WO2023274398A1

WO2023274398A1 - Security detection method and apparatus for smart contract

Info

Publication number: WO2023274398A1
Application number: PCT/CN2022/103305
Authority: WO
Inventors: 傅滢; 王海军; 邬萌
Original assignee: 支付宝(杭州)信息技术有限公司; 蚂蚁区块链科技(上海)有限公司
Priority date: 2021-07-01
Filing date: 2022-07-01
Publication date: 2023-01-05
Also published as: CN113435893B; CN113435893A

Abstract

A security detection method and apparatus for a smart contract, related to the technical field of blockchains. The method comprises: determining an inter-process control flow graph corresponding to a smart contract (101); in the inter-process control flow graph, determining a plurality of target parameters for performing assignment operation by using external data, and operation frequencies of respective assignment operation of the plurality of target parameters (103), the plurality of target parameters belonging to world state parameters of the block chain, and the external data being input into the smart contract from the outside; and determining whether the smart contract has a security risk according to the operation frequencies corresponding to the plurality of target parameters (105).

Description

Smart contract security detection method and device

This application claims priority to a Chinese patent application filed with the State Intellectual Property Office of China on July 1, 2021, with application number 202110749282.5 and titled "Security Detection Method and Device for Smart Contracts", the entire contents of which are hereby incorporated by reference In this application.

technical field

One or more embodiments of this specification relate to the field of blockchain technology, and in particular to a method and device for security detection of smart contracts.

Background technique

Smart contracts are commitments defined in digital form. Among them, the blockchain system that deploys smart contracts can ensure that the storage, reading and execution of smart contracts can be traced and cannot be tampered with. Both parties to the transaction registered in the blockchain system can deploy or call smart contracts to exchange resources or other business purposes in a transparent and conflict-free manner.

The inventor found that the smart contract itself may have security risks. For example, the smart contract itself may be at risk of being rearranged, and intruders may use rearrangement attacks on the smart contract, so that the transaction initiated by the user to call the smart contract cannot achieve its expected effect.

It is hoped that there will be a new technical solution to detect whether the smart contract itself has security risks.

Contents of the invention

One or more embodiments of this specification provide a smart contract security detection method and device, which can more efficiently detect whether there is a security risk in a smart contract.

In the first aspect, a security detection method for smart contracts is provided, including: determining an inter-procedural control flow graph corresponding to the smart contract; in the inter-procedural control flow graph, determining a number of target parameters for assignment operations using external data, And the operating frequency of the assignment operation for each of the several target parameters, the several target parameters belong to the world state parameters of the block chain, and the external data is input from the outside to the smart contract; according to the corresponding operations of the several target parameters Frequency, to determine whether the smart contract has a security risk.

In a possible implementation manner, according to the operating frequencies corresponding to the several target parameters, determining whether there is a security risk in the smart contract specifically includes: when there is an operating frequency greater than 1, determining that there is a security risk in the smart contract risk.

In a possible implementation manner, the smart contract includes a code segment and a data segment stored in the memory, the data segment includes the several target parameters, and the code segment uses the data segment in the The base address in memory and the address offset corresponding to the target parameter identify the target parameter.

In a possible implementation manner, the determining, in the inter-procedural control flow diagram, several target parameters for performing assignment operations using external data includes: determining a plurality of first instruction sequences in the inter-procedural control flow diagram , the first instruction sequence is used to represent the propagation process of external data; a number of second instruction sequences are determined in the inter-process control flow graph, and the second instruction sequence is used to represent the process in which the world state parameters are assigned; according to The overlapping assignment instructions between the plurality of first instruction sequences and the plurality of second instruction sequences determine the plurality of target parameters.

In a possible implementation manner, determining several first instruction sequences in the inter-procedural control flow graph includes: adding taint marks to parameters corresponding to external data in the inter-procedural control flow graph, along the data The flow direction performs taint tracking, and the instructions corresponding to the parameters added with taint marks are included in the first instruction sequence.

In a possible implementation manner, determining several second instruction sequences in the inter-procedural control flow graph includes: determining a world state parameter in the inter-procedural control flow graph; , determining an assignment instruction for assigning a world state parameter, and including the assignment instruction into the second instruction sequence.

In a possible implementation manner, determining the world state parameter in the inter-procedural control flow graph includes: determining a plurality of function call instructions for calling a predetermined function in the inter-procedural control flow graph; calling the function The parameter that the instruction requests to update is determined as the world state parameter.

In a possible implementation manner, according to the overlapping assignment instructions between the several first instruction sequences and the several second instruction sequences, determining the several target parameters includes: setting the world to which the overlapping assignment instructions point The state parameter is used as the target parameter.

In a possible implementation manner, the process in which a single world state parameter represented by the second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process, or the third type of assignment process; wherein the first type of assignment The process indicates that external data is used for direct assignment, the second type of assignment process indicates that the operation result of logic operation is used for external data, and the third type of assignment process is used for assignment without external data. The determining whether there is a security risk in the smart contract according to the operation frequency corresponding to each of the several target parameters specifically includes: determining whether there is a first parameter among the several target parameters, wherein the operation frequency corresponding to the first parameter If it is greater than 1, the process in which the first parameter is assigned includes the first type of assignment process and the second type of assignment process; if yes, it is determined that the smart contract has a security risk.

