// Copyright 2016 The Rust Project Developers. See the COPYRIGHT // file at the top-level directory of this distribution and at // http://rust-lang.org/COPYRIGHT. // // Licensed under the Apache License, Version 2.0 or the MIT license // , at your // option. This file may not be copied, modified, or distributed // except according to those terms. //! Inlining pass for MIR functions use rustc::hir; use rustc::hir::def_id::DefId; use rustc_data_structures::bitvec::BitVector; use rustc_data_structures::indexed_vec::{Idx, IndexVec}; use rustc::mir::*; use rustc::mir::visit::*; use rustc::ty::{self, Instance, Ty, TyCtxt, TypeFoldable}; use rustc::ty::layout::LayoutOf; use rustc::ty::subst::{Subst,Substs}; use std::collections::VecDeque; use std::iter; use transform::{MirPass, MirSource}; use super::simplify::{remove_dead_blocks, CfgSimplifier}; use syntax::{attr}; use syntax::abi::Abi; const DEFAULT_THRESHOLD: usize = 50; const HINT_THRESHOLD: usize = 100; const INSTR_COST: usize = 5; const CALL_PENALTY: usize = 25; const UNKNOWN_SIZE_COST: usize = 10; pub struct Inline; #[derive(Copy, Clone, Debug)] struct CallSite<'tcx> { callee: DefId, substs: &'tcx Substs<'tcx>, bb: BasicBlock, location: SourceInfo, } impl MirPass for Inline { fn run_pass<'a, 'tcx>(&self, tcx: TyCtxt<'a, 'tcx, 'tcx>, source: MirSource, mir: &mut Mir<'tcx>) { if tcx.sess.opts.debugging_opts.mir_opt_level >= 2 { Inliner { tcx, source }.run_pass(mir); } } } struct Inliner<'a, 'tcx: 'a> { tcx: TyCtxt<'a, 'tcx, 'tcx>, source: MirSource, } impl<'a, 'tcx> Inliner<'a, 'tcx> { fn run_pass(&self, caller_mir: &mut Mir<'tcx>) { // Keep a queue of callsites to try inlining on. We take // advantage of the fact that queries detect cycles here to // allow us to try and fetch the fully optimized MIR of a // call; if it succeeds, we can inline it and we know that // they do not call us. Otherwise, we just don't try to // inline. // // We use a queue so that we inline "broadly" before we inline // in depth. It is unclear if this is the best heuristic, // really, but that's true of all the heuristics in this // file. =) let mut callsites = VecDeque::new(); let param_env = self.tcx.param_env(self.source.def_id); // Only do inlining into fn bodies. let id = self.tcx.hir.as_local_node_id(self.source.def_id).unwrap(); let body_owner_kind = self.tcx.hir.body_owner_kind(id); if let (hir::BodyOwnerKind::Fn, None) = (body_owner_kind, self.source.promoted) { for (bb, bb_data) in caller_mir.basic_blocks().iter_enumerated() { // Don't inline calls that are in cleanup blocks. if bb_data.is_cleanup { continue; } // Only consider direct calls to functions let terminator = bb_data.terminator(); if let TerminatorKind::Call { func: Operand::Constant(ref f), .. } = terminator.kind { if let ty::TyFnDef(callee_def_id, substs) = f.ty.sty { if let Some(instance) = Instance::resolve(self.tcx, param_env, callee_def_id, substs) { callsites.push_back(CallSite { callee: instance.def_id(), substs: instance.substs, bb, location: terminator.source_info }); } } } } } else { return; } let mut local_change; let mut changed = false; loop { local_change = false; while let Some(callsite) = callsites.pop_front() { debug!("checking whether to inline callsite {:?}", callsite); if !self.tcx.is_mir_available(callsite.callee) { debug!("checking whether to inline callsite {:?} - MIR unavailable", callsite); continue; } let callee_mir = match ty::queries::optimized_mir::try_get(self.tcx, callsite.location.span, callsite.callee) { Ok(ref callee_mir) if self.should_inline(callsite, callee_mir) => { subst_and_normalize(callee_mir, self.tcx, &callsite.substs, param_env) } Ok(_) => continue, Err(mut bug) => { // FIXME(#43542) shouldn't have to cancel an error bug.cancel(); continue } }; let start = caller_mir.basic_blocks().len(); debug!("attempting to inline callsite {:?} - mir={:?}", callsite, callee_mir); if !self.inline_call(callsite, caller_mir, callee_mir) { debug!("attempting to inline callsite {:?} - failure", callsite); continue; } debug!("attempting to inline callsite {:?} - success", callsite); // Add callsites from inlined function for (bb, bb_data) in caller_mir.basic_blocks().iter_enumerated().skip(start) { // Only consider direct calls to functions let terminator = bb_data.terminator(); if let TerminatorKind::Call { func: Operand::Constant(ref f), .. } = terminator.kind { if let ty::TyFnDef(callee_def_id, substs) = f.ty.sty { // Don't inline the same function multiple times. if callsite.callee != callee_def_id { callsites.push_back(CallSite { callee: callee_def_id, substs, bb, location: terminator.source_info }); } } } } local_change = true; changed = true; } if !local_change { break; } } // Simplify if we inlined anything. if changed { debug!("Running simplify cfg on {:?}", self.source); CfgSimplifier::new(caller_mir).simplify(); remove_dead_blocks(caller_mir); } } fn should_inline(&self, callsite: CallSite<'tcx>, callee_mir: &Mir<'tcx>) -> bool { debug!("should_inline({:?})", callsite); let tcx = self.tcx; // Don't inline closures that have captures // FIXME: Handle closures better if callee_mir.upvar_decls.len() > 0 { debug!(" upvar decls present - not inlining"); return false; } // Cannot inline generators which haven't been transformed yet if callee_mir.yield_ty.is_some() { debug!(" yield ty present - not inlining"); return false; } let attrs = tcx.get_attrs(callsite.callee); let hint = attr::find_inline_attr(None, &attrs[..]); let hinted = match hint { // Just treat inline(always) as a hint for now, // there are cases that prevent inlining that we // need to check for first. attr::InlineAttr::Always => true, attr::InlineAttr::Never => { debug!("#[inline(never)] present - not inlining"); return false } attr::InlineAttr::Hint => true, attr::InlineAttr::None => false, }; // Only inline local functions if they would be eligible for cross-crate // inlining. This is to ensure that the final crate doesn't have MIR that // reference unexported symbols if callsite.callee.is_local() { if callsite.substs.types().count() == 0 && !hinted { debug!(" callee is an exported function - not inlining"); return false; } } let mut threshold = if hinted { HINT_THRESHOLD } else { DEFAULT_THRESHOLD }; // Significantly lower the threshold for inlining cold functions if attr::contains_name(&attrs[..], "cold") { threshold /= 5; } // Give a bonus functions with a small number of blocks, // We normally have two or three blocks for even // very small functions. if callee_mir.basic_blocks().len() <= 3 { threshold += threshold / 4; } debug!(" final inline threshold = {}", threshold); // FIXME: Give a bonus to functions with only a single caller let param_env = tcx.param_env(self.source.def_id); let mut first_block = true; let mut cost = 0; // Traverse the MIR manually so we can account for the effects of // inlining on the CFG. let mut work_list = vec![START_BLOCK]; let mut visited = BitVector::new(callee_mir.basic_blocks().len()); while let Some(bb) = work_list.pop() { if !visited.insert(bb.index()) { continue; } let blk = &callee_mir.basic_blocks()[bb]; for stmt in &blk.statements { // Don't count StorageLive/StorageDead in the inlining cost. match stmt.kind { StatementKind::StorageLive(_) | StatementKind::StorageDead(_) | StatementKind::Nop => {} _ => cost += INSTR_COST } } let term = blk.terminator(); let mut is_drop = false; match term.kind { TerminatorKind::Drop { ref location, target, unwind } | TerminatorKind::DropAndReplace { ref