Language processors (Chapter 2) 1 Course Overview PART I: overview material 1Introduction 2Language processors (tombstone diagrams, bootstrapping) 3Architecture of a compiler PART II: inside a compiler 4Syntax analysis 5Contextual analysis 6Runtime organization 7Code generation PART III: conclusion 8Interpretation 9Review

Language processors (Chapter 2) 2 Chapter 2 Language Processors RECAP –Different levels of programming languages low-level high-level –different language processors to bridge the semantic gap GOAL this lecture: –A high level understanding of what language processors are. what do they do? what kinds are there? Topics: –translators (compilers, assemblers,...) –tombstone diagrams –bootstrapping

Language processors (Chapter 2) 3 Compilers and other translators Examples: Chinese => English Java => JVM byte codes Scheme => C C => Scheme x86 Assembly Language => x86 binary codes Other non-traditional examples: disassembler, decompiler (e.g. JVM => Java)

Language processors (Chapter 2) 4 Assemblers versus Compilers An assembler and a compiler both translate a “more high level” language into a “more low-level” one. C x86 assembly x86 executable binary Scheme More high level More low level Scheme (to C) compiler C compiler x86 assembler

Language processors (Chapter 2) 5 Assemblers versus Compilers Q: What is the difference between a compiler and an assembler? Assembler: very direct mapping (almost 1 to 1 mapping) from the source language onto the target language. Compiler: more complex “conceptual” mapping from a HL language to a LL language. 1 construct in the HL language => many instructions in the LL language

Language processors (Chapter 2) 6 Terminology Translatorinputoutput source program object program is expressed in the source language is expressed in the implementation language is expressed in the target language Q: Which programming languages play a role in this picture?

Language processors (Chapter 2) 7 Tombstone Diagrams What are they? –diagrams consist of a set of “puzzle pieces” we can use to reason about language processors and programs –different kinds of pieces –combination rules (not all diagrams are “well formed”) M Machine implemented in hardware S --> T L Translator implemented in L MLML Language interpreter in L Program P implemented in L L P

Language processors (Chapter 2) 8 Tombstone diagrams: Combination rules S PP T S --> T M M L P M WRONG!OK! M M P M L P WRONG!

Language processors (Chapter 2) 9 Tetris x86C Tetris Compilation x86 Example: Compilation of C programs on an x86 machine C --> x86 x86 Tetris x86

Language processors (Chapter 2) 10 Tetris PPCC Tetris Cross compilation x86 Example: A C “cross compiler” from x86 to PPC C --> PPC x86 A cross compiler is a compiler which runs on one machine (the host machine) but emits code for another machine (the target machine). Host ≠ Target Q: Are cross compilers useful? Why would/could we use them? PPC Tetris PPC download

Language processors (Chapter 2) 11 Tetris x86 Tetris JVMJava Tetris Two Stage Compilation x86 Java-->JVM x86 A two-stage translator is a composition of two translators. The output of the first translator is provided as input to the second translator. x86 JVM-->x86 x86

Language processors (Chapter 2) 12 x86 Java-->x86 Compiling a Compiler Observation: A compiler is a program! Therefore it can be provided as input to a language processor. Example: compiling a compiler. Java-->x86 C ++ x86 C ++ -->x86 x86

Language processors (Chapter 2) 13 Interpreters An interpreter is a language processor implemented in software, i.e. as a program. Terminology: abstract (or virtual) machine versus real machine Example: The Java Virtual Machine JVM x86 JVM Tetris Q: Why are abstract machines useful?

Language processors (Chapter 2) 14 Interpreters Q: Why are abstract machines useful? 1) Abstract machines provide better platform independence JVM x86 PPC JVM Tetris JVM PPC JVM Tetris

Language processors (Chapter 2) 15 Interpreters Q: Why are abstract machines useful? 2) Abstract machines are useful for testing and debugging. Example: Testing the “Ultima” processor using hardware emulation Ultima x86 Ultima  P P Functional equivalence Note: we don’t have to implement Ultima emulator directly in x86; we can use a high-level language and compile it to x86.

