5.3. Optimisation (code improvement)

The -O* options specify convenient “packages” of optimisation flags; the -f* options described later on specify individual optimisations to be turned on/off; the -m* options specify machine-specific optimisations to be turned on/off.

Most of these options are boolean and have options to turn them both “on” and “off” (beginning with the prefix no-). For instance, while -fspecialise enables specialisation, -fno-specialise disables it. When multiple flags for the same option appear in the command-line they are evaluated from left to right. For instance, -fno-specialise -fspecialise will enable specialisation.

It is important to note that the -O* flags are roughly equivalent to combinations of -f* flags. For this reason, the effect of the -O* and -f* flags is dependent upon the order in which they occur on the command line.

For instance, take the example of -fno-specialise -O1. Despite the -fno-specialise appearing in the command line, specialisation will still be enabled. This is the case as -O1 implies -fspecialise, overriding the previous flag. By contrast, -O1 -fno-specialise will compile without specialisation, as one would expect.

5.3.1. -O*: convenient “packages” of optimisation flags.

There are many options that affect the quality of code produced by GHC. Most people only have a general goal, something like “Compile quickly” or “Make my program run like greased lightning.” The following “packages” of optimisations (or lack thereof) should suffice.

Note that higher optimisation levels cause more cross-module optimisation to be performed, which can have an impact on how much of your program needs to be recompiled when you change something. This is one reason to stick to no-optimisation when developing code.

No ``-O*``-type option specified: This is taken to mean “Please compile quickly; I’m not over-bothered about compiled-code quality.” So, for example, ghc -c Foo.hs

-O0

Means “turn off all optimisation”, reverting to the same settings as if no -O options had been specified. Saying -O0 can be useful if e.g. make has inserted a -O on the command line already.

-O
-O1

Means: “Generate good-quality code without taking too long about it.” Thus, for example: ghc -c -O Main.lhs

-O2

Means: “Apply every non-dangerous optimisation, even if it means significantly longer compile times.”

The avoided “dangerous” optimisations are those that can make runtime or space worse if you’re unlucky. They are normally turned on or off individually.

-O⟨n⟩

Any -On where n > 2 is the same as -O2.

We don’t use a -O* flag for day-to-day work. We use -O to get respectable speed; e.g., when we want to measure something. When we want to go for broke, we tend to use -O2 (and we go for lots of coffee breaks).

The easiest way to see what -O (etc.) “really mean” is to run with -v, then stand back in amazement.

5.3.2. -f*: platform-independent flags

These flags turn on and off individual optimisations. Flags marked as on by default are enabled by -O, and as such you shouldn’t need to set any of them explicitly. A flag -fwombat can be negated by saying -fno-wombat.

-fcore-constant-folding
Default:on

Enables Core-level constant folding, i.e. propagation of values that can be computed at compile time.

-fcase-merge
Default:on

Merge immediately-nested case expressions that scrutinise the same variable. For example,

case x of
   Red -> e1
   _   -> case x of
            Blue -> e2
            Green -> e3

Is transformed to,

case x of
   Red -> e1
   Blue -> e2
   Green -> e2
-fcase-folding
Default:on

Allow constant folding in case expressions that scrutinise some primops: For example,

case x `minusWord#` 10## of
   10## -> e1
   20## -> e2
   v    -> e3

Is transformed to,

case x of
   20## -> e1
   30## -> e2
   _    -> let v = x `minusWord#` 10## in e3
-fcall-arity
Default:on

Enable call-arity analysis.

-fexitification
Default:on

Enables the floating of exit paths out of recursive functions.

-fcmm-elim-common-blocks
Default:on

Enables the common block elimination optimisation in the code generator. This optimisation attempts to find identical Cmm blocks and eliminate the duplicates.

-fcmm-sink
Default:on

Enables the sinking pass in the code generator. This optimisation attempts to find identical Cmm blocks and eliminate the duplicates attempts to move variable bindings closer to their usage sites. It also inlines simple expressions like literals or registers.

-fcmm-static-pred
Default:off but enabled with -O.

This enables static control flow prediction on the final Cmm code. If enabled GHC will apply certain heuristics to identify loops and hot code paths. This information is then used by the register allocation and code layout passes.

-fcmm-control-flow
Default:on

Enables some control flow optimisations in the Cmm code generator, merging basic blocks and avoiding jumps right after jumps.

-fasm-shortcutting
Default:off

This enables shortcutting at the assembly stage of the code generator. In simpler terms shortcutting means if a block of instructions A only consists of a unconditionally jump, we replace all jumps to A by jumps to the successor of A.

