convert_to_int32 Class — pytorch Architecture
Architecture documentation for the convert_to_int32 class in vec512_int.h from the pytorch codebase.
Entity Profile
Source Code
aten/src/ATen/cpu/vec/vec512/vec512_int.h lines 1603–1615
template <typename T>
std::
enable_if_t<std::is_same_v<T, int8_t>, Vectorized<int32_t>> inline convert_to_int32(
const int8_t* ptr,
int count = Vectorized<int32_t>::size()) {
if (count == Vectorized<int32_t>::size()) {
return _mm512_cvtepi8_epi32(
_mm_loadu_si128(reinterpret_cast<const __m128i*>(ptr)));
} else {
auto a = Vectorized<int8_t>::loadu(ptr, count);
return _mm512_cvtepi8_epi32(_mm512_castsi512_si128(a));
}
}
Source
Analyze Your Own Codebase
Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.
Try Supermodel Free