Filters








73,073 Hits in 4.5 sec

Wish Branches: Enabling Adaptive and Aggressive Predicated Execution

Hyesoon Kim, O. Mutlu, Y.N. Patt, J. Stark
2006 IEEE Micro  
The compiler decides to keep a branch as a conditional branch or to predicate it based on compiletime profile information.  ...  A predicated branch remains predicated for all its dynamic instances even if it turns out to be very easy to predict at runtime.  ...  ., and the Advanced Technology Program of the Texas Higher Education Coordinating Board for supporting our research at the University of Texas at Austin.  ... 
doi:10.1109/mm.2006.27 fatcat:xt4tpnx6b5djdj2qgoyqeypc5m

Control flow optimization for supercomputer scalar processing

Pohua P. Chang, Wen-mei W. Hwu
1989 Proceedings of the 3rd international conference on Supercomputing - ICS '89  
To obtain program control flow information, a system independent profiler has been integrated into the IMPACT-I C compiler.  ...  Based on the weighted control graph, function inline expansion, multi-way branch layout, and software branch prediction can be implemented.  ...  Fundation (NSF) under Grant MIP-8809478, a donation from NCR, the National Aeronautics and Space Administration (NASA) under Contract NASA NAG 1-613 in cooperation with the Illinois Computer laboratory for  ... 
doi:10.1145/318789.318806 dblp:conf/ics/ChangH89 fatcat:trjyclu63nd4joypvcgc3quizy

Profile-assisted Compiler Support for Dynamic Predication in Diverge-Merge Processors

Hyesoon Kim, Jos?e A. Joao, Onur Mutlu, Yale N. Patt
2007 International Symposium on Code Generation and Optimization (CGO'07)  
We describe new profile-driven algorithms and heuristics to select branch instructions that are suitable and profitable for dynamic predication.  ...  However, previous work on dynamic predication did not extensively examine the tradeoffs involved in profiling and code generation for dynamic predication architectures.  ...  Acknowledgments We thank Robert Cohn, Moinuddin Qureshi, Aater Suleman, Mark Oskin, members of the HPS research group, and the anonymous reviewers for their comments and suggestions.  ... 
doi:10.1109/cgo.2007.31 dblp:conf/cgo/KimJMP07 fatcat:hzg23256qfdsnhoml4zeu6hf5u

Dispersing proprietary applications as benchmarks through code mutation

Luk Van Ertvelde, Lieven Eeckhout
2008 ACM SIGOPS Operating Systems Review  
To this end, we compute program slices for memory access operations and/or control flow operations trimmed through constant value and branch profiles; and subsequently mutate the instructions not appearing  ...  Our experimental results using SPEC CPU2000 and MiBench benchmarks show that code mutation is a promising technique that mutates up to 90% of the static binary, up to 50% of the dynamically executed instructions  ...  Lieven Eeckhout is supported by a Postdoctoral Fellowship with the Fund for Scientific Research -Flanders (Belgium).  ... 
doi:10.1145/1353535.1346307 fatcat:ag4khoqmgnfodo6s5d66e3u6ke

Dispersing proprietary applications as benchmarks through code mutation

Luk Van Ertvelde, Lieven Eeckhout
2008 SIGPLAN notices  
To this end, we compute program slices for memory access operations and/or control flow operations trimmed through constant value and branch profiles; and subsequently mutate the instructions not appearing  ...  Our experimental results using SPEC CPU2000 and MiBench benchmarks show that code mutation is a promising technique that mutates up to 90% of the static binary, up to 50% of the dynamically executed instructions  ...  Lieven Eeckhout is supported by a Postdoctoral Fellowship with the Fund for Scientific Research -Flanders (Belgium).  ... 
doi:10.1145/1353536.1346307 fatcat:nts6bauhtjckfla4ezrrho4xmq

Dispersing proprietary applications as benchmarks through code mutation

Luk Van Ertvelde, Lieven Eeckhout
2008 SIGARCH Computer Architecture News  
To this end, we compute program slices for memory access operations and/or control flow operations trimmed through constant value and branch profiles; and subsequently mutate the instructions not appearing  ...  Our experimental results using SPEC CPU2000 and MiBench benchmarks show that code mutation is a promising technique that mutates up to 90% of the static binary, up to 50% of the dynamically executed instructions  ...  Lieven Eeckhout is supported by a Postdoctoral Fellowship with the Fund for Scientific Research -Flanders (Belgium).  ... 
doi:10.1145/1353534.1346307 fatcat:i3htochuhjadrh2tcyo7yd7q7a

Dispersing proprietary applications as benchmarks through code mutation

Luk Van Ertvelde, Lieven Eeckhout
2008 Proceedings of the 13th international conference on Architectural support for programming languages and operating systems - ASPLOS XIII  
To this end, we compute program slices for memory access operations and/or control flow operations trimmed through constant value and branch profiles; and subsequently mutate the instructions not appearing  ...  Our experimental results using SPEC CPU2000 and MiBench benchmarks show that code mutation is a promising technique that mutates up to 90% of the static binary, up to 50% of the dynamically executed instructions  ...  Lieven Eeckhout is supported by a Postdoctoral Fellowship with the Fund for Scientific Research -Flanders (Belgium).  ... 
doi:10.1145/1346281.1346307 dblp:conf/asplos/ErtveldeE08 fatcat:uevr44bagzfejebxlqedfeaz6y