In a possible implementation manner, the process in which a single world state parameter represented by the second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process, or the third type of assignment process; wherein the first type of assignment The process indicates that external data is used for direct assignment, the second type of assignment process indicates that the operation result of logical operation is used for external data, and the third type of assignment process indicates that no external data is used for assignment. The determining whether there is a security risk in the smart contract according to the operation frequency corresponding to each of the several target parameters specifically includes: determining whether there is a second parameter among the several target parameters, wherein the operation frequency corresponding to the second parameter is greater than 1, the process in which the second parameter is assigned includes a first-type assignment process and a second-type assignment process, the first-type assignment process and the second-type assignment process of the second parameter, and the control flow between the processes The figure corresponds to different method functions; if so, it is determined that the smart contract has a security risk.

In a possible implementation manner, the smart contract is a smart contract to be deployed in the blockchain; and/or, the security risk specifically includes the risk of being rearranged and attacked.

In a second aspect, a security detection device for a smart contract is provided, including: an analysis processing unit configured to determine an inter-procedural control flow graph corresponding to the smart contract; an assignment detection unit configured to, in the inter-procedural control flow graph, Determine a number of target parameters for the assignment operation using external data, and the operation frequency of each of the number of target parameters for the assignment operation, the number of target parameters belong to the world state parameters of the blockchain, and the external data is input from the outside to the smart A contract; a risk determination unit configured to determine whether the smart contract has a security risk according to the operation frequency corresponding to each of the several target parameters.

In a possible implementation manner, the risk determining unit is configured to determine that the smart contract has a security risk when there is an operation frequency greater than 1.

In a possible implementation manner, the assignment detection unit includes: a first determination subunit configured to determine a number of first instruction sequences in the inter-procedural control flow graph, and the first instruction sequences are used to represent external The propagation process of data; the second determination subunit is configured to determine a number of second instruction sequences in the inter-process control flow graph, and the second instruction sequences are used to represent the process in which the world state parameters are assigned; the assignment detection subunit , configured to determine the plurality of target parameters according to overlapping assignment instructions between the plurality of first instruction sequences and the plurality of second instruction sequences.

In a possible implementation manner, the first determining subunit is configured to add taint marks to parameters corresponding to external data in the inter-procedural control flow graph, and perform taint tracking along the data flow direction, adding Instructions corresponding to parameters marked with taints are included in the first instruction sequence.

In a possible implementation manner, the second determining subunit is configured to determine the world state parameter in the inter-procedural control flow graph; in the inter-procedural control flow graph, determine the An assignment instruction that assigns a value, and puts the assignment instruction into the second instruction sequence.

In a possible implementation manner, the second determination subunit is specifically configured to determine, in the inter-procedural control flow graph, several function call instructions that call predetermined functions; and request the updated parameters of the function call instructions Determined as a world state parameter.

In a possible implementation manner, the assignment detection subunit is specifically configured to use the world state parameter pointed to by the overlapping assignment instruction as the target parameter.

In a possible implementation manner, the process in which a single world state parameter represented by the second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process, or the third type of assignment process; wherein the first type of assignment The process indicates that external data is used for direct assignment, the second type of assignment process indicates that the operation result of logic operation is used for external data, and the third type of assignment process is used for assignment without external data. The risk determination unit is specifically configured to determine whether there is a first parameter among the several target parameters, wherein the operation frequency corresponding to the first parameter is greater than 1, and the process of assigning the first parameter includes a first-type assignment process and the second type of value assignment process; if yes, it is determined that the smart contract has a security risk.

In a possible implementation manner, the process in which a single world state parameter represented by the second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process, or the third type of assignment process; wherein the first type of assignment The process indicates that external data is used for direct assignment, the second type of assignment process indicates that the operation result of logical operation is used for external data, and the third type of assignment process indicates that no external data is used for assignment. The risk determination unit is specifically configured to determine whether there is a second parameter among the several target parameters, wherein the operation frequency corresponding to the second parameter is greater than 1, and the process of assigning the second parameter includes the first type of assignment process and the second type of value assignment process, the first type of value assignment process and the second type of value assignment process of the second parameter correspond to different method functions in the inter-process control flow graph; if yes, determine that the smart contract exists security risk.

In a possible implementation manner, the smart contract is a smart contract to be deployed in a blockchain; and/or, the security risk specifically includes the risk of being subjected to a rearrangement attack.

In a third aspect, a computer-readable storage medium is provided, on which a computer program/instruction is stored, and when the computer program/instruction is executed in a computing device, the computing device executes the method as described in any one of the first aspect .

In a fourth aspect, a computing device is provided, including a memory and a processor, where computer programs/instructions are stored in the memory, and when the processor executes the computer programs/instructions, the method according to any one of the first aspect is implemented.

Through the methods and devices provided in one or more embodiments of this specification, it is possible to determine the inter-procedural control flow graph corresponding to the smart contract, and then based on the inter-procedural control flow graph, determine a number of assignment operations using external data in the smart contract. The operation frequency corresponding to each target parameter, and then based on the corresponding operation frequency of several target parameters, quickly and efficiently determine whether there is a security risk in the smart contract.

Description of drawings

In order to illustrate the technical solutions of the embodiments of this specification more clearly, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For Those of ordinary skill in the art can also obtain other drawings based on these drawings without making creative efforts.

