From: Sanjay Patel <spatel@rotateright.com>
Date: Wed, 1 Jul 2015 18:10:20 +0000 (+0000)
Subject: add a cl::opt override for TargetLoweringBase's JumpIsExpensive
X-Git-Url: http://plrg.eecs.uci.edu/git/?p=oota-llvm.git;a=commitdiff_plain;h=7472456fe973adfd371890fddf16d087569e90a5

add a cl::opt override for TargetLoweringBase's JumpIsExpensive

This patch is not intended to change existing codegen behavior for any target.
It just exposes the JumpIsExpensive setting on the command-line to allow for
easier testing and emergency overrides.

Also, change the existing regression test to use FileCheck, explicitly specify
the jump-is-expensive option, and use more precise checks.

Differential Revision: http://reviews.llvm.org/D10846



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@241179 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/include/llvm/Target/TargetLowering.h b/include/llvm/Target/TargetLowering.h
index 80826cb6c3e..277487fee6b 100644
--- a/include/llvm/Target/TargetLowering.h
+++ b/include/llvm/Target/TargetLowering.h
@@ -1235,11 +1235,10 @@ protected:
     HasExtractBitsInsn = hasExtractInsn;
   }
 
-  /// Tells the code generator not to expand sequence of operations into a
-  /// separate sequences that increases the amount of flow control.
-  void setJumpIsExpensive(bool isExpensive = true) {
-    JumpIsExpensive = isExpensive;
-  }
+  /// Tells the code generator not to expand logic operations on comparison
+  /// predicates into separate sequences that increase the amount of flow
+  /// control.
+  void setJumpIsExpensive(bool isExpensive = true);
 
   /// Tells the code generator that integer divide is expensive, and if
   /// possible, should be replaced by an alternate sequence of instructions not
diff --git a/lib/CodeGen/TargetLoweringBase.cpp b/lib/CodeGen/TargetLoweringBase.cpp
index 53f3b00789c..78492a6e881 100644
--- a/lib/CodeGen/TargetLoweringBase.cpp
+++ b/lib/CodeGen/TargetLoweringBase.cpp
@@ -38,6 +38,11 @@
 #include <cctype>
 using namespace llvm;
 
+static cl::opt<bool> JumpIsExpensiveOverride(
+    "jump-is-expensive", cl::init(false),
+    cl::desc("Do not create extra branches to split comparison logic."),
+    cl::Hidden);
+
 /// InitLibcallNames - Set default libcall names.
 ///
 static void InitLibcallNames(const char **Names, const Triple &TT) {
@@ -757,7 +762,7 @@ TargetLoweringBase::TargetLoweringBase(const TargetMachine &tm) : TM(tm) {
   IntDivIsCheap = false;
   FsqrtIsCheap = false;
   Pow2SDivIsCheap = false;
-  JumpIsExpensive = false;
+  JumpIsExpensive = JumpIsExpensiveOverride;
   PredictableSelectIsExpensive = false;
   MaskAndBranchFoldingIsLegal = false;
   EnableExtLdPromotion = false;
@@ -915,6 +920,12 @@ bool TargetLoweringBase::canOpTrap(unsigned Op, EVT VT) const {
   }
 }
 
+void TargetLoweringBase::setJumpIsExpensive(bool isExpensive) {
+  // If the command-line option was specified, ignore this request.
+  if (!JumpIsExpensiveOverride.getNumOccurrences())
+    JumpIsExpensive = isExpensive;
+}
+
 TargetLoweringBase::LegalizeKind
 TargetLoweringBase::getTypeConversion(LLVMContext &Context, EVT VT) const {
   // If this is a simple type, use the ComputeRegisterProp mechanism.
diff --git a/test/CodeGen/X86/or-branch.ll b/test/CodeGen/X86/or-branch.ll
index ae3ed3f8344..9db948adb46 100644
--- a/test/CodeGen/X86/or-branch.ll
+++ b/test/CodeGen/X86/or-branch.ll
@@ -1,19 +1,28 @@
-; RUN: llc < %s -march=x86  | not grep set
+; RUN: llc < %s -mtriple=i386-unknown-unknown -jump-is-expensive=0 | FileCheck %s --check-prefix=JUMP2
+; RUN: llc < %s -mtriple=i386-unknown-unknown -jump-is-expensive=1 | FileCheck %s --check-prefix=JUMP1
 
 define void @foo(i32 %X, i32 %Y, i32 %Z) nounwind {
+; JUMP2-LABEL: foo:
+; JUMP2-DAG:     jl
+; JUMP2-DAG:     je
+;
+; JUMP1-LABEL: foo:
+; JUMP1-DAG:     sete
+; JUMP1-DAG:     setl
+; JUMP1:         orb
+; JUMP1:         jne
 entry:
-	%tmp = tail call i32 (...) @bar( )		; <i32> [#uses=0]
-	%tmp.upgrd.1 = icmp eq i32 %X, 0		; <i1> [#uses=1]
-	%tmp3 = icmp slt i32 %Y, 5		; <i1> [#uses=1]
-	%tmp4 = or i1 %tmp3, %tmp.upgrd.1		; <i1> [#uses=1]
-	br i1 %tmp4, label %cond_true, label %UnifiedReturnBlock
+  %tmp1 = icmp eq i32 %X, 0
+  %tmp3 = icmp slt i32 %Y, 5
+  %tmp4 = or i1 %tmp3, %tmp1
+  br i1 %tmp4, label %cond_true, label %UnifiedReturnBlock
 
-cond_true:		; preds = %entry
-	%tmp5 = tail call i32 (...) @bar( )		; <i32> [#uses=0]
-	ret void
+cond_true:
+  %tmp5 = tail call i32 (...) @bar( )
+  ret void
 
-UnifiedReturnBlock:		; preds = %entry
-	ret void
+UnifiedReturnBlock:
+  ret void
 }
 
 declare i32 @bar(...)