WO2014186938A1 - 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 - Google Patents
一种计算机系统、高速外围组件互联端点设备的访问方法和装置 Download PDFInfo
- Publication number
- WO2014186938A1 WO2014186938A1 PCT/CN2013/075920 CN2013075920W WO2014186938A1 WO 2014186938 A1 WO2014186938 A1 WO 2014186938A1 CN 2013075920 W CN2013075920 W CN 2013075920W WO 2014186938 A1 WO2014186938 A1 WO 2014186938A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- access request
- endpoint device
- pcie endpoint
- response message
- processor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3041—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is an input/output interface
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/004—Error avoidance
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3027—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a bus
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/10—Program control for peripheral devices
- G06F13/105—Program control for peripheral devices where the program performs an input/output emulation function
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4204—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
- G06F13/4221—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus being an input/output bus, e.g. ISA bus, EISA bus, PCI bus, SCSI bus
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3089—Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2213/00—Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F2213/0026—PCI express
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2213/00—Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F2213/40—Bus coupling
- G06F2213/4004—Universal serial bus hub with a plurality of upstream ports
Definitions
- Embodiments of the present invention relate to computer technology, and in particular, to a high-speed peripheral component interconnection endpoint device access method, computer system, and apparatus.
- PCIe Peripheral Component Interconnect Express
- the PCIe bus is widely used in the interconnection system of CPUs and peripherals as a core service channel in computing and storage devices.
- peripheral devices that can be interconnected with the CPU through the PCIe bus, such as network card devices or Solid State Disks (SSDs). These devices are collectively referred to as PCIe endpoint devices in this document.
- the PCIe bus is widely used as the bus interface of the server or storage system.
- the need for online expansion and maintenance requires the addition or removal of PCIe endpoint devices, that is, hot swapping requirements.
- the existing PCIe hot plug follows the following operation process: The operator initiates a hot pull request by pressing the button, and the hot plug controller knows that after the hot plug event, all the drivers in the system that may access the PCIe endpoint device are stopped. The PCIe endpoint device is accessed, and the resources of the PCIe endpoint device that needs to be hot-swapped are offloaded. Thereafter, the PCIe endpoint device is powered off, and the operator pulls out the PCIe endpoint device.
- Hot swapping of existing PCIe endpoint devices requires advance notice to ensure system operation.
- the PCIe bus has gradually evolved from interconnection within the system to interconnection between systems.
- Applications such as external cables have increased, cables have been easily dropped, and PCIe endpoint devices that are not notified are abnormally offline.
- the scenario in which the user directly accesses the system using the SSD is more widely used. Due to user habits, the user may directly insert and remove the SSD without prior notice.
- the PCIe endpoint device mentioned above suddenly suddenly goes abnormally offline, if the CPU has initiated a read/write instruction to the PCIe endpoint device, the related instruction will remain in a pending state, when the CPU is to the PCIe endpoint device.
- the access command continues to accumulate to a certain extent, and the CPU considers the entire system. Abnormal, a Machine check exception (MCE) error is reported and reset.
- MCE Machine check exception
- Embodiments of the present invention provide an access method, a computer system, and a device for interconnecting an endpoint device of a high-speed peripheral component. After the PCIe endpoint device is abnormally offline, the CPU is guaranteed not to be reset.
- an embodiment of the present invention provides a computer system, where the computer system includes: a processor;
- the processor is configured to obtain an operation instruction for accessing the PCIe endpoint device, and send an access request to the PCIe endpoint device according to the operation instruction;
- the computer system further includes an analog response module, configured to monitor a message status of the access request after the processor sends the access request to the PCIe endpoint device, when a preset trigger condition is reached Determining a message status of the access request, if the message status of the access request is "not completed", acquiring an analog response message of the access request, and sending the analog response message to the processor;
- the message status of the access request includes "completed” or "not completed", and the "not completed" indicates that the PCIe endpoint device does not send the response message of the access request.
- the analog response module is disposed in a signal transmission path of the processor and the PCIe endpoint device, and is specifically configured to receive an access request sent by the processor. And forwarding the access request to the PCIe endpoint device, and monitoring whether the response message of the access request sent by the PCIe endpoint device is received by the PCIe endpoint device, and receiving the response message of the access request sent by the PCIe endpoint device , the message of the access request is recorded as "completed".
- the computer system further includes: a PCIe switch, configured to use the multiple PCIe An endpoint device is connected to the processor, and a downstream port of the PCIe switch is connected to a plurality of the PCIe endpoint devices through the PCIe bus;
- the simulated response message of the access request includes the first a type analog response message indicating that the PCIe endpoint device does not support the access request, or a second type analog response message indicating that the PCIe endpoint device is abnormal;
- the processor is further configured to suspend subsequent access to the PCIe endpoint device according to the first type of analog response message or the second type of analog response message.
- the preset triggering condition is the PCIe
- the endpoint device is abnormally offline.
- the analog response module is further configured to: monitor whether the PCIe endpoint device is abnormally offline, and determine a packet status of the access request when determining that the PCIe endpoint device is abnormally offline.
- the preset triggering condition is the access The requested timer expires
- the analog response module is further configured to: perform timing on the access request start timer sent by the processor, and determine a message status of the access request after the timer expires.
- a method for accessing a PCIe endpoint device of a high-speed peripheral component is also provided.
- the PCIe endpoint device is connected to the computer system through the PCIe bus, and the method includes: After the processor in the computer system sends an access request to the PCIe endpoint device, the packet status of the access request is monitored, where the message status of the access request includes "completed” or "not”"Complete", said "not completed” indicating that the PCIe endpoint device has not sent the response message of the access request;
- the method further includes: receiving an access request sent by the processor, and forwarding the access request to the PCIe endpoint device;
- the monitoring the status of the message of the access request includes:
- the preset triggering condition is that the PCIe endpoint device is abnormally offline; the method further includes:
- the preset The trigger condition is that the timer of the access request times out
- the method further includes:
- the access request start timer is sent to the processor for timing.
- an analog response device is proposed, which is applied to a computer system, and a high-speed peripheral
- the component interconnect PCIe endpoint device is accessed into the computer system via a PCIe bus, and the device includes:
- a monitoring unit configured to monitor, after the processor in the computer system sends an access request to the PCIe endpoint device, the packet status of the access request, where the message status of the access request includes Completing "or” is not completed, the "not completed” indicating that the PCIe endpoint device has not sent the response message of the access request;
- a determining unit configured to determine, according to the monitoring result of the monitoring unit, a message status of the access request according to a monitoring result of the monitoring unit; when the time is "", obtain an analog response message of the access request;
- a response unit configured to send, to the processor, the simulated response message acquired by the acquiring unit.
