14.1. Haskell standards vs. Glasgow Haskell: language non-compliance
This section lists Glasgow Haskell infelicities in its implementation of Haskell 98 and Haskell 2010. See also the “when things go wrong” section (What to do when something goes wrong) for information about crashes, space leaks, and other undesirable phenomena.
The limitations here are listed in Haskell Report order (roughly).
14.1.1. Divergence from Haskell 98 and Haskell 2010
GHC aims to be able to behave (mostly) like a Haskell 98 or Haskell 2010 compiler, if you tell it to try to behave like that with the Haskell98
and Haskell2010
flags. The known deviations from the standards are described below. Unless otherwise stated, the deviation applies in both Haskell 98 and Haskell 2010 mode.
14.1.1.1. Lexical syntax
- Certain lexical rules regarding qualified identifiers are slightly different in GHC compared to the Haskell report. When you have ⟨module⟩
.
⟨reservedop⟩, such asM.\
, GHC will interpret it as a single qualified operator rather than the two lexemesM
and.\
. forall
is always a reserved keyword at the type level, contrary to the Haskell Report, which allows type variables to be namedforall
. Note that this does not imply that GHC always enables theExplicitForAll
extension. Even without this extension enabled, reservingforall
as a keyword has significance. For instance, GHC will not parse the type signaturefoo :: forall x
.- The
(!)
operator, when written in prefix form (preceded by whitespace and not followed by whitespace, as inf !x = ...
), is interpreted as a bang pattern, contrary to the Haskell Report, which prescribes to treat!
as an operator regardless of surrounding whitespace. Note that this does not imply that GHC always enablesBangPatterns
. Without the extension, GHC will issue a parse error onf !x
, asking to enable the extension. Irrefutable patterns must be written in prefix form:
f ~a ~b = ... -- accepted by both GHC and the Haskell Report f ~ a ~ b = ... -- accepted by the Haskell Report but not GHC
When written in non-prefix form,
(~)
is treated by GHC as a regular infix operator.See GHC Proposal #229 for the precise rules.
Strictness annotations in data declarations must be written in prefix form:
data T = MkT !Int -- accepted by both GHC and the Haskell Report data T = MkT ! Int -- accepted by the Haskell Report but not GHC
See GHC Proposal #229 for the precise rules.
As-patterns must not be surrounded by whitespace on either side:
f p@(x, y, z) = ... -- accepted by both GHC and the Haskell Report -- accepted by the Haskell Report but not GHC: f p @ (x, y, z) = ... f p @(x, y, z) = ... f p@ (x, y, z) = ...
When surrounded by whitespace on both sides,
(@)
is treated by GHC as a regular infix operator.When preceded but not followed by whitespace,
(@)
is treated as a visible type application.See GHC Proposal #229 for the precise rules.
Haskell Report allows any Unicode Decimal Number in decimal literals. However, GHC accepts only ASCII numbers:
ascDigit → 0 | 1 | … | 9 decimal → ascDigit {ascDigit}
GHC is more lenient in which characters are allowed in the identifiers. Unicode Other Letters are considered to be small letters, therefore variable identifiers can begin with them. Digit class contains all Unicode numbers instead of just Decimal Numbers. Modifier Letters and Non-Spacing Marks can appear in the tail of the identifiers.:
uniSmall → any Unicode Lowercase Letter or Other Letter uniDigit → any Unicode Decimal Number, Letter Number or Other Number uniIdchar → any Unicode Modifier Letter or Non-Spacing Mark idchar → small | large | digit | uniIdchar | ' varid → small {idchar} ⟨reservedid⟩ conid → large {idchar}
14.1.1.2. Context-free syntax
In Haskell 98 mode (but not in Haskell 2010 mode), GHC is a little less strict about the layout rule when used in
do
expressions. Specifically, the restriction that “a nested context must be indented further to the right than the enclosing context” is relaxed to allow the nested context to be at the same level as the enclosing context, if the enclosing context is ado
expression.For example, the following code is accepted by GHC:
main = do args <- getArgs if null args then return [] else do ps <- mapM process args mapM print ps
This behaviour is controlled by the
NondecreasingIndentation
extension.
NondecreasingIndentation
-
Since: 7.2.1 Allow nested contexts to be at the same indentation level as its enclosing context.
GHC doesn’t do the fixity resolution in expressions during parsing as required by Haskell 98 (but not by Haskell 2010). For example, according to the Haskell 98 report, the following expression is legal:
let x = 42 in x == 42 == True
and parses as:
(let x = 42 in x == 42) == True
because according to the report, the
let
expression “extends as far to the right as possible”. Since it can’t extend past the second equals sign without causing a parse error (==
is non-fix), thelet
-expression must terminate there. GHC simply gobbles up the whole expression, parsing like this:(let x = 42 in x == 42 == True)
14.1.1.3. Expressions and patterns
By default, GHC makes some programs slightly more defined than they should be. For example, consider
f :: [a] -> b -> b
f [] = error "urk"
f (x:xs) = \v -> v
main = print (f [] `seq` True)
This should call error
but actually prints True
. Reason: GHC eta-expands f
to
f :: [a] -> b -> b
f [] v = error "urk"
f (x:xs) v = v
For most programs this improves efficiency enough to be enabled & bad only in few rare cases. To suppress this optimisation use -fpedantic-bottoms
.
14.1.1.4. Failable patterns
Since the MonadFail Proposal (MFP), do-notation blocks that contain a failable pattern need a MonadFail constraint.
For example
mayFail :: (MonadIO m) => m ()
mayFail = do
(Just value) <- fetchData
putStrLn value
Will warn you with
• Could not deduce (MonadFail m)
arising from a do statement
with the failable pattern ‘(Just x)’
from the context: MonadIO m
bound by the type signature for:
mayFail :: forall (m :: * -> *). MonadIO m => m ()
And indeed, since the Monad class does not have the fail
method anymore, we need to explicitly add (MonadFail m)
to the constraints of the function.
14.1.1.5. Typechecking of recursive binding groups
The Haskell Report specifies that a group of bindings (at top level, or in a let
or where
) should be sorted into strongly-connected components, and then type-checked in dependency order (Haskell Report, Section 4.5.1). As each group is type-checked, any binders of the group that have an explicit type signature are put in the type environment with the specified polymorphic type, and all others are monomorphic until the group is generalised (Haskell Report, Section 4.5.2).
Following a suggestion of Mark Jones, in his paper Typing Haskell in Haskell, GHC implements a more general scheme. In GHC the dependency analysis ignores references to variables that have an explicit type signature. As a result of this refined dependency analysis, the dependency groups are smaller, and more bindings will typecheck. For example, consider:
f :: Eq a => a -> Bool
f x = (x == x) || g True || g "Yes"
g y = (y <= y) || f True
This is rejected by Haskell 98, but under Jones’s scheme the definition for g
is typechecked first, separately from that for f
, because the reference to f
in g
’s right hand side is ignored by the dependency analysis. Then g
’s type is generalised, to get
g :: Ord a => a -> Bool
Now, the definition for f
is typechecked, with this type for g
in the type environment.
The same refined dependency analysis also allows the type signatures of mutually-recursive functions to have different contexts, something that is illegal in Haskell 98 (Section 4.5.2, last sentence). GHC only insists that the type signatures of a refined group have identical type signatures; in practice this means that only variables bound by the same pattern binding must have the same context. For example, this is fine:
f :: Eq a => a -> Bool
f x = (x == x) || g True
g :: Ord a => a -> Bool
g y = (y <= y) || f True
14.1.1.6. Default Module headers with -main-is
The Haskell2010 Report specifies in <https://www.haskell.org/onlinereport/haskell2010/haskellch5.html#x11-990005.1> that
- “An abbreviated form of module, consisting only of the module body,
-
is permitted. If this is used, the header is assumed to be
module Main(main) where
.”
GHC’s -main-is
option can be used to change the name of the top-level entry point from main
to any other variable. When compiling the main module and -main-is
has been used to rename the default entry point, GHC will also use the alternate name in the default export list.
Consider the following program:
-- file: Main.hs
program :: IO ()
program = return ()
GHC will successfully compile this module with ghc -main-is Main.program Main.hs
, because the default export list will include program
rather than main
, as the Haskell Report typically requires.
This change only applies to the main module. Other modules will still export main
from a default export list, regardless of the -main-is
flag. This allows use of -main-is
with existing modules that export main
via a default export list, even when -main-is
points to a different entry point, as in this example (compiled with -main-is MainWrapper.program
).
-- file MainWrapper.hs
module MainWrapper where
import Main
program :: IO ()
program = putStrLn "Redirecting..." >> main
-- file Main.hs
main :: IO ()
main = putStrLn "I am main."
14.1.1.7. Module system and interface files
GHC requires the use of hs-boot
files to cut the recursive loops among mutually recursive modules as described in How to compile mutually recursive modules. This more of an infelicity than a bug: the Haskell Report says (Section 5.7)
14.1.1.8. Numbers, basic types, and built-in classes
Num superclasses
-
The
Num
class does not haveShow
orEq
superclasses.You can make code that works with both Haskell98/Haskell2010 and GHC by:
- Whenever you make a
Num
instance of a type, also makeShow
andEq
instances, and - Whenever you give a function, instance or class a
Num t
constraint, also give itShow t
andEq t
constraints.
- Whenever you make a
Bits superclass
-
The
Bits
class does not have aNum
superclass. It therefore does not have default methods for thebit
,testBit
andpopCount
methods.You can make code that works with both Haskell 2010 and GHC by:
- Whenever you make a
Bits
instance of a type, also make aNum
instance, and - Whenever you give a function, instance or class a
Bits t
constraint, also give it aNum t
constraint, and - Always define the
bit
,testBit
andpopCount
methods inBits
instances.
- Whenever you make a
Read class methods
-
The
Read
class has two extra methods,readPrec
andreadListPrec
, that are not found in the Haskell 2010 since they rely on theReadPrec
data type, which requires theRankNTypes
extension. GHC also derivesRead
instances by implementingreadPrec
instead ofreadsPrec
, and relies on a default implementation ofreadsPrec
that is defined in terms ofreadPrec
. GHC adds these two extra methods simply becauseReadPrec
is more efficient thanReadS
(the type on whichreadsPrec
is based). Monad superclass
-
The
Monad
class has anApplicative
superclass. You cannot writeMonad
instances that work for GHC and also for a Haskell 2010 implementation that does not defineApplicative
. - Extra instances
-
The following extra instances are defined:
instance Functor ((->) r) instance Monad ((->) r) instance Functor ((,) a) instance Functor (Either a) instance Monad (Either e)
- Multiply-defined array elements not checked
-
This code fragment should elicit a fatal error, but it does not:
main = print (array (1,1) [(1,2), (1,3)])
GHC’s implementation of
array
takes the value of an array slot from the last (index,value) pair in the list, and does no checking for duplicates. The reason for this is efficiency, pure and simple.
14.1.1.9. In Prelude
support
splitAt semantics
-
Data.List.splitAt
is more strict than specified in the Report. Specifically, the Report specifies thatsplitAt n xs = (take n xs, drop n xs)
which implies that
splitAt undefined undefined = (undefined, undefined)
but GHC’s implementation is strict in its first argument, so
splitAt undefined [] = undefined
Showing records
-
The Haskell 2010 definition of
Show
stipulates that the rendered string should only include parentheses which are necessary to unambiguously parse the result. For historical reasons,Show
instances derived by GHC include parentheses around records despite the fact that record syntax binds more tightly than function application; e.g.,data Hello = Hello { aField :: Int } deriving (Show) -- GHC produces... show (Just (Hello {aField=42})) == "Just (Hello {aField=42})" -- whereas Haskell 2010 calls for... show (Just (Hello {aField=42})) == "Just Hello {aField=42}"
Reading integers
-
GHC’s implementation of the
Read
class for integral types accepts hexadecimal and octal literals (the code in the Haskell 98 report doesn’t). So, for example,read "0xf00" :: Int
works in GHC.
A possible reason for this is that
readLitChar
accepts hex and octal escapes, so it seems inconsistent not to do so for integers too. isAlpha
-
The Haskell 98 definition of
isAlpha
is:isAlpha c = isUpper c || isLower c
GHC’s implementation diverges from the Haskell 98 definition in the sense that Unicode alphabetic characters which are neither upper nor lower case will still be identified as alphabetic by
isAlpha
. hGetContents
-
Lazy I/O throws an exception if an error is encountered, in contrast to the Haskell 98 spec which requires that errors are discarded (see Section 21.2.2 of the Haskell 98 report). The exception thrown is the usual IO exception that would be thrown if the failing IO operation was performed in the IO monad, and can be caught by
System.IO.Error.catch
orControl.Exception.catch
.
14.1.1.10. The Foreign Function Interface
hs_init(), hs_exit()
-
The FFI spec requires the implementation to support re-initialising itself after being shut down with
hs_exit()
, but GHC does not currently support that. See #13693.
14.1.2. GHC’s interpretation of undefined behaviour in Haskell 98 and Haskell 2010
This section documents GHC’s take on various issues that are left undefined or implementation specific in Haskell 98.
Char
-
Following the ISO-10646 standard,
maxBound :: Char
in GHC is0x10FFFF
. Int
-
In GHC the
Int
type follows the size of an address on the host architecture; in other words it holds 32 bits on a 32-bit machine, and 64-bits on a 64-bit machine.Arithmetic on
Int
is unchecked for overflowInt
, so all operations onInt
happen modulo 2⟨n⟩ where ⟨n⟩ is the size in bits of theInt
type.The
fromInteger
(and hence alsofromIntegral
) is a special case when converting toInt
. The value offromIntegral x :: Int
is given by taking the lower ⟨n⟩ bits of(abs x)
, multiplied by the sign ofx
(in 2’s complement ⟨n⟩-bit arithmetic). This behaviour was chosen so that for example writing0xffffffff :: Int
preserves the bit-pattern in the resultingInt
.Negative literals, such as
-3
, are specified by (a careful reading of) the Haskell Report as meaningPrelude.negate (Prelude.fromInteger 3)
. So-2147483648
meansnegate (fromInteger 2147483648)
. SincefromInteger
takes the lower 32 bits of the representation,fromInteger (2147483648::Integer)
, computed at typeInt
is-2147483648::Int
. Thenegate
operation then overflows, but it is unchecked, sonegate (-2147483648::Int)
is just-2147483648
. In short, one can writeminBound::Int
as a literal with the expected meaning (but that is not in general guaranteed).The
fromIntegral
function also preserves bit-patterns when converting between the sized integral types (Int8
,Int16
,Int32
,Int64
and the unsignedWord
variants), see the modulesData.Int
andData.Word
in the library documentation. - Unchecked floating-point arithmetic
-
Operations on
Float
andDouble
numbers are unchecked for overflow, underflow, and other sad occurrences. (note, however, that some architectures trap floating-point overflow and loss-of-precision and report a floating-point exception, probably terminating the program) Large tuple support
-
The Haskell Report only requires implementations to provide tuple types and their accompanying standard instances up to size 15. GHC limits the size of tuple types to 62 and provides instances of
Eq
,Ord
,Bounded
,Read
,Show
, andIx
for tuples up to size 15.
14.2. Known bugs or infelicities
The bug tracker lists bugs that have been reported in GHC but not yet fixed: see the GHC issue tracker. In addition to those, GHC also has the following known bugs or infelicities. These bugs are more permanent; it is unlikely that any of them will be fixed in the short term.
14.2.1. Bugs in GHC
GHC’s runtime system implements cooperative multitasking, with context switching potentially occurring only when a program allocates. This means that programs that do not allocate may never context switch. This is especially true of programs using STM, which may deadlock after observing inconsistent state. See #367 for further discussion.
If you are hit by this, you may want to compile the affected module with
-fno-omit-yields
(see -f*: platform-independent flags). This flag ensures that yield points are inserted at every function entrypoint (at the expense of a bit of performance).GHC does not allow you to have a data type with a context that mentions type variables that are not data type parameters. For example:
data C a b => T a = MkT a
so that
MkT
’s type isMkT :: forall a b. C a b => a -> T a
In principle, with a suitable class declaration with a functional dependency, it’s possible that this type is not ambiguous; but GHC nevertheless rejects it. The type variables mentioned in the context of the data type declaration must be among the type parameters of the data type.
GHC’s inliner can be persuaded into non-termination using the standard way to encode recursion via a data type:
data U = MkU (U -> Bool) russel :: U -> Bool russel u@(MkU p) = not $ p u x :: Bool x = russel (MkU russel)
The non-termination is reported like this:
ghc: panic! (the 'impossible' happened) (GHC version 8.2.1 for x86_64-unknown-linux): Simplifier ticks exhausted When trying UnfoldingDone x_alB To increase the limit, use -fsimpl-tick-factor=N (default 100)
with the panic being reported no matter how high a
-fsimpl-tick-factor
you supply.We have never found another class of programs, other than this contrived one, that makes GHC diverge, and fixing the problem would impose an extra overhead on every compilation. So the bug remains un-fixed. There is more background in Secrets of the GHC inliner.
On 32-bit x86 platforms when using the native code generator, the
-fexcess-precision
option is always on. This means that floating-point calculations are non-deterministic, because depending on how the program is compiled (optimisation settings, for example), certain calculations might be done at 80-bit precision instead of the intended 32-bit or 64-bit precision. Floating-point results may differ when optimisation is turned on. In the worst case, referential transparency is violated, because for examplelet x = E1 in E2
can evaluate to a different value thanE2[E1/x]
.One workaround is to use the
-msse2
option (see Platform-specific Flags), which generates code to use the SSE2 instruction set instead of the x87 instruction set. SSE2 code uses the correct precision for all floating-point operations, and so gives deterministic results. However, note that this only works with processors that support SSE2 (Intel Pentium 4 or AMD Athlon 64 and later), which is why the option is not enabled by default. The libraries that come with GHC are probably built without this option, unless you built GHC yourself.The
state hack
optimization can result in non-obvious changes in evaluation ordering which may hide exceptions, even with-fpedantic-bottoms
(see, e.g., #7411). For instance,import Control.Exception import Control.DeepSeq main = do evaluate (('a' : undefined) `deepseq` return () :: IO ()) putStrLn "Hello"
Compiling this program with
-O
results inHello
to be printed, despite the fact thatevaluate
should have bottomed. Compiling with-O -fno-state-hack
results in the exception one would expect.Programs compiled with
-fdefer-type-errors
may fail a bit more eagerly than one might expect. For instance,{-# OPTIONS_GHC -fdefer-type-errors #-} main = do putStrLn "Hi there." putStrLn True
Will emit no output, despite the fact that the ill-typed term appears after the well-typed
putStrLn "Hi there."
. See #11197.- Despite appearances
*
andConstraint
aren’t really distinct kinds in the compiler’s internal representation and can be unified producing unexpected results. See #11715 for one example. - Because of a toolchain limitation we are unable to support full Unicode paths on Windows. On Windows we support up to Latin-1. See #12971 for more.
14.2.2. Bugs in GHCi (the interactive GHC)
GHCi does not respect the
default
declaration in the module whose scope you are in. Instead, for expressions typed at the command line, you always get the default default-type behaviour; that is,default(Int,Double)
.It would be better for GHCi to record what the default settings in each module are, and use those of the ‘current’ module (whatever that is).
On Windows, there’s a GNU ld/BFD bug whereby it emits bogus PE object files that have more than 0xffff relocations. When GHCi tries to load a package affected by this bug, you get an error message of the form
Loading package javavm ... linking ... WARNING: Overflown relocation field (# relocs found: 30765)
The last time we looked, this bug still wasn’t fixed in the BFD codebase, and there wasn’t any noticeable interest in fixing it when we reported the bug back in 2001 or so.
The workaround is to split up the .o files that make up your package into two or more .o’s, along the lines of how the
base
package does it.