Language processors (Chapter 2) 16 Interpreters versus Compilers Q: What are the tradeoffs between compilation and interpretation? Compilers typically offer more advantages when –programs are deployed in a production setting –programs are “repetitive” –the instructions of the programming language are complex Interpreters typically are a better choice when –we are in a development/testing/debugging stage –programs are run once and then discarded –the instructions of the language are simple

Language processors (Chapter 2) 17 Interpretive Compilers Why? A tradeoff between fast(er) compilation and a reasonable runtime performance. How? Use an “intermediate language” more high-level than machine code => easier to compile to more low-level than source language => easier to implement as an interpreter Example: A “Java Development Kit” for machine M Java-->JVM M JVM M

Language processors (Chapter 2) 18 P JVMJava P Interpretive Compilers Example: Here is how to use a “Java Development Kit” to run a Java program P Java-->JVM M JVM M M JVM P M

Language processors (Chapter 2) 19 Portable Compilers Example: Two different “Java Development Kits” Java-->JVM JVM M Kit 2: Java-->JVM M JVM M Kit 1: Q: Which one is “more portable”?

Language processors (Chapter 2) 20 Portable Compilers In the previous example we have seen that portability is not an “all or nothing” kind of deal. It is useful to talk about a “degree of portability” as the percentage of code that needs to be re-written when moving to a dissimilar machine. In practice 100% portability is not possible.

Language processors (Chapter 2) 21 Example: a “portable” compiler kit Java-->JVM Java JVM Java Java-->JVM JVM Q: Suppose we want to run this kit on some machine M. How could we go about realizing that goal? (with the least amount of effort) Portable Compiler Kit:

Language processors (Chapter 2) 22 Example: a “portable” compiler kit Java-->JVM Java JVM Java Java-->JVM JVM Q: Suppose we want to run this kit on our some machine M. How could we go about realizing that goal? (with the least amount of effort) JVM Java JVM C ++ reimplement C ++ -->M M JVM M M

Language processors (Chapter 2) 23 Example: a “portable” compiler kit Java-->JVM Java JVM Java Java-->JVM JVM M This is what we have now: Now, how do we run our Tetris program? Tetris JVMJava Tetris M Java-->JVM JVM M JVM Tetris JVM M M

Language processors (Chapter 2) 24 Bootstrapping Java-->JVM Java JVM Java Java-->JVM JVM Remember our “portable compiler kit”: We haven’t used this yet! Java-->JVM Java Same language! Q: What can we do with a compiler written in itself? Is that useful at all? JVM M

Language processors (Chapter 2) 25 Bootstrapping Java-->JVM Java Same language! Q: What can we do with a compiler written in itself? Is that useful at all? By implementing the compiler in (a subset of) its own language, we become less dependent on the target platform => more portable implementation. But… “chicken and egg problem”? How can we get around that? => BOOTSTRAPPING: requires some work to make the first “egg”. There are many possible variations on how to bootstrap a compiler written in its own language.

Language processors (Chapter 2) 26 Bootstrapping an Interpretive Compiler to Generate M code Java-->JVM Java JVM Java Java-->JVM JVM Our “portable compiler kit”: P M Java P Goal: we want to get a “completely native” Java compiler on machine M Java-->M M JVM M M

Language processors (Chapter 2) 27 P M P MJava P Bootstrapping an Interpretive Compiler to Generate M code Idea: we will build a two-stage Java --> M compiler. P JVM M Java-->JVM M M JVM-->M M We will make this by compiling To get this we implement JVM-->M Java Java-->JVM JVM and compile it

Language processors (Chapter 2) 28 Bootstrapping an Interpretive Compiler to Generate M code Step 1: implement JVM-->M JavaJVM JVM-->M Java-->JVM JVM M M JVM-->M Java Step 2: compile it Step 3: compile this

Language processors (Chapter 2) 29 Bootstrapping an Interpretive Compiler to Generate M code Step 3: “Self compile” the JVM (in JVM) compiler M JVM-->M JVM M M JVM-->M JVM JVM-->M JVMThis is the second stage of our compiler! Step 4: use this to compile the Java compiler

Language processors (Chapter 2) 30 Bootstrapping an Interpretive Compiler to Generate M code Step 4: Compile the Java-->JVM compiler into machine code M Java-->JVM M JVM JVM-->M MThe first stage of our compiler! We are DONE!

