Rollup merge of #134497 - Zalathar:spans, r=jieyouxu

coverage: Store coverage source regions as `Span` until codegen (take 2)

This is an attempt to re-land #133418:

> Historically, coverage spans were converted into line/column coordinates during the MIR instrumentation pass.

> This PR moves that conversion step into codegen, so that coverage spans spend most of their time stored as Span instead.

> In addition to being conceptually nicer, this also reduces the size of coverage mappings in MIR, because Span is smaller than 4x u32.

That PR was reverted by #133608, because in some circumstances not covered by our test suite we were emitting coverage metadata that was causing `llvm-cov` to exit with an error (#133606).

---

The implementation here is *mostly* the same, but adapted for subsequent changes in the relevant code (e.g. #134163).

I believe that the changes in #134163 should be sufficient to prevent the problem that required the original PR to be reverted. But I haven't been able to reproduce the original breakage in a regression test, and the `llvm-cov` error message is extremely unhelpful, so I can't completely rule out the possibility of this breaking again.

r? jieyouxu (reviewer of the original PR)

compiler/rustc_codegen_llvm/src/coverageinfo/ffi.rs

@@ -1,6 +1,4 @@
-use rustc_middle::mir::coverage::{CounterId, CovTerm, ExpressionId, SourceRegion};
+use rustc_middle::mir::coverage::{CounterId, CovTerm, ExpressionId};
-
-use crate::coverageinfo::mapgen::LocalFileId;
 
 /// Must match the layout of `LLVMRustCounterKind`.
 #[derive(Copy, Clone, Debug)]
@@ -126,30 +124,16 @@ pub(crate) struct CoverageSpan {
     /// Local index into the function's local-to-global file ID table.
     /// The value at that index is itself an index into the coverage filename
     /// table in the CGU's `__llvm_covmap` section.
-    file_id: u32,
+    pub(crate) file_id: u32,
 
     /// 1-based starting line of the source code span.
-    start_line: u32,
+    pub(crate) start_line: u32,
     /// 1-based starting column of the source code span.
-    start_col: u32,
+    pub(crate) start_col: u32,
     /// 1-based ending line of the source code span.
-    end_line: u32,
+    pub(crate) end_line: u32,
     /// 1-based ending column of the source code span. High bit must be unset.
-    end_col: u32,
+    pub(crate) end_col: u32,
-}
-
-impl CoverageSpan {
-    pub(crate) fn from_source_region(
-        local_file_id: LocalFileId,
-        code_region: &SourceRegion,
-    ) -> Self {
-        let file_id = local_file_id.as_u32();
-        let &SourceRegion { start_line, start_col, end_line, end_col } = code_region;
-        // Internally, LLVM uses the high bit of `end_col` to distinguish between
-        // code regions and gap regions, so it can't be used by the column number.
-        assert!(end_col & (1u32 << 31) == 0, "high bit of `end_col` must be unset: {end_col:#X}");
-        Self { file_id, start_line, start_col, end_line, end_col }
-    }
 }
 
 /// Holds tables of the various region types in one struct.
@@ -184,7 +168,7 @@ impl Regions {
 #[derive(Clone, Debug)]
 #[repr(C)]
 pub(crate) struct CodeRegion {
-    pub(crate) span: CoverageSpan,
+    pub(crate) cov_span: CoverageSpan,
     pub(crate) counter: Counter,
 }
 
@@ -192,7 +176,7 @@ pub(crate) struct CodeRegion {
 #[derive(Clone, Debug)]
 #[repr(C)]
 pub(crate) struct BranchRegion {
-    pub(crate) span: CoverageSpan,
+    pub(crate) cov_span: CoverageSpan,
     pub(crate) true_counter: Counter,
     pub(crate) false_counter: Counter,
 }
@@ -201,7 +185,7 @@ pub(crate) struct BranchRegion {
 #[derive(Clone, Debug)]
 #[repr(C)]
 pub(crate) struct MCDCBranchRegion {
-    pub(crate) span: CoverageSpan,
+    pub(crate) cov_span: CoverageSpan,
     pub(crate) true_counter: Counter,
     pub(crate) false_counter: Counter,
     pub(crate) mcdc_branch_params: mcdc::BranchParameters,
@@ -211,6 +195,6 @@ pub(crate) struct MCDCBranchRegion {
 #[derive(Clone, Debug)]
 #[repr(C)]
 pub(crate) struct MCDCDecisionRegion {
-    pub(crate) span: CoverageSpan,
+    pub(crate) cov_span: CoverageSpan,
     pub(crate) mcdc_decision_params: mcdc::DecisionParameters,
 }

compiler/rustc_codegen_llvm/src/coverageinfo/llvm_cov.rs

@@ -40,11 +40,10 @@ pub(crate) fn create_pgo_func_name_var<'ll>(
     }
 }
 
-pub(crate) fn write_filenames_to_buffer<'a>(
+pub(crate) fn write_filenames_to_buffer(filenames: &[impl AsRef<str>]) -> Vec<u8> {
-    filenames: impl IntoIterator<Item = &'a str>,
-) -> Vec<u8> {
     let (pointers, lengths) = filenames
         .into_iter()
+        .map(AsRef::as_ref)
         .map(|s: &str| (s.as_c_char_ptr(), s.len()))
         .unzip::<_, _, Vec<_>, Vec<_>>();
 

compiler/rustc_codegen_llvm/src/coverageinfo/mapgen.rs

@@ -1,11 +1,11 @@
-use std::iter;
+use std::sync::Arc;
 
 use itertools::Itertools;
 use rustc_abi::Align;
 use rustc_codegen_ssa::traits::{
     BaseTypeCodegenMethods, ConstCodegenMethods, StaticCodegenMethods,
 };
-use rustc_data_structures::fx::{FxHashSet, FxIndexMap, FxIndexSet};
+use rustc_data_structures::fx::{FxHashSet, FxIndexMap};
 use rustc_hir::def_id::{DefId, LocalDefId};
 use rustc_index::IndexVec;
 use rustc_middle::mir;
@@ -13,7 +13,7 @@ use rustc_middle::ty::{self, TyCtxt};
 use rustc_session::RemapFileNameExt;
 use rustc_session::config::RemapPathScopeComponents;
 use rustc_span::def_id::DefIdSet;
-use rustc_span::{Span, Symbol};
+use rustc_span::{SourceFile, StableSourceFileId};
 use tracing::debug;
 
 use crate::common::CodegenCx;
@@ -22,6 +22,7 @@ use crate::coverageinfo::mapgen::covfun::prepare_covfun_record;
 use crate::llvm;
 
 mod covfun;
+mod spans;
 
 /// Generates and exports the coverage map, which is embedded in special
 /// linker sections in the final binary.
@@ -131,45 +132,51 @@ pub(crate) fn finalize(cx: &CodegenCx<'_, '_>) {
     generate_covmap_record(cx, covmap_version, &filenames_buffer);
 }
 
-/// Maps "global" (per-CGU) file ID numbers to their underlying filenames.
+/// Maps "global" (per-CGU) file ID numbers to their underlying source files.
 struct GlobalFileTable {
-    /// This "raw" table doesn't include the working dir, so a filename's
+    /// This "raw" table doesn't include the working dir, so a file's
     /// global ID is its index in this set **plus one**.
-    raw_file_table: FxIndexSet<Symbol>,
+    raw_file_table: FxIndexMap<StableSourceFileId, Arc<SourceFile>>,
 }
 
 impl GlobalFileTable {
     fn new() -> Self {
-        Self { raw_file_table: FxIndexSet::default() }
+        Self { raw_file_table: FxIndexMap::default() }
     }
 
-    fn global_file_id_for_file_name(&mut self, file_name: Symbol) -> GlobalFileId {
+    fn global_file_id_for_file(&mut self, file: &Arc<SourceFile>) -> GlobalFileId {
         // Ensure the given file has a table entry, and get its index.
-        let (raw_id, _) = self.raw_file_table.insert_full(file_name);
+        let entry = self.raw_file_table.entry(file.stable_id);
+        let raw_id = entry.index();
+        entry.or_insert_with(|| Arc::clone(file));
+
         // The raw file table doesn't include an entry for the working dir
         // (which has ID 0), so add 1 to get the correct ID.
         GlobalFileId::from_usize(raw_id + 1)
     }
 
     fn make_filenames_buffer(&self, tcx: TyCtxt<'_>) -> Vec<u8> {
+        let mut table = Vec::with_capacity(self.raw_file_table.len() + 1);
+
         // LLVM Coverage Mapping Format version 6 (zero-based encoded as 5)
         // requires setting the first filename to the compilation directory.
         // Since rustc generates coverage maps with relative paths, the
         // compilation directory can be combined with the relative paths
         // to get absolute paths, if needed.
-        use rustc_session::RemapFileNameExt;
+        table.push(
-        use rustc_session::config::RemapPathScopeComponents;
+            tcx.sess
-        let working_dir: &str = &tcx
+                .opts
-            .sess
+                .working_dir
-            .opts
+                .for_scope(tcx.sess, RemapPathScopeComponents::MACRO)
-            .working_dir
+                .to_string_lossy(),
-            .for_scope(tcx.sess, RemapPathScopeComponents::MACRO)
+        );
-            .to_string_lossy();
 
-        // Insert the working dir at index 0, before the other filenames.
+        // Add the regular entries after the base directory.
-        let filenames =
+        table.extend(self.raw_file_table.values().map(|file| {
-            iter::once(working_dir).chain(self.raw_file_table.iter().map(Symbol::as_str));
+            file.name.for_scope(tcx.sess, RemapPathScopeComponents::MACRO).to_string_lossy()
-        llvm_cov::write_filenames_to_buffer(filenames)
+        }));
+
+        llvm_cov::write_filenames_to_buffer(&table)
     }
 }
 
@@ -182,7 +189,7 @@ rustc_index::newtype_index! {
     /// An index into a function's list of global file IDs. That underlying list
     /// of local-to-global mappings will be embedded in the function's record in
     /// the `__llvm_covfun` linker section.
-    pub(crate) struct LocalFileId {}
+    struct LocalFileId {}
 }
 
 /// Holds a mapping from "local" (per-function) file IDs to "global" (per-CGU)
@@ -208,13 +215,6 @@ impl VirtualFileMapping {
     }
 }
 
-fn span_file_name(tcx: TyCtxt<'_>, span: Span) -> Symbol {
-    let source_file = tcx.sess.source_map().lookup_source_file(span.lo());
-    let name =
-        source_file.name.for_scope(tcx.sess, RemapPathScopeComponents::MACRO).to_string_lossy();
-    Symbol::intern(&name)
-}
-
 /// Generates the contents of the covmap record for this CGU, which mostly
 /// consists of a header and a list of filenames. The record is then stored
 /// as a global variable in the `__llvm_covmap` section.

compiler/rustc_codegen_llvm/src/coverageinfo/mapgen/covfun.rs

@@ -10,16 +10,16 @@ use rustc_abi::Align;
 use rustc_codegen_ssa::traits::{
     BaseTypeCodegenMethods, ConstCodegenMethods, StaticCodegenMethods,
 };
-use rustc_middle::bug;
 use rustc_middle::mir::coverage::{
     CovTerm, CoverageIdsInfo, Expression, FunctionCoverageInfo, Mapping, MappingKind, Op,
 };
 use rustc_middle::ty::{Instance, TyCtxt};
+use rustc_span::Span;
 use rustc_target::spec::HasTargetSpec;
 use tracing::debug;
 
 use crate::common::CodegenCx;
-use crate::coverageinfo::mapgen::{GlobalFileTable, VirtualFileMapping, span_file_name};
+use crate::coverageinfo::mapgen::{GlobalFileTable, VirtualFileMapping, spans};
 use crate::coverageinfo::{ffi, llvm_cov};
 use crate::llvm;
 