- the method further includes: an interaction unit, configured to receive an access request sent by the processor, and forward the access to the PCIe endpoint device Request
- the monitoring unit is specifically configured to monitor whether the interaction unit receives the response message of the access request sent by the PCIe endpoint device, and if receiving the response message of the access request sent by the PCIe endpoint device, The message status of the access request is recorded as "completed".
- a computer comprising: a processor, a memory, a bus, and a communication interface;
- the memory is configured to store a computer execution instruction
- the processor is connected to the memory through the bus, and when the computer is running, the processor executes the computer execution instruction stored in the memory to make The computer performs the access method of a high speed peripheral component interconnect PCIe endpoint device according to the second aspect.
- a computer readable medium comprising computer executed instructions when computing
- the processor of the machine executes the computer-executed instructions
- the computer performs the access method of a high-speed peripheral component interconnect PCIe endpoint device of the second aspect.
- the analog response message of the access request is sent to the processor.
- the processor can confirm that the flow corresponding to the previously issued access request has been completed, thereby closing the timer for timing the access request, and clearing the cached access request, and maintaining the internal processing of the processor is normal. This is done to avoid the MCE reset problem caused by the accumulation of access requests in the processor.
- FIG. 1 is a composition diagram of a computer system according to an embodiment of the present invention.
- FIG. 2 is a schematic diagram of a computer system according to an embodiment of the present invention
- FIG. 3 is a flowchart of a method according to an embodiment of the present invention
- FIG. 5 is a flowchart of still another method according to an embodiment of the present invention.
- FIG. 6 is a composition diagram of a simulation response apparatus according to an embodiment of the present invention
- FIG. 7 is a composition diagram of still another analog response apparatus according to an embodiment of the present invention
- FIG. 8 is a diagram according to an embodiment of the present invention.
- the embodiment of the invention provides a method, a computer system and a device for accessing a high-speed peripheral component interconnection endpoint device.
- a PCIe endpoint device needs to be hot-swapped, it may not be pre-passed. Knowing that the system is pre-processed, the connection between the PCIe endpoint device and the processor is directly disconnected, and the processor does not risk the MCE reset.
- the case where the PCIe endpoint device is directly pulled out of the system or the fault is dropped is collectively referred to as abnormal offline of the PCIe endpoint device.
- FIG. 1 is a block diagram of a computer system provided by an embodiment of the present invention.
- the computer system shown in FIG. 1 includes a CPU 110 and a memory 120.
- the PCIe endpoint device 130 is connected to the CPU 110 through a PCIe bus 140, and can be from the computer system. Plug in and out.
- the PCIe endpoint device 130 includes a plurality of types, such as a graphics processing unit 131, a network adapter 132, a solid state hard disk 133, and a video acceleration component 134.
- the PCIe endpoint device 130 and the CPU 110 in the computer system shown in Fig. 1 form a PCIe domain, and all devices in the PCIe domain are connected to the CPU 110 via the PCIe bus 140 and are controlled by the CPU 110.
- the CPU 110 After the CPU 110 obtains an operation instruction for accessing the PCIe endpoint device 130, the CPU 110 issues an access request to the PCIe endpoint device 130 according to the operation instruction, requesting a read operation or a write operation to the register of the PCIe endpoint device 130, if the PCIe endpoint device 130 If an abnormality occurs, the CPU 110 will not receive a response message from the PCIe endpoint device 130 to the CPU 110. At this time, the CPU considers the access request to be an unfinished task, and if such an unfinished task accumulates in the CPU. To the extent, the CPU will consider the entire system abnormal and report an MCE error to reset.
- a new analog response module 160 is added to the system, and the analog response response module 160 is configured to report the access request sent by the CPU 110 after the CPU 110 sends the access request to the PCIe endpoint device 130. The status of the message is monitored. If the preset trigger condition is reached, the message status of the access request is still "incomplete", that is, the PCIe endpoint device 130 does not send a response message to the access request, the simulation The response response module 160 obtains the analog response message of the access request and sends the analog response message to the CPU 110.
- the analog response message has the same message header as the access request, and the access request and the analog response message correspond to each other, so that the CPU considers that the flow corresponding to the previously issued access request has been completed, thereby closing the Access the timer that requested the timing, and clear the access request cached by the CPU itself.
- the message status of the access request includes "completed” or "not completed”, "completed” indicates that the PCIe endpoint device has sent a response message of the access request, and "not completed” indicates the PCIe endpoint The device did not send a response message for the access request.
- analog response module 160 is further configured to: after the processor sends the access request to the PCIe endpoint device, generate an analog response message of the access request, and cache an analog response message of the access request.
- the analog response module 160 may directly obtain the simulated response message of the access request cached by itself.
- the analog response module can monitor the packet status of the access request sent by the CPU, and send an access request to the CPU when determining that the PCIe endpoint device does not send the response message to the access request. Simulating the response message, enabling the CPU to confirm that the flow corresponding to the previously issued access request has been completed according to the analog response message, thereby closing the timer for timing the access request, and clearing the cached access request, and maintaining the internal processing of the CPU Normally, the MCE reset problem caused by the accumulation of access requests in the CPU is avoided.
- the analog response module 160 monitors the packet status of the access request, and may implement the response message by monitoring whether the access request sent by the PCIe endpoint device 130 is received by the PCIe endpoint device 130. And if the analog response module receives the response message of the access request sent by the PCIe endpoint device, the analog response module records the status of the access request as “completed”.
- the analog response module 160 may be disposed in a transmission path between the CPU 110 and each PCIe endpoint device 130 to ensure that messages between the CPU 110 and each PCIe endpoint device 130 need to pass through an analog response module. 160, whereby the analog response module 160 can monitor whether the access request sent by the PCIe endpoint device 130 is received by itself. The response message is used to monitor the message status of the access request.
- the PCIe endpoint device 130 may be configured in advance, so that the PCIe endpoint device 130 sends the response request response message to the CPU 110.
- the response message is also sent to the analog response module, so that the analog response module 160 can also ensure that the message status of the access request is monitored according to whether the response message of the access request is received by itself.
- the analog response module 160 can implement the analog response module 160 to monitor the status of the message of the access request according to the implementation provided by the embodiment of the present invention by using other equivalent or similar implementation manners.