location, target, unwind, .. } => { is_drop = true; work_list.push(target); // If the location doesn't actually need dropping, treat it like // a regular goto. let ty = location.ty(callee_mir, tcx).subst(tcx, callsite.substs); let ty = ty.to_ty(tcx); if ty.needs_drop(tcx, param_env) { cost += CALL_PENALTY; if let Some(unwind) = unwind { work_list.push(unwind); } } else { cost += INSTR_COST; } } TerminatorKind::Unreachable | TerminatorKind::Call { destination: None, .. } if first_block => { // If the function always diverges, don't inline // unless the cost is zero threshold = 0; } TerminatorKind::Call {func: Operand::Constant(ref f), .. } => { if let ty::TyFnDef(def_id, _) = f.ty.sty { // Don't give intrinsics the extra penalty for calls let f = tcx.fn_sig(def_id); if f.abi() == Abi::RustIntrinsic || f.abi() == Abi::PlatformIntrinsic { cost += INSTR_COST; } else { cost += CALL_PENALTY; } } } TerminatorKind::Assert { .. } => cost += CALL_PENALTY, _ => cost += INSTR_COST } if !is_drop { for &succ in &term.successors()[..] { work_list.push(succ); } } first_block = false; } // Count up the cost of local variables and temps, if we know the size // use that, otherwise we use a moderately-large dummy cost. let ptr_size = tcx.data_layout.pointer_size.bytes(); for v in callee_mir.vars_and_temps_iter() { let v = &callee_mir.local_decls[v]; let ty = v.ty.subst(tcx, callsite.substs); // Cost of the var is the size in machine-words, if we know // it. if let Some(size) = type_size_of(tcx, param_env.clone(), ty) { cost += (size / ptr_size) as usize; } else { cost += UNKNOWN_SIZE_COST; } } if let attr::InlineAttr::Always = hint { debug!("INLINING {:?} because inline(always) [cost={}]", callsite, cost); true } else { if cost <= threshold { debug!("INLINING {:?} [cost={} <= threshold={}]", callsite, cost, threshold); true } else { debug!("NOT inlining {:?} [cost={} > threshold={}]", callsite, cost, threshold); false } } } fn inline_call(&self, callsite: CallSite<'tcx>, caller_mir: &mut Mir<'tcx>, mut callee_mir: Mir<'tcx>) -> bool { let terminator = caller_mir[callsite.bb].terminator.take().unwrap(); match terminator.kind { // FIXME: Handle inlining of diverging calls TerminatorKind::Call { args, destination: Some(destination), cleanup, .. } => { debug!("Inlined {:?} into {:?}", callsite.callee, self.source); let is_box_free = Some(callsite.callee) == self.tcx.lang_items().box_free_fn(); let mut local_map = IndexVec::with_capacity(callee_mir.local_decls.len()); let mut scope_map = IndexVec::with_capacity(callee_mir.visibility_scopes.len()); let mut promoted_map = IndexVec::with_capacity(callee_mir.promoted.len()); for mut scope in callee_mir.visibility_scopes.iter().cloned() { if scope.parent_scope.is_none() { scope.parent_scope = Some(callsite.location.scope); scope.span = callee_mir.span; } scope.span = callsite.location.span; let idx = caller_mir.visibility_scopes.push(scope); scope_map.push(idx); } for loc in callee_mir.vars_and_temps_iter() { let mut local = callee_mir.local_decls[loc].clone(); local.source_info.scope = scope_map[local.source_info.scope]; local.source_info.span = callsite.location.span; let idx = caller_mir.local_decls.push(local); local_map.push(idx); } for p in callee_mir.promoted.iter().cloned() { let idx = caller_mir.promoted.push(p); promoted_map.push(idx); } // If the call is something like `a[*i] = f(i)`, where // `i : &mut usize`, then just duplicating the `a[*i]` // Lvalue could result in two different locations if `f` // writes to `i`. To prevent this we need to create a temporary // borrow of the lvalue and pass the destination as `*temp` instead. fn dest_needs_borrow(lval: &Lvalue) -> bool { match *lval { Lvalue::Projection(ref p) => { match p.elem { ProjectionElem::Deref | ProjectionElem::Index(_) => true, _ => dest_needs_borrow(&p.base) } } // Static variables need a borrow because the callee // might modify the same static. Lvalue::Static(_) => true, _ => false } } let dest = if dest_needs_borrow(&destination.0) { debug!