Merge pull request #1575 from cfallin/test-fixes

MachInst backend: pass through SourceLoc information.

Cargo.lock

@@ -1538,12 +1538,13 @@ dependencies = [
 
 [[package]]
 name = "regalloc"
-version = "0.0.17"
+version = "0.0.18"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "89ce0cd835fa6e91bbf5d010beee19d0c2e97e4ad5e13c399a31122cfc83bdd6"
+checksum = "c6d7df180ed313488abdbeec2e3398f16cbf5ea6ce20db968c8cd6fc410400a9"
 dependencies = [
  "log",
  "rustc-hash",
+ "smallvec",
 ]
 
 [[package]]

cranelift/codegen/Cargo.toml

@@ -24,7 +24,7 @@ gimli = { version = "0.20.0", default-features = false, features = ["write"], op
 smallvec = { version = "1.0.0" }
 thiserror = "1.0.4"
 byteorder = { version = "1.3.2", default-features = false }
-regalloc = "0.0.17"
+regalloc = "0.0.18"
 # It is a goal of the cranelift-codegen crate to have minimal external dependencies.
 # Please don't add any unless they are essential to the task of creating binary
 # machine code. Integration tests that need external dependencies can be

cranelift/codegen/src/machinst/lower.rs

@@ -345,6 +345,7 @@ impl<'func, I: VCodeInst> Lower<'func, I> {
                             "lower_branch_group: targets = {:?} branches = {:?}",
                             targets, branches
                         );
+                        self.vcode.set_srcloc(self.srcloc(branches[0]));
                         backend.lower_branch_group(
                             &mut self,
                             &branches[..],
@@ -361,6 +362,7 @@ impl<'func, I: VCodeInst> Lower<'func, I> {
                     let num_uses = self.num_uses[inst];
                     let side_effect = has_side_effect(self.f, inst);
                     if side_effect || num_uses > 0 {
+                        self.vcode.set_srcloc(self.srcloc(inst));
                         backend.lower(&mut self, inst);
                         self.vcode.end_ir_inst();
                     } else {
@@ -389,6 +391,7 @@ impl<'func, I: VCodeInst> Lower<'func, I> {
                     "lower_branch_group: targets = {:?} branches = {:?}",
                     targets, branches
                 );
+                self.vcode.set_srcloc(self.srcloc(branches[0]));
                 backend.lower_branch_group(&mut self, &branches[..], &targets[..], fallthrough);
                 self.vcode.end_ir_inst();
                 branches.clear();

cranelift/codegen/src/machinst/mod.rs

@@ -97,7 +97,6 @@
 //! ```
 
 use crate::binemit::{CodeInfo, CodeOffset};
-use crate::entity::SecondaryMap;
 use crate::ir::condcodes::IntCC;
 use crate::ir::{Function, Type};
 use crate::result::CodegenResult;

cranelift/codegen/src/machinst/sections.rs

@@ -74,6 +74,11 @@ impl MachSections {
         sink.end_codegen();
     }
 
+    /// Get a list of source location mapping tuples in sorted-by-start-offset order.
+    pub fn get_srclocs_sorted<'a>(&'a self) -> MachSectionsSrcLocs<'a> {
+        MachSectionsSrcLocs::new(&self.sections)
+    }
+
     /// Get the total required size for these sections.
     pub fn total_size(&self) -> CodeOffset {
         if self.sections.len() == 0 {
@@ -90,6 +95,58 @@ impl MachSections {
     }
 }
 
