intrinsics

问题 What I want to do is: Multiply the input floating point number by a fixed factor. Convert them to 8-bit signed char. Note that most of the inputs have a small absolute range of values, like [-6, 6], so that the fixed factor can map them to [-127, 127]. I work on avx2 instruction set only, so intrinsics function like _mm256_cvtepi32_epi8 can't be used. I would like to use _mm256_packs_epi16 but it mixes two inputs together. :( I also wrote some code that converts 32-bit float to 16-bit int,

问题 What is the best practice for swapping __m128i variables? The background is a compile error under Sun Studio 12.2, which is a C++03 compiler. __m128i is an opaque type used with MMX and SSE instructions, and its usually and unsigned long long[2] . C++03 does not provide the support for swapping arrays, and std:swap(__m128i a, __m128i b) fails under the compiler. Here are some related questions that don't quite hit the mark. They don't apply because std::vector is not available. How can we

问题 Can anyone explain what they are and why I would need them? What kind of applications am I building if I need to use intrinsics? 回答1: Normally, "intrinsics" refers to functions that are built-in -- i.e. most standard library functions that the compiler can/will generate inline instead of calling an actual function in the library. For example, a call like: memset(array1, 10, 0) could be compiled for an x86 as something like: mov ecx, 10 xor eax, eax mov edi, offset FLAT:array1 rep stosb

问题 Does Intel C++ compiler and/or GCC support the following intrinsics, like MSVC does since 2012 / 2013? int _rdrand16_step(uint16_t*); int _rdrand32_step(uint32_t*); int _rdrand64_step(uint64_t*); int _rdseed16_step(uint16_t*); int _rdseed32_step(uint32_t*); int _rdseed64_step(uint64_t*); And if these intrinsics are supported, since which version are they supported (with compile-time-constant please)? 回答1: Both GCC and Intel compiler support them. GCC support was introduced at the end of 2010.

问题 Does Intel C++ compiler and/or GCC support the following intrinsics, like MSVC does since 2012 / 2013? int _rdrand16_step(uint16_t*); int _rdrand32_step(uint32_t*); int _rdrand64_step(uint64_t*); int _rdseed16_step(uint16_t*); int _rdseed32_step(uint32_t*); int _rdseed64_step(uint64_t*); And if these intrinsics are supported, since which version are they supported (with compile-time-constant please)? 回答1: Both GCC and Intel compiler support them. GCC support was introduced at the end of 2010.

问题 I really can't get what "keyword" like __m128d is in C++. I'm using MSVC, and it says: The __m128d data type, for use with the Streaming SIMD Extensions 2 instructions intrinsics, is defined in <emmintrin.h> . So, is it a Data Type? typedef ? If I do: #include <emmintrin.h> int main() { __m128d x; } I can't see the defination on <emmintrin.h> . It seems a keyword of compiler? Does it automatically convert that keyword to somethings like "move register xmm0" etc? Or which kind of operation

问题 I'm wondering if Microsofts SSE intrinsics are a little different than the norm because I tried compiling this code with GCC with flags -msse -msse2 -msse3 -msse4 #include <stdio.h> #include <smmintrin.h> int main () { __m128i a, b; a.m128i_u64[0] = 0x000000000000000; b.m128i_u64[0] = 0xFFFFFFFFFFFFFFF; a.m128i_u64[1] = 0x000000000000000; b.m128i_u64[1] = 0x000000000000000; int res1 = _mm_testnzc_si128(a, b); a.m128i_u64[0] = 0x000000000000001; int res2 = _mm_testnzc_si128(a, b); printf_s(

问题 We have the intrinsic _mm_storeu_ps to store __m128 into a float array. However, I don't see any equivalent for integers. I was expecting something like _mm_storeu_epi32 , but that doesn't exist. So, what is the way of storing a _m128i into an int array? 回答1: Its name is _mm_storeu_si128(). 来源： https://stackoverflow.com/questions/43018299/what-is-the-method-of-storing-contents-of-m128i-into-an-int-array

问题 We are testing under Sun Studio 12.3. We are catching a compiler error that's not present under 12.4 and later. Its not present under 12.1 and earlier, but that's because the compiler has trouble with AES instructions. Its also not present under other compilers, like Clang, GCC, ICPC and VC++. The error is: /opt/solarisstudio12.3/bin/CC -DDEBUG -g3 -xO0 -D__SSE2__ -D__SSE3__ -D__SSSE3__ \ -D__SSE4_1__ -D__SSE4_2__ -D__AES__ -D__PCLMUL__ -D__RDRND__ -D__RDSEED__ -D__AVX__ \ -D__AVX2__ -D__BMI_

问题 Here is the test code for broadcasting variable: #include <stdio.h> #include <cuda_runtime.h> __global__ void broadcast(){ int lane_id = threadIdx.x & 0x1f; int value = 31 - lane_id; //let all lanes within the warp be broadcasted the value //whose laneID is 2 less than that of current lane int broadcasted_value = __shfl_up_sync(0xffffffff, value, 2) value = n; printf("thread %d final value = %d\n", threadIdx.x, value); } int main() { broadcast<<<1,32>>>(); cudaDeviceSynchronize(); return 0; }