– Modern architectures provide mechanisms to allow a process to manipulate a portion of its memor... more – Modern architectures provide mechanisms to allow a process to manipulate a portion of its memory as a stack through special registers and instructions. If the portion of memory that represents the stack were kept in a cache hierarchy separate from the traditional data cache, the so-called “stack cache ” could be tuned to the specialized behavior of stack access patterns. We have simulated a simple stack cache within the SimpleScalar framework. For four of the SimpleScalar benchmark applications the stack cache we implemented produced speedups in the range of 1– 4%. 1.
GNU Compiler Collection (GCC) needs a strategy to support future multicore architectures, which w... more GNU Compiler Collection (GCC) needs a strategy to support future multicore architectures, which will probably include heterogeneous accelerator-like designs with explicit management of scratchpad memories. Some have further restrictions; for example, SIMD has limited synchronization capabilities. Some platforms will probably offer hardware support for streaming, transactions, and speculation. The purpose of this paper is to survey and evaluate some automatic and manual techniques for improving support for such targets in GCC. We focus on translation of sequential code for such platforms, i.e., the translation to task graphs and their communication and memory access operations. The paper provides an evaluation of the communication library support on an AMD Phenom T M X4 9550 quad-core processor. We use these experiments to tune the automatic task partitioning algorithm implemented in GCC. The paper concludes with recommendations for strategic developments of GCC to support a stream programming language and improve the automatic generation of streamized tasks.
Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a ... more Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a detailed description of the design and future directions of this infrastructure. Graphite uses the polyhedral model as the internal representation (GPOLY). The plan is to create a polyhedral compilation package (PCP) that will provide loop optimization and analysis capabilities to GCC. This package will be separated from GIMPLE via an interface language that is restricted to express only what GPOLY can represent. The interface language is a set of data structures that encodes the control flow and memory accesses of a code region. A syntax for the language is also defined to facilitate debugging and testing.
ABSTRACT Memory size is an important economic factor in the development of embedded systems. It i... more ABSTRACT Memory size is an important economic factor in the development of embedded systems. It is therefore desirable to find compiler optimization techniques that reduce the size of the generated code. One such technique is procedural abstraction, where repeated occurrences of equivalent code fragments are factored out into new subroutines. Previously, procedural abstraction has been applied at the machine code level in optimizing linkers or binary rewriting tools. We investigate the prospects of applying procedural abstraction at the intermediate code level of a whole-program optimizing C compiler. Placing the optimization before register allocation makes it target indepentdent, and will allow us to take full advantage of powerful code selection, register allocation and code scheduling techniques.
Multi-cores and multi-processors became ubiquitous during the last few years, and the trend is to... more Multi-cores and multi-processors became ubiquitous during the last few years, and the trend is to increase the number of simple, power-efficient, and slower cores per chip. One of the results is that the performance of single-threaded applications did not significantly im- prove, or even declined, on new processors, which heightened the interest in compiler automatic paral- lelization techniques. Our objective
Our objective is to develop a framework in GCC to transform loops into pipelines of concurrent ta... more Our objective is to develop a framework in GCC to transform loops into pipelines of concurrent tasks using streams to communicate and synchronize. This trans-formation can either rely on user hints (pragmas) or on static analysis of control and data dependences. Our fo-cus is ...
Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a ... more Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a detailed description of the design and future directions of this infrastructure. Graphite uses the polyhedral model as the internal representation (GPOLY). The plan is to create a polyhedral compilation package (PCP) that will provide loop optimization and analysis capabilities to GCC. This package will be separated from GIMPLE via an interface language that is restricted to express only what GPOLY can represent. The interface language is a set of data structures that encodes the control flow and memory accesses of a code region. A syntax for the language is also defined to facilitate debugging and testing.
Modern compilers are responsible for adapting the semantics of source programs into a form that m... more Modern compilers are responsible for adapting the semantics of source programs into a form that makes efficient use of a highly complex, heterogeneous machine. This adaptation amounts to solve an optimization problem in a huge and unstructured search space, while predicting the performance outcome of complex sequences of program transformations. The polyhedral model of compilation is aimed at these challenges. Its geometrical, non-inductive semantics enables the construction of better-structured optimization problems and precise analytical models. Recent work demonstrated the scalability of the main polyhedral algorithms to real-world programs. Its integration into production compilers is under way, pioneered by the graphite branch of the GNU Compiler Collection (GCC). Two years after the effective beginning of the project, this paper reports on original questions and innovative solutions that arose during the design and implementation of graphite.
– Modern architectures provide mechanisms to allow a process to manipulate a portion of its memor... more – Modern architectures provide mechanisms to allow a process to manipulate a portion of its memory as a stack through special registers and instructions. If the portion of memory that represents the stack were kept in a cache hierarchy separate from the traditional data cache, the so-called “stack cache ” could be tuned to the specialized behavior of stack access patterns. We have simulated a simple stack cache within the SimpleScalar framework. For four of the SimpleScalar benchmark applications the stack cache we implemented produced speedups in the range of 1– 4%. 1.
GNU Compiler Collection (GCC) needs a strategy to support future multicore architectures, which w... more GNU Compiler Collection (GCC) needs a strategy to support future multicore architectures, which will probably include heterogeneous accelerator-like designs with explicit management of scratchpad memories. Some have further restrictions; for example, SIMD has limited synchronization capabilities. Some platforms will probably offer hardware support for streaming, transactions, and speculation. The purpose of this paper is to survey and evaluate some automatic and manual techniques for improving support for such targets in GCC. We focus on translation of sequential code for such platforms, i.e., the translation to task graphs and their communication and memory access operations. The paper provides an evaluation of the communication library support on an AMD Phenom T M X4 9550 quad-core processor. We use these experiments to tune the automatic task partitioning algorithm implemented in GCC. The paper concludes with recommendations for strategic developments of GCC to support a stream programming language and improve the automatic generation of streamized tasks.
Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a ... more Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a detailed description of the design and future directions of this infrastructure. Graphite uses the polyhedral model as the internal representation (GPOLY). The plan is to create a polyhedral compilation package (PCP) that will provide loop optimization and analysis capabilities to GCC. This package will be separated from GIMPLE via an interface language that is restricted to express only what GPOLY can represent. The interface language is a set of data structures that encodes the control flow and memory accesses of a code region. A syntax for the language is also defined to facilitate debugging and testing.
ABSTRACT Memory size is an important economic factor in the development of embedded systems. It i... more ABSTRACT Memory size is an important economic factor in the development of embedded systems. It is therefore desirable to find compiler optimization techniques that reduce the size of the generated code. One such technique is procedural abstraction, where repeated occurrences of equivalent code fragments are factored out into new subroutines. Previously, procedural abstraction has been applied at the machine code level in optimizing linkers or binary rewriting tools. We investigate the prospects of applying procedural abstraction at the intermediate code level of a whole-program optimizing C compiler. Placing the optimization before register allocation makes it target indepentdent, and will allow us to take full advantage of powerful code selection, register allocation and code scheduling techniques.
Multi-cores and multi-processors became ubiquitous during the last few years, and the trend is to... more Multi-cores and multi-processors became ubiquitous during the last few years, and the trend is to increase the number of simple, power-efficient, and slower cores per chip. One of the results is that the performance of single-threaded applications did not significantly im- prove, or even declined, on new processors, which heightened the interest in compiler automatic paral- lelization techniques. Our objective
Our objective is to develop a framework in GCC to transform loops into pipelines of concurrent ta... more Our objective is to develop a framework in GCC to transform loops into pipelines of concurrent tasks using streams to communicate and synchronize. This trans-formation can either rely on user hints (pragmas) or on static analysis of control and data dependences. Our fo-cus is ...
Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a ... more Graphite is the loop transformation framework that was introduced in GCC 4.4. This paper gives a detailed description of the design and future directions of this infrastructure. Graphite uses the polyhedral model as the internal representation (GPOLY). The plan is to create a polyhedral compilation package (PCP) that will provide loop optimization and analysis capabilities to GCC. This package will be separated from GIMPLE via an interface language that is restricted to express only what GPOLY can represent. The interface language is a set of data structures that encodes the control flow and memory accesses of a code region. A syntax for the language is also defined to facilitate debugging and testing.
Modern compilers are responsible for adapting the semantics of source programs into a form that m... more Modern compilers are responsible for adapting the semantics of source programs into a form that makes efficient use of a highly complex, heterogeneous machine. This adaptation amounts to solve an optimization problem in a huge and unstructured search space, while predicting the performance outcome of complex sequences of program transformations. The polyhedral model of compilation is aimed at these challenges. Its geometrical, non-inductive semantics enables the construction of better-structured optimization problems and precise analytical models. Recent work demonstrated the scalability of the main polyhedral algorithms to real-world programs. Its integration into production compilers is under way, pioneered by the graphite branch of the GNU Compiler Collection (GCC). Two years after the effective beginning of the project, this paper reports on original questions and innovative solutions that arose during the design and implementation of graphite.
Uploads
Papers by Jan Sjodin