第一處理器110之功率,並在脈寬調變信號PWM2處於第一邏輯準位時以升壓提高第二處理器120之功率。由於脈寬調變信號PWM1與PWM2互為反相,因此可避免在同時將第一處理器110與第二處理器120的功率提高至上限的情況。 In this embodiment, the controller 150 boosts the power of the first processor 110 with the boost when the pulse width modulation signal PWM 1 is at the first logic level (eg, a high logic level), and adjusts the pulse width. When the variable signal PWM 2 is at the first logic level, the power of the second processor 120 is boosted by the boost. Since the pulse width modulation signal PWM 1 and the PWM 2 are mutually inverted, it is possible to avoid the case where the power of the first processor 110 and the second processor 120 is simultaneously raised to the upper limit.

綜上所述,本發明所述之電腦系統及其功率分享方法能在系統效能與溫度間取得平衡。進一步來說,由於電腦系統的表面溫度係取決於內部熱量的累積程度,而依據處 理器之使用率動態分配系統散熱功率並藉由脈寬調變信號控制兩處理器高速交替切換至功率上限,則可確保處理器在短時間高速運作,同時將電腦系統的熱源控制在一定程度之內,據此讓電腦系統在不失去系統效能的同時而有更佳的溫度表現。 In summary, the computer system and the power sharing method thereof according to the present invention can achieve a balance between system performance and temperature. Further, since the surface temperature of the computer system depends on the accumulation of internal heat, the basis is The usage rate of the processor dynamically allocates the cooling power of the system and controls the high-speed switching of the two processors to the upper limit of the power by the pulse width modulation signal. This ensures that the processor operates at a high speed in a short time while controlling the heat source of the computer system to a certain extent. Within this, the computer system can be used to achieve better temperature performance without losing system performance.

雖然本發明已以實施例揭露如上,然其並非用以限定本發明,任何所屬技術領域中具有通常知識者,在不脫離本發明之精神和範圍內,當可作些許之更動與潤飾,故本發明之保護範圍當視後附之申請專利範圍所界定者為準。 Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

100‧‧‧電腦系統 100‧‧‧ computer system

110‧‧‧第一處理器 110‧‧‧First processor

120‧‧‧第二處理器 120‧‧‧second processor

130‧‧‧晶片組 130‧‧‧chipset

140‧‧‧溫度感測器 140‧‧‧temperature sensor

150‧‧‧控制器 150‧‧‧ Controller

PWM1、PWM2‧‧‧脈寬調變信號 PWM 1 , PWM 2 ‧‧‧ pulse width modulation signal

S210~S270‧‧‧本發明之一實施例所述之功率分享方法的各步驟 S210~S270‧‧‧ steps of the power sharing method according to an embodiment of the present invention

T‧‧‧時脈週期 T‧‧‧ clock cycle

H‧‧‧高邏輯準位 H‧‧‧High logic level

L‧‧‧低邏輯準位 L‧‧‧Low logic level

S405~S480‧‧‧本發明之另一實施例所述之功率分享方法的各步驟 S405~S480‧‧‧ steps of the power sharing method according to another embodiment of the present invention

圖1是依照本發明之一實施例所繪示之電腦系統的方塊圖。 1 is a block diagram of a computer system in accordance with an embodiment of the present invention.

圖2是依照本發明之一實施例所繪示之功率分享方法的流程圖。 2 is a flow chart of a power sharing method according to an embodiment of the invention.

圖3A、3B是依照本發明之一實施例所繪示之脈寬調變信號之責任週期的示意圖。 3A and 3B are schematic diagrams showing a duty cycle of a pulse width modulation signal according to an embodiment of the invention.

圖4是依照本發明之另一實施例所繪示之功率分享方法的流程圖。 4 is a flow chart of a power sharing method according to another embodiment of the present invention.

圖5是依照本發明之另一實施例所繪示之不同脈寬調變信號的責任週期的示意圖。 FIG. 5 is a schematic diagram of a duty cycle of different pulse width modulation signals according to another embodiment of the present invention.

S210~S270‧‧‧本發明之一實施例所述之功率分享方法的各步驟 S210~S270‧‧‧ steps of the power sharing method according to an embodiment of the present invention

Claims (12)

