make matching on NaN a hard error

Author: RalfJung

compiler/rustc_middle/src/mir/interpret/value.rs

@@ -413,7 +413,7 @@ impl<'tcx, Prov: Provenance> Scalar<Prov> {
     #[inline]
     pub fn to_float<F: Float>(self) -> InterpResult<'tcx, F> {
         // Going through `to_uint` to check size and truncation.
-        Ok(F::from_bits(self.to_uint(Size::from_bits(F::BITS))?))
+        Ok(F::from_bits(self.to_bits(Size::from_bits(F::BITS))?))
     }
 
     #[inline]

compiler/rustc_middle/src/ty/consts/int.rs

@@ -249,11 +249,6 @@ impl ScalarInt {
         }
     }
 
-    #[inline]
-    pub fn try_to_target_usize(&self, tcx: TyCtxt<'_>) -> Result<u64, Size> {
-        Ok(self.to_bits(tcx.data_layout.pointer_size)? as u64)
-    }
-
     /// Tries to convert the `ScalarInt` to an unsigned integer of the given size.
     /// Fails if the size of the `ScalarInt` is not equal to `size` and returns the
     /// `ScalarInt`s size in that case.
@@ -262,56 +257,61 @@ impl ScalarInt {
         self.to_bits(size)
     }
 
-    // Tries to convert the `ScalarInt` to `bool`. Fails if the `size` of the `ScalarInt`
-    // in not equal to `Size { raw: 1 }` or if the value is not 0 or 1 and returns the `size`
-    // value of the `ScalarInt` in that case.
-    #[inline]
-    pub fn try_to_bool(self) -> Result<bool, Size> {
-        match self.try_to_u8()? {
-            0 => Ok(false),
-            1 => Ok(true),
-            _ => Err(self.size()),
-        }
-    }
-
     // Tries to convert the `ScalarInt` to `u8`. Fails if the `size` of the `ScalarInt`
     // in not equal to `Size { raw: 1 }` and returns the `size` value of the `ScalarInt` in
     // that case.
     #[inline]
     pub fn try_to_u8(self) -> Result<u8, Size> {
-        self.to_bits(Size::from_bits(8)).map(|v| u8::try_from(v).unwrap())
+        self.try_to_uint(Size::from_bits(8)).map(|v| u8::try_from(v).unwrap())
     }
 
     /// Tries to convert the `ScalarInt` to `u16`. Fails if the size of the `ScalarInt`
     /// in not equal to `Size { raw: 2 }` and returns the `size` value of the `ScalarInt` in
     /// that case.
     #[inline]
     pub fn try_to_u16(self) -> Result<u16, Size> {
-        self.to_bits(Size::from_bits(16)).map(|v| u16::try_from(v).unwrap())
+        self.try_to_uint(Size::from_bits(16)).map(|v| u16::try_from(v).unwrap())
     }
 
     /// Tries to convert the `ScalarInt` to `u32`. Fails if the `size` of the `ScalarInt`
     /// in not equal to `Size { raw: 4 }` and returns the `size` value of the `ScalarInt` in
     /// that case.
     #[inline]
     pub fn try_to_u32(self) -> Result<u32, Size> {
-        self.to_bits(Size::from_bits(32)).map(|v| u32::try_from(v).unwrap())
+        self.try_to_uint(Size::from_bits(32)).map(|v| u32::try_from(v).unwrap())
     }
 
     /// Tries to convert the `ScalarInt` to `u64`. Fails if the `size` of the `ScalarInt`
     /// in not equal to `Size { raw: 8 }` and returns the `size` value of the `ScalarInt` in
     /// that case.
     #[inline]
     pub fn try_to_u64(self) -> Result<u64, Size> {
-        self.to_bits(Size::from_bits(64)).map(|v| u64::try_from(v).unwrap())
+        self.try_to_uint(Size::from_bits(64)).map(|v| u64::try_from(v).unwrap())
     }
 
     /// Tries to convert the `ScalarInt` to `u128`. Fails if the `size` of the `ScalarInt`
     /// in not equal to `Size { raw: 16 }` and returns the `size` value of the `ScalarInt` in
     /// that case.
     #[inline]
     pub fn try_to_u128(self) -> Result<u128, Size> {
-        self.to_bits(Size::from_bits(128))
+        self.try_to_uint(Size::from_bits(128))
+    }
+
+    #[inline]
+    pub fn try_to_target_usize(&self, tcx: TyCtxt<'_>) -> Result<u64, Size> {
+        self.try_to_uint(tcx.data_layout.pointer_size).map(|v| u64::try_from(v).unwrap())
+    }
+
+    // Tries to convert the `ScalarInt` to `bool`. Fails if the `size` of the `ScalarInt`
+    // in not equal to `Size { raw: 1 }` or if the value is not 0 or 1 and returns the `size`
+    // value of the `ScalarInt` in that case.
+    #[inline]
+    pub fn try_to_bool(self) -> Result<bool, Size> {
+        match self.try_to_u8()? {
+            0 => Ok(false),
+            1 => Ok(true),
+            _ => Err(self.size()),
+        }
     }
 
