本發明係關於一資訊處理平台，尤其關於一種多中央處理單元偵錯切換之方法。The present invention relates to an information processing platform, and more particularly to a method for multi-central processing unit debug switching.
隨著資訊科技的進步，開發出了多中央處理單元之運算平台，以應付日趨複雜的資料運算及處理需求。相較於一般單中央處理單元之運算平台，多中央處理單元之運算平台不僅較為穩定，整體運算處理能力亦提升許多。因此，不論是企業之伺服器或者家庭之個人電腦，皆適合使用多中央處理單元。With the advancement of information technology, a computing platform for multiple central processing units has been developed to cope with the increasingly complex data computing and processing needs. Compared with the computing platform of a single central processing unit, the computing platform of the multi-central processing unit is not only stable, but also the overall computing processing capability is much improved. Therefore, it is suitable to use a multi-central processing unit whether it is a server of a company or a personal computer of a home.
然而，在多中央處理單元系統中，不免遭遇其中一中央處理單元故障之情形。以目前技術水平而言，多中央處理單元系統可分為主從式架構及對等式架構。在對等式架構中，作業系統可以在任何一個中央處理單元上執行，每一個中央處理單元亦可自行進行排程，因此，若其中一中央處理單元故障，此運算平台仍可以其他中央處理單元繼續執行程序。然而，此種架構將使作業系統的安排更複雜，且在設計系統時，必須確定此多中央處理單元不會執行相同的程序。However, in a multi-central processing unit system, it is inevitable that one of the central processing units will fail. At the current state of the art, the multi-central processing unit system can be divided into a master-slave architecture and a peer-to-peer architecture. In a peer-to-peer architecture, the operating system can be executed on any central processing unit, and each central processing unit can also schedule itself. Therefore, if one of the central processing units fails, the computing platform can still have other central processing units. Continue to execute the program. However, such an architecture would complicate the scheduling of the operating system and, when designing the system, it must be determined that the multi-central processing unit would not perform the same procedure.
另一方面，相較於對等式架構，主從式架構之設計較簡單，成本亦較低廉。只是在主從式架構中，作業系統及主要核心功能皆在一主要之中央處理器上執行，此中央處理器必須負責工作的排程。因此，若主要之中央處理器故障，其他中央處理器並無法單獨運作，因而造成系統當機，甚至無法開機。例如INTEL Romeley/Grantley等系列的雙中央處理器系統，在中央處理器與PCH(Platform Controller Hub)晶片組之間係利用DMI(Direct Media Interface)介面溝通，倘若連接DMI之中央處理器故障，即使另一中央處理器也無法單獨運作。如此，不僅無法發揮雙中央處理單元平台之優點，亦可能對企業或個人帶來極大的損失。On the other hand, compared with the peer-to-peer architecture, the master-slave architecture is simpler in design and less expensive. In the master-slave architecture, the operating system and the main core functions are all in one mainTo be executed on the central processor, this central processor must be responsible for the scheduling of the work. Therefore, if the main central processor fails, other central processing units cannot operate independently, causing the system to crash or even fail to boot. For example, the dual-CPU system of the INTEL Romeley/Grantley series uses the DMI (Direct Media Interface) interface between the central processing unit and the PCH (Platform Controller Hub) chipset, even if the central processor connected to the DMI fails, even Another central processor cannot operate alone. In this way, not only can not play the advantages of the dual central processing unit platform, but also can bring great losses to enterprises or individuals.
本發明之一範疇在於提供一種多中央處理單元偵錯切換之方法。One aspect of the present invention is to provide a method for multi-central processing unit debug switching.
根據本發明之一具體實施例，一種多中央處理單元偵錯切換之方法包含下列步驟。首先，利用一微控制器模組，偵測第一處理器是否故障，包含無法偵測到或無法正常開機。接著，當偵測到第一處理器故障時，微控制器模組發出一故障訊號，並將此故障訊號傳輸至一切換器模組，以將預設處理器由第一處理器切換至第二處理器。最後，微控制器模組偵測第二處理器是否故障，若第二處理器故障，則微控制器模組再次將第一處理器設定為預設處理器，並重新開機。In accordance with an embodiment of the present invention, a method for multi-central processing unit debug switching includes the following steps. First, a microcontroller module is used to detect whether the first processor is faulty, including undetectable or unable to boot normally. Then, when the first processor failure is detected, the microcontroller module sends a fault signal and transmits the fault signal to a switch module to switch the preset processor from the first processor to the first processor. Two processors. Finally, the microcontroller module detects whether the second processor is faulty. If the second processor fails, the microcontroller module again sets the first processor as a preset processor and reboots.
在本發明之一具體實施例中，微控制器模組係透過一通用型輸入輸出模組而判斷第一或第二處理器是否能夠正常開機。通用型輸入輸出模組與一PCH(Platform Controller Hub)晶片組電性連接，當開機成功時將發出一開機成功訊號。In an embodiment of the invention, the microcontroller module determines whether the first or second processor can be powered on normally through a general-purpose input/output module. The universal input/output module is electrically connected to a PCH (Platform Controller Hub) chipset, and a power-on success signal is issued when the power-on is successful.
在本發明之另一具體實施例中，微控制器模組係透過一計時器而判斷第一或第二處理器是否能夠正常開機。第一或第二處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則判斷第一或第二處理器故障。In another embodiment of the present invention, the microcontroller module is passed through a timingThe device determines whether the first or second processor can be powered on normally. The first or second processor resets the timer at a predetermined time interval, and when the timer counts down to zero, determines that the first or second processor is faulty.
