Library Float.FnElem.FArgReduct

IEEE754 : FArgReduct
Sylvie Boldo
Require Export FnormI.
Require Export Classical.

Section SterbenzApprox.
Variable radix : Z.
Hypothesis radixMoreThanOne : (1 < radix)%Z.

Let FtoRradix := FtoR radix.
Coercion FtoRradix : float >-> R.
Variables (b1 : Fbound) (b2 : Fbound).
Variables (prec1 : nat) (prec2 : nat).
Hypothesis prec1MoreThanOne : 1 < prec1.
Hypothesis p1GivesBound : Zpos (vNum b1) = Zpower_nat radix prec1.
Hypothesis prec2MoreThanOne : 1 < prec2.
Hypothesis p2GivesBound : Zpos (vNum b2) = Zpower_nat radix prec2.

Theorem Rmin_1 : x y : R, (x y)%RRmin x y = x.
intros x y H; unfold Rmin in |- ×.
case (Rle_dec x y); auto with real.
Qed.

Theorem Rmin_2 : x y : R, (y x)%RRmin x y = y.
intros x y H; unfold Rmin in |- ×.
case (Rle_dec x y); auto with real.
Qed.

Theorem Rmin_eq : x : R, Rmin x x = x.
intros x; unfold Rmin in |- ×.
case (Rle_dec x x); auto with real.
Qed.

Theorem Rdiv_Rle :
  a b c d : R,
 (0 < a)%R
 (0 < b)%R
 (0 < c)%R → (0 < d)%R → (a c)%R → (d b)%R → (a / b c / d)%R.
intros a b c d Ha Hb Hc Hd H1 H2; unfold Rdiv in |- ×.
apply Rle_trans with (c × / b)%R; auto with real.
apply Rmult_le_compat_l; auto with real.
apply Rle_Rinv; auto with real.
Qed.

Theorem SterbenzApprox :
  (rho : R) (x y : float),
 (0 < rho)%R
 IZR (Zpos (vNum b1)) = (rho × Zsucc (Zpos (vNum b2)))%R
 (- dExp b2 - dExp b1)%Z
 Fbounded b1 x
 Fbounded b1 y
 (/ (1 + / rho) × y x)%R
 (x (1 + / rho) × y)%R
  u : float, u = (x - y)%R :>R Fbounded b2 u.
intros rho x y Hrho Hrho' Hv0 Fx Fy H1 H2.
cut
 ( b : Fbound,
  Z_of_nat
    (vNumInf
       (BoundI (pred (nat_of_P (vNum b))) (nat_of_P (vNum b))
          (Zabs_nat (dExp b)))) = Zpred (Zpos (vNum b)));
 [ intros V1 | idtac ].
2: intros b; apply trans_eq with (Z_of_nat (pred (nat_of_P (vNum b))));
    [ simpl in |- *; auto | idtac ].
2: apply trans_eq with (Zpred (nat_of_P (vNum b)));
    [ apply inj_pred; auto with zarith | idtac ].
2: rewrite inject_nat_convert with (Zpos (vNum b)) (vNum b);
    auto with zarith arith.
cut
 ( b : Fbound,
  Z_of_nat
    (vNumSup
       (BoundI (pred (nat_of_P (vNum b))) (nat_of_P (vNum b))
          (Zabs_nat (dExp b)))) = Zpos (vNum b)); [ intros V2 | idtac ].
2: intros b; apply trans_eq with (Z_of_nat (nat_of_P (vNum b)));
    [ simpl in |- *; auto | idtac ].
2: apply inject_nat_convert; auto with zarith arith.
elim
 SterbenzApproxI_pos
  with
    radix
    (BoundI (pred (nat_of_P (vNum b1))) (nat_of_P (vNum b1))
       (Zabs_nat (dExp b1)))
    (BoundI (pred (nat_of_P (vNum b2))) (nat_of_P (vNum b2))
       (Zabs_nat (dExp b2)))
    rho
    x
    y; auto with real zarith.
2: rewrite (V1 b1); rewrite (V1 b2); rewrite (V2 b1); rewrite (V2 b2).
2: repeat rewrite <- Zsucc_pred.
2: rewrite
    Rmin_1
           with
           (x := (Zpos (vNum b1) / Zsucc (Zpos (vNum b2)))%R);
    [ idtac | apply Rdiv_Rle; auto with real zarith ].
2: rewrite Rmin_2; [ rewrite Hrho' | idtac ].
2: unfold Rdiv in |- *; rewrite Rmult_assoc; rewrite Rinv_r;
    auto with real zarith arith.
2: case
    (Rle_or_lt (Zpos (vNum b1) / Zpos (vNum b2))
       (Zsucc (Zpos (vNum b1)) / Zsucc (Zpos (vNum b2))));
    intros H.
2: rewrite Rmin_1; auto with real; apply Rdiv_Rle; auto with real zarith.
2: rewrite Rmin_2; auto with real; apply Rdiv_Rle; auto with real zarith.
2: simpl; repeat rewrite <- Zabs_absolu.
2: repeat rewrite Zabs_eq; auto with zarith.
2: case (dExp b1); auto with zarith.
2: case (dExp b2); auto with zarith.
2: intros z Hz.
2: apply FoppBoundedI2 with (Zabs_nat (Zpower_nat radix (pred prec2)));
    auto with zarith.
2: rewrite (V1 b2); rewrite p2GivesBound; rewrite <- Zabs_absolu;
    rewrite Zabs_eq; auto with zarith arith.
2: replace (radix × Zpower_nat radix (pred prec2))%Z with
    (Zpower_nat radix prec2); auto with zarith arith.
2: pattern prec2 at 1 in |- *; replace prec2 with (1 + pred prec2);
    auto with zarith arith.
2: rewrite (V2 b2); rewrite p2GivesBound; rewrite <- Zabs_absolu;
    rewrite Zabs_eq; auto with zarith arith.
2: pattern prec2 at 1 in |- *; replace prec2 with (1 + pred prec2);
    auto with zarith arith.
2: intros z Hz.
2: apply FoppBoundedI2 with (Zabs_nat (Zpower_nat radix (pred prec1)));
    auto with zarith.
2: rewrite (V1 b1); rewrite p1GivesBound; rewrite <- Zabs_absolu;
    rewrite Zabs_eq; auto with zarith arith.
2: replace (radix × Zpower_nat radix (pred prec1))%Z with
    (Zpower_nat radix prec1); auto with zarith arith.
2: pattern prec1 at 1 in |- *; replace prec1 with (1 + pred prec1);
    auto with zarith arith.
2: rewrite (V2 b1); rewrite p1GivesBound; rewrite <- Zabs_absolu;
    rewrite Zabs_eq; auto with zarith arith.
2: pattern prec1 at 1 in |- *; replace prec1 with (1 + pred prec1);
    auto with zarith arith.
2: split; simpl in |- *; auto with zarith float.
2: split.
2: apply Zle_Zopp_Inv; rewrite Zopp_involutive;
    apply Zle_trans with (Zabs (Fnum x)).
2: apply Zle_trans with (Zabs (- Fnum x)); auto with zarith.
2: rewrite Zabs_Zopp; auto with zarith.
2: rewrite inj_pred; auto with float zarith arith.
2: rewrite inject_nat_convert with (Zpos (vNum b1)) (vNum b1);
    auto with zarith float.
2: apply Zle_trans with (Zabs (Fnum x)); auto with zarith float.
2: rewrite inject_nat_convert with (Zpos (vNum b1)) (vNum b1);
    auto with zarith float.
2: rewrite <- Zabs_absolu; rewrite Zabs_eq; auto with zarith float.
2: case (dExp b1); auto with zarith.
2: split; simpl in |- *; auto with zarith arith float.
2: split.
2: apply Zle_Zopp_Inv; rewrite Zopp_involutive;
    apply Zle_trans with (Zabs (Fnum y)).
2: apply Zle_trans with (Zabs (- Fnum y)); auto with zarith float.
2: rewrite Zabs_Zopp; auto with zarith.
2: rewrite inj_pred; auto with float zarith arith.
2: rewrite inject_nat_convert with (Zpos (vNum b1)) (vNum b1);
    auto with zarith float.
2: apply Zle_trans with (Zabs (Fnum y)); auto with zarith float.
2: rewrite inject_nat_convert with (Zpos (vNum b1)) (vNum b1);
    auto with zarith float.
2: rewrite <- Zabs_absolu; rewrite Zabs_eq; auto with zarith float.
2: case (dExp b1); auto with zarith.
intros u tmp; elim tmp; intros Hu1 Hu2; clear tmp.
elim
 ReductRangeI
  with
    radix
    (BoundI (pred (nat_of_P (vNum b2))) (nat_of_P (vNum b2))
       (Zabs_nat (dExp b2)))
    (BoundI (pred (nat_of_P (vNum b2)))
       (pred (nat_of_P (vNum b2))) (Zabs_nat (dExp b2)))
    (Zabs_nat (Zpower_nat radix (pred prec2)))
    u; auto with real zarith.
2: rewrite (V2 b2);
    apply trans_eq with (pred (nat_of_P (vNum b2)) + 1)%Z;
    [ idtac | simpl in |- *; auto ].
2: rewrite inj_pred; auto with zarith.
2: rewrite inject_nat_convert with (Zpos (vNum b2)) (vNum b2);
    auto with zarith arith; unfold Zpred in |- *; ring.
2: rewrite (V2 b2); rewrite p2GivesBound; rewrite <- Zabs_absolu;
    rewrite Zabs_eq; auto with zarith arith.
2: pattern prec2 at 1 in |- *; replace prec2 with (1 + pred prec2);
    auto with zarith arith.
intros v tmp; elim tmp; intros Hv1 Hv2; clear tmp.
v; split.
unfold FtoRradix in |- *; rewrite <- Hv2; rewrite Hu1; auto with real.
elim Hv1; intros tmp H3; elim tmp; intros H4 H5; clear tmp; split;
 auto with zarith.
case (Zle_or_lt 0 (Fnum v)); intros H'.
rewrite Zabs_eq; auto with zarith.
replace (Zpos (vNum b2)) with
 (Zsucc
    (vNumSup
       (BoundI (pred (nat_of_P (vNum b2)))
          (pred (nat_of_P (vNum b2))) (Zabs_nat (dExp b2)))));
 auto with zarith.
apply trans_eq with (Zsucc (pred (nat_of_P (vNum b2))));
 [ simpl in |- *; auto | idtac ].
rewrite inj_pred; auto with zarith.
rewrite inject_nat_convert with (Zpos (vNum b2)) (vNum b2);
 auto with zarith; rewrite <- Zsucc_pred; auto.
rewrite <- Zabs_Zopp; rewrite Zabs_eq; auto with zarith.
replace (Zpos (vNum b2)) with
 (Zsucc
    (vNumInf
       (BoundI (pred (nat_of_P (vNum b2)))
          (pred (nat_of_P (vNum b2))) (Zabs_nat (dExp b2)))));
 auto with zarith.
apply trans_eq with (Zsucc (pred (nat_of_P (vNum b2))));
 [ simpl in |- *; auto | idtac ].
rewrite inj_pred; auto with zarith.
rewrite inject_nat_convert with (Zpos (vNum b2)) (vNum b2);
 auto with zarith.
apply Zle_trans with (2:=H3); simpl.
rewrite <- Zabs_absolu; rewrite Zabs_eq; auto with zarith;
  case (dExp b2); auto with zarith.
Qed.

Theorem SterbenzApprox_weak_1 :
  (rho : R) (x y : float),
 (0 < rho)%R
 IZR (Zpos (vNum b1)) = (rho × Zpos (vNum b2))%R
 (- dExp b2 - dExp b1)%Z
 Fbounded b1 x
 Fbounded b1 y
 (0 y)%R
 (y x)%R
 (x (1 + / rho) × y)%R
 Fbounded b2
   (Fminus radix (Fnormalize radix b1 prec1 x) (Fnormalize radix b1 prec1 y)).
intros rho x y H1 H2 H'' H3 H4 H5 H6 H7.
cut
 (0
  Fminus radix (Fnormalize radix b1 prec1 x) (Fnormalize radix b1 prec1 y))%R;
 [ intros Rle1 | idtac ].
2: rewrite (Fminus_correct radix); auto with zarith; fold FtoRradix in |- ×.
2: unfold FtoRradix in |- *; repeat rewrite FnormalizeCorrect; auto.
2: apply Rplus_le_reg_l with (r := FtoR radix y); auto.
2: fold FtoRradix in |- *; ring_simplify; auto with real.
cut
 (Fexp
    (Fminus radix (Fnormalize radix b1 prec1 x) (Fnormalize radix b1 prec1 y)) =
  Fexp (Fnormalize radix b1 prec1 y)); [ intros He | idtac ].
2: unfold Fminus in |- *; simpl in |- *; apply Zmin_le2.
2: apply Fcanonic_Rle_Zle with radix b1 prec1; auto with real zarith float.
2: repeat rewrite FnormalizeCorrect; auto.
2: fold FtoRradix in |- *; repeat rewrite Rabs_right; auto with real.
2: apply Rle_ge; apply Rle_trans with (FtoRradix y); auto with real.
split.
rewrite Zabs_eq; [ idtac | apply LeR0Fnum with radix; auto with real zarith ].
apply Zlt_Rlt;
 apply
  Rle_lt_trans
   with ((x - y) × powerRZ radix (- Fexp (Fnormalize radix b1 prec1 y)))%R.
right;
 apply
  trans_eq
   with
     (Fminus radix (Fnormalize radix b1 prec1 x)
        (Fnormalize radix b1 prec1 y) ×
      powerRZ radix
        (-
         Fexp
           (Fminus radix (Fnormalize radix b1 prec1 x)
              (Fnormalize radix b1 prec1 y))))%R.
unfold FtoRradix, FtoR in |- *; rewrite Rmult_assoc.
rewrite <- powerRZ_add; auto with zarith real.
ring_simplify
    (Fexp
       (Fminus radix (Fnormalize radix b1 prec1 x)
          (Fnormalize radix b1 prec1 y)) +
     -
     Fexp
       (Fminus radix (Fnormalize radix b1 prec1 x)
          (Fnormalize radix b1 prec1 y)))%Z; simpl in |- *;
 ring.
unfold FtoRradix in |- *; rewrite Fminus_correct; auto with zarith.
rewrite <- FnormalizeCorrect with radix b1 prec1 x; auto.
rewrite <- FnormalizeCorrect with radix b1 prec1 y; auto.
rewrite He; ring.
apply
 Rle_lt_trans
  with
    (/ rho × FtoRradix y ×
     powerRZ radix (- Fexp (Fnormalize radix b1 prec1 y)))%R.
apply Rmult_le_compat_r; auto with real zarith.
apply Rplus_le_reg_l with (FtoRradix y); auto.
ring_simplify (FtoRradix y + (FtoRradix x - FtoRradix y))%R.
apply Rle_trans with (1 := H7); right; ring.
apply Rle_lt_trans with (/ rho × Fnum (Fnormalize radix b1 prec1 y))%R.
unfold FtoRradix in |- *; rewrite <- FnormalizeCorrect with radix b1 prec1 y;
 auto.
right; unfold FtoR in |- ×.
apply
 trans_eq
  with
    (/ rho ×
     (powerRZ radix (- Fexp (Fnormalize radix b1 prec1 y)) ×
      powerRZ radix (Fexp (Fnormalize radix b1 prec1 y))) ×
     Fnum (Fnormalize radix b1 prec1 y))%R.
ring.
rewrite <- powerRZ_add; auto with zarith real.
ring_simplify
    (- Fexp (Fnormalize radix b1 prec1 y) +
     Fexp (Fnormalize radix b1 prec1 y))%Z; simpl in |- *;
 ring.
apply Rlt_le_trans with (/ rho × Zpos (vNum b1))%R.
apply Rmult_lt_compat_l; auto with real.
rewrite <- (Zabs_eq (Fnum (Fnormalize radix b1 prec1 y)));
 auto with real zarith float.
apply LeR0Fnum with radix; auto with zarith real.
rewrite FnormalizeCorrect; auto with real float zarith.
right; rewrite H2; rewrite <- Rmult_assoc; rewrite Rinv_l;
 [ ring | auto with zarith real ].
apply Zle_trans with (1 := H''); auto with zarith; rewrite He;
 auto with float zarith.
Qed.

Theorem SterbenzApprox2 :
  (rho : R) (x y : float),
 (0 < rho)%R
 IZR (Zpos (vNum b1)) = (rho × Zpos (vNum b2))%R
 (- dExp b2 - dExp b1)%Z
 Fbounded b1 x
 Fbounded b1 y
 (/ (1 + / rho) × y x)%R
 (x (1 + / rho) × y)%R
 Fbounded b2
   (Fminus radix (Fnormalize radix b1 prec1 x) (Fnormalize radix b1 prec1 y)).
intros rho x y Hrho Hv1 Hv2 F1x F1y H1 H2.
cut (0 < 1 + / rho)%R; [ intros H' | idtac ].
2: apply Rlt_trans with (1 + 0)%R;
    [ ring_simplify (1 + 0)%R | apply Rplus_lt_compat_l ]; auto with real.
cut (y (1 + / rho) × x)%R; [ intros H'1 | idtac ].
2: apply Rmult_le_reg_l with (/ (1 + / rho))%R; auto with real.
2: rewrite <- Rmult_assoc; rewrite Rinv_l; auto with real; ring_simplify (1 × x)%R;
    auto with real.
cut (0 y)%R; [ intros H'2 | idtac ].
2: apply Rmult_le_reg_l with (1 + / rho - / (1 + / rho))%R.
2: apply Rplus_lt_reg_r with (/ (1 + / rho))%R.
2: ring_simplify.
2: apply Rmult_lt_reg_l with (1 + / rho)%R; auto with real.
2: rewrite Rinv_r; auto with real.
2: apply Rlt_le_trans with (1+(/rho+(/rho+/rho×/rho)))%R;[idtac|right; ring].
2: apply Rle_lt_trans with (1 + (0 + (0 + 0)))%R;
    [ right; ring | apply Rplus_lt_compat_l; auto with real ].
2: apply Rplus_lt_compat; auto with real.
2: apply Rplus_lt_compat; auto with real.
2: apply Rmult_lt_0_compat; auto with real.
2: ring_simplify ((1 + / rho - / (1 + / rho)) × 0)%R.
2: apply Rplus_le_reg_l with (/ (1 + / rho) × FtoRradix y)%R.
2: ring_simplify; auto with real.
2: apply Rle_trans with (FtoRradix x); auto with real.
2: apply Rle_trans with (1 := H2); right; ring.
cut (0 x)%R; [ intros H'3 | idtac ].
2: apply Rle_trans with (2 := H1); apply Rmult_le_pos; auto with real.
case (Rle_or_lt y x); intros H3.
apply SterbenzApprox_weak_1 with rho; auto with real float.
apply oppBoundedInv; rewrite Fopp_Fminus.
apply SterbenzApprox_weak_1 with rho; auto with real float.
Qed.

End SterbenzApprox.

Section RinvProps.
Variable radix : Z.
Hypothesis radixMoreThanOne : (1 < radix)%Z.

Let FtoRradix := FtoR radix.
Coercion FtoRradix : float >-> R.
Variable b : Fbound.
Variable prec : nat.
Hypothesis precMoreThanOne : 1 < prec.
Hypothesis pGivesBound : Zpos (vNum b) = Zpower_nat radix prec.

Theorem Rle_floats_isMax_Pos :
  (f g : float) (r : R),
 Fcanonic radix b f
 Fcanonic radix b g
 (0 < f)%R
 isMax b radix r g
 (f - Fulp b radix prec (FPred b radix prec f) < r)%R → (f g)%R.
intros f g r Hf Hg Hf' H1 H.
rewrite <- FSucPred with b radix prec f; auto with arith.
unfold FtoRradix in |- *; apply FSuccProp; fold FtoRradix in |- *;
 auto with arith float.
apply Rlt_le_trans with r; auto with real float.
2: elim H1; intros H2 H3; elim H3; auto with real.
apply Rle_lt_trans with (2 := H).
apply Rplus_le_reg_l with (Fulp b radix prec (FPred b radix prec f)).
right; ring_simplify (Fulp b radix prec (FPred b radix prec f) +
 (f - Fulp b radix prec (FPred b radix prec f)))%R.
rewrite Rplus_comm; unfold FtoRradix in |- *; apply FpredUlpPos;
 auto with float real zarith.
Qed.

Theorem Rle_floats_isMax_Neg :
  (f g : float) (r : R),
 Fcanonic radix b f
 Fcanonic radix b g
 (f < 0)%R
 isMax b radix r g → (f - Fulp b radix prec f < r)%R → (f g)%R.
intros f g r Hf Hg Hf' H1 H.
rewrite <- FSucPred with b radix prec f; auto with arith.
unfold FtoRradix in |- *; apply FSuccProp; fold FtoRradix in |- *;
 auto with arith float.
apply Rlt_le_trans with r; auto with real float.
2: elim H1; intros H2 H3; elim H3; auto with real.
apply Rle_lt_trans with (2 := H).
apply
 Rplus_le_reg_l
  with (Fulp b radix prec f - FtoRradix (FPred b radix prec f))%R.
right; ring_simplify.
apply trans_eq with (Fulp b radix prec (Fopp f)).
repeat rewrite CanonicFulp; auto with float zarith.
rewrite <- FSuccUlpPos; auto with float real;
 [ idtac | rewrite Fopp_correct; auto with real ].
rewrite Fminus_correct; auto with zarith; rewrite FPredFopFSucc;
 auto with arith.
unfold FtoRradix in |- *; repeat rewrite Fopp_correct; ring.
Qed.

Theorem FulpFPred_not_pow :
  f : float,
 ( e : Z, FtoRradix f powerRZ radix e) →
 (0 < f)%R
 Fcanonic radix b f
 Fulp b radix prec (FPred b radix prec f) = Fulp b radix prec f.
intros f H Hfpos Hf.
cut (Fcanonic radix b (FPred b radix prec f));
 [ intros H1 | apply FPredCanonic; auto with arith ].
repeat rewrite CanonicFulp; auto.
unfold FtoR in |- *; simpl in |- *; ring_simplify.
generalize (Z_eq_bool_correct (Fnum f) (- pPred (vNum b)));
 case (Z_eq_bool (Fnum f) (- pPred (vNum b))); intros H2.
Contradict Hfpos; apply Rle_not_lt; unfold FtoRradix in |- *;
 apply LeZEROFnum; auto.
rewrite H2; unfold pPred in |- *; replace 0%Z with (- (0))%Z;
 auto with zarith float.
generalize (Z_eq_bool_correct (Fnum f) (nNormMin radix prec));
 case (Z_eq_bool (Fnum f) (nNormMin radix prec)); intros H3.
unfold nNormMin in H3; Contradict H.
apply
 ex_not_not_all
  with
    (U := Z)
    (P := fun t : ZFtoRradix f powerRZ radix t).
(Fexp f + Zpred prec)%Z.
cut (FtoRradix f = powerRZ radix (Fexp f + Zpred prec)); auto with real.
unfold FtoRradix, FtoR in |- *; simpl in |- *; rewrite powerRZ_add;
 auto with zarith real.
rewrite H3; rewrite Zpower_nat_Z_powerRZ; rewrite inj_pred; auto with zarith;
 ring.
rewrite FPredSimpl4; auto.
Qed.

Theorem RinvClosestRinvMaxRle_Pos :
 radix = 2%Z
  a u v : float,
 ( e : Z, FtoRradix a powerRZ radix e) →
 Fbounded b a
 Fbounded b u
 Fnormal radix b (Fnormalize radix b prec u) →
 Fcanonic radix b a
 Fcanonic radix b v
 Closest b radix (/ a) u
 isMax b radix (/ u) v → (0 < a)%R → (0 < u)%R → (a v)%R.
intros Hradix a u v Hpow Fa Fu Nu Ca Cv Hu Hv H H'.
apply Rle_floats_isMax_Pos with (/ u)%R; auto; unfold Rminus in |- ×.
rewrite FulpFPred_not_pow; auto.
cut (0 < powerRZ radix prec - 1)%R; [ intros V0 | idtac ].
2: apply Rplus_lt_reg_r with 1%R.
2: ring_simplify.
2: replace 1%R with (powerRZ radix 0); auto with real zarith arith.
apply Rle_lt_trans with (FtoRradix a + - (a × / (powerRZ radix prec - 1)))%R;
 [ apply Rplus_le_compat_l; apply Ropp_le_contravar | idtac ].
apply Rmult_le_reg_l with (powerRZ radix prec - 1)%R; auto with real zarith.
rewrite Rmult_comm; rewrite Rmult_assoc; rewrite Rinv_l; auto with real.
ring_simplify (FtoRradix a × 1)%R.
apply Rle_trans with (Rabs a); [ apply RRle_abs | idtac ].
unfold FtoRradix in |- *; apply FulpGe; auto with real zarith.
apply Rle_lt_trans with (FtoRradix a × (1 + - / (powerRZ radix prec - 1)))%R;
 [ right; ring; ring | idtac ].
cut (0 < 1 + - / (powerRZ radix prec - 1))%R; [ intros V1 | idtac ].
2: apply Rplus_lt_reg_r with (/ (powerRZ radix prec - 1))%R.
2: ring_simplify.
2: pattern 1%R at 2 in |- *; rewrite <- Rinv_1; apply Rinv_lt_contravar;
    auto with real.
2: apply Rlt_le_trans with (1 := V0); right; ring.
2: apply Rplus_lt_reg_r with 1%R; ring_simplify (1 + (powerRZ radix prec - 1))%R.
2: apply Rle_lt_trans with (IZR radix); auto with zarith real.
2: replace 2%R with (IZR 2); auto with real zarith.
2: pattern (IZR radix) at 1 in |- *;
    replace (IZR radix) with (powerRZ radix 1); auto with real zarith.
cut (0 < 1 + - / (2%nat × powerRZ radix (Zpred prec)))%R;
 [ intros V2 | idtac ].
2: apply Rplus_lt_reg_r with (/ (2%nat × powerRZ radix (Zpred prec)))%R.
2: ring_simplify.
2: rewrite <- Rinv_1; apply Rinv_lt_contravar; auto with real.
2: repeat apply Rmult_lt_0_compat; auto with arith zarith real.
2: apply Rlt_trans with (powerRZ radix (Zpred prec)).
2: replace 1%R with (powerRZ radix (Zpred 1)); auto with real zarith arith.
2: apply Rle_lt_trans with (1 × powerRZ radix (Zpred prec))%R;
    [ right; ring | auto with real arith ].
apply
 Rle_lt_trans
  with
    (/ FtoRradix u × / (1 + - / (2%nat × powerRZ radix (Zpred prec))) ×
     (1 + - / (powerRZ radix prec - 1)))%R.
apply Rmult_le_compat_r; auto with real.
rewrite <- Rinv_involutive with (FtoRradix a); auto with real.
rewrite <- Rinv_mult_distr; auto with real.
apply Rle_Rinv; [ apply Rmult_lt_0_compat; auto with real | idtac ].
rewrite Rinv_mult_distr; auto with real arith zarith.
apply Rle_trans with (FtoRradix u + - (/ 2%nat × Fulp b radix prec u))%R.
apply Rle_trans with (u + - (/ 2%nat *( /powerRZ radix (Zpred prec)×u)))%R;
   [right; ring|idtac].
apply Rplus_le_compat_l; apply Ropp_le_contravar.
apply Rmult_le_compat_l; auto with real arith.
apply Rle_trans with (Rabs u × powerRZ radix (Zsucc (- prec)))%R.
unfold FtoRradix in |- *; apply FulpLe2; auto with float.
rewrite Rinv_powerRZ; auto with zarith real.
rewrite Rabs_right with (FtoRradix u);
 [ idtac | apply Rle_ge; auto with real ].
replace (Zsucc (- prec)) with (- Zpred prec)%Z;
 [ right | unfold Zsucc, Zpred in |- × ]; ring.
apply Rplus_le_reg_l with (/ 2%nat × Fulp b radix prec u + - / FtoRradix a)%R.
ring_simplify.
apply Rle_trans with (Rabs (- / FtoRradix a + FtoRradix u));
 [ apply RRle_abs | idtac ].
rewrite <- Rabs_Ropp.
replace (- (- / FtoRradix a + FtoRradix u))%R with (/ FtoRradix a - u)%R;
 [ idtac | ring ].
apply Rmult_le_reg_l with (INR 2); auto with arith real.
apply Rle_trans with (Fulp b radix prec u).
unfold FtoRradix in |- *; apply ClosestUlp; auto.
right; simpl; field.
rewrite Rmult_assoc;
  apply Rlt_le_trans with (/u × 1)%R;[ apply Rmult_lt_compat_l; auto with real| right; ring].
apply Rmult_lt_reg_l with (1 + - / (2%nat × powerRZ radix (Zpred prec)))%R;
 auto with real.
rewrite <- Rmult_assoc; rewrite Rinv_r; auto with real.
ring_simplify.
apply Rplus_lt_compat_r; apply Ropp_lt_contravar; apply Rinv_lt_contravar.
apply Rmult_lt_0_compat; auto with real arith zarith.
apply Rmult_lt_0_compat; auto with real arith zarith.
apply Rlt_le_trans with (powerRZ radix prec - 0)%R; unfold Rminus in |- *;
 auto with real.
ring_simplify (powerRZ radix prec + -0)%R.
pattern (Z_of_nat prec) at 1 in |- *;
 replace (Z_of_nat prec) with (1 + Zpred prec)%Z;
 [ rewrite powerRZ_add; auto with real zarith | unfold Zpred in |- *; ring ].
apply Rmult_le_compat_r; auto with real zarith; simpl in |- ×.
ring_simplify (radix × 1)%R; rewrite Hradix; auto with real zarith.
Qed.

Theorem RinvClosestRinvMaxRle_Neg :
 radix = 2%Z
  a u v : float,
 Fbounded b a
 Fbounded b u
 Fnormal radix b (Fnormalize radix b prec u) →
 Fcanonic radix b a
 Fcanonic radix b v
 Closest b radix (/ a) u
 isMax b radix (/ u) v → (a < 0)%R → (u < 0)%R → (a v)%R.
intros Hradix a u v Fa Fu Nu Ca Cv Hu Hv H H'.
apply Rle_floats_isMax_Neg with (/ u)%R; auto; unfold Rminus in |- ×.
cut (0 < powerRZ radix prec - 1)%R; [ intros V0 | idtac ].
2: apply Rplus_lt_reg_r with 1%R.
2: ring_simplify.
2: replace 1%R with (powerRZ radix 0); auto with real zarith arith.
apply Rle_lt_trans with (FtoRradix a + a × / (powerRZ radix prec - 1))%R;
 [ apply Rplus_le_compat_l | idtac ].
apply Ropp_le_cancel; rewrite Ropp_involutive;
 rewrite <- Ropp_mult_distr_l_reverse.
apply Rmult_le_reg_l with (powerRZ radix prec - 1)%R; auto with real zarith.
rewrite Rmult_comm; rewrite Rmult_assoc; rewrite Rinv_l; auto with real.
ring_simplify (- FtoRradix a × 1)%R; rewrite <- Rabs_left; auto with real.
unfold FtoRradix in |- *; apply FulpGe; auto with real zarith.
apply Rle_lt_trans with (FtoRradix a × (1 + / (powerRZ radix prec - 1)))%R;
 [ right; ring; ring | idtac ].
cut (0 < 1 + / (powerRZ radix prec - 1))%R; [ intros V1 | idtac ].
2: apply Rlt_trans with (1 + 0)%R;
    [ ring_simplify (1 + 0)%R | apply Rplus_lt_compat_l ]; auto with real.
cut (0 < 1 + / (2%nat × powerRZ radix (Zpred prec)))%R; [ intros V2 | idtac ].
2: apply Rlt_trans with (1 + 0)%R;
    [ ring_simplify (1 + 0)%R | apply Rplus_lt_compat_l ]; auto with real.
2: apply Rinv_0_lt_compat; apply Rmult_lt_0_compat;
    auto with real arith zarith.
apply
 Rle_lt_trans
  with
    (/ FtoRradix u × / (1 + / (2%nat × powerRZ radix (Zpred prec))) ×
     (1 + / (powerRZ radix prec - 1)))%R.
apply Rmult_le_compat_r; auto with real.
rewrite <- Rinv_involutive with (FtoRradix a); auto with real.
rewrite <- Rinv_mult_distr; auto with real.
apply Ropp_le_cancel; rewrite Ropp_inv_permute; auto with real.
rewrite Ropp_inv_permute with (/ FtoRradix a)%R; auto with real.
apply Rle_Rinv; [ auto with real | apply Ropp_le_contravar ].
rewrite Rinv_mult_distr; auto with real arith zarith.
apply Rle_trans with (FtoRradix u + - (/ 2%nat × Fulp b radix prec u))%R.
apply Rle_trans with (u+(/2%nat*(/ powerRZ radix (Zpred prec) × u)))%R;
  [right; ring|idtac].
apply Rplus_le_compat_l; apply Ropp_le_cancel; rewrite Ropp_involutive.
apply
 Rle_trans with (/ 2%nat × (/ powerRZ radix (Zpred prec) × - FtoRradix u))%R;
 [ apply Rmult_le_compat_l; auto with real arith | right; ring ].
apply Rle_trans with (Rabs u × powerRZ radix (Zsucc (- prec)))%R.
unfold FtoRradix in |- *; apply FulpLe2; auto with float.
rewrite Rinv_powerRZ; auto with zarith real.
rewrite Rabs_left with (FtoRradix u); [ idtac | auto with real ].
replace (Zsucc (- prec)) with (- Zpred prec)%Z;
 [ right | unfold Zsucc, Zpred in |- × ]; ring.
apply Rplus_le_reg_l with (/ 2%nat × Fulp b radix prec u + - / FtoRradix a)%R.
ring_simplify.
apply Rle_trans with (Rabs (- / FtoRradix a + FtoRradix u));
 [ apply RRle_abs | idtac ].
rewrite <- Rabs_Ropp.
replace (- (- / FtoRradix a + FtoRradix u))%R with (/ FtoRradix a - u)%R;
 [ idtac | ring ].
apply Rmult_le_reg_l with (INR 2); auto with arith real.
apply Rle_trans with (Fulp b radix prec u).
unfold FtoRradix in |- *; apply ClosestUlp; auto.
right; simpl; field.
apply Ropp_lt_cancel; rewrite Ropp_inv_permute; auto with real.
rewrite <- Ropp_mult_distr_l_reverse; rewrite <- Ropp_mult_distr_l_reverse;
 rewrite Ropp_inv_permute; auto with real.
pattern (/ - FtoRradix u)%R at 1 in |- *;
 replace (/ - FtoRradix u)%R with (/ - FtoRradix u × 1)%R;
 [ idtac | ring ].
rewrite Rmult_assoc; apply Rmult_lt_compat_l; auto with real.
apply Rmult_lt_reg_l with (1 + / (2%nat × powerRZ radix (Zpred prec)))%R;
 auto with real.
rewrite <- Rmult_assoc; rewrite Rinv_r; auto with real.
ring_simplify.
apply Rplus_lt_compat_r; apply Rinv_lt_contravar; auto with real.
apply Rmult_lt_0_compat; auto with real arith zarith.
apply Rmult_lt_0_compat; auto with real arith zarith.
apply Rlt_le_trans with (powerRZ radix prec - 0)%R; unfold Rminus in |- *;
 auto with real.
ring_simplify (powerRZ radix prec + -0)%R.
pattern (Z_of_nat prec) at 1 in |- *;
 replace (Z_of_nat prec) with (1 + Zpred prec)%Z;
 [ rewrite powerRZ_add; auto with real zarith | unfold Zpred in |- *; ring ].
apply Rmult_le_compat_r; auto with real zarith; simpl in |- ×.
ring_simplify (radix × 1)%R; rewrite Hradix; auto with real zarith.
Qed.