Language processors (Chapter 2) 31 Full Bootstrap A full bootstrap is necessary when we are building a new compiler from scratch. Example: We want to implement an Ada compiler for machine M. We don’t currently have access to any Ada compiler (not on M, nor on any other machine). Idea: Ada is very large, we will implement the compiler in a subset of Ada and bootstrap it from a subset of Ada compiler in another language. (e.g. C) Ada-S-->M C v1 Step 1a: Build a compiler for Ada-S in another language

Language processors (Chapter 2) 32 Full Bootstrap Ada-S-->M C v1 Step 1a: Build a compiler (v1) for Ada-S in another language. Ada-S-->M C v1 M Ada-S-->M v1 Step 1b: Compile v1 compiler on M M C-->M M This compiler can be used for bootstrapping on machine M but we do not want to rely on it permanently!

Language processors (Chapter 2) 33 Full Bootstrap Ada-S-->M Ada-S v2 Step 2a: Implement v2 of Ada-S compiler in Ada-S Ada-S-->M Ada-S v2 M M Ada-S-->M v2 Step 2b: Compile v2 compiler with v1 compiler Ada-S-->M M v1 Q: Is it hard to rewrite the compiler in Ada-S? We are now no longer dependent on the availability of a C compiler!

Language processors (Chapter 2) 34 Full Bootstrap Step 3a: Build a full Ada compiler in Ada-S Ada-->M Ada-S v3 M M Ada-->M v3 Ada-S-->M M v2 Step 3b: Compile with v2 compiler Ada-->M Ada-S v3 From this point on we can maintain the compiler in Ada. Subsequent versions v4,v5,... of the compiler in Ada and compile each with the the previous version.

Language processors (Chapter 2) 35 Half Bootstrap We discussed full bootstrap which is required when we have no access to a compiler for our language at all. Q: What if we have access to a compiler for our language on a different machine HM (host machine) but want to develop one for TM (target machine)? Ada-->HM HM We have: Ada-->TM TM We want: Idea: We can use cross compilation from HM to TM to bootstrap the TM compiler. Ada-->HM Ada

Language processors (Chapter 2) 36 HM Ada-->TM Half Bootstrap Idea: We can use cross compilation from HM to TM to bootstrap the TM compiler. Step 1: Implement Ada-->TM compiler in Ada Ada-->TM Ada Step 2: Compile on HM Ada-->TM Ada Ada-->HM HM Cross compiler: running on HM but emits TM code

Language processors (Chapter 2) 37 TM Ada-->TM Half Bootstrap Step 3: Cross compile our Ada-->TM compiler. Ada-->TM Ada Ada-->TM HM DONE! From now on we can develop subsequent versions of the compiler completely on TM

Language processors (Chapter 2) 38 Bootstrapping to Improve Efficiency The efficiency of programs and compilers: Efficiency of programs: - memory usage - runtime Efficiency of compilers: - Efficiency of the compiler itself - Efficiency of the emitted code Idea: We start from a simple compiler (generating inefficient code) and develop more sophisticated version of it. We can then use bootstrapping to improve performance of the compiler.

Language processors (Chapter 2) 39 Bootstrapping to Improve Efficiency We have: Ada-->M slow Ada Ada-->M slow M slow We implement: Ada-->M fast Ada Ada-->M fast Ada M Ada-->M fast M slow Step 1 Ada-->M slow M slow Step 2 Ada-->M fast Ada M Ada-->M fast M fast Ada-->M fast M slow Fast compiler that emits fast code!