PCIe device error handling system
A PCIe device error handling system includes a BIOS subsystem coupled to a PCIe device and a BMC device. The BIOS subsystem identifies an error in the PCIe device and, in response, begins an SMM that suspends the performance of at least one workload in an operating system, and generates and transmits a PCIe device error information collection instruction associated with the PCIe device to the BMC device. Subsequent to transmitting the PCIe device error information collection instruction, the BIOS subsystem ends the SMM such that the performance of at least one workload is resumed in the operating system. In response to receiving the PCIe device error information collection instruction from the BIOS subsystem, the BMC device retrieves PCIe device error information from the PCIe device while the operating system performs the at least one workload, and stores the PCIe device error information.
1. A Peripheral Component Interconnect express (PCIe) device error handling system, comprising:
an operating system;
a Peripheral Component Interconnect express (PCIe) device;
a Basic Input/Output System (BIOS) subsystem that is coupled to the PCIe device and that is configured to:
identify an error in the PCIe device and, in response, begin a System Management Mode (SMM) that suspends the performance of at least one workload in the operating system;
generate and transmit a PCIe device error information collection instruction associated with the PCIe device; and
end, subsequent to transmitting the PCIe device error information collection instruction, the SMM such that the performance of at least one workload is resumed in the operating system; and
a Baseboard Management Controller (BMC) device that is coupled to the BIOS and that is configured to:
receive, from the BIOS subsystem, the PCIe device error information collection instruction and, in response, retrieve PCIe device error information from the PCIe device while the operating system performs the at least one workload; and
store the PCIe device error information.
2. The system of claim 1 , wherein the PCIe device error information retrieved by the BMC device includes a first subset of PCIe device error information that is included in at least one register in the PCIe device, and wherein the BIOS subsystem is configured to:
retrieve, in response to identifying the error in the PCIe device, a second subset of the PCIe device error information that is included in at least one register in the PCIe device and that is different than the first subset of the PCIe device error information; and
store the second subset of the PCIe device error information.
3. The system of claim 1 , wherein the BMC device is configured to:
identify, in response to receiving the PCIe device error information collection instruction, at least one port that is connected to the PCIe device; and
store a port identifier for each of the at least one port.
4. The system of claim 1 , wherein the BMC device is configured to:
identify, in response to receiving the PCIe device error information collection instruction, at least one bridge device that is connected to the PCIe device; and
store a bridge device identifier for each of the at least one bridge device.
5. The system of claim 1 , wherein the PCIe device is a Non-Volatile Memory express (NVMe) storage device and the PCIe device error information is retrieved from a queue included in the NVMe storage device.
6. The system of claim 1 , wherein the BMC device is configured to:
perform error causation analysis operations using the PCIe device error information to determine at least one cause of the error identified in the PCIe device; and
provide the at least one cause of the error identified in the PCIe device for display on a display device.
7. An Information Handling System (IHS), comprising:
a Basic Input/Output System (BIOS) processing system;
a BIOS memory system that is coupled to the BIOS processing system and that includes instructions that, when executed by the BIOS processing system, cause the BIOS processing system to provide a BIOS engine that is configured to:
identify an error in a Peripheral Component Interconnect express (PCIe) device that is coupled to the BIOS processing system and, in response, begin a System Management Mode (SMM) that suspends the performance of at least one workload in an operating system;
generate and transmit a PCIe device error information collection instruction associated with the PCIe device; and
end, subsequent to transmitting the PCIe device error information collection instruction, the SMM such that the performance of at least one workload is resumed in the operating system;
a Baseboard Management Controller (BMC) processing system that is coupled to the BIOS processing system and the PCIe device; and
a BMC memory system that is coupled to the BMC processing system and that includes instructions that, when executed by the BMC processing system, cause the BMC processing system to provide a BMC engine that is configured to:
receive, from the BIOS engine, the PCIe device error information collection instruction and, in response, retrieve PCIe device error information from the PCIe device while the operating system performs the at least one workload; and
store the PCIe device error information.
8. The IHS of claim 7 , wherein the PCIe device error information retrieved by the BMC engine includes a first subset of PCIe device error information that is included in at least one register in the PCIe device, and wherein the BIOS engine is configured to:
retrieve, in response to identifying the error in the PCIe device, a second subset of the PCIe device error information that is included in at least one register in the PCIe device and that is different than the first subset of the PCIe device error information; and
store the second subset of the PCIe device error information.
9. The IHS of claim 8 , wherein the first subset of the PCIe device error information includes Base Address Register (BAR) information, command (CMD) register information, and status register information.
10. The IHS of claim 7 , wherein the BMC engine is configured to:
identify, in response to receiving the PCIe device error information collection instruction, at least one port that is connected to the PCIe device; and
store a port identifier for each of the at least one port.
11. The IHS of claim 7 , wherein the BMC engine is configured to:
identify, in response to receiving the PCIe device error information collection instruction, at least one bridge device that is connected to the PCIe device; and
store a bridge device identifier for each of the at least one bridge device.
12. The IHS of claim 7 , wherein the PCIe device is a Non-Volatile Memory express (NVMe) storage device and the PCIe device error information is retrieved from a queue included in the NVMe storage device.
13. The IHS of claim 7 , wherein the BMC engine is configured to:
perform error causation analysis operations using the PCIe device error information to determine at least one cause of the error identified in the PCIe device; and
provide the at least one cause of the error identified in the PCIe device for display on a display device.
14. A method for handling errors in a Peripheral Component Interconnect express (PCIe) device, comprising:
identifying, by a Basic Input/Output System (BIOS) subsystem, an error in a Peripheral Component Interconnect express (PCIe) device that is coupled to the BIOS subsystem and, in response, beginning a System Management Mode (SMM) that suspends the performance of at least one workload in an operating system;
generating and transmitting, by the BIOS subsystem, a PCIe device error information collection instruction associated with the PCIe device;
ending, by the BIOS subsystem subsequent to transmitting the PCIe device error information collection instruction, the SMM such that the performance of at least one workload is resumed in the operating system;
receiving, by a Baseboard Management Controller (BMC) device from the BIOS subsystem, the PCIe device error information collection instruction and, in response, retrieving PCIe device error information from the PCIe device while the operating system performs the at least one workload; and
storing, by the BMC device, the PCIe device error information.
15. The method of claim 14 , wherein the PCIe device error information retrieved by the BMC subsystem includes a first subset of PCIe device error information that is included in at least one register in the PCIe device, and wherein the method includes:
retrieving, by the BIOS subsystem in response to identifying the error in the PCIe device, a second subset of the PCIe device error information that is included in at least one register in the PCIe device and that is different than the first subset of the PCIe device error information; and
storing, by the BIOS subsystem, the second subset of the PCIe device error information.
16. The method of claim 15 , wherein the first subset of the PCIe device error information includes Base Address Register (BAR) information, command (CMD) register information, and status register information.
17. The method of claim 14 , further comprising:
identifying, by the BMC device in response to receiving the PCIe device error information collection instruction, at least one port that is connected to the PCIe device; and
storing, by the BMC device, a port identifier for each of the at least one port.
18. The method of claim 14 , further comprising:
identifying, by the BMC device in response to receiving the PCIe device error information collection instruction, at least one bridge device that is connected to the PCIe device; and
storing, by the BMC device, a bridge device identifier for each of the at least one bridge device.
19. The method of claim 14 , wherein the PCIe device is a Non-Volatile Memory express (NVMe) storage device and the PCIe device error information is retrieved from a queue included in the NVMe storage device.
20. The method of claim 14 , further comprising:
performing, by the BMC device, error causation analysis operations using the PCIe device error information to determine at least one cause of the error identified in the PCIe device; and
providing, by the BMC device, the at least one cause of the error identified in the PCIe device for display on a display device.