[SimplifyLibcalls] Atoi, strtol replacements

Reviewers: spatel, lebedev.ri, xbolva00, efriedma

Reviewed By: xbolva00, efriedma

Subscribers: efriedma, llvm-commits

Differential Revision: https://reviews.llvm.org/D45418

llvm-svn: 330860

4 files changed, 322 insertions, 0 deletions

diff --git a/llvm/include/llvm/Transforms/Utils/SimplifyLibCalls.h b/llvm/include/llvm/Transforms/Utils/SimplifyLibCalls.h
index eaed60b78e7..87ff7ff3c2f 100644
--- a/llvm/include/llvm/Transforms/Utils/SimplifyLibCalls.h
+++ b/llvm/include/llvm/Transforms/Utils/SimplifyLibCalls.h
@@ -151,6 +151,8 @@ private:
   Value *optimizeIsDigit(CallInst *CI, IRBuilder<> &B);
   Value *optimizeIsAscii(CallInst *CI, IRBuilder<> &B);
   Value *optimizeToAscii(CallInst *CI, IRBuilder<> &B);
+  Value *optimizeAtoi(CallInst *CI, IRBuilder<> &B);
+  Value *optimizeStrtol(CallInst *CI, IRBuilder<> &B);
 
   // Formatting and IO Library Call Optimizations
   Value *optimizeErrorReporting(CallInst *CI, IRBuilder<> &B,
diff --git a/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp b/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
index e65ecc55c83..ce804aef7da 100644
--- a/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
+++ b/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
@@ -102,6 +102,31 @@ static bool callHasFloatingPointArgument(const CallInst *CI) {
   });
 }
 
+static Value *convertStrToNumber(CallInst *CI, StringRef &Str, int64_t Base) {
+  if (Base < 2 || Base > 36)
+    // handle special zero base
+    if (Base != 0)
+      return nullptr;
+
+  char *End;
+  std::string nptr = Str.str();
+  errno = 0;
+  long long int Result = strtoll(nptr.c_str(), &End, Base);
+  if (errno)
+    return nullptr;
+
+  // if we assume all possible target locales are ASCII supersets,
+  // then if strtoll successfully parses a number on the host,
+  // it will also successfully parse the same way on the target
+  if (*End != '\0')
+    return nullptr;
+
+  if (!isIntN(CI->getType()->getPrimitiveSizeInBits(), Result))
+    return nullptr;
+
+  return ConstantInt::get(CI->getType(), Result);
+}
+
 //===----------------------------------------------------------------------===//
 // String and Memory Library Call Optimizations
 //===----------------------------------------------------------------------===//
@@ -1663,6 +1688,29 @@ Value *LibCallSimplifier::optimizeToAscii(CallInst *CI, IRBuilder<> &B) {
                      ConstantInt::get(CI->getType(), 0x7F));
 }
 