Fig. 1 is a flowchart of a security detection method for a smart contract provided in the embodiment of this specification;

Fig. 2 is a schematic diagram of the process of determining the target parameters of the assignment operation using external data in the embodiment of this specification;

FIG. 3 is an exemplary inter-process control flow diagram provided in the embodiment of this specification;

Fig. 4 is a schematic diagram of a smart contract security detection device provided in the embodiment of this specification.

detailed description

Various non-limiting embodiments provided in this specification will be described in detail below with reference to the accompanying drawings.

The inventor found that the smart contract itself may have security risks. For example, the smart contract itself may be at risk of being rearranged, and intruders may use the smart contract to rearrange the attack, so that the transaction used to call the smart contract cannot achieve the expected effect. As another example, the smart contract itself allows the use of external data to assign values to the world state parameters, and the intruder may maliciously modify the world state parameters by calling different method functions in the smart contract, so that the transaction used to call the smart contract cannot achieve its expected Effect.

According to the research of the inventors, after determining the inter-procedural control flow graph (ICFG) of the smart contract, it is possible to statically analyze the ICFG to know that the external data input into the smart contract from the outside is used in the smart contract. The operation frequency of the assignment operation for several target parameters of the world state parameters of the blockchain. The operation frequency corresponding to several target parameters can reflect the possibility of security risks in the smart contract itself to some extent, where the security risks include but not limited to the risk of being reordered.

Correspondingly, in order to quickly and efficiently detect whether there is a security risk in a smart contract, the embodiments of this specification provide a method and device for detecting security of a smart contract. As shown in Figure 1, the method may at least include: step 101, determine the inter-procedural control flow graph corresponding to the smart contract; step 103, in the inter-procedural control flow graph, determine several target parameters for the assignment operation using external data , and the operating frequency of the assignment operation for each of the several target parameters, the several target parameters belong to the world state parameters of the block chain, and the external data is input from the outside to the smart contract; step 105, according to the several target parameters The respective operation frequencies are used to determine whether the smart contract has a security risk.

The smart contract for security detection can be the smart contract to be deployed to the blockchain, or it can also be the smart contract that has been deployed to the blockchain. In view of the fact that the smart contract is unmodifiable or difficult to modify after it is deployed to the blockchain, it is preferable to perform security testing on the smart contract to be deployed to the blockchain; thus, when it is determined that the smart contract has a security risk, It is helpful for staff to quickly and efficiently modify smart contracts, and it can avoid deploying smart contracts with security risks to the blockchain and causing other security problems for the blockchain.

Each method step in the embodiment shown in FIG. 1 will be described in detail below.

First, in step 101, the inter-process control flow graph corresponding to the smart contract is determined.

It is possible to disassemble the smart contract loaded into the memory, and first construct the corresponding control flow graph (control flow graph, CFG) for each function in the smart contract based on the disassembly result. Among them, CFG is a directed graph with basic blocks as nodes, and a single basic block includes multiple assembly instructions that need to be executed sequentially; the directed edge connecting two nodes in CFG indicates that there is a control transfer between the two nodes, and in CFG Has a unique entry node and exit node. Then, based on the disassembly results, the function call relationship in the smart contract is determined, and the ICFG is constructed using the CFG corresponding to each function according to the function call relationship. Among them, ICFG is a directed graph composed of CFG corresponding to each function in the smart contract, several Call edges and several Return edges; Call edges are function call instructions in the CFG corresponding to the calling function, pointing to the corresponding The entry node in the CFG; the Return edges point to the next instruction of the function call instruction in the CFG corresponding to the calling function from the Ret instruction in the CFG corresponding to the called function. In other words, the ICFG corresponding to the smart contract is composed of multiple assembly instructions, which at least characterize the arrangement/execution order of the assembly instructions corresponding to the same or different functions.

The CFG composed of assembly instructions usually does not directly contain the assigned world state parameters, but represents the world state parameters through the memory address or the address pointer pointing to the memory address, and completes the world state parameter by writing data on the memory address assignment. If it is necessary to ensure that in the subsequent process, based on ICFG, all the assignment operations that use external data to assign values to each world state parameter can be found, it is necessary to ensure that a single world state parameter has a fixed and unique memory address in the memory, that is, the memory address needs to be used as An identifier for the world state parameter. In view of the above needs, in some embodiments, for the detected smart contract loaded into the memory, it includes the code segment and the data segment stored in the memory, and the data segment includes but not limited to all the world state parameters involved in the smart contract ; For a single world state parameter, the code segment identifies the world state parameter through the base address of the data segment in memory and the address offset corresponding to the world state parameter. Correspondingly, since ICFG is essentially obtained by disassembling the code segment of the smart contract, the memory address identified by the base address and offset in ICFG or the address pointer pointing to the memory address can be used as The identifier of the corresponding world state parameter.

Next, in step 103, in the inter-procedural control flow graph, determine a number of target parameters that use external data to perform assignment operations, and the operation frequency of each of the number of target parameters to perform the assignment operation, and the number of target parameters belong to The world state parameters of the blockchain, and external data are input to the smart contract from the outside.

Through the analysis of ICFG, it is possible to discover the process of dissemination of external data and the process of assigning values to world state parameters, and then based on the process of dissemination of external data and the process of assigning values to world state parameters, from all the world state parameters involved in the smart contract, determine The target parameter of the assignment operation using external data. In a more specific example, please refer to FIG. 2 , through the following steps 1031 to 1035 , several target parameters for assignment operations using external data may be determined in the ICFG.