("Creating temp for return destination"); let dest = Rvalue::Ref( self.tcx.types.re_erased, BorrowKind::Mut, destination.0); let ty = dest.ty(caller_mir, self.tcx); let temp = LocalDecl::new_temp(ty, callsite.location.span); let tmp = caller_mir.local_decls.push(temp); let tmp = Lvalue::Local(tmp); let stmt = Statement { source_info: callsite.location, kind: StatementKind::Assign(tmp.clone(), dest) }; caller_mir[callsite.bb] .statements.push(stmt); tmp.deref() } else { destination.0 }; let return_block = destination.1; let args : Vec<_> = if is_box_free { assert!(args.len() == 1); // box_free takes a Box, but is defined with a *mut T, inlining // needs to generate the cast. // FIXME: we should probably just generate correct MIR in the first place... let arg = if let Operand::Move(ref lval) = args[0] { lval.clone() } else { bug!("Constant arg to \"box_free\""); }; let ptr_ty = args[0].ty(caller_mir, self.tcx); vec![self.cast_box_free_arg(arg, ptr_ty, &callsite, caller_mir)] } else { // Copy the arguments if needed. self.make_call_args(args, &callsite, caller_mir) }; let bb_len = caller_mir.basic_blocks().len(); let mut integrator = Integrator { block_idx: bb_len, args: &args, local_map, scope_map, promoted_map, _callsite: callsite, destination: dest, return_block, cleanup_block: cleanup, in_cleanup_block: false }; for (bb, mut block) in callee_mir.basic_blocks_mut().drain_enumerated(..) { integrator.visit_basic_block_data(bb, &mut block); caller_mir.basic_blocks_mut().push(block); } let terminator = Terminator { source_info: callsite.location, kind: TerminatorKind::Goto { target: BasicBlock::new(bb_len) } }; caller_mir[callsite.bb].terminator = Some(terminator); true } kind => { caller_mir[callsite.bb].terminator = Some(Terminator { source_info: terminator.source_info, kind, }); false } } } fn cast_box_free_arg(&self, arg: Lvalue<'tcx>, ptr_ty: Ty<'tcx>, callsite: &CallSite<'tcx>, caller_mir: &mut Mir<'tcx>) -> Local { let arg = Rvalue::Ref( self.tcx.types.re_erased, BorrowKind::Mut, arg.deref()); let ty = arg.ty(caller_mir, self.tcx); let ref_tmp = LocalDecl::new_temp(ty, callsite.location.span); let ref_tmp = caller_mir.local_decls.push(ref_tmp); let ref_tmp = Lvalue::Local(ref_tmp); let ref_stmt = Statement { source_info: callsite.location, kind: StatementKind::Assign(ref_tmp.clone(), arg) }; caller_mir[callsite.bb] .statements.push(ref_stmt); let pointee_ty = match ptr_ty.sty { ty::TyRawPtr(tm) | ty::TyRef(_, tm) => tm.ty, _ if ptr_ty.is_box() => ptr_ty.boxed_ty(), _ => bug!("Invalid type `{:?}` for call to box_free", ptr_ty) }; let ptr_ty = self.tcx.mk_mut_ptr(pointee_ty); let raw_ptr = Rvalue::Cast(CastKind::Misc, Operand::Move(ref_tmp), ptr_ty); let cast_tmp = LocalDecl::new_temp(ptr_ty, callsite.location.span); let cast_tmp = caller_mir.local_decls.push(cast_tmp); let cast_stmt = Statement { source_info: callsite.location, kind: StatementKind::Assign(Lvalue::Local(cast_tmp), raw_ptr) }; caller_mir[callsite.bb] .statements.push(cast_stmt); cast_tmp } fn make_call_args( &self, args: Vec>, callsite: &CallSite<'tcx>, caller_mir: &mut Mir<'tcx>, ) -> Vec { let tcx = self.tcx; // There is a bit of a mismatch between the *caller* of a closure and the *callee*. // The caller provides the arguments wrapped up in a tuple: // // tuple_tmp = (a, b, c) // Fn::call(closure_ref, tuple_tmp) // // meanwhile the closure body expects the arguments (here, `a`, `b`, and `c`) // as distinct arguments. (This is the "rust-call" ABI