[EABI] Add LLVM support for -meabi flag

"GCC requires the freestanding environment provide memcpy, memmove, memset
and memcmp": https://gcc.gnu.org/onlinedocs/gcc-5.2.0/gcc/Standards.html

Hence in GNUEABI targets LLVM should not convert 'memops' to their equivalent
'__aeabi_memops'. This convertion violates GCC contract.

The -meabi flag controls whether or not LLVM will modify 'memops' in GNUEABI
targets.

Without -meabi: use the triple default EABI.
With -meabi=default: use the triple default EABI.
With -meabi=gnu: use 'memops'.
With -meabi=4 or -meabi=5: use '__aeabi_memops'.
With -meabi set to an unknown value: same as -meabi=default.

Patch by Vinicius Tinti.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@252462 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/docs/CommandGuide/llc.rst b/docs/CommandGuide/llc.rst
index 8d5c9ce8f8a1a62b05f43c755b2cf7a9a5a3076e..5094259f9f952775536709da9287306f691a1d54 100644 (file)
--- a/docs/CommandGuide/llc.rst
+++ b/docs/CommandGuide/llc.rst
@@ -127,6 +127,12 @@ End-user Options
  implements an LLVM target.  This will permit the target name to be used with
  the :option:`-march` option so that code can be generated for that target.
 
+.. option:: -meabi=[default|gnu|4|5]
+
+ Specify which EABI version should conform to.  Valid EABI versions are *gnu*,
+ *4* and *5*.  Default value (*default*) depends on the triple.
+
+
 Tuning/Configuration Options
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

diff --git a/include/llvm/CodeGen/CommandFlags.h b/include/llvm/CodeGen/CommandFlags.h
index c97d186e512223cd703b645326a1f64df7c59ad1..bc70a447af267f63dd28321c3ae60d4c03e37595 100644 (file)
--- a/include/llvm/CodeGen/CommandFlags.h
+++ b/include/llvm/CodeGen/CommandFlags.h
@@ -247,6 +247,15 @@ JTableType("jump-table-type",
                          "Create one table per unique function type."),
               clEnumValEnd));
 
+cl::opt<llvm::EABI> EABIVersion(
+    "meabi", cl::desc("Set EABI type (default depends on triple):"),
+    cl::init(EABI::Default),
+    cl::values(clEnumValN(EABI::Default, "default",
+                          "Triple default EABI version"),
+               clEnumValN(EABI::EABI4, "4", "EABI version 4"),
+               clEnumValN(EABI::EABI5, "5", "EABI version 5"),
+               clEnumValN(EABI::GNU, "gnu", "EABI GNU"), clEnumValEnd));
+
 // Common utility function tightly tied to the options listed here. Initializes
 // a TargetOptions object with CodeGen flags and returns it.
 static inline TargetOptions InitTargetOptionsFromCodeGenFlags() {
@@ -275,6 +284,7 @@ static inline TargetOptions InitTargetOptionsFromCodeGenFlags() {
   Options.JTType = JTableType;
 
   Options.ThreadModel = TMModel;
+  Options.EABIVersion = EABIVersion;
 
   return Options;
 }

diff --git a/include/llvm/Target/TargetOptions.h b/include/llvm/Target/TargetOptions.h
index 2b087af85906388e64d261b4fdd1c39b4037a2cd..af5a45a856c99f869907b0495da94c993ec81d38 100644 (file)
--- a/include/llvm/Target/TargetOptions.h
+++ b/include/llvm/Target/TargetOptions.h
@@ -58,24 +58,29 @@ namespace llvm {
     };
   }
 
+  enum class EABI {
+    Unknown,
+    Default, // Default means not specified
+    EABI4,   // Target-specific (either 4, 5 or gnu depending on triple).
+    EABI5,
+    GNU
+  };
+
   class TargetOptions {
   public:
     TargetOptions()
-        : PrintMachineCode(false),
-          LessPreciseFPMADOption(false), UnsafeFPMath(false),
-          NoInfsFPMath(false), NoNaNsFPMath(false),
-          HonorSignDependentRoundingFPMathOption(false),
-          NoZerosInBSS(false),
-          GuaranteedTailCallOpt(false),
-          StackAlignmentOverride(0),
+        : PrintMachineCode(false), LessPreciseFPMADOption(false),
+          UnsafeFPMath(false), NoInfsFPMath(false), NoNaNsFPMath(false),
+          HonorSignDependentRoundingFPMathOption(false), NoZerosInBSS(false),
+          GuaranteedTailCallOpt(false), StackAlignmentOverride(0),
           EnableFastISel(false), PositionIndependentExecutable(false),
           UseInitArray(false), DisableIntegratedAS(false),
           CompressDebugSections(false), FunctionSections(false),
           DataSections(false), UniqueSectionNames(true), TrapUnreachable(false),
           EmulatedTLS(false), FloatABIType(FloatABI::Default),
           AllowFPOpFusion(FPOpFusion::Standard), Reciprocals(TargetRecip()),
-          JTType(JumpTable::Single),
-          ThreadModel(ThreadModel::POSIX) {}
+          JTType(JumpTable::Single), ThreadModel(ThreadModel::POSIX),
+          EABIVersion(EABI::Default) {}
 
     /// PrintMachineCode - This flag is enabled when the -print-machineinstrs
     /// option is specified on the command line, and should enable debugging
@@ -213,6 +218,9 @@ namespace llvm {
     /// for things like atomics
     ThreadModel::Model ThreadModel;
 
+    /// EABIVersion - This flag specifies the EABI version
+    EABI EABIVersion;
+
     /// Machine level options.
     MCTargetOptions MCOptions;
   };
@@ -241,6 +249,7 @@ inline bool operator==(const TargetOptions &LHS,
     ARE_EQUAL(Reciprocals) &&
     ARE_EQUAL(JTType) &&
     ARE_EQUAL(ThreadModel) &&
+    ARE_EQUAL(EABIVersion) &&
     ARE_EQUAL(MCOptions);
 #undef ARE_EQUAL
 }

diff --git a/lib/Target/ARM/ARMISelLowering.cpp b/lib/Target/ARM/ARMISelLowering.cpp
index 8f9ca96e5d95d922d24a89f8aea91965f3f4924f..f6a382b44fcbdf2e116bf2921dbd9df62e3e7393 100644 (file)
--- a/lib/Target/ARM/ARMISelLowering.cpp
+++ b/lib/Target/ARM/ARMISelLowering.cpp
@@ -256,8 +256,10 @@ ARMTargetLowering::ARMTargetLowering(const TargetMachine &TM,
   setLibcallName(RTLIB::SRL_I128, nullptr);
   setLibcallName(RTLIB::SRA_I128, nullptr);
 
-  if (Subtarget->isAAPCS_ABI() && !Subtarget->isTargetMachO() &&
-      !Subtarget->isTargetWindows()) {
+  // RTLIB
+  if (Subtarget->isAAPCS_ABI() &&
+      (Subtarget->isTargetAEABI() || Subtarget->isTargetGNUAEABI() ||
+       Subtarget->isTargetAndroid())) {
     static const struct {
       const RTLIB::Libcall Op;
       const char * const Name;
@@ -345,12 +347,6 @@ ARMTargetLowering::ARMTargetLowering(const TargetMachine &TM,
       { RTLIB::UDIV_I16, "__aeabi_uidiv",    CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
       { RTLIB::UDIV_I32, "__aeabi_uidiv",    CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
       { RTLIB::UDIV_I64, "__aeabi_uldivmod", CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
-
-      // Memory operations
-      // RTABI chapter 4.3.4
-      { RTLIB::MEMCPY,  "__aeabi_memcpy",  CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
-      { RTLIB::MEMMOVE, "__aeabi_memmove", CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
-      { RTLIB::MEMSET,  "__aeabi_memset",  CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
     };
 
     for (const auto &LC : LibraryCalls) {
@@ -359,6 +355,30 @@ ARMTargetLowering::ARMTargetLowering(const TargetMachine &TM,
       if (LC.Cond != ISD::SETCC_INVALID)
         setCmpLibcallCC(LC.Op, LC.Cond);
     }
+
+    // EABI dependent RTLIB
+    if (TM.Options.EABIVersion == EABI::EABI4 ||
+        TM.Options.EABIVersion == EABI::EABI5) {
+      static const struct {
+        const RTLIB::Libcall Op;
+        const char *const Name;
+        const CallingConv::ID CC;
+        const ISD::CondCode Cond;
+      } MemOpsLibraryCalls[] = {
+        // Memory operations
+        // RTABI chapter 4.3.4
+        { RTLIB::MEMCPY,  "__aeabi_memcpy",  CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
+        { RTLIB::MEMMOVE, "__aeabi_memmove", CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
+        { RTLIB::MEMSET,  "__aeabi_memset",  CallingConv::ARM_AAPCS, ISD::SETCC_INVALID },
+      };
+
+      for (const auto &LC : MemOpsLibraryCalls) {
+        setLibcallName(LC.Op, LC.Name);
+        setLibcallCallingConv(LC.Op, LC.CC);
+        if (LC.Cond != ISD::SETCC_INVALID)
+          setCmpLibcallCC(LC.Op, LC.Cond);
+      }
+    }
   }
 
   if (Subtarget->isTargetWindows()) {

diff --git a/lib/Target/ARM/ARMSubtarget.h b/lib/Target/ARM/ARMSubtarget.h
index 66b56d10faaa78cd579bbcdab1c37305659eb667..530aa63fff43717a5efb9a962b1b9e77f8a62c6a 100644 (file)
--- a/lib/Target/ARM/ARMSubtarget.h
+++ b/lib/Target/ARM/ARMSubtarget.h
@@ -380,6 +380,11 @@ public:
             TargetTriple.getEnvironment() == Triple::EABIHF) &&
            !isTargetDarwin() && !isTargetWindows();
   }
+  bool isTargetGNUAEABI() const {
+    return (TargetTriple.getEnvironment() == Triple::GNUEABI ||
+            TargetTriple.getEnvironment() == Triple::GNUEABIHF) &&
+           !isTargetDarwin() && !isTargetWindows();
+  }
 
   // ARM Targets that support EHABI exception handling standard
   // Darwin uses SjLj. Other targets might need more checks.

diff --git a/lib/Target/ARM/ARMTargetMachine.cpp b/lib/Target/ARM/ARMTargetMachine.cpp
index 9a5efc8d0b9e88426d3fe2504ba8e11771be7295..fca1901dc57c0f9b833b6726fe9585cb4e6f7bd9 100644 (file)
--- a/lib/Target/ARM/ARMTargetMachine.cpp
+++ b/lib/Target/ARM/ARMTargetMachine.cpp
@@ -188,6 +188,15 @@ ARMBaseTargetMachine::ARMBaseTargetMachine(const Target &T, const Triple &TT,
   if (Options.FloatABIType == FloatABI::Default)
     this->Options.FloatABIType =
         Subtarget.isTargetHardFloat() ? FloatABI::Hard : FloatABI::Soft;
+
+  // Default to triple-appropriate EABI
+  if (Options.EABIVersion == EABI::Default ||
+      Options.EABIVersion == EABI::Unknown) {
+    if (Subtarget.isTargetGNUAEABI())
+      this->Options.EABIVersion = EABI::GNU;
+    else
+      this->Options.EABIVersion = EABI::EABI5;
+  }
 }
 
 ARMBaseTargetMachine::~ARMBaseTargetMachine() {}

diff --git a/test/CodeGen/ARM/arm-eabi.ll b/test/CodeGen/ARM/arm-eabi.ll
new file mode 100644 (file)
index 0000000..d1e7a94
--- /dev/null
+++ b/test/CodeGen/ARM/arm-eabi.ll
@@ -0,0 +1,63 @@
+; RUN: llc < %s -mtriple=arm-none-eabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-eabihf -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-androideabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-gnueabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-gnueabihf -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-eabi -meabi=gnu -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-eabihf -meabi=gnu -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-androideabi -meabi=gnu -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-gnueabi -meabi=gnu -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-gnueabihf -meabi=gnu -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-eabi -meabi=4 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-eabihf -meabi=4 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-androideabi -meabi=4 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-gnueabi -meabi=4 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-gnueabihf -meabi=4 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-eabi -meabi=5 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-eabihf -meabi=5 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-androideabi -meabi=5 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-gnueabi -meabi=5 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-gnueabihf -meabi=5 -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+
+%struct.my_s = type { [18 x i32] }
+
+define void @foo(i32* %t) {
+  ; CHECK-LABEL: foo
+
+  %1 = alloca i32*, align 4
+  store i32* %t, i32** %1, align 4
+  %2 = load i32*, i32** %1, align 4
+  %3 = bitcast i32* %2 to %struct.my_s*
+  %4 = bitcast %struct.my_s* %3 to i8*
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %4, i8* inttoptr (i32 1 to i8*), i32 72, i32 4, i1 false)
+  ret void
+}
+
+define void @f1(i8* %dest, i8* %src) {
+entry:
+  ; CHECK-LABEL: f1
+
+  ; memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
+  call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 0, i1 false)
+
+  ; memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 0, i1 false)
+
+  ; memset
+  ; CHECK-EABI: mov r2, #1
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
+  call void @llvm.memset.p0i8.i32(i8* %dest, i8 1, i32 500, i32 0, i1 false)
+  ret void
+}
+
+declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind

diff --git a/test/CodeGen/ARM/memfunc.ll b/test/CodeGen/ARM/memfunc.ll
index 5223983a7f3020da79f649ce4dd2c5ee0537ea5d..66743f3e9d5e3b32fa52574daf63b5f2239b6d9d 100644 (file)
--- a/test/CodeGen/ARM/memfunc.ll
+++ b/test/CodeGen/ARM/memfunc.ll
@@ -1,79 +1,95 @@
-; RUN: llc < %s -mtriple=armv7-apple-ios -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-IOS --check-prefix=CHECK
-; RUN: llc < %s -mtriple=thumbv7m-none-macho -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-DARWIN --check-prefix=CHECK
-; RUN: llc < %s -mtriple=arm-none-eabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI --check-prefix=CHECK
-; RUN: llc < %s -mtriple=arm-none-eabihf -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI --check-prefix=CHECK
+; RUN: llc < %s -mtriple=armv7-apple-ios -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-IOS
+; RUN: llc < %s -mtriple=thumbv7m-none-macho -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-DARWIN
+; RUN: llc < %s -mtriple=arm-none-eabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-eabihf -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-androideabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-EABI
+; RUN: llc < %s -mtriple=arm-none-gnueabi -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
+; RUN: llc < %s -mtriple=arm-none-gnueabihf -disable-post-ra -o - | FileCheck %s --check-prefix=CHECK-GNUEABI
 
 define void @f1