Step 1031 , determining a number of first instruction sequences in the inter-procedural control flow graph, where the first instruction sequences are used to characterize the propagation process of external data.

By adding taint marks to the parameters corresponding to the external data in ICFG, and then performing taint tracking along the data flow direction, multiple instructions corresponding to the parameters added with taint marks can be sequentially determined, and multiple instructions determined in sequence can be composed A first sequence of instructions for characterizing the propagation process of external data. As an example, please refer to the example and simplified ICFG in Figure 3. The CFGs corresponding to the method functions Function_A and Int Add(x, y) in the smart contract are composed of different instruction sequences, which are not shown in the CFG. The basic block, not shown, represents the directed edge of the control transfer; the direction of the straight arrow indicates the arrangement/execution order of the instructions in the CFG, and the broken line arrow represents the Call edge or Return edge; in addition, it will correspond to the parameter a and parameter b of the external data The instantiated parameter/parameter value is exemplarily expressed as "0Ah (i.e. instantiated parameter/parameter value 10)" and "14h (i.e. instantiated parameter/parameter value 20), which can obviously be other instantiations in actual business scenarios Parameter/parameter value. Then, after analyzing the ICFG to determine that parameter a and parameter b are parameters corresponding to external data, parameter a and parameter b can be marked, that is, "mov dword ptr[a], 0Ah" and "mov dword ptr[b], 14h" in ptr[a] and ptr[b] are tainted, and then taint tracking is performed along the data flow direction, and the starting point is determined to be "mov dword ptr[a], 0Ah " and the end point is the first instruction sequence of "mov dword ptr[c], eax". It should be noted that in the first instruction sequence, the instruction sequence that constitutes the CFG corresponding to Int Add(x, y) is located in " call_Add" and "add esp,8".

Step 1033, determine a number of second instruction sequences in the inter-process control flow diagram, the second instruction sequences are used to represent the process in which the world state parameters are assigned values.

First, all world state parameters involved in smart contracts can be determined in ICFG. In a more specific example, several function call instructions for calling a predetermined function are firstly determined in the ICFG. Among them, the predetermined function can be used to update the world state parameters of the blockchain in the non-volatile storage medium. For example, the predetermined function can write the assigned world state parameters into the non-volatile Volatile storage medium; or, the predetermined function can be a method function used in the block chain node to realize the predetermined service, for example, the predetermined service realized by the predetermined function is that when it is called by the smart contract, the block chain node can know The smart contract has completed the assignment of a certain world state parameter, so that the blockchain node can write the assigned world state parameter into the non-volatile storage medium according to the corresponding business logic. Next, for each determined function call instruction, search upwards in the ICFG starting from the function call instruction, determine the parameter that the function call instruction requests to update, and determine the parameter as the world state parameter. Exemplarily, please continue to refer to FIG. 3 , the function call instruction "call_Rt" that calls the predetermined function "_Rt" can be determined first in the ICFG, and then search upwards in the ICFG starting from "call_Rt", and determine that "call_Rt" requests an update The parameter is the parameter c added to the function stack by "mov eax, dword ptr[c]" and "push eax", and then the parameter c is determined as the world state parameter, where the parameter c is passed through the address pointer ptr[c in Figure 3 ] for identification.

Next, since the world state parameter is identified by the memory address or address pointer in ICFG, for each determined world state parameter, the memory address or address pointer used to identify the world state parameter can be used to determine the world state parameter in ICFG. All assignment instructions for assigning values to the world state parameters. Exemplarily, please continue to refer to FIG. 3 , based on the determined address pointer ptr[c] used to identify the world state parameter c, the assignment instruction "mov dword" for assigning ptr[c] can be determined in ICFG ptr[c],eax".

Finally, for each determined world state parameter, all assignment instructions for assigning the world state parameter may be classified into different second instruction sequences. Specifically, for each assignment instruction used to assign a value to a single world state parameter, the assignment instruction can be used as a starting point to search upwards in ICFG until the source of the parameter value directly or indirectly used by the assignment instruction is found, Then, one of the second instruction sequences used to characterize the process in which the world state parameter is assigned is determined. Exemplarily, please continue to refer to Figure 3. When the "mov dword ptr[c], eax" used to assign a value to ptr[c] is determined, you can use "mov dword ptr[c], eax" in ICFG Search upwards for the starting point until the parameters a and b indirectly used by "mov dword ptr[c], eax" are found. Among them, "mov eax, dword ptr[b]" and "push eax" are used to add the instantiation parameter ptr[b] of parameter b to the function stack, and "mov ecx, dword ptr[a]" and "push ecx" are used to The instantiation parameter ptr[a] of the parameter a is added to the function stack, so the searched four instructions can be used as the source of the parameter value indirectly used by "mov dword ptr[c], eax"; or, the above four instructions are used for Add the instantiation parameters ptr[a] and ptr[b] to the function stack, but it does not indicate how ptr[a] and ptr[b] are created in ICFG, so you can continue to search upwards in ICFG until you find "mov dword ptr[a], 0Ah" and "mov dword ptr[b], 14h" used to create ptr[a] and ptr[b] in the function stack, and then "mov dword ptr[a], 0Ah " and "mov dword ptr[b], 14h" as the source of the parameter value indirectly used by "mov dword ptr[c], eax". Correspondingly, for the assignment instruction "mov dword ptr[c], eax" used to assign values to the world state parameter ptr[c], the starting point can be determined from ICFG as "mov eax, dword ptr[b]" Or "mov dword ptr[a], 0Ah", the end point is the second instruction sequence of "mov dword ptr[c], eax", the second instruction sequence can completely represent the process of the world state parameter ptr[c] being assigned .