- the analog response module 160 is disposed in a signal transmission path between the CPU 110 and each PCIe endpoint device 130.
- an access request issued by the CPU 110 needs to pass through the analog response module 160 to reach the PCIe endpoint device 130, thus simulating The response module 160 can receive the access request, and since the message sent by the PCIe endpoint device needs to pass through the analog response module 160 to reach the CPU 110, the analog response module 160 can receive the access request sent by the PCIe endpoint device 130 according to whether it receives the access request. And responding to the message, determining whether the PCIe endpoint device 130 sends the response message of the access request, and recording the message status of the access request.
- the access request may also be cached, if the access sent by the PCIe endpoint device 130 is subsequently received.
- the response response message may be: the analog response module 160 may mark the cached access request with a "completed" flag, or delete the cached access request, and the analog response module 160 may query the access request when the preset trigger condition arrives.
- the message state that is, the tag of the cached access request or the query cached access request.
- the analog response module 160 determines that The message status of the access request is "incomplete", that is, the PCIe endpoint device 130 does not send the response message of the access request, and the analog response module 160 acquires the analog response message of the access request, and sends the analog response message to the CPU 110. .
- the analog response module 160 may generate an analog response message of the access request in advance when receiving the access request sent by the CPU 110, and generate a simulated response of the generated access request.
- the message should be cached in association with the access request, and the subsequent analog response module 160 can directly obtain the simulated response message of the cached access request.
- the analog response module 160 can also generate the analog response message of the access request when the response message of the access request sent by the PCIe endpoint device 130 is not received.
- the embodiment of the present invention does not limit the specific implementation manner.
- multiple PCIe endpoint devices 130 can be connected to the CPU 110 through the PCIe switch 150.
- a PCIe switch 150 is also included, the upstream port of the PCIe switch 150 is connected to the CPU 110 through the PCIe bus 140, and the downstream port is provided with one PCIe port for each PCIe endpoint device, and each PCIe port is connected to the PCIe bus 140.
- Each PCIe endpoint device, PCIe switch 150 is used to route data downstream to the corresponding PCIe port, and to route data upstream to CPU 110 from each individual PCIe port.
- the analog response module 160 may be disposed in a signal transmission path between the CPU 110 and the PCIe switch 150 for receiving an access request issued by the CPU 110, and buffering an access request issued by the CPU 110 to the PCIe endpoint.
- the device 130 forwards the access request, monitors the message status of the access request, and if it determines that it receives the response message of the access request sent by the PCIe endpoint device 130, the access request cached by itself is marked.
- the analog response module 160 determines that the PCIe endpoint device 130 has not sent the response message of the access request, acquires the analog response message of the access request, and sends the analog response message to the CPU 110.
- the preset triggering condition may be that the analog response module 160 determines that the PCIe endpoint device 130 is abnormally offline.
- the analog response module 160 is further configured to monitor whether the PCIe endpoint device is abnormally offline.
- the analog response module 160 determines whether it has received an abnormal offline notification of the PCIe endpoint device 130 (e.g., an interrupt notification of the PCIe endpoint device being pulled out of the system;), according to the PCIe endpoint device 130
- the abnormal offline notification determines that the PCIe endpoint device 130 is abnormally offline.
- the preset trigger condition may also be that the timer of the access request times out.
- the analog response module 160 is further configured to start a timer for the access request sent by the PCIe endpoint device 130, after the timer expires, Determining whether the PCIe endpoint device 130 sends a response message to the access request.
- the analog response module 160 after receiving the access request sent by the CPU 110 and caching the access request, is directed to the access Request to start the timer.
- the analog response module monitors the status of the message of the access request, and when the preset trigger condition is reached, if the message status of the access request is still "unfinished", the analog response module will simulate generation.
- the analog response message of the access request is sent to the CPU, so that the CPU does not continue to count the completion of the flow corresponding to the access request, and clears the access request cached by itself, keeps the internal processing of the CPU running normally, and avoids the CPU access request.
- the analog response message of the access request sent by the analog response module 160 to the CPU 110 may include a first type of analog response message or a second type of analog response message, the first type of analog response message indicating the PCIe endpoint The device 130 does not support the access request, and the second type of analog response message indicates that the PCIe endpoint device 130 is abnormal.
- the CPU 110 After receiving the analog response message of the access request, the CPU 110 is further configured to use the first type of analog response.
- the message or the second type of analog response message determines that the access to the PCIe endpoint device 130 has failed, and accordingly the operation of reading the access data is no longer initiated.
- the CPU confirms that the access process is completed according to the simulated response message, and can confirm that the current access is not a successful access, and does not perform the present
- the next step of accessing the related data for example, the read operation
- the embodiment of the present invention prevents the CPU from performing unnecessary operations while maintaining the normal processing flow of the CPU in the case where the PCIe endpoint device is abnormally offline. To avoid wasting system resources.
- the CPU 110 may specifically have more than one access request for the accessed PCIe endpoint device, for example, sending multiple access requests, in this case, simulating
- the response module 160 may monitor the message status of each of the multiple access requests sent by the CPU, for example, buffer each of the multiple access requests, and respond to each of the access requests. The message is monitored, and when the preset trigger condition is reached, the number of access requests that do not receive the response message is counted, and the number of access requests in the plurality of access requests whose message status is "uncompleted" is obtained. And if the quantity exceeds the first threshold, acquiring an analog response message of the access request, and sending the analog response message to the processor.
- the CPU does not reset the MCE as long as there is a response message that is not received by the access request.
- the CPU has a certain tolerance, and only after the number of access requests that do not return the response message is accumulated to a certain value.
- the MCE reset is generated. Therefore, in this embodiment, the first threshold is set, and the analog response module 160 sends the analog response message to the CPU only when the number of access requests that do not receive the response message exceeds the first threshold. , to avoid the MCE reset of the CPU.
- the above-mentioned analog response module 160 can be implemented by software or hardware.
- an FPGA Field-Programmable Gate Array
- a field programmable gate array can be implemented.
- the CPU 110 depicted in Figures 1 and 2 is also only an example, and may be, for example, a specific integrated circuit, in either form, which implements the functions of the processor in a computer system.
- the computer system according to the embodiment of the present invention may be a computing server or a server that manages routes, such as a switch.
- the specific implementation form of the computer system is not limited.