本發明之功效在於：針對主從式架構多中央處理單元之資訊處理平台，藉由獨立或內含微控制器的晶片模組，自動偵測第一中央處理器是否故障，若是，則切換至第二中央處理器，並以第二中央處理器繼續程序執行，以期將損失降到最低。另外可以注意的是，此種多中央處理單元之資訊處理平台可包含兩個以上之中央處理器，因此，藉由獨立或內含微控制器的晶片模組，自動偵測第一中央處理器是否故障，若是，則切換至第二中央處理器，若偵測到第二中央處理器故障，則切換至第三中央處理器，依此類推，直至偵測到有未故障之處理器，則以該處理器繼續執行程序。The utility model has the advantages that: for the information processing platform of the multi-central processing unit of the master-slave architecture, the chip module of the independent or embedded microcontroller automatically detects whether the first central processing unit is faulty, and if so, switches to The second central processor continues execution with the second central processor to minimize losses. In addition, it can be noted that the information processing platform of the multi-central processing unit may include more than two central processing units, and therefore, the first central processing unit is automatically detected by a chip module that is independent or includes a microcontroller. Whether it is faulty, if yes, switch to the second central processor, switch to the third central processor if a second central processor failure is detected, and so on, until a faulty processor is detected, The program continues to execute with the processor.
100‧‧‧步驟：開機100‧‧‧Steps: Power on
105‧‧‧步驟：偵測第一處理器是否連接至電源105‧‧‧Step: Detect if the first processor is connected to the power supply
110‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第一處理器是否開機成功110‧‧‧Step: The microcontroller module confirms whether the first processor is powered on successfully through an input/output module or a timer
115‧‧‧步驟：正常開機115‧‧‧Steps: Normal boot
120‧‧‧步驟：重新開機120‧‧‧Steps: Reboot
125‧‧‧步驟：偵測第二處理器是否連接至電源125‧‧‧Step: Detect if the second processor is connected to the power supply
130‧‧‧步驟：微控制器模組透過一切換器模組，切換至第二處理器執行開機130‧‧‧Step: The microcontroller module switches to the second processor to perform booting through a switcher module
135‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第二處理器是否開機成功135‧‧‧Step: The microcontroller module confirms whether the second processor is powered on successfully through an input/output module or a timer
140‧‧‧步驟：正常開機140‧‧‧Steps: Normal boot
145‧‧‧步驟：關機145‧‧‧Steps: Shutdown
200‧‧‧步驟：重新開機200‧‧‧Steps: Reboot
205‧‧‧步驟：偵測第三處理器是否連接至電源205‧‧‧Step: Detect if the third processor is connected to the power supply
210‧‧‧步驟：微控制器模組透過一切換器模組，切換至第三處理器執行開機210‧‧‧Step: The microcontroller module switches to the third processor to perform booting through a switcher module
215‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第三處理器是否開機成功215‧‧‧Step: The microcontroller module confirms whether the third processor is powered on successfully through an input/output module or a timer
220‧‧‧步驟：正常開機220‧‧‧Steps: Normal boot
225‧‧‧步驟：重新開機225‧‧‧Steps: Reboot
230‧‧‧步驟：偵測第四處理器是否連接至電源230‧‧‧Step: Detect if the fourth processor is connected to the power supply
235‧‧‧步驟：微控制器模組透過一切換器模組，切換至第四處理器執行開機235‧‧‧Step: The microcontroller module switches to the fourth processor to perform booting through a switcher module
240‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第四處理器是否開機成功240‧‧‧Step: The microcontroller module confirms whether the fourth processor is powered on successfully through an input/output module or a timer.
245‧‧‧步驟：正常開機245‧‧‧Steps: Normal boot
250‧‧‧步驟：關機250‧‧‧Steps: Shutdown
300‧‧‧步驟：重新開機次數是否達一預定次數300‧‧‧Steps: Is the number of reboots a predetermined number of times?
305‧‧‧步驟：關機305‧‧‧Steps: Shutdown
310‧‧‧步驟：重新開機310‧‧‧Steps: Reboot
400‧‧‧步驟：開機400‧‧‧Steps: Power on
405‧‧‧步驟：偵測第一處理器是否連接至電源405‧‧‧Step: Detect if the first processor is connected to the power supply
410‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第一處理器是否開機成功410‧‧‧Step: The microcontroller module confirms whether the first processor is powered on successfully through an input/output module or a timer.
415‧‧‧步驟：正常開機415‧‧‧Steps: Normal boot
420‧‧‧步驟：由微控制器模組發出一訊號，令切換器模組將第二處理器改為預設處理器420‧‧‧Step: A signal is sent from the microcontroller module to change the second processor to the default processor
425‧‧‧步驟：重新開機425‧‧‧Steps: Reboot
430‧‧‧步驟：偵測第二處理器是否連接至電源430‧‧‧Step: Detect if the second processor is connected to the power supply
435‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第二處理器是否開機成功435‧‧‧Step: The microcontroller module confirms whether the second processor is powered on successfully through an input/output module or a timer
440‧‧‧步驟：正常開機440‧‧‧Steps: Normal boot
圖1係根據本發明第一實施例之多中央處理單元偵錯切換之方法流程圖。1 is a flow chart of a method for multi-central processing unit debug switching according to a first embodiment of the present invention.
圖2係根據本發明第二實施例之多中央處理單元偵錯切換之方法流程圖。2 is a flow chart of a method for multi-central processing unit debug switching according to a second embodiment of the present invention.
圖3係根據本發明第三實施例之多中央處理單元偵錯切換之方法流程圖。3 is a flow chart of a method for multi-central processing unit debug switching according to a third embodiment of the present invention.
圖4係根據本發明第四實施例之多中央處理單元偵錯切換之方法流程圖。4 is a flow chart of a method for multi-central processing unit debug switching according to a fourth embodiment of the present invention.
圖1係繪示根據本發明一具體實施例之多中央處理單元偵錯切換之方法流程圖。此實施例係以兩個中央處理單元為說明。1 is a flow chart of a method for multi-central processing unit debug switching according to an embodiment of the invention. This embodiment is illustrated by two central processing units.
如圖1所示，首先執行步驟100之開機程序，於此實施例中，開機程序泛指一般開機程序，例如包括但不限於冷啟動(cold boot)及熱啟動(warm boot)。As shown in FIG. 1 , the booting process of step 100 is first performed. In this embodiment, the booting process generally refers to a general booting process, such as, but not limited to, a cold boot and a warm boot.