It should be noted that in the ICFG of the example in Figure 3, a second instruction sequence used to represent the process of ptr[c] being assigned can be determined, but in actual business scenarios, the ICFG corresponding to the smart contract may determine A plurality of second instruction sequences used to characterize the process in which ptr[c] is assigned. As an example, assume that the smart contract includes not only the aforementioned method functions Function_A and Int Add(x, y), but also the method function Function_B, where the function implemented by Function_B is to directly assign external data to the world state parameter c, and pass Calling the predetermined function "_Rt" enables the blockchain node to update the world state parameter c in the non-volatile storage medium. Then, the ICFG corresponding to the smart contract may include the CFG corresponding to Function_B on the basis of the ICFG shown in Figure 3, and the CFG corresponding to Function_B also includes an assignment instruction for assigning ptr[c]. Correspondingly, since the CFG corresponding to Function_A and the CFG corresponding to Function_B both include assignment instructions for assigning ptr[c], it can be determined from ICFG that two parameters representing the world state c are assigned The second instruction sequence of the process.

Step 1035: Determine the number of target parameters according to overlapping assignment instructions between the number of first instruction sequences and the number of second instruction sequences.

Referring to the previous example, it is not difficult to find that the last instruction in the second instruction sequence is an assignment instruction used to assign values to world state parameters, and the first instruction in the first instruction sequence is used to create external data in the function stack or set External data is added to the function stack. Therefore, when the last instruction in the second instruction sequence coincides/is the same as the last instruction in the first instruction sequence, it means that the second instruction sequence may directly or indirectly use external data to assign values to the world state parameters, Therefore, the parameter pointed to by the overlapping assignment instruction can be determined as the target parameter for assignment using external data, that is, the parameter requested by the overlapping assignment instruction for assignment can be determined as the target parameter for assignment using external data. In a more specific example, for each world state parameter involved in the smart contract, after determining all the second instruction sequences used to characterize the process of assigning the world state parameter, each of the world state parameters corresponding to the The second instruction sequence is to detect whether there is a target instruction sequence in all the determined first instruction sequences, and the last instruction in the target instruction sequence is the same as the last instruction in the second instruction sequence; if so, the world state The parameter is determined as a target parameter for an assignment operation using external data. Correspondingly, the aforementioned processing is performed on each second sequence of instructions used to characterize the process in which the world state parameter is assigned a value, and the cumulative number of times the world state parameter is determined as the target parameter can be counted. The number of times is the operation frequency of the assignment operation of the world state parameter using external data.

Returning to Fig. 1, in step 105, it is determined whether the smart contract has a security risk according to the operation frequency corresponding to each of the target parameters.

In a possible implementation, when there is an operation frequency greater than 1 among the operation frequencies corresponding to several target parameters, it can be determined that the smart contract has a security risk.

In some embodiments, the process of assigning the world state parameters represented by the second instruction sequence can also be divided into the first type of assignment process, the second type of assignment process or the third type of assignment process; wherein the first type of assignment process indicates The external data is used for direct assignment, the second type of assignment process indicates that the operation result of the logical operation is used for external data, and the third type of assignment process indicates that no external data is used for assignment. For a single world state parameter, when determining a number of second instruction sequences that characterize the assignment process of the world state parameter, the respective assignment types corresponding to the number of second instruction sequences can be determined, and then the statistics corresponding to the world state parameter All assignment types of . Correspondingly, in step 105, it can specifically be determined whether there is a first parameter among several target parameters, the operation frequency corresponding to the first parameter is greater than 1, and the process of assigning the first parameter includes both the first type of assignment process and the second type of assignment process. There are two types of assignments in the class assignment process, that is, the assignment type corresponding to the first parameter includes both the first type of assignment process and the second type of assignment process; if the first parameter exists in several target parameters, it is determined that the smart contract has a security risk. In this way, by comprehensively considering the operation frequency of world state parameters using external data for assignment operations, and the assignment type of the process of world state parameters being assigned, it is possible to more accurately determine whether there is a security risk in the smart contract.

For example, please continue to refer to FIG. 3 . A second instruction sequence starting from "mov dword ptr[a], 0Ah" and ending with "mov dword ptr[c], eax" represents the process of assigning the world state parameter c. Since the second instruction sequence indicates to use the external data corresponding to the parameters a and b to perform logic operations, and use the operation results obtained after the logic operations to assign values to the world state parameter c, it can be determined that the second instruction sequence belongs to The assignment type of is the second type of assignment process, that is, the assignment process of determining the world state parameter c includes the second type of assignment process, or in other words, the assignment type corresponding to the determination of the world state parameter c includes the second type of assignment process.