- Accessing the PCIe Endpoint Device The following describes the access flow embodiment of the PCIe endpoint device. As shown in FIG. 3, the flow of accessing the PCIe endpoint device is provided by the PCIe endpoint device. Access to computer systems, including:
- the message status of the access request includes “completed” or “not completed”, “completed” "Represents a response message indicating that the PCIe endpoint device has sent the access request, and "not completed” indicates that the PCIe endpoint device did not send a response message to the access request.
- the analog response module monitors the status of the message of the access request, and may be implemented by monitoring whether it receives the response message of the access request sent by the PCIe endpoint device, if the analog response module receives the The response message of the access request sent by the PCIe endpoint device, the analog response module records the message status of the access request as "completed".
- the analog response module can be disposed on the signal transmission path of the CPU and the PCIe endpoint device, so that the analog response module can receive the access request from the CPU, and can also receive the response message of the access request from the PCIe endpoint device, and simulate The response module may cache the access request after receiving the access request, and if the response message of the access request is received, the cached access request is marked with "completed", or the simulated response module deletes the cached access.
- Request in this case, the status of the message of the access request is marked by whether the access request is cached. If the access request is cached, the status of the message of the access request is not completed, and if the cached access request is deleted, the access is indicated. The requested message status has been completed).
- the preset triggering condition may be that the timer of the access request is timed out, or may be notified that the PCIe endpoint device is abnormally offline.
- the embodiment will be specifically described in the following embodiments.
- the analog response module may obtain an analog response message of the access request generated by itself, where the simulated response message has the same packet header as the access request, and the access request and the simulated response message correspond to each other.
- the CPU can be made to think that the flow corresponding to the access request issued by the CPU has been completed, thereby closing the timer for timing the access request, and clearing
- the access request cached by the CPU itself keeps the internal processing of the CPU normal, and avoids the MCE reset problem of the CPU caused by the accumulation of the access request in the CPU.
- the analog response message of the access request in the embodiment of the present invention is not generated by the PCIe endpoint device, and is not a real response message of the access request, and may be generated by other modules, which may be generated by the analog response module in the embodiment of the present invention.
- the analog response module may pre-generate the analog response message of the access request, or may temporarily generate an analog response message of the access request when determining that the message status of the access request is "uncompleted".
- the analog response module monitors the status of the message of the access request, and when the preset trigger condition is reached, if the message status of the access request is still "unfinished", the analog response module will simulate generation.
- the analog response message of the access request is sent to the CPU, so that the CPU does not continue to count the completion of the flow corresponding to the access request, and clears the access request cached by itself, keeps the internal processing of the CPU running normally, and avoids the CPU access request.
- FIG. 4 a specific access process of the embodiment of the present invention is as shown in FIG. 4, including :
- S401 The CPU in the computer system obtains an operation instruction, the operation instruction instructing the CPU to access the PCIe endpoint device, and the CPU sends an access request to the PCIe endpoint device.
- the driver module of the PCIe endpoint device receives the call of the upstream endpoint, and generates an operation instruction for accessing the PCIe endpoint device according to the pre-configured access interface.
- the CPU obtains the operation instruction.
- the analog response module receives the access request from the CPU, and caches the access request.
- the analog response module is disposed in a signal transmission path between the CPU and the PCIe endpoint device, it is capable of receiving an access request from the CPU;
- the analog response module generates an analog response message of the access request, and the An analog response message is cached in association with the access request.
- the simulation response module may generate the simulation response message of the access request in advance in the step 403, that is, when the access request is cached, or may generate the simulated response when the simulation response message is required in the subsequent step. Message.
- S404 The analog response module forwards the access request to the PCIe endpoint device.
- S405 The analog response module monitors a message status of the access request.
- the analog response module monitors whether it receives the response message of the access request sent by the PCIe endpoint device, and if the analog response module receives the response message of the access request sent by the PCIe endpoint device The analog response module marks the cached access request with a "completed" flag, or the analog response module deletes the cached access request.
- the analog response module may start the monitoring after the step 404, that is, after forwarding the access request to the PCIe endpoint device, or after the step 401, the CPU may send the access request to the PCIe endpoint device. The monitoring is then initiated.
- the analog response module receives a notification that the PCIe endpoint device is abnormally offline, and determines a packet status of the access request.
- the preset trigger condition is that the notification that the PCIe endpoint device is abnormally disconnected is received. Since the analog response module sets the signal transmission path between the CPU and the PCIe endpoint device, if the PCIe endpoint device is pulled out, the notification that the PCIe endpoint device is abnormally offline to the CPU passes through the analog response module, and the analog response module receives the PCIe. The notification that the endpoint device is abnormally offline determines that the preset trigger condition is reached, queries the packet status of the access request, and determines the packet status.
- the analog response message may be a first type of analog response message, indicating that the PCIe endpoint device 130 does not support the access request, for example, the simulated response message carries The Unsupported Request (UR) indicates, or the analog response message may be a second type of analog response message indicating that the PCIe endpoint device 130 is abnormal, for example, the analog response message carries a Complete Abort (CA) indication.
- UR Unsupported Request
- CA Complete Abort
- the CPU receives the analog response message, and determines, according to the simulated response message, that the process corresponding to the access request is completed, and clears the access request cached by itself.
- the access request and the analog response message correspond to each other, so that the CPU can consider that the flow corresponding to the previously issued access request has been completed, thereby closing the access request.
- the timing timer is executed, and the access request of the CPU itself is cleared, and the internal processing of the CPU is maintained normally, thereby avoiding the MCE reset problem of the CPU caused by the accumulation of the access request in the CPU.
- the analog response module may further carry the UR or the CA indication, and the analog response module may further determine, according to the indication information carried in the analog response message, that the access to the PCIe endpoint device fails, and the access error processing is performed.
- the access error processing may include: inhibiting further operations (such as reading of data) related to the access request by the CPU, or recording a status of the PCIe endpoint device as abnormal offline, thereby causing the CPU to abort the The PCIe endpoint device performs subsequent access operations.
- S409 The analog response module forwards the notification that the PCIe endpoint device is abnormally offline to the CPU.
- the CPU may perform a power-off operation and a related resource release operation on the physical device corresponding to the PCIe endpoint device according to the notification that the PCIe endpoint device is abnormally offline.
- the analog response module monitors the status of the message of the access request, and when receiving the notification that the accessed PCIe endpoint device is abnormally offline, if the status of the access request message is still in an "incomplete" state, The analog response module sends the simulated response message of the simulated access request to the CPU, so that the CPU does not timeout the completion of the process corresponding to the access request, keeps the internal processing of the CPU running normally, and avoids the CPU internal access request timeout.
