Auto merge of #1776 - hyd-dev:1170, r=RalfJung

Implement calls to exported symbols (#1170)

Closes #1170.

Author: bors

README.md

@@ -200,6 +200,8 @@ environment variable:
   `compare_exchange_weak` operations. The default is `0.8` (so 4 out of 5 weak ops will fail).
   You can change it to any value between `0.0` and `1.0`, where `1.0` means it
   will always fail and `0.0` means it will never fail.
+* `-Zmiri-disable-abi-check` disables checking [function ABI]. Using this flag
+  is **unsound**.
 * `-Zmiri-disable-alignment-check` disables checking pointer alignment, so you
   can focus on other failures, but it means Miri can miss bugs in your program.
   Using this flag is **unsound**.
@@ -263,6 +265,8 @@ environment variable:
   with `-Zmiri-track-raw-pointers` also works without
   `-Zmiri-track-raw-pointers`, but for the vast majority of code, this will be the case.
 
+[function ABI]: https://doc.rust-lang.org/reference/items/functions.html#extern-function-qualifier
+
 Some native rustc `-Z` flags are also very relevant for Miri:
 
 * `-Zmir-opt-level` controls how many MIR optimizations are performed.  Miri

src/bin/miri.rs

@@ -1,29 +1,52 @@
-#![feature(rustc_private)]
+#![feature(rustc_private, bool_to_option, stmt_expr_attributes)]
 
 extern crate rustc_driver;
 extern crate rustc_errors;
 extern crate rustc_hir;
 extern crate rustc_interface;
+extern crate rustc_metadata;
 extern crate rustc_middle;
 extern crate rustc_session;
 
 use std::convert::TryFrom;
 use std::env;
+use std::path::PathBuf;
+use std::rc::Rc;
 use std::str::FromStr;
 
 use hex::FromHexError;
 use log::debug;
 
 use rustc_driver::Compilation;
 use rustc_errors::emitter::{ColorConfig, HumanReadableErrorType};
-use rustc_middle::ty::TyCtxt;
-use rustc_session::{config::ErrorOutputType, CtfeBacktrace};
+use rustc_hir::def_id::LOCAL_CRATE;
+use rustc_interface::interface::Config;
+use rustc_middle::{
+    middle::exported_symbols::{ExportedSymbol, SymbolExportLevel},
+    ty::{query::Providers, TyCtxt},
+};
+use rustc_session::{config::ErrorOutputType, search_paths::PathKind, CtfeBacktrace};
 
 struct MiriCompilerCalls {
     miri_config: miri::MiriConfig,
 }
 
 impl rustc_driver::Callbacks for MiriCompilerCalls {
+    fn config(&mut self, config: &mut Config) {
+        config.override_queries = Some(|_, _, external_providers| {
+            external_providers.used_crate_source = |tcx, cnum| {
+                let mut providers = Providers::default();
+                rustc_metadata::provide_extern(&mut providers);
+                let mut crate_source = (providers.used_crate_source)(tcx, cnum);
+                // HACK: rustc will emit "crate ... required to be available in rlib format, but
+                // was not found in this form" errors once we use `tcx.dependency_formats()` if
+                // there's no rlib provided, so setting a dummy path here to workaround those errors.
+                Rc::make_mut(&mut crate_source).rlib = Some((PathBuf::new(), PathKind::All));
+                crate_source
+            };
+        });
+    }
+
     fn after_analysis<'tcx>(
         &mut self,
         compiler: &rustc_interface::interface::Compiler,
@@ -67,6 +90,39 @@ impl rustc_driver::Callbacks for MiriCompilerCalls {
     }
 }
 
