Module `Numeric_common.Common`Source

Common constructs for numeric abstractions.

module I = Mopsa.ItvUtils.IntItv

module C = Mopsa.CongUtils.IntCong

Integer intervals

*********************

Sourcetype int_itv = I.t_with_bot

Integer intervals

Sourcetype Mopsa.avalue_kind +=

| V_int_interval : int_itv Mopsa.avalue_kind
(*
Query to evaluate the integer interval of an expression
*)

Sourcetype Mopsa.avalue_kind +=

| V_int_interval_fast : int_itv Mopsa.avalue_kind
(*
Same as V_int_interval but should be handled by optimized domains, such Boxes
*)

Source

val mk_int_interval_query : 
  ?fast:bool ->
  Ast.Expr.expr ->
  ('a, int_itv) Core.Query.query

Source

val pp_int_interval : 
  Stdlib.Format.formatter ->
  I.t Utils_core.Bot.with_bot ->
  unit

Source

val compare_int_interval : 
  I.t Utils_core.Bot.with_bot ->
  I.t Utils_core.Bot.with_bot ->
  int

Integer intervals with congruence

*************************************

Sourcetype int_congr_itv = int_itv * C.t_with_bot

Integer step intervals

Sourcetype Mopsa.avalue_kind +=

| V_int_congr_interval : int_congr_itv Mopsa.avalue_kind

Query to evaluate the integer interval of an expression

Source

val mk_int_congr_interval_query : 
  Ast.Expr.expr ->
  ('a, int_congr_itv) Core.Query.query

Rounding mode of floats

***************************

Sourceval opt_float_rounding : Apron.Texpr1.round Stdlib.ref

Sourceval rounding_option_name : string

Float intervals

*******************

module F = Mopsa.ItvUtils.FloatItvNan

Sourcetype float_itv = F.t

Float intervals

Sourcetype Mopsa.avalue_kind +=

| V_float_interval : Lang.Ast.float_prec -> float_itv Mopsa.avalue_kind
(*
Query to evaluate the float interval of an expression, with infinities and NaN
*)
| V_float_maybenan : Lang.Ast.float_prec -> bool Mopsa.avalue_kind

Source

val mk_float_interval_query : 
  ?prec:Lang.Ast.float_prec ->
  Ast.Expr.expr ->
  ('a, float_itv) Core.Query.query

Source

val mk_float_maybenan_query : 
  ?prec:Lang.Ast.float_prec ->
  Ast.Expr.expr ->
  ('a, bool) Core.Query.query

Sourceval pp_float_interval : Stdlib.Format.formatter -> F.t -> unit

Sourceval compare_float_interval : F.t -> F.t -> int

Sourceval prec : Lang.Ast.float_prec -> Mopsa.ItvUtils.FloatItvNan.prec

Sourceval round : unit -> Mopsa.ItvUtils.FloatItvNan.round

Fast assume on numeric conditions

*************************************

Source

val interval_of_num_expr : 
  Mopsa.expr ->
  ('a, 'b) Mopsa.man ->
  'a Core.Flow.flow ->
  int_itv

Get the intervals of a numeric expression

Source

val eval_num_cond : 
  Ast.Expr.expr ->
  ('a, 'b) Mopsa.man ->
  'a Core.Flow.flow ->
  bool option

Evaluate a numeric condition using intervals

Source

val assume_num : 
  Ast.Expr.expr ->
  fthen:('a Core.Flow.flow -> ('a, 'b) Core.Cases.cases) ->
  felse:('a Core.Flow.flow -> ('a, 'b) Core.Cases.cases) ->
  ?route:Mopsa.route ->
  ('a, 'c) Mopsa.man ->
  'a Core.Flow.flow ->
  ('a, 'b) Mopsa.Cases.cases

Optimized assume function that uses intervals to check a condition or falls back to classic assume

Widening thresholds

***********************

module K : sig ... end

Source

val widening_thresholds_ctx_key : 
  ('a, Mopsa.SetExt.ZSet.t) Core__Context.ctx_key

Key for accessing widening thresholds

Source

val add_widening_threshold : 
  Core.All.var ->
  Mopsa.SetExt.ZSet.elt ->
  'a Core.All.ctx ->
  'a Core.All.ctx

Add a constant to the widening thresholds of a variable

Source

val remove_widening_thresholds : 
  Core.All.var ->
  'a Core.All.ctx ->
  'a Core.All.ctx

Remove all widening thresholds of a variable

Install

Dune Dependency

Authors

Maintainers

Sources

doc/numeric_common/Numeric_common/Common/index.html

Module `Numeric_common.Common`Source

Integer intervals

Integer intervals with congruence

Rounding mode of floats

Float intervals

Fast assume on numeric conditions

Widening thresholds

package mopsa

Install

Dune Dependency

Authors

Maintainers

Sources

doc/numeric_common/Numeric_common/Common/index.html

Module Numeric_common.CommonSource

Integer intervals

Integer intervals with congruence

Rounding mode of floats

Float intervals

Fast assume on numeric conditions

Widening thresholds

Module `Numeric_common.Common`Source