Continuing to assume that the smart contract also includes the method function Function_B of the preceding example, then based on the assignment instruction used to assign the world state parameter c in the CFG corresponding to Function_B, another method used to represent the process of assigning the world state parameter c can be determined. A second instruction sequence, since the second instruction sequence uses external data to directly assign a value to the world state parameter c, it can be determined that the assignment type to which the second instruction sequence belongs is the first type of assignment process, that is, the process of determining the world state parameter c The assignment process includes the first type of assignment process, or in other words, determining the assignment type corresponding to the world state parameter c includes the first type of assignment process.

Continue to assume that the smart contract also includes the method function Function_C, where the function implemented by Function_C is to assign the preset value to the world state parameter c, and realize the world state parameter c in the non-volatile storage medium by calling the predetermined function "_Rt". to update. Then the ICFG corresponding to the smart contract may include the CFG corresponding to Function_C, and the CFG corresponding to Function_C includes an assignment instruction for assigning ptr[c]. Correspondingly, based on the assignment instruction for assigning the world state parameter c in the CFG corresponding to Function_C, another second instruction sequence for representing the process of assigning the world state parameter c can be determined, the second instruction The sequence uses a preset constant to directly assign a value to the world state parameter c, and does not use external data to assign a value to the world state parameter c, so it can be determined that the assignment type of the second instruction sequence belongs to the third type of assignment process, that is, to determine the world state parameter The assignment process of c includes the third type of assignment process, or in other words, determining the assignment type corresponding to the world state parameter c includes the third type of assignment process.

In some embodiments, for a single world state parameter, when several second instruction sequences representing the process of assigning the world state parameter are determined, the method functions corresponding to each of the several second instruction sequences can also be determined, for example, the The method function to which the assignment instruction for assigning the world state parameter in the single second instruction sequence belongs is determined as the function corresponding to the second instruction sequence. Correspondingly, in step 105, it can be specifically determined whether there is a second parameter among several target parameters, the operation frequency corresponding to the second parameter is greater than 1, and the process of assigning the second parameter includes the first type of assignment process and the second type Assignment process, the first type of assignment process and the second type of assignment process included in the second parameter correspond to different method functions in the inter-procedural control flow graph; if there is a second parameter in several target parameters, it is determined that the smart contract has a security risk . In this way, by comprehensively considering the operation frequency of the world state parameter using external data for assignment operations, the assignment type of the world state parameter assigned process, and the method functions corresponding to the instruction sequences of different assignment types, it is possible to more accurately determine whether the smart contract is There is a security risk.

For example, please continue to refer to FIG. 3 . Assume that the smart contract includes the aforementioned method function Function_A, Int Add(x, y) and method function Function_B. Then, the second instruction sequence determined based on the assignment instruction used to assign the world state parameter c in the CFG corresponding to Function_A, its assignment type is the second type of assignment process, and its corresponding function is Function_A, that is, the world state The method function corresponding to the second type of assignment process included in the parameter c is Function_A; the second instruction sequence determined based on the assignment instruction used to assign the world state parameter c in the CFG corresponding to Function_B, and its assignment type is the first One type of assignment process, the corresponding function includes Function_B, that is, the function corresponding to the first type of assignment process of the world state parameter c is Function_B. In this case, the world state parameter c is used as the target parameter, and the operation frequency of its assignment operation using external data is 2 times, and its assignment process includes the first type of assignment process and the second type of assignment process, which includes the first The method function Function_A corresponding to the class assignment process is different from the method function Function_B corresponding to the second class assignment process included in it, so the world state parameter c will be determined as the second state parameter, and then it can be determined that there is a security risk in the smart contract .

It should be noted that the exemplified and simplified ICFG in Figure 3 above is only used to assist in describing the technical solutions provided in the embodiments of this specification. The specification does not describe in detail the functions that some commonly used instructions in Figure 3 need to implement, but this does not It will affect those skilled in the art to correctly understand the ICFG illustrated in FIG. 3 . In an actual business scenario, instructions that implement the same function may have different expressions, and more or fewer instructions may be used to implement the functions required by one or more instructions in FIG. 3 .

Based on the same idea as the foregoing method embodiments, the embodiments of this specification also provide a smart contract security detection device. As shown in Figure 4, the device includes: a parsing processing unit 401 configured to determine the inter-procedural control flow graph corresponding to the smart contract; an assignment detection unit 403 configured to determine, in the inter-procedural control flow graph, the Several target parameters of the assignment operation, and the operation frequency of each of the several target parameters for the assignment operation, the several target parameters belong to the world state parameters of the block chain, and external data is input from the outside to the smart contract; the risk determination unit 405, configured to determine whether there is a security risk in the smart contract according to the operation frequency corresponding to each of the several target parameters.

In a possible implementation manner, the risk determination unit 405 is configured to determine that the smart contract has a security risk when there is an operation frequency greater than 1.

In a possible implementation manner, the assignment detection unit 403 includes: a first determination subunit configured to determine a number of first instruction sequences in the interprocedural control flow graph, and the first instruction sequences are used to represent The propagation process of external data; the second determination subunit is configured to determine a number of second instruction sequences in the inter-process control flow diagram, and the second instruction sequences are used to represent the process in which the world state parameters are assigned; the assignment detector A unit configured to determine the plurality of target parameters according to overlapping assignment instructions between the plurality of first instruction sequences and the plurality of second instruction sequences.

