Unrolled build for #158501

Rollup merge of #158501 - RalfJung:miri, r=RalfJung

miri subtree update

Subtree update of `miri` to rust-lang/miri@97eec7c.

Created using https://github.com/rust-lang/josh-sync.

r? @ghost

Author: rust-timer

src/tools/miri/README.md

@@ -171,6 +171,38 @@ available (which affects `cfg(target_feature)`), and it tells Miri to consider t
 that the interpreted program runs on as having the feature available (meaning the code is allowed to
 invoke the corresponding intrinsics).
 
+### Nextest integration
+
+Miri can be combined with [`cargo-nextest`](https://nexte.st):
+
+```
+cargo install --locked cargo-nextest
+cargo miri nextest run
+```
+
+Nextest spawns a separate instance of Miri for each test, which has several advantages:
+- Tests can run in parallel. Miri itself only uses a single thread per interpreter so this can
+  give a massive speedup (but see the caveat below).
+- Tests do not stop when a single problem is found. Miri aborts execution when it encounters
+  Undefined Behavior or an unsupported operation (there is often not really any way to continue),
+  so once a single test fails, the remaining tests cannot be executed. Nextest's process-per-test
+  model means that you end up with a full list of which tests worked in Miri and which tests had a
+  problem.
+
+However, there is also a big caveat: Miri will [re-compile the test crate every time it is
+invoked](https://github.com/rust-lang/miri/issues/5013), which means a crate with N tests will be
+compiled N+1 times. If the test crate takes a long time to build, this can outweigh the benefits of
+parallelization.
+
+For more information about nextest, see the [`cargo-nextest` Miri
+documentation](https://nexte.st/book/miri.html).
+
+Note: Nextest's one-test-per-process model means that `cargo miri test` is able to detect data
+races where two tests race on a shared resource, but `cargo miri nextest run` will not detect
+such races.
+
+Note: `cargo-nextest` [does not support doctests](https://github.com/nextest-rs/nextest/issues/16).
+
 ### Testing multiple different executions
 
