A method, device, and medium for firmware anomaly localization based on burning logs

By constructing address chains and block sequence chains, generating write overwrite, read back return, and verification difference sets, calculating difference signatures, dividing source segments and comparison objects, and finding abnormal block boundaries, the problems of insufficient differentiation of abnormal source types and imprecise address ranges during firmware flashing are solved, achieving accurate abnormal location and tracing.

CN123114809APending Publication Date: 2026-11-17SUZHOU DAYAO TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611147370.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-30
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies lack sufficient differentiation of anomaly source types and fine-grained anomaly address ranges during firmware flashing, making it difficult to accurately determine the boundaries and forms of anomaly addresses in multi-round flashing and retesting scenarios.

Method used

By reading the firmware burning log, a write overwrite set, a read-back return set, and a verification difference set are generated. An address chain and a block sequence chain are constructed, a difference signature is calculated, source segments are divided, source records are formed, comparison objects for adjacent burning rounds are established, a retention set and a migration set are generated, the nearest normal blocks on both sides of the difference block are found, outer boundaries and inner boundaries are generated, and the type and form of the anomaly source are determined.

Benefits of technology

It improves the precision of firmware anomaly location and the completeness of burning quality traceability, and realizes the accurate location of anomaly address range and the itemized expression of anomaly source type.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123114809A_ABST
    Figure CN123114809A_ABST
Patent Text Reader

Abstract

This invention discloses a firmware anomaly localization method, device, and medium based on burning logs, relating to the field of firmware burning anomaly localization technology. The method includes: reading firmware burning logs, generating a write overlay set, a read-back return set, and a verification difference set; constructing an address chain and a block sequence chain; generating block states; extracting the verification difference set; comparing it with the write overlay set and the read-back return set; generating a write-not-overlay set, a read-back-not-return set, and a value difference set; calculating the difference signature; forming a source record; establishing comparison objects for adjacent burning rounds; generating a retention set and a migration set; generating a signature retention set; forming an evolution record; finding the nearest normal blocks on both sides of the difference block; generating an outer boundary; aggregating the difference address set; generating an inner boundary; finding the intersection to form a localization boundary; determining the anomaly source type and anomaly form; and generating a localization result, thereby improving the precision of firmware anomaly localization and the completeness of burning quality traceability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of firmware burning anomaly location technology, and in particular to a firmware anomaly location method, device and medium based on burning logs. Background Technology

[0002] During the firmware burning process of application-specific integrated circuits, embedded controllers, and memory chips, the burning device typically performs write, read, and verification according to the target address range. The firmware burning log records the data block number, target start address, target end address, write response, read response, verification result, and verification difference address. Conventional firmware burning quality analysis methods usually use the log records to complete the burning round organization, address range verification, verification difference extraction, and abnormal address prompts, providing a data foundation for firmware version verification, production line retesting, and chip storage area verification.

[0003] Conventional methods still have room for further optimization. On the one hand, in the process of conventional log analysis, the verification difference address is usually recorded as an anomaly prompt object, and the distinction between different anomaly source types such as write not overwritten anomaly, read back not returned anomaly, and value difference anomaly is relatively limited. On the other hand, in the scenario of multi-round burning and retesting, the retention, migration, and persistence of difference addresses between adjacent burning rounds usually need to be further judged by manual review, and there is still room for improvement in the granularity of the expression of anomaly address boundaries and anomaly patterns. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a firmware anomaly location method based on burning logs to solve the problems of insufficient differentiation of anomaly source types and imprecise anomaly address range in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a firmware anomaly localization method based on burning logs, comprising: reading firmware burning logs, collecting data block records, dividing burning rounds, generating a write overlay set, a read-back return set, and a verification difference set, constructing an address chain and a block sequence chain, and generating block states; extracting the verification difference set based on the address chain and block states, comparing it with the write overlay set and the read-back return set, generating a write not overlay set, a read-back not return set, and a value difference set, calculating the difference signature, dividing the source segment, and forming a source record; establishing comparison objects for adjacent burning rounds based on the source record, comparing the difference sets of the same type, generating a retention set and a migration set, comparing the difference signatures, generating a signature retention set, registering the state without round comparison, and forming an evolution record; and finding the nearest normal blocks on both sides of the difference block based on the block sequence chain, block states, source records, and evolution records, generating an outer boundary, collecting the difference address set, generating an inner boundary, finding the intersection to form a localization boundary, determining the anomaly source type and anomaly form, and generating a localization result.

[0007] As a preferred embodiment of the firmware anomaly localization method based on burning logs described in this invention, the construction of the address chain and block sequence chain, and the generation of block status, include: reading the firmware burning log, dividing write records, read-back records, and verification records according to the burning stage name, registering write completion markers and read-back completion markers, and jointly aggregating them according to the data block number, target start address, and target end address to generate data block records; dividing the burning rounds of the data block records according to the burning round number, and generating write coverage sets, read-back return sets, and verification difference sets according to the target address range, write completion markers, read-back completion markers, and verification results; binding the write coverage sets, read-back return sets, and verification difference sets under the same ASIC number, firmware version, data block number, target address range, and burning round to generate an address chain, generating a block sequence chain according to the target address order, and generating block status according to the record integrity and verification difference set.

[0008] As a preferred embodiment of the firmware anomaly localization method based on burning logs described in this invention, the generation of the write-overwritten set, the read-back-not-returned set, and the value difference set includes: filtering difference blocks based on address chains and block states; extracting address chain units according to data block numbers, target address ranges, and burning rounds; reading the write-overwritten set, the read-back-returned set, and the verification difference set from the address chain units; performing address attribution verification on the verification difference set; retaining the verification difference addresses located between the target start address and the target end address; generating the verification difference set after attribution verification; comparing the verification difference addresses in the verification difference set after attribution verification with the write-overwritten set and the read-back-returned set in sequence; performing anomaly source type classification of the verification difference addresses; and generating the write-overwritten set, the read-back-not-returned set, and the value difference set.