@@ -67,12 +67,8 @@ pub(crate) fn prepare_covfun_record<'tcx>(
     fill_region_tables(tcx, global_file_table, fn_cov_info, ids_info, &mut covfun);
 
     if covfun.regions.has_no_regions() {
-        if covfun.is_used {
+        debug!(?covfun, "function has no mappings to embed; skipping");
-            bug!("a used function should have had coverage mapping data but did not: {covfun:?}");
+        return None;
-        } else {
-            debug!(?covfun, "unused function had no coverage mapping data");
-            return None;
-        }
     }
 
     Some(covfun)
@@ -121,41 +117,58 @@ fn fill_region_tables<'tcx>(
     covfun: &mut CovfunRecord<'tcx>,
 ) {
     // Currently a function's mappings must all be in the same file as its body span.
-    let file_name = span_file_name(tcx, fn_cov_info.body_span);
+    let source_map = tcx.sess.source_map();
+    let source_file = source_map.lookup_source_file(fn_cov_info.body_span.lo());
 
-    // Look up the global file ID for that filename.
+    // Look up the global file ID for that file.
-    let global_file_id = global_file_table.global_file_id_for_file_name(file_name);
+    let global_file_id = global_file_table.global_file_id_for_file(&source_file);
 
     // Associate that global file ID with a local file ID for this function.
     let local_file_id = covfun.virtual_file_mapping.local_id_for_global(global_file_id);
-    debug!("  file id: {local_file_id:?} => {global_file_id:?} = '{file_name:?}'");
 
     let ffi::Regions { code_regions, branch_regions, mcdc_branch_regions, mcdc_decision_regions } =
         &mut covfun.regions;
 
+    let make_cov_span = |span: Span| {
+        spans::make_coverage_span(local_file_id, source_map, fn_cov_info, &source_file, span)
+    };
+    let discard_all = tcx.sess.coverage_discard_all_spans_in_codegen();
+
     // For each counter/region pair in this function+file, convert it to a
     // form suitable for FFI.
     let is_zero_term = |term| !covfun.is_used || ids_info.is_zero_term(term);
-    for Mapping { kind, ref source_region } in &fn_cov_info.mappings {
+    for &Mapping { ref kind, span } in &fn_cov_info.mappings {
         // If the mapping refers to counters/expressions that were removed by
         // MIR opts, replace those occurrences with zero.
         let kind = kind.map_terms(|term| if is_zero_term(term) { CovTerm::Zero } else { term });
 
-        let span = ffi::CoverageSpan::from_source_region(local_file_id, source_region);
+        // Convert the `Span` into coordinates that we can pass to LLVM, or
+        // discard the span if conversion fails. In rare, cases _all_ of a
+        // function's spans are discarded, and the rest of coverage codegen
+        // needs to handle that gracefully to avoid a repeat of #133606.
+        // We don't have a good test case for triggering that organically, so
+        // instead we set `-Zcoverage-options=discard-all-spans-in-codegen`
+        // to force it to occur.
+        let Some(cov_span) = make_cov_span(span) else { continue };
+        if discard_all {
+            continue;
+        }
+
         match kind {
             MappingKind::Code(term) => {
-                code_regions.push(ffi::CodeRegion { span, counter: ffi::Counter::from_term(term) });
+                code_regions
+                    .push(ffi::CodeRegion { cov_span, counter: ffi::Counter::from_term(term) });
             }
             MappingKind::Branch { true_term, false_term } => {
                 branch_regions.push(ffi::BranchRegion {
-                    span,
+                    cov_span,
                     true_counter: ffi::Counter::from_term(true_term),
                     false_counter: ffi::Counter::from_term(false_term),
                 });
             }
             MappingKind::MCDCBranch { true_term, false_term, mcdc_params } => {
                 mcdc_branch_regions.push(ffi::MCDCBranchRegion {
-                    span,
+                    cov_span,
                     true_counter: ffi::Counter::from_term(true_term),
                     false_counter: ffi::Counter::from_term(false_term),
                     mcdc_branch_params: ffi::mcdc::BranchParameters::from(mcdc_params),
@@ -163,7 +176,7 @@ fn fill_region_tables<'tcx>(
             }
             MappingKind::MCDCDecision(mcdc_decision_params) => {
                 mcdc_decision_regions.push(ffi::MCDCDecisionRegion {
-                    span,
+                    cov_span,
                     mcdc_decision_params: ffi::mcdc::DecisionParameters::from(mcdc_decision_params),
                 });
             }

