Pin (computer program)

Last updated
Pin
Developer(s) Intel
Stable release
3.31 / June 30, 2024 (2024-06-30)
Operating system Linux, Windows, macOS
Platform IA-32, x86-64, Intel Xeon Phi
Type Instrumentation framework, Profiler
License Proprietary, Pin is given free of charge for any use under the Intel Simplified Software License .
Website www.intel.com/software/pintool

Pin is a platform for creating analysis tools. A pin tool comprises instrumentation, analysis and callback routines. Instrumentation routines are called when code that has not yet been recompiled is about to be run, and enable the insertion of analysis routines. Analysis routines are called when the code associated with them is run. Callback routines are only called when specific conditions are met, or when a certain event has occurred. Pin provides an extensive application programming interface (API) for instrumentation at different abstraction levels, from one instruction to an entire binary module. It also supports callbacks for many events such as library loads, system calls, signals/exceptions and thread creation events.

Contents

In 2020, it received the Programming Languages Software Award from ACM SIGPLAN. [1]

Pin performs instrumentation by taking control of the program just after it loads into the memory. Then just-in-time recompiles (JIT) small sections of the binary code using pin just before it is run. New instructions to perform analysis are added to the recompiled code. These new instructions come from the Pintool. A large array of optimization techniques are used to obtain the lowest possible running time and memory use overhead. As of June 2010, Pin's average base overhead is 30 percent (without running a pintool). [2]

Features

Instrumentation modes

Pin supports two modes of instrumentation called JIT mode and Probe mode. JIT mode supports all features of Pin, while Probe mode supports a limited feature set but is far faster, adding almost no overhead to program's running time. JIT mode uses a just-in-time compiler to recompile all program code and insert instrumentation, while Probe mode uses code trampolines for instrumentation.

Platform independence

Pin was designed for tool portability, and despite JIT compiling from one ISA to the same ISA (and not using a single intermediate representation for all code), most of its APIs are architecture and operating system independent. It was also designed to be portable itself, carefully isolating platform-specific code from generic code, allowing the fast adaptation of Pin to new platforms. Approximately half of the code is generic and the rest is either architecture or OS dependent. [3]

Optimizations

Pin uses many techniques to optimize instrumentation and analysis code, using techniques such as inlining, liveness analysis and smart register spilling. Pin performs these optimizations automatically whenever possible, without needing users to insert any extra code to allow inlining. Naturally, some optimizations still require user hints, and some code structures are easier to inline than others. Direct linking of jitted code sections, a technique called trace linking, and register binding reconciliation, which minimizes register spilling and remapping, are also used.

Ease of use

Pin’s API and implementation are focused on making pin tools easy to write. Pin takes full responsibility for assuring that the instrumentation code from the pin tool does not affect the application state. Also, the API enables instrumentation code to request many pieces of information from Pin. For example, the instrumentation code in the pin tool can use the Pin API to get the memory address being accessed by an instruction, without having to examine the instruction in detail.

Tools

There are many Pintools that are used for varying tasks.

Alternatives to Pin Tool

There are many other tools available to collect resource usage of running programs on the system such as Bell Lab’s strapon tool and Dyninst tool etc. Bell Lab’s tool uses the strap on technology which runs a tool to collect resources simultaneously with the program but this tool is only compatible with the programs which allow other programs to run simultaneously with them. [6] Furthermore, Dyninst tool uses binary rewriting of the program’s executable and implementable commands inside the program to check for resource usage and is very efficient. However, it is very unstable as it is a relatively new tool and crashes on large scale programs. [7] Lastly, Intel Pin tool uses static binary instrumentation and runs the program as a part of itself while keeping track of all its resources. [8] This approach is more suitable for an antivirus as it can easily run all the processes under itself and can kill programs if they reach a maximum allocated limit as defined by the antivirus.

See also

Notes

  1. "Programming Languages Software Award". www.sigplan.org.
  2. Analyzing Parallel Programs with Pin
  3. Pin: Building Customized Program Analysis Tools with Dynamic Instrumentation
  4. CMP$im: A Pin-Based On-The-Fly Multi-Core Cache Simulator
  5. PinPlay: a framework for deterministic replay and reproducible analysis of parallel programs
  6. Gupta, Chandrashekhar (2007). "Building secure products and solutions. Bell Labs Technical Journal". doi:10.1002/bltj.20247. S2CID   30412754.{{cite journal}}: Cite journal requires |journal= (help)
  7. Lee, Schulz (2007). "Dynamic binary instrumentation and data aggregation on large scale systems". International Journal of Parallel Programming. 35 (3): 207–232. doi:10.1007/s10766-007-0036-3. S2CID   6653468.
  8. Bach, M.; Charney, M.; Cohn, R.; Demikhovsky, E.; Devor, T.; Hazelwood, K.; Jaleel, A.; Luk, Chi-Keung; Lyons, G. (March 2010). "Analyzing Parallel Programs with PIN". Computer. 43 (3): 34–41. doi:10.1109/MC.2010.60. ISSN   0018-9162. S2CID   15155077.

Related Research Articles

<span class="mw-page-title-main">Interpreter (computing)</span> Program that executes source code without a separate compilation step

In computer science, an interpreter is a computer program that directly executes instructions written in a programming or scripting language, without requiring them previously to have been compiled into a machine language program. An interpreter generally uses one of the following strategies for program execution:

  1. Parse the source code and perform its behavior directly;
  2. Translate source code into some efficient intermediate representation or object code and immediately execute that;
  3. Explicitly execute stored precompiled bytecode made by a compiler and matched with the interpreter's virtual machine.