This is mostly done during Cmm passes. However this can miss corner cases. So at -O2 we run the pass again at the asm stage to catch these.

-fblock-layout-cfg
Default:off but enabled with -O.

The new algorithm considers all outgoing edges of a basic blocks for code layout instead of only the last jump instruction. It also builds a control flow graph for functions, tries to find hot code paths and place them sequentially leading to better cache utilization and performance.

This is expected to improve performance on average, but actual performance difference can vary.

If you find cases of significant performance regressions, which can be traced back to obviously bad code layout please open a ticket.

-fblock-layout-weights

This flag is hacker territory. The main purpose of this flag is to make it easy to debug and tune the new code layout algorithm. There is no guarantee that values giving better results now won’t be worse with the next release.

If you feel your code warrants modifying these settings please consult the source code for default values and documentation. But I strongly advise against this.

-fblock-layout-weightless
Default:off

When not using the cfg based blocklayout layout is determined either by the last jump in a basic block or the heaviest outgoing edge of the block in the cfg.

With this flag enabled we use the last jump instruction in blocks. Without this flags the old algorithm also uses the heaviest outgoing edge.

When this flag is enabled and -fblock-layout-cfg is disabled block layout behaves the same as in 8.6 and earlier.

-fcpr-anal
Default:on

Turn on CPR analysis, which enables the worker/wrapper transformation (cf. -fworker-wrapper) to unbox the result of a function, such as

sum :: [Int] -> Int
sum []     = 0
sum (x:xs) = x + sum xs

CPR analysis will see that each code path produces a constructed product such as I# 0# in the first branch (where GHC.Exts.I# is the data constructor of Int, boxing up the primitive integer literal 0# of type Int#) and optimise to

sum xs = I# ($wsum xs)
$wsum []        = 0#
$wsum (I# x:xs) = x# +# $wsum xs

and then sum can inline to potentially cancel away the I# box.

Here’s an example of the function that does not return a constructed product:

f :: [Int] -> (Int -> Int) -> Int
f []     g = g 0
f (x:xs) g = x + f xs g

The expression g 0 is not a constructed product, because we don’t know anything about g.

CPR analysis also works nestedly, for example

sumIO :: [Int] -> IO Int
sumIO []     = return 0
sumIO (x:xs) = do
  r <- sumIO xs
  return $! x + r

Note the use of $!: Without it, GHC would be unable to see that evaluation of r and x terminates (and rapidly, at that). An alternative would be to evaluate both with a bang pattern or a seq, but the return $! <res> idiom should work more reliably and needs less thinking. The above example will be optimised to

sumIO :: [Int] -> IO Int
sumIO xs = IO $ \s -> case $wsum xs s of
  (# s', r #) -> (# s', I# r #)
$wsumIO :: [Int] -> (# RealWorld#, Int# #)
$wsumIO []        s = (# s, 0# #)
$wsumIO (I# x:xs) s = case $wsumIO xs of
  (# s', r #) -> (# s', x +# r#)

And the latter can inline sumIO and cancel away the I# constructor. Unboxing the result of a State action should work similarly.

-fcse
Default:on

Enables the common-sub-expression elimination optimisation. Switching this off can be useful if you have some unsafePerformIO expressions that you don’t want commoned-up.

-fstg-cse
Default:on

Enables the common-sub-expression elimination optimisation on the STG intermediate language, where it is able to common up some subexpressions that differ in their types, but not their representation.

-fdicts-cheap
Default:off

A very experimental flag that makes dictionary-valued expressions seem cheap to the optimiser.

-fdicts-strict
Default:off

Make dictionaries strict.

This enables WW to fire on dictionary constraints which usually results in better runtime. In niche cases it can lead to significant compile time regressions because of changed inlining behaviour. Rarely this can also affect runtime negatively.

If enabling this flag leads to regressions try increasing the unfolding threshold using -funfolding-use-threshold=⟨n⟩ by a modest amount (~30) as this is likely a result of a known limitation described in #18421.

-fdmd-tx-dict-sel
Default:on

Use a special demand transformer for dictionary selectors. Behaviour is unconditionally enabled starting with 9.2

-fdo-eta-reduction
Default:on

Eta-reduce lambda expressions, if doing so gets rid of a whole group of lambdas.

-fdo-lambda-eta-expansion
Default:on

Eta-expand let-bindings to increase their arity.

-feager-blackholing
Default:off

Usually GHC black-holes a thunk only when it switches threads. This flag makes it do so as soon as the thunk is entered. See Haskell on a shared-memory multiprocessor.

See Compile-time options for SMP parallelism for a discussion on its use.

