Auto merge of #145259 - nikic:read-only-capture, r=wesleywiser

Tell LLVM about read-only captures

`&Freeze` parameters are not only `readonly` within the function, but any captures of the pointer can also only be used for reads. This can now be encoded using the `captures(address, read_provenance)` attribute.

compiler/rustc_codegen_llvm/src/abi.rs

@@ -42,12 +42,13 @@ trait ArgAttributesExt {
 const ABI_AFFECTING_ATTRIBUTES: [(ArgAttribute, llvm::AttributeKind); 1] =
     [(ArgAttribute::InReg, llvm::AttributeKind::InReg)];
 
-const OPTIMIZATION_ATTRIBUTES: [(ArgAttribute, llvm::AttributeKind); 5] = [
+const OPTIMIZATION_ATTRIBUTES: [(ArgAttribute, llvm::AttributeKind); 6] = [
     (ArgAttribute::NoAlias, llvm::AttributeKind::NoAlias),
     (ArgAttribute::NoCapture, llvm::AttributeKind::NoCapture),
     (ArgAttribute::NonNull, llvm::AttributeKind::NonNull),
     (ArgAttribute::ReadOnly, llvm::AttributeKind::ReadOnly),
     (ArgAttribute::NoUndef, llvm::AttributeKind::NoUndef),
+    (ArgAttribute::CapturesReadOnly, llvm::AttributeKind::CapturesReadOnly),
 ];
 
 fn get_attrs<'ll>(this: &ArgAttributes, cx: &CodegenCx<'ll, '_>) -> SmallVec<[&'ll Attribute; 8]> {
@@ -83,6 +84,10 @@ fn get_attrs<'ll>(this: &ArgAttributes, cx: &CodegenCx<'ll, '_>) -> SmallVec<[&'
         }
         for (attr, llattr) in OPTIMIZATION_ATTRIBUTES {
             if regular.contains(attr) {
+                // captures(address, read_provenance) is only available since LLVM 21.
+                if attr == ArgAttribute::CapturesReadOnly && llvm_util::get_version() < (21, 0, 0) {
+                    continue;
+                }
                 attrs.push(llattr.create_attr(cx.llcx));
             }
         }

compiler/rustc_codegen_llvm/src/llvm/ffi.rs

@@ -251,6 +251,7 @@ pub(crate) enum AttributeKind {
     Writable = 42,
     DeadOnUnwind = 43,
     DeadOnReturn = 44,
+    CapturesReadOnly = 45,
 }
 
 /// LLVMIntPredicate

compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp

@@ -278,6 +278,7 @@ enum class LLVMRustAttributeKind {
   Writable = 42,
   DeadOnUnwind = 43,
   DeadOnReturn = 44,
+  CapturesReadOnly = 45,
 };
 
 static Attribute::AttrKind fromRust(LLVMRustAttributeKind Kind) {
@@ -376,6 +377,8 @@ static Attribute::AttrKind fromRust(LLVMRustAttributeKind Kind) {
 #else
     report_fatal_error("DeadOnReturn attribute requires LLVM 21 or later");
 #endif
+  case LLVMRustAttributeKind::CapturesReadOnly:
+    report_fatal_error("Should be handled separately");
   }
   report_fatal_error("bad LLVMRustAttributeKind");
 }
@@ -430,6 +433,11 @@ LLVMRustCreateAttrNoValue(LLVMContextRef C, LLVMRustAttributeKind RustAttr) {
   if (RustAttr == LLVMRustAttributeKind::NoCapture) {
     return wrap(Attribute::getWithCaptureInfo(*unwrap(C), CaptureInfo::none()));
   }
+  if (RustAttr == LLVMRustAttributeKind::CapturesReadOnly) {
+    return wrap(Attribute::getWithCaptureInfo(
+        *unwrap(C), CaptureInfo(CaptureComponents::Address |
+                                CaptureComponents::ReadProvenance)));
+  }
 #endif
   return wrap(Attribute::get(*unwrap(C), fromRust(RustAttr)));
 }

compiler/rustc_target/src/callconv/mod.rs

@@ -119,6 +119,7 @@ mod attr_impl {
             const ReadOnly  = 1 << 4;
             const InReg     = 1 << 5;
             const NoUndef = 1 << 6;
+            const CapturesReadOnly = 1 << 7;
         }
     }
     rustc_data_structures::external_bitflags_debug! { ArgAttribute }

compiler/rustc_ty_utils/src/abi.rs

@@ -356,6 +356,7 @@ fn arg_attrs_for_rust_scalar<'tcx>(
 
             if matches!(kind, PointerKind::SharedRef { frozen: true }) && !is_return {
                 attrs.set(ArgAttribute::ReadOnly);
+                attrs.set(ArgAttribute::CapturesReadOnly);
             }
         }
     }

tests/codegen-llvm/function-arguments.rs

@@ -80,7 +80,7 @@ pub fn option_nonzero_int(x: Option<NonZero<u64>>) -> Option<NonZero<u64>> {
     x
 }
 
