CWE-440 Base Rascunho

Expected Behavior Violation

This weakness occurs when a software component, such as a function, API, or feature, fails to act as documented or intended. The system's actual behavior deviates from its promised specification,…

Definição

What is CWE-440?

This weakness occurs when a software component, such as a function, API, or feature, fails to act as documented or intended. The system's actual behavior deviates from its promised specification, leading to unpredictable results.
At its core, this violation is a trust issue between the developer and the component's interface. When you call a function or use an API, you rely on its documented contract—what inputs it accepts, what processing it performs, and what outputs or side effects it guarantees. If the component silently breaks this contract, your application logic can fail, security assumptions can be invalidated, and the entire system's stability is compromised. This often stems from ambiguous documentation, implementation bugs, or unintended side effects that the spec didn't account for. For developers, mitigating this requires a proactive approach. First, treat specifications as critical requirements, not suggestions. Implement rigorous input validation and error handling even for 'trusted' components. Second, employ defensive programming practices: write comprehensive unit and integration tests that verify both the happy path and edge cases against the documented behavior. Fuzz testing can be particularly effective in uncovering unexpected behaviors. Finally, when designing your own APIs, ensure your specifications are precise, complete, and tested, as unclear docs are a primary cause of downstream violations.
Impacto no mundo real

Real-world CVEs caused by CWE-440

  • Program uses large timeouts on unconfirmed connections resulting from inconsistency in linked lists implementations.

  • "strncpy" in Linux kernel acts different than libc on x86, leading to expected behavior difference - sort of a multiple interpretation error?

  • Buffer overflow in product stems the use of a third party library function that is expected to have internal protection against overflows, but doesn't.

Como os atacantes a exploram

Trajeto do atacante passo a passo

  1. 1

    The provided code is extracted from the Control and Status Register (CSR), csr_regfile, module within the Hack@DAC'21 OpenPiton System-on-Chip (SoC). This module is designed to implement CSR registers in accordance with the RISC-V specification. The mie (machine interrupt enable) register is a 64-bit register [REF-1384], where bits correspond to different interrupt sources. As the name suggests, mie is a machine-level register that determines which interrupts are enabled. Note that in the example below the mie_q and mie_d registers represent the conceptual mie reigster in the RISC-V specification. The mie_d register is the value to be stored in the mie register while the mie_q register holds the current value of the mie register [REF-1385].

  2. 2

    The mideleg (machine interrupt delegation) register, also 64-bit wide, enables the delegation of specific interrupt sources from machine privilege mode to lower privilege levels. By setting specific bits in the mideleg register, the handling of certain interrupts can be delegated to lower privilege levels without engaging the machine-level privilege mode. For example, in supervisor mode, the mie register is limited to a specific register called the sie (supervisor interrupt enable) register. If delegated, an interrupt becomes visible in the sip (supervisor interrupt pending) register and can be enabled or blocked using the sie register. If no delegation occurs, the related bits in sip and sie are set to zero.

  3. 3

    The sie register value is computed based on the current value of mie register, i.e., mie_q, and the mideleg register.

  4. 4

    The above code snippet illustrates an instance of a vulnerable implementation of the sie register update logic, where users can tamper with the mie_d register value through the utval (user trap value) register. This behavior violates the RISC-V specification.

  5. 5

    The code shows that the value of utval, among other signals, is used in updating the mie_d value within the sie update logic. While utval is a register accessible to users, it should not influence or compromise the integrity of sie. Through manipulation of the utval register, it becomes feasible to manipulate the sie register's value. This opens the door for potential attacks, as an adversary can gain control over or corrupt the sie value. Consequently, such manipulation empowers an attacker to enable or disable critical supervisor-level interrupts, resulting in various security risks such as privilege escalation or denial-of-service attacks.

Exemplo de código vulnerável

Vulnerable Verilog

The sie register value is computed based on the current value of mie register, i.e., mie_q, and the mideleg register.

Vulnerável Verilog
module csr_regfile #(...)(...);
 ...
 // ---------------------------
 // CSR Write and update logic
 // ---------------------------
 ...

```
   if (csr_we) begin
  	 unique case (csr_addr.address)
  	 ...
  		 riscv::CSR_SIE: begin
  			 // the mideleg makes sure only delegate-able register
  			 //(and therefore also only implemented registers) are written
```
mie_d = (mie_q & ~mideleg_q) | (csr_wdata & mideleg_q) | utval_q;** 
  			 end
  		 ...
  		 endcase
  	 end
   endmodule
Exemplo de código seguro

Secure Verilog

A fix to this issue is to remove the utval from the right-hand side of the assignment. That is the value of the mie_d should be updated as shown in the good code example [REF-1386].