-fexcess-precision
Default:off

When this option is given, intermediate floating point values can have a greater precision/range than the final type. Generally this is a good thing, but some programs may rely on the exact precision/range of Float/Double values and should not use this option for their compilation.

Note that the 32-bit x86 native code generator only supports excess-precision mode, so neither -fexcess-precision nor -fno-excess-precision has any effect. This is a known bug, see Bugs in GHC.

-fexpose-all-unfoldings
Default:off

An experimental flag to expose all unfoldings, even for very large or recursive functions. This allows for all functions to be inlined while usually GHC would avoid inlining larger functions.

-ffloat-in
Default:on

Float let-bindings inwards, nearer their binding site. See Let-floating: moving bindings to give faster programs (ICFP‘96).

This optimisation moves let bindings closer to their use site. The benefit here is that this may avoid unnecessary allocation if the branch the let is now on is never executed. It also enables other optimisation passes to work more effectively as they have more information locally.

This optimisation isn’t always beneficial though (so GHC applies some heuristics to decide when to apply it). The details get complicated but a simple example is that it is often beneficial to move let bindings outwards so that multiple let bindings can be grouped into a larger single let binding, effectively batching their allocation and helping the garbage collector and allocator.

-ffull-laziness
Default:on

Run the full laziness optimisation (also known as let-floating), which floats let-bindings outside enclosing lambdas, in the hope they will be thereby be computed less often. See Let-floating: moving bindings to give faster programs (ICFP‘96). Full laziness increases sharing, which can lead to increased memory residency.

Note

GHC doesn’t implement complete full laziness. Although GHC’s full-laziness optimisation does enable some transformations which would be performed by a fully lazy implementation (such as extracting repeated computations from loops), these transformations are not applied consistently, so don’t rely on them.

-ffun-to-thunk
Default:off

Worker/wrapper removes unused arguments, but usually we do not remove them all, lest it turn a function closure into a thunk, thereby perhaps creating a space leak and/or disrupting inlining. This flag allows worker/wrapper to remove all value lambdas.

This flag was ineffective in the presence of -ffull-laziness, which would flout a thunk out of a constant worker function even though -ffun-to-thunk was off.

Hence use of this flag is deprecated since GHC 9.4.1 and we rather suggest to pass -fno-full-laziness instead. That implies there’s no way for worker/wrapper to turn a function into a thunk in the presence of -fno-full-laziness. If that is inconvenient for you, please leave a comment on the issue tracker (#21204).

-fignore-asserts
Default:on

Causes GHC to ignore uses of the function Exception.assert in source code (in other words, rewriting Exception.assert p e to e (see Assertions).

-fignore-interface-pragmas
Default:off

Tells GHC to ignore all inessential information when reading interface files. That is, even if M.hi contains unfolding or strictness information for a function, GHC will ignore that information.

-flate-dmd-anal
Default:off

Run demand analysis again, at the end of the simplification pipeline. We found some opportunities for discovering strictness that were not visible earlier; and optimisations like -fspec-constr can create functions with unused arguments which are eliminated by late demand analysis. Improvements are modest, but so is the cost. See notes on the wiki page.

-fliberate-case
Default:off but enabled with -O2.

Turn on the liberate-case transformation. This unrolls recursive function once in its own RHS, to avoid repeated case analysis of free variables. It’s a bit like the call-pattern specialiser (-fspec-constr) but for free variables rather than arguments.

-fliberate-case-threshold=⟨n⟩
Default:2000

Set the size threshold for the liberate-case transformation.

-floopification
Default:on

When this optimisation is enabled the code generator will turn all self-recursive saturated tail calls into local jumps rather than function calls.

-fllvm-pass-vectors-in-regs
Default:on

This flag has no effect since GHC 8.8 - its behavior is always on. It used to instruct GHC to use the platform’s native vector registers to pass vector arguments during function calls.

-fmax-inline-alloc-size=⟨n⟩
Default:128

Set the maximum size of inline array allocations to n bytes. GHC will allocate non-pinned arrays of statically known size in the current nursery block if they’re no bigger than n bytes, ignoring GC overheap. This value should be quite a bit smaller than the block size (typically: 4096).

-fmax-inline-memcpy-insns=⟨n⟩
Default:32

Inline memcpy calls if they would generate no more than ⟨n⟩ pseudo-instructions.

-fmax-inline-memset-insns=⟨n⟩
Default:32

Inline memset calls if they would generate no more than n pseudo instructions.

