Add @llvm.assume, lowering, and some basic properties

This is the first commit in a series that add an @llvm.assume intrinsic which
can be used to provide the optimizer with a condition it may assume to be true
(when the control flow would hit the intrinsic call). Some basic properties are added here:

 - llvm.invariant(true) is dead.
 - llvm.invariant(false) is unreachable (this directly corresponds to the
   documented behavior of MSVC's __assume(0)), so is llvm.invariant(undef).

The intrinsic is tagged as writing arbitrarily, in order to maintain control
dependencies. BasicAA has been updated, however, to return NoModRef for any
particular location-based query so that we don't unnecessarily block code
motion.

llvm-svn: 213973

docs/LangRef.rst

@@ -9427,6 +9427,46 @@ Semantics:
 
 This intrinsic is lowered to the ``val``.
 
+'``llvm.assume``' Intrinsic
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+
+::
+
+      declare void @llvm.assume(i1 %cond)
+
+Overview:
+"""""""""
+
+The ``llvm.assume`` allows the optimizer to assume that the provided
+condition is true. This information can then be used in simplifying other parts
+of the code.
+
+Arguments:
+""""""""""
+
+The condition which the optimizer may assume is always true.
+
+Semantics:
+""""""""""
+
+The intrinsic allows the optimizer to assume that the provided condition is
+always true whenever the control flow reaches the intrinsic call. No code is
+generated for this intrinsic, and instructions that contribute only to the
+provided condition are not used for code generation. If the condition is
+violated during execution, the behavior is undefined.
+
+Please note that optimizer might limit the transformations performed on values
+used by the ``llvm.assume`` intrinsic in order to preserve the instructions
+only used to form the intrinsic's input argument. This might prove undesirable
+if the extra information provided by the ``llvm.assume`` intrinsic does cause
+sufficient overall improvement in code quality. For this reason,
+``llvm.assume`` should not be used to document basic mathematical invariants
+that the optimizer can otherwise deduce or facts that are of little use to the
+optimizer.
+
 '``llvm.donothing``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 

include/llvm/IR/Intrinsics.td

@@ -277,6 +277,10 @@ def int_pcmarker      : Intrinsic<[], [llvm_i32_ty]>;
 
 def int_readcyclecounter : Intrinsic<[llvm_i64_ty]>;
 
+// The assume intrinsic is marked as arbitrarily writing so that proper
+// control dependencies will be maintained.
+def int_assume        : Intrinsic<[], [llvm_i1_ty], []>;
+
 // Stack Protector Intrinsic - The stackprotector intrinsic writes the stack
 // guard to the correct place on the stack frame.
 def int_stackprotector : Intrinsic<[], [llvm_ptr_ty, llvm_ptrptr_ty], []>;

lib/Analysis/BasicAliasAnalysis.cpp

@@ -471,10 +471,7 @@ namespace {
                                const Location &Loc) override;
 
     ModRefResult getModRefInfo(ImmutableCallSite CS1,
-                               ImmutableCallSite CS2) override {
-      // The AliasAnalysis base class has some smarts, lets use them.
-      return AliasAnalysis::getModRefInfo(CS1, CS2);
-    }
+                               ImmutableCallSite CS2) override;
 
     /// pointsToConstantMemory - Chase pointers until we find a (constant
     /// global) or not.
@@ -788,6 +785,14 @@ BasicAliasAnalysis::getArgLocation(ImmutableCallSite CS, unsigned ArgIdx,
   return Loc;
 }
 
+static bool isAssumeIntrinsic(ImmutableCallSite CS) {
+  const IntrinsicInst *II = dyn_cast<IntrinsicInst>(CS.getInstruction());
+  if (II && II->getIntrinsicID() == Intrinsic::assume)
+    return true;
+
+  return false;
+}
+
 /// getModRefInfo - Check to see if the specified callsite can clobber the
 /// specified memory object.  Since we only look at local properties of this
 /// function, we really can't say much about this query.  We do, however, use
@@ -840,10 +845,29 @@ BasicAliasAnalysis::getModRefInfo(ImmutableCallSite CS,
       return NoModRef;
   }
 
+  // While the assume intrinsic is marked as arbitrarily writing so that
+  // proper control dependencies will be maintained, it never aliases any
+  // particular memory location.
+  if (isAssumeIntrinsic(CS))
+    return NoModRef;
+
   // The AliasAnalysis base class has some smarts, lets use them.
   return AliasAnalysis::getModRefInfo(CS, Loc);
 }
 
+AliasAnalysis::ModRefResult
+BasicAliasAnalysis::getModRefInfo(ImmutableCallSite CS1,
+                                  ImmutableCallSite CS2) {
+  // While the assume intrinsic is marked as arbitrarily writing so that
+  // proper control dependencies will be maintained, it never aliases any
+  // particular memory location.
+  if (isAssumeIntrinsic(CS1) || isAssumeIntrinsic(CS2))
+    return NoModRef;
+
+  // The AliasAnalysis base class has some smarts, lets use them.
+  return AliasAnalysis::getModRefInfo(CS1, CS2);
+}
+
 /// aliasGEP - Provide a bunch of ad-hoc rules to disambiguate a GEP instruction
 /// against another pointer.  We know that V1 is a GEP, but we don't know
 /// anything about V2.  UnderlyingV1 is GetUnderlyingObject(GEP1, DL),

lib/Analysis/TargetTransformInfo.cpp

@@ -385,6 +385,7 @@ struct NoTTI final : ImmutablePass, TargetTransformInfo {
       // FIXME: This is wrong for libc intrinsics.
       return TCC_Basic;
 
+    case Intrinsic::assume:
     case Intrinsic::dbg_declare:
     case Intrinsic::dbg_value:
     case Intrinsic::invariant_start:

lib/CodeGen/BasicTargetTransformInfo.cpp

@@ -572,6 +572,7 @@ unsigned BasicTTI::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
   case Intrinsic::pow:     ISD = ISD::FPOW;   break;
   case Intrinsic::fma:     ISD = ISD::FMA;    break;
   case Intrinsic::fmuladd: ISD = ISD::FMA;    break;
+  // FIXME: We should return 0 whenever getIntrinsicCost == TCC_Free.
   case Intrinsic::lifetime_start:
   case Intrinsic::lifetime_end:
     return 0;

lib/CodeGen/IntrinsicLowering.cpp

@@ -459,9 +459,10 @@ void IntrinsicLowering::LowerIntrinsicCall(CallInst *CI) {
     CI->replaceAllUsesWith(CI->getOperand(0));
     break;
 
+  case Intrinsic::assume:
   case Intrinsic::var_annotation:
-    break;   // Strip out annotate intrinsic
-    
+    break;   // Strip out these intrinsics
+ 
   case Intrinsic::memcpy: {
     Type *IntPtr = DL.getIntPtrType(Context);
     Value *Size = Builder.CreateIntCast(CI->getArgOperand(2), IntPtr,

lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

@@ -5304,8 +5304,9 @@ SelectionDAGBuilder::visitIntrinsicCall(const CallInst &I, unsigned Intrinsic) {
     // Drop the intrinsic, but forward the value
     setValue(&I, getValue(I.getOperand(0)));
     return nullptr;
+  case Intrinsic::assume:
   case Intrinsic::var_annotation:
-    // Discard annotate attributes
+    // Discard annotate attributes and assumptions
     return nullptr;
 
   case Intrinsic::init_trampoline: {

lib/Transforms/Utils/Local.cpp

@@ -301,6 +301,14 @@ bool llvm::isInstructionTriviallyDead(Instruction *I,
     if (II->getIntrinsicID() == Intrinsic::lifetime_start ||
         II->getIntrinsicID() == Intrinsic::lifetime_end)
       return isa<UndefValue>(II->getArgOperand(1));
+
+    // Assumptions are dead if their condition is trivially true.
+    if (II->getIntrinsicID() == Intrinsic::assume) {
+      if (ConstantInt *Cond = dyn_cast<ConstantInt>(II->getArgOperand(0)))
+        return !Cond->isZero();
+
+      return false;
+    }
   }
 