     /// Tries to convert the `ScalarInt` to a signed integer of the given size.
@@ -357,6 +357,27 @@ impl ScalarInt {
     pub fn try_to_i128(self) -> Result<i128, Size> {
         self.try_to_int(Size::from_bits(128))
     }
+
+    #[inline]
+    pub fn try_to_target_isize(&self, tcx: TyCtxt<'_>) -> Result<i64, Size> {
+        self.try_to_int(tcx.data_layout.pointer_size).map(|v| i64::try_from(v).unwrap())
+    }
+
+    #[inline]
+    pub fn try_to_float<F: Float>(self) -> Result<F, Size> {
+        // Going through `to_uint` to check size and truncation.
+        Ok(F::from_bits(self.to_bits(Size::from_bits(F::BITS))?))
+    }
+
+    #[inline]
+    pub fn try_to_f32(self) -> Result<Single, Size> {
+        self.try_to_float()
+    }
+
+    #[inline]
+    pub fn try_to_f64(self) -> Result<Double, Size> {
+        self.try_to_float()
+    }
 }
 
 macro_rules! from {
@@ -399,11 +420,7 @@ impl TryFrom<ScalarInt> for bool {
     type Error = Size;
     #[inline]
     fn try_from(int: ScalarInt) -> Result<Self, Size> {
-        int.to_bits(Size::from_bytes(1)).and_then(|u| match u {
-            0 => Ok(false),
-            1 => Ok(true),
-            _ => Err(Size::from_bytes(1)),
-        })
+        int.try_to_bool()
     }
 }
 

compiler/rustc_mir_build/messages.ftl

@@ -212,6 +212,8 @@ mir_build_mutation_of_layout_constrained_field_requires_unsafe_unsafe_op_in_unsa
     .note = mutating layout constrained fields cannot statically be checked for valid values
     .label = mutation of layout constrained field
 
+mir_build_nan_pattern = cannot use NaN in patterns
+
 mir_build_non_const_path = runtime values cannot be referenced in patterns
 
 mir_build_non_exhaustive_match_all_arms_guarded =

compiler/rustc_mir_build/src/errors.rs

@@ -724,6 +724,13 @@ pub struct UnsizedPattern<'tcx> {
     pub non_sm_ty: Ty<'tcx>,
 }
 
+#[derive(Diagnostic)]
+#[diag(mir_build_nan_pattern)]
+pub struct NaNPattern {
+    #[primary_span]
+    pub span: Span,
+}
+
 #[derive(LintDiagnostic)]
 #[diag(mir_build_float_pattern)]
 pub struct FloatPattern;

compiler/rustc_mir_build/src/thir/pattern/const_to_pat.rs

@@ -1,3 +1,4 @@
+use rustc_apfloat::Float;
 use rustc_hir as hir;
 use rustc_hir::def_id::DefId;
 use rustc_index::Idx;
@@ -16,7 +17,7 @@ use std::cell::Cell;
 
 use super::PatCtxt;
 use crate::errors::{
-    FloatPattern, IndirectStructuralMatch, InvalidPattern, NonPartialEqMatch,
+    FloatPattern, IndirectStructuralMatch, InvalidPattern, NaNPattern, NonPartialEqMatch,
     NontrivialStructuralMatch, PointerPattern, TypeNotStructural, UnionPattern, UnsizedPattern,
 };
 
@@ -310,16 +311,6 @@ impl<'tcx> ConstToPat<'tcx> {
         let param_env = self.param_env;
 