-fmax-relevant-binds=⟨n⟩
Default:6

The type checker sometimes displays a fragment of the type environment in error messages, but only up to some maximum number, set by this flag. Turning it off with -fno-max-relevant-binds gives an unlimited number. Syntactically top-level bindings are also usually excluded (since they may be numerous), but -fno-max-relevant-binds includes them too.

-fmax-uncovered-patterns=⟨n⟩
Default:4

Maximum number of unmatched patterns to be shown in warnings generated by -Wincomplete-patterns and -Wincomplete-uni-patterns.

-fmax-simplifier-iterations=⟨n⟩
Default:4

Sets the maximal number of iterations for the simplifier.

-flocal-float-out
Default:on

Enable local floating of bindings from the RHS of a let(rec) in the simplifier. For example

let x = let y = rhs_y in rhs_x in blah
==>
let y = rhs_y in let x = rhs_x in blah

See the paper “Let-floating: moving bindings to give faster programs”, Partain, Santos, and Peyton Jones; ICFP 1996. https://www.microsoft.com/en-us/research/publication/let-floating-moving-bindings-to-give-faster-programs/

Note

This is distinct from the global floating pass which can be disabled with -fno-full-laziness.

-flocal-float-out-top-level
Default:on

Enable local floating of top-level bindings from the RHS of a let(rec) in the simplifier. For example

x = let y = e in (a,b) ===> y = e; x = (a,b)

See the paper “Let-floating: moving bindings to give faster programs”, Partain, Santos, and Peyton Jones; ICFP 1996. https://www.microsoft.com/en-us/research/publication/let-floating-moving-bindings-to-give-faster-programs/

Note that if -fno-local-float-out is set, that will take precedence.

Note

This is distinct from the global floating pass which can be disabled with -fno-full-laziness.

-fmax-worker-args=⟨n⟩
Default:10

A function will not be split into worker and wrapper if the number of value arguments of the resulting worker exceeds both that of the original function and this setting.

-fno-opt-coercion
Default:coercion optimisation enabled.

Turn off the coercion optimiser.

-fno-pre-inlining
Default:pre-inlining enabled

Turn off pre-inlining.

-fno-state-hack
Default:state hack is enabled

Turn off the “state hack” whereby any lambda with a State# token as argument is considered to be single-entry, hence it is considered okay to inline things inside it. This can improve performance of IO and ST monad code, but it runs the risk of reducing sharing.

-fomit-interface-pragmas
Default:Implied by -O0, otherwise off.

Tells GHC to omit all inessential information from the interface file generated for the module being compiled (say M). This means that a module importing M will see only the types of the functions that M exports, but not their unfoldings, strictness info, etc. Hence, for example, no function exported by M will be inlined into an importing module. The benefit is that modules that import M will need to be recompiled less often (only when M’s exports change their type, not when they change their implementation).

-fomit-yields
Default:on (yields are not inserted)

Tells GHC to omit heap checks when no allocation is being performed. While this improves binary sizes by about 5%, it also means that threads run in tight non-allocating loops will not get preempted in a timely fashion. If it is important to always be able to interrupt such threads, you should turn this optimization off. Consider also recompiling all libraries with this optimization turned off, if you need to guarantee interruptibility.

-fpedantic-bottoms
Default:off

Make GHC be more precise about its treatment of bottom (but see also -fno-state-hack). In particular, stop GHC eta-expanding through a case expression, which is good for performance, but bad if you are using seq on partial applications.

-fregs-graph
Default:off due to a performance regression bug (%s7679)

Only applies in combination with the native code generator. Use the graph colouring register allocator for register allocation in the native code generator. By default, GHC uses a simpler, faster linear register allocator. The downside being that the linear register allocator usually generates worse code.

-fregs-iterative
Default:off

Only applies in combination with the native code generator. Use the iterative coalescing graph colouring register allocator for register allocation in the native code generator. This is the same register allocator as the -fregs-graph one but also enables iterative coalescing during register allocation.

-fsimplifier-phases=⟨n⟩
Default:2

Set the number of phases for the simplifier. Ignored with -O0.

-fsimpl-tick-factor=⟨n⟩
Default:100

GHC’s optimiser can diverge if you write rewrite rules (Rewrite rules) that don’t terminate, or (less satisfactorily) if you code up recursion through data types (Bugs in GHC). To avoid making the compiler fall into an infinite loop, the optimiser carries a “tick count” and stops inlining and applying rewrite rules when this count is exceeded. The limit is set as a multiple of the program size, so bigger programs get more ticks. The -fsimpl-tick-factor flag lets you change the multiplier. The default is 100; numbers larger than 100 give more ticks, and numbers smaller than 100 give fewer.

