为什么 NVRTC 不优化我的整数除法和模运算? [英] Why isn't NVRTC optimizing out my integer division and modulo operations?

问题描述

我在 NVRTC 中编译了一个内核:

I compiled a kernel in NVRTC:

__global__ void kernel_A(/* args */) {
    unsigned short idx = threadIdx.x;
    unsigned char warp_id = idx / 32;
    unsigned char lane_id = idx % 32;
    /* ... */
}

我知道整数除法和取模在 CUDA GPU 上非常昂贵.但是我认为这种除以 2 的幂应该优化为位运算，直到我发现它不是:

I know integer division and modulo are very costly on CUDA GPUs. However I thought this kind of division-by-power-of-2 should be optimized into bit operations, until I found it isn't:

__global__ void kernel_B(/* args */) {
    unsigned short idx = threadIdx.x;
    unsigned char warp_id = idx >> 5;
    unsigned char lane_id = idx & 31;
    /* ... */
}

看起来 kernel_B 只是运行得更快.当忽略内核中的所有其他代码时，以 1024 个大小为 1024 的块启动时，nvprof 显示 kernel_A 平均运行 15.2us，而 kernel_B 平均运行 7.4us.我推测 NVRTC 没有优化整数除法和取模.

it seems kernel_B just runs faster. When omitting all other codes in kernel, launching with 1024 blocks of size 1024, nvprof shows kernel_A runs for 15.2us in average, while kernel_B runs 7.4us in average. I speculate NVRTC did not optimize out the integer division and modulo.

结果是在 GeForce 750 Ti、CUDA 8.0 上获得的，平均 100 次调用.赋予 nvrtcCompileProgram() 的编译器选项是 -arch compute_50.

The result is obtained on a GeForce 750 Ti, CUDA 8.0, averaged from 100 calls. The compiler options given to nvrtcCompileProgram() is -arch compute_50.

这是预期的吗?

推荐答案

彻底清除了代码库中的错误.原来我的应用程序是在 DEBUG 模式下构建的.这会导致额外的标志 -G 和 -lineinfo 传递给 nvrtcCompileProgram()

Did a thorough bugsweep in the codebase. Turns out my app was built in DEBUG mode. This causes additional flags -G and -lineinfo passed to nvrtcCompileProgram()

来自 nvcc 手册页:

--device-debug (-G)

为设备代码生成调试信息.关闭所有优化.不要用于分析；改用 -lineinfo.

Generate debug information for device code. Turns off all optimizations. Don't use for profiling; use -lineinfo instead.

这篇关于为什么 NVRTC 不优化我的整数除法和模运算?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！