Skip to content

Incorrect Usage of cudaDeviceGetLimit for Retrieving Heap Size in interpreter.cu #39

@Shuwei-Song

Description

@Shuwei-Song

Description

Hello 👋,

I am currently studying your code and came across a potential issue in the interpreter.cu file. It appears that the following two lines of code are repeated, both retrieving the stack size instead of one retrieving the heap size:

Location:

  • File: interpreter/interpreter.cu
  • Lines: 32-35
    cudaDeviceGetLimit(&size_value, cudaLimitStackSize);
    printf("current stack size %zu\n", size_value);
    cudaDeviceGetLimit(&size_value, cudaLimitStackSize);
    printf("current heap size %zu\n", size_value);

It seems like the second call to cudaDeviceGetLimit should be retrieving the heap size instead of the stack size. Could you please confirm if this is a mistake?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions