The Glorious Glasgow Haskell Compilation System User's Guide, Version 6.4
Table of Contents
- The Glasgow Haskell Compiler License
- 1. Introduction to GHC
- 1.1. Meta-information: Web sites, mailing lists, etc.
- 1.2. Reporting bugs in GHC
- 1.2.1. How do I tell if I should report my bug?
- 1.2.2. What to put in a bug report
- 1.3. GHC version numbering policy
- 1.4. Release notes for version 6.4
- 1.4.1. User-visible compiler changes
- 1.4.2. User-visible interpreter (GHCi) changes
- 1.4.3. User-visible library changes
- 1.4.4. Experimental features
- 1.4.5. Internal changes
- 2. Installing GHC
- 2.1. Installing on Unix-a-likes
- 2.1.1. When a platform-specific package is available
- 2.1.2. GHC binary distributions
- 2.1.2.1. Installing
- 2.1.2.2. What bundles there are
- 2.1.2.3. Testing that GHC seems to be working
- 2.2. Installing on Windows
- 2.2.1. Installing GHC on Windows
- 2.2.2. Moving GHC around
- 2.2.3. Installing ghc-win32 FAQ
- 2.3. The layout of installed files
- 2.3.1. The binary directory
- 2.3.2. The library directory
- 3. Using GHCi
- 3.1. Introduction to GHCi
- 3.2. Loading source files
- 3.2.1. Modules vs. filenames
- 3.2.2. Making changes and recompilation
- 3.3. Loading compiled code
- 3.4. Interactive evaluation at the prompt
- 3.4.1. What's really in scope at the prompt?
- 3.4.1.1. Qualified names
- 3.4.2. Using
do-notation at the prompt - 3.4.3. The
it variable - 3.4.4. Type defaulting in GHCi
- 3.5. Invoking GHCi
- 3.5.1. Packages
- 3.5.2. Extra libraries
- 3.6. GHCi commands
- 3.7. The
:set command - 3.7.1. GHCi options
- 3.7.2. Setting GHC command-line options in GHCi
- 3.8. The
.ghci file - 3.9. FAQ and Things To Watch Out For
- 4. Using GHC
- 4.1. Options overview
- 4.1.1. command-line arguments
- 4.1.2. command line options in source files
- 4.1.3. Setting options in GHCi
- 4.2. Static vs. Dynamic options
- 4.3. Meaningful file suffixes
- 4.4. Modes of operation
- 4.4.1. Using ghc
––make - 4.4.2. Expression evaluation mode
- 4.4.3. Batch compiler mode
- 4.5. Help and verbosity options
- 4.6. Filenames and separate compilation
- 4.6.1. Haskell source files
- 4.6.2. Output files
- 4.6.3. The search path
- 4.6.4. Redirecting the compilation output(s)
- 4.6.5. Keeping Intermediate Files
- 4.6.6. Redirecting temporary files
- 4.6.7. Other options related to interface files
- 4.6.8. The recompilation checker
- 4.6.9. How to compile mutually recursive modules
- 4.6.10. Using make
- 4.6.11. Dependency generation
- 4.6.12. Orphan modules and instance declarations
- 4.7. Warnings and sanity-checking
- 4.8.
Packages
- 4.8.1. Using Packages
- 4.8.2. Package Databases
- 4.8.3. Building a package from Haskell source
- 4.8.4. Package management (the
ghc-pkg command) - 4.8.5.
InstalledPackageInfo: a package specification
- 4.9. Optimisation (code improvement)
- 4.9.1.
-O*: convenient “packages” of optimisation flags. - 4.9.2.
-f*: platform-independent flags
- 4.10. Options related to a particular phase
- 4.10.1. Replacing the program for one or more phases
- 4.10.2. Forcing options to a particular phase
- 4.10.3. Options affecting the C pre-processor
- 4.10.3.1. CPP and string gaps
- 4.10.4. Options affecting a Haskell pre-processor
- 4.10.5. Options affecting the C compiler (if applicable)
- 4.10.6. Options affecting code generation
- 4.10.7. Options affecting linking
- 4.11. Using Concurrent Haskell
- 4.12. Using parallel Haskell
- 4.12.1. Dummy's guide to using PVM
- 4.12.2. parallelism profiles
- 4.12.3. Other useful info about running parallel programs
- 4.12.4. RTS options for Concurrent/parallel Haskell
- 4.13. Platform-specific Flags
- 4.14. Running a compiled program
- 4.14.1. Setting global RTS options
- 4.14.2. RTS options to control the garbage collector
- 4.14.3. RTS options for profiling and Concurrent/Parallel Haskell
- 4.14.4. RTS options for hackers, debuggers, and over-interested
souls
- 4.14.5. “Hooks” to change RTS behaviour
- 4.15. Generating and compiling External Core Files
- 4.16. Debugging the compiler
- 4.16.1. Dumping out compiler intermediate structures
- 4.16.2. Checking for consistency
- 4.16.3. How to read Core syntax (from some
-ddump
flags) - 4.16.4. Unregisterised compilation
- 4.17. Flag reference
- 4.17.1. Help and verbosity options
- 4.17.2. Which phases to run
- 4.17.3. Alternative modes of operation
- 4.17.4. Redirecting output
- 4.17.5. Keeping intermediate files
- 4.17.6. Temporary files
- 4.17.7. Finding imports
- 4.17.8. Interface file options
- 4.17.9. Recompilation checking
- 4.17.10. Interactive-mode options
- 4.17.11. Packages
- 4.17.12. Language options
- 4.17.13. Warnings
- 4.17.14. Optimisation levels
- 4.17.15. Individual optimisations
- 4.17.16. Profiling options
- 4.17.17. Parallelism options
- 4.17.18. C pre-processor options
- 4.17.19. C compiler options
- 4.17.20. Code generation options
- 4.17.21. Linking options
- 4.17.22. Replacing phases
- 4.17.23. Forcing options to particular phases
- 4.17.24. Platform-specific options
- 4.17.25. External core file options
- 4.17.26. Compiler debugging options
- 4.17.27. Misc compiler options
- 5. Profiling
- 5.1. Cost centres and cost-centre stacks
- 5.1.1. Inserting cost centres by hand
- 5.1.2. Rules for attributing costs
- 5.2. Compiler options for profiling
- 5.3. Time and allocation profiling
- 5.4. Profiling memory usage
- 5.4.1. RTS options for heap profiling
- 5.4.2. Retainer Profiling
- 5.4.2.1. Hints for using retainer profiling
- 5.4.3. Biographical Profiling
- 5.5. Graphical time/allocation profile
- 5.6. hp2ps––heap profile to PostScript
- 5.6.1. Manipulating the hp file
- 5.6.2. Zooming in on regions of your profile
- 5.6.3. Viewing the heap profile of a running program
- 5.6.4. Viewing a heap profile in real time
- 5.7. Using “ticky-ticky” profiling (for implementors)
- 6. Advice on: sooner, faster, smaller, thriftier
- 6.1. Sooner: producing a program more quickly
- 6.2. Faster: producing a program that runs quicker
- 6.3. Smaller: producing a program that is smaller
- 6.4. Thriftier: producing a program that gobbles less heap space
- 7. GHC Language Features
- 7.1. Language options
- 7.2. Unboxed types and primitive operations
- 7.2.1. Unboxed types
- 7.2.2. Unboxed Tuples
- 7.3. Syntactic extensions
- 7.3.1. Hierarchical Modules
- 7.3.2. Pattern guards
- 7.3.3. The recursive do-notation
- 7.3.4. Parallel List Comprehensions
- 7.3.5. Rebindable syntax
- 7.4. Type system extensions
- 7.4.1. Data types and type synonyms
- 7.4.1.1. Data types with no constructors
- 7.4.1.2. Infix type constructors, classes, and type variables
- 7.4.1.3. Liberalised type synonyms
- 7.4.1.4. Existentially quantified data constructors
- 7.4.2. Class declarations
- 7.4.2.1. Class method types
- 7.4.3. Type signatures
- 7.4.3.1. The context of a type signature
- 7.4.3.2. For-all hoisting
- 7.4.4. Instance declarations
- 7.4.4.1. Overlapping instances
- 7.4.4.2. Type synonyms in the instance head
- 7.4.4.3. Undecidable instances
- 7.4.5. Implicit parameters
- 7.4.5.1. Implicit-parameter type constraints
- 7.4.5.2. Implicit-parameter bindings
- 7.4.5.3. Implicit parameters and polymorphic recursion
- 7.4.5.4. Implicit parameters and monomorphism
- 7.4.6. Linear implicit parameters
- 7.4.6.1. Warnings
- 7.4.6.2. Recursive functions
- 7.4.7. Functional dependencies
- 7.4.8. Explicitly-kinded quantification
- 7.4.9. Arbitrary-rank polymorphism
- 7.4.9.1. Examples
- 7.4.9.2. Type inference
- 7.4.9.3. Implicit quantification
- 7.4.10. Scoped type variables
- 7.4.10.1. What a scoped type variable means
- 7.4.10.2. Scope and implicit quantification
- 7.4.10.3. Declaration type signatures
- 7.4.10.4. Where a pattern type signature can occur
- 7.4.10.5. Result type signatures
- 7.4.11. Deriving clause for classes
Typeable and Data - 7.4.12. Generalised derived instances for newtypes
- 7.4.12.1. Generalising the deriving clause
- 7.4.12.2. A more precise specification
- 7.5. Generalised Algebraic Data Types
- 7.6. Template Haskell
- 7.6.1. Syntax
- 7.6.2. Using Template Haskell
- 7.6.3. A Template Haskell Worked Example
- 7.7. Arrow notation
- 7.7.1. do-notation for commands
- 7.7.2. Conditional commands
- 7.7.3. Defining your own control structures
- 7.7.4. Primitive constructs
- 7.7.5. Differences with the paper
- 7.7.6. Portability
- 7.8. Assertions
- 7.9. Pragmas
- 7.9.1. DEPRECATED pragma
- 7.9.2. INCLUDE pragma
- 7.9.3. INLINE and NOINLINE pragmas
- 7.9.3.1. INLINE pragma
- 7.9.3.2. NOINLINE pragma
- 7.9.3.3. Phase control
- 7.9.4. LINE pragma
- 7.9.5. OPTIONS_GHC pragma
- 7.9.6. RULES pragma
- 7.9.7. SPECIALIZE pragma
- 7.9.8. SPECIALIZE instance pragma
- 7.9.9. UNPACK pragma
- 7.10. Rewrite rules
- 7.10.1. Syntax
- 7.10.2. Semantics
- 7.10.3. List fusion
- 7.10.4. Specialisation
- 7.10.5. Controlling what's going on
- 7.10.6. CORE pragma
- 7.11. Generic classes
- 7.11.1. Using generics
- 7.11.2. Changes wrt the paper
- 7.11.3. Terminology and restrictions
- 7.11.4. Another example
- 7.12. Concurrent and Parallel Haskell
- 7.12.1. Features specific to Parallel Haskell
- 7.12.1.1. The
Parallel interface (recommended)
- 7.12.1.2. Underlying functions and primitives
- 7.12.1.3. Scheduling policy for concurrent threads
- 7.12.1.4. Scheduling policy for parallel threads
- 8.
Foreign function interface (FFI)
- 8.1. GHC extensions to the FFI Addendum
- 8.1.1. Unboxed types
- 8.2. Using the FFI with GHC
- 8.2.1. Using
foreign export and foreign import ccall "wrapper" with GHC - 8.2.1.1. Using your own
main() - 8.2.1.2. Using
foreign import ccall "wrapper" with GHC
- 8.2.2. Using function headers
- 8.2.2.1. Finding Header files
- 8.2.3. Memory Allocation
- 9. What to do when something goes wrong
- 9.1. When the compiler “does the wrong thing”
- 9.2. When your program “does the wrong thing”
- 10. Other Haskell utility programs
- 10.1. Ctags and Etags for Haskell: hasktags
- 10.1.1. Using tags with your editor
- 10.2. “Yacc for Haskell”: happy
- 10.3. Writing Haskell interfaces to C code:
hsc2hs
- 10.3.1. command line syntax
- 10.3.2. Input syntax
- 10.3.3. Custom constructs
- 11. Running GHC on Win32 systems
- 11.1.
Starting GHC on Win32 platforms
- 11.2.
Interacting with the terminal
- 11.3.
Differences in library behaviour
- 11.4.
Using GHC (and other GHC-compiled executables) with cygwin
- 11.4.1. Background
- 11.4.2. The problem
- 11.4.3. Things to do
- 11.5. Building and using Win32 DLLs
- 11.5.1. Linking with DLLs
- 11.5.2. Not linking with DLLs
- 11.5.3. Creating a DLL
- 11.5.4. Making DLLs to be called from other languages
- 12. Known bugs and infelicities
- 12.1. Haskell 98 vs. Glasgow Haskell: language non-compliance
- 12.1.1. Divergence from Haskell 98
- 12.1.1.1. Lexical syntax
- 12.1.1.2. Context-free syntax
- 12.1.1.3. Expressions and patterns
- 12.1.1.4. Declarations and bindings
- 12.1.1.5. Module system and interface files
- 12.1.1.6. Numbers, basic types, and built-in classes
- 12.1.1.7. In
Prelude support
- 12.1.2. GHC's interpretation of undefined behaviour in
Haskell 98
- 12.2. Known bugs or infelicities
- 12.2.1. Bugs in GHC
- 12.2.2. Bugs in GHCi (the interactive GHC)
- 13. GHC FAQ
- Index