[0009] As a preferred embodiment of the firmware anomaly location method based on burning logs described in this invention, the calculation of the difference signature, the division of the source segment, and the formation of the source record include: reading the expected verification value and the read-back verification value corresponding to the value difference set, performing a bitwise XOR operation to generate the difference signature, dividing the write-uncovered set, the read-back-not-returned set, and the value difference set according to the continuous relationship of the target address, generating the source segment, and forming the source record.

[0010] As a preferred embodiment of the firmware anomaly localization method based on burning logs described in this invention, the step of establishing comparison objects for adjacent burning rounds based on source records, comparing similar difference sets, and generating a retention set and a migration set includes: grouping source records according to ASIC number, firmware version, data block number, target start address, and target end address to generate round grouping groups; arranging source records according to burning rounds within the round grouping groups; pairing adjacent source records to form comparison objects for adjacent burning rounds; and clarifying the source records of the previous burning round and the subsequent burning rounds. One programming cycle source record; based on the comparison objects of adjacent programming cycles, extract the write-uncovered set, read-back-not-returned set, and value difference set from the source record of the previous programming cycle and the source record of the next programming cycle respectively. Perform an intersection operation on the write-uncovered set of the previous programming cycle and the write-uncovered set of the next programming cycle to generate the write-uncovered retention set. Register the non-common addresses in the write-uncovered sets of the two programming cycles as the write-uncovered migration set. Process the read-back-not-returned set and the value difference set in the same way to generate the retention set and the migration set.

[0011] As a preferred embodiment of the firmware anomaly localization method based on burning logs described in this invention, the steps of comparing differential signatures, generating a signature retention set, registering a no-round comparison state, and forming an evolution record include: for the value difference set in the comparison objects of adjacent burning rounds, extracting the common target address in the value difference set of the previous burning round and the value difference set of the next burning round, reading the differential signature of the previous burning round and the differential signature of the next burning round corresponding to the common target address, comparing the differential signatures of the common target address in the two burning rounds one by one, registering the common target address with consistent differential signatures as a signature retention set, and registering a no-round comparison state when the round aggregation group cannot form an adjacent burning round comparison object; binding the round aggregation group, adjacent burning round comparison objects, retention set, migration set, signature retention set, and no-round comparison state according to the anomaly source type to generate an evolution record.

[0012] As a preferred embodiment of the firmware anomaly localization method based on burning logs described in this invention, the step of finding the nearest normal blocks on both sides of the difference block based on the block sequence chain, block status, source record, and evolution record to generate the outer boundary includes: reading the block sequence chain and block status, taking the difference block as the localization object, searching for the nearest normal block on the front side and the nearest normal block on the back side along the previous neighbor block and the next neighbor block to generate the nearest normal block record; and determining the starting point of the outer boundary, the ending point of the outer boundary, and the set of data blocks within the outer boundary based on the nearest normal block record and the block sequence chain to generate the outer boundary.

[0013] As a preferred embodiment of the firmware anomaly localization method based on burning logs described in this invention, the steps of collecting the set of difference addresses, generating an inner boundary, finding the intersection to form a localization boundary, determining the anomaly source type and anomaly form, and generating a localization result include: based on the outer boundary and the set of data blocks within the outer boundary, filtering data blocks with a difference block status within the outer boundary, retrieving the source records corresponding to the difference blocks, collecting the write-uncovered set, read-back-not-returned set, and value difference set for each burning round, merging them to generate a set of difference addresses, extracting the inner boundary start point and inner boundary end point from the set of difference addresses according to the target address order, and generating an inner boundary; reading the outer boundary and inner boundary, performing an address range intersection operation to generate a localization boundary, dividing the localization unit and determining the anomaly source type based on the source segments and difference signatures within the localization boundary according to the write-uncovered source segments, read-back-not-returned source segments, and value difference source segments according to the source segments and difference signatures within the localization boundary, according to the write-uncovered source segments, read-back-not-returned source segments, and value difference source segments according to the source segments and difference signatures within the localization boundary, according to the evolution records within the localization boundary, registering the anomaly form according to the retained set, migration set, signature retained set, and no round comparison state, binding the localization boundary, anomaly source type, anomaly form, source record, evolution record, and missing block information, and generating a localization result.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements any step of the firmware anomaly location method based on burning logs as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the firmware anomaly location method based on burning logs as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By reading the address chain and block state, the verification difference set is extracted and compared with the write-overlay set and read-back return set to generate the write-uncovered set, read-back-not-return set, and value difference set. Combined with the difference signature and source segment, a source record is formed, realizing the itemized expression of the abnormal source type of the verification difference address, which facilitates firmware data review; based on the block sequence chain, block state, source record, and evolution record, outer boundary, inner boundary, and positioning boundary are generated, and the abnormal source type and abnormal form are registered. The positioning result simultaneously reflects the address range, abnormal source type, and round evolution relationship, improving the precision of firmware abnormal positioning and the integrity of burning quality traceability. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a firmware anomaly localization method based on burning logs.

[0019] Figure 2 A flowchart for generating source records.

[0020] Figure 3 A flowchart for obtaining evolutionary records.

[0021] Figure 4 A flowchart for generating the localization results. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4This is one embodiment of the present invention, which provides a firmware anomaly location method based on burning logs, including the following steps: S1. Read firmware burning logs, collect data block records, divide burning rounds, generate write overwrite set, read back return set and verification difference set, construct address chain and block order chain, and generate block status.

[0026] Read the firmware burning log, divide it into write records, read-back records and verification records according to the burning stage name, register the write completion mark and read-back completion mark, and combine them according to the data block number, target start address and target end address to generate data block records.

[0027] Furthermore, after reading the firmware burning log, the log records are divided into write records, read-back records, and verification records according to the burning stage name.

[0028] The write record stores the data block number, target start address, target end address, write start address, write end address, and write response; the readback record stores the data block number, target start address, target end address, readback start address, readback end address, and readback response; the verification record stores the data block number, target start address, target end address, verification result, verification difference address, verification expectation value, and verification readback value.

[0029] Furthermore, the write response is registered to indicate completion status. Specifically, when the write response indicates completion, a write completion flag is generated; when the write response does not indicate completion or the write record is missing, a write incomplete flag is generated.

