1 / 20

Pin : Building Customized Program Analysis Tools with Dynamic Instrumentation

Pin : Building Customized Program Analysis Tools with Dynamic Instrumentation . Chi-Keung Luk, Robert Cohn, Robert Muth, Harish Patil, Artur Klauser, Geoff Lowney, Steven Wallace, Vijay Janapa Reddi, Kim Hazelwood. What is Pin ? . Dynamic binary instrumentation system. JIT compiler.

Télécharger la présentation

Pin : Building Customized Program Analysis Tools with Dynamic Instrumentation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Pin : Building Customized Program Analysis Tools with Dynamic Instrumentation Chi-Keung Luk, Robert Cohn, Robert Muth, Harish Patil, Artur Klauser, Geoff Lowney, Steven Wallace, Vijay Janapa Reddi, Kim Hazelwood

  2. What is Pin ? • Dynamic binary instrumentation system. • JIT compiler. • Pintools for writing instrumentation routines. • Rich API for pintools. • Call-based model of instrumentation.

  3. Design goals • Transparency • Application observes same addresses (code/data) and values (register/memory). • Ease-of-use • Architecture knowledge not required. • Manual inlining of instrumentation instructions not required. • Manual save/restore of architectural state not required. • Portability • Architecture independent API for pintools. • Efficiency • Optimized instrumentation. • Robustness • Handle binaries with mixed code and data. • Handle variable length instructions. • Process attaching/detaching • Support instrumentation at instruction/basic block/routine levels.

  4. System Overview

  5. Instrumentation with Pin • Attach to process using ptrace. • Intercept execution of first/next instruction. • Loop until process terminate or detach from process • Generate new code(trace) for straight-line code sequence starting from instruction. • Insert calls to instrumentation routines into jitted trace. • Trace stored in code cache and executed. • Branche(s) in trace transfer control back to Pin. • Repeat starting with branch target instruction.

  6. Trace code management • Software based cache:<entryIaddr, entrySct> • entryIaddr : original instruction address of trace entry. • entrySct: static context of trace. • Register bindings. • Recent call sites (call stack). • Two traces are compatible if they have same entryIaddr and same entrySct or only register binding differences. • JIT generates new trace only if no compatible trace exists in code cache. • Hash table:<entryIaddr> • Trace entry address. • Trace entry liveness information.

  7. Support for multithreaded applications • Thread local storage for virtual register spilling. • Pin steals physical register(%ebx,%r7) as pointer to spill area. • Application is assumed to be single threaded until thread-create syscall is intercepted. • Spill area accessed using absolute addressing for single threaded application.

  8. Optimized Instrumentation • Trace linking • Register re-allocation • Inlining • X86 eflags liveness analysis • Instruction scheduling

  9. Trace Linking • Branch directly from trace exit to target trace. • Trivial for direct branches but difficult for indirect branches. • Optimization techniques • Target prediction. • Per indirect jump hashtable. • Function cloning for returns using call stack.

  10. Register re-allocation • Obtain registers for JIT without overwriting application’s scratch registers. • Interprocedural register allocation. • Register liveness analysis. • Reconciliation of register bindings.

  11. Other instrumentation optimizations • Inline analysis routines • Avoid call/return to/from bridge routine. • Avoid call/return to/from analysis routine. • Rename caller-save registers, avoid explicit save/restore. • x86 eflags liveness analysis • Avoid save/restore of dead eflags. • Pintool API (IPOINT_ANYWHERE) • Schedule analysis routine to avoid save/restore of eflags.

  12. Experimental Evaluation • IA32, EM64T, Itanium, and ARM ports. • Instrumentation optimizations. • Comparison with Valgrind and DynamoRIO. • Performance without instrumentation. • Performance with basic block counting instrumentation.

  13. Sample Pintools • Opcodemix. • Determine dynamic mix of opcode of execution. • Useful for architectural and compiler comparison studies. • PinPoints. • Automated collection and validation of representative instruction traces.

  14. Questions and Discussions

More Related