hack.) Normally, trans has // the job of unpacking this tuple. But here, we are trans. =) So we want to create // a vector like // // [closure_ref, tuple_tmp.0, tuple_tmp.1, tuple_tmp.2] // // Except for one tiny wrinkle: we don't actually want `tuple_tmp.0`. It's more convenient // if we "spill" that into *another* temporary, so that we can map the argument // variable in the callee MIR directly to an argument variable on our side. // So we introduce temporaries like: // // tmp0 = tuple_tmp.0 // tmp1 = tuple_tmp.1 // tmp2 = tuple_tmp.2 // // and the vector is `[closure_ref, tmp0, tmp1, tmp2]`. if tcx.is_closure(callsite.callee) { let mut args = args.into_iter(); let self_ = self.create_temp_if_necessary(args.next().unwrap(), callsite, caller_mir); let tuple = self.create_temp_if_necessary(args.next().unwrap(), callsite, caller_mir); assert!(args.next().is_none()); let tuple = Lvalue::Local(tuple); let tuple_tys = if let ty::TyTuple(s, _) = tuple.ty(caller_mir, tcx).to_ty(tcx).sty { s } else { bug!("Closure arguments are not passed as a tuple"); }; // The `closure_ref` in our example above. let closure_ref_arg = iter::once(self_); // The `tmp0`, `tmp1`, and `tmp2` in our example abonve. let tuple_tmp_args = tuple_tys.iter().enumerate().map(|(i, ty)| { // This is e.g. `tuple_tmp.0` in our example above. let tuple_field = Operand::Move(tuple.clone().field(Field::new(i), ty)); // Spill to a local to make e.g. `tmp0`. self.create_temp_if_necessary(tuple_field, callsite, caller_mir) }); closure_ref_arg.chain(tuple_tmp_args).collect() } else { args.into_iter() .map(|a| self.create_temp_if_necessary(a, callsite, caller_mir)) .collect() } } /// If `arg` is already a temporary, returns it. Otherwise, introduces a fresh /// temporary `T` and an instruction `T = arg`, and returns `T`. fn create_temp_if_necessary( &self, arg: Operand<'tcx>, callsite: &CallSite<'tcx>, caller_mir: &mut Mir<'tcx>, ) -> Local { // FIXME: Analysis of the usage of the arguments to avoid // unnecessary temporaries. if let Operand::Move(Lvalue::Local(local)) = arg { if caller_mir.local_kind(local) == LocalKind::Temp { // Reuse the operand if it's a temporary already return local; } } debug!("Creating temp for argument {:?}", arg); // Otherwise, create a temporary for the arg let arg = Rvalue::Use(arg); let ty = arg.ty(caller_mir, self.tcx); let arg_tmp = LocalDecl::new_temp(ty, callsite.location.span); let arg_tmp = caller_mir.local_decls.push(arg_tmp); let stmt = Statement { source_info: callsite.location, kind: StatementKind::Assign(Lvalue::Local(arg_tmp), arg), }; caller_mir[callsite.bb].statements.push(stmt); arg_tmp } } fn type_size_of<'a, 'tcx>(tcx: TyCtxt<'a, 'tcx, 'tcx>, param_env: ty::ParamEnv<'tcx>, ty: Ty<'tcx>) -> Option { (tcx, param_env).layout_of(ty).ok().map(|layout| layout.size.bytes()) } fn subst_and_normalize<'a, 'tcx: 'a>( mir: &Mir<'tcx>, tcx: TyCtxt<'a, 'tcx, 'tcx>, substs: &'tcx ty::subst::Substs<'tcx>, param_env: ty::ParamEnv<'tcx>, ) -> Mir<'tcx> { struct Folder<'a, 'tcx: 'a> { tcx: TyCtxt<'a, 'tcx, 'tcx>, param_env: ty::ParamEnv<'tcx>, substs: &'tcx ty::subst::Substs<'tcx>, } impl<'a, 'tcx: 'a> ty::fold::TypeFolder<'tcx, 'tcx> for Folder<'a, 'tcx> { fn tcx<'b>(&'b self) -> TyCtxt<'b, 'tcx, 'tcx> { self.tcx } fn fold_ty(&mut self, t: Ty<'tcx>) -> Ty<'tcx> { self.tcx.trans_apply_param_substs_env(&self.substs, self.param_env, &t) } } let mut f = Folder { tcx, param_env, substs }; mir.fold_with(&mut f) } /** * Integrator. * * Integrates blocks from the callee function into the calling function. * Updates block indices, references to locals and other control flow * stuff. */ struct Integrator<'a, 'tcx: 'a> { block_idx: usize, args: &'a [Local], local_map: IndexVec, scope_map: IndexVec, promoted_map: IndexVec, _callsite: CallSite<'tcx>, destination: Lvalue<'tcx>, return_block: BasicBlock, cleanup_block: Option, in_cleanup_block: bool, } impl<'a, 'tcx> Integrator<'a, 'tcx> { fn update_target(&self, tgt: BasicBlock) -> BasicBlock { let new = BasicBlock::new(tgt.index() + self.block_idx); debug!