接著，利用微控制器模組，偵測第一處理器是否連接至電源(步驟105)，其中微控制器模組係與第一及第二處理器相連接。在此步驟中，微控制器模組由第一處理器讀取是否能成功連接至電源之訊號。假設此訊號之預設值為高準位，若可成功連接至電源，則此訊號會被更改成低準位，因此，若微控制器模組接收到之訊號為低準位，表示第一處理器可成功連接至電源，且由第一處理器執行開機程序；反之，若訊息為高準位，則表示第一處理器並未成功連接至電源。Next, the microcontroller module is used to detect whether the first processor is connected to the power source (step 105), wherein the microcontroller module is connected to the first and second processors. In this step, the microcontroller module reads from the first processor whether the signal can be successfully connected to the power source. Assume that the default value of this signal is high. If it can be successfully connected to the power supply, the signal will be changed to the low level. Therefore, if the signal received by the microcontroller module is low, it means the first. The processor can be successfully connected to the power source, and the boot process is executed by the first processor; otherwise, if the message is high level, the first processor is not successfully connected to the power source.
接著，在步驟110中，微控制器模組藉由一通用型輸入輸出模組，以判斷第一處理器是否開機成功。通用型輸入輸出模組與微控制器模組及一PCH(Platform Controller Hub)晶片組相連接，此PCH晶片組可於開機成功時發出一開機成功訊號。假設此訊號之預設值為高準位，若開機成功，則此訊號會被更改成低準位，因此若微控制器模組接收到低準位之訊號，表示開機成功；反之，若微控制器模組接收到高準位之訊號，則表示開機失敗，此時確認第一處理器故障。Next, in step 110, the microcontroller module uses a general-purpose input/output module to determine whether the first processor is powered on successfully. The universal input/output module is connected to the microcontroller module and a PCH (Platform Controller Hub) chipset, and the PCH chipset can send a power-on success signal when the power is turned on successfully. Assume that the default value of this signal is high. If the power is turned on successfully, the signal will be changed to low level. Therefore, if the microcontroller module receives the low level signal, it indicates that the power is turned on successfully. When the controller module receives the signal of the high level, it indicates that the power fails, and the first processor is confirmed to be faulty.
或者，在步驟110中，微控制器模組亦可藉由一計時器，以確認第一處理器是否開機成功。此計時器可例如設置於微控制器模組內，且分別連接至第一處理器及第二處理器。在此狀況下，第一處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則表示第一處理器開機失敗，可確認第一處理器故障。若否，則表示第一處理器開機成功。Alternatively, in step 110, the microcontroller module can also use a timer to confirm whether the first processor is powered on successfully. The timer can be, for example, disposed within the microcontroller module and coupled to the first processor and the second processor, respectively. In this case, the first processor takes a pre-The timer is reset at a fixed time interval. When the timer counts down to zero, it indicates that the first processor fails to boot, and the first processor failure can be confirmed. If not, it means that the first processor is powered on successfully.
於步驟110中，若偵測到第一處理器並未故障，則由第一處理器完成開機程序(步驟115)；若偵測到第一處理器故障，並且等待一預定時間，令主機板充分放電之後，重新開機(步驟120)。In step 110, if it is detected that the first processor is not faulty, the first processor completes the booting process (step 115); if the first processor fault is detected, and waits for a predetermined time, the motherboard is After fully discharging, the power is turned back on (step 120).
在偵測到第一處理器並未連接至電源(步驟105)或第一處理器故障且執行重新開機(步驟110及120)之情況下，接著，微控制器模組偵測第二處理器是否連接至電源(步驟125)。此時，微控制器模組由第二處理器讀取是否成功連接至電源之訊號，假設此訊號之預設值為高準位，若成功連接至電源，則此訊號會被更改成低準位。因此，若微控制器模組接收到之訊號為低準位，表示第二處理器可成功連接至電源；反之，若訊息為高準位，則表示第二處理器並未成功連接至電源。After detecting that the first processor is not connected to the power source (step 105) or the first processor is faulty and performing a reboot (steps 110 and 120), then the microcontroller module detects the second processor Whether to connect to the power source (step 125). At this time, the microcontroller module reads whether the signal is successfully connected to the power source by the second processor, and assumes that the preset value of the signal is a high level. If the signal is successfully connected to the power source, the signal is changed to a low level. Bit. Therefore, if the signal received by the microcontroller module is low, the second processor can be successfully connected to the power supply; otherwise, if the message is high, the second processor is not successfully connected to the power supply.
於步驟125中，若偵測到第二處理器已成功連接至電源，則微控制器模組透過一切換器模組，將電路切換至第二處理器並執行開機(步驟130)。切換器模組連接至第一及第二處理器，亦透過通用型輸入輸出模組連接至微控制器模組。因此，當微控制器模組偵測到第二處理器已成功連接至電源時，將發出一訊號，此訊號透過通用型之輸入輸出模組而傳輸至切換器模組，切換器模組在接收訊號後，便將電路切換至第二處理器，由第二處理器執行開機。In step 125, if it is detected that the second processor has successfully connected to the power source, the microcontroller module switches the circuit to the second processor and performs booting through a switch module (step 130). The switch module is connected to the first and second processors, and is also connected to the microcontroller module through a universal input and output module. Therefore, when the microcontroller module detects that the second processor has successfully connected to the power supply, a signal is sent, and the signal is transmitted to the switch module through the universal input and output module, and the switch module is After receiving the signal, the circuit is switched to the second processor, and the second processor performs power-on.