+/// An iterator over the srclocs in each section.
+/// Returns MachSrcLocs in an order sorted by start location.
+pub struct MachSectionsSrcLocs<'a> {
+    sections: &'a [MachSection],
+    cur_section: usize,
+    cur_srcloc: usize,
+    // For validation:
+    last_offset: CodeOffset,
+}
+
+impl<'a> MachSectionsSrcLocs<'a> {
+    fn new(sections: &'a [MachSection]) -> MachSectionsSrcLocs<'a> {
+        MachSectionsSrcLocs {
+            sections,
+            cur_section: 0,
+            cur_srcloc: 0,
+            last_offset: 0,
+        }
+    }
+}
+
+impl<'a> Iterator for MachSectionsSrcLocs<'a> {
+    type Item = &'a MachSrcLoc;
+
+    fn next(&mut self) -> Option<&'a MachSrcLoc> {
+        // We simply iterate through sections and srcloc records in order. This produces a
+        // sorted order naturally because sections are in starting-offset-order, and srclocs
+        // are produced as a section is emitted into, so are in order as well.
+
+        // If we're out of sections, we're done.
+        if self.cur_section >= self.sections.len() {
+            return None;
+        }
+
+        // Otherwise, make sure we have a srcloc in the current section left to return, and
+        // advance to the next section if not. Done if we run out of sections.
+        while self.cur_srcloc >= self.sections[self.cur_section].srclocs.len() {
+            self.cur_srcloc = 0;
+            self.cur_section += 1;
+            if self.cur_section >= self.sections.len() {
+                return None;
+            }
+        }
+
+        let loc = &self.sections[self.cur_section].srclocs[self.cur_srcloc];
+        self.cur_srcloc += 1;
+        debug_assert!(loc.start >= self.last_offset);
+        self.last_offset = loc.start;
+        Some(loc)
+    }
+}
+
 /// An abstraction over MachSection and MachSectionSize: some
 /// receiver of section data.
 pub trait MachSectionOutput {
@@ -143,6 +200,12 @@ pub trait MachSectionOutput {
     /// Add a call return address record at the current offset.
     fn add_call_site(&mut self, loc: SourceLoc, opcode: Opcode);
 
+    /// Start the output for the given source-location at the current offset.
+    fn start_srcloc(&mut self, loc: SourceLoc);
+
+    /// End the output for the previously-given source-location at the current offset.
+    fn end_srcloc(&mut self);
+
     /// Align up to the given alignment.
     fn align_to(&mut self, align_to: CodeOffset) {
         assert!(align_to.is_power_of_two());
@@ -168,8 +231,13 @@ pub struct MachSection {
     pub relocs: Vec<MachReloc>,
     /// Any trap records referring to this section.
     pub traps: Vec<MachTrap>,
-    /// Any call site record referring to this section.
+    /// Any call site records referring to this section.
     pub call_sites: Vec<MachCallSite>,
+    /// Any source location mappings referring to this section.
+    pub srclocs: Vec<MachSrcLoc>,
+    /// The current source location in progress (after `start_srcloc()` and before `end_srcloc()`).
+    /// This is a (start_offset, src_loc) tuple.
+    pub cur_srcloc: Option<(CodeOffset, SourceLoc)>,
 }
 
 impl MachSection {
@@ -182,6 +250,8 @@ impl MachSection {
             relocs: vec![],
             traps: vec![],
             call_sites: vec![],
+            srclocs: vec![],
+            cur_srcloc: None,
         }
     }
 
@@ -266,6 +336,23 @@ impl MachSectionOutput for MachSection {
             opcode,
         });
     }
+
+    fn start_srcloc(&mut self, loc: SourceLoc) {
+        self.cur_srcloc = Some((self.cur_offset_from_start(), loc));
+    }
+
+    fn end_srcloc(&mut self) {
+        let (start, loc) = self
+            .cur_srcloc
+            .take()
+            .expect("end_srcloc() called without start_srcloc()");
+        let end = self.cur_offset_from_start();
+        // Skip zero-length extends.
+        debug_assert!(end >= start);
+        if end > start {
+            self.srclocs.push(MachSrcLoc { start, end, loc });
+        }
+    }
 }
 
 /// A MachSectionOutput implementation that records only size.
@@ -315,6 +402,10 @@ impl MachSectionOutput for MachSectionSize {
     fn add_trap(&mut self, _: SourceLoc, _: TrapCode) {}
 
     fn add_call_site(&mut self, _: SourceLoc, _: Opcode) {}
+
+    fn start_srcloc(&mut self, _: SourceLoc) {}
+
+    fn end_srcloc(&mut self) {}
 }
 
 /// A relocation resulting from a compilation.
@@ -352,3 +443,18 @@ pub struct MachCallSite {
     /// The call's opcode.
     pub opcode: Opcode,
 }
+
+/// A source-location mapping resulting from a compilation.
+#[derive(Clone, Debug)]
+pub struct MachSrcLoc {
+    /// The start of the region of code corresponding to a source location.
+    /// This is relative to the start of the function, not to the start of the
+    /// section.
+    pub start: CodeOffset,
+    /// The end of the region of code corresponding to a source location.
+    /// This is relative to the start of the section, not to the start of the
+    /// section.
+    pub end: CodeOffset,
+    /// The source location.
+    pub loc: SourceLoc,
+}

cranelift/codegen/src/machinst/vcode.rs

@@ -17,7 +17,9 @@
 //! See the main module comment in `mod.rs` for more details on the VCode-based
 //! backend pipeline.
 