+struct MiriBeRustCompilerCalls {
+    target_crate: bool,
+}
+
+impl rustc_driver::Callbacks for MiriBeRustCompilerCalls {
+    fn config(&mut self, config: &mut Config) {
+        if config.opts.prints.is_empty() && self.target_crate {
+            // Queries overriden here affect the data stored in `rmeta` files of dependencies,
+            // which will be used later in non-`MIRI_BE_RUSTC` mode.
+            config.override_queries = Some(|_, local_providers, _| {
+                // `exported_symbols()` provided by rustc always returns empty result if
+                // `tcx.sess.opts.output_types.should_codegen()` is false.
+                local_providers.exported_symbols = |tcx, cnum| {
+                    assert_eq!(cnum, LOCAL_CRATE);
+                    tcx.arena.alloc_from_iter(
+                        // This is based on:
+                        // https://github.com/rust-lang/rust/blob/2962e7c0089d5c136f4e9600b7abccfbbde4973d/compiler/rustc_codegen_ssa/src/back/symbol_export.rs#L62-L63
+                        // https://github.com/rust-lang/rust/blob/2962e7c0089d5c136f4e9600b7abccfbbde4973d/compiler/rustc_codegen_ssa/src/back/symbol_export.rs#L174
+                        tcx.reachable_set(()).iter().filter_map(|&local_def_id| {
+                            tcx.codegen_fn_attrs(local_def_id)
+                                .contains_extern_indicator()
+                                .then_some((
+                                    ExportedSymbol::NonGeneric(local_def_id.to_def_id()),
+                                    SymbolExportLevel::C,
+                                ))
+                        }),
+                    )
+                }
+            });
+        }
+    }
+}
+
 fn init_early_loggers() {
     // Note that our `extern crate log` is *not* the same as rustc's; as a result, we have to
     // initialize them both, and we always initialize `miri`'s first.
@@ -179,11 +235,7 @@ fn main() {
     if let Some(crate_kind) = env::var_os("MIRI_BE_RUSTC") {
         rustc_driver::init_rustc_env_logger();
 
-        // Don't insert `MIRI_DEFAULT_ARGS`, in particular, `--cfg=miri`, if we are building a
-        // "host" crate. That may cause procedural macros (and probably build scripts) to depend
-        // on Miri-only symbols, such as `miri_resolve_frame`:
-        // https://github.com/rust-lang/miri/issues/1760
-        let insert_default_args = if crate_kind == "target" {
+        let target_crate = if crate_kind == "target" {
             true
         } else if crate_kind == "host" {
             false
@@ -192,8 +244,16 @@ fn main() {
         };
 
         // We cannot use `rustc_driver::main` as we need to adjust the CLI arguments.
-        let mut callbacks = rustc_driver::TimePassesCallbacks::default();
-        run_compiler(env::args().collect(), &mut callbacks, insert_default_args)
+        run_compiler(
+            env::args().collect(),
+            &mut MiriBeRustCompilerCalls { target_crate },
+            // Don't insert `MIRI_DEFAULT_ARGS`, in particular, `--cfg=miri`, if we are building
+            // a "host" crate. That may cause procedural macros (and probably build scripts) to
+            // depend on Miri-only symbols, such as `miri_resolve_frame`:
+            // https://github.com/rust-lang/miri/issues/1760
+            #[rustfmt::skip]
+            /* insert_default_args: */ target_crate,
+        )
     }
 
     // Init loggers the Miri way.
@@ -227,6 +287,9 @@ fn main() {
                 "-Zmiri-symbolic-alignment-check" => {
                     miri_config.check_alignment = miri::AlignmentCheck::Symbolic;
                 }
+                "-Zmiri-disable-abi-check" => {
+                    miri_config.check_abi = false;
+                }
                 "-Zmiri-disable-isolation" => {
                     miri_config.communicate = true;
                 }

src/diagnostics.rs

@@ -5,7 +5,7 @@ use std::num::NonZeroU64;
 use log::trace;
 
 use rustc_middle::ty::{self, TyCtxt};
-use rustc_span::{source_map::DUMMY_SP, Span};
+use rustc_span::{source_map::DUMMY_SP, Span, SpanData, Symbol};
 
 use crate::*;
 
@@ -14,8 +14,18 @@ pub enum TerminationInfo {
     Exit(i64),
     Abort(String),
     UnsupportedInIsolation(String),
-    ExperimentalUb { msg: String, url: String },
+    ExperimentalUb {
+        msg: String,
+        url: String,
+    },
     Deadlock,
+    MultipleSymbolDefinitions {
+        link_name: Symbol,
+        first: SpanData,
+        first_crate: Symbol,
+        second: SpanData,
+        second_crate: Symbol,
+    },
 }
 
