TWI807521B - Warning method - Google Patents

Abstract

A warning method includes: checking whether there is an garbled string in a plurality of types of an SMBIOS in the process of power on self-test(POST); when it is determined that there is the garbled string in the type, saving a corresponding error message in a system event log(SEL); when the error message is recorded in the SEL, changing an error flag stored in a non-volatile memory unit from a first logical value to a second logical value; continuing to execute a boot procedure. In this way, the user or developer can instantly know that the SMBIOS contains the garbled string through the SEL, and can correct and debug as soon as possible.

Description

Warning method

The present invention relates to a warning method, in particular to a warning method for identifying tampered system management basic input/output system (SMBIOS).

Smbios (System management bios) is a unified specification followed by motherboard or system manufacturers to display product information in a standard format. In computer systems, the input or update of SMBIOS is usually performed by third-part software (Third-part tool). However, sometimes due to human or software factors, unrecognizable characters (ie ASCII 0x00~0x1F) will be generated during the writing process, resulting in SMBIOS being tampered with (garbled). For example, when unreadable characters appear in the first structure table (Type 1) of SMBIOS, it will cause system information error, and this error is usually not easy to be noticed by users or developers. For another example, when unreadable characters appear in the 16th structure table (Type 16) or the 19th structure table (Type 19), it will also cause the system to fail to pass the WHQL (Windows Hardware Quality Labs testing) certification. Because these errors often do not lead to serious problems in the system immediately and make the system unable to operate, users or developers lose the opportunity to debug errors at the first time, which in turn leads to difficulties and time-consuming subsequent debugging, thus becoming an undetectable potential problem.

Therefore, the object of the present invention is to provide a warning method for SMBIOS being tampered with.

Therefore, the present invention provides a warning method, which is suitable for a computer system and implemented by the computer system. The computer system includes a BIOS, a first non-volatile memory corresponding to the BIOS, a baseboard management controller, and a second non-volatile memory corresponding to the baseboard management controller. The warning method includes steps (A) to (D).

In step (A), in the process of Power On Self-Test (POST), check whether there is a Garbled string in multiple structure tables (Type) of a system management basic input/output system (SMBIOS).

In step (B), when it is determined that the unreadable string exists in one of the structure tables, a corresponding error message is recorded in a system event log (System event log, SEL), and the system event log is stored in the second non-volatile memory.

In step (C), when the error message is recorded in the system event log, an error flag stored in the first non-volatile memory is changed from a first logic value to a second logic value.

In step (D), continue to execute a boot procedure.

In some implementation aspects, the warning method also includes a step (E) between steps (A) and (D), when it is judged that the unreadable character string does not exist in any of the structure tables, and when it is judged that the logical value of the error flag is equal to the first logical value, step (D) is executed.

In some implementations, the warning method further includes a step (F) between steps (A) and (D). When it is judged that the unreadable character string does not exist in any of the structure tables, and it is judged that the logic value of the error flag is equal to the second logic value, the error flag stored in the first non-volatile memory is changed from the second logic value to the first logic value, and then step (D) is performed.

In some implementation aspects, wherein, in step (F), after the error flag stored in the first non-volatile memory is changed from the second logic value to the first logic value, a corresponding recovery message is recorded in the system event record, and the system event record is stored in the second non-volatile memory.

In some implementation aspects, wherein, in step (A), the ASCII code corresponding to the unreadable string is any one of 0x00 to 0x1F.

In some implementation aspects, the SMBIOS is stored in the BIOS, and the SMBIOS is stored in the second volatile memory. Wherein, in step (A), the computer system checks the SMBIOS stored in the BIOS and the SMBIOS stored in the second volatile memory, and determines that the SMBIOS with the unreadable string is from the BIOS or the second volatile memory. In step (B), the error message includes a second event data (Event data 2), and the second event data indicates that the SMBIOS is from the second non-volatile memory or the BIOS.

In some implementation aspects, wherein, in step (B), the computer system also displays the error message to indicate that the SMBIOS with the unreadable string is from the second non-volatile memory or the BIOS.

In some other implementations, wherein, in step (B), the error message further includes a 3rd event data (Event data 3), when the 3rd event data is equal to 30h, 31h, 32h, 33h, or 34h, indicating that the unreadable character string occurs in a product manufacturer information, a product name information, a product module number information, a product version information, or a product serial number information in a first structure table (Type 1) in the structure tables respectively .