+use crate::entity::SecondaryMap;
 use crate::ir;
+use crate::ir::SourceLoc;
 use crate::machinst::*;
 use crate::settings;
 
@@ -59,6 +61,10 @@ pub struct VCode<I: VCodeInst> {
     /// Lowered machine instructions in order corresponding to the original IR.
     insts: Vec<I>,
 
+    /// Source locations for each instruction. (`SourceLoc` is a `u32`, so it is
+    /// reasonable to keep one of these per instruction.)
+    srclocs: Vec<SourceLoc>,
+
     /// Entry block.
     entry: BlockIndex,
 
@@ -115,13 +121,16 @@ pub struct VCodeBuilder<I: VCodeInst> {
 
     /// Current basic block instructions, in reverse order (because blocks are
     /// built bottom-to-top).
-    bb_insns: SmallVec<[I; 32]>,
+    bb_insns: SmallVec<[(I, SourceLoc); 32]>,
 
     /// Current IR-inst instructions, in forward order.
-    ir_inst_insns: SmallVec<[I; 4]>,
+    ir_inst_insns: SmallVec<[(I, SourceLoc); 4]>,
 
     /// Start of succs for the current block in the concatenated succs list.
     succ_start: usize,
+
+    /// Current source location.
+    cur_srcloc: SourceLoc,
 }
 
 impl<I: VCodeInst> VCodeBuilder<I> {
@@ -133,6 +142,7 @@ impl<I: VCodeInst> VCodeBuilder<I> {
             bb_insns: SmallVec::new(),
             ir_inst_insns: SmallVec::new(),
             succ_start: 0,
+            cur_srcloc: SourceLoc::default(),
         }
     }
 
@@ -179,8 +189,8 @@ impl<I: VCodeInst> VCodeBuilder<I> {
     /// End the current IR instruction. Must be called after pushing any
     /// instructions and prior to ending the basic block.
     pub fn end_ir_inst(&mut self) {
-        while let Some(i) = self.ir_inst_insns.pop() {
+        while let Some(pair) = self.ir_inst_insns.pop() {
-            self.bb_insns.push(i);
+            self.bb_insns.push(pair);
         }
     }
 
@@ -191,8 +201,9 @@ impl<I: VCodeInst> VCodeBuilder<I> {
         let block_num = self.vcode.block_ranges.len() as BlockIndex;
         // Push the instructions.
         let start_idx = self.vcode.insts.len() as InsnIndex;
-        while let Some(i) = self.bb_insns.pop() {
+        while let Some((i, loc)) = self.bb_insns.pop() {
             self.vcode.insts.push(i);
+            self.vcode.srclocs.push(loc);
         }
         let end_idx = self.vcode.insts.len() as InsnIndex;
         // Add the instruction index range to the list of blocks.
@@ -224,7 +235,12 @@ impl<I: VCodeInst> VCodeBuilder<I> {
                 }
             }
         }
-        self.ir_inst_insns.push(insn);
+        self.ir_inst_insns.push((insn, self.cur_srcloc));
+    }
+
+    /// Set the current source location.
+    pub fn set_srcloc(&mut self, srcloc: SourceLoc) {
+        self.cur_srcloc = srcloc;
     }
 
     /// Build the final VCode.
@@ -286,6 +302,7 @@ impl<I: VCodeInst> VCode<I> {
             liveouts: abi.liveouts(),
             vreg_types: vec![],
             insts: vec![],
+            srclocs: vec![],
             entry: 0,
             block_ranges: vec![],
             block_succ_range: vec![],
@@ -349,6 +366,7 @@ impl<I: VCodeInst> VCode<I> {
             block_ranges(result.target_map.elems(), result.insns.len());
         let mut final_insns = vec![];
         let mut final_block_ranges = vec![(0, 0); self.num_blocks()];
+        let mut final_srclocs = vec![];
 
         for block in &self.final_block_order {
             let (start, end) = block_ranges[*block as usize];
@@ -356,7 +374,10 @@ impl<I: VCodeInst> VCode<I> {
 
             if *block == self.entry {
                 // Start with the prologue.
-                final_insns.extend(self.abi.gen_prologue().into_iter());
+                let prologue = self.abi.gen_prologue();
+                let len = prologue.len();
+                final_insns.extend(prologue.into_iter());
+                final_srclocs.extend(iter::repeat(SourceLoc::default()).take(len));
             }
 
             for i in start..end {
@@ -368,13 +389,27 @@ impl<I: VCodeInst> VCode<I> {
                     continue;
                 }
 