[0030] Furthermore, the readback response is registered to indicate completion status. Specifically, when the readback response indicates completion, a readback completion flag is generated; when the readback response does not indicate completion or the readback record is missing, a readback incomplete flag is generated.

[0031] Furthermore, the write records, readback records, and verification records are read and aggregated according to the data block number, target start address, and target end address to generate data block records.

[0032] Each data block record includes a data block number, target start address, target end address, write record, read-back record, and verification record.

[0033] Specifically, during the aggregation process, the data block number is matched first, followed by the target start address and target end address.

[0034] It should be noted that only write records, read-back records, and verification records with the same data block number, target start address, and target end address are classified into the same data block record.

[0035] It should be noted that if a data block record is missing a write record, read back record, or verification record, the data block record will not be deleted. The missing information will be retained to prevent the evidence for anomaly location from being discarded prematurely due to missing log entries.

[0036] The data block records are divided into burning rounds according to the burning round number. Based on the target address range, write completion flag, read back completion flag, and verification result, a write overwrite set, a read back return set, and a verification difference set are generated.

[0037] Bind the write overlay set, read back return set, and verification difference set under the same application-specific integrated circuit number, firmware version, data block number, target address range, and programming round to generate an address chain. Generate a block sequence chain according to the target address order. Generate the block status based on the record integrity and verification difference set.

[0038] Furthermore, read data block records, and within the same data block number and the same target address range, arrange the written records, read-back records, and verification records according to log time.

[0039] Specifically, when recording the burning round number in the firmware burning log, write records, read records, and verification records under the same burning round number, the same data block number, the same target start address, and the same target end address are grouped into the same burning round.

[0040] When the firmware burning log does not record the burning round number, read the write records, read back records, and verification records under the same data block number, the same target start address, and the same target end address, and arrange them according to the log time; take the log time of the verification record as the end position of the burning round, and classify the write records, read back records, and the current verification record between the end position of the previous burning round and the current verification record log time into the same burning round.

[0041] When multiple write records exist within the same programming cycle, these multiple write records together generate the write overlay set for the same programming cycle; when multiple readback records exist within the same programming cycle, these multiple readback records together generate the readback return set for the same programming cycle; when write records, readback records, and verification records appear alternately according to log time, the write records, readback records, and verification records located before the end position of the same programming cycle are classified into the same programming cycle.

[0042] It should be noted that if there are still write or read records after the last verification log time, the write or read records after the last verification log time will be registered for the next burning round, and the situation of missing verification records will be retained.

[0043] When there is no verification record in the firmware burning log, write records and read-back records under the same data block number, the same target start address, and the same target end address are grouped into one burning round according to the log time, and the missing verification record is retained.

[0044] It should be noted that if a certain burning cycle lacks write records, read records, or verification records, the burning cycle will not be deleted, and the missing information will be retained.

[0045] Furthermore, the target start address, target end address, write start address, write end address, and write completion marker are read from the same data block and the same burning cycle.

[0046] When the write completion mark is set, the target addresses that are located between the target start address and the target end address, and simultaneously between the write start address and the write end address, are registered as the write overwrite set. The write overwrite set represents the target addresses that the write record confirms overwrite.

[0047] When a write record is missing or the write completion is marked as incomplete, the corresponding target address is not registered in the write overwrite set, and the write overwrite set is registered as an empty set.

[0048] Furthermore, the target start address, target end address, readback start address, readback end address, and readback completion flag are read from the same data block and the same burning cycle.

[0049] When the readback is marked as complete, the target address that is located between the target start address and the target end address, and simultaneously between the readback start address and the readback end address, is registered as the readback return set. The readback return set is the target address for the confirmed return of the readback record.

[0050] When a readback record is missing or the readback completion is marked as incomplete, the corresponding target address is not registered in the readback return set, and the readback return set is registered as an empty set.

[0051] Furthermore, the target start address, target end address, verification difference address, verification expected value, and verification readback value are read from the same data block and the same burning round.

[0052] When a verification record exists, and a certain verification difference address is located between the target start address and the target end address, and the verification expected value is inconsistent with the verification read-back value, the corresponding verification difference address is registered as a verification difference set. The verification difference set represents the target address where the verification expected value and the verification read-back value are inconsistent.

[0053] When a verification record exists and the verification result is no difference, the set of verification differences is registered as an empty set.

[0054] When a verification record is missing, a verification difference set is not generated, but the missing verification record information is retained.

[0055] Furthermore, the write overlay set, read back return set, and verification difference set under the same application-specific integrated circuit number, same firmware version, same data block number, same target address range, and same burning round are read, and the three sets are bound with the verification expectation value, verification read back value, and log time to generate an address chain.

[0056] It should be noted that the address chain unit in the address chain includes the application-specific integrated circuit number, firmware version, data block number, target start address, target end address, programming round, write overwrite set, read-back return set, verification difference set, verification expected value, verification read-back value, and log time.

[0057] Furthermore, the data block number, target start address, and target end address are read from the data block record, and the data block records are arranged from front to back according to the target start address; when multiple data blocks have the same target start address, they are arranged from front to back according to the target end address.

[0058] After sorting, register the preceding and following neighbor blocks for each data block; the preceding neighbor block is the data block that is located one position before the current data block in the address order; the following neighbor block is the data block that is located one position after the current data block in the address order.

[0059] The block order chain includes the data block number, the target start address, the target end address, the previous neighboring block, the next neighboring block, and the address order.

[0060] Furthermore, the address chain and block sequence chain are read, and the block state is generated according to the write record, read-back record, verification record, and verification difference set of each data block in each burning round.

[0061] Specifically, when the write record, read-back record, and verification record all exist, and the verification difference set is empty, it is registered as a normal block.

[0062] When the written record, read-back record, and verification record all exist, and the verification difference set is not empty, it is registered as a difference block.

[0063] When there are missing records in the write, read, or verification processes, they are registered as missing blocks.

[0064] The block status includes the data block number, target address range, programming round, normal block marker, difference block marker, missing block marker, and corresponding address chain unit.

