Module Simplify_common

const_*_expr expr v annot, where the expression expr is known to evaluate to the value v, attempt to produce a more simple expression together with its approximation and the benefit gained by replacing expr with this new expression. This simplification is only performed if expr is known to have no side effects. Otherwise, expr itself is returned, with an appropriate approximation but zero benefit.

const_boxed_int_expr takes an additional argument specifying the kind of boxed integer to which the given expression evaluates.

val const_int_expr : Flambda.named -> int -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_char_expr : Flambda.named -> char -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_bool_expr : Flambda.named -> bool -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_ptr_expr : Flambda.named -> int -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_float_expr : Flambda.named -> float -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_boxed_int_expr : Flambda.named -> 'a Simple_value_approx.boxed_int -> 'a -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_integer_comparison_expr : Flambda.named -> Lambda.integer_comparison -> 'a -> 'a -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val const_float_comparison_expr : Flambda.named -> Lambda.float_comparison -> float -> float -> Flambda.named * Simple_value_approx.t * Inlining_cost.Benefit.t
val swap16 : int -> int

Functions for transposing the order of bytes within words of various sizes.

val swap32 : int32 -> int32
val swap64 : int64 -> int64
val swapnative : nativeint -> nativeint