 Certain parts of the execution are picked randomly by Miri, such as the exact base address
@@ -184,6 +216,7 @@ MIRIFLAGS="-Zmiri-many-seeds" cargo miri test # tries the seeds in 0..64
 MIRIFLAGS="-Zmiri-many-seeds=0..16" cargo miri test
 ```
 
+Miri will test the given range of seeds with parallel interpreter instances.
 The default of 64 different seeds can be quite slow, so you often want to specify a smaller range.
 
 ### Running Miri on CI
@@ -243,26 +276,6 @@ However, even for targets that we do support, the degree of support for accessin
 (such as the file system) differs between targets: generally, Linux targets have the best support,
 and macOS targets are usually on par. Windows is supported less well.
 
-### Running tests in parallel
-
-Though it implements Rust threading, Miri itself is a single-threaded interpreter
-(it works like a multi-threaded OS on a single-core CPU).
-This means that when running `cargo miri test`, you will probably see a dramatic
-increase in the amount of time it takes to run your whole test suite due to the
-inherent interpreter slowdown and a loss of parallelism.
-
-You can get your test suite's parallelism back by running `cargo miri nextest run -jN`
-(note that you will need [`cargo-nextest`](https://nexte.st) installed).
-This works because `cargo-nextest` collects a list of all tests then launches a
-separate `cargo miri run` for each test. For more information about nextest, see the
-[`cargo-nextest` Miri documentation](https://nexte.st/book/miri.html).
-
-Note: This one-test-per-process model means that `cargo miri test` is able to detect data
-races where two tests race on a shared resource, but `cargo miri nextest run` will not detect
-such races.
-
-Note: `cargo-nextest` does not support doctests, see https://github.com/nextest-rs/nextest/issues/16
-
 ### Directly invoking the `miri` driver
 
 The recommended way to invoke Miri is via `cargo miri`. Directly invoking the underlying `miri`

src/tools/miri/cargo-miri/src/phases.rs

@@ -36,6 +36,9 @@ Examples:
         This will print the path to the generated sysroot (and nothing else) on stdout.
         stderr will still contain progress information about how the build is doing.
 
+For documentation on `-Zmiri-...` flags, see Miri's README.md, available at:
+- $(rustc --print sysroot)/share/doc/miri/README.md
+- https://github.com/rust-lang/miri/blob/master/README.md
 ";
 
 fn show_help() {

src/tools/miri/ci/ci.sh

@@ -168,6 +168,7 @@ case $HOST_TARGET in
     MANY_SEEDS=16 TEST_TARGET=x86_64-unknown-freebsd run_tests
     MANY_SEEDS=16 TEST_TARGET=i686-unknown-freebsd run_tests
     MANY_SEEDS=16 TEST_TARGET=x86_64-unknown-illumos run_tests
+    MANY_SEEDS=16 TEST_TARGET=x86_64-unknown-netbsd run_tests_minimal hello
     ;;
   armv7-unknown-linux-gnueabihf)
     # Host

src/tools/miri/priroda/src/main.rs

@@ -6,6 +6,7 @@ extern crate rustc_data_structures;
 extern crate rustc_driver;
 extern crate rustc_hir;
 extern crate rustc_hir_analysis;
+extern crate rustc_index;
 extern crate rustc_interface;
 extern crate rustc_log;
 extern crate rustc_middle;
@@ -19,13 +20,15 @@ use std::path::PathBuf;
 use miri::*;
 use rustc_driver::Compilation;
 use rustc_hir::attrs::CrateType;
+use rustc_index::IndexVec;
 use rustc_interface::interface;
 use rustc_middle::mir;
+use rustc_middle::mir::{Local, VarDebugInfoContents};
 use rustc_middle::ty::TyCtxt;
 use rustc_session::EarlyDiagCtxt;
 use rustc_session::config::ErrorOutputType;
-use rustc_span::Span;
 use rustc_span::source_map::SourceMap;
+use rustc_span::{Span, Symbol};
 
 fn find_sysroot() -> String {
     std::env::var("MIRI_SYSROOT")
@@ -129,6 +132,11 @@ struct PrirodaContext<'tcx> {
     last_location: Option<SourceLocation>,
 }
 
+struct LocalDesc {
+    name: Option<Symbol>,
+    local: Local,
+    ty: String,
+}
 /// Controls when execution returns to the frontend.
 enum ResumeMode {
     /// Stop at the next visible MIR instruction.
@@ -336,15 +344,49 @@ impl<'tcx> PrirodaContext<'tcx> {
         }
     }
 
-    /// Returns the names of all user-visible locals in the innermost stack frame.
+    /// Returns structured descriptions for locals in the innermost stack frame.
     ///
-    /// Uses `var_debug_info` from the MIR body, which is the same source that
-    /// DWARF debug info is built from, so the names match what the user wrote.
-    fn list_locals(&self) -> Vec<String> {
+    /// Starts from all MIR locals, then enriches them with source names from
+    /// `var_debug_info` when a debug entry maps directly to a whole local.
+    fn list_locals(&self) -> Vec<LocalDesc> {
         let Some(frame) = self.ecx.active_thread_stack().last() else {
             return Vec::new();
         };
-        frame.body().var_debug_info.iter().map(|info| info.name.to_string()).collect()
+
+        self.local_desc_map(frame).into_iter().collect()
+    }
+
+    fn local_desc_map(
+        &self,
+        frame: &Frame<'tcx, Provenance, FrameExtra<'tcx>>,
+    ) -> IndexVec<Local, LocalDesc> {
+        // Initialize one description per MIR local so the table can be indexed by Local.
+        let mut locals: IndexVec<Local, LocalDesc> = frame
+            .body()
