def f (x : Nat) := x

theorem ex1 (h₁ : a = 0) (h₂ : a + b = 0) : f (b + c) = c := by
  simp_all
  simp [f]

theorem ex2 (h₁ : a = 0) (h₂ : a + b = 0) : f (b + c) = c := by
  simp_all [f]

theorem ex3 (h₁ : a = 0) (h₂ : a + b = 0) : f (b + c) = c := by
  simp_all only [f]
  rw [Nat.zero_add] at h₂
  simp [h₂]

theorem ex4 (h₁ : a = f a) (h₂ : b + 0 = 0) : f b = 0 := by
  simp_all (config := { decide := true }) [-h₁]

theorem ex5 (h₁ : a = f a) (h₂ : b + 0 = 0) : f (b + a) = a := by
  simp_all [-h₁, f]

/-!
Prior to lean4#2334, `simp_all` would unnecessarily reorder hypotheses,
even when it did not do any simplification.
See lean4#2402.

This test verifies that the last hypothesis stays in the last position
through the `simp_all`.
-/
example : ∀ {A : Prop} (_ : A) (_ : W), W := by
  intros
  simp_all (config := { failIfUnchanged := false })
  rename_i w
  exact w