   if (isAllocLikeFn(I, TLI)) return true;
@@ -1183,6 +1191,26 @@ static bool markAliveBlocks(BasicBlock *BB,
     // instructions into LLVM unreachable insts.  The instruction combining pass
     // canonicalizes unreachable insts into stores to null or undef.
     for (BasicBlock::iterator BBI = BB->begin(), E = BB->end(); BBI != E;++BBI){
+      // Assumptions that are known to be false are equivalent to unreachable.
+      // Also, if the condition is undefined, then we make the choice most
+      // beneficial to the optimizer, and choose that to also be unreachable.
+      if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(BBI))
+        if (II->getIntrinsicID() == Intrinsic::assume) {
+          bool MakeUnreachable = false;
+          if (isa<UndefValue>(II->getArgOperand(0)))
+            MakeUnreachable = true;
+          else if (ConstantInt *Cond =
+                   dyn_cast<ConstantInt>(II->getArgOperand(0)))
+            MakeUnreachable = Cond->isZero();
+
+          if (MakeUnreachable) { 
+            // Don't insert a call to llvm.trap right before the unreachable.
+            changeToUnreachable(BBI, false);
+            Changed = true;
+            break;
+          }
+        }
+
       if (CallInst *CI = dyn_cast<CallInst>(BBI)) {
         if (CI->doesNotReturn()) {
           // If we found a call to a no-return function, insert an unreachable

test/Analysis/BasicAA/assume.ll

@@ -0,0 +1,23 @@
+; RUN: opt < %s -basicaa -aa-eval -print-all-alias-modref-info -disable-output 2>&1 | FileCheck %s
+target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:32-f32:32:32-f64:32:32-v64:32:64-v128:32:128-a0:0:32-n32"
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) #0
+declare void @llvm.assume(i1) #0
+
+define void @test1(i8* %P, i8* %Q) nounwind ssp {
+  tail call void @llvm.assume(i1 true)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+  ret void
+
+; CHECK-LABEL: Function: test1:
+
+; CHECK: MayAlias:	i8* %P, i8* %Q
+; CHECK: NoModRef:  Ptr: i8* %P	<->  tail call void @llvm.assume(i1 true)
+; CHECK: NoModRef:  Ptr: i8* %Q	<->  tail call void @llvm.assume(i1 true)
+; CHECK: Both ModRef:  Ptr: i8* %P	<->  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+; CHECK: Both ModRef:  Ptr: i8* %Q	<->  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+; CHECK: NoModRef:   tail call void @llvm.assume(i1 true) <->   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+; CHECK: NoModRef:   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false) <->   tail call void @llvm.assume(i1 true)
+}
+
+attributes #0 = { nounwind }

test/CodeGen/Generic/assume.ll

@@ -0,0 +1,9 @@
+; RUN: llc < %s
+
+define void @main() {
+        call void @llvm.assume(i1 1)
+        ret void
+}
+
+declare void @llvm.assume(i1) nounwind
+

test/Transforms/InstSimplify/assume.ll

@@ -0,0 +1,13 @@
+; RUN: opt -instsimplify -S < %s | FileCheck %s
+
+define void @test1() {
+  call void @llvm.assume(i1 1)
+  ret void
+
+; CHECK-LABEL: @test1
+; CHECK-NOT: llvm.assume
+; CHECK: ret void
+}
+
+declare void @llvm.assume(i1) nounwind
+

test/Transforms/SimplifyCFG/assume.ll

@@ -0,0 +1,22 @@
+; RUN: opt -simplifycfg -S < %s | FileCheck %s
+
+define void @test1() {
+        call void @llvm.assume(i1 0)
+        ret void
+
+; CHECK-LABEL: @test1
+; CHECK-NOT: llvm.assume
+; CHECK: unreachable
+}
+
+define void @test2() {
+        call void @llvm.assume(i1 undef)
+        ret void
+
+; CHECK-LABEL: @test2
+; CHECK-NOT: llvm.assume
+; CHECK: unreachable
+}
+
+declare void @llvm.assume(i1) nounwind
+