CUDA-4.2

Never invoke a non-kernel with a configuration

Required inputs: IR

CUDA 4.2 [kernel.configured_call] Never invoke a non-kernel with a configuration

The postfix expression of configured calls must always refer to a global function.

Scope: Host, Device.

Audience: CUDA C++, CUDA Libraries.

Category: Mandatory.

Hardware Applicability: All Compute Capabilities.

Rationale

Non global functions will not be properly compiled to execute on the device. Calling these functions using the postfix expression of configured calls will result in an incorrect kernel invocation.

Example 1 (Bad)

void not_a_kernel();

int main() {
  auto pfn = &not_a_kernel;
  pfn<<<1,1>>>(); // Non-compliant: configured call does not refer to a
// __global__ function.
}

Excerpt from NVIDIA CUDA C++ Guidelines for robust and safety-critical programming, Version 3.0.1, Copyright (C) 2018-2023 NVIDIA Corporation.

Possible Messages

Key	Text	Severity	Disabled
cuda_kernel_configured_call	Never invoke a non-kernel with a configuration	None	False

Options

• This rule shares the following common options: exclude_in_macros, exclude_messages_in_system_headers, excludes, extend_exclude_to_macro_invocations, includes, justification_checker, languages, post_processing, provider, report_at, severity

• The following places define options that affect this rule: Stylechecks, Analysis-GlobalOptions

• This rule has no individual options.