In some implementations, wherein, in step (B), when the 3rd event data is equal to 20h, 21h, 22h, or 23h, respectively indicate that the unreadable string occurs in a motherboard manufacturer information, a motherboard name information, a motherboard serial number information, or a motherboard material number information of a second structure table (Type 2) in the structure tables.

In some implementations, wherein, in step (B), when the third event data is equal to 10h, 11h, or 12h, it indicates that the unreadable character string occurs in a casing type information, a casing material number information, or a casing serial number information of a third structure table (Type 3) in the structure tables, respectively.

The effect of the present invention is that: when the computer system judges that the structure table has the unreadable character string in the process of self-testing at power-on, it records the corresponding error message and the error flag in the system event record stored in the non-volatile memory unit, so that the user or developer can know the occurrence of the error in real time, and then can correct and debug it as soon as possible.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same numerals.

Referring to FIG. 1 , an embodiment of the warning method of the present invention is applicable to a computer system. The computer system is, for example, a server host, and includes a chip set (PCH) 1, a central processing unit (CPU) 2 electrically connected to the chip set 1, a baseboard management controller (BMC) 3 electrically connected to the chip set 1, a basic input output system (BIOS) 4 electrically connected to the chip set 1, a first non-volatile memory 5 electrically connected to the BIOS 4, and a second non-volatile memory 6 electrically connected to the baseboard management controller 3. The BIOS 4 is, for example, a chip (Chip or IC), and stores a system management basic input/output system (SMBIOS). The first non-volatile memory 5 and the second non-volatile memory 6 are, for example, an electronically erasable rewritable read-only memory (EEPROM). The first non-volatile memory 5 also stores the SMBIOS and also stores an error flag.

When the SMBIOS stored in the basic input output system 4 needs to be updated, a user or a developer can establish a connection with the baseboard management controller 3 of the computer system through another computer host, and use a third-part tool in a remote login mode to write a newer version of the SMBIOS in the second non-volatile memory 6. After that, the newer version of the SMBIOS is synchronously updated to the basic input output system 4 again, or, through the computer system on the ground The mode utilizes another third-party software to execute the write program for the SMBIOS stored in the basic input output system 4 to update. At this time, the SMBIOS stored in the second non-volatile memory 6 or the BIOS 4 may have an error problem as in the prior art. The warning method includes steps S1-S11.

In step S1, during the booting process, the CPU 2 reads the program code of the BIOS 4 to execute a Power On Self-Test (POST) of a booting procedure.

In step S2, the central processing unit 2 checks whether there is a garbage string in the structure tables (Type) of the SMBIOS. More specifically, the CPU 2 checks the SMBIOS stored in the BIOS 4 and the SMBIOS stored in the second volatile memory 6 . The SMBIOS includes 128 structure tables in total, namely a 0th structure table (Type 0) to a 127th structure table (Type 127), and the warning method can pre-set which structure tables among all 128 structure tables, or all structure tables, to be checked. In this embodiment, the structure tables to be checked are the first structure table (Type 1), a second structure table (Type 2), and a third structure table (Type 3).

The ASCII code corresponding to the unreadable string is any one of 0x00 to 0x1F. Specifically, the 16 unreadable strings corresponding to ASCII codes 0x00~0x0F are NUL, SOH, STX, ETX, EOT, ENQ, ACK, BEL, BS, HT, LF, VT, FF, CR, SO, and S1, and the 16 unreadable strings corresponding to ASCII codes 0x10~0x1F are DLE, DC1, DC2, DC3, DC4, NAK, SYN, ETB , CAN, EM, SUB, ESC, FS, GS, RS, and US.

When the central processing unit 2 judges that the unreadable character string exists in one of the structure tables, step S3 is executed. And when the central processing unit 2 judges that the unreadable character string does not exist in any of the structure tables, step S8 is executed.

In step S3, the central processing unit 2 determines which structure table the unreadable character string is in, and which information in the structure table corresponds to it. For example, the first structure table (Type 1) includes a product manufacturer information (Product manufacturer), a product name information (Product name), a product module number information (Product part/model number), a product version information (Product version), and a product serial number information (Product serial number). The second structure table (Type 2) includes a board manufacturer information (Board manufacturer), a board name information (Board product name), a board serial number information (Board serial number), and a board part number information (Board part number). The third structure table (Type 3) includes a chassis type information (Chassis type), a chassis part number information (Chassis part number), and a chassis serial number information (Chassis serial number). Next, step S4 is executed.