[0065] It should be noted that when generating block status, missing blocks and difference blocks are registered separately; a missing block indicates that the address chain unit is incomplete; a difference block indicates that a complete write record, read-back record and verification record have been formed, and a set of verification differences exists.

[0066] S2. Based on the address chain and block state, extract the verification difference set, compare it with the write-overwritten set and the read-back return set, generate the write-not-overwritten set, the read-back-not-return set and the value difference set, calculate the difference signature, divide the source segment, and form the source record.

[0067] Based on the address chain and block status, the differential blocks are filtered, and the address chain units are extracted according to the data block number, target address range and burning round. The write overwrite set, read back return set and verification difference set in the address chain unit are read. The address ownership verification is performed on the verification difference set, and the verification difference addresses located between the target start address and the target end address are retained to generate the verification difference set after ownership verification.

[0068] After attribution verification, the check difference addresses in the check difference set are compared with the write overwrite set and the read back return set in turn. The check difference address anomaly source type is classified to generate the write not overwrite set, the read back not return set, and the value difference set.

[0069] Furthermore, the address chain and block status are read, and the difference blocks are filtered from the block status.

[0070] Furthermore, based on the application-specific integrated circuit number, firmware version, data block number, target address range, and programming round, the address chain unit corresponding to the difference block is extracted from the address chain.

[0071] The address chain unit includes the write overwrite set, read back return set, check difference set, check expected value, check read back value, target start address, target end address, and log time.

[0072] Furthermore, the check difference set in the address chain unit corresponding to the difference block is read, and the check difference address in the check difference set is compared with the target start address and the target end address for attribution verification.

[0073] Specifically, the check difference addresses located between the target start address and the target end address continue to participate in the anomaly source type classification; check difference addresses located outside the target start address and the target end address are not written into the source record.

[0074] Furthermore, the write overwrite set and readback return set in the same address chain unit are read. For the same data block and the same burning round, the verification difference set, write overwrite set and readback return set are divided into sets to generate the write not overwritten set, readback not returned set and value difference set.

[0075] Specifically, it determines whether the verification difference address belongs to the write overwrite set, and registers the verification difference addresses that do not belong to the write overwrite set to the write unoverwrite set.

[0076] If the verification difference address belongs to the write overwrite set, then determine whether the verification difference address belongs to the read back return set. If the verification difference address does not belong to the read back return set, register it to the read back non-return set.

[0077] Register the verification difference address that belongs to both the write-overwrite set and the read-back return set to the value difference set.

[0078] It should be noted that the write-not-overwritten set represents the check difference address that is not covered by the write-overwritten set; the read-back not-returned set represents the check difference address that has been covered by the write-overwritten set but has not been returned by the read-back return set; and the value difference set represents the target address that belongs to both the write-overwritten set and the read-back return set but still has a check difference.

[0079] Specifically, the sets of items written to but not covered, read back but not returned, and sets of values ​​differing are represented as follows: ; ; ; in, Indicates the first The data block number The set of writes not overwritten in each programming cycle Indicates the first The data block number A set of verification differences from each programming round. Indicates the first The data block number The write overwrite set for each programming cycle Indicates the first The data block number The set of read-backs that were not returned in each programming cycle. Indicates the first The data block number The set returned by each burning cycle. Indicates the first The data block number The set of value differences for each burning cycle Indicates the data block number. Indicates the programming cycle number. This represents the set difference operation. This represents the intersection operation of sets.

[0080] Read the expected value and read-back value corresponding to the value difference set, perform a bitwise XOR operation to generate a difference signature, divide the write-uncovered set, read-back-not-returned set and value difference set according to the continuous relationship of the target address, generate the source segment, and form the source record.

[0081] Furthermore, the target address in the value difference set is read, and the expected verification value and the verification readback value corresponding to the target address are read from the address chain unit.

[0082] For each target address in the value difference set, perform a bitwise XOR operation to generate a difference signature.

[0083] Specifically, the difference signature is represented as: ; in, Indicates the first The data block number Target address in each burning round Difference signature, Indicates the target address. Indicates the first The data block number Target address in each burning round The expected value of the verification Indicates the first The data block number Target address in each burning round The verification readback value, This indicates a bitwise XOR operation.

[0084] It should be noted that the difference signature represents the bit difference between the expected verification value and the read-back value, and the bit width of the difference signature is the same as the bit width of the expected verification value and the read-back value.

[0085] It should be noted that the value difference set can only indicate that there is a data value inconsistency at the target address, while the difference signature can record the bit form of the data value inconsistency.

[0086] Furthermore, the sets of data that are not overwritten, not returned, and have different values ​​are read and written, and sorted from front to back according to the target address. Each set is then divided into source segments according to the continuity of the target address.

[0087] Specifically, consecutive target addresses are registered as the same source segment. When a target address is interrupted, the current source segment ends, and a new source segment is registered starting from the next target address.

[0088] The source segment includes the source segment type, source segment start address, source segment end address, data block number, and programming round. Source segment types include write-not-overwritten source segments, read-back-not-return source segments, and value-difference source segments. Write-not-overwritten source segments originate from the write-not-overwritten set, read-back-not-return source segments originate from the read-back-not-return set, and value-difference source segments originate from the value-difference set.

[0089] It should be noted that the target address continuity relationship is directly determined by the target address in the address chain, which can objectively reflect the distribution of differential addresses in the target memory area of ​​the application-specific integrated circuit.

[0090] It should be noted that if a certain type of collection is empty, the corresponding source segment will not be generated, and the empty collection field will be retained in the source record.

[0091] Furthermore, by reading and writing the overwritten set, reading back the unreturned set, the value difference set, the difference signature, and the source segment, data under the same ASIC number, firmware version, data block number, target address range, and programming round are bound to form a source record.

[0092] The source record includes the application-specific integrated circuit number, firmware version, data block number, target start address, target end address, programming round, set of write-not-overwritten items, set of read-back-not-returned items, set of value differences, difference signature, source segment, and corresponding address chain unit.