compiler/rustc_codegen_llvm/src/coverageinfo/mapgen/spans.rs

@@ -0,0 +1,126 @@
+use rustc_middle::mir::coverage::FunctionCoverageInfo;
+use rustc_span::source_map::SourceMap;
+use rustc_span::{BytePos, Pos, SourceFile, Span};
+use tracing::debug;
+
+use crate::coverageinfo::ffi;
+use crate::coverageinfo::mapgen::LocalFileId;
+
+/// Converts the span into its start line and column, and end line and column.
+///
+/// Line numbers and column numbers are 1-based. Unlike most column numbers emitted by
+/// the compiler, these column numbers are denoted in **bytes**, because that's what
+/// LLVM's `llvm-cov` tool expects to see in coverage maps.
+///
+/// Returns `None` if the conversion failed for some reason. This shouldn't happen,
+/// but it's hard to rule out entirely (especially in the presence of complex macros
+/// or other expansions), and if it does happen then skipping a span or function is
+/// better than an ICE or `llvm-cov` failure that the user might have no way to avoid.
+pub(crate) fn make_coverage_span(
+    file_id: LocalFileId,
+    source_map: &SourceMap,
+    fn_cov_info: &FunctionCoverageInfo,
+    file: &SourceFile,
+    span: Span,
+) -> Option<ffi::CoverageSpan> {
+    let span = ensure_non_empty_span(source_map, fn_cov_info, span)?;
+
+    let lo = span.lo();
+    let hi = span.hi();
+
+    // Column numbers need to be in bytes, so we can't use the more convenient
+    // `SourceMap` methods for looking up file coordinates.
+    let line_and_byte_column = |pos: BytePos| -> Option<(usize, usize)> {
+        let rpos = file.relative_position(pos);
+        let line_index = file.lookup_line(rpos)?;
+        let line_start = file.lines()[line_index];
+        // Line numbers and column numbers are 1-based, so add 1 to each.
+        Some((line_index + 1, (rpos - line_start).to_usize() + 1))
+    };
+
+    let (mut start_line, start_col) = line_and_byte_column(lo)?;
+    let (mut end_line, end_col) = line_and_byte_column(hi)?;
+
+    // Apply an offset so that code in doctests has correct line numbers.
+    // FIXME(#79417): Currently we have no way to offset doctest _columns_.
+    start_line = source_map.doctest_offset_line(&file.name, start_line);
+    end_line = source_map.doctest_offset_line(&file.name, end_line);
+
+    check_coverage_span(ffi::CoverageSpan {
+        file_id: file_id.as_u32(),
+        start_line: start_line as u32,
+        start_col: start_col as u32,
+        end_line: end_line as u32,
+        end_col: end_col as u32,
+    })
+}
+
+fn ensure_non_empty_span(
+    source_map: &SourceMap,
+    fn_cov_info: &FunctionCoverageInfo,
+    span: Span,
+) -> Option<Span> {
+    if !span.is_empty() {
+        return Some(span);
+    }
+
+    let lo = span.lo();
+    let hi = span.hi();
+
+    // The span is empty, so try to expand it to cover an adjacent '{' or '}',
+    // but only within the bounds of the body span.
+    let try_next = hi < fn_cov_info.body_span.hi();
+    let try_prev = fn_cov_info.body_span.lo() < lo;
+    if !(try_next || try_prev) {
+        return None;
+    }
+
+    source_map
+        .span_to_source(span, |src, start, end| try {
+            // Adjusting span endpoints by `BytePos(1)` is normally a bug,
+            // but in this case we have specifically checked that the character
+            // we're skipping over is one of two specific ASCII characters, so
+            // adjusting by exactly 1 byte is correct.
+            if try_next && src.as_bytes()[end] == b'{' {
+                Some(span.with_hi(hi + BytePos(1)))
+            } else if try_prev && src.as_bytes()[start - 1] == b'}' {
+                Some(span.with_lo(lo - BytePos(1)))
+            } else {
+                None
+            }
+        })
+        .ok()?
+}
+
+/// If `llvm-cov` sees a source region that is improperly ordered (end < start),
+/// it will immediately exit with a fatal error. To prevent that from happening,
+/// discard regions that are improperly ordered, or might be interpreted in a
+/// way that makes them improperly ordered.
+fn check_coverage_span(cov_span: ffi::CoverageSpan) -> Option<ffi::CoverageSpan> {
+    let ffi::CoverageSpan { file_id: _, start_line, start_col, end_line, end_col } = cov_span;
+
+    // Line/column coordinates are supposed to be 1-based. If we ever emit
+    // coordinates of 0, `llvm-cov` might misinterpret them.
+    let all_nonzero = [start_line, start_col, end_line, end_col].into_iter().all(|x| x != 0);
+    // Coverage mappings use the high bit of `end_col` to indicate that a
+    // region is actually a "gap" region, so make sure it's unset.
+    let end_col_has_high_bit_unset = (end_col & (1 << 31)) == 0;
+    // If a region is improperly ordered (end < start), `llvm-cov` will exit
+    // with a fatal error, which is inconvenient for users and hard to debug.
+    let is_ordered = (start_line, start_col) <= (end_line, end_col);
+
+    if all_nonzero && end_col_has_high_bit_unset && is_ordered {
+        Some(cov_span)
+    } else {
+        debug!(
+            ?cov_span,
+            ?all_nonzero,
+            ?end_col_has_high_bit_unset,
+            ?is_ordered,
+            "Skipping source region that would be misinterpreted or rejected by LLVM"
+        );
+        // If this happens in a debug build, ICE to make it easier to notice.
+        debug_assert!(false, "Improper source region: {cov_span:?}");
+        None
+    }
+}

compiler/rustc_codegen_llvm/src/lib.rs

@@ -17,6 +17,7 @@
 #![feature(iter_intersperse)]
 #![feature(let_chains)]
 #![feature(rustdoc_internals)]
+#![feature(try_blocks)]
 #![warn(unreachable_pub)]
 // tidy-alphabetical-end
 

compiler/rustc_interface/src/tests.rs

@@ -766,7 +766,11 @@ fn test_unstable_options_tracking_hash() {
         })
     );
     tracked!(codegen_backend, Some("abc".to_string()));
-    tracked!(coverage_options, CoverageOptions { level: CoverageLevel::Mcdc, no_mir_spans: true });
+    tracked!(coverage_options, CoverageOptions {
+        level: CoverageLevel::Mcdc,
+        no_mir_spans: true,
+        discard_all_spans_in_codegen: true
+    });
     tracked!(crate_attr, vec!["abc".to_string()]);
     tracked!(cross_crate_inline_threshold, InliningThreshold::Always);
     tracked!(debug_info_for_profiling, true);

compiler/rustc_middle/src/mir/coverage.rs

@@ -154,22 +154,6 @@ impl Debug for CoverageKind {
     }
 }
 
-#[derive(Clone, TyEncodable, TyDecodable, Hash, HashStable, PartialEq, Eq, PartialOrd, Ord)]
-#[derive(TypeFoldable, TypeVisitable)]
-pub struct SourceRegion {
-    pub start_line: u32,
-    pub start_col: u32,
-    pub end_line: u32,
-    pub end_col: u32,
-}
-
-impl Debug for SourceRegion {
-    fn fmt(&self, fmt: &mut Formatter<'_>) -> fmt::Result {
-        let &Self { start_line, start_col, end_line, end_col } = self;
-        write!(fmt, "{start_line}:{start_col} - {end_line}:{end_col}")
-    }
-}
-
 #[derive(Copy, Clone, Debug, PartialEq, Eq, Hash, HashStable)]
 #[derive(TyEncodable, TyDecodable, TypeFoldable, TypeVisitable)]
 pub enum Op {
@@ -231,7 +215,7 @@ impl MappingKind {
 #[derive(TyEncodable, TyDecodable, Hash, HashStable, TypeFoldable, TypeVisitable)]
 pub struct Mapping {
     pub kind: MappingKind,
-    pub source_region: SourceRegion,
+    pub span: Span,
 }
 
 /// Stores per-function coverage information attached to a `mir::Body`,

compiler/rustc_middle/src/mir/pretty.rs

@@ -603,8 +603,8 @@ fn write_function_coverage_info(
     for (id, expression) in expressions.iter_enumerated() {
         writeln!(w, "{INDENT}coverage {id:?} => {expression:?};")?;
     }
-    for coverage::Mapping { kind, source_region } in mappings {
+    for coverage::Mapping { kind, span } in mappings {
-        writeln!(w, "{INDENT}coverage {kind:?} => {source_region:?};")?;
+        writeln!(w, "{INDENT}coverage {kind:?} => {span:?};")?;
     }
     writeln!(w)?;
 

compiler/rustc_mir_transform/src/coverage/mod.rs

@@ -13,16 +13,15 @@ use rustc_hir::intravisit::{Visitor, walk_expr};
 use rustc_middle::hir::map::Map;
 use rustc_middle::hir::nested_filter;
 use rustc_middle::mir::coverage::{
-    CoverageKind, DecisionInfo, FunctionCoverageInfo, Mapping, MappingKind, SourceRegion,
+    CoverageKind, DecisionInfo, FunctionCoverageInfo, Mapping, MappingKind,
 };
 use rustc_middle::mir::{
     self, BasicBlock, BasicBlockData, SourceInfo, Statement, StatementKind, Terminator,
     TerminatorKind,
 };
 use rustc_middle::ty::TyCtxt;
+use rustc_span::Span;
 use rustc_span::def_id::LocalDefId;
-use rustc_span::source_map::SourceMap;
-use rustc_span::{BytePos, Pos, SourceFile, Span};
 use tracing::{debug, debug_span, trace};
 
 use crate::coverage::counters::{CoverageCounters, Site};
@@ -97,7 +96,7 @@ fn instrument_function_for_coverage<'tcx>(tcx: TyCtxt<'tcx>, mir_body: &mut mir:
     let coverage_counters =
         CoverageCounters::make_bcb_counters(&basic_coverage_blocks, &bcbs_with_counter_mappings);
 
-    let mappings = create_mappings(tcx, &hir_info, &extracted_mappings, &coverage_counters);
+    let mappings = create_mappings(&extracted_mappings, &coverage_counters);
     if mappings.is_empty() {
         // No spans could be converted into valid mappings, so skip this function.
         debug!("no spans could be converted into valid mappings; skipping");
@@ -136,18 +135,12 @@ fn instrument_function_for_coverage<'tcx>(tcx: TyCtxt<'tcx>, mir_body: &mut mir:
 ///
 /// Precondition: All BCBs corresponding to those spans have been given
 /// coverage counters.
-fn create_mappings<'tcx>(
+fn create_mappings(
-    tcx: TyCtxt<'tcx>,
-    hir_info: &ExtractedHirInfo,
     extracted_mappings: &ExtractedMappings,
     coverage_counters: &CoverageCounters,
 ) -> Vec<Mapping> {
-    let source_map = tcx.sess.source_map();
-    let file = source_map.lookup_source_file(hir_info.body_span.lo());
-
     let term_for_bcb =
         |bcb| coverage_counters.term_for_bcb(bcb).expect("all BCBs with spans were given counters");
-    let region_for_span = |span: Span| make_source_region(source_map, hir_info, &file, span);
 
     // Fully destructure the mappings struct to make sure we don't miss any kinds.
     let ExtractedMappings {
@@ -160,22 +153,20 @@ fn create_mappings<'tcx>(
     } = extracted_mappings;
     let mut mappings = Vec::new();
 
-    mappings.extend(code_mappings.iter().filter_map(
+    mappings.extend(code_mappings.iter().map(
         // Ordinary code mappings are the simplest kind.
         |&mappings::CodeMapping { span, bcb }| {
-            let source_region = region_for_span(span)?;
             let kind = MappingKind::Code(term_for_bcb(bcb));
-            Some(Mapping { kind, source_region })
+            Mapping { kind, span }
         },
     ));
 
-    mappings.extend(branch_pairs.iter().filter_map(
+    mappings.extend(branch_pairs.iter().map(
         |&mappings::BranchPair { span, true_bcb, false_bcb }| {
             let true_term = term_for_bcb(true_bcb);
             let false_term = term_for_bcb(false_bcb);
             let kind = MappingKind::Branch { true_term, false_term };
-            let source_region = region_for_span(span)?;
+            Mapping { kind, span }
-            Some(Mapping { kind, source_region })
         },
     ));
 
@@ -183,7 +174,7 @@ fn create_mappings<'tcx>(
         |bcb| coverage_counters.term_for_bcb(bcb).expect("all BCBs with spans were given counters");
 
     // MCDC branch mappings are appended with their decisions in case decisions were ignored.
-    mappings.extend(mcdc_degraded_branches.iter().filter_map(
+    mappings.extend(mcdc_degraded_branches.iter().map(
         |&mappings::MCDCBranch {
              span,
              true_bcb,
@@ -192,10 +183,9 @@ fn create_mappings<'tcx>(
              true_index: _,
              false_index: _,
          }| {
-            let source_region = region_for_span(span)?;
             let true_term = term_for_bcb(true_bcb);
             let false_term = term_for_bcb(false_bcb);
-            Some(Mapping { kind: MappingKind::Branch { true_term, false_term }, source_region })
+            Mapping { kind: MappingKind::Branch { true_term, false_term }, span }
         },
     ));
 