+            .local_decls
+            .iter_enumerated()
+            .map(|(id, local_decl)| {
+                LocalDesc { name: None, local: id, ty: local_decl.ty.to_string() }
+            })
+            .collect();
+
+        // FIXME: Some debug-info entries do not have a backing MIR local, for example
+        // because the source variable was optimized out or is represented as a
+        // projection. This local-indexed table cannot represent those entries yet;
+        // the final locals list should become a `Vec<LocalDesc>` with `id : Option<Local>`, `id`
+        // could be renamed to `local`.
+
+        // Attach source names from debug info when the debug entry maps directly to a whole MIR local.
+        for var_debug_info in &frame.body().var_debug_info {
+            if let VarDebugInfoContents::Place(place) = var_debug_info.value
+                && let Some(local) = place.as_local()
+                && locals[local].name.is_none()
+            {
+                locals[local].name = Some(var_debug_info.name);
+            }
+        }
+
+        locals
     }
 }
 
@@ -366,7 +408,7 @@ enum BreakpointSetResult {
 enum CommandResult {
     ExecutionStopped(StepResult),
     BreakpointResult(BreakpointSetResult),
-    Locals(Vec<String>),
+    Locals(Vec<LocalDesc>),
     // FIXME: distinguish terminating the debugger session from disconnecting a
     // frontend and terminating the interpreted program once multiple frontends exist.
     TerminateSession,
@@ -403,12 +445,21 @@ impl Cli {
 
                             BreakpointSetResult::Duplicate => println!("Duplicate breakpoint"),
                         },
-                    CommandResult::Locals(names) =>
-                        if names.is_empty() {
+                    CommandResult::Locals(locals_desc) =>
+                        if locals_desc.is_empty() {
                             println!("no locals");
                         } else {
-                            for name in &names {
-                                println!("{name}");
+                            for local_desc in &locals_desc {
+                                let mut name_str = "None".to_string();
+                                if let Some(name) = local_desc.name {
+                                    name_str = name.to_string();
+                                }
+                                println!(
+                                    "Name: {}, Id: _{}, Ty: {}",
+                                    name_str,
+                                    local_desc.local.index(),
+                                    local_desc.ty,
+                                );
                             }
                         },
                     CommandResult::TerminateSession => {

src/tools/miri/priroda/tests/ui/locals_in_function.stdout

@@ -1,6 +1,10 @@
 (priroda) breakpoint added: {MANIFEST_DIR}/tests/ui/locals_in_function.rs:5
 (priroda) Hit breakpoint
 {MANIFEST_DIR}/tests/ui/locals_in_function.rs:5
-(priroda) x
-y
+(priroda) Name: None, Id: _0, Ty: ()
+Name: x, Id: _1, Ty: i32
+Name: y, Id: _2, Ty: bool
+Name: None, Id: _3, Ty: (i32, bool)
+Name: None, Id: _4, Ty: i32
+Name: None, Id: _5, Ty: bool
 (priroda) quitting

src/tools/miri/rust-version

@@ -1 +1 @@
-01f54e80e888b66d6486a3a95d481b87353016df
+16761606d606b6ec4d0c88fc9251670742ad9fd2

src/tools/miri/src/borrow_tracker/tree_borrows/tree.rs

@@ -574,7 +574,9 @@ impl<'tcx> Tree {
                                 // Don't check for protector if it is a Cell (see `unsafe_cell_deallocate` in `interior_mutability.rs`).
                                 // Related to https://github.com/rust-lang/rust/issues/55005.
                                 && !perm.permission.is_cell()
-                                // Only trigger UB if the accessed bit is set, i.e. if the protector is actually protecting this offset. See #4579.
+                                // Only trigger UB if the accessed bit is set, i.e. if the protector
+                                // is actually protecting this offset. See #4579. Note that this
+                                // takes into account the access we just did above!
                                 && perm.accessed
                             {
                                 Err(TbError {

src/tools/miri/src/helpers.rs

@@ -129,7 +129,13 @@ pub fn iter_exported_symbols<'tcx>(
                 || codegen_attrs.flags.contains(CodegenFnAttrFlags::USED_COMPILER)
                 || codegen_attrs.flags.contains(CodegenFnAttrFlags::USED_LINKER)
         };
-        if exported {
+        // FIXME: `#[no_mangle]` makes no sense on a generic item, but still causes it to be
+        // considered "extern". Remove this once `no_mangle_generic_items` is a hard error.
+        let exported_mono = exported && {
+            let generics = tcx.generics_of(def_id);
+            !generics.requires_monomorphization(tcx)
+        };
+        if exported_mono {
             f(LOCAL_CRATE, def_id.into())?;
         }
     }

src/tools/miri/src/shims/aarch64.rs

@@ -146,6 +146,45 @@ pub(super) trait EvalContextExt<'tcx>: crate::MiriInterpCxExt<'tcx> {