接著，在步驟135中，微控制器模組藉由一通用型輸入輸出模組，以判斷第二處理器是否開機成功。通用型輸入輸出模組與微控制器模組及一PCH(Platform Controller Hub)晶片組相連接，此PCH晶片組用以於開機成功時發出一開機成功訊號。假設此訊號預設值為高準位，若開機成功，則此訊號會被更改成低準位，因此若微控制器模組接收到低準位之訊號，表示開機成功；反之若接收到高準位之訊號，則表示開機失敗，此時確認第二處理器故障。Next, in step 135, the microcontroller module uses a general-purpose input/output module to determine whether the second processor is powered on successfully. The universal input/output module is connected to the microcontroller module and a PCH (Platform Controller Hub) chip set, and the PCH chip set is used forA power-on success signal is sent when the power is turned on successfully. Assume that the default value of this signal is high. If the power is turned on successfully, the signal will be changed to low level. Therefore, if the microcontroller module receives the low level signal, it indicates that the power is turned on successfully; otherwise, if it receives high The signal of the level indicates that the power fails, and the second processor is confirmed to be faulty.
或者，同步驟135，微控制器模組可藉由一計時器，以確認第二處理器是否開機成功。此計時器係位於微控制器模組之內，且分別連接至第一處理器及第二處理器，第二處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則表示第二處理器開機失敗，可確認第二處理器故障。若否，則表示第二處理器開機成功。Alternatively, in the same step 135, the microcontroller module can use a timer to confirm whether the second processor is powered on successfully. The timer is located in the microcontroller module and is respectively connected to the first processor and the second processor. The second processor resets the timer at a predetermined time interval. When the timer counts down to zero, Indicates that the second processor fails to boot, and can confirm that the second processor is faulty. If not, it means that the second processor is powered on successfully.
於步驟135中，若偵測到第二處理器並未故障，則接著由第二處理器完成開機程序(步驟140)；若偵測到第二處理器故障，則啟動關機(步驟145)。In step 135, if it is detected that the second processor is not faulty, then the boot process is completed by the second processor (step 140); if the second processor fault is detected, the shutdown is initiated (step 145).
圖2係根據本發明第二具體實施例之多中央處理單元偵錯切換之方法流程圖。2 is a flow chart of a method for multi-central processing unit debug switching according to a second embodiment of the present invention.
如圖2所示，第二實施例係以四個中央處理單元為實施例，可視為圖1之第一實施例之延伸，因此大致上與第一實施例相同，差別在於當第二處理器未能正常開機時(步驟135)，並且等待一預定時間，令主機板充分放電之後，重新開機(步驟200)。As shown in FIG. 2, the second embodiment adopts four central processing units as an embodiment, which can be regarded as an extension of the first embodiment of FIG. 1, and thus is substantially the same as the first embodiment, with the difference that when the second processor When the power is not turned on normally (step 135), and after waiting for a predetermined time to fully discharge the motherboard, the power is turned on again (step 200).
接著，微控制器模組偵測第三處理器是否連接至電源(步驟205)。此時，微控制器模組由第三處理器讀取是否成功連接至電源之訊號，假設此訊號之預設值為高準位，若成功連接至電源，則此訊號會被更改成低準位。因此，若微控制器模組接收到之訊號為低準位，表示第三處理器可成功連接至電源；反之，若訊息為高準位，則表示第三處理器並未成功連接至電源。Next, the microcontroller module detects if the third processor is connected to the power source (step 205). At this time, the microcontroller module reads whether the signal is successfully connected to the power source by the third processor, assuming that the preset value of the signal is a high level, and if the connection is successfully connected to the power source, the signal is changed to a low level. Bit. Therefore, if the signal received by the microcontroller module is low, the third processor is represented.Successfully connected to the power supply; conversely, if the message is high, the third processor is not successfully connected to the power supply.
於步驟205中，若偵測到第三處理器已成功連接至電源，則微控制器模組透過一切換器模組，將電路切換至第三處理器並執行開機(步驟210)。切換器模組連接至第一、第二、第三及第四處理器，亦透過通用型輸入輸出模組連接至微控制器模組。因此，當微控制器模組偵測到第三處理器已成功連接至電源時，將發出一訊號，此訊號透過通用型之輸入輸出模組而傳輸至切換器模組，切換器模組在接收訊號後，便將電路切換至第三處理器，由第三處理器執行開機。In step 205, if it is detected that the third processor has successfully connected to the power source, the microcontroller module switches the circuit to the third processor and performs booting through a switch module (step 210). The switch module is connected to the first, second, third and fourth processors, and is also connected to the microcontroller module through a universal input and output module. Therefore, when the microcontroller module detects that the third processor has successfully connected to the power supply, a signal is sent, and the signal is transmitted to the switch module through the universal input/output module, and the switch module is After receiving the signal, the circuit is switched to the third processor, and the third processor performs power-on.
接著，在步驟215中，微控制器模組藉由一通用型輸入輸出模組，以判斷第三處理器是否開機成功。通用型輸入輸出模組與微控制器模組及一PCH(Platform Controller Hub)晶片組相連接，此PCH晶片組用以於開機成功時發出一開機成功訊號。假設此訊號預設值為高準位，若開機成功，則此訊號會被更改成低準位，因此若微控制器模組接收到低準位之訊號，表示開機成功；反之若接收到高準位之訊號，則表示開機失敗，此時確認第三處理器故障。Next, in step 215, the microcontroller module uses a general-purpose input/output module to determine whether the third processor is powered on successfully. The universal input/output module is connected to the microcontroller module and a PCH (Platform Controller Hub) chipset, and the PCH chipset is used to send a power-on success signal when the power is turned on successfully. Assume that the default value of this signal is high. If the power is turned on successfully, the signal will be changed to low level. Therefore, if the microcontroller module receives the low level signal, it indicates that the power is turned on successfully; otherwise, if it receives high The signal of the level indicates that the power-on failed, and the third processor is confirmed to be faulty.
或者，同步驟215，微控制器模組可藉由一計時器，以確認第三處理器是否開機成功。此計時器係位於微控制器模組之內，且分別連接至第一、第二、第三及第四處理器，第三處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則表示第三處理器開機失敗，可確認第三處理器故障。若否，則表示第三處理器開機成功。Alternatively, in the same step 215, the microcontroller module can use a timer to confirm whether the third processor is powered on successfully. The timer is located within the microcontroller module and is coupled to the first, second, third, and fourth processors, respectively, and the third processor resets the timer at a predetermined time interval, when the timer counts down to Zero time means that the third processor fails to boot, and the third processor failure can be confirmed. If not, it means that the third processor is powered on successfully.