("Updating target `{:?}`, new: `{:?}`", tgt, new); new } } impl<'a, 'tcx> MutVisitor<'tcx> for Integrator<'a, 'tcx> { fn visit_local(&mut self, local: &mut Local, _ctxt: LvalueContext<'tcx>, _location: Location) { if *local == RETURN_POINTER { match self.destination { Lvalue::Local(l) => { *local = l; return; }, ref lval => bug!("Return lvalue is {:?}, not local", lval) } } let idx = local.index() - 1; if idx < self.args.len() { *local = self.args[idx]; return; } *local = self.local_map[Local::new(idx - self.args.len())]; } fn visit_lvalue(&mut self, lvalue: &mut Lvalue<'tcx>, _ctxt: LvalueContext<'tcx>, _location: Location) { if let Lvalue::Local(RETURN_POINTER) = *lvalue { // Return pointer; update the lvalue itself *lvalue = self.destination.clone(); } else { self.super_lvalue(lvalue, _ctxt, _location); } } fn visit_basic_block_data(&mut self, block: BasicBlock, data: &mut BasicBlockData<'tcx>) { self.in_cleanup_block = data.is_cleanup; self.super_basic_block_data(block, data); self.in_cleanup_block = false; } fn visit_terminator_kind(&mut self, block: BasicBlock, kind: &mut TerminatorKind<'tcx>, loc: Location) { self.super_terminator_kind(block, kind, loc); match *kind { TerminatorKind::GeneratorDrop | TerminatorKind::Yield { .. } => bug!(), TerminatorKind::Goto { ref mut target} => { *target = self.update_target(*target); } TerminatorKind::SwitchInt { ref mut targets, .. } => { for tgt in targets { *tgt = self.update_target(*tgt); } } TerminatorKind::Drop { ref mut target, ref mut unwind, .. } | TerminatorKind::DropAndReplace { ref mut target, ref mut unwind, .. } => { *target = self.update_target(*target); if let Some(tgt) = *unwind { *unwind = Some(self.update_target(tgt)); } else if !self.in_cleanup_block { // Unless this drop is in a cleanup block, add an unwind edge to // the orignal call's cleanup block *unwind = self.cleanup_block; } } TerminatorKind::Call { ref mut destination, ref mut cleanup, .. } => { if let Some((_, ref mut tgt)) = *destination { *tgt = self.update_target(*tgt); } if let Some(tgt) = *cleanup { *cleanup = Some(self.update_target(tgt)); } else if !self.in_cleanup_block { // Unless this call is in a cleanup block, add an unwind edge to // the orignal call's cleanup block *cleanup = self.cleanup_block; } } TerminatorKind::Assert { ref mut target, ref mut cleanup, .. } => { *target = self.update_target(*target); if let Some(tgt) = *cleanup { *cleanup = Some(self.update_target(tgt)); } else if !self.in_cleanup_block { // Unless this assert is in a cleanup block, add an unwind edge to // the orignal call's cleanup block *cleanup = self.cleanup_block; } } TerminatorKind::Return => { *kind = TerminatorKind::Goto { target: self.return_block }; } TerminatorKind::Resume => { if let Some(tgt) = self.cleanup_block { *kind = TerminatorKind::Goto { target: tgt } } } TerminatorKind::Unreachable => { } TerminatorKind::FalseEdges { ref mut real_target, ref mut imaginary_targets } => { *real_target = self.update_target(*real_target); for target in imaginary_targets { *target = self.update_target(*target); } } } } fn visit_visibility_scope(&mut self, scope: &mut VisibilityScope) { *scope = self.scope_map[*scope]; } fn visit_literal(&mut self, literal: &mut Literal<'tcx>, loc: Location) { if let Literal::Promoted { ref mut index } = *literal { if let Some(p) = self.promoted_map.get(*index).cloned() { *index = p; } } else { self.super_literal(literal, loc); } } }