+                // Is there a srcloc associated with this insn? Look it up based on original
+                // instruction index (if new insn corresponds to some original insn, i.e., is not
+                // an inserted load/spill/move).
+                let orig_iix = result.orig_insn_map[InstIx::new(i as u32)];
+                let srcloc = if orig_iix.is_invalid() {
+                    SourceLoc::default()
+                } else {
+                    self.srclocs[orig_iix.get() as usize]
+                };
+
                 // Whenever encountering a return instruction, replace it
                 // with the epilogue.
                 let is_ret = insn.is_term() == MachTerminator::Ret;
                 if is_ret {
-                    final_insns.extend(self.abi.gen_epilogue().into_iter());
+                    let epilogue = self.abi.gen_epilogue();
+                    let len = epilogue.len();
+                    final_insns.extend(epilogue.into_iter());
+                    final_srclocs.extend(iter::repeat(srcloc).take(len));
                 } else {
                     final_insns.push(insn.clone());
+                    final_srclocs.push(srcloc);
                 }
             }
 
@@ -382,7 +417,10 @@ impl<I: VCodeInst> VCode<I> {
             final_block_ranges[*block as usize] = (final_start, final_end);
         }
 
+        debug_assert!(final_insns.len() == final_srclocs.len());
+
         self.insts = final_insns;
+        self.srclocs = final_srclocs;
         self.block_ranges = final_block_ranges;
     }
 
@@ -512,6 +550,7 @@ impl<I: VCodeInst> VCode<I> {
         let code_section = sections.get_section(code_idx);
 
         let flags = self.abi.flags();
+        let mut cur_srcloc = SourceLoc::default();
         for &block in &self.final_block_order {
             let new_offset = I::align_basic_block(code_section.cur_offset_from_start());
             while new_offset > code_section.cur_offset_from_start() {
@@ -523,8 +562,24 @@ impl<I: VCodeInst> VCode<I> {
 
             let (start, end) = self.block_ranges[block as usize];
             for iix in start..end {
+                let srcloc = self.srclocs[iix as usize];
+                if srcloc != cur_srcloc {
+                    if !cur_srcloc.is_default() {
+                        code_section.end_srcloc();
+                    }
+                    if !srcloc.is_default() {
+                        code_section.start_srcloc(srcloc);
+                    }
+                    cur_srcloc = srcloc;
+                }
+
                 self.insts[iix as usize].emit(code_section, flags);
             }
+
+            if !cur_srcloc.is_default() {
+                code_section.end_srcloc();
+                cur_srcloc = SourceLoc::default();
+            }
         }
 
         sections

crates/environ/src/cranelift.rs

@@ -93,6 +93,7 @@ use crate::compilation::{
 use crate::func_environ::{get_func_name, FuncEnvironment};
 use crate::{CacheConfig, FunctionBodyData, ModuleLocal, ModuleTranslation, Tunables};
 use cranelift_codegen::ir::{self, ExternalName};
+use cranelift_codegen::machinst::sections::MachSrcLoc;
 use cranelift_codegen::print_errors::pretty_error;
 use cranelift_codegen::{binemit, isa, Context};
 use cranelift_entity::PrimaryMap;
@@ -207,13 +208,22 @@ fn get_function_address_map<'data>(
 ) -> FunctionAddressMap {
     let mut instructions = Vec::new();
 
+    if let Some(ref mcr) = &context.mach_compile_result {
+        // New-style backend: we have a `MachCompileResult` that will give us `MachSrcLoc` mapping
+        // tuples.
+        for &MachSrcLoc { start, end, loc } in mcr.sections.get_srclocs_sorted() {
+            instructions.push(InstructionAddressMap {
+                srcloc: loc,
+                code_offset: start as usize,
+                code_len: (end - start) as usize,
+            });
+        }
+    } else {
+        // Old-style backend: we need to traverse the instruction/encoding info in the function.
         let func = &context.func;
         let mut blocks = func.layout.blocks().collect::<Vec<_>>();
         blocks.sort_by_key(|block| func.offsets[*block]); // Ensure inst offsets always increase
 
-    // FIXME(#1523): New backend does not support debug info or instruction-address mapping
-    // yet.
-    if !isa.get_mach_backend().is_some() {
         let encinfo = isa.encoding_info();
         for block in blocks {
             for (offset, inst, size) in func.inst_offsets(block, &encinfo) {