於步驟215中，若偵測到第三處理器並未故障，則接著由第三處理器完成開機程序(步驟220)；若偵測到第三處理器故障，並且等待一預定時間，令主機板充分放電之後，重新開機(步驟225)。In step 215, if it is detected that the third processor is not faulty, thenThe three processors complete the boot process (step 220); if a third processor failure is detected and a predetermined time is elapsed, the motherboard is fully discharged and then powered back on (step 225).
接著，微控制器模組偵測第四處理器是否連接至電源(步驟230)。此時，微控制器模組由第四處理器讀取是否成功連接至電源之訊號，假設此訊號之預設值為高準位，若成功連接至電源，則此訊號會被更改成低準位。因此，若微控制器模組接收到之訊號為低準位，表示第四處理器可成功連接至電源；反之，若訊息為高準位，則表示第四處理器並未成功連接至電源。Next, the microcontroller module detects if the fourth processor is connected to the power source (step 230). At this time, the microcontroller module reads whether the signal is successfully connected to the power source by the fourth processor, and assumes that the preset value of the signal is a high level. If the signal is successfully connected to the power source, the signal is changed to a low level. Bit. Therefore, if the signal received by the microcontroller module is low, the fourth processor can be successfully connected to the power supply; otherwise, if the message is high, the fourth processor is not successfully connected to the power supply.
於步驟230中，若偵測到第四處理器已成功連接至電源，則微控制器模組透過一切換器模組，將電路切換至第四處理器並執行開機(步驟235)。切換器模組連接至第一、第二、第三及第四處理器，亦透過通用型輸入輸出模組連接至微控制器模組。因此，當微控制器模組偵測到第四處理器已成功連接至電源時，將發出一訊號，此訊號透過通用型之輸入輸出模組而傳輸至切換器模組，切換器模組在接收訊號後，便將電路切換至第四處理器，由第四處理器執行開機。In step 230, if it is detected that the fourth processor has successfully connected to the power source, the microcontroller module switches the circuit to the fourth processor and performs booting through a switch module (step 235). The switch module is connected to the first, second, third and fourth processors, and is also connected to the microcontroller module through a universal input and output module. Therefore, when the microcontroller module detects that the fourth processor has successfully connected to the power supply, a signal is sent, and the signal is transmitted to the switch module through the universal input/output module, and the switch module is After receiving the signal, the circuit is switched to the fourth processor, and the fourth processor performs power-on.
接著，在步驟240中，微控制器模組藉由一通用型輸入輸出模組，以判斷第四處理器是否開機成功。通用型輸入輸出模組與微控制器模組及一PCH(Platform Controller Hub)晶片組相連接，此PCH晶片組用以於開機成功時發出一開機成功訊號。假設此訊號預設值為高準位，若開機成功，則此訊號會被更改成低準位，因此若微控制器模組接收到低準位之訊號，表示開機成功；反之若接收到高準位之訊號，則表示開機失敗，此時確認第四處理器故障。Next, in step 240, the microcontroller module uses a general-purpose input/output module to determine whether the fourth processor is powered on successfully. The universal input/output module is connected to the microcontroller module and a PCH (Platform Controller Hub) chipset, and the PCH chipset is used to send a power-on success signal when the power is turned on successfully. Assume that the default value of this signal is high. If the power is turned on successfully, the signal will be changed to low level. Therefore, if the microcontroller module receives the low level signal, it indicates that the power is turned on successfully; otherwise, if it receives high The signal of the level indicates that the power fails, and the fourth processor is confirmed to be faulty.
或者，同步驟240，微控制器模組可藉由一計時器，以確認第四處理器是否開機成功。此計時器係位於微控制器模組之內，且分別連接至第一、第二、第三及第四處理器，第四處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則表示第四處理器開機失敗，可確認第四處理器故障。若否，則表示第四處理器開機成功。Alternatively, in the same step 240, the microcontroller module can use a timer to confirm whether the fourth processor is powered on successfully. The timer is located in the microcontroller module and is respectively connected to the first, second, third and fourth processors, and the fourth processor resets the timer at a predetermined time interval, when the timer counts down to Zero time means that the fourth processor fails to boot, and the fourth processor failure can be confirmed. If not, it means that the fourth processor is powered on successfully.
於步驟240中，若偵測到第四處理器並未故障，則接著由第四處理器完成開機程序(步驟245)；若偵測到第四處理器故障，則啟動關機(步驟250)。In step 240, if it is detected that the fourth processor is not faulty, then the boot process is completed by the fourth processor (step 245); if the fourth processor fault is detected, the shutdown is initiated (step 250).
圖3係根據本發明第三具體實施例之多中央處理單元偵錯切換之方法流程圖。3 is a flow chart of a method for multi-central processing unit debug switching according to a third embodiment of the present invention.
如圖3所示，第三實施例之方法大致上與第一實施例相同，差別僅在於當第二處理器未能正常開機時(步驟135)，再次重新開機，並增加判斷重新開機之次數是否已達一預定次數之步驟(步驟300)。倘若已達一預定次數，則進行關機(步驟305)；倘若未達預定次數，並且等待一預定時間，令主機板充分放電之後，再重新開機(步驟310)。As shown in FIG. 3, the method of the third embodiment is substantially the same as that of the first embodiment, except that when the second processor fails to boot normally (step 135), the power is turned back on again, and the number of times of restarting is increased. Whether the step has been reached a predetermined number of times (step 300). If the predetermined number of times has elapsed, the shutdown is performed (step 305); if the predetermined number of times has not elapsed, and the predetermined time has elapsed, the motherboard is fully discharged, and then turned back on (step 310).