If the tick-count expires, GHC summarises what simplifier steps it has done; you can use -fddump-simpl-stats to generate a much more detailed list. Usually that identifies the loop quite accurately, because some numbers are very large.

-fdmd-unbox-width=⟨n⟩
Default:3

Boxity analysis optimistically pretends that a function returning a record with at most -fdmd-unbox-width fields has only call sites that don’t need the box of the returned record. That may in turn allow more argument unboxing to happen. Set to 0 to be completely conservative (which guarantees that no reboxing will happen due to this mechanism).

-fspec-constr
Default:off but enabled by -O2.

Turn on call-pattern specialisation; see Call-pattern specialisation for Haskell programs.

This optimisation specializes recursive functions according to their argument “shapes”. This is best explained by example so consider:

last :: [a] -> a
last [] = error "last"
last (x : []) = x
last (x : xs) = last xs

In this code, once we pass the initial check for an empty list we know that in the recursive case this pattern match is redundant. As such -fspec-constr will transform the above code to:

last :: [a] -> a
last []       = error "last"
last (x : xs) = last' x xs
    where
      last' x []       = x
      last' x (y : ys) = last' y ys

As well avoid unnecessary pattern matching it also helps avoid unnecessary allocation. This applies when a argument is strict in the recursive call to itself but not on the initial entry. As strict recursive branch of the function is created similar to the above example.

It is also possible for library writers to instruct GHC to perform call-pattern specialisation extremely aggressively. This is necessary for some highly optimized libraries, where we may want to specialize regardless of the number of specialisations, or the size of the code. As an example, consider a simplified use-case from the vector library:

import GHC.Types (SPEC(..))

foldl :: (a -> b -> a) -> a -> Stream b -> a
{-# INLINE foldl #-}
foldl f z (Stream step s _) = foldl_loop SPEC z s
  where
    foldl_loop !sPEC z s = case step s of
                            Yield x s' -> foldl_loop sPEC (f z x) s'
                            Skip       -> foldl_loop sPEC z s'
                            Done       -> z

Here, after GHC inlines the body of foldl to a call site, it will perform call-pattern specialisation very aggressively on foldl_loop due to the use of SPEC in the argument of the loop body. SPEC from GHC.Types is specifically recognised by the compiler.

(NB: it is extremely important you use seq or a bang pattern on the SPEC argument!)

In particular, after inlining this will expose f to the loop body directly, allowing heavy specialisation over the recursive cases.

-fspec-constr-keen
Default:off

If this flag is on, call-pattern specialisation will specialise a call (f (Just x)) with an explicit constructor argument, even if the argument is not scrutinised in the body of the function. This is sometimes beneficial; e.g. the argument might be given to some other function that can itself be specialised.

-fspec-constr-count=⟨n⟩
Default:3

Set the maximum number of specialisations that will be created for any one function by the SpecConstr transformation.

-fspec-constr-threshold=⟨n⟩
Default:2000

Set the size threshold for the SpecConstr transformation.

-fspecialise
Default:on

Specialise each type-class-overloaded function defined in this module for the types at which it is called in this module. If -fcross-module-specialise is set imported functions that have an INLINABLE pragma (INLINABLE pragma) will be specialised as well.

-fspecialise-aggressively
Default:off

By default only type class methods and methods marked INLINABLE or INLINE are specialised. This flag will specialise any overloaded function regardless of size if its unfolding is available. This flag is not included in any optimisation level as it can massively increase code size. It can be used in conjunction with -fexpose-all-unfoldings if you want to ensure all calls are specialised.

-fcross-module-specialise
Default:on

Specialise INLINABLE (INLINABLE pragma) type-class-overloaded functions imported from other modules for the types at which they are called in this module. Note that specialisation must be enabled (by -fspecialise) for this to have any effect.

-flate-specialise
Default:off

Runs another specialisation pass towards the end of the optimisation pipeline. This can catch specialisation opportunities which arose from the previous specialisation pass or other inlining.

You might want to use this if you are you have a type class method which returns a constrained type. For example, a type class where one of the methods implements a traversal.

-finline-generics
Default:on
Since:9.2.1

Annotate methods of derived Generic and Generic1 instances with INLINE[1] pragmas based on heuristics dependent on the size of the data type in question. Improves performance of generics-based algorithms as GHC is able to optimize away intermediate representation more often.

-finline-generics-aggressively
Default:off
Since:9.2.1

Annotate methods of all derived Generic and Generic1 instances with INLINE[1] pragmas.

This flag should only be used in modules deriving Generic instances that weren’t considered appropriate for INLINE[1] annotations by heuristics of -finline-generics, yet you know that doing so would be beneficial.