- the MCE reset problem caused by not completing.
- the analog response module sets the CPU In the signal transmission path between the PCIe and the endpoint device, the status of the access request packet can be monitored by the response message of the access request sent by the accessed PCIe endpoint device, which is convenient, flexible, and efficient.
- the CPU confirms that the access process is completed according to the simulated response message, and can confirm that the current access is not a successful access, and does not perform according to this.
- the embodiment of the present invention prevents the CPU from performing some unnecessary operations while maintaining the normal CPU processing flow when the PCIe endpoint device is abnormally offline. Operation to avoid wasting system resources.
- Embodiment 2 of the method of the present invention the analog response module in this embodiment is still set in the CPU and
- the method includes:
- S501 The CPU in the computer system obtains an operation instruction, the operation instruction instructing the CPU to access the PCIe endpoint device, and the CPU sends an access request to the PCIe endpoint device.
- the analog response module receives the access request from the CPU, and caches the access request.
- the analog response module Since the analog response module sets a signal transmission path between the CPU and the PCIe endpoint device, it can receive an access request from the CPU.
- S503 The analog response module starts a timer for the access request, and starts timing.
- S504 The analog response module generates an analog response message of the access request, and caches the simulated response message with the access request.
- the simulation response module may generate the simulation response message of the access request in advance in the step 503, that is, when the access request is cached, or may generate the simulated response when the simulation response message is needed in the subsequent step. Message.
- the analog response module may start executing steps 503 and 504 simultaneously after receiving the access request, or may perform 503 and then perform 504.
- S505 The analog response module forwards the access request to the PCIe endpoint device.
- S506 The analog response module monitors a message status of the access request.
- the analog response module monitors whether it receives the response message of the access request sent by the PCIe endpoint device, and if the analog response module receives the response message of the access request sent by the PCIe endpoint device The analog response module marks the cached access request with a "completed" flag, or the analog response module deletes the cached access request.
- the analog response module determines a message status of the access request after the timer expires.
- the preset trigger condition is whether the timer of the access request times out, and the timing of the timer is set according to an empirical value, where the experience value is the maximum tolerance time of the response message sent by the PCIe endpoint device to the access request.
- the analog response module must receive a response message for the access request it sent during the empirical time.
- the analog response module After the timer expires, the analog response module queries the message status of the access request. If it is determined that the message status is "incomplete", it can be determined that the accessed PCIe endpoint device is abnormally offline, and the CPU cannot receive the message. The response message of the request is accessed. At this time, in order to avoid the MCE reset problem caused by the CPU cache access request timeout, the analog response module sends the analog response message generated in advance to the CPU.
- the analog response message may be a first type of analog response message, indicating that the PCIe endpoint device 130 does not support the access request, for example, the analog response message carries an Unsupported Request (UR) indication, or The analog response message may be a second type of analog response message indicating that the PCIe endpoint device 130 is abnormal, eg, the analog response message carries a Completer Abort (CA) indication.
- UR Unsupported Request
- CA Completer Abort
- the access request and the analog response message correspond to each other, so that the CPU can consider that the flow corresponding to the previously issued access request has been completed, thereby closing the access request.
- the timing timer is executed, and the access request of the CPU itself is cleared, and the internal processing of the CPU is maintained normally, thereby avoiding the MCE reset problem of the CPU caused by the accumulation of the access request in the CPU.
- the analog response module may further carry an access error process according to the indication information carried in the analog response message, and the access error processing may include: suppressing the CPU pair.
- the access request is related to further operations (such as reading of data), or the status of the PCIe endpoint device is recorded as abnormal offline, such that the CPU suspends subsequent access operations to the PCIe endpoint device.
- the analog response module determines a status of the PCIe endpoint device.
- the analog response module further determines whether the status of the PCIe endpoint device is abnormally offline or fails. If the PCIe endpoint device is abnormally offline, step 511 is performed. If the PCIe endpoint device fails, step 512 is performed.
- the analog response module may determine whether it receives a notification that the PCIe endpoint device is abnormally offline, and determines whether the PCIe endpoint device is abnormally offline according to whether the notification is received, or may send a probe message to the PCIe endpoint device to determine The status of the PCIe endpoint device.
- S511 If the analog response module determines that the PCIe endpoint device is abnormally offline, sending a notification that the PCIe endpoint device is abnormally offline to the CPU, so that the CPU is configured to notify the PCIe endpoint device that the offline device is abnormally offline.
- the physical device corresponding to the PCIe endpoint device performs a power-off operation and a resource release operation.
- the analog response module monitors the status of the message by accessing the request, After the timer expires, if the message status of the access request is still "incomplete", the analog response module sends an analog response message simulating the generated access request to the CPU, so that the CPU does not process the corresponding request for the access request. The timeout is completed, and the internal processing of the CPU is maintained, which avoids the MCE reset problem caused by the incomplete CPU internal access request timeout.
- the CPU confirms that the access process is completed according to the simulated response message, and can confirm that the current access is not a successful access, and does not perform according to this.
- the embodiment of the present invention prevents the CPU from performing some unnecessary operations while maintaining the normal CPU processing flow when the PCIe endpoint device is abnormally offline. Operation to avoid wasting system resources.
- the analog response device provided by the embodiment of the present invention is applied to a computer system, and the PCIe endpoint device accesses the computer system through a PCIe bus. As shown in FIG. 6, the device interacts with the PCIe endpoint device 130 and the CPU 110. Includes:
- the monitoring unit 601 is configured to monitor, after the CPU 110 in the computer system sends an access request to the PCIe endpoint device, the packet status of the access request, where the message status of the access request includes Completing "or” is not completed, the "not completed” indicating that the PCIe endpoint device does not send the response message of the access request;
- a determining unit 602 configured to determine, according to a monitoring result of the monitoring unit 601, a packet status of the access request when a preset trigger condition is met;
- the obtaining unit 603 is configured to: when the determining unit 602 determines that the message status of the access request is “uncompleted”, obtain an analog response message of the access request;
- the response unit 604 is configured to send, to the CPU 110, the analog response message acquired by the acquiring unit 603.
- the apparatus may further include:
- the interaction unit 605 is configured to receive an access request sent by the CPU 110, and forward the access request to the PCIe endpoint device 130.