@@ -203,7 +193,7 @@ fn create_mappings<'tcx>(
         let num_conditions = branches.len() as u16;
         let conditions = branches
             .into_iter()
-            .filter_map(
+            .map(
                 |&mappings::MCDCBranch {
                      span,
                      true_bcb,
@@ -212,32 +202,28 @@ fn create_mappings<'tcx>(
                      true_index: _,
                      false_index: _,
                  }| {
-                    let source_region = region_for_span(span)?;
                     let true_term = term_for_bcb(true_bcb);
                     let false_term = term_for_bcb(false_bcb);
-                    Some(Mapping {
+                    Mapping {
                         kind: MappingKind::MCDCBranch {
                             true_term,
                             false_term,
                             mcdc_params: condition_info,
                         },
-                        source_region,
+                        span,
-                    })
+                    }
                 },
             )
             .collect::<Vec<_>>();
 
-        if conditions.len() == num_conditions as usize
+        if conditions.len() == num_conditions as usize {
-            && let Some(source_region) = region_for_span(decision.span)
-        {
             // LLVM requires end index for counter mapping regions.
             let kind = MappingKind::MCDCDecision(DecisionInfo {
                 bitmap_idx: (decision.bitmap_idx + decision.num_test_vectors) as u32,
                 num_conditions,
             });
-            mappings.extend(
+            let span = decision.span;
-                std::iter::once(Mapping { kind, source_region }).chain(conditions.into_iter()),
+            mappings.extend(std::iter::once(Mapping { kind, span }).chain(conditions.into_iter()));
-            );
         } else {
             mappings.extend(conditions.into_iter().map(|mapping| {
                 let MappingKind::MCDCBranch { true_term, false_term, mcdc_params: _ } =
@@ -245,10 +231,7 @@ fn create_mappings<'tcx>(
                 else {
                     unreachable!("all mappings here are MCDCBranch as shown above");
                 };
-                Mapping {
+                Mapping { kind: MappingKind::Branch { true_term, false_term }, span: mapping.span }
-                    kind: MappingKind::Branch { true_term, false_term },
-                    source_region: mapping.source_region,
-                }
             }))
         }
     }
@@ -391,121 +374,6 @@ fn inject_statement(mir_body: &mut mir::Body<'_>, counter_kind: CoverageKind, bb
     data.statements.insert(0, statement);
 }
 
-fn ensure_non_empty_span(
-    source_map: &SourceMap,
-    hir_info: &ExtractedHirInfo,
-    span: Span,
-) -> Option<Span> {
-    if !span.is_empty() {
-        return Some(span);
-    }
-
-    let lo = span.lo();
-    let hi = span.hi();
-
-    // The span is empty, so try to expand it to cover an adjacent '{' or '}',
-    // but only within the bounds of the body span.
-    let try_next = hi < hir_info.body_span.hi();
-    let try_prev = hir_info.body_span.lo() < lo;
-    if !(try_next || try_prev) {
-        return None;
-    }
-
-    source_map
-        .span_to_source(span, |src, start, end| try {
-            // We're only checking for specific ASCII characters, so we don't
-            // have to worry about multi-byte code points.
-            if try_next && src.as_bytes()[end] == b'{' {
-                Some(span.with_hi(hi + BytePos(1)))
-            } else if try_prev && src.as_bytes()[start - 1] == b'}' {
-                Some(span.with_lo(lo - BytePos(1)))
-            } else {
-                None
-            }
-        })
-        .ok()?
-}
-
-/// Converts the span into its start line and column, and end line and column.
-///
-/// Line numbers and column numbers are 1-based. Unlike most column numbers emitted by
-/// the compiler, these column numbers are denoted in **bytes**, because that's what
-/// LLVM's `llvm-cov` tool expects to see in coverage maps.
-///
-/// Returns `None` if the conversion failed for some reason. This shouldn't happen,
-/// but it's hard to rule out entirely (especially in the presence of complex macros
-/// or other expansions), and if it does happen then skipping a span or function is
-/// better than an ICE or `llvm-cov` failure that the user might have no way to avoid.
-fn make_source_region(
-    source_map: &SourceMap,
-    hir_info: &ExtractedHirInfo,
-    file: &SourceFile,
-    span: Span,
-) -> Option<SourceRegion> {
-    let span = ensure_non_empty_span(source_map, hir_info, span)?;
-
-    let lo = span.lo();
-    let hi = span.hi();
-
-    // Column numbers need to be in bytes, so we can't use the more convenient
-    // `SourceMap` methods for looking up file coordinates.
-    let line_and_byte_column = |pos: BytePos| -> Option<(usize, usize)> {
-        let rpos = file.relative_position(pos);
-        let line_index = file.lookup_line(rpos)?;
-        let line_start = file.lines()[line_index];
-        // Line numbers and column numbers are 1-based, so add 1 to each.
-        Some((line_index + 1, (rpos - line_start).to_usize() + 1))
-    };
-
-    let (mut start_line, start_col) = line_and_byte_column(lo)?;
-    let (mut end_line, end_col) = line_and_byte_column(hi)?;
-
-    // Apply an offset so that code in doctests has correct line numbers.
-    // FIXME(#79417): Currently we have no way to offset doctest _columns_.
-    start_line = source_map.doctest_offset_line(&file.name, start_line);
-    end_line = source_map.doctest_offset_line(&file.name, end_line);
-
-    check_source_region(SourceRegion {
-        start_line: start_line as u32,
-        start_col: start_col as u32,
-        end_line: end_line as u32,
-        end_col: end_col as u32,
-    })
-}
-
-/// If `llvm-cov` sees a source region that is improperly ordered (end < start),
-/// it will immediately exit with a fatal error. To prevent that from happening,
-/// discard regions that are improperly ordered, or might be interpreted in a
-/// way that makes them improperly ordered.
-fn check_source_region(source_region: SourceRegion) -> Option<SourceRegion> {
-    let SourceRegion { start_line, start_col, end_line, end_col } = source_region;
-
-    // Line/column coordinates are supposed to be 1-based. If we ever emit
-    // coordinates of 0, `llvm-cov` might misinterpret them.
-    let all_nonzero = [start_line, start_col, end_line, end_col].into_iter().all(|x| x != 0);
-    // Coverage mappings use the high bit of `end_col` to indicate that a
-    // region is actually a "gap" region, so make sure it's unset.
-    let end_col_has_high_bit_unset = (end_col & (1 << 31)) == 0;
-    // If a region is improperly ordered (end < start), `llvm-cov` will exit
-    // with a fatal error, which is inconvenient for users and hard to debug.
-    let is_ordered = (start_line, start_col) <= (end_line, end_col);
-
-    if all_nonzero && end_col_has_high_bit_unset && is_ordered {
-        Some(source_region)
-    } else {
-        debug!(
-            ?source_region,
-            ?all_nonzero,
-            ?end_col_has_high_bit_unset,
-            ?is_ordered,
-            "Skipping source region that would be misinterpreted or rejected by LLVM"
-        );
-        // If this happens in a debug build, ICE to make it easier to notice.
-        debug_assert!(false, "Improper source region: {source_region:?}");
-        None
-    }
-}
-
 /// Function information extracted from HIR by the coverage instrumentor.
 #[derive(Debug)]
 struct ExtractedHirInfo {

