Auto merge of #121662 - saethlin:precondition-unification, r=RalfJung

Distinguish between library and lang UB in assert_unsafe_precondition

As described in #121583 (comment), `assert_unsafe_precondition` now explicitly distinguishes between language UB (conditions we explicitly optimize on) and library UB (things we document you shouldn't do, and maybe some library internals assume you don't do).

`debug_assert_nounwind` was originally added to avoid the "only at runtime" aspect of `assert_unsafe_precondition`. Since then the difference between the macros has gotten muddied. This totally revamps the situation.

Now _all_ preconditions shall be checked with `assert_unsafe_precondition`. If you have a precondition that's only checkable at runtime, do a `const_eval_select` hack, as done in this PR.

r? RalfJung

Author: bors

compiler/rustc_borrowck/src/type_check/mod.rs

@@ -2000,7 +2000,7 @@ impl<'a, 'tcx> TypeChecker<'a, 'tcx> {
                     ConstraintCategory::SizedBound,
                 );
             }
-            &Rvalue::NullaryOp(NullOp::DebugAssertions, _) => {}
+            &Rvalue::NullaryOp(NullOp::UbCheck(_), _) => {}
 
             Rvalue::ShallowInitBox(operand, ty) => {
                 self.check_operand(operand, location);

compiler/rustc_codegen_cranelift/src/base.rs

@@ -779,7 +779,7 @@ fn codegen_stmt<'tcx>(
                         NullOp::OffsetOf(fields) => {
                             layout.offset_of_subfield(fx, fields.iter()).bytes()
                         }
-                        NullOp::DebugAssertions => {
+                        NullOp::UbCheck(_) => {
                             let val = fx.tcx.sess.opts.debug_assertions;
                             let val = CValue::by_val(
                                 fx.bcx.ins().iconst(types::I8, i64::try_from(val).unwrap()),

compiler/rustc_codegen_ssa/src/mir/rvalue.rs

@@ -685,7 +685,8 @@ impl<'a, 'tcx, Bx: BuilderMethods<'a, 'tcx>> FunctionCx<'a, 'tcx, Bx> {
                         let val = layout.offset_of_subfield(bx.cx(), fields.iter()).bytes();
                         bx.cx().const_usize(val)
                     }
-                    mir::NullOp::DebugAssertions => {
+                    mir::NullOp::UbCheck(_) => {
+                        // In codegen, we want to check for language UB and library UB
                         let val = bx.tcx().sess.opts.debug_assertions;
                         bx.cx().const_bool(val)
                     }

compiler/rustc_const_eval/src/interpret/step.rs

@@ -258,10 +258,16 @@ impl<'mir, 'tcx: 'mir, M: Machine<'mir, 'tcx>> InterpCx<'mir, 'tcx, M> {
                         let val = layout.offset_of_subfield(self, fields.iter()).bytes();
                         Scalar::from_target_usize(val, self)
                     }
-                    mir::NullOp::DebugAssertions => {
-                        // The checks hidden behind this are always better done by the interpreter
-                        // itself, because it knows the runtime state better.
-                        Scalar::from_bool(false)
+                    mir::NullOp::UbCheck(kind) => {
+                        // We want to enable checks for library UB, because the interpreter doesn't
+                        // know about those on its own.
+                        // But we want to disable checks for language UB, because the interpreter
+                        // has its own better checks for that.
+                        let should_check = match kind {
+                            mir::UbKind::LibraryUb => self.tcx.sess.opts.debug_assertions,
+                            mir::UbKind::LanguageUb => false,
+                        };
+                        Scalar::from_bool(should_check)
                     }
                 };
                 self.write_scalar(val, &dest)?;

compiler/rustc_const_eval/src/transform/check_consts/check.rs

@@ -558,7 +558,7 @@ impl<'tcx> Visitor<'tcx> for Checker<'_, 'tcx> {
             Rvalue::Cast(_, _, _) => {}
 