In step S4, the CPU 2 judges whether the SMBIOS with the unreadable character string is from the BIOS 4. When the judgment is "No", step S5 is executed, and when the judgment is "Yes", step S6 is executed.

In step S5, the CPU 2 records a corresponding error message in a system event log (System event log, SEL), and stores the system event log in the second non-volatile memory 6 . More specifically, the system event record stores various information and can be parsed by an intelligent platform management interface (IPMI) command (Command). The format of each message is, for example, 0003h∣02h∣60742B6Bh∣01h∣00h∣04h∣2Ch∣FCh∣6Fh∣06h∣80h∣01h. Among them, 06h | 80h | 01h is sequentially a first event data (Event data 1), a second event data (Event data 2), and a third event data (Event data 3).

When the first event data included in one of the messages is equal to 06h, that one of the messages is the error message. Furthermore, the second event data included in the error message in step S5 will be equal to 81h, indicating that the SMBIOS is from the second non-volatile memory 6 corresponding to the baseboard management controller 3, that is, the SMBIOS of the basic input output system 4 is updated synchronously by a newer version in the second non-volatile memory 6. Produced. In addition, in other embodiments, in the boot process, the computer system also displays the error message on a display (not shown), and the error message indicates that the SMBIOS with the unreadable string is from the second non-volatile memory 6, to remind the user or the developer. Next, step S7 is executed.

In step S6, the CPU 2 records the corresponding error message in the system event log (SEL), and stores the system event log in the second non-volatile memory 6 . The second event data included in the error message in step S6 will be equal to 80h, to indicate that the SMBIOS is from the corresponding BIOS 4, that is, the SMBIOS of the BIOS 4 is not updated synchronously through the second non-volatile memory 6, but is directly updated to a newer version. In other embodiments, in the booting process, the computer system also displays the error message on the display (not shown), and the error message indicates that the SMBIOS with the unreadable string is from the basic input output system 4, to remind the user or the developer. Next, step S7 is executed.

In addition, in the error messages of steps S5 and S6, when the third event data included is equal to 30h, 31h, 32h, 33h, or 34h, it indicates that the unreadable character string occurs in the product manufacturer information, the product name information, the product module number information, the product version information, or the product serial number information of the first structure table (Type 1) in the structure tables respectively. And when the 3rd event data is equal to 20h, 21h, 22h, or 23h, respectively indicate that the unreadable string occurs in the motherboard manufacturer information, the motherboard name information, the motherboard serial number information, or the motherboard material number information of the second structure table in the structure tables. And when the 3rd event data is equal to 10h, 11h, or 12h, respectively indicate that the unreadable string occurs in the casing type information, the casing material number information, or the casing serial number information of the third structure table (Type 3) in the structure tables.

In step S7, the CPU 2 changes the error flag stored in the first non-volatile memory 5 from a first logic value to a second logic value. In this embodiment, the first logic value and the second logic value are respectively equal to logic 0 and logic 1, while in other embodiments, they may be equal to logic 1 and logic 0 respectively. Next, step S11 is executed.

In step S8, the CPU 2 judges whether the logic value of the error flag stored in the first non-volatile memory 5 is equal to the second logic value (ie logic 1). When the error flag is equal to the second logical value, step S9 is executed. And when the error flag is equal to the first logic value, execute step S11

In step S9, the CPU 2 changes the error flag stored in the first non-volatile memory 5 from the second logic value (ie logic 1) to the first logic value (ie logic 0). In other words, the logic value of the error flag indicates whether the unrecognizable character string occurred in the power-on self-test of the previous boot process, but in the power-on self-test of the current boot process, step S2 did not check any of the unrecognizable character strings. Therefore, the error flag that has been set to the second logic value (i.e. logic 1) is corrected to the first logic value (i.e. logic 0). Next, step S10 is executed.

In step S10 , the CPU 2 records a corresponding recovery message in the system event record, and stores the system event record in the second non-volatile memory 6 . When one of these messages stored in the system event record is the repair message, the first event data included in it is equal to 06h, and the second event data included in the repair message will be equal to 00h, to indicate that the unreadable character string has been corrected or debugged by the user or the developer. In addition, in other embodiments, during the boot process, the computer system also displays the repair message on the display (not shown) to remind the user or the developer. Next, step S11 is executed.

In step S11, the CPU 2 continues to execute the booting procedure. In this embodiment, when the version of the SMBIOS stored in the second volatile memory 6 is newer than the version of the SMBIOS stored in the BIOS 4, the SMBIOS stored in the second volatile memory 6 will be updated synchronously to the SMBIOS stored in the BIOS 4. Therefore, the warning method is implemented before the two SMBIOS stored in the second volatile memory 6 and the BIOS 4 are updated synchronously.