[0093] It should be noted that if the same difference block exists in multiple burning rounds, multiple source records will be generated respectively, and each source record corresponds to one burning round.

[0094] S3. Based on the source record, establish comparison objects for adjacent burning rounds, compare the sets of differences of the same type, generate a retention set and a migration set, compare the difference signatures, generate a signature retention set, register the state without round comparison, and form an evolution record.

[0095] Source records are aggregated according to the application-specific integrated circuit number, firmware version, data block number, target start address, and target end address. Round aggregation groups are generated, and source records are arranged according to the burning round within the round aggregation group. Adjacent source records are paired to form comparison objects for adjacent burning rounds, thus clarifying the source records of the previous burning round and the source records of the next burning round.

[0096] Based on the comparison objects of adjacent programming rounds, the write-uncovered set, read-back-not-returned set, and value difference set are extracted from the source records of the previous programming round and the source records of the next programming round, respectively. The write-uncovered set of the previous programming round and the write-uncovered set of the next programming round are intersected to generate the write-uncovered retention set. The non-common addresses in the write-uncovered sets of the two programming rounds are registered as the write-uncovered migration set. The read-back-not-returned set and the value difference set are processed in the same way to generate the retention set and the migration set.

[0097] Furthermore, the source records are read and aggregated according to the ASIC number, firmware version, data block number, target start address, and target end address to generate round aggregation groups.

[0098] Each round of aggregation group corresponds to all source records formed under the same application-specific integrated circuit, the same firmware version, the same data block, and the same target address range.

[0099] Furthermore, the source records in the round collection group are read, and the source records are arranged from front to back according to the burning round. Adjacent source records are used as comparison objects for adjacent burning rounds. Specifically, the source record of the first burning round and the source record of the second burning round establish an adjacent burning round comparison object, the source record of the second burning round and the source record of the third burning round establish the next adjacent burning round comparison object, and so on. The adjacent burning round comparison object includes the source record of the previous burning round and the source record of the next burning round.

[0100] Furthermore, the source records of the previous and next burning cycles in the comparison objects of adjacent burning cycles are read, and the sets of write-not-overwritten, read-back-not-returned, and value-difference in the source records of the previous burning cycle are extracted respectively. The sets of write-not-overwritten, read-back-not-returned, and value-difference in the source records of the next burning cycle are extracted respectively.

[0101] It should be noted that comparisons are limited to sets of differences of the same type. Specifically, the write-not-overwritten set is compared only with the write-not-overwritten set, the read-back-not-returned set is compared only with the read-back-not-returned set, and the value difference set is compared only with the value difference set; no cross-comparison is performed between different exception source types.

[0102] Furthermore, for each adjacent burning round comparison object, the write-uncovered set, the read-back-not-returned set, and the value difference set are compared using the same type of difference set to generate the retain set and the migration set.

[0103] Specifically, the write-uncovered set from the previous programming round is first intersected with the write-uncovered set from the next programming round to obtain the write-uncovered retention set; then, the non-common addresses in the write-uncovered sets from the two programming rounds are registered as the write-uncovered migration set.

[0104] Furthermore, the readback non-return set and the value difference set are processed in the same way to generate the readback non-return retain set, the readback non-return migration set, the value difference retain set, and the value difference migration set, respectively.

[0105] Specifically, the retention set and the migration set are represented as follows: ; ; in, Indicates the first The data block number The set of records retained between each programming cycle and the next programming cycle. Indicates the first The data block number The migration set between each programming cycle and the next programming cycle. Indicates the first The data block number The set of differences of the same type for each burning round, and one of the following sets: the set of overwritten items in a single write operation, the set of items not returned during readback, and the set of value differences. Indicates the first The set of similar differences in the next burning round after each data block, and the single substitution type. Maintain consistency. This indicates the set union operation.

[0106] It should be noted that the retention set represents the same type of difference address that persists in adjacent burning rounds, indicating that the same type of difference address exists in both adjacent burning rounds; the migration set represents the same type of difference address that disappears or is added in adjacent burning rounds, indicating that the same type of difference address exists only in one of the two adjacent burning rounds.

[0107] For the set of value differences in the comparison objects of adjacent burning rounds, extract the common target address in the set of value differences of the previous burning round and the set of value differences of the next burning round. Read the difference signature of the previous burning round and the difference signature of the next burning round corresponding to the common target address. Compare the difference signatures of the common target address in the two burning rounds one by one. Register the common target addresses with the same difference signature as the signature retention set. When the round collection group cannot form a comparison object for adjacent burning rounds, register the state of no round comparison.

[0108] The round collection group, adjacent burning round comparison objects, retention set, migration set, signature retention set, and no round comparison state are bound to each item according to the anomaly source type to generate an evolution record.

[0109] Furthermore, the difference signatures corresponding to the value difference sets in the comparison objects of adjacent burning rounds are read, and the target addresses that exist in both the value difference sets of the previous burning round and the value difference set of the next burning round are extracted.

[0110] For a target address that exists in both sets of value differences, compare the difference signature from the previous burning round with the difference signature from the next burning round; register the target addresses with matching difference signatures in the signature retention set.

[0111] Specifically, the signature retention set is represented as: ; in, Indicates the first The data block number The signature retention set between each programming round and the next programming round Indicates the first The data block number The set of value differences for each burning cycle Indicates the first The set of value differences in the next burning round after each data block Indicates the first The data block number Target address in each burning round Difference signature, Indicates the first The target address in the next programming round after each data block The difference signature.

[0112] It should be noted that the signature retention set not only indicates that the target address continues to show value differences in adjacent burning rounds, but also that the bit difference pattern between the verification expected value and the verification read-back value remains consistent in adjacent burning rounds.

[0113] If the target address exists in two adjacent value difference sets but the difference signatures are inconsistent, the target address is not registered in the signature retention set; if two adjacent value difference sets do not have a common target address, the signature retention set is registered as an empty set.

[0114] Furthermore, the cycle collection group is read. If there is only one source record of a burning cycle in a certain cycle collection group, and it is impossible to establish a comparison object for adjacent burning cycles, then the state of no cycle comparison is registered. The state of no cycle comparison indicates that the firmware burning log has not formed a comparison condition for adjacent burning cycles.