-// CHECK: @readonly_borrow(ptr noalias noundef readonly align 4 dereferenceable(4) %_1)
+// CHECK: @readonly_borrow(ptr noalias noundef readonly align 4{{( captures\(address, read_provenance\))?}} dereferenceable(4) %_1)
 // FIXME #25759 This should also have `nocapture`
 #[no_mangle]
 pub fn readonly_borrow(_: &i32) {}
@@ -91,12 +91,12 @@ pub fn readonly_borrow_ret() -> &'static i32 {
     loop {}
 }
 
-// CHECK: @static_borrow(ptr noalias noundef readonly align 4 dereferenceable(4) %_1)
+// CHECK: @static_borrow(ptr noalias noundef readonly align 4{{( captures\(address, read_provenance\))?}} dereferenceable(4) %_1)
 // static borrow may be captured
 #[no_mangle]
 pub fn static_borrow(_: &'static i32) {}
 
-// CHECK: @named_borrow(ptr noalias noundef readonly align 4 dereferenceable(4) %_1)
+// CHECK: @named_borrow(ptr noalias noundef readonly align 4{{( captures\(address, read_provenance\))?}} dereferenceable(4) %_1)
 // borrow with named lifetime may be captured
 #[no_mangle]
 pub fn named_borrow<'r>(_: &'r i32) {}
@@ -129,7 +129,7 @@ pub fn mutable_borrow_ret() -> &'static mut i32 {
 // <https://github.com/rust-lang/unsafe-code-guidelines/issues/381>.
 pub fn mutable_notunpin_borrow(_: &mut NotUnpin) {}
 
-// CHECK: @notunpin_borrow(ptr noalias noundef readonly align 4 dereferenceable(4) %_1)
+// CHECK: @notunpin_borrow(ptr noalias noundef readonly align 4{{( captures\(address, read_provenance\))?}} dereferenceable(4) %_1)
 // But `&NotUnpin` behaves perfectly normal.
 #[no_mangle]
 pub fn notunpin_borrow(_: &NotUnpin) {}
@@ -138,12 +138,12 @@ pub fn notunpin_borrow(_: &NotUnpin) {}
 #[no_mangle]
 pub fn indirect_struct(_: S) {}
 
-// CHECK: @borrowed_struct(ptr noalias noundef readonly align 4 dereferenceable(32) %_1)
+// CHECK: @borrowed_struct(ptr noalias noundef readonly align 4{{( captures\(address, read_provenance\))?}} dereferenceable(32) %_1)
 // FIXME #25759 This should also have `nocapture`
 #[no_mangle]
 pub fn borrowed_struct(_: &S) {}
 
-// CHECK: @option_borrow(ptr noalias noundef readonly align 4 dereferenceable_or_null(4) %_x)
+// CHECK: @option_borrow(ptr noalias noundef readonly align 4{{( captures\(address, read_provenance\))?}} dereferenceable_or_null(4) %_x)
 #[no_mangle]
 pub fn option_borrow(_x: Option<&i32>) {}
 
@@ -185,7 +185,7 @@ pub fn _box(x: Box<i32>) -> Box<i32> {
 // With a custom allocator, it should *not* have `noalias`. (See
 // <https://github.com/rust-lang/miri/issues/3341> for why.) The second argument is the allocator,
 // which is a reference here that still carries `noalias` as usual.
-// CHECK: @_box_custom(ptr noundef nonnull align 4 %x.0, ptr noalias noundef nonnull readonly align 1 %x.1)
+// CHECK: @_box_custom(ptr noundef nonnull align 4 %x.0, ptr noalias noundef nonnull readonly align 1{{( captures\(address, read_provenance\))?}} %x.1)
 #[no_mangle]
 pub fn _box_custom(x: Box<i32, &std::alloc::Global>) {
     drop(x)
@@ -208,7 +208,7 @@ pub fn struct_return() -> S {
 #[no_mangle]
 pub fn helper(_: usize) {}
 
-// CHECK: @slice(ptr noalias noundef nonnull readonly align 1 %_1.0, [[USIZE]] noundef %_1.1)
+// CHECK: @slice(ptr noalias noundef nonnull readonly align 1{{( captures\(address, read_provenance\))?}} %_1.0, [[USIZE]] noundef %_1.1)
 // FIXME #25759 This should also have `nocapture`
 #[no_mangle]
 pub fn slice(_: &[u8]) {}
@@ -227,7 +227,7 @@ pub fn unsafe_slice(_: &[UnsafeInner]) {}
 #[no_mangle]
 pub fn raw_slice(_: *const [u8]) {}
 
