Data Dependence Profiling for Speculative Optimizations

Data dependence analysis is the foundation to many reordering related compiler optimizations and loop parallelization. Traditional data dependence analysis algorithms are developed primarily for Fortran-like subscripted array variables. They are not very effective for pointer-based references in C or C++. With more advanced hardware support for speculative execution, such as the advanced load instructions in Intel’s IA64 architecture, some data dependences with low probability can be speculatively ignored. However, such speculative optimizations must be carefully applied to avoid excessive cost associated with potential mis-speculations. Data dependence profiling is one way to provide probabilistic information on data dependences to guide such speculative optimizations and speculative thread generation. Software-based data dependence profiling requires detailed tracing of memory accesses, therefore, could be very time consuming. In this paper, we examine issues related to data dependence profiling, and propose various techniques to improve the efficiency of data dependence profiling. We use the Open Research Compiler (ORC) [15,16] to test the efficiency of our data profiling techniques. We also study the effectiveness of data dependence profiling on data speculative optimizations on Itanium systems. Our results show that efficient data dependence profiling could improve the performance for data speculative optimizations.

[1]  J. Larus.  Whole program paths , 1999, PLDI '99.

[2]  William Pugh,et al.  A practical algorithm for exact array dependence analysis , 1992, CACM.

[3]  Ken Kennedy,et al.  Estimating Interlock and Improving Balance for Pipelined Architectures , 1988, J. Parallel Distributed Comput..

[4]  William Pugh,et al.  The Omega test: A fast and practical integer programming algorithm for dependence analysis , 1991, Proceedings of the 1991 ACM/IEEE Conference on Supercomputing (Supercomputing '91).

[5]  Scott A. Mahlke,et al.  Dynamic memory disambiguation using the memory conflict buffer , 1994, ASPLOS VI.

[6]  Laurie J. Hendren,et al.  Is it a tree, a DAG, or a cyclic graph? A shape analysis for heap-directed pointers in C , 1996, POPL '96.

[7]  Raymond Lo,et al.  Partial redundancy elimination in SSA form , 1999, TOPL.

[8]  Thomas Ball,et al.  Edge profiling versus path profiling: the showdown , 1998, POPL '98.

[9]  Michael Hind,et al.  Pointer analysis: haven't we solved this problem yet? , 2001, PASTE '01.

[10]  Carole Dulong,et al.  The IA-64 Architecture at Work , 1998, Computer.

[11]  Kunle Olukotun,et al.  TEST: a Tracer for Extracting Speculative Threads , 2003, International Symposium on Code Generation and Optimization, 2003. CGO 2003..

[12]  Rudolf Eigenmann,et al.  Automatic Parallelization of C by Means of Language Transcription , 1998, LCPC.

[13]  David C. Sehr,et al.  On the importance of points-to analysis and other memory disambiguation methods for C programs , 2001, PLDI '01.

[14]  Alexandru Nicolau,et al.  A general data dependence test for dynamic, pointer-based data structures , 1994, PLDI '94.

[15]  Matthew Arnold,et al.  A framework for reducing the cost of instrumented code , 2001, PLDI '01.

[16]  Jian Huang,et al.  The Superthreaded Processor Architecture , 1999, IEEE Trans. Computers.

[17]  Gurindar S. Sohi,et al.  A programmable co-processor for profiling , 2001, Proceedings HPCA Seventh International Symposium on High-Performance Computer Architecture.

[18]  Utpal Banerjee,et al.  Dependence analysis for supercomputing , 1988, The Kluwer international series in engineering and computer science.

[19]  Jin Lin,et al.  Speculative register promotion using advanced load address table (ALAT) , 2003, International Symposium on Code Generation and Optimization, 2003. CGO 2003..

[20]  Roy Dz-Ching Ju,et al.  Probabilistic memory disambiguation and its application to data speculation , 1999, CARN.

[21]  Roy Dz-Ching Ju,et al.  A compiler framework for speculative analysis and optimizations , 2003, PLDI '03.