[0115] Specifically, if there are source records from multiple burning rounds within a certain round of aggregation group, a retention set, a migration set, and a signature retention set are generated for each adjacent burning round comparison object, and the state of no round comparison is not registered.

[0116] Furthermore, the round collection group, the comparison objects of adjacent burning rounds, the retention set, the migration set, the signature retention set, and the no-round comparison state are read, and the comparison results under the same application-specific integrated circuit number, the same firmware version, the same data block number, and the same target address range are bound to form an evolution record.

[0117] The evolution record includes the application-specific integrated circuit number, firmware version, data block number, target start address, target end address, programming round relationship, anomaly source type, reserved set, migration set, signature reserved set, no round comparison status, and corresponding source record.

[0118] The evolution records are registered according to the type of anomaly source. The types of anomaly source include write-not-overwritten anomaly, read-back-not-returned anomaly, and value difference anomaly. Write-not-overwritten anomaly corresponds to the write-not-overwritten set, and the write-not-overwritten retained set and write-not-overwritten migration set are registered. Read-back-not-returned anomaly corresponds to the read-back-not-returned set, and the read-back-not-returned retained set and read-back-not-returned migration set are registered. Value difference anomaly corresponds to the value difference set, and the value difference retained set, value difference migration set, and signature retained set are registered.

[0119] S4. Based on the block sequence chain, block state, source record, and evolution record, find the nearest normal blocks on both sides of the difference block, generate the outer boundary, collect the difference address set, generate the inner boundary, find the intersection to form the location boundary, determine the anomaly source type and anomaly form, and generate the location result.

[0120] Read the block sequence chain and block status, take the difference block as the location object, search for the nearest normal block before and after the previous and next neighbor blocks, and generate the nearest normal block record.

[0121] Based on the most recent normal block record and block sequence chain, the starting point of the outer boundary, the ending point of the outer boundary, and the set of data blocks within the outer boundary are determined, and the outer boundary is generated.

[0122] Furthermore, the block sequence chain and block status are read; the block status includes normal blocks, difference blocks, missing blocks, and corresponding address chain units.

[0123] Read source records, which include the write-not-overwritten set, the read-back-not-returned set, the value difference set, the difference signature, and the source segment; wherein, the source segment includes the write-not-overwritten source segment, the read-back-not-returned source segment, and the value difference source segment.

[0124] Read the evolution record, which includes the retention set, migration set, signature retention set, anomaly source type, burning round relationship, and no round comparison state.

[0125] Use the difference blocks in the block status as the location objects; normal blocks are used to generate the outer boundary, and missing item blocks are used to indicate missing items in the log in the location results.

[0126] Furthermore, for each difference block, the address order in the block sequence chain is read; data blocks are read one by one along the direction of the previous neighbor block, and the first normal block encountered is registered as the nearest normal block in front.

[0127] It should be noted that if a normal block is not encountered after reading along the direction of the previous neighbor block to the start of the block sequence chain, the nearest normal block to the front is registered as empty, and the start of the block sequence chain is used as the basis for the open boundary to the front.

[0128] Furthermore, data blocks are read one by one along the direction of the next adjacent block, and the first normal block encountered is registered as the nearest normal block behind it.

[0129] It should be noted that if a normal block is not encountered after reading along the direction of the next adjacent block to the end of the block sequence chain, the nearest normal block behind it is registered as empty, and the end of the block sequence chain is used as the basis for the open boundary behind it.

[0130] During the search process, if a difference block is encountered, the search continues; if a missing block is encountered, the search also continues. At the same time, the data block number and target address range of the missing block are recorded.

[0131] It should be noted that missing entries blocks are entered into the location results as missing log entries, without truncating the outer boundary; when a missing entry block corresponds to a missing write record, a missing read record, or a missing check record, the missing write record information, missing read record information, or missing check record information are respectively registered in the location results.

[0132] Furthermore, the nearest normal block before and after the block are read, and the outer boundary is generated according to the address order in the block order chain.

[0133] Specifically, when the nearest normal block exists on the current side, the target start address of the first data block in the block sequence chain that is located after the nearest normal block on the previous side is taken as the starting point of the outer boundary; when the nearest normal block on the current side is empty, the target start address corresponding to the starting point of the block sequence chain is taken as the starting point of the outer boundary.

[0134] When the nearest normal block exists, the target end address of the last data block in the block sequence chain that precedes the nearest normal block is taken as the outer boundary endpoint; when the nearest normal block is empty, the target end address corresponding to the end of the block sequence chain is taken as the outer boundary endpoint.

[0135] The outer boundary includes the location object number, the outer boundary start point, the outer boundary end point, the nearest normal block on the front side, the nearest normal block on the back side, and the set of data blocks within the outer boundary; the set of data blocks within the outer boundary includes difference blocks and missing blocks located between the outer boundary start point and the outer boundary end point, but does not include the nearest normal block on the front side and the nearest normal block on the back side.

[0136] Based on the outer boundary and the set of data blocks within the outer boundary, filter the data blocks with the status of difference blocks within the outer boundary, retrieve the source records corresponding to the difference blocks, and collect the write-not-overwritten set, read-back-not-returned set, and value difference set of each burning round. Merge them to generate a difference address set, and extract the inner boundary start point and inner boundary end point from the difference address set according to the target address order to generate the inner boundary.

[0137] Furthermore, the source records corresponding to all difference blocks within the outer boundary are read, and the difference address set is generated based on the difference blocks within the outer boundary, the write-uncovered set of each burning round, the read-back-unreturned set, and the value difference set.

[0138] Specifically, the set of differing addresses is represented as: ; in, No. Each difference block corresponds to a set of difference addresses within its outer boundary. Indicates the location object number. Indicates the first Each difference block corresponds to a set of difference blocks within the outer boundary. Indicates the first The set of burning cycles in which each data block exists in the source record. Indicates the first The data block number The set of writes not overwritten in each programming cycle Indicates the first The data block number The set of read-backs that were not returned in each programming cycle. Indicates the first The data block number The set of value differences for each burning cycle.