 impl fmt::Display for TerminationInfo {
@@ -27,6 +37,8 @@ impl fmt::Display for TerminationInfo {
             UnsupportedInIsolation(msg) => write!(f, "{}", msg),
             ExperimentalUb { msg, .. } => write!(f, "{}", msg),
             Deadlock => write!(f, "the evaluated program deadlocked"),
+            MultipleSymbolDefinitions { link_name, .. } =>
+                write!(f, "multiple definitions of symbol `{}`", link_name),
         }
     }
 }
@@ -55,19 +67,25 @@ pub fn report_error<'tcx, 'mir>(
             use TerminationInfo::*;
             let title = match info {
                 Exit(code) => return Some(*code),
-                Abort(_) => "abnormal termination",
-                UnsupportedInIsolation(_) => "unsupported operation",
-                ExperimentalUb { .. } => "Undefined Behavior",
-                Deadlock => "deadlock",
+                Abort(_) => Some("abnormal termination"),
+                UnsupportedInIsolation(_) => Some("unsupported operation"),
+                ExperimentalUb { .. } => Some("Undefined Behavior"),
+                Deadlock => Some("deadlock"),
+                MultipleSymbolDefinitions { .. } => None,
             };
             #[rustfmt::skip]
             let helps = match info {
                 UnsupportedInIsolation(_) =>
-                    vec![format!("pass the flag `-Zmiri-disable-isolation` to disable isolation")],
+                    vec![(None, format!("pass the flag `-Zmiri-disable-isolation` to disable isolation"))],
                 ExperimentalUb { url, .. } =>
                     vec![
-                        format!("this indicates a potential bug in the program: it performed an invalid operation, but the rules it violated are still experimental"),
-                        format!("see {} for further information", url),
+                        (None, format!("this indicates a potential bug in the program: it performed an invalid operation, but the rules it violated are still experimental")),
+                        (None, format!("see {} for further information", url)),
+                    ],
+                MultipleSymbolDefinitions { first, first_crate, second, second_crate, .. } =>
+                    vec![
+                        (Some(*first), format!("it's first defined here, in crate `{}`", first_crate)),
+                        (Some(*second), format!("then it's defined here again, in crate `{}`", second_crate)),
                     ],
                 _ => vec![],
             };
@@ -92,26 +110,26 @@ pub fn report_error<'tcx, 'mir>(
             #[rustfmt::skip]
             let helps = match e.kind() {
                 Unsupported(UnsupportedOpInfo::NoMirFor(..)) =>
-                    vec![format!("make sure to use a Miri sysroot, which you can prepare with `cargo miri setup`")],
+                    vec![(None, format!("make sure to use a Miri sysroot, which you can prepare with `cargo miri setup`"))],
                 Unsupported(UnsupportedOpInfo::ReadBytesAsPointer | UnsupportedOpInfo::ThreadLocalStatic(_) | UnsupportedOpInfo::ReadExternStatic(_)) =>
                     panic!("Error should never be raised by Miri: {:?}", e.kind()),
                 Unsupported(_) =>
-                    vec![format!("this is likely not a bug in the program; it indicates that the program performed an operation that the interpreter does not support")],
+                    vec![(None, format!("this is likely not a bug in the program; it indicates that the program performed an operation that the interpreter does not support"))],
                 UndefinedBehavior(UndefinedBehaviorInfo::AlignmentCheckFailed { .. })
                     if ecx.memory.extra.check_alignment == AlignmentCheck::Symbolic
                 =>
                     vec![
-                        format!("this usually indicates that your program performed an invalid operation and caused Undefined Behavior"),
-                        format!("but due to `-Zmiri-symbolic-alignment-check`, alignment errors can also be false positives"),
+                        (None, format!("this usually indicates that your program performed an invalid operation and caused Undefined Behavior")),
+                        (None, format!("but due to `-Zmiri-symbolic-alignment-check`, alignment errors can also be false positives")),
                     ],
                 UndefinedBehavior(_) =>
                     vec![
-                        format!("this indicates a bug in the program: it performed an invalid operation, and caused Undefined Behavior"),
-                        format!("see https://doc.rust-lang.org/nightly/reference/behavior-considered-undefined.html for further information"),
+                        (None, format!("this indicates a bug in the program: it performed an invalid operation, and caused Undefined Behavior")),
+                        (None, format!("see https://doc.rust-lang.org/nightly/reference/behavior-considered-undefined.html for further information")),
                     ],
                 _ => vec![],
             };
-            (title, helps)
+            (Some(title), helps)
         }
     };
 