+Value *LibCallSimplifier::optimizeAtoi(CallInst *CI, IRBuilder<> &B) {
+  StringRef Str;
+  if (!getConstantStringInfo(CI->getArgOperand(0), Str))
+    return nullptr;
+
+  return convertStrToNumber(CI, Str, 10);
+}
+
+Value *LibCallSimplifier::optimizeStrtol(CallInst *CI, IRBuilder<> &B) {
+  StringRef Str;
+  if (!getConstantStringInfo(CI->getArgOperand(0), Str))
+    return nullptr;
+
+  if (!isa<ConstantPointerNull>(CI->getArgOperand(1)))
+    return nullptr;
+
+  if (ConstantInt *CInt = dyn_cast<ConstantInt>(CI->getArgOperand(2))) {
+    return convertStrToNumber(CI, Str, CInt->getSExtValue());
+  }
+
+  return nullptr;
+}
+
 //===----------------------------------------------------------------------===//
 // Formatting and IO Library Call Optimizations
 //===----------------------------------------------------------------------===//
@@ -2259,6 +2307,13 @@ Value *LibCallSimplifier::optimizeCall(CallInst *CI) {
       return optimizeIsAscii(CI, Builder);
     case LibFunc_toascii:
       return optimizeToAscii(CI, Builder);
+    case LibFunc_atoi:
+    case LibFunc_atol:
+    case LibFunc_atoll:
+      return optimizeAtoi(CI, Builder);
+    case LibFunc_strtol:
+    case LibFunc_strtoll:
+      return optimizeStrtol(CI, Builder);
     case LibFunc_printf:
       return optimizePrintF(CI, Builder);
     case LibFunc_sprintf:
diff --git a/llvm/test/Transforms/InstCombine/str-int-2.ll b/llvm/test/Transforms/InstCombine/str-int-2.ll
new file mode 100644
index 00000000000..37e64a31ee8
--- /dev/null
+++ b/llvm/test/Transforms/InstCombine/str-int-2.ll
@@ -0,0 +1,131 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+@.str = private unnamed_addr constant [3 x i8] c"12\00", align 1
+@.str.1 = private unnamed_addr constant [2 x i8] c"0\00", align 1
+@.str.2 = private unnamed_addr constant [11 x i8] c"4294967296\00", align 1
+@.str.3 = private unnamed_addr constant [24 x i8] c"10000000000000000000000\00", align 1
+@.str.4 = private unnamed_addr constant [20 x i8] c"9923372036854775807\00", align 1
+@.str.5 = private unnamed_addr constant [11 x i8] c"4994967295\00", align 1
+@.str.6 = private unnamed_addr constant [10 x i8] c"499496729\00", align 1
+@.str.7 = private unnamed_addr constant [11 x i8] c"4994967295\00", align 1
+
+declare i64 @strtol(i8*, i8**, i32)
+declare i32 @atoi(i8*)
+declare i64 @atol(i8*)
+declare i64 @atoll(i8*)
+declare i64 @strtoll(i8*, i8**, i32)
+
+define i64 @strtol_dec() #0 {
+; CHECK-LABEL: @strtol_dec(
+; CHECK-NEXT:    ret i64 12
+;
+  %call = call i64 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 10) #2
+  ret i64 %call
+}
+
+define i64 @strtol_base_zero() #0 {
+; CHECK-LABEL: @strtol_base_zero(
+; CHECK-NEXT:    ret i64 12
+;
+  %call = call i64 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 0) #2
+  ret i64 %call
+}
+
+define i64 @strtol_hex() #0 {
+; CHECK-LABEL: @strtol_hex(
+; CHECK-NEXT:    ret i64 18
+;
+  %call = call i64 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 16) #2
+  ret i64 %call
+}
+
+define i64 @strtol_endptr_not_null() #0 {
+; CHECK-LABEL: @strtol_endptr_not_null(
+; CHECK-NEXT:    [[END:%.*]] = alloca i8*, align 4
+; CHECK-NEXT:    [[CALL:%.*]] = call i64 @strtol(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str.1, i64 0, i64 0), i8** nonnull [[END]], i32 10)
+; CHECK-NEXT:    ret i64 [[CALL]]
+;
+  %end = alloca i8*, align 4
+  %call = call i64 @strtol(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str.1, i32 0, i32 0), i8** %end, i32 10) #2
+  ret i64 %call
+}
+
+define i32 @atoi_test() #0 {
+; CHECK-LABEL: @atoi_test(
+; CHECK-NEXT:    ret i32 12
+;
+  %call = call i32 @atoi(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0)) #4
+  ret i32 %call
+}
+
+define i64 @strtol_not_const_str(i8* %s) #0 {