In a possible implementation manner, the process in which a single world state parameter represented by the second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process, or the third type of assignment process; wherein the first type of assignment The process indicates that external data is used for direct assignment, the second type of assignment process indicates that the operation result of logic operation is used for external data, and the third type of assignment process is used for assignment without external data. The risk determination unit 405 is specifically configured to determine whether there is a first parameter among the several target parameters, wherein the operation frequency corresponding to the first parameter is greater than 1, and the process of assigning the first parameter includes the first type of assignment process and the second type of assignment process; if so, it is determined that the smart contract has a security risk.

In a possible implementation manner, the process in which a single world state parameter represented by the second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process, or the third type of assignment process; wherein the first type of assignment The process indicates that external data is used for direct assignment, the second type of assignment process indicates that the operation result of logical operation is used for external data, and the third type of assignment process indicates that no external data is used for assignment. The risk determination unit 405 is specifically configured to determine whether there is a second parameter among the several target parameters, wherein the operation frequency corresponding to the second parameter is greater than 1, and the process of assigning the second parameter includes the first type of assignment process and the second type of value assignment process, the first type of value assignment process and the second type of value assignment process of the second parameter correspond to different method functions in the inter-procedural control flow graph; if so, determine that the smart contract exists Security Risk.

Those skilled in the art should be aware that in one or more of the above examples, the functions described in this specification may be implemented by hardware, software, firmware or any combination thereof. When implemented in software, the computer programs corresponding to these functions can be stored in a computer-readable medium or transmitted as one or more instructions/codes on a computer-readable medium, so that the computer programs corresponding to these functions can be read by the computer During execution, the method described in any one of the embodiments of this specification is realized by a computer.

The embodiments of this specification also provide a computer-readable storage medium on which computer programs/instructions are stored. When the computer programs/instructions are executed in a computing device, the computing device executes the A security detection method for smart contracts.

The embodiments of this specification also provide a computing device, including a memory and a processor, the memory stores computer programs/instructions, and when the processor executes the computer programs/instructions, any one of the embodiments of this specification is realized The security detection method of the smart contract provided in .

Each embodiment in this specification is described in a progressive manner, the same and similar parts in each embodiment can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for relevant parts, please refer to part of the description of the method embodiment.

The foregoing describes specific embodiments of this specification. Other implementations are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or may be advantageous in certain embodiments.

The specific embodiments described above have further described the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above descriptions are only specific embodiments of the present invention and are not intended to limit the scope of the present invention. Protection scope, any modification, equivalent replacement, improvement, etc. made on the basis of the technical solution of the present invention shall be included in the protection scope of the present invention.

Claims

A security detection method for smart contracts, comprising:

Determine the inter-process control flow graph corresponding to the smart contract;

In the inter-process control flow diagram, determine a number of target parameters using external data for assignment operations, and the operation frequency of each of the number of target parameters for the assignment operation, and the number of target parameters belong to the world state of the blockchain Parameters, external data is input from the outside to the smart contract;

According to the operating frequency corresponding to each of the several target parameters, it is determined whether there is a security risk in the smart contract.
The method according to claim 1, wherein, according to the operating frequency corresponding to each of the several target parameters, determining whether the smart contract has a security risk, specifically includes: when there is an operating frequency greater than 1, determining whether the smart contract There is a security risk.
The method according to claim 1, wherein the smart contract includes a code segment and a data segment stored in the memory, the data segment includes the several target parameters, and the code segment passes the data segment in the The base address in the memory and the address offset corresponding to the target parameter identify the target parameter.
The method according to claim 1, wherein, in the inter-procedural control flow graph, determining a number of target parameters using external data for assignment operations includes:

Determining several first instruction sequences in the inter-procedural control flow graph, the first instruction sequences are used to characterize the propagation process of external data;

Determining a number of second instruction sequences in the inter-process control flow graph, the second instruction sequences being used to represent a process in which world state parameters are assigned values;

The plurality of target parameters are determined according to overlapping assignment instructions between the plurality of first instruction sequences and the plurality of second instruction sequences.
The method according to claim 4, wherein determining a plurality of first instruction sequences in the inter-procedural control flow graph comprises: adding taint marks to parameters corresponding to external data in the inter-procedural control flow graph, along the Perform taint tracking along the data flow, and include instructions corresponding to parameters with taint marks added into the first instruction sequence.
The method of claim 4, wherein determining a plurality of second instruction sequences in the interprocedural control flow graph comprises: determining a world state parameter in the interprocedural control flow graph; In the figure, an assignment instruction for assigning a world state parameter is determined, and the assignment instruction is included in the second instruction sequence.
The method according to claim 6, wherein determining the world state parameters in the inter-procedural control flow graph comprises: determining a plurality of function call instructions for calling a predetermined function in the inter-procedural control flow graph; The parameters requested by the function call instruction to be updated are determined as world state parameters.
The method according to claim 4, wherein, according to the overlapping assignment instructions between the plurality of first instruction sequences and the plurality of second instruction sequences, determining the plurality of target parameters includes: directing the overlapping assignment instructions to The world state parameter of is used as the target parameter.
The method according to claim 4, wherein the process in which the world state parameter represented by the single second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process or the third type of assignment process; wherein the first The class assignment process indicates direct assignment using external data, the second class assignment process indicates assignment of the operation result after logical operation using external data, and the third class assignment process is used to indicate that no external data is used for assignment;