- the monitoring unit 601 is specifically configured to monitor whether the interaction unit 605 receives the response message of the access request sent by the PCIe endpoint device, and if the response message of the access request sent by the PCIe endpoint device is received, , the status of the message of the access request is recorded as "completed,".
- the monitoring unit 601 is further configured to cache the access request received by the interaction unit 605;
- the monitoring unit 601 is specifically configured to: when the interaction unit 605 receives the response message of the access request sent by the PCIe endpoint device, the cached access request is
- the determining unit 602 is specifically configured to query, when the preset trigger condition is reached, the access request cached by the monitoring unit 601, if the cached access request is marked as “completed” or the cached access If the request has been deleted, it is determined that the message status of the access request is "completed”, if the cached access request is marked as "incomplete” or the cached access request has no "completed” flag, then The message status of the access request is determined to be "incomplete”.
- the monitoring unit 601 is further configured to: after the interaction unit 605 receives the access request sent by the CPU 110, generate an analog response message of the access request. And buffering the simulated response message of the access request;
- the obtaining unit 603 is specifically configured to acquire an analog response message of the access request buffered by the monitoring unit 601.
- the analog response message of the access request includes a first type analog response message or a second type analog response message, and the first type analog response message indicates the PCIe endpoint device The access request is not supported, and the second type of analog response message indicates that the PCIe endpoint device is abnormal.
- the preset triggering condition is The PCIe endpoint device is abnormally offline
- the monitoring unit 601 is further configured to: monitor whether the PCIe endpoint device is abnormally offline, and notify the determining unit 602 to perform the determining the packet status of the access request when determining that the PCIe endpoint device is abnormally offline. A step of.
- the monitoring unit 601 is specifically configured to: determine whether the interaction unit 605 receives an abnormal offline notification of the PCIe endpoint device, and if yes, determine the PCIe endpoint device. An exception occurred offline.
- the interaction unit 605 is further configured to: forward an abnormal offline notification of the PCIe endpoint device to the CPU 110, so that the CPU 110 is configured according to the received PCIe endpoint.
- the abnormal offline notification of the device 130 performs a power-off operation on the physical device corresponding to the PCIe endpoint device 130.
- the preset trigger condition is that the timer of the access request times out
- the monitoring unit 601 is further configured to:
- the access request start timer sent by the CPU 110 is timed, in the step of counting.
- the monitoring unit 601 is further configured to:
- Determining a status of the PCIe endpoint device if it is determined that the status of the PCIe endpoint device is abnormally offline, notifying the interaction unit 605 to send an abnormal offline notification of the PCIe endpoint device 130 to the CPU 110, so that the CPU 110
- the physical device corresponding to the PCIe endpoint device 130 is powered off according to the received abnormal offline notification of the PCIe endpoint device 130.
- the CPU 110 sends multiple access requests to the PCIe endpoint device 130;
- the monitoring unit 601 is specifically configured to send a message for each access request of the multiple access requests. Status monitoring;
- the determining unit 602 is specifically configured to: when the preset trigger condition is reached, count the number of access requests that do not receive the response message in the multiple access requests, and determine whether the quantity exceeds the first threshold;
- the obtaining unit 603 is specifically configured to acquire a simulated response message of the access request when the quantity exceeds the first threshold.
- FIG. 8 is a schematic structural diagram of a computer according to an embodiment of the present invention.
- a computer of an embodiment of the invention may include:
- the processor 801, the memory 802, and the communication interface 805 are connected by the system bus 804 and communicate with each other.
- Processor 801 may be a single core or multi-core central processing unit, or a particular integrated circuit, or one or more integrated circuits configured to implement embodiments of the present invention.
- the memory 802 may be a high speed RAM memory or a non-volatile memory such as at least one disk memory.
- Memory 802 is used by computer to execute instructions 803. Specifically, the program code may be included in the computer execution instruction 803.
- the processor 801 runs the computer execution instruction 803, and the method flow described in any one of FIG. 3 to FIG. 5 may be performed, and details are not repeatedly described herein.
- aspects of the present invention, or possible implementations of various aspects can be embodied as a system, method, or computer program product.
- aspects of the invention, or possible implementations of various aspects may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, etc.), or a combination of software and hardware aspects, They are collectively referred to herein as "circuits," “modules,” or “systems.”
- aspects of the invention, or possible implementations of various aspects may take the form of a computer program product, which is a computer readable program code stored in a computer readable medium.
- the computer readable medium can be a computer readable signal medium or a computer readable storage medium.
- the computer readable storage medium includes, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing, such as a random access memory.
- RAM Read Only Memory
- ROM Read Only Memory
- EPROM Erasable Programmable Read Only Memory
- CD-ROM Portable Read Only Memory
- the processor in the computer reads the computer readable program code stored in the computer readable medium, such that the processor can perform the functional actions specified in each step or combination of steps in the flowchart; A device that functions as specified in each block, or combination of blocks.
- the computer readable program code can execute entirely on the user's computer, partly on the user's computer, as a separate software package, partly on the user's computer and partly on the remote computer, or entirely on the remote computer or server.