Theorem RinvClosestRinvMaxRle_pow :
  (a u v : float) (e : Z),
 a = powerRZ radix e :>R
 (e dExp b)%Z
 Fbounded b a
 Closest b radix (/ a) uisMax b radix (/ u) va = v :>R.
intros a u v e Ha He Fa Hu Hv.
cut (ProjectorP b radix (isMax b radix));
 [ unfold ProjectorP in |- *; intros V | apply ProjectMax ].
unfold FtoR in |- *; apply V; auto.
replace (FtoR radix a) with (/ FtoRradix u)%R; auto.
replace (FtoRradix u) with (powerRZ radix (- e)); fold FtoRradix in |- ×.
rewrite Ha; rewrite Rinv_powerRZ; auto with zarith real; ring_simplify (- - e)%Z; auto.
apply trans_eq with (FtoRradix (Float 1 (- e)));
 [ unfold FtoRradix, FtoR in |- *; simpl in |- *; ring | idtac ].
unfold FtoRradix in |- *;
 apply RoundedModeProjectorIdemEq with b prec (Closest b radix);
 auto.
apply ClosestRoundedModeP with prec; auto.
split; simpl in |- *; auto with zarith; rewrite pGivesBound;
 replace 1%Z with (Zpower_nat radix 0); auto with zarith.
replace (FtoR radix (Float 1 (- e))) with (/ FtoRradix a)%R; auto.
rewrite Ha; rewrite Rinv_powerRZ; auto with zarith real; unfold FtoR in |- *;
 simpl in |- *; ring.
Qed.