一種功率分享方法,用於包括一第一處理器與一第二處理器的一電腦系統,該方法包括下列步驟:a.接收一第一脈寬調變信號,其中該第一脈寬調變信號具有一責任週期;b.定義一系統散熱功率為該第一處理器與該第二處理器個別預設之一溫度設計功率(Thermal Design Power,TDP)的一總和;c.判斷該系統散熱功率是否大於或等於一預設上限值;d.若是,利用該第一脈寬調變信號調整該第一處理器與該第二處理器的功率,其中利用升壓提高該第一處理器之功率的時間與利用升壓提高該第二處理器之功率的時間互不重疊;e.判斷該電腦系統的一系統溫度是否上升;f.若是,減少該責任週期或該系統散熱功率以縮短利用升壓以提高功率的時間;g.若否,增加該責任週期或該系統散熱功率以增長利用升壓以提高功率的時間;以及h.重複步驟c至步驟g直到該系統散熱功率小於該預設上限值。 A power sharing method for a computer system including a first processor and a second processor, the method comprising the steps of: a. receiving a first pulse width modulation signal, wherein the first pulse width modulation The signal has a duty cycle; b. defines a system heat dissipation power as a sum of a temperature design power (TDP) of the first processor and the second processor; c. determining the heat dissipation of the system Whether the power is greater than or equal to a predetermined upper limit; d. if yes, adjusting the power of the first processor and the second processor by using the first pulse width modulation signal, wherein the first processor is boosted by boosting The time of the power does not overlap with the time for boosting the power of the second processor; e. determining whether the temperature of a system of the computer system is rising; f. if yes, reducing the duty cycle or the heat dissipation power of the system to shorten Use boost to increase power time; g. If not, increase the duty cycle or the system heat dissipation power to increase the time to use boost to increase power; and h. repeat steps c through g until the system heat dissipation power Less than the preset upper limit. 如申請專利範圍第1項所述之功率分享方法,其中該步驟f包括:判斷該責任週期是否到達一週期下限; 若否,則減少該責任週期;以及若是,則減少該系統散熱功率。 The power sharing method of claim 1, wherein the step f comprises: determining whether the responsibility period reaches a lower limit of a period; If not, the duty cycle is reduced; and if so, the system cooling power is reduced. 如申請專利範圍第1項所述之功率分享方法,其中該步驟g包括:判斷該責任週期是否到達一週期上限;若否,則增加該責任週期;以及若是,則增加該系統散熱功率。 The power sharing method of claim 1, wherein the step g comprises: determining whether the responsibility period reaches an upper limit of the period; if not, increasing the responsibility period; and if so, increasing the heat dissipation power of the system. 如申請專利範圍第1項所述之功率分享方法,其中在該步驟d至該步驟e之間,該方法更包括:在需要判斷該第一處理器的使用率時,取得該第一處理器的一目前使用功率;根據該第一處理器的該目前使用功率與該溫度設計功率判斷該第一處理器處於一高使用率狀態或一低使用率狀態;若該第一處理器處於該低使用率狀態,則降低該第一處理器的該溫度設計功率,並將該第二處理器的該溫度設計功率設定為該系統散熱功率與該第一處理器目前之該溫度設計功率的差值;以及若該第一處理器處於該高使用率狀態,則增加該第一處理器的該溫度設計功率,並將該第二處理器的該溫度設計功率設定為該系統散熱功率與該第一處理器目前之該溫度設計功率的差值。 The power sharing method of claim 1, wherein the method further comprises: when the usage of the first processor needs to be determined, obtaining the first processor a current usage power; determining, according to the current usage power of the first processor and the temperature design power, that the first processor is in a high usage state or a low usage state; if the first processor is at the low The usage state, the temperature design power of the first processor is decreased, and the temperature design power of the second processor is set to a difference between the system heat dissipation power and the current temperature design power of the first processor. And if the first processor is in the high usage state, increasing the temperature design power of the first processor, and setting the temperature design power of the second processor to the system cooling power and the first The difference between the current design power of the processor. 如申請專利範圍第1項所述之功率分享方法,其中該步驟d包括: 在該第一脈寬調變信號處於一第一邏輯準位時以升壓提高該第一處理器之功率;以及在該第一脈寬調變信號處於一第二邏輯準位時以升壓提高該第二處理器之功率。 The power sharing method of claim 1, wherein the step d includes: Raising the power of the first processor by boosting when the first pulse width modulation signal is at a first logic level; and boosting when the first pulse width modulation signal is at a second logic level Increase the power of the second processor. 如申請專利範圍第1項所述之功率分享方法,其中該步驟d包括:提供一第二脈寬調變信號,其中該第二脈寬調變信號與該第一脈寬調變信號互為反相;在該第一脈寬調變信號處於一第一邏輯準位時以升壓提高該第一處理器之功率;以及在該第二脈寬調變信號處於該第一邏輯準位時以升壓提高該第二處理器之功率。 The power sharing method of claim 1, wherein the step d includes: providing a second pulse width modulation signal, wherein the second pulse width modulation signal and the first pulse width modulation signal are mutually Inverting; increasing the power of the first processor by boosting when the first pulse width modulation signal is at a first logic level; and when the second pulse width modulation signal is at the first logic level The power of the second processor is increased by boosting. 一種電腦系統,包括:一第一處理器;一第二處理器;一晶片組,耦接該第一處理器與一第二處理器;一溫度感測器,以感測該電腦系統的一系統溫度;以及一控制器,耦接該晶片組與該溫度感測器,該控制器接收一第一脈寬調變信號,其中該第一脈寬調變信號具有一責任週期,以及定義一系統散熱功率為該第一處理器與該第二處理器個別預設之一溫度設計功率的一總和;該控制器在判斷該系統散熱功率大於或等於一預設上限值時,利用該第一脈寬調變信號調整該第一處理器與 該第二處理器的功率,其中利用升壓提高該第一處理器之功率的時間與利用升壓提高該第二處理器之功率的時間互不重疊,並判斷該電腦系統的該系統溫度是否上升,在判定該系統溫度上升時減少該責任週期或該系統散熱功率以縮短利用升壓以提高功率的時間,以及在判定該系統溫度下降時增加該責任週期或該系統散熱功率以增長利用升壓以提高功率的時間,該控制器重複前述操作直到該系統散熱功率小於該預設上限值。 