- the functions noted in the various steps of the flowchart, or in the blocks in the block diagrams may not occur in the order noted. For example, two steps, or two blocks shown in succession may be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
Claims
Priority Applications (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP13791889.2A EP2829986B1 (en) | 2013-05-20 | 2013-05-20 | Computer system, access method and apparatus for peripheral component interconnect express endpoint devices |
| JP2015517591A JP5932146B2 (ja) | 2013-05-20 | 2013-05-20 | ピーシーアイエクスプレスのエンドポイントデバイスにアクセスするための方法、コンピューターシステム、および、装置 |
| PCT/CN2013/075920 WO2014186938A1 (zh) | 2013-05-20 | 2013-05-20 | 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 |
| CN201380002624.0A CN103797469B (zh) | 2013-05-20 | 2013-05-20 | 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 |
| US14/141,154 US9384110B2 (en) | 2013-05-20 | 2013-12-26 | Method, computer system, and apparatus for accessing peripheral component interconnect express endpoint device |
| US14/706,667 US9323635B2 (en) | 2013-05-20 | 2015-05-07 | Method, computer system, and apparatus for accessing peripheral component interconnect express endpoint device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2013/075920 WO2014186938A1 (zh) | 2013-05-20 | 2013-05-20 | 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US14/141,154 Continuation US9384110B2 (en) | 2013-05-20 | 2013-12-26 | Method, computer system, and apparatus for accessing peripheral component interconnect express endpoint device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2014186938A1 true WO2014186938A1 (zh) | 2014-11-27 |
Family
ID=50671623
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2013/075920 Ceased WO2014186938A1 (zh) | 2013-05-20 | 2013-05-20 | 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 |
Country Status (5)
| Country | Link |
|---|---|
| US (2) | US9384110B2 (zh) |
| EP (1) | EP2829986B1 (zh) |
| JP (1) | JP5932146B2 (zh) |
| CN (1) | CN103797469B (zh) |
| WO (1) | WO2014186938A1 (zh) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104133759A (zh) * | 2014-08-06 | 2014-11-05 | 浪潮(北京)电子信息产业有限公司 | 一种实现扩展模组移除的方法及装置 |
| CN105893293B (zh) * | 2014-12-18 | 2019-10-29 | 伊姆西公司 | 用于管理外部组件快速互连设备热插拔的系统和方法 |
| CN104657317B (zh) * | 2015-03-06 | 2017-12-26 | 北京百度网讯科技有限公司 | 服务器 |
| TWI559148B (zh) * | 2015-05-11 | 2016-11-21 | 廣達電腦股份有限公司 | 自動硬體恢復方法及自動硬體恢復系統 |
| CN105022716A (zh) * | 2015-06-24 | 2015-11-04 | 浪潮电子信息产业股份有限公司 | 一种多数据链路的gpu服务器 |
| JP7103231B2 (ja) * | 2017-01-13 | 2022-07-20 | 日本電気株式会社 | コネクション管理ユニット、およびコネクション管理方法 |
| CN108733479B (zh) | 2017-04-24 | 2021-11-02 | 上海宝存信息科技有限公司 | 卸载固态硬盘卡的方法以及使用该方法的装置 |
| CN107682397B (zh) | 2017-08-28 | 2019-02-26 | 平安科技(深圳)有限公司 | 客户资源获取方法、装置、终端设备及存储介质 |
| CN107678994B (zh) * | 2017-09-15 | 2021-04-20 | 华为技术有限公司 | PCIe设备热拔方法及装置 |
| CN110413554B (zh) * | 2018-04-28 | 2021-05-28 | 杭州海康威视数字技术股份有限公司 | 热插拔系统及热插拔方法 |
| CN109324991B (zh) * | 2018-09-21 | 2021-12-03 | 郑州云海信息技术有限公司 | 一种pcie设备的热插拔装置、方法、介质及系统 |
| CN112015580A (zh) * | 2019-05-31 | 2020-12-01 | 瑞昱半导体股份有限公司 | 具资料传收机制的电子装置以及资料传收方法 |
| CN110851330B (zh) * | 2019-11-13 | 2023-04-21 | 重庆秦嵩科技有限公司 | 一种cpu在线监控方法 |
| CN110928827B (zh) * | 2019-11-29 | 2021-06-22 | 苏州浪潮智能科技有限公司 | 一种访问PCIe设备的优化的方法和设备 |
| US11449456B2 (en) | 2020-06-02 | 2022-09-20 | Microchip Technology Inc. | System and method for scheduling sharable PCIe endpoint devices |
| WO2024152345A1 (en) * | 2023-01-20 | 2024-07-25 | Beijing Baidu Netcom Science Technology Co., Ltd. | Peripheral component interconnect express device reset for autonomous driving systems |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996146A (zh) * | 2009-08-26 | 2011-03-30 | 戴尔产品有限公司 | 多模式处理模块及其使用方法 |
| CN102023941A (zh) * | 2009-09-23 | 2011-04-20 | 联想(北京)有限公司 | 一种笔记本电脑 |
| US20120079312A1 (en) * | 2010-09-24 | 2012-03-29 | Sridhar Muthrasanallur | Method and system of live error recovery |
Family Cites Families (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0283649A (ja) * | 1988-09-20 | 1990-03-23 | Fujitsu Ltd | バス監視方式 |
| US5337413A (en) * | 1992-02-06 | 1994-08-09 | Tandem Computers Incorporated | Environment monitoring system for standard interface bus computer systems |
| DE69507360T2 (de) * | 1994-04-06 | 1999-06-17 | Advanced Micro Devices, Inc., Sunnyvale, Calif. | Parallelschlussschnittstellenschaltkreise in Rechnersystemen |
| US5594875A (en) * | 1994-07-01 | 1997-01-14 | Digital Equipment Corporation | Method and apparatus to provide pended transaction on a non-pended system bus |
| US5819027A (en) * | 1996-02-28 | 1998-10-06 | Intel Corporation | Bus patcher |
| US6070207A (en) * | 1998-01-26 | 2000-05-30 | Intel Corporation | Hot plug connected I/O bus for computer system |
| US6751698B1 (en) * | 1999-09-29 | 2004-06-15 | Silicon Graphics, Inc. | Multiprocessor node controller circuit and method |
| JP4144986B2 (ja) * | 1999-12-13 | 2008-09-03 | 富士通株式会社 | 情報処理装置、及び、トランザクション処理方法、プロセッサ、並びに、リトライ方法 |
| US7058750B1 (en) * | 2000-05-10 | 2006-06-06 | Intel Corporation | Scalable distributed memory and I/O multiprocessor system |
| GB2367647B (en) * | 2000-10-03 | 2002-11-20 | Sun Microsystems Inc | Resource access control for a processor |