             Rvalue::NullaryOp(
-                NullOp::SizeOf | NullOp::AlignOf | NullOp::OffsetOf(_) | NullOp::DebugAssertions,
+                NullOp::SizeOf | NullOp::AlignOf | NullOp::OffsetOf(_) | NullOp::UbCheck(_),
                 _,
             ) => {}
             Rvalue::ShallowInitBox(_, _) => {}

compiler/rustc_const_eval/src/transform/validate.rs

@@ -1157,7 +1157,7 @@ impl<'a, 'tcx> Visitor<'tcx> for TypeChecker<'a, 'tcx> {
             Rvalue::Repeat(_, _)
             | Rvalue::ThreadLocalRef(_)
             | Rvalue::AddressOf(_, _)
-            | Rvalue::NullaryOp(NullOp::SizeOf | NullOp::AlignOf | NullOp::DebugAssertions, _)
+            | Rvalue::NullaryOp(NullOp::SizeOf | NullOp::AlignOf | NullOp::UbCheck(_), _)
             | Rvalue::Discriminant(_) => {}
         }
         self.super_rvalue(rvalue, location);

compiler/rustc_hir_analysis/src/check/intrinsic.rs

@@ -127,7 +127,8 @@ pub fn intrinsic_operation_unsafety(tcx: TyCtxt<'_>, intrinsic_id: LocalDefId) -
         | sym::variant_count
         | sym::is_val_statically_known
         | sym::ptr_mask
-        | sym::debug_assertions
+        | sym::check_language_ub
+        | sym::check_library_ub
         | sym::fadd_algebraic
         | sym::fsub_algebraic
         | sym::fmul_algebraic
@@ -584,7 +585,7 @@ pub fn check_intrinsic_type(
                 (0, 0, vec![Ty::new_imm_ptr(tcx, Ty::new_unit(tcx))], tcx.types.usize)
             }
 
-            sym::debug_assertions => (0, 1, Vec::new(), tcx.types.bool),
+            sym::check_language_ub | sym::check_library_ub => (0, 1, Vec::new(), tcx.types.bool),
 
             sym::simd_eq
             | sym::simd_ne

compiler/rustc_middle/src/mir/pretty.rs

@@ -915,7 +915,7 @@ impl<'tcx> Debug for Rvalue<'tcx> {
                     NullOp::SizeOf => write!(fmt, "SizeOf({t})"),
                     NullOp::AlignOf => write!(fmt, "AlignOf({t})"),
                     NullOp::OffsetOf(fields) => write!(fmt, "OffsetOf({t}, {fields:?})"),
-                    NullOp::DebugAssertions => write!(fmt, "cfg!(debug_assertions)"),
+                    NullOp::UbCheck(kind) => write!(fmt, "UbCheck({kind:?})"),
                 }
             }
             ThreadLocalRef(did) => ty::tls::with(|tcx| {

compiler/rustc_middle/src/mir/syntax.rs

@@ -1366,8 +1366,16 @@ pub enum NullOp<'tcx> {
     AlignOf,
     /// Returns the offset of a field
     OffsetOf(&'tcx List<(VariantIdx, FieldIdx)>),
-    /// cfg!(debug_assertions), but expanded in codegen
-    DebugAssertions,
+    /// Returns whether we want to check for library UB or language UB at monomorphization time.
+    /// Both kinds of UB evaluate to `true` in codegen, and only library UB evalutes to `true` in
+    /// const-eval/Miri, because the interpreter has its own better checks for language UB.
+    UbCheck(UbKind),
+}
+
+#[derive(Clone, Copy, Debug, PartialEq, Eq, TyEncodable, TyDecodable, Hash, HashStable)]
+pub enum UbKind {
+    LanguageUb,
+    LibraryUb,
 }
 
 #[derive(Copy, Clone, Debug, PartialEq, Eq, PartialOrd, Ord, Hash)]

compiler/rustc_middle/src/mir/tcx.rs

@@ -194,7 +194,7 @@ impl<'tcx> Rvalue<'tcx> {
             Rvalue::NullaryOp(NullOp::SizeOf | NullOp::AlignOf | NullOp::OffsetOf(..), _) => {
                 tcx.types.usize
             }
-            Rvalue::NullaryOp(NullOp::DebugAssertions, _) => tcx.types.bool,
+            Rvalue::NullaryOp(NullOp::UbCheck(_), _) => tcx.types.bool,
             Rvalue::Aggregate(ref ak, ref ops) => match **ak {
                 AggregateKind::Array(ty) => Ty::new_array(tcx, ty, ops.len() as u64),
                 AggregateKind::Tuple => {