When enabled globally it will most likely lead to worse compile times and code size blowup without runtime performance gains.

-fsolve-constant-dicts
Default:on

When solving constraints, try to eagerly solve super classes using available dictionaries.

For example:

class M a b where m :: a -> b

type C a b = (Num a, M a b)

f :: C Int b => b -> Int -> Int
f _ x = x + 1

The body of f requires a Num Int instance. We could solve this constraint from the context because we have C Int b and that provides us a solution for Num Int. However, we can often produce much better code by directly solving for an available Num Int dictionary we might have at hand. This removes potentially many layers of indirection and crucially allows other optimisations to fire as the dictionary will be statically known and selector functions can be inlined.

The optimisation also works for GADTs which bind dictionaries. If we statically know which class dictionary we need then we will solve it directly rather than indirectly using the one passed in at run time.

-fstatic-argument-transformation
Default:off

Turn on the static argument transformation, which turns a recursive function into a non-recursive one with a local recursive loop. See Chapter 7 of Andre Santos’s PhD thesis.

-fstg-lift-lams
Default:on

Enables the late lambda lifting optimisation on the STG intermediate language. This selectively lifts local functions to top-level by converting free variables into function parameters.

-fstg-lift-lams-known
Default:off

Allow turning known into unknown calls while performing late lambda lifting. This is deemed non-beneficial, so it’s off by default.

-fstg-lift-lams-non-rec-args
Default:5

Create top-level non-recursive functions with at most <n> parameters while performing late lambda lifting. The default is 5, the number of available parameter registers on x86_64.

-fstg-lift-lams-rec-args
Default:5

Create top-level recursive functions with at most <n> parameters while performing late lambda lifting. The default is 5, the number of available parameter registers on x86_64.

-fstrictness
Default:on

Turn on demand analysis.

A Demand describes an evaluation context of an expression. Demand analysis tries to find out what demands a function puts on its arguments when called: If an argument is scrutinised on every code path, the function is strict in that argument and GHC is free to use the more efficient call-by-value calling convention, as well as pass parameters unboxed.

Apart from strictness analysis, demand analysis also performs usage analysis: Where strict translates to “evaluated at least once”, usage analysis asks whether arguments and bindings are “evaluated at most once” or not at all (“evaluated at most zero times”), e.g. absent. For the former, GHC may use call-by-name instead of call-by-need, effectively turning thunks into non-memoised functions. For the latter, no code needs to be generated at all: An absent argument can simply be replaced by a dummy value at the call site or omitted altogether.

The worker/wrapper transformation (-fworker-wrapper) is responsible for exploiting unboxing opportunities and replacing absent arguments by dummies. For arguments that can’t be unboxed, opportunities for call-by-value and call-by-name are exploited in CorePrep when translating to STG.

It’s not only interesting to look at how often a binding is evaluated, but also how often a function is called. If a function is called at most once, we may freely eta-expand it, even if doing so destroys shared work if the function was called multiple times. This information translates into OneShotInfo annotations that the Simplifier acts on.

Notation

So demand analysis is about conservatively inferring lower and upper bounds about how many times something is evaluated/called. We call the “how many times” part a cardinality. In the compiler and debug output we differentiate the following cardinality intervals as approximations to cardinality:

Interval Set of denoted cardinalities Syntax Explanation tying syntax to semantics
[1,0] {} B Bottom element
[0,0] {0} A Absent
[0,1] {0,1} M Used at most once (“Maybe”)
[0,ω] {0,1,ω} L Lazy. Top element, no information, used at least 0, at most many times
[1,1] {1} 1 Strict, used exactly once
[1,ω] {1,ω} S Strict, used possibly many times

Note that it’s never interesting to differentiate between a cardinality of 2 and 3, or even 4232123. We just approximate the >1 case with ω, standing for “many times”.

Apart from the cardinality describing how often an argument is evaluated, a demand also carries a sub-demand, describing how deep something is evaluated beyond a simple seq-like evaluation.

This is the full syntax for cardinalities, demands and sub-demands in BNF:

card ::= B | A | M | L | 1 | S    semantics as in the table above

d    ::= card sd                  card = how often, sd = how deep
      |  card                     abbreviation: Same as "card card"

sd   ::= card                     polymorphic sub-demand, card at every level
      |  P(d,d,..)                product sub-demand
      |  C(card,sd)               call sub-demand

For example, fst is strict in its argument, and also in the first component of the argument. It will not evaluate the argument’s second component. That is expressed by the demand 1P(1L,A). The P is for “product sub-demand”, which has a demand for each product field. The notation 1L just says “evaluated strictly (1), with everything nested inside evaluated according to L” – e.g., no information, because that would depend on the evaluation context of the call site of fst. The role of L in 1L is that of a polymorphic sub-demand, being semantically equivalent to the sub-demand P(LP(..)), which we simply abbreviate by the (consequently overloaded) cardinality notation L.