To sum up, when the computer system judges that the structure table has the unreadable character string during the self-testing process, it records the error message and the error flag in the system event record stored in the second non-volatile memory 6, so that the user or developer can immediately know the occurrence of the error, and then can correct and debug it as soon as possible. In addition, the repair message indicating that the repair has been completed can also be recorded in the system event record, so the object of the present invention can be indeed achieved.

But the above are only embodiments of the present invention, and should not limit the scope of the present invention. All simple equivalent changes and modifications made according to the patent scope of the present invention and the content of the patent specification are still within the scope of the patent of the present invention.

1: Chipset 2: CPU 3: Baseboard Management Controller 4: Basic input and output system 5: The first non-volatile memory 6: Second non-volatile memory S1~S11: Steps

Other features and effects of the present invention will be clearly presented in the implementation manner with reference to the drawings, wherein: Fig. 1 is a block diagram illustrating a computer system to which the warning method of the present invention is applicable; and FIG. 2 is a flowchart illustrating an embodiment of the warning method of the present invention.

S1~S11: Steps

Claims (9)

A warning method is applicable to a computer system and implemented by the computer system. The computer system includes a basic input output system, a first non-volatile memory corresponding to the basic input output system, a baseboard management controller, and a second non-volatile memory corresponding to the baseboard management controller. The warning method includes: (A) checking a system management basic input/output system (System management basic input/output) system, SMBIOS) whether there is an unreadable string (Garbled string) in a plurality of structure tables (Type), and the ASCII code corresponding to the unreadable string is any one of 0x00 to 0x1F; (B) when it is judged that the unreadable string exists in one of the structure tables, record a corresponding error message in a system event log (System event log, SEL), and store the system event log in the second non-volatile memory; When recording to the system event record, an error flag stored in the first non-volatile memory is changed from a first logic value to a second logic value; and (D) continue to execute a boot procedure. The warning method as described in claim 1, further comprising a step (E) between steps (A) and (D), when it is judged that the unreadable character string does not exist in any of the structure tables, and when it is judged that the logical value of the error flag is equal to the first logical value, step (D) is executed. The warning method as described in claim 2, further comprising a step (F) between steps (A) and (D), when it is judged that the unreadable character string does not exist in any of the structure tables, and when the logic value of the error flag is judged to be equal to the second logic value, the error flag stored in the first non-volatile memory is changed from the second logic value to the first logic value, and then step (D) is performed. The warning method as described in claim 3, wherein, in step (F), after the error flag stored in the first non-volatile memory is changed from the second logical value to the first logical value, a corresponding repair message is recorded in the system event record, and the system event record is stored in the second non-volatile memory. In the warning method described in claim 4, the SMBIOS is stored in the basic input output system, and the SMBIOS is stored in the second volatile memory. In step (A), the computer system checks the SMBIOS stored in the basic input output system and the SMBIOS stored in the second volatile memory, and determines that the SMBIOS with the unreadable string is from the basic input output system or the second volatile memory. In step (B), the error message includes a second volatile memory. Event data (Event data 2), the second event data indicates that the SMBIOS is from the second non-volatile memory or the BIOS. The warning method according to claim 5, wherein, in step (B), the computer system also displays the error message to indicate that the SMBIOS with the unreadable string is from the second non-volatile memory or the basic input output system. The warning method as described in claim item 5, wherein, in step (B), the error The error message also includes a 3rd event data (Event data 3). When the 3rd event data is equal to 30h, 31h, 32h, 33h, or 34h, it indicates that the unreadable string occurs in a product manufacturer information, a product name information, a product module number information, a product version information, or a product serial number information in a first structure table (Type 1) in the structure tables respectively. As the warning method described in claim item 7, wherein, in step (B), when the 3rd event data is equal to 20h, 21h, 22h, or 23h, respectively indicate that the unreadable character string occurs in a motherboard manufacturer information, a motherboard name information, a motherboard serial number information, or a motherboard material number information of a 2nd structure table (Type 2) in these structure tables. The warning method as described in claim item 8, wherein, in step (B), when the 3rd event data is equal to 10h, 11h, or 12h, respectively indicate that the unreadable string occurs in a casing type information, a casing material number information, or a casing serial number information of a 3rd structure table (Type 3) in these structure tables.

Images