Seguro Verilog
module csr_regfile #(...)(...);
 ...
 // ---------------------------
 // CSR Write and update logic
 // ---------------------------
 ...

```
   if (csr_we) begin
  	 unique case (csr_addr.address)
  	 ...
  		 riscv::CSR_SIE: begin
  			 // the mideleg makes sure only delegate-able register
  			 //(and therefore also only implemented registers) are written
```
mie_d = (mie_q & ~mideleg_q) | (csr_wdata & mideleg_q);** 
  			 end
  		 ...
  		 endcase
  	 end
   endmodule
What changed: the unsafe sink is replaced (or the input is validated/escaped) so the same payload no longer triggers the weakness.
Lista de verificação de prevenção

How to prevent CWE-440

  • Architecture Use safe-by-default frameworks and APIs that prevent the unsafe pattern from being expressible.
  • Implementation Validate input at trust boundaries; use allowlists, not denylists.
  • Implementation Apply the principle of least privilege to credentials, file paths, and runtime permissions.
  • Testing Cover this weakness in CI: SAST rules + targeted unit tests for the data flow.
  • Operation Monitor logs for the runtime signals listed in the next section.
Sinais de deteção

How to detect CWE-440

SAST High

Executar análise estática (SAST) na base de código à procura do padrão inseguro no fluxo de dados.

DAST Moderate

Executar testes dinâmicos de segurança de aplicações (DAST) contra o endpoint em execução.

Runtime Moderate

Monitorizar os registos em tempo de execução para traços de exceção invulgares, input malformado ou tentativas de contornar a autorização.

Code review Moderate

Revisão de código: sinalizar qualquer novo código que trate input desta superfície sem usar os ajudantes validados do framework.

Correção automática do Plexicus

O Plexicus deteta automaticamente o CWE-440 e abre um PR de correção em menos de 60 segundos.

O Codex Remedium analisa cada commit, identifica esta fraqueza exata e entrega um pull request pronto para revisão com o patch. Sem tickets. Sem transferências.

Perguntas frequentes

Frequently asked questions

O que é o CWE-440?

This weakness occurs when a software component, such as a function, API, or feature, fails to act as documented or intended. The system's actual behavior deviates from its promised specification, leading to unpredictable results.

Qual a gravidade do CWE-440?

A MITRE não publicou uma classificação de probabilidade de exploração para esta fraqueza. Trate-a como impacto médio até o seu modelo de ameaças provar o contrário.

Que linguagens ou plataformas são afetadas pelo CWE-440?

MITRE lists the following affected platforms: ICS/OT.

Como posso prevenir o CWE-440?

Use safe-by-default frameworks, validate untrusted input at trust boundaries, and apply the principle of least privilege. Cover the data-flow signature in CI with SAST.

Como é que o Plexicus deteta e corrige o CWE-440?

O motor SAST do Plexicus correlaciona a assinatura de fluxo de dados do CWE-440 em cada commit. Quando é encontrada uma correspondência, o nosso agente Codex Remedium abre um PR de correção com o código corrigido, testes e um resumo de uma linha para o revisor.

Onde posso saber mais sobre o CWE-440?

A MITRE publica a definição canónica em https://cwe.mitre.org/data/definitions/440.html. Pode também consultar a documentação da OWASP e do NIST para orientações adjacentes.

Fraquezas relacionadas

Weaknesses related to CWE-440

CWE-684 Pai

Incorrect Provision of Specified Functionality

This weakness occurs when software behaves differently than its documented specifications, which can mislead users and create security…

CWE-1245 Irmão

Improper Finite State Machines (FSMs) in Hardware Logic

This vulnerability occurs when hardware logic contains flawed Finite State Machines (FSMs). Attackers can exploit these design errors to…

CWE-392 Irmão

Missing Report of Error Condition

This vulnerability occurs when a system fails to properly signal that an error has happened. Instead of returning a clear error code,…

CWE-393 Irmão

Return of Wrong Status Code

This vulnerability occurs when a function returns an inaccurate status code or value that misrepresents the actual outcome of an…

CWE-446 Irmão

UI Discrepancy for Security Feature

This vulnerability occurs when a user interface incorrectly displays a security feature as active or properly configured, misleading users…

CWE-451 Irmão

User Interface (UI) Misrepresentation of Critical Information

This vulnerability occurs when a user interface fails to accurately display or highlight crucial information, potentially misleading users…

CWE-912 Irmão

Hidden Functionality

Hidden functionality refers to undocumented features, commands, or code within a product that are not part of its official specification…

CWE-1434 Filho

Insecure Setting of Generative AI/ML Model Inference Parameters

This vulnerability occurs when a generative AI or ML model is deployed with inference parameters that are too permissive, causing it to…

Pronto quando você estiver

Pare de pagar por desenvolvedor.
Comece a fechar o ciclo.

O Plexicus é o ASPM nativo de IA que verifica, filtra, corrige, pentesta e explica — de forma autónoma. Programadores ilimitados, repos ilimitados, ações de IA de utilização justa. Nível gratuito real, €269/mo anual quando estiver pronto.