Schnauzer: scalable profiling for likely security bug sites

W. Arthur, B. Mammo, R. Rodriguez, T. Austin, V. Bertacco
2013 Proceedings of the 2013 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)  
In this paper we propose a novel approach to security vulnerability analysis: dynamic control frontier profiling.  ...  Therefore, we develop Schnauzer, a distributed sampling technology to discover the dynamic control frontier, which forms the line of demarcation between dynamically executed and unseen paths.  ...  As such, the conditional branches of an application are preferred to model paths for the dynamic control frontier.  ... 
doi:10.1109/cgo.2013.6494998 dblp:conf/cgo/ArthurMRAB13 fatcat:qbogcaz47rcrxceqkkuyeqfp64

Adaptive scheduling and voltage scaling for multiprocessor real-time applications with non-deterministic workload

Parth Malani, Prakash Mukre, Qinru Qiu, Qing Wu
2008 Proceedings of the conference on Design, automation and test in Europe - DATE '08  
The proposed framework maintains workload awareness using dynamic profiling of branch probability.  ...  (DVFS) for a multiprocessor system.  ...  This low complexity enables the algorithm to be used for dynamic scheduling in a system with the capability of runtime branch prediction.  ... 
doi:10.1145/1403375.1403532 fatcat:z43lxrps4nfiznlilflrggpv7u

Adaptive Scheduling and Voltage Scaling for Multiprocessor Real-time Applications with Non-deterministic Workload

Parth Malani, Prakash Mukre, Qinru Qiu, Qing Wu
2008 2008 Design, Automation and Test in Europe  
The proposed framework maintains workload awareness using dynamic profiling of branch probability.  ...  (DVFS) for a multiprocessor system.  ...  This low complexity enables the algorithm to be used for dynamic scheduling in a system with the capability of runtime branch prediction.  ... 
doi:10.1109/date.2008.4484751 dblp:conf/date/MalaniMQW08 fatcat:ajrhhmeexbgvdoqb4uhckqxuju

Automated Detection of Logical Errors in Programs [chapter]

George Stergiopoulos, Panagiotis Katsaros, Dimitris Gritzalis
2015 Lecture Notes in Computer Science  
In this paper, we present a new code profiling method that combines an information flow analysis, the crosschecking of dynamic invariants with symbolic execution, and the use of fuzzy logic.  ...  Automation can be based on profiling the intended behavior behind the source code.  ...  APP_LogGIC uses Severity ranks and focuses on dynamic invariants that refer to variables used in conditional statements (branch conditions), which are responsible for execution path deviations; if there  ... 
doi:10.1007/978-3-319-17127-2_3 fatcat:diu6nly47bfkvjoepdbjmvyeay

Efficient remote profiling for resource-constrained devices

Priya Nagpurkar, Hussam Mousa, Chandra Krintz, Timothy Sherwood
2006 ACM Transactions on Architecture and Code Optimization (TACO)  
HPS efficiently inserts profiling instructions into an executing program using a novel extension to Dynamic Instruction Stream Editing(DISE).  ...  We compare our system to random and periodic sampling for a number of widely used performance profile types.  ...  ACKNOWLEDGMENT We thank the anonymous reviewers for providing useful comments on this paper. This work was funded in part by Intel, Microsoft, and NSF grants Nos.  ... 
doi:10.1145/1132462.1132465 fatcat:onemupfjfrcdpej7w5c6p2w62u

HPS: hybrid profiling support

H. Mousa, C. Krintz
2005 14th International Conference on Parallel Architectures and Compilation Techniques (PACT'05)  
Key to understanding and optimizing complex applications, is our ability to dynamically monitor executing programs with low overhead and high accuracy.  ...  HPS toggles profiling to sample the executing program as required by the profile consumer, e.g. a dynamic optimizer.  ...  Acknowledgments We thank Marc Corliss, E Lewis, and Amir Roth for providing us with the baseline DISE simulator.  ... 
doi:10.1109/pact.2005.24 dblp:conf/IEEEpact/MousaK05 fatcat:kj2s5g6v35caxechpwei4hnibi

Diverge-Merge Processor: Generalized and Energy-Efficient Dynamic Predication

Hyesoon Kim, Jose A. Joao, Onur Mutlu, Yale N. Patt
2007 IEEE Micro  
the branch instruction from the program and adding the branch condition (that is, predicate) to each of the instructions on both true and false paths following the original branch.  ...  Thus, the processor fetches instructions from both sides of the original branch, but commits results from only the correct side, as determined by the branch condition.  ...  In the Figure 1 example, DMP can predicate Overview of DMP operation The compiler identifies conditional branches with control flow suitable for dynamic predication as diverge branches.  ... 
doi:10.1109/mm.2007.9 fatcat:g2zfwwrfgjfk7fhe7emwxd27iu

Periodic Adaptive Branch Prediction and its Application in Superscalar Processing in Prolog

R.-L. Ma, C.-P. Chung
1995 Computer journal  
Branch instructions create barriers to instruction prefetching, greatly reducing the fine-grained parallelism of programs. Branch prediction is a common method for solving this problem.  ...  We first present four lemmata in this paper describing the relationships among branch prediction hit rate and system performance, hardware efficiency, and branch prediction overhead.  ...  A boundary condition is then used to choose between periodic prediction or dynamic profiling.  ... 
doi:10.1093/comjnl/38.6.457 fatcat:ubiueegspjhk7ohfvvqojehfnu
« Previous Showing results 1 — 15 out of 73,073 results