Theorem RinvClosestRinvMaxRle_pow2 :
  (a u v : float) (e : Z),
 a = powerRZ radix e :>R
 Fnormal radix b (Fnormalize radix b prec u) →
 Fbounded b a
 Closest b radix (/ a) uisMax b radix (/ u) va = v :>R.
intros a u v e Ha Nu Fa Hu Hv.
apply RinvClosestRinvMaxRle_pow with u e; auto.
case (Zle_or_lt e (dExp b)); intros H; auto with zarith.
absurd (u powerRZ radix (- dExp b))%R.
apply Rlt_not_le;
 apply Rlt_le_trans with (FtoRradix (firstNormalPos radix b prec)).
unfold firstNormalPos, nNormMin, FtoRradix, FtoR in |- *; simpl in |- *;
 rewrite Zpower_nat_Z_powerRZ.
apply Rle_lt_trans with (powerRZ radix 0 × powerRZ radix (- dExp b))%R;
 [ right; simpl in |- *; ring | idtac ].
apply Rmult_lt_compat_r; auto with real zarith; rewrite inj_pred;
 auto with zarith arith.
apply Rlt_powerRZ; unfold Zpred in |- *; auto with arith zarith real.
unfold FtoRradix in |- *;
 rewrite <- FnormalizeCorrect with (p := u) (b := b) (precision := prec);
 auto.
apply FnormalLtFirstNormalPos; auto with arith.
rewrite FnormalizeCorrect; auto.
apply RleRoundedR0 with b prec (Closest b radix) (/ a)%R; auto;
 [ apply ClosestRoundedModeP with prec; auto
 | rewrite Ha; auto with real zarith ].
cut (MonotoneP radix (Closest b radix));
 [ unfold MonotoneP in |- *; intros V | apply ClosestMonotone ].
apply Rle_trans with (FtoRradix (Float 1 (- dExp b)));
 unfold FtoRradix in |- *;
 [ idtac | right; unfold FtoR in |- *; simpl in |- *; ring ].
unfold FtoRradix in |- *;
 apply V with (/ a)%R (FtoRradix (Float 1 (- dExp b)));
 auto.
rewrite Ha; rewrite Rinv_powerRZ; unfold FtoRradix, FtoR in |- *;
 simpl in |- *; auto with real zarith.
ring_simplify (1 × powerRZ radix (- dExp b))%R; auto with real zarith.
unfold FtoRradix in |- *; apply RoundedModeProjectorIdem with b;
 auto with float zarith.
apply ClosestRoundedModeP with prec; auto.
split; simpl in |- *; auto with zarith float.
rewrite pGivesBound; replace 1%Z with (Zpower_nat radix 0);
 auto with arith zarith real.
Qed.

Theorem RinvClosestRinvMaxRle :
 radix = 2%Z
  a u v : float,
 Fbounded b a
 Fbounded b u
 Fnormal radix b (Fnormalize radix b prec u) →
 Fcanonic radix b a
 Fcanonic radix b v
 Closest b radix (/ a) u
 isMax b radix (/ u) va 0%R :>R → (a v)%R.
intros Hradix a u v Fa Fu Nu Ca Cv Hu Hv Ha.
cut ( a b : R, (a b)%Ra b → (a < b)%R); [ intros V | idtac ].
2: intros x y H'1 H'2; case H'1; auto with real; intros H'3.
2: absurd (x = y :>R); auto with real.
cut (u 0%R :>R); [ intros H' | idtac ].
2: unfold FtoRradix in |- *;
    rewrite <- FnormalizeCorrect with (b := b) (precision := prec);
    auto.
2: cut
    (¬ is_Fzero (Fnormalize radix b prec u) →
     FtoR radix (Fnormalize radix b prec u) 0%R :>R);
    [ intros V'; apply V' | idtac ].
2: apply FnormalNotZero with radix b; auto with float real.
2: unfold is_Fzero in |- *; intros H2; unfold FtoR in |- *; simpl in |- ×.
2: apply prod_neq_R0; auto with real zarith.
case (Rle_or_lt a 0); intros H.
case H; intros H1; [ idtac | absurd (FtoRradix a = 0%R :>R); auto with real ].
apply RinvClosestRinvMaxRle_Neg with u; auto.
apply V; auto; unfold FtoRradix in |- ×.
apply RleRoundedLessR0 with b prec (Closest b radix) (/ a)%R;
 auto with float zarith real.
apply ClosestRoundedModeP with prec; auto.
case (classic ( e : Z, FtoRradix a powerRZ radix e :>R));
 intros H1.
apply RinvClosestRinvMaxRle_Pos with u; auto with real.
apply V; auto; unfold FtoRradix in |- ×.
apply RleRoundedR0 with b prec (Closest b radix) (/ a)%R;
 auto with float zarith real.
apply ClosestRoundedModeP with prec; auto.
cut ( e : Z, FtoRradix a = powerRZ radix e :>R);
 [ intros H2 | idtac ].
elim H2; intros e H3.
right; apply RinvClosestRinvMaxRle_pow2 with u e; auto.
apply
 not_all_not_ex
  with
    (U := Z)
    (P := fun t : ZFtoRradix a = powerRZ radix t :>R);
 auto.
Qed.

End RinvProps.