The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Este artigo apresenta técnicas de implementação eficazes e eficientes para eliminação de buffer overflow de DMA no Cell Broadband EngineTM Processador (Célula/BE). No modelo de programação Cell/BE, os desenvolvedores de aplicativos emitem manualmente comandos DMA para transferir dados da memória do sistema para as memórias locais dos núcleos Cell/BE. Embora isso nos permita eliminar falhas de cache ou sobrecarga de invalidação de cache, requer um gerenciamento cuidadoso das matrizes de buffer para DMA nos programas aplicativos para evitar estouros de buffer de DMA. Para se proteger contra estouros de buffer de DMA, introduzimos funções seguras de manipulação de DMA para uso dos aplicativos. Para melhorar e minimizar a sobrecarga de desempenho da prevenção de buffer overflow, usamos três técnicas de otimização diferentes que aproveitam as operações SIMD: otimizações baseadas em dicas de ramificação, otimizações baseadas em tabelas de salto e otimizações baseadas em automodificação. Nossa implementação otimizada evita todos os estouros de buffer DMA com sobrecarga mínima de desempenho, apenas 2.93% de desaceleração média em comparação ao código sem a proteção contra estouro de buffer.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Masana MURASE, "Eliminating Cell Broadband EngineTM DMA Buffer Overflows" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 5, pp. 1062-1069, May 2010, doi: 10.1587/transinf.E93.D.1062.
Abstract: This paper presents effective and efficient implementation techniques for DMA buffer overflow elimination on the Cell Broadband EngineTM (Cell/B.E.) processor. In the Cell/B.E. programming model, application developers manually issue DMA commands to transfer data from the system memory to the local memories of the Cell/B.E. cores. Although this allows us to eliminate cache misses or cache invalidation overhead, it requires careful management of the buffer arrays for DMA in the application programs to prevent DMA buffer overflows. To guard against DMA buffer overflows, we introduced safe DMA handling functions for the applications to use. To improve and minimize the performance overhead of buffer overflow prevention, we used three different optimization techniques that take advantage of SIMD operations: branch-hint-based optimizations, jump-table-based optimizations and self-modifying-based optimizations. Our optimized implementation prevents all DMA buffer overflows with minimal performance overhead, only 2.93% average slowdown in comparison to code without the buffer overflow protection.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.1062/_p
Copiar
@ARTICLE{e93-d_5_1062,
author={Masana MURASE, },
journal={IEICE TRANSACTIONS on Information},
title={Eliminating Cell Broadband EngineTM DMA Buffer Overflows},
year={2010},
volume={E93-D},
number={5},
pages={1062-1069},
abstract={This paper presents effective and efficient implementation techniques for DMA buffer overflow elimination on the Cell Broadband EngineTM (Cell/B.E.) processor. In the Cell/B.E. programming model, application developers manually issue DMA commands to transfer data from the system memory to the local memories of the Cell/B.E. cores. Although this allows us to eliminate cache misses or cache invalidation overhead, it requires careful management of the buffer arrays for DMA in the application programs to prevent DMA buffer overflows. To guard against DMA buffer overflows, we introduced safe DMA handling functions for the applications to use. To improve and minimize the performance overhead of buffer overflow prevention, we used three different optimization techniques that take advantage of SIMD operations: branch-hint-based optimizations, jump-table-based optimizations and self-modifying-based optimizations. Our optimized implementation prevents all DMA buffer overflows with minimal performance overhead, only 2.93% average slowdown in comparison to code without the buffer overflow protection.},
keywords={},
doi={10.1587/transinf.E93.D.1062},
ISSN={1745-1361},
month={May},}
Copiar
TY - JOUR
TI - Eliminating Cell Broadband EngineTM DMA Buffer Overflows
T2 - IEICE TRANSACTIONS on Information
SP - 1062
EP - 1069
AU - Masana MURASE
PY - 2010
DO - 10.1587/transinf.E93.D.1062
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2010
AB - This paper presents effective and efficient implementation techniques for DMA buffer overflow elimination on the Cell Broadband EngineTM (Cell/B.E.) processor. In the Cell/B.E. programming model, application developers manually issue DMA commands to transfer data from the system memory to the local memories of the Cell/B.E. cores. Although this allows us to eliminate cache misses or cache invalidation overhead, it requires careful management of the buffer arrays for DMA in the application programs to prevent DMA buffer overflows. To guard against DMA buffer overflows, we introduced safe DMA handling functions for the applications to use. To improve and minimize the performance overhead of buffer overflow prevention, we used three different optimization techniques that take advantage of SIMD operations: branch-hint-based optimizations, jump-table-based optimizations and self-modifying-based optimizations. Our optimized implementation prevents all DMA buffer overflows with minimal performance overhead, only 2.93% average slowdown in comparison to code without the buffer overflow protection.
ER -