compiler/rustc_session/src/config.rs

@@ -147,18 +147,24 @@ pub enum InstrumentCoverage {
     Yes,
 }
 
-/// Individual flag values controlled by `-Z coverage-options`.
+/// Individual flag values controlled by `-Zcoverage-options`.
 #[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Default)]
 pub struct CoverageOptions {
     pub level: CoverageLevel,
 
-    /// `-Z coverage-options=no-mir-spans`: Don't extract block coverage spans
+    /// `-Zcoverage-options=no-mir-spans`: Don't extract block coverage spans
     /// from MIR statements/terminators, making it easier to inspect/debug
     /// branch and MC/DC coverage mappings.
     ///
     /// For internal debugging only. If other code changes would make it hard
     /// to keep supporting this flag, remove it.
     pub no_mir_spans: bool,
+
+    /// `-Zcoverage-options=discard-all-spans-in-codegen`: During codgen,
+    /// discard all coverage spans as though they were invalid. Needed by
+    /// regression tests for #133606, because we don't have an easy way to
+    /// reproduce it from actual source code.
+    pub discard_all_spans_in_codegen: bool,
 }
 
 /// Controls whether branch coverage or MC/DC coverage is enabled.

compiler/rustc_session/src/options.rs

@@ -1037,6 +1037,7 @@ pub mod parse {
                 "condition" => slot.level = CoverageLevel::Condition,
                 "mcdc" => slot.level = CoverageLevel::Mcdc,
                 "no-mir-spans" => slot.no_mir_spans = true,
+                "discard-all-spans-in-codegen" => slot.discard_all_spans_in_codegen = true,
                 _ => return false,
             }
         }

compiler/rustc_session/src/session.rs

@@ -352,6 +352,11 @@ impl Session {
         self.opts.unstable_opts.coverage_options.no_mir_spans
     }
 
+    /// True if `-Zcoverage-options=discard-all-spans-in-codegen` was passed.
+    pub fn coverage_discard_all_spans_in_codegen(&self) -> bool {
+        self.opts.unstable_opts.coverage_options.discard_all_spans_in_codegen
+    }
+
     pub fn is_sanitizer_cfi_enabled(&self) -> bool {
         self.opts.unstable_opts.sanitizer.contains(SanitizerSet::CFI)
     }

tests/coverage/auxiliary/discard_all_helper.rs

@@ -0,0 +1,6 @@
+//@ edition: 2021
+
+// Force this function to be generated in its home crate, so that it ends up
+// with normal coverage metadata.
+#[inline(never)]
+pub fn external_function() {}

tests/coverage/discard-all-issue-133606.coverage

@@ -0,0 +1,7 @@
+   LL|       |//@ edition: 2021
+   LL|       |
+   LL|       |// Force this function to be generated in its home crate, so that it ends up
+   LL|       |// with normal coverage metadata.
+   LL|       |#[inline(never)]
+   LL|      1|pub fn external_function() {}
+

tests/coverage/discard-all-issue-133606.rs

@@ -0,0 +1,24 @@
+//! Regression test for <https://github.com/rust-lang/rust/issues/133606>.
+//!
+//! In rare cases, all of a function's coverage spans are discarded at a late
+//! stage during codegen. When that happens, the subsequent code needs to take
+//! special care to avoid emitting coverage metadata that would cause `llvm-cov`
+//! to fail with a fatal error.
+//!
+//! We currently don't know of a concise way to reproduce that scenario with
+//! ordinary Rust source code, so instead we set a special testing-only flag to
+//! force it to occur.
+
+//@ edition: 2021
+//@ compile-flags: -Zcoverage-options=discard-all-spans-in-codegen
+
+// The `llvm-cov` tool will complain if the test binary ends up having no
+// coverage metadata at all. To prevent that, we also link to instrumented
+// code in an auxiliary crate that doesn't have the special flag set.
+
+//@ aux-build: discard_all_helper.rs
+extern crate discard_all_helper;
+
+fn main() {
+    discard_all_helper::external_function();
+}