+; CHECK-LABEL: @strtol_not_const_str(
+; CHECK-NEXT:    [[CALL:%.*]] = call i64 @strtol(i8* nocapture [[S:%.*]], i8** null, i32 10)
+; CHECK-NEXT:    ret i64 [[CALL]]
+;
+  %call = call i64 @strtol(i8* %s, i8** null, i32 10) #3
+  ret i64 %call
+}
+
+define i32 @atoi_not_const_str(i8* %s) #0 {
+; CHECK-LABEL: @atoi_not_const_str(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @atoi(i8* [[S:%.*]])
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @atoi(i8* %s) #4
+  ret i32 %call
+}
+
+define i64 @strtol_not_const_base(i32 %b) #0 {
+; CHECK-LABEL: @strtol_not_const_base(
+; CHECK-NEXT:    [[CALL:%.*]] = call i64 @strtol(i8* nocapture getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i64 0, i64 0), i8** null, i32 [[B:%.*]])
+; CHECK-NEXT:    ret i64 [[CALL]]
+;
+  %call = call i64 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 %b) #2
+  ret i64 %call
+}
+
+define i64 @strtol_long_int() #0 {
+; CHECK-LABEL: @strtol_long_int(
+; CHECK-NEXT:    ret i64 4294967296
+;
+  %call = call i64 @strtol(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.2, i32 0, i32 0), i8** null, i32 10) #3
+  ret i64 %call
+}
+
+
+define i64 @strtol_big_overflow() #0 {
+; CHECK-LABEL: @strtol_big_overflow(
+; CHECK-NEXT:    [[CALL:%.*]] = call i64 @strtol(i8* nocapture getelementptr inbounds ([24 x i8], [24 x i8]* @.str.3, i64 0, i64 0), i8** null, i32 10)
+; CHECK-NEXT:    ret i64 [[CALL]]
+;
+  %call = call i64 @strtol(i8* nocapture getelementptr inbounds ([24 x i8], [24 x i8]* @.str.3, i64 0, i64 0), i8** null, i32 10) #2
+  ret i64 %call
+}
+
+define i64 @atol_test() #0 {
+; CHECK-LABEL: @atol_test(
+; CHECK-NEXT:    ret i64 499496729
+;
+; CHECK-NEXT
+  %call = call i64 @atol(i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str.6, i32 0, i32 0)) #4
+  ret i64 %call
+}
+
+define i64 @atoll_test() #0 {
+; CHECK-LABEL: @atoll_test(
+; CHECK-NEXT:    ret i64 4994967295
+;
+  %call = call i64 @atoll(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.5, i32 0, i32 0)) #3
+  ret i64 %call
+}
+
+define i64 @strtoll_test() #0 {
+; CHECK-LABEL: @strtoll_test(
+; CHECK-NEXT:    ret i64 4994967295
+;
+; CHECK-NEXT
+  %call = call i64 @strtoll(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.7, i32 0, i32 0), i8** null, i32 10) #5
+  ret i64 %call
+}
diff --git a/llvm/test/Transforms/InstCombine/str-int.ll b/llvm/test/Transforms/InstCombine/str-int.ll
new file mode 100644
index 00000000000..ac5b6ceccce
--- /dev/null
+++ b/llvm/test/Transforms/InstCombine/str-int.ll
@@ -0,0 +1,134 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+@.str = private unnamed_addr constant [3 x i8] c"12\00", align 1
+@.str.1 = private unnamed_addr constant [2 x i8] c"0\00", align 1
+@.str.2 = private unnamed_addr constant [11 x i8] c"4294967296\00", align 1
+@.str.3 = private unnamed_addr constant [24 x i8] c"10000000000000000000000\00", align 1
+@.str.4 = private unnamed_addr constant [20 x i8] c"9923372036854775807\00", align 1
+@.str.5 = private unnamed_addr constant [11 x i8] c"4994967295\00", align 1
+@.str.6 = private unnamed_addr constant [10 x i8] c"499496729\00", align 1
+@.str.7 = private unnamed_addr constant [11 x i8] c"4994967295\00", align 1
+
+declare i32 @strtol(i8*, i8**, i32)
+declare i32 @atoi(i8*)
+declare i32 @atol(i8*)
+declare i32 @atoll(i8*)
+declare i32 @strtoll(i8*, i8**, i32)
+
+define i32 @strtol_dec() #0 {
+; CHECK-LABEL: @strtol_dec(
+; CHECK-NEXT:    ret i32 12
+;
+  %call = call i32 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 10) #2
+  ret i32 %call
+}
+
+define i32 @strtol_base_zero() #0 {
+; CHECK-LABEL: @strtol_base_zero(
+; CHECK-NEXT:    ret i32 12
+;
+  %call = call i32 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 0) #2
+  ret i32 %call
+}