         let kind = match ty.kind() {
-            ty::Float(_) => {
-                self.saw_const_match_lint.set(true);
-                tcx.emit_spanned_lint(
-                    lint::builtin::ILLEGAL_FLOATING_POINT_LITERAL_PATTERN,
-                    id,
-                    span,
-                    FloatPattern,
-                );
-                return Err(FallbackToOpaqueConst);
-            }
             // If the type is not structurally comparable, just emit the constant directly,
             // causing the pattern match code to treat it opaquely.
             // FIXME: This code doesn't emit errors itself, the caller emits the errors.
@@ -469,6 +460,26 @@ impl<'tcx> ConstToPat<'tcx> {
                     }
                 }
             },
+            ty::Float(flt) => {
+                let v = cv.unwrap_leaf();
+                let is_nan = match flt {
+                    ty::FloatTy::F32 => v.try_to_f32().unwrap().is_nan(),
+                    ty::FloatTy::F64 => v.try_to_f64().unwrap().is_nan(),
+                };
+                if is_nan {
+                    self.saw_const_match_error.set(true);
+                    tcx.sess.emit_err(NaNPattern { span });
+                } else {
+                    self.saw_const_match_lint.set(true);
+                    tcx.emit_spanned_lint(
+                        lint::builtin::ILLEGAL_FLOATING_POINT_LITERAL_PATTERN,
+                        id,
+                        span,
+                        FloatPattern,
+                    );
+                }
+                return Err(FallbackToOpaqueConst);
+            }
             ty::Bool | ty::Char | ty::Int(_) | ty::Uint(_) => {
                 PatKind::Constant { value: mir::Const::Ty(ty::Const::new_value(tcx, cv, ty)) }
             }

tests/ui/rfcs/rfc-1445-restrict-constants-in-patterns/issue-6804-nan-match.rs

@@ -0,0 +1,32 @@
+// Matching against NaN should result in an error
+#![feature(exclusive_range_pattern)]
+#![allow(unused)]
+#![allow(illegal_floating_point_literal_pattern)]
+
+const NAN: f64 = f64::NAN;
+
+fn main() {
+    let x = NAN;
+    match x {
+        NAN => {}, //~ ERROR cannot use NaN in patterns
+        _ => {},
+    };
+
+    match [x, 1.0] {
+        [NAN, _] => {}, //~ ERROR cannot use NaN in patterns
+        _ => {},
+    };
+
+    // Also cover range patterns
+    match x {
+        NAN..=1.0 => {}, //~ ERROR cannot use NaN in patterns
+        //~^ ERROR lower range bound must be less than or equal to upper
+        -1.0..=NAN => {}, //~ ERROR cannot use NaN in patterns
+        //~^ ERROR lower range bound must be less than or equal to upper
+        NAN.. => {}, //~ ERROR cannot use NaN in patterns
+        //~^ ERROR lower range bound must be less than or equal to upper
+        ..NAN => {}, //~ ERROR cannot use NaN in patterns
+        //~^ ERROR lower range bound must be less than upper
+        _ => {},
+    };
+}

tests/ui/rfcs/rfc-1445-restrict-constants-in-patterns/issue-6804-nan-match.stderr

@@ -0,0 +1,64 @@
+error: cannot use NaN in patterns
+  --> $DIR/issue-6804-nan-match.rs:11:9
+   |
+LL |         NAN => {},
+   |         ^^^
+
+error: cannot use NaN in patterns
+  --> $DIR/issue-6804-nan-match.rs:16:10
+   |
+LL |         [NAN, _] => {},
+   |          ^^^
+
+error: cannot use NaN in patterns
+  --> $DIR/issue-6804-nan-match.rs:22:9
+   |
+LL |         NAN..=1.0 => {},
+   |         ^^^
+
+error[E0030]: lower range bound must be less than or equal to upper
+  --> $DIR/issue-6804-nan-match.rs:22:9
+   |
+LL |         NAN..=1.0 => {},
+   |         ^^^ lower bound larger than upper bound
+
+error: cannot use NaN in patterns
+  --> $DIR/issue-6804-nan-match.rs:24:16
+   |
+LL |         -1.0..=NAN => {},
+   |                ^^^
+
+error[E0030]: lower range bound must be less than or equal to upper
+  --> $DIR/issue-6804-nan-match.rs:24:9
+   |
+LL |         -1.0..=NAN => {},
+   |         ^^^^ lower bound larger than upper bound
+
+error: cannot use NaN in patterns
+  --> $DIR/issue-6804-nan-match.rs:26:9
+   |
+LL |         NAN.. => {},
+   |         ^^^
+
+error[E0030]: lower range bound must be less than or equal to upper
+  --> $DIR/issue-6804-nan-match.rs:26:9
+   |
+LL |         NAN.. => {},
+   |         ^^^ lower bound larger than upper bound
+
+error: cannot use NaN in patterns
+  --> $DIR/issue-6804-nan-match.rs:28:11
+   |
+LL |         ..NAN => {},
+   |           ^^^
+
+error[E0579]: lower range bound must be less than upper
+  --> $DIR/issue-6804-nan-match.rs:28:9
+   |
+LL |         ..NAN => {},
+   |         ^^^^^
+
+error: aborting due to 10 previous errors
+
+Some errors have detailed explanations: E0030, E0579.
+For more information about an error, try `rustc --explain E0030`.