| DE60213616T2 (de) | 2001-08-24 | 2007-08-09 | Intel Corporation, Santa Clara | Eine allgemeine eingabe-/ausgabearchitektur, protokoll und entsprechende verfahren zur umsetzung der flusssteuerung |
| JP3665624B2 (ja) * | 2002-01-25 | 2005-06-29 | 沖電気工業株式会社 | マイクロコントローラ |
| US7260749B2 (en) * | 2003-01-22 | 2007-08-21 | Red Hat, Inc. | Hot plug interfaces and failure handling |
| US20070033314A1 (en) * | 2005-08-04 | 2007-02-08 | Arad Rostampour | Event generation for device hotplug |
| JP4630766B2 (ja) * | 2005-08-26 | 2011-02-09 | キヤノン株式会社 | プライマリシステムとセカンダリシステムを有するシステム |
| US20070088987A1 (en) * | 2005-10-13 | 2007-04-19 | Hiroshi Kimizuka | System and method for handling information transfer errors between devices |
| JP4555791B2 (ja) * | 2006-03-16 | 2010-10-06 | 富士通株式会社 | データ読出方法及びデータ読出装置 |
| US8935437B2 (en) * | 2006-06-14 | 2015-01-13 | Dell Products L.P. | Peripheral component health monitoring apparatus |
| CN100511162C (zh) * | 2006-09-29 | 2009-07-08 | 华为技术有限公司 | 一种隔离总线故障的方法、装置与一种单板 |
| CN100504823C (zh) * | 2007-04-19 | 2009-06-24 | 杭州华三通信技术有限公司 | 实现周边元件接口加速总线插卡热插拔的方法及装置 |
| US8151145B2 (en) * | 2008-04-03 | 2012-04-03 | Oracle America, Inc. | Flow control timeout mechanism to detect PCI-express forward progress blockage |
| JP4988671B2 (ja) * | 2008-08-29 | 2012-08-01 | インターナショナル・ビジネス・マシーンズ・コーポレーション | シリアルバスシステム及びハングアップスレーブリセット方法 |
| JP2010140361A (ja) * | 2008-12-12 | 2010-06-24 | Fujitsu Microelectronics Ltd | コンピュータシステム及び異常検出回路 |
| JP5293165B2 (ja) | 2008-12-25 | 2013-09-18 | 富士通セミコンダクター株式会社 | シミュレーション支援プログラム、シミュレーション装置、およびシミュレーション支援方法 |
| JP5617429B2 (ja) | 2010-08-19 | 2014-11-05 | ソニー株式会社 | バスシステムおよびバスシステムと接続機器とを接続するブリッジ回路 |
| JP5316502B2 (ja) | 2010-09-16 | 2013-10-16 | 日本電気株式会社 | I/oブリッジ装置、応答通知方法、及びプログラム |
| US8443126B2 (en) * | 2010-09-22 | 2013-05-14 | Wilocity, Ltd. | Hot plug process in a distributed interconnect bus |
| CN102193799A (zh) * | 2011-03-25 | 2011-09-21 | 上海磁浮交通发展有限公司 | 一种用于安全计算机通信的安全中间系统 |
| US9086945B2 (en) * | 2011-09-01 | 2015-07-21 | Dell Products, Lp | System and method to correlate errors to a specific downstream device in a PCIe switching network |
| JP5745996B2 (ja) * | 2011-10-28 | 2015-07-08 | 京セラドキュメントソリューションズ株式会社 | 情報処理装置および画像形成装置 |
| CN102662903B (zh) * | 2012-03-31 | 2016-09-28 | 浪潮电子信息产业股份有限公司 | 一种通过cpld或fpga实现pcie设备热插拔的方法 |
-
2013
- 2013-05-20 EP EP13791889.2A patent/EP2829986B1/en active Active
- 2013-05-20 JP JP2015517591A patent/JP5932146B2/ja active Active
- 2013-05-20 WO PCT/CN2013/075920 patent/WO2014186938A1/zh not_active Ceased
- 2013-05-20 CN CN201380002624.0A patent/CN103797469B/zh active Active
- 2013-12-26 US US14/141,154 patent/US9384110B2/en active Active
-
2015
- 2015-05-07 US US14/706,667 patent/US9323635B2/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996146A (zh) * | 2009-08-26 | 2011-03-30 | 戴尔产品有限公司 | 多模式处理模块及其使用方法 |
| CN102023941A (zh) * | 2009-09-23 | 2011-04-20 | 联想(北京)有限公司 | 一种笔记本电脑 |
| US20120079312A1 (en) * | 2010-09-24 | 2012-03-29 | Sridhar Muthrasanallur | Method and system of live error recovery |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP2829986A4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| US20150242341A1 (en) | 2015-08-27 |
| JP2015524122A (ja) | 2015-08-20 |
| US20140344632A1 (en) | 2014-11-20 |
| EP2829986B1 (en) | 2017-07-12 |
| JP5932146B2 (ja) | 2016-06-08 |
| CN103797469A (zh) | 2014-05-14 |
| US9323635B2 (en) | 2016-04-26 |
| EP2829986A1 (en) | 2015-01-28 |
| US9384110B2 (en) | 2016-07-05 |
| CN103797469B (zh) | 2016-08-31 |
| EP2829986A4 (en) | 2015-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103797469B (zh) | 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 | |
| JP5953573B2 (ja) | ペリフェラル・コンポーネント・インターコネクト・エクスプレス・エンドポイントデバイスにアクセスするためのコンピュータシステム、方法、および装置 | |
| US8443237B2 (en) | Storage apparatus and method for controlling the same using loopback diagnosis to detect failure | |
| CN109766302B (zh) | 设备管理的方法和装置 | |
| CN110535692B (zh) | 故障处理方法、装置、计算机设备、存储介质及存储系统 | |
| WO2015035574A1 (zh) | 一种故障处理的方法、计算机系统和装置 | |
| CN115550291B (zh) | 交换机的复位系统及方法、存储介质、电子设备 | |
| TWI512490B (zh) | 擷取控制台訊息的系統及其方法及非暫態電腦可讀媒體 | |
| WO2021098485A1 (zh) | PCIe设备的上下电控制方法以及系统 | |
| CN106126454B (zh) | 一种计算机系统、高速外围组件互联端点设备的访问方法和装置 | |
| US7676701B2 (en) | Computer readable medium storing an error recovery program, error recovery method, error recovery apparatus, and computer system | |
| CN112804115B (zh) | 一种虚拟网络功能的异常检测方法、装置及设备 | |
| WO2021136099A1 (zh) | 数据传输方法、装置、设备及存储介质 | |
| TWI483125B (zh) | 基板管理控制器恢復系統及其使用方法 | |
| JP2016225943A (ja) | 装置状態の復旧システム,及びシステムの装置状態復旧方法 | |
| US20130238284A1 (en) | Testing method | |
| TW201820137A (zh) | 具有重啟功能的裝置 | |
| JP6112205B2 (ja) | 情報処理システム、装置、方法及びプログラム | |
| CN121233403A (zh) | PCIe Switch端口复位方法和装置 | |
| JP6032024B2 (ja) | コンピュータシステム、コンピュータシステムの起動方法、及びプログラム | |
| TW201205437A (en) | Server system | |
| WO2014115293A1 (ja) | 情報処理装置、情報処理方法及び情報処理システム |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| REEP | Request for entry into the european phase |
Ref document number: 2013791889 Country of ref document: EP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2013791889 Country of ref document: EP |
|
| ENP | Entry into the national phase |
Ref document number: 2015517591 Country of ref document: JP Kind code of ref document: A |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13791889 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |