SSD front-end error processing method and device, computer equipment and storage medium

Abstract

The invention relates to an SSD front-end error processing method and device, computer equipment and a storage medium, and the method comprises the steps: detecting an uncorrectable error through an error detection mechanism, and transmitting error report information to a root assembly in a graded manner according to the error severity of the uncorrectable error; checking the current link state and the TLP information, triggering a corresponding interrupt request, and controlling other functional modules in the SSD to pause, so as to enable the NVMe hardware in the SSD to enter a pause state;querying error report information of the current link, and judging whether the current link state is normal or not; if not, reinitializing the PCIe physical layer and the error report control register, and retraining the link; if yes, eliminating the error state of the NVMe hardware; clearing an error state bit through firmware in cooperation with NVMe hardware, and updating a current error statepointer; and resetting the host, and enabling the SSD to enter a normal working state. The intervention of firmware ensures the efficiency and reliability of error processing, and further ensures thestability of SSD interface communication.

Description

technical field [0001] The present invention relates to the field of SSD solid state hard disks, and more specifically refers to an SSD front-end error handling method, device, computer equipment and storage medium. Background technique [0002] The NVMe (Non-Volatile Memory express) standard is a logical device interface standard based on the M.2 interface. NVMe SSD is based on PCIe (peripheral component interconnect express, a high-speed serial computer expansion bus standard) and inherits the PCIe The error reporting mechanism (AER) can well realize error detection, error registration, error reporting and error message sending, but when dealing with some special application scenarios, for example, when AER encounters an uncorrectable error (Uncorrect error) , if the error report function is enabled, the error report information will be sent to the root component (Rootcomplex, hereinafter referred to as RC), and if other errors occur to the link at this time, the previousl...

AI Technical Summary

Problems solved by technology

[0002] The NVMe (Non-Volatile Memory express) standard is a logical device interface standard based on the M.2 interface. NVMe SSD is based on PCIe (peripheral component interconnect express, a high-speed serial computer expansion bus standard) and inherits the PCIe The error reporting mechanism (AER) can well realize error detection, error registration, error reporting and error information transmission, but when dealing with some special application scenarios, for example, when AER encounters an uncorrectable error (Uncorrect error) , if the error report function is turned on, the error report information will be sent to the root component (Rootcomplex, hereinafter referred to as RC). If the error processing request from RC is not reached, CPLtimeout will be generated, and the SSD will enter an abnormal state and cannot work normally, which will affect the communication stability of the SSD interface.

Images