583L9 - EECS 583 Class 9 Classic and ILP Optimization...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
EECS 583 – Class 9 Classic and ILP Optimization University of Michigan October 5, 2011
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
- 1 - Announcements & Reading Material Look on Phorum for help with HW #2 » Daya’s post on adding/deleting instructions » Splitting a BB Today’s class » “Compiler Code Transformations for Superscalar-Based High- Performance Systems,” Scott Mahlke, William Chen, John Gyllenhaal, Wen-mei Hwu, Pohua Chang, and Tokuzo Kiyohara, Proceedings of Supercomputing '92 , Nov. 1992, pp. 808-817 Next class (instruction scheduling) » “Machine Description Driven Compilers for EPIC Processors”, B. Rau, V. Kathail, and S. Aditya, HP Technical Report, HPL- 98-40, 1998. (long paper but informative)
Background image of page 2
- 2 - HW#2 – Best Times from Last Semester Times in seconds for each application » no spec LICM with spec LICM » Perf1 .008453 .000095 » Perf2 .023796 .000133 » Perf3 .012391 .000099 » 583wc .002813 .001644 This translates to: 89x, 178x and 125x and 1.7x speedup Note: These generated with an older LLVM » But, gives you an idea of what to expect.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
- 3 - Course Project – Time to Start Thinking About This Mission statement: Design and implement something “interesting” in a compiler » LLVM preferred, but others are fine » Groups of 2-3 people (1 or 4 persons is possible in some cases) » Extend existing research paper or go out on your own Topic areas » Automatic parallelization/SIMDization » Optimizing for GPUs » Targeting VLIW processors » Memory system optimization » Reliability » Energy » For the adventurous Dynamic optimization Streaming applications Binary optimization
Background image of page 4
- 4 - Course Projects – Timetable Now » Start thinking about potential topics, identify teammates Oct 24-28: Project proposals » No class that week » Daya/I will meet with each group » Informal proposal discussed at meeting » Written proposal (a paragraph or 2 plus some references) due Oct 31 Last 2 wks of Nov: Project checkpoint » Class as usual » Daya/I will again meet with each group » Update on your progress, what left to do Dec 13-16: Project demos
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
- 5 - Sample Project Ideas (in random order) Memory system » Cache profiler for LLVM IR – miss rates, stride determination » Data cache prefetching » Data layout for improved cache behavior Optimizing for GPUs » Dumb OpenCL/CUDA smart OpenCL/CUDA – selection of threads/blocks and managing on-chip memory » Reducing uncoalesced memory accesses – measurement of uncoalesced accesses, code restructuring to reduce these » Matlab CUDA/OpenCL » StreamIt to CUDA/OpenCL (single or multiple GPUs) Reliability » AVF profiling, reducing vulnerability to soft errors » Coarse-grain code duplication for soft error protection
Background image of page 6
- 6 - More Project Ideas Parallelization/SIMDization » CUDA/OpenCL/Matlab/OpenMP to x86+SSE or Arm+NEON » DOALL loop parallelization, dependence breaking transformations » Stream parallelism analysis tool - profiling tool to identify maximal
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 29

583L9 - EECS 583 Class 9 Classic and ILP Optimization...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online