由於資訊處理平台未能開機成功的原因眾多，除了中央處理器故障，還包括電源供應之電壓不穩等問題。因此，吾人設定一重新開機之預定次數，以嘗試排除此類因電源供應電壓不穩而導致未能開機成功之情況。Due to the failure of the information processing platform to boot successfully, in addition to the failure of the central processing unit, the voltage supply of the power supply is unstable. Therefore, we set a predetermined number of reboots in an attempt to eliminate such a failure to power on due to unstable power supply voltage.
於第二實施例中，係以如下方式計算重新開機之次數。若判斷第一及第二處理器皆為故障，則將此時所執行之重新開機計數為一次(當執行至步驟310時，計數一次)，以此類推。然而，應瞭解的是，計算重新開機次數及判斷重新開機是否已達一預定次數，可由程式自動或以手動方式完成，惟本發明不應以此為限。此外，吾人可視實際狀況而定義重新開機次數，例如，偵測到第一處理器故障之後的重新開機，即計數為一次(當執行至步驟120時，即計數一次)，且偵測到第二處理器故障之後的重新開機，亦計數為一次(當執行至步驟310時，亦計數一次)。或者，對於偵測到第一處理器及第二處理器皆故障之後的重新開機，方才計數為一次(當執行至步驟310時，總共計數一次)。In the second embodiment, the number of restarts is calculated in the following manner. If it is determined that both the first and second processors are faulty, the restart count performed at this time is counted once (when executed to step 310, counted once), and so on. However, it should be understood that the calculation is reThe number of power-on and the judgment of whether the power-on has reached a predetermined number of times can be completed automatically or manually by the program, but the invention should not be limited thereto. In addition, we can define the number of reboots according to the actual situation, for example, detecting the reboot after the first processor failure, that is, counting once (once when executing to step 120, counting once), and detecting the second The reboot after the processor failure is also counted once (also counted once when executed to step 310). Alternatively, it is counted once for detecting that the first processor and the second processor are both restarted after the failure (when executed to step 310, the total count is once).
圖4係根據本發明之第四具體實施例之多中央處理單元偵錯切換之方法之步驟流程圖。4 is a flow chart showing the steps of a method for multi-central processing unit debug switching according to a fourth embodiment of the present invention.
如圖4所示，首先執行步驟400之開機程序，於此實施例中，開機程序泛指一般開機程序，例如包括但不限於冷啟動(cold boot)及熱啟動(warm boot)。As shown in FIG. 4, the booting process of step 400 is first performed. In this embodiment, the booting process generally refers to a general booting process, such as, but not limited to, a cold boot and a warm boot.
接著，利用微控制器模組，偵測第一處理器是否連接至電源(步驟405)，其中微控制器模組係與第一及第二處理器相連接。在此步驟中，微控制器模組由第一處理器讀取是否能成功連接至電源之訊號。假設此訊號之預設值為高準位，若可成功連接至電源，則此訊號會被更改成低準位，因此，若微控制器模組接收到之訊號為低準位，表示第一處理器可成功連接至電源，且由第一處理器執行開機程序；反之，若訊息為高準位，則表示第一處理器並未成功連接至電源。Next, the microcontroller module is used to detect whether the first processor is connected to the power source (step 405), wherein the microcontroller module is connected to the first and second processors. In this step, the microcontroller module reads from the first processor whether the signal can be successfully connected to the power source. Assume that the default value of this signal is high. If it can be successfully connected to the power supply, the signal will be changed to the low level. Therefore, if the signal received by the microcontroller module is low, it means the first. The processor can be successfully connected to the power source, and the boot process is executed by the first processor; otherwise, if the message is high level, the first processor is not successfully connected to the power source.
接著，在步驟410中，微控制器模組藉由一通用型輸入輸出模組，以判斷第一處理器是否開機成功。通用型輸入輸出模組與微控制器模組及一PCH(Platform Controller Hub)晶片組相連接，此PCH晶片組可於開機成功時發出一開機成功訊號。假設此訊號之預設值為高準位，若開機成功，則此訊號會被更改成低準位，因此若微控制器模組接收到低準位之訊號，表示開機成功；反之，若微控制器模組接收到高準位之訊號，則表示開機失敗，此時確認第一處理器故障。Next, in step 410, the microcontroller module determines whether the first processor is powered on successfully by using a general-purpose input/output module. The universal input/output module is connected to the microcontroller module and a PCH (Platform Controller Hub) chip set, and the PCH chip set can be opened.A successful start signal is sent when the machine is successful. Assume that the default value of this signal is high. If the power is turned on successfully, the signal will be changed to low level. Therefore, if the microcontroller module receives the low level signal, it indicates that the power is turned on successfully. When the controller module receives the signal of the high level, it indicates that the power fails, and the first processor is confirmed to be faulty.
或者，在步驟410中，微控制器模組亦可藉由一計時器，以確認第一處理器是否開機成功。此計時器可例如設置於微控制器模組內，且分別連接至第一處理器及第二處理器。在此狀況下，第一處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則表示第一處理器開機失敗，可確認第一處理器故障。若否，則表示第一處理器開機成功。Alternatively, in step 410, the microcontroller module can also use a timer to confirm whether the first processor is powered on successfully. The timer can be, for example, disposed within the microcontroller module and coupled to the first processor and the second processor, respectively. In this case, the first processor resets the timer at a predetermined time interval. When the timer counts down to zero, it indicates that the first processor fails to boot, and the first processor may be confirmed to be faulty. If not, it means that the first processor is powered on successfully.
於步驟410中，若偵測到第一處理器並未故障，則由第一處理器完成開機程序(步驟415)；若偵測到第一處理器並未連接到電源或故障，則微控制器模組透過一切換器模組，將預設處理器切換為第二處理器(步驟420)。切換器模組連接至第一及第二處理器，亦透過通用型輸入輸出模組連接至微控制器模組。因此，當微控制器偵測到第一處理器並未連接至電源或故障，即發出一訊號，此訊號透過通用型之輸入輸出模組，而傳輸至切換器模組，切換器模組在接收訊號後，便將預設處理器由第一處理器切換至第二處理器。In step 410, if it is detected that the first processor is not faulty, the boot process is completed by the first processor (step 415); if it is detected that the first processor is not connected to the power source or the fault, the micro-control The module switches the preset processor to the second processor through a switch module (step 420). The switch module is connected to the first and second processors, and is also connected to the microcontroller module through a universal input and output module. Therefore, when the microcontroller detects that the first processor is not connected to the power source or fails, a signal is sent, and the signal is transmitted to the switch module through the universal input/output module, and the switch module is After receiving the signal, the preset processor is switched from the first processor to the second processor.
在步驟425中，並且等待一預定時間，令主機板充分放電之後，重新開機。In step 425, and waiting for a predetermined time, the motherboard is fully discharged and then turned back on.
接著，微控制器模組偵測第二處理器是否連接至電源(步驟430)。此時，微控制器模組由第二處理器讀取是否成功連接至電源之訊號，假設此訊號之預設值為高準位，若成功連接至電源，則此訊號會被更改成低準位。因此，若微控制器模組接收到之訊號為低準位，表示第二處理器可成功連接至電源；反之，若訊息為高準位，則表示第二處理器並未成功連接至電源。Next, the microcontroller module detects if the second processor is connected to the power source (step 430). At this time, the microcontroller module reads from the second processor whether the signal is successfully connected to the power supply, and the preset value of the signal is assumed to be a high level. If the power is successfully connected, the signal is changed toLow level. Therefore, if the signal received by the microcontroller module is low, the second processor can be successfully connected to the power supply; otherwise, if the message is high, the second processor is not successfully connected to the power supply.
於步驟430中，若偵測到第二處理器已成功連接至電源，則微控制器模組藉由一通用型輸入輸出模組，以判斷第二處理器是否開機成功(在步驟435)。通用型輸入輸出模組與微控制器模組及一PCH(Platform Controller Hub)晶片組相連接，此PCH晶片組用以於開機成功時發出一開機成功訊號。假設此訊號預設值為高準位，若開機成功，則此訊號會被更改成低準位，因此若微控制器模組接收到低準位之訊號，表示開機成功；反之若接收到高準位之訊號，則表示開機失敗，此時確認第二處理器故障。In step 430, if it is detected that the second processor has successfully connected to the power source, the microcontroller module uses a general-purpose input/output module to determine whether the second processor is powered on successfully (at step 435). The universal input/output module is connected to the microcontroller module and a PCH (Platform Controller Hub) chipset, and the PCH chipset is used to send a power-on success signal when the power is turned on successfully. Assume that the default value of this signal is high. If the power is turned on successfully, the signal will be changed to low level. Therefore, if the microcontroller module receives the low level signal, it indicates that the power is turned on successfully; otherwise, if it receives high The signal of the level indicates that the power fails, and the second processor is confirmed to be faulty.
或者，同步驟435，微控制器模組可藉由一計時器，以確認第二處理器是否開機成功。此計時器係位於微控制器模組之內，且分別連接至第一處理器及第二處理器，第二處理器以一預定時間間隔重置計時器，當計時器倒數至零時，則表示第二處理器開機失敗，可確認第二處理器故障。若否，則表示第二處理器開機成功。Alternatively, in the same step 435, the microcontroller module can use a timer to confirm whether the second processor is powered on successfully. The timer is located in the microcontroller module and is respectively connected to the first processor and the second processor. The second processor resets the timer at a predetermined time interval. When the timer counts down to zero, Indicates that the second processor fails to boot, and can confirm that the second processor is faulty. If not, it means that the second processor is powered on successfully.
於步驟435中，若偵測到第二處理器並未故障，則接著由第二處理器完成開機程序(步驟440)。若偵測到第二處理器故障，則接著判斷重新開機之次數是否已達一預定次數之步驟(圖3步驟300)。倘若已達一預定次數，則進行關機(圖3步驟305)；倘若未達預定次數，並且等待一預定時間，令主機板充分放電之後，再重新開機(圖3步驟310)。In step 435, if it is detected that the second processor is not faulty, then the boot process is completed by the second processor (step 440). If the second processor failure is detected, then it is determined whether the number of restarts has reached a predetermined number of times (step 300 of FIG. 3). If the predetermined number of times has elapsed, the shutdown is performed (step 305 of FIG. 3); if the predetermined number of times has not elapsed, and the predetermined time has elapsed, the motherboard is fully discharged, and then turned back on (step 310 of FIG. 3).
在上述具體實施例中，主要藉由微控制器模組以及切換器模組，再搭配其他的必要組件，使得兩中央處理器皆可作為預設處理器，執行開機及其他運算。因此，在多中央處理器之資訊處理系統中，可利用此方法以偵測預設處理器是否故障。若是，則切換到其他處理器並執行開機及其他運算，如此便可維持此資訊處理系統之功能。In the above specific embodiment, the microcontroller module and the switch module are mainly used together with other necessary components, so that both CPUs can be used as preset processors.Start up and other operations. Therefore, in an information processing system of a multi-CPU, this method can be utilized to detect whether a preset processor is faulty. If so, switch to another processor and perform boot and other operations to maintain the functionality of this information processing system.
由於電腦的運算日趨複雜，商用之資訊處理系統大多使用數目為兩個以上的中央處理器，例如4,6，或8個。儘管本發明係以2及4個CPU之系統為例說明如上，然應注意者，本發明並不限制於此。換言之，本發明之方法可使用於多個中央處理器之資訊處理系統，而在不同的中央處理器之間進行切換。如此一來，即使其中一中央處理器故障，系統仍能正常運作，如此可將損失降至最低。Due to the increasing complexity of computer computing, commercial information processing systems mostly use more than two central processors, such as 4, 6, or 8. Although the present invention is described above by taking a system of two or four CPUs as an example, it should be noted that the present invention is not limited thereto. In other words, the method of the present invention can be used for information processing systems of multiple central processors while switching between different central processors. In this way, even if one of the central processors fails, the system can still operate normally, so that the loss can be minimized.