[0139] It should be noted that the difference address set represents the target addresses within the outer boundary that actually have verification differences, and is a subset of the target address set within the outer boundary; missing blocks do not have complete source records, and the target addresses of missing blocks are not included in the difference address set.

[0140] Furthermore, the set of differencing addresses is read and arranged from front to back according to the target address. The first target address in the set of differencing addresses is used as the starting point of the inner boundary, and the last target address in the set of differencing addresses is used as the ending point of the inner boundary to generate the inner boundary. The inner boundary includes the starting point, the ending point, and the corresponding set of differencing addresses.

[0141] It should be noted that when the differential address set is empty, no inner boundary is generated, no intersection operation of the address ranges of the outer and inner boundaries is performed, the outer boundary, block status, and source record missing conditions are preserved, a location result that cannot form an inner boundary is generated, and the abnormal address range is not expanded.

[0142] Furthermore, when the set of differing addresses is not empty and an inner boundary is generated, the outer and inner boundaries are read, the address range intersection operation is performed, and a location boundary is generated. Based on the source segments and differing signatures within the location boundary, the location unit is divided according to the source segments written without being covered, the source segments read back without being returned, and the source segments with differing values, and the anomaly source type is determined. According to the evolution record within the location boundary, the anomaly form is registered according to the retention set, migration set, signature retention set, and no round comparison state. The location boundary, anomaly source type, anomaly form, source record, evolution record, and missing block information are bound together to generate the location result.

[0143] Furthermore, when the set of different addresses is not empty and an inner boundary is generated, the outer boundary and the inner boundary are read, and the intersection of the outer boundary and the inner boundary is calculated to form the positioning boundary.

[0144] Specifically, the positioning boundary is represented as: ; in, No. The positioning boundaries corresponding to each difference block. Indicates the first The outer boundary corresponding to each difference block Indicates the first The inner boundaries corresponding to each difference block Indicates the location object number.

[0145] It should be noted that the location boundary is the target address range used for anomaly location, which is the address sub-interval of the target storage area of ​​the application-specific integrated circuit. The location boundary is obtained by the intersection of the outer boundary and the inner boundary, and is jointly limited by the set of data blocks and the set of differential addresses within the outer boundary.

[0146] Furthermore, the source segments within the positioning boundary are read, and positioning units are generated according to the source segment type and the continuity of the target address.

[0147] It should be noted that different source segment types cannot be merged into the same location unit; source segments of the same type with non-contiguous addresses should generate location units separately.

[0148] Furthermore, the anomaly source type for each positioning unit is determined. Specifically, when a positioning unit writes to an uncovered source segment, an "overwritten" anomaly is registered; when a positioning unit reads back a source segment that has not been returned, a "readback" anomaly is registered; and when a positioning unit corresponds to a value difference source segment, a "value difference" anomaly is registered.

[0149] Furthermore, the evolution records within the positioning boundary are read, including the retention set, migration set, signature retention set, and round-free comparison state.

[0150] It should be noted that when there is an address intersection between the location unit and the retention set, the retention difference is recorded; when there is an address intersection between the location unit and the migration set, the migration difference is recorded; when there is an address intersection between the location unit and the signature retention set, the signature retention difference is recorded; when the location unit is only associated with a state without round comparison and there is no retention set, migration set, or signature retention set, a single round difference is recorded.

[0151] By separating the source and the form, it is possible to avoid mixing write-not-overwritten exceptions, read-back-not-return exceptions, and value difference exceptions into a single conclusion.

[0152] Furthermore, the system reads the positioning boundary, positioning unit, anomaly source type, and anomaly form, backfills the source record, evolution record, block status, and missing block information, and generates positioning results. The positioning results are used for firmware data verification, burning process review, application-specific integrated circuit target storage area verification, and mass production burning quality traceability.

[0153] The location results include the application-specific integrated circuit number, firmware version, data block number, abnormal start address, abnormal end address, abnormal source type, abnormal morphology, source segment, difference signature, the nearest normal block before and after, outer boundary, inner boundary, location boundary, burning cycle relationship, missing block information, corresponding source record, and corresponding evolution record.

[0154] The exception start address is taken from the start address of the positioning unit, and the exception end address is taken from the end address of the positioning unit; the exception source type is registered as write not overwritten exception, read back not returned exception, or value difference exception; the exception form is registered as retained difference, migration difference, signature retained difference, or single round difference.

[0155] It should be noted that the difference signature is only registered when the exception source type is a value difference exception. When the exception source type is a write not overwritten exception or a read back not returned exception, the difference signature field is registered as empty.

[0156] It should be noted that when multiple positioning units exist within the same positioning boundary, multiple positioning results are generated respectively. Each positioning result corresponds to a positioning unit with a continuous address and has an independent anomaly source type and anomaly pattern.

[0157] This embodiment also provides a computer device applicable to the firmware anomaly location method based on burning logs, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the firmware anomaly location method based on burning logs as proposed in the above embodiment.

[0158] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0159] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the firmware anomaly location method based on burning logs as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0160] In summary, this invention extracts the verification difference set by reading the address chain and block state, and compares it with the write-overlay set and read-back return set to generate the write-uncovered set, read-back-not-return set, and value difference set. Combining the difference signature and source segment to form the source record, it realizes the itemized expression of the abnormal source type of the verification difference address, which facilitates firmware data verification. Based on the block sequence chain, block state, source record, and evolution record, it generates outer boundary, inner boundary, and positioning boundary, registers the abnormal source type and abnormal form, and the positioning result simultaneously reflects the address range, abnormal source type, and round evolution relationship, improving the precision of firmware abnormal positioning and the integrity of burning quality traceability.