In computer science, dynamic recompilation is a feature of some emulators and virtual machines, where the system may recompile some part of a program during execution. By compiling during execution, the system can tailor the generated code to reflect the program's run-time environment, and potentially produce more efficient code by exploiting information that is not available to a traditional static compiler.

In computing, binary translation is a form of binary recompilation where sequences of instructions are translated from a source instruction set to the target instruction set. In some cases such as instruction set simulation, the target instruction set may be the same as the source instruction set, providing testing and debugging features such as instruction trace, conditional breakpoints and hot spot detection.

In computing, just-in-time (JIT) compilation is compilation during execution of a program rather than before execution. This may consist of source code translation but is more commonly bytecode translation to machine code, which is then executed directly. A system implementing a JIT compiler typically continuously analyses the code being executed and identifies parts of the code where the speedup gained from compilation or recompilation would outweigh the overhead of compiling that code.

In compiler design, static single assignment form is a type of intermediate representation (IR) where each variable is assigned exactly once. SSA is used in most high-quality optimizing compilers for imperative languages, including LLVM, the GNU Compiler Collection, and many commercial compilers.

<span class="mw-page-title-main">Valgrind</span> Programming tool for profiling, memory debugging and memory leak detection

Valgrind is a programming tool for memory debugging, memory leak detection, and profiling.

In compiler theory, dead-code elimination is a compiler optimization to remove dead code. Removing such code has several benefits: it shrinks program size, an important consideration in some contexts, it reduces resource usage such as the number of bytes to be transferred and it allows the running program to avoid executing irrelevant operations, which reduces its running time. It can also enable further optimizations by simplifying program structure. Dead code includes code that can never be executed, and code that only affects dead variables, that is, irrelevant to the program.

SIGPLAN is the Association for Computing Machinery's Special Interest Group on programming languages.

In computer programming, a runtime system or runtime environment is a sub-system that exists both in the computer where a program is created, as well as in the computers where the program is intended to be run. The name comes from the compile time and runtime division from compiled languages, which similarly distinguishes the computer processes involved in the creation of a program (compilation) and its execution in the target machine.

In software engineering, profiling is a form of dynamic program analysis that measures, for example, the space (memory) or time complexity of a program, the usage of particular instructions, or the frequency and duration of function calls. Most commonly, profiling information serves to aid program optimization, and more specifically, performance engineering.

Thread Level Speculation (TLS), also known as Speculative Multi-threading, or Speculative Parallelization, is a technique to speculatively execute a section of computer code that is anticipated to be executed later in parallel with the normal execution on a separate independent thread. Such a speculative thread may need to make assumptions about the values of input variables. If these prove to be invalid, then the portions of the speculative thread that rely on these input variables will need to be discarded and squashed. If the assumptions are correct the program can complete in a shorter time provided the thread was able to be scheduled efficiently.

In computer programming, instrumentation is the act of modifying software so that analysis can be performed on it.

Binary-code compatibility is a property of a computer system, meaning that it can run the same executable code, typically machine code for a general-purpose computer central processing unit (CPU), that another computer system can run. Source-code compatibility, on the other hand, means that recompilation or interpretation is necessary before the program can be run on the compatible system.

Gprof is a performance analysis tool for Unix applications. It used a hybrid of instrumentation and sampling and was created as an extended version of the older "prof" tool. Unlike prof, gprof is capable of limited call graph collecting and printing.

Profile-guided optimization, also known as profile-directed feedback (PDF), and feedback-directed optimization (FDO) is a compiler optimization technique in computer programming that uses profiling to improve program runtime performance.

<span class="mw-page-title-main">Object code optimizer</span> Aspect of software compilation

An object code optimizer, sometimes also known as a post pass optimizer or, for small sections of code, peephole optimizer, forms part of a software compiler. It takes the output from the source language compile step - the object code or binary file - and tries to replace identifiable sections of the code with replacement code that is more algorithmically efficient.

Tracing just-in-time compilation is a technique used by virtual machines to optimize the execution of a program at runtime. This is done by recording a linear sequence of frequently executed operations, compiling them to native machine code and executing them. This is opposed to traditional just-in-time (JIT) compilers that work on a per-method basis.

<span class="mw-page-title-main">DynamoRIO</span> Software framework

DynamoRIO is a BSD-licensed dynamic binary instrumentation framework for the development of dynamic program analysis tools. DynamoRIO targets user space applications under the Android, Linux, and Windows operating systems running on the AArch32, IA-32, and x86-64 instruction set architectures.

<span class="mw-page-title-main">GraalVM</span> Virtual machine software

GraalVM is a Java Development Kit (JDK) written in Java. The open-source distribution of GraalVM is based on OpenJDK, and the enterprise distribution is based on Oracle JDK. As well as just-in-time (JIT) compilation, GraalVM can compile a Java application ahead of time. This allows for faster initialization, greater runtime performance, and decreased resource consumption, but the resulting executable can only run on the platform it was compiled for.

The Computer Language Benchmarks Game is a free software project for comparing how a given subset of simple algorithms can be implemented in various popular programming languages.

References