For another example, the expression x + 1 evaluates x according to demand 1P(L). We have seen single letters stand for cardinalities and polymorphic sub-demands, but what does the single letter L mean for a demand? Such a single letter demand simply expands to a cardinality and a polymorphic sub-demand of the same letter: E.g. L is equivalent to LL by expansion of the single letter demand, which is equivalent to LP(LP(..)), so Ls all the way down. It is always clear from context whether we talk about about a cardinality, sub-demand or demand.

Demand signatures

We summarise a function’s demand properties in its demand signature. This is the general syntax:

{x->dx,y->dy,z->dz...}<d1><d2><d3>...<dn>div
        ^              ^   ^   ^      ^   ^
        |              |   |   |      |   |
        |              \---+---+------/   |
        |                  |              |
   demand on free        demand on      divergence
     variables           arguments      information
 (omitted if empty)                     (omitted if
                                      no information)

We summarise fst’s demand properties in its demand signature <1P(1L,A)>, which just says “If fst is applied to one argument, that argument is evaluated according to 1P(1L,A)”. For another example, the demand signature of seq would be <1A><1L> and that of + would be <1P(L)><1P(L)>.

If not omitted, the divergence information can be b (surely diverges) or x (surely diverges or throws a precise exception). For example, error has demand signature <S>b and throwIO (which is the only way to throw precise exceptions) has demand signature <_><L><L>x (leaving out the complicated demand on the Exception dictionary).

Call sub-demands

Consider maybe:

maybe :: b -> (a -> b) -> Maybe a -> b
maybe n _ Nothing  = n
maybe _ s (Just a) = s a

We give it demand signature <L><MC(M,L)><1L>. The C(M,L) is a call sub-demand that says “Called at most once, where the result is used according to L”. The expression f `seq` f 1 puts f under demand SC(1,L) and serves as an example where the upper bound on evaluation cardinality doesn’t coincide with that of the call cardinality.

Cardinality is always relative to the enclosing call cardinality, so g 1 2 + g 3 4 puts g under demand SC(S,C(1,L)), which says “called multiple times (S), but every time it is called with one argument, it is applied exactly once to another argument (1)”.

-fstrictness-before=⟨n⟩

Run an additional demand analysis before simplifier phase ⟨n⟩.

-funbox-small-strict-fields
Default:on

This option causes all constructor fields which are marked strict (i.e. “!”) and which representation is smaller or equal to the size of a pointer to be unpacked, if possible. It is equivalent to adding an UNPACK pragma (see UNPACK pragma) to every strict constructor field that fulfils the size restriction.

For example, the constructor fields in the following data types

data A = A !Int
data B = B !A
newtype C = C B
data D = D !C

would all be represented by a single Int# (see Unboxed types and primitive operations) value with -funbox-small-strict-fields enabled.

This option is less of a sledgehammer than -funbox-strict-fields: it should rarely make things worse. If you use -funbox-small-strict-fields to turn on unboxing by default you can disable it for certain constructor fields using the NOUNPACK pragma (see NOUNPACK pragma).

Note that for consistency Double, Word64, and Int64 constructor fields are unpacked on 32-bit platforms, even though they are technically larger than a pointer on those platforms.

-funbox-strict-fields
Default:off

This option causes all constructor fields which are marked strict (i.e. !) to be unpacked if possible. It is equivalent to adding an UNPACK pragma to every strict constructor field (see UNPACK pragma).

This option is a bit of a sledgehammer: it might sometimes make things worse. Selectively unboxing fields by using UNPACK pragmas might be better. An alternative is to use -funbox-strict-fields to turn on unboxing by default but disable it for certain constructor fields using the NOUNPACK pragma (see NOUNPACK pragma).

Alternatively you can use -funbox-small-strict-fields to only unbox strict fields which are “small”.

-funfolding-creation-threshold=⟨n⟩
Default:750

Governs the maximum size that GHC will allow a function unfolding to be. (An unfolding has a “size” that reflects the cost in terms of “code bloat” of expanding (aka inlining) that unfolding at a call site. A bigger function would be assigned a bigger cost.)

Consequences:

  1. nothing larger than this will be inlined (unless it has an INLINE pragma)
  2. nothing larger than this will be spewed into an interface file.

Increasing this figure is more likely to result in longer compile times than faster code. The -funfolding-use-threshold=⟨n⟩ is more useful.