[0161] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A firmware anomaly localization method based on burning logs, characterized in that, include: Read firmware burning logs, collect data block records, divide burning rounds, generate write overwrite set, read back return set and verification difference set, construct address chain and block sequence chain, and generate block status; Based on the address chain and block status, extract the verification difference set, compare it with the write-overwritten set and the read-back returned set, generate the write-not-overwritten set, the read-back-not-returned set and the value difference set, calculate the difference signature, divide the source segment, and form the source record; Based on the source record, establish comparison objects for adjacent burning rounds, compare the same set of differences to generate a retention set and a migration set, compare the difference signatures to generate a signature retention set, register the state without round comparison, and form an evolution record; Based on the block sequence chain, block state, source record, and evolution record, find the nearest normal blocks on both sides of the difference block to generate the outer boundary, collect the difference address set to generate the inner boundary, find the intersection to form the location boundary, determine the anomaly source type and anomaly form, and generate the location result.

2. The firmware anomaly location method based on burning logs as described in claim 1, characterized in that, The construction of the address chain and block order chain, and the generation of block states, include: Read the firmware burning log, divide it into write records, read back records and verification records according to the burning stage name, register the write completion mark and read back completion mark, and combine them according to the data block number, target start address and target end address to generate data block records; The data block records are divided into burning rounds according to the burning round number. Based on the target address range, write completion flag, read back completion flag, and verification result, a write overwrite set, a read back return set, and a verification difference set are generated. Bind the write overlay set, read back return set, and verification difference set under the same application-specific integrated circuit number, firmware version, data block number, target address range, and programming round to generate an address chain. Generate a block sequence chain according to the target address order. Generate the block status based on the record integrity and verification difference set.

3. The firmware anomaly location method based on burning logs as described in claim 1 or 2, characterized in that, The generation of the write-not-overwritten set, the read-back-not-returned set, and the value difference set includes: Based on the address chain and block status, filter the difference blocks, extract the address chain unit according to the data block number, target address range and burning round, read the write overlay set, read back return set and verification difference set in the address chain unit, perform address ownership verification on the verification difference set, retain the verification difference address located between the target start address and the target end address, and generate the ownership verification verification difference set; After attribution verification, the check difference addresses in the check difference set are compared with the write overwrite set and the read back return set in turn. The check difference address anomaly source type is classified to generate the write not overwrite set, the read back not return set, and the value difference set.

4. The firmware anomaly location method based on burning logs as described in claim 3, characterized in that, The calculation of the difference signature, the division of the source segment, and the formation of the source record include: Read the expected value and read-back value corresponding to the value difference set, perform a bitwise XOR operation to generate a difference signature, divide the write-uncovered set, read-back-not-returned set and value difference set according to the continuous relationship of the target address, generate the source segment, and form the source record.

5. The firmware anomaly location method based on burning logs as described in claim 1, characterized in that, The step of establishing comparison objects for adjacent burning rounds based on the source record, comparing the sets of differences of the same type, and generating a retention set and a migration set includes: Source records are collected according to the application-specific integrated circuit number, firmware version, data block number, target start address, and target end address. Round collection groups are generated. Within the round collection groups, source records are arranged according to the burning round. Adjacent source records are paired to form comparison objects for adjacent burning rounds, thus clarifying the source records of the previous burning round and the source records of the next burning round. Based on the comparison objects of adjacent programming rounds, the write-uncovered set, read-back-not-returned set, and value difference set are extracted from the source records of the previous programming round and the source records of the next programming round, respectively. The write-uncovered set of the previous programming round and the write-uncovered set of the next programming round are intersected to generate the write-uncovered retention set. The non-common addresses in the write-uncovered sets of the two programming rounds are registered as the write-uncovered migration set. The read-back-not-returned set and the value difference set are processed in the same way to generate the retention set and the migration set.

6. The firmware anomaly location method based on burning logs as described in claim 5, characterized in that, The process of comparing and contrasting signatures, generating a signature retention set, registering states without rounds of comparison, and forming an evolutionary record includes: For the set of value differences in the comparison objects of adjacent burning rounds, extract the common target address in the set of value differences of the previous burning round and the set of value differences of the next burning round. Read the difference signature of the previous burning round and the difference signature of the next burning round corresponding to the common target address. Compare the difference signature of the common target address in the two burning rounds one by one. Register the common target addresses with the same difference signature as the signature retention set. When the round collection group cannot form the comparison objects of adjacent burning rounds, register the state of no round comparison. The round collection group, adjacent burning round comparison objects, retention set, migration set, signature retention set, and no round comparison state are bound to each item according to the anomaly source type to generate an evolution record.

7. The firmware anomaly location method based on burning logs as described in claim 6, characterized in that, The process of finding the nearest normal blocks on both sides of a differing block and generating the outer boundary based on the block sequence chain, block state, source record, and evolution record includes: Read the block sequence chain and block status, take the difference block as the positioning object, search for the nearest normal block before and after the previous and next neighbor blocks, and generate the nearest normal block record; Based on the most recent normal block record and block sequence chain, the starting point of the outer boundary, the ending point of the outer boundary, and the set of data blocks within the outer boundary are determined, and the outer boundary is generated.

8. The firmware anomaly location method based on burning logs as described in claim 1, characterized in that, The aggregated set of differing addresses is used to generate an inner boundary. Intersections of these boundaries form a location boundary. The anomaly source type and anomaly morphology are determined, and the resulting location results include: Based on the outer boundary and the set of data blocks within the outer boundary, filter the data blocks with the status of difference blocks within the outer boundary, retrieve the source records corresponding to the difference blocks, collect the write-not-overwritten set, read-back-not-returned set, and value difference set of each burning round, merge them to generate a difference address set, extract the inner boundary start point and inner boundary end point in the difference address set according to the target address order, and generate the inner boundary. Read the outer and inner boundaries, perform address range intersection operation to generate the location boundary, and divide the location unit according to the source segment and difference signature within the location boundary, according to the source segment written without coverage, the source segment read back without return, and the source segment with value difference, and determine the anomaly source type. According to the evolution record within the location boundary, register the anomaly form according to the retention set, migration set, signature retention set, and no round comparison state, bind the location boundary, anomaly source type, anomaly form, source record, evolution record, and missing block information, and generate the location result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the firmware anomaly location method based on burning logs as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the firmware anomaly location method based on burning logs as described in any one of claims 1 to 8.