+
+define i32 @strtol_hex() #0 {
+; CHECK-LABEL: @strtol_hex(
+; CHECK-NEXT:    ret i32 18
+;
+  %call = call i32 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 16) #2
+  ret i32 %call
+}
+
+define i32 @strtol_endptr_not_null() #0 {
+; CHECK-LABEL: @strtol_endptr_not_null(
+; CHECK-NEXT:    [[END:%.*]] = alloca i8*, align 4
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @strtol(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str.1, i64 0, i64 0), i8** nonnull [[END]], i32 10)
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %end = alloca i8*, align 4
+  %call = call i32 @strtol(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str.1, i32 0, i32 0), i8** %end, i32 10) #2
+  ret i32 %call
+}
+
+define i32 @atoi_test() #0 {
+; CHECK-LABEL: @atoi_test(
+; CHECK-NEXT:    ret i32 12
+;
+  %call = call i32 @atoi(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0)) #4
+  ret i32 %call
+}
+
+define i32 @strtol_not_const_str(i8* %s) #0 {
+; CHECK-LABEL: @strtol_not_const_str(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @strtol(i8* nocapture [[S:%.*]], i8** null, i32 10)
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @strtol(i8* %s, i8** null, i32 10) #3
+  ret i32 %call
+}
+
+define i32 @atoi_not_const_str(i8* %s) #0 {
+; CHECK-LABEL: @atoi_not_const_str(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @atoi(i8* [[S:%.*]])
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @atoi(i8* %s) #4
+  ret i32 %call
+}
+
+define i32 @strtol_not_const_base(i32 %b) #0 {
+; CHECK-LABEL: @strtol_not_const_base(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @strtol(i8* nocapture getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i64 0, i64 0), i8** null, i32 [[B:%.*]])
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @strtol(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i8** null, i32 %b) #2
+  ret i32 %call
+}
+
+define i32 @strtol_long_int() #0 {
+; CHECK-LABEL: @strtol_long_int(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @strtol(i8* nocapture getelementptr inbounds ([11 x i8], [11 x i8]* @.str.2, i64 0, i64 0), i8** null, i32 10)
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @strtol(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.2, i32 0, i32 0), i8** null, i32 10) #3
+  ret i32 %call
+}
+
+
+define i32 @strtol_big_overflow() #0 {
+; CHECK-LABEL: @strtol_big_overflow(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @strtol(i8* nocapture getelementptr inbounds ([24 x i8], [24 x i8]* @.str.3, i64 0, i64 0), i8** null, i32 10)
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @strtol(i8* nocapture getelementptr inbounds ([24 x i8], [24 x i8]* @.str.3, i32 0, i32 0), i8** null, i32 10) #2
+  ret i32 %call
+}
+
+define i32 @atol_test() #0 {
+; CHECK-LABEL: @atol_test(
+; CHECK-NEXT:    ret i32 499496729
+;
+; CHECK-NEXT
+  %call = call i32 @atol(i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str.6, i32 0, i32 0)) #4
+  ret i32 %call
+}
+
+define i32 @atoll_test() #0 {
+; CHECK-LABEL: @atoll_test(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @atoll(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.5, i64 0, i64 0))
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+  %call = call i32 @atoll(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.5, i32 0, i32 0)) #3
+  ret i32 %call
+}
+
+define i32 @strtoll_test() #0 {
+; CHECK-LABEL: @strtoll_test(
+; CHECK-NEXT:    [[CALL:%.*]] = call i32 @strtoll(i8* nocapture getelementptr inbounds ([11 x i8], [11 x i8]* @.str.7, i64 0, i64 0), i8** null, i32 10)
+; CHECK-NEXT:    ret i32 [[CALL]]
+;
+; CHECK-NEXT
+  %call = call i32 @strtoll(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str.7, i32 0, i32 0), i8** null, i32 10) #5
+  ret i32 %call
+}