100‧‧‧步驟：開機100‧‧‧Steps: Power on
105‧‧‧步驟：偵測第一處理器是否連接至電源105‧‧‧Step: Detect if the first processor is connected to the power supply
110‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第一處理器是否開機成功110‧‧‧Step: The microcontroller module confirms whether the first processor is powered on successfully through an input/output module or a timer
115‧‧‧步驟：正常開機115‧‧‧Steps: Normal boot
120‧‧‧步驟：重新開機120‧‧‧Steps: Reboot
125‧‧‧步驟：偵測第二處理器是否連接至電源125‧‧‧Step: Detect if the second processor is connected to the power supply
130‧‧‧步驟：微控制器模組透過一切換器模組，切換至第二處理器執行開機130‧‧‧Step: The microcontroller module switches to the second processor to perform booting through a switcher module
135‧‧‧步驟：微控制器模組透過一輸入輸出模組，或一計時器，確認第二處理器是否開機成功135‧‧‧Step: The microcontroller module confirms whether the second processor is powered on successfully through an input/output module or a timer
140‧‧‧步驟：正常開機140‧‧‧Steps: Normal boot
300‧‧‧步驟：重新開機次數是否達一預定次數300‧‧‧Steps: Is the number of reboots a predetermined number of times?
305‧‧‧步驟：關機305‧‧‧Steps: Shutdown
310‧‧‧步驟；重新開機310‧‧‧Steps; reboot
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW102123298ATWI551983B (en) | 2013-06-28 | 2013-06-28 | Failure detection and switch method for multi-cpu system |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW102123298ATWI551983B (en) | 2013-06-28 | 2013-06-28 | Failure detection and switch method for multi-cpu system |
| Publication Number | Publication Date |
|---|---|
| TW201500914A TW201500914A (en) | 2015-01-01 |
| TWI551983Btrue TWI551983B (en) | 2016-10-01 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW102123298ATWI551983B (en) | 2013-06-28 | 2013-06-28 | Failure detection and switch method for multi-cpu system |
| Country | Link |
|---|---|
| TW (1) | TWI551983B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113778210B (en)* | 2021-08-20 | 2023-03-10 | 江苏嘉擎信息技术有限公司 | MCU-based ACPI management method, system and equipment |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5506987A (en)* | 1991-02-01 | 1996-04-09 | Digital Equipment Corporation | Affinity scheduling of processes on symmetric multiprocessing systems |
| TW200923671A (en)* | 2007-11-27 | 2009-06-01 | Inventec Corp | Multiprocessor system and method for switching CPUs |
| CN101452437A (en)* | 2007-12-03 | 2009-06-10 | 英业达股份有限公司 | Multiprocessor system and method for switching CPU |
| TW201107963A (en)* | 2009-08-24 | 2011-03-01 | Inventec Corp | Test method |
| TW201227269A (en)* | 2010-12-30 | 2012-07-01 | Ibm | Apparatus and method for handling a failed processor of a multiprocessor information handling system |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5506987A (en)* | 1991-02-01 | 1996-04-09 | Digital Equipment Corporation | Affinity scheduling of processes on symmetric multiprocessing systems |
| TW200923671A (en)* | 2007-11-27 | 2009-06-01 | Inventec Corp | Multiprocessor system and method for switching CPUs |
| CN101452437A (en)* | 2007-12-03 | 2009-06-10 | 英业达股份有限公司 | Multiprocessor system and method for switching CPU |
| TW201107963A (en)* | 2009-08-24 | 2011-03-01 | Inventec Corp | Test method |
| TW201227269A (en)* | 2010-12-30 | 2012-07-01 | Ibm | Apparatus and method for handling a failed processor of a multiprocessor information handling system |
| Publication number | Publication date |
|---|---|
| TW201500914A (en) | 2015-01-01 |
| Publication | Publication Date | Title |
|---|---|---|
| US8898517B2 (en) | Handling a failed processor of a multiprocessor information handling system | |
| TWI588649B (en) | Hardware recovery methods, hardware recovery systems, and computer-readable storage device | |
| CN104050061B (en) | A kind of Based PC Ie bus many master control board redundancies standby system | |
| CN107943603B (en) | Running state detection method, detection circuit and electronic equipment | |
| US20250286836A1 (en) | Switch reset system and method, non-volatile readable storage medium, and electronic device | |
| CN111767244A (en) | Dual-redundancy computer equipment based on domestic Loongson platform | |
| CN105242980A (en) | Complementary watchdog system and complementary watchdog monitoring method | |
| CN103425553A (en) | Duplicated hot-standby system and method for detecting faults of duplicated hot-standby system | |
| CN106776054A (en) | A deadlock detection method, device and circuit | |
| CN102508746A (en) | Management method for triple configurable fault-tolerant computer system | |
| CN108958987A (en) | A kind of Low earth orbit satellite tolerant system and method | |
| CN110764829B (en) | Multi-path server CPU isolation method and system | |
| CN212541329U (en) | Dual-redundancy computer equipment based on domestic Loongson platform | |
| CN103178977A (en) | Computer system and boot management method of computer system | |
| CN107291207B (en) | A power switching method and system | |
| US8560867B2 (en) | Server system and method for processing power off | |
| CN101009684A (en) | Monitoring device and method for single-board operation state in the distributed system | |
| CN204406385U (en) | The management devices of computer system | |
| TWI551983B (en) | Failure detection and switch method for multi-cpu system | |
| CN103034511B (en) | Changing method between a kind of Basic Input or Output System (BIOS) and device | |
| CN104252396B (en) | Method for switching multiple CPU debugging | |
| JP6256087B2 (en) | Dump system and dump processing method | |
| CN107179911A (en) | A kind of method and apparatus for restarting management engine | |
| CN112783696A (en) | Network card redundancy system and server | |
| CN111414066A (en) | A server expansion system and power control method thereof |
| Date | Code | Title | Description |
|---|---|---|---|
| MM4A | Annulment or lapse of patent due to non-payment of fees |