The determining whether there is a security risk in the smart contract according to the operation frequency corresponding to each of the target parameters includes:

Determining whether there is a first parameter among the plurality of target parameters, wherein the operation frequency corresponding to the first parameter is greater than 1, and the process of assigning the first parameter includes a first type of assignment process and a second type of assignment process;

If yes, it is determined that the smart contract has a security risk.
The method according to claim 4, wherein the process in which the world state parameter represented by the single second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process or the third type of assignment process; wherein the first The class assignment process indicates direct assignment using external data, the second class assignment process indicates the use of external data to perform logical operation results for assignment, and the third class assignment process indicates not using external data for assignment;

The determining whether there is a security risk in the smart contract according to the operation frequency corresponding to each of the target parameters includes:

Determining whether there is a second parameter among the several target parameters, wherein the operation frequency corresponding to the second parameter is greater than 1, the process of assigning the second parameter includes a first type of assignment process and a second type of assignment process, the The first type of assignment process and the second type of assignment process of the second parameter correspond to different method functions in the inter-procedural control flow graph;

If yes, it is determined that the smart contract has a security risk.
The method according to any one of claims 1-10, wherein the smart contract is a smart contract to be deployed in the blockchain; and/or, the security risk specifically includes the risk of being rearranged and attacked .
A security detection device for smart contracts, comprising:

The analysis processing unit is configured to determine the inter-process control flow graph corresponding to the smart contract;

The assignment detection unit is configured to determine, in the inter-procedural control flow graph, a number of target parameters that use external data to perform assignment operations, and an operation frequency for each of the number of target parameters to perform the assignment operation, and the number of target parameters belong to The world state parameters of the blockchain, external data are input from the outside to the smart contract;

The risk determining unit is configured to determine whether the smart contract has a security risk according to the operating frequency corresponding to each of the several target parameters.
The device according to claim 12, wherein the risk determining unit is configured to determine that the smart contract has a security risk when there is an operation frequency greater than 1.
The device according to claim 12, wherein the smart contract includes a code segment and a data segment stored in the memory, the data segment includes the plurality of target parameters, and the code segment passes the data segment in the The base address in the memory and the address offset corresponding to the target parameter identify the target parameter.
The apparatus according to claim 12, wherein the assignment detection unit comprises:

The first determination subunit is configured to determine a plurality of first instruction sequences in the inter-process control flow graph, and the first instruction sequences are used to represent the propagation process of external data;

The second determination subunit is configured to determine a number of second instruction sequences in the inter-process control flow diagram, and the second instruction sequences are used to represent the process in which the world state parameter is assigned;

The assignment detection subunit is configured to determine the plurality of target parameters according to overlapping assignment instructions between the plurality of first instruction sequences and the plurality of second instruction sequences.
The device according to claim 15, wherein the first determining subunit is configured to add taint marks to parameters corresponding to external data in the inter-procedural control flow graph, and perform taint tracking along the data flow direction, and Instructions corresponding to parameters added with taint marks are included in the first instruction sequence.
The device according to claim 15, wherein the second determining subunit is configured to determine the world state parameter in the inter-procedural control flow graph; in the inter-procedural control flow graph, determine An assignment instruction for assigning values to state parameters, and the assignment instruction is included in the second instruction sequence.
The device according to claim 17, wherein the second determination subunit is specifically configured to determine a number of function call instructions for calling a predetermined function in the inter-procedural control flow graph; and request an update of the function call instructions The parameters of are determined as world state parameters.
The device according to claim 15, wherein the assignment detection subunit is specifically configured to use the world state parameter pointed to by the overlapping assignment instruction as the target parameter.
The device according to claim 15, wherein the process in which the world state parameter represented by the single second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process or the third type of assignment process; wherein the first The class assignment process indicates direct assignment using external data, the second class assignment process indicates assignment of the operation result after logical operation using external data, and the third class assignment process is used to indicate that no external data is used for assignment;

The risk determination unit is specifically configured to determine whether there is a first parameter among the several target parameters, wherein the operation frequency corresponding to the first parameter is greater than 1, and the process of assigning the first parameter includes a first-type assignment process and the second type of value assignment process; if yes, it is determined that the smart contract has a security risk.
The device according to claim 15, wherein the process in which the world state parameter represented by the single second instruction sequence is assigned a value belongs to the first type of assignment process, the second type of assignment process or the third type of assignment process; wherein the first The class assignment process indicates direct assignment using external data, the second class assignment process indicates the use of external data to perform logical operation results for assignment, and the third class assignment process indicates not using external data for assignment;

The risk determination unit is specifically configured to determine whether there is a second parameter among the several target parameters, wherein the operation frequency corresponding to the second parameter is greater than 1, and the process of assigning the second parameter includes the first type of assignment process and the second type of value assignment process, the first type of value assignment process and the second type of value assignment process of the second parameter correspond to different method functions in the inter-process control flow graph; if yes, determine that the smart contract exists security risk.
The device according to any one of claims 12-21, wherein the smart contract is a smart contract to be deployed in a blockchain; and/or, the security risk specifically includes the risk of being subjected to a rearrangement attack .
A computer-readable storage medium on which a computer program is stored, and when the computer program is executed in a computing device, the computing device executes the method according to any one of claims 1-11.
A computing device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the method according to any one of claims 1-11 is realized.