-funfolding-dict-discount=⟨n⟩
Default:30

How eager should the compiler be to inline dictionaries?

-funfolding-fun-discount=⟨n⟩
Default:60

How eager should the compiler be to inline functions?

-funfolding-keeness-factor=⟨n⟩

This factor was deprecated in GHC 9.0.1. See %s15304 for details. Users who need to control inlining should rather consider -funfolding-use-threshold=⟨n⟩.

-funfolding-use-threshold=⟨n⟩
Default:80

This is the magic cut-off figure for unfolding (aka inlining): below this size, a function definition will be unfolded at the call-site, any bigger and it won’t. The size computed for a function depends on two things: the actual size of the expression minus any discounts that apply depending on the context into which the expression is to be inlined.

The difference between this and -funfolding-creation-threshold=⟨n⟩ is that this one determines if a function definition will be inlined at a call site. The other option determines if a function definition will be kept around at all for potential inlining.

-funfolding-case-threshold=⟨n⟩
Default:2

GHC is in general quite eager to inline small functions. However sometimes these functions will be expanded by more inlining after inlining. Since they are now applied to “interesting” arguments. Even worse, their expanded form might reference again a small function, which will be inlined and expanded afterwards. This can repeat often and lead to explosive growth of programs.

As it happened in #18730.

Starting with GHC 9.0 we will be less eager to inline deep into nested cases. We achieve this by applying a inlining penalty that increases as the nesting gets deeper. However sometimes a specific (maybe quite high!) threshold of nesting is to be expected.

In such cases this flag can be used to ignore the first ⟨n⟩ levels of nesting when computing the penalty.

This flag in combination with -funfolding-case-scaling=⟨n⟩ can be used to break inlining loops without disabling inlining completely. For this purpose a smaller value is more likely to break such loops although often adjusting the scaling is enough and preferably.

-funfolding-case-scaling=⟨n⟩
Default:30

GHC is in general quite eager to inline small functions. However sometimes these functions will be expanded by more inlining after inlining. Since they are now applied to “interesting” arguments. Even worse, their expanded form might reference again a small function, which will be inlined and expanded afterwards. This can repeat often and lead to explosive growth of programs.

As it happened in #18730.

Starting with GHC 9.0 we will be less eager to inline deep into nested cases. We achieve this by applying a inlining penalty that increases as the nesting gets deeper. However sometimes we are ok with inlining a lot in the name of performance.

In such cases this flag can be used to tune how hard we penalize inlining into deeply nested cases beyond the threshold set by -funfolding-case-threshold=⟨n⟩. Cases are only counted against the nesting level if they have more than one alternative.

We use 1/n to scale the penalty. That is a higher value gives a lower penalty.

This can be used to break inlining loops. For this purpose a lower value is recommended. Values in the range 10 <= n <= 20 allow some inlining to take place while still allowing GHC to compile modules containing such inlining loops.

-fworker-wrapper

Enable the worker/wrapper transformation after a demand analysis pass.

Exploits strictness and absence information by unboxing strict arguments and replacing absent fields by dummy values in a wrapper function that will inline in all relevant scenarios and thus expose a specialised, unboxed calling convention of the worker function.

Implied by -O, and by -fstrictness. Disabled by -fno-strictness. Enabling -fworker-wrapper while demand analysis is disabled (by -fno-strictness) has no effect.

-fworker-wrapper-cbv

Disabling this flag prevents a W/W split if the only benefit would be call-by-value for some arguments.

Otherwise this exploits strictness information by passing strict value arguments call-by-value to the functions worker. Even for functions who would otherwise not get a worker.

This avoids (potentially repeated) checks for evaluatedness of arguments in the rhs of the worker by pushing this check to the call site. If the argument is statically visible to be a value at the call site the overhead for the check disappears completely.

This can cause slight codesize increases. It will also cause many more functions to get a worker/wrapper split which can play badly with rules (see Ticket #20364) which is why it’s currently disabled by default. In particular if you depend on rules firing on functions marked as NOINLINE without marking use sites of these functions as INLINE or INLINEABLE then things will break unless this flag is disabled.

While WorkerWrapper is disabled this has no effect.

-fbinary-blob-threshold=⟨n⟩
Default:500000

The native code-generator can either dump binary blobs (e.g. string literals) into the assembly file (by using “.asciz” or “.string” assembler directives) or it can dump them as binary data into a temporary file which is then included by the assembler (using the “.incbin” assembler directive).

This flag sets the size (in bytes) threshold above which the second approach is used. You can disable the second approach entirely by setting the threshold to 0.