OpenVec Portable SIMD Intrinsics

Souza, Paulo Sérgio Lopes de; Borges, L.; Andreolli, C.; Thierry, Philippe

doi:10.3997/2214-4609.201414038

Cited by 3 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…(6) and (7). Then, we compare two vectors at lines (10) and (11). Finally, we test if two vectors are different at lines (15) and (16).…”

Section: Abstraction Overheadmentioning

confidence: 99%

“…In Table 3, we compare different aspects of Inastemp and two well-known libraries, OpenVec [11] and VCL [10]. All three packages are written on top of intrinsics and intend to separate the algorithms from their vectorization.…”

Section: A Brief Comparison Of Inastemp and Some Related Workmentioning

confidence: 99%

“…(1) // Scalar interaction (2) (4) RealType * restrict C, const size t size){ (5) (6) const int BlockSize = VecType::VecLength; (7) (8) static assert(PanelSizeK >= BlockSize, "PanelSizeK must be greater than block"); (9) static assert(PanelSizeiA >= BlockSize, "PanelSizeiA must be greater than block"); (10) static assert(PanelSizejB >= BlockSize, "PanelSizejB must be greater than block"); (11) for(size t jp = 0 ; jp < size ; jp += PanelSizejB){ (21) (22) for(size t kp = 0 ; kp < size ; kp += PanelSizeK){ (23) (24) alignas ( expected pattern and using SSE or AVX provides a significant speedup; see Figures 4(a), 4(b), 4(c), and 4(d). However, the Clang compiler has not been able to apply the same degree of optimization for the AVX in Float, making the Inastemp-based kernel slower.…”

Section: Natural Exponential Test Casementioning

confidence: 99%

“…The idea of using templates to provide a generic layer above intrinsic functions, as we propose in this paper, is not new and was first proposed a decade ago [3]. It has more recently appeared in dedicated tools [4][5][6][7][8][9][10][11] and in the scope of HPC projects [12,13]. However, this approach has not become widespread because it is incompatible with the two most widely used languages in HPC, which are C and Fortran.…”

Section: Introductionmentioning

confidence: 99%

“…Most of the existing vectorization libraries have performance similar to pure intrinsic-based code, and their differences mainly concern design, features, and code expression. As an example, OpenVec [11] is an open-source library that provides its own data type but chooses at compile time which instruction set to use. It is a C/C++ code that remains at low level and allows abstraction of the hardware in a simple way.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Inastemp: A Novel Intrinsics-as-Template Library for Portable SIMD-Vectorization

Bramas

2017

Scientific Programming

View full text Add to dashboard Cite

The development of scientific applications requires highly optimized computational kernels to benefit from modern hardware. In recent years, vectorization has gained key importance in exploiting the processing capabilities of modern CPUs, whose evolution is characterized by increasing register-widths and core numbers, but stagnating clock frequencies. In particular, vectorization allows floating point operations to be performed at a higher rate than the processor's frequency. However, compilers often fail to vectorize complex codes and pure assembly/intrinsic implementations often suffer from software engineering issues, such as readability and maintainability. Moreover, it is difficult for domain scientists to write optimized code without technical support. To address these issues, we propose Inastemp, a lightweight open-source C++ library. Inastemp offers a solution to develop hardware-independent computational kernels for the CPU. These kernels are portable across compilers and floating point precision and vectorized targeting SSE(3,4.1,4.2), AVX(2), AVX512, or ALTIVEC/VMX instructions. Inastemp provides advanced features, such as an if-else statement that vectorizes branches that cannot be removed. Our performance study shows that Inastemp has the same efficiency as pure intrinsic approaches on modern architectures. As side-results, this study provides micro benchmarks on the latest HPC architectures for three different computational kernels, emphasizing comparisons between scalar and intrinsic-based codes.

show abstract

“…(6) and (7). Then, we compare two vectors at lines (10) and (11). Finally, we test if two vectors are different at lines (15) and (16).…”

Section: Abstraction Overheadmentioning

confidence: 99%

Section: A Brief Comparison Of Inastemp and Some Related Workmentioning

confidence: 99%

Section: Natural Exponential Test Casementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Inastemp: A Novel Intrinsics-as-Template Library for Portable SIMD-Vectorization

Bramas

2017

Scientific Programming

View full text Add to dashboard Cite

show abstract

Autovesk: Automatic Vectorized Code Generation from Unstructured Static Kernels Using Graph Transformations

Tayeb,

Paillat,

Bramas

2023

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Leveraging the SIMD capability of modern CPU architectures is mandatory to take full advantage of their increased performance. To exploit this capability, binary executables must be vectorized, either manually by developers or automatically by a tool. For this reason, the compilation research community has developed several strategies for transforming scalar code into a vectorized implementation. However, most existing automatic vectorization techniques in modern compilers are designed for regular codes, leaving irregular applications with non-contiguous data access patterns at a disadvantage. In this paper, we present a new tool, Autovesk, that automatically generates vectorized code from scalar code, specifically targeting irregular data access patterns. We describe how our method transforms a graph of scalar instructions into a vectorized one, using different heuristics to reduce the number or cost of instructions. Finally, we demonstrate the effectiveness of our approach on various computational kernels using Intel AVX-512 and ARM SVE. We compare the speedups of Autovesk vectorized code over GCC, Clang LLVM and Intel automatic vectorization optimizations. We achieve competitive results on linear kernels and up to 11x speedups on irregular kernels.

show abstract

Auxílio a Portabilidade De Código Em Aplicações De Alto Desempenho

FILHO¹

View full text Add to dashboard Cite

show abstract

OpenVec Portable SIMD Intrinsics

Cited by 3 publications

References 0 publications

Inastemp: A Novel Intrinsics-as-Template Library for Portable SIMD-Vectorization

Inastemp: A Novel Intrinsics-as-Template Library for Portable SIMD-Vectorization

Autovesk: Automatic Vectorized Code Generation from Unstructured Static Kernels Using Graph Transformations

Auxílio a Portabilidade De Código Em Aplicações De Alto Desempenho

Contact Info

Product

Resources

About