tests/mir-opt/coverage/branch_match_arms.main.InstrumentCoverage.diff

@@ -30,12 +30,12 @@
 +     coverage ExpressionId(0) => Expression { lhs: Counter(1), op: Add, rhs: Counter(2) };
 +     coverage ExpressionId(1) => Expression { lhs: Expression(0), op: Add, rhs: Counter(3) };
 +     coverage ExpressionId(2) => Expression { lhs: Counter(0), op: Subtract, rhs: Expression(1) };
-+     coverage Code(Counter(0)) => 13:1 - 14:21;
++     coverage Code(Counter(0)) => $DIR/branch_match_arms.rs:13:1: 14:21 (#0);
-+     coverage Code(Counter(1)) => 15:17 - 15:33;
++     coverage Code(Counter(1)) => $DIR/branch_match_arms.rs:15:17: 15:33 (#0);
-+     coverage Code(Counter(2)) => 16:17 - 16:33;
++     coverage Code(Counter(2)) => $DIR/branch_match_arms.rs:16:17: 16:33 (#0);
-+     coverage Code(Counter(3)) => 17:17 - 17:33;
++     coverage Code(Counter(3)) => $DIR/branch_match_arms.rs:17:17: 17:33 (#0);
-+     coverage Code(Expression(2)) => 18:17 - 18:33;
++     coverage Code(Expression(2)) => $DIR/branch_match_arms.rs:18:17: 18:33 (#0);
-+     coverage Code(Counter(0)) => 20:1 - 20:2;
++     coverage Code(Counter(0)) => $DIR/branch_match_arms.rs:20:2: 20:2 (#0);
 + 
       bb0: {
 +         Coverage::CounterIncrement(0);

tests/mir-opt/coverage/instrument_coverage.bar.InstrumentCoverage.diff

@@ -5,7 +5,7 @@
       let mut _0: bool;
   
 +     coverage body span: $DIR/instrument_coverage.rs:19:18: 21:2 (#0)
-+     coverage Code(Counter(0)) => 19:1 - 21:2;
++     coverage Code(Counter(0)) => $DIR/instrument_coverage.rs:19:1: 21:2 (#0);
 + 
       bb0: {
 +         Coverage::CounterIncrement(0);

tests/mir-opt/coverage/instrument_coverage.main.InstrumentCoverage.diff

@@ -9,11 +9,11 @@
   
 +     coverage body span: $DIR/instrument_coverage.rs:10:11: 16:2 (#0)
 +     coverage ExpressionId(0) => Expression { lhs: Counter(0), op: Add, rhs: Counter(1) };
-+     coverage Code(Counter(0)) => 10:1 - 10:11;
++     coverage Code(Counter(0)) => $DIR/instrument_coverage.rs:10:1: 10:11 (#0);
-+     coverage Code(Expression(0)) => 12:12 - 12:17;
++     coverage Code(Expression(0)) => $DIR/instrument_coverage.rs:12:12: 12:17 (#0);
-+     coverage Code(Counter(0)) => 13:13 - 13:18;
++     coverage Code(Counter(0)) => $DIR/instrument_coverage.rs:13:13: 13:18 (#0);
-+     coverage Code(Counter(1)) => 14:9 - 14:10;
++     coverage Code(Counter(1)) => $DIR/instrument_coverage.rs:14:10: 14:10 (#0);
-+     coverage Code(Counter(0)) => 16:1 - 16:2;
++     coverage Code(Counter(0)) => $DIR/instrument_coverage.rs:16:2: 16:2 (#0);
 + 
       bb0: {
 +         Coverage::CounterIncrement(0);

tests/mir-opt/coverage/instrument_coverage_cleanup.main.CleanupPostBorrowck.diff

@@ -9,11 +9,11 @@
   
       coverage body span: $DIR/instrument_coverage_cleanup.rs:13:11: 15:2 (#0)
       coverage ExpressionId(0) => Expression { lhs: Counter(0), op: Subtract, rhs: Counter(1) };
-      coverage Code(Counter(0)) => 13:1 - 14:36;
+      coverage Code(Counter(0)) => $DIR/instrument_coverage_cleanup.rs:13:1: 14:36 (#0);
-      coverage Code(Expression(0)) => 14:37 - 14:39;
+      coverage Code(Expression(0)) => $DIR/instrument_coverage_cleanup.rs:14:37: 14:39 (#0);
-      coverage Code(Counter(1)) => 14:38 - 14:39;
+      coverage Code(Counter(1)) => $DIR/instrument_coverage_cleanup.rs:14:39: 14:39 (#0);
-      coverage Code(Counter(0)) => 15:1 - 15:2;
+      coverage Code(Counter(0)) => $DIR/instrument_coverage_cleanup.rs:15:2: 15:2 (#0);
-      coverage Branch { true_term: Expression(0), false_term: Counter(1) } => 14:8 - 14:36;
+      coverage Branch { true_term: Expression(0), false_term: Counter(1) } => $DIR/instrument_coverage_cleanup.rs:14:8: 14:36 (#0);
   
       bb0: {
           Coverage::CounterIncrement(0);

tests/mir-opt/coverage/instrument_coverage_cleanup.main.InstrumentCoverage.diff

@@ -9,11 +9,11 @@
   
 +     coverage body span: $DIR/instrument_coverage_cleanup.rs:13:11: 15:2 (#0)
 +     coverage ExpressionId(0) => Expression { lhs: Counter(0), op: Subtract, rhs: Counter(1) };
-+     coverage Code(Counter(0)) => 13:1 - 14:36;
++     coverage Code(Counter(0)) => $DIR/instrument_coverage_cleanup.rs:13:1: 14:36 (#0);
-+     coverage Code(Expression(0)) => 14:37 - 14:39;
++     coverage Code(Expression(0)) => $DIR/instrument_coverage_cleanup.rs:14:37: 14:39 (#0);
-+     coverage Code(Counter(1)) => 14:38 - 14:39;
++     coverage Code(Counter(1)) => $DIR/instrument_coverage_cleanup.rs:14:39: 14:39 (#0);
-+     coverage Code(Counter(0)) => 15:1 - 15:2;
++     coverage Code(Counter(0)) => $DIR/instrument_coverage_cleanup.rs:15:2: 15:2 (#0);
-+     coverage Branch { true_term: Expression(0), false_term: Counter(1) } => 14:8 - 14:36;
++     coverage Branch { true_term: Expression(0), false_term: Counter(1) } => $DIR/instrument_coverage_cleanup.rs:14:8: 14:36 (#0);
 + 
       bb0: {
 +         Coverage::CounterIncrement(0);