-// CHECK: @str(ptr noalias noundef nonnull readonly align 1 %_1.0, [[USIZE]] noundef %_1.1)
+// CHECK: @str(ptr noalias noundef nonnull readonly align 1{{( captures\(address, read_provenance\))?}} %_1.0, [[USIZE]] noundef %_1.1)
 // FIXME #25759 This should also have `nocapture`
 #[no_mangle]
 pub fn str(_: &[u8]) {}
@@ -259,7 +259,7 @@ pub fn trait_option(x: Option<Box<dyn Drop + Unpin>>) -> Option<Box<dyn Drop + U
     x
 }
 
-// CHECK: { ptr, [[USIZE]] } @return_slice(ptr noalias noundef nonnull readonly align 2 %x.0, [[USIZE]] noundef %x.1)
+// CHECK: { ptr, [[USIZE]] } @return_slice(ptr noalias noundef nonnull readonly align 2{{( captures\(address, read_provenance\))?}} %x.0, [[USIZE]] noundef %x.1)
 #[no_mangle]
 pub fn return_slice(x: &[u16]) -> &[u16] {
     x

tests/codegen-llvm/range-attribute.rs

@@ -67,7 +67,7 @@ pub fn enum2_value(x: Enum2) -> Enum2 {
     x
 }
 
-// CHECK: noundef [[USIZE]] @takes_slice(ptr noalias noundef nonnull readonly align 4 %x.0, [[USIZE]] noundef %x.1)
+// CHECK: noundef [[USIZE]] @takes_slice(ptr {{.*}} %x.0, [[USIZE]] noundef %x.1)
 #[no_mangle]
 pub fn takes_slice(x: &[i32]) -> usize {
     x.len()

tests/codegen-llvm/read-only-capture-opt.rs

@@ -0,0 +1,18 @@
+//@ compile-flags: -C opt-level=3 -Z mir-opt-level=0
+//@ min-llvm-version: 21
+
+#![crate_type = "lib"]
+
+unsafe extern "C" {
+    safe fn do_something(p: &i32);
+}
+
+#[unsafe(no_mangle)]
+pub fn test() -> i32 {
+    // CHECK-LABEL: @test(
+    // CHECK: ret i32 0
+    let i = 0;
+    do_something(&i);
+    do_something(&i);
+    i
+}

tests/ui/abi/debug.generic.stderr

@@ -939,7 +939,7 @@ error: fn_abi_of(assoc_test) = FnAbi {
                    },
                    mode: Direct(
                        ArgAttributes {
-                           regular: NoAlias | NonNull | ReadOnly | NoUndef,
+                           regular: NoAlias | NonNull | ReadOnly | NoUndef | CapturesReadOnly,
                            arg_ext: None,
                            pointee_size: Size(2 bytes),
                            pointee_align: Some(

tests/ui/abi/debug.riscv64.stderr

@@ -939,7 +939,7 @@ error: fn_abi_of(assoc_test) = FnAbi {
                    },
                    mode: Direct(
                        ArgAttributes {
-                           regular: NoAlias | NonNull | ReadOnly | NoUndef,
+                           regular: NoAlias | NonNull | ReadOnly | NoUndef | CapturesReadOnly,
                            arg_ext: None,
                            pointee_size: Size(2 bytes),
                            pointee_align: Some(

tests/ui/codegen/equal-pointers-unequal/as-cast/segfault.rs

@@ -55,6 +55,7 @@ fn main() {
     // The `Box` has been deallocated by now, so this is a dangling reference!
     let r: &u8 = &*r;
     println!("{:p}", r);
+    println!("{}", i);
 
     // The following might segfault. Or it might not.
     // Depends on the platform semantics

tests/ui/codegen/equal-pointers-unequal/exposed-provenance/segfault.rs

@@ -58,6 +58,7 @@ fn main() {
     // The `Box` has been deallocated by now, so this is a dangling reference!
     let r: &u8 = &*r;
     println!("{:p}", r);
+    println!("{}", i);
 
     // The following might segfault. Or it might not.
     // Depends on the platform semantics

tests/ui/codegen/equal-pointers-unequal/strict-provenance/segfault.rs

@@ -58,6 +58,7 @@ fn main() {
     // The `Box` has been deallocated by now, so this is a dangling reference!
     let r: &u8 = &*r;
     println!("{:p}", r);
+    println!("{}", i);
 
     // The following might segfault. Or it might not.
     // Depends on the platform semantics