                 }
             }
 
+            // Signed saturating doubling multiply returning the high half.
+            //
+            // Used by the `vqdmulh*` functions.
+            //
+            // This LLVM intrinsic multiplies the values of corresponding elements of the two source
+            // vector registers (which are signed integers), doubles the results, places the most significant half of the
+            // final results (using a saturating cast to fit the element type) into a vector, and writes the vector to the destination register.
+            //
+            // https://developer.arm.com/architectures/instruction-sets/intrinsics#f:@navigationhierarchiessimdisa=[Neon]&q=vqdmulh
+            name if name.starts_with("neon.sqdmulh.") => {
+                let [left, right] =
+                    this.check_shim_sig_lenient(abi, CanonAbi::C, link_name, args)?;
+
+                let (left, left_len) = this.project_to_simd(left)?;
+                let (right, right_len) = this.project_to_simd(right)?;
+                let (dest, dest_len) = this.project_to_simd(dest)?;
+                assert_eq!(left_len, right_len);
+                assert_eq!(left_len, dest_len);
+
+                let elem_size = dest.layout.field(this, 0).size;
+                let bits = elem_size.bits();
+                let min = elem_size.signed_int_min();
+                let max = elem_size.signed_int_max();
+
+                for i in 0..dest_len {
+                    let a = this.read_scalar(&this.project_index(&left, i)?)?.to_int(elem_size)?;
+                    let b = this.read_scalar(&this.project_index(&right, i)?)?.to_int(elem_size)?;
+
+                    // Uses i128 arithmetic, which cannot overflow because the intrinsic takes at most i32.
+                    let doubled = a.strict_mul(b).strict_mul(2);
+                    let res = (doubled >> bits).clamp(min, max);
+
+                    this.write_scalar(
+                        Scalar::from_int(res, elem_size),
+                        &this.project_index(&dest, i)?,
+                    )?;
+                }
+            }
+
             // Vector table lookup: each index selects a byte from the 16-byte table, out-of-range -> 0.
             // Used to implement vtbl1_u8 function.
             // LLVM does not have a portable shuffle that takes non-const indices

src/tools/miri/src/shims/unix/foreign_items.rs

@@ -840,18 +840,46 @@ pub trait EvalContextExt<'tcx>: crate::MiriInterpCxExt<'tcx> {
             }
 
             "mmap" => {
-                let [addr, length, prot, flags, fd, offset] =
-                    this.check_shim_sig_lenient(abi, CanonAbi::C, link_name, args)?;
+                let [addr, length, prot, flags, fd, offset] = this.check_shim_sig(
+                    shim_sig!(extern "C" fn(*mut _, usize, i32, i32, i32, libc::off_t) -> *mut _),
+                    link_name,
+                    abi,
+                    args,
+                )?;
                 let offset = this.read_scalar(offset)?.to_int(this.libc_ty_layout("off_t").size)?;
                 let ptr = this.mmap(addr, length, prot, flags, fd, offset)?;
                 this.write_scalar(ptr, dest)?;
             }
             "munmap" => {
-                let [addr, length] =
-                    this.check_shim_sig_lenient(abi, CanonAbi::C, link_name, args)?;
+                let [addr, length] = this.check_shim_sig(
+                    shim_sig!(extern "C" fn(*mut _, usize) -> i32),
+                    link_name,
+                    abi,
+                    args,
+                )?;
                 let result = this.munmap(addr, length)?;
                 this.write_scalar(result, dest)?;
             }
+            "mprotect" => {
+                let [addr, length, prot] = this.check_shim_sig(
+                    shim_sig!(extern "C" fn(*mut _, usize, i32) -> i32),
+                    link_name,
+                    abi,
+                    args,
+                )?;
+                let result = this.mprotect(addr, length, prot)?;
+                this.write_scalar(result, dest)?;
+            }
+            "madvise" => {
+                let [addr, length, advice] = this.check_shim_sig(
+                    shim_sig!(extern "C" fn(*mut _, usize, i32) -> i32),
+                    link_name,
+                    abi,
+                    args,
+                )?;
+                let result = this.madvise(addr, length, advice)?;
+                this.write_scalar(result, dest)?;
+            }
 
             "reallocarray" => {
                 // Currently this function does not exist on all Unixes, e.g. on macOS.
@@ -1410,7 +1438,7 @@ pub trait EvalContextExt<'tcx>: crate::MiriInterpCxExt<'tcx> {
                 let [_, _] = this.check_shim_sig_lenient(abi, CanonAbi::C, link_name, args)?;
                 this.write_null(dest)?;
             }
-            "sigaction" | "mprotect" if this.frame_in_std() => {
+            "sigaction" if this.frame_in_std() => {
                 let [_, _, _] = this.check_shim_sig_lenient(abi, CanonAbi::C, link_name, args)?;
                 this.write_null(dest)?;
             }