@@ -120,7 +138,7 @@ pub fn report_error<'tcx, 'mir>(
     report_msg(
         *ecx.tcx,
         /*error*/ true,
-        &format!("{}: {}", title, msg),
+        &if let Some(title) = title { format!("{}: {}", title, msg) } else { msg.clone() },
         msg,
         helps,
         &ecx.generate_stacktrace(),
@@ -159,7 +177,7 @@ fn report_msg<'tcx>(
     error: bool,
     title: &str,
     span_msg: String,
-    mut helps: Vec<String>,
+    mut helps: Vec<(Option<SpanData>, String)>,
     stacktrace: &[FrameInfo<'tcx>],
 ) {
     let span = stacktrace.first().map_or(DUMMY_SP, |fi| fi.span);
@@ -179,9 +197,13 @@ fn report_msg<'tcx>(
     // Show help messages.
     if !helps.is_empty() {
         // Add visual separator before backtrace.
-        helps.last_mut().unwrap().push_str("\n");
-        for help in helps {
-            err.help(&help);
+        helps.last_mut().unwrap().1.push_str("\n");
+        for (span_data, help) in helps {
+            if let Some(span_data) = span_data {
+                err.span_help(span_data.span(), &help);
+            } else {
+                err.help(&help);
+            }
         }
     }
     // Add backtrace

src/eval.rs

@@ -31,6 +31,8 @@ pub struct MiriConfig {
     pub stacked_borrows: bool,
     /// Controls alignment checking.
     pub check_alignment: AlignmentCheck,
+    /// Controls function [ABI](Abi) checking.
+    pub check_abi: bool,
     /// Determines if communication with the host environment is enabled.
     pub communicate: bool,
     /// Determines if memory leaks should be ignored.
@@ -65,6 +67,7 @@ impl Default for MiriConfig {
             validate: true,
             stacked_borrows: true,
             check_alignment: AlignmentCheck::Int,
+            check_abi: true,
             communicate: false,
             ignore_leaks: false,
             excluded_env_vars: vec![],

src/helpers.rs

@@ -165,7 +165,7 @@ pub trait EvalContextExt<'mir, 'tcx: 'mir>: crate::MiriEvalContextExt<'mir, 'tcx
         let this = self.eval_context_mut();
         let param_env = ty::ParamEnv::reveal_all(); // in Miri this is always the param_env we use... and this.param_env is private.
         let callee_abi = f.ty(*this.tcx, param_env).fn_sig(*this.tcx).abi();
-        if callee_abi != caller_abi {
+        if this.machine.enforce_abi && callee_abi != caller_abi {
             throw_ub_format!(
                 "calling a function with ABI {} using caller ABI {}",
                 callee_abi.name(),
@@ -616,6 +616,18 @@ pub trait EvalContextExt<'mir, 'tcx: 'mir>: crate::MiriEvalContextExt<'mir, 'tcx
 
         Ok(wchars)
     }
+
+    /// Check that the ABI is what we expect.
+    fn check_abi<'a>(&self, abi: Abi, exp_abi: Abi) -> InterpResult<'a, ()> {
+        if self.eval_context_ref().machine.enforce_abi && abi != exp_abi {
+            throw_ub_format!(
+                "calling a function with ABI {} using caller ABI {}",
+                exp_abi.name(),
+                abi.name()
+            )
+        }
+        Ok(())
+    }
 }
 
 /// Check that the number of args is what we expect.
@@ -631,19 +643,6 @@ where
     throw_ub_format!("incorrect number of arguments: got {}, expected {}", args.len(), N)
 }
 
-/// Check that the ABI is what we expect.
-pub fn check_abi<'a>(abi: Abi, exp_abi: Abi) -> InterpResult<'a, ()> {
-    if abi == exp_abi {
-        Ok(())
-    } else {
-        throw_ub_format!(
-            "calling a function with ABI {} using caller ABI {}",
-            exp_abi.name(),
-            abi.name()
-        )
-    }
-}
-
 pub fn isolation_error(name: &str) -> InterpResult<'static> {
     throw_machine_stop!(TerminationInfo::UnsupportedInIsolation(format!(
         "{} not available when isolation is enabled",

src/lib.rs

@@ -3,6 +3,7 @@
 #![feature(map_try_insert)]
 #![feature(never_type)]
 #![feature(try_blocks)]
+#![feature(bool_to_option)]
 #![warn(rust_2018_idioms)]
 #![allow(clippy::cast_lossless)]
 
@@ -14,6 +15,7 @@ extern crate rustc_data_structures;
 extern crate rustc_hir;
 extern crate rustc_index;
 extern crate rustc_mir;
+extern crate rustc_session;
 extern crate rustc_span;
 extern crate rustc_target;