A computer system comprising: a first processor; a second processor; a chipset coupled to the first processor and a second processor; and a temperature sensor to sense one of the computer systems a system temperature coupled to the chip set and the temperature sensor, the controller receiving a first pulse width modulation signal, wherein the first pulse width modulation signal has a duty cycle, and defining a The system heat dissipation power is a sum of one of the temperature preset design powers of the first processor and the second processor; the controller uses the first when determining that the system heat dissipation power is greater than or equal to a predetermined upper limit value a pulse width modulation signal is adjusted to the first processor and The power of the second processor, wherein the time for boosting the power of the first processor by boosting does not overlap with the time for boosting the power of the second processor, and determining whether the temperature of the system of the computer system is Rising, reducing the duty cycle or the heat dissipation power of the system when determining the temperature rise of the system to shorten the time for using the boost to increase the power, and increasing the duty cycle or the heat dissipation power of the system to increase the utilization of the rise when the temperature of the system is determined to decrease. When the voltage is increased to increase the power, the controller repeats the foregoing operation until the system heat dissipation power is less than the preset upper limit value. 如申請專利範圍第7項所述之電腦系統,其中若該控制器判斷該責任週期未到達一週期下限,則減少該責任週期以縮短利用升壓來提高功率的時間,而若該控制器判斷該責任週期到達該週期下限,則減少該系統散熱功率以縮短利用升壓來提高功率的時間。 The computer system of claim 7, wherein if the controller determines that the duty cycle has not reached a lower limit of a cycle, reducing the duty cycle to shorten the time for boosting power, and if the controller determines When the duty cycle reaches the lower limit of the cycle, the heat dissipation power of the system is reduced to shorten the time for boosting power by boosting. 如申請專利範圍第7項所述之電腦系統,其中若該控制器判斷該責任週期未到達一週期上限,則增加該責任週期以增長利用升壓來提高功率的時間,而若該控制器判斷該責任週期到達該週期上限,則增加該系統散熱功率以增長利用升壓來提高功率的時間。 The computer system of claim 7, wherein if the controller determines that the duty cycle does not reach a cycle upper limit, increasing the responsibility cycle to increase the time for boosting power by using the boost, and if the controller determines When the duty cycle reaches the upper limit of the cycle, the heat dissipation power of the system is increased to increase the time for boosting power by using boost. 如申請專利範圍第7項所述之電腦系統,其中該控制器在利用該第一脈寬調變信號調整該第一處理器與該第二處理器的功率以及判斷該電腦系統的該系統溫度是否上升之間,更包括在需要判斷該第一處理器的使用率時取得該第一處理器的一目前使用功率,根據該第一處理器的該目前使用功率與該溫度設計功率判斷該第一處理器是處 於一高使用率狀態或一低使用率狀態,若該第一處理器處於該低使用率狀態,則降低該第一處理器的該溫度設計功率,並將該第二處理器的該溫度設計功率設定為該系統散熱功率與該第一處理器目前之該溫度設計功率的差值,若該第一處理器處於該高使用率狀態,則增加該第一處理器的該溫度設計功率,並將該第二處理器的該溫度設計功率設定為該系統散熱功率與該第一處理器目前之該溫度設計功率的差值。 The computer system of claim 7, wherein the controller adjusts the power of the first processor and the second processor by using the first pulse width modulation signal and determines the system temperature of the computer system Whether it is rising, or not, when determining the usage rate of the first processor, obtaining a current usage power of the first processor, and determining the first according to the current usage power of the first processor and the temperature design power of the first processor a processor is at In a high usage state or a low usage state, if the first processor is in the low usage state, reducing the temperature design power of the first processor, and designing the temperature of the second processor The power is set to a difference between the system cooling power and the current temperature design power of the first processor, and if the first processor is in the high usage state, increasing the temperature design power of the first processor, and The temperature design power of the second processor is set to a difference between the system heat dissipation power and the current temperature design power of the first processor. 如申請專利範圍第7項所述之電腦系統,其中該控制器在該第一脈寬調變信號處於一第一邏輯準位時以升壓提高該第一處理器之功率,並在該第一脈寬調變信號處於一第二邏輯準位時以升壓提高該第二處理器之功率。 The computer system of claim 7, wherein the controller increases the power of the first processor by boosting when the first pulse width modulation signal is at a first logic level, and When the pulse width modulation signal is at a second logic level, the power of the second processor is increased by boosting. 如申請專利範圍第7項所述之電腦系統,其中該控制器接收一第二脈寬調變信號,其中該第二脈寬調變信號與該第一脈寬調變信號互為反相,該控制器在該第一脈寬調變信號處於一第一邏輯準位時以升壓提高該第一處理器之功率,並在該第二脈寬調變信號處於該第一邏輯準位時以升壓提高該第二處理器之功率。 The computer system of claim 7, wherein the controller receives a second pulse width modulation signal, wherein the second pulse width modulation signal and the first pulse width modulation signal are mutually inverted. The controller increases the power of the first processor by boosting when the first pulse width modulation signal is at a first logic level, and when the second pulse width modulation signal is at the first logic level The power of the second processor is increased by boosting.
TW101106064A 2012-02-23 2012-02-23 Computer system and power sharing method thereof TWI483098B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW101106064A TWI483098B (en) 2012-02-23 2012-02-23 Computer system and power sharing method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW101106064A TWI483098B (en) 2012-02-23 2012-02-23 Computer system and power sharing method thereof

Publications (2)

Publication Number Publication Date
TW201335735A TW201335735A (en) 2013-09-01
TWI483098B true TWI483098B (en) 2015-05-01

Family

ID=49627405

Family Applications (1)

Application Number Title Priority Date Filing Date
TW101106064A TWI483098B (en) 2012-02-23 2012-02-23 Computer system and power sharing method thereof

Country Status (1)

Country Link
TW (1) TWI483098B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6526491B2 (en) * 2001-03-22 2003-02-25 Sony Corporation Entertainment Inc. Memory protection system and method for computer architecture for broadband networks
TWI317468B (en) * 2006-02-20 2009-11-21 Ite Tech Inc Method for controlling power consumption and multi-processor system using the same
TWI348641B (en) * 2006-07-31 2011-09-11 Intel Corp System and method for controlling processor low power states

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6526491B2 (en) * 2001-03-22 2003-02-25 Sony Corporation Entertainment Inc. Memory protection system and method for computer architecture for broadband networks
TWI317468B (en) * 2006-02-20 2009-11-21 Ite Tech Inc Method for controlling power consumption and multi-processor system using the same
TWI348641B (en) * 2006-07-31 2011-09-11 Intel Corp System and method for controlling processor low power states

Also Published As

Publication number Publication date
TW201335735A (en) 2013-09-01

Similar Documents

Publication Publication Date Title
JP5189921B2 (en) Computer heat dissipation system
JP4376897B2 (en) Memory controller considering processor power state
US10444819B2 (en) Techniques to control computational resources for an electronic device
JP2007233782A (en) Control method for heating value, and computer
JP2009522688A (en) System and method for operating integrated circuit components at independent frequencies and / or voltages
CN102314213A (en) Computer system capable of adjusting operating frequency dynamically
JP2015130035A (en) Semiconductor device and control method
KR101707096B1 (en) Generic host-based controller latency method and apparatus
JP6103783B2 (en) Power control device
WO2016138765A1 (en) Method for controlling a plurality of hardware modules and associated controller and system
US11507176B2 (en) Information processing apparatus and control method
TW201602771A (en) System on chip, method thereof, and device including the same
TW201447544A (en) System and method for temperature driven selection of voltage modes in a portable computing device
JP7218410B1 (en) Information processing device and control method
TWI483098B (en) Computer system and power sharing method thereof
US11163345B2 (en) Electronic device to control temperature and computing performance of at least one processing unit and system and method thereof
CN103309423B (en) Computer system and power sharing method thereof
US20140089695A1 (en) Energy-saving device
EP2775395B1 (en) Integrated circuit, electronic device and instruction scheduling method
KR20160085029A (en) Computer-system power control unit and method thereof
JP2004164530A (en) Battery condition reading method by operation system of portable computer
TWI487844B (en) Method for fan monitoring and server system using the same
CN111381643A (en) Component power consumption switching circuit and mainboard thereof
JP2010205363A (en) Sdram control system, sdram control method, sdram control program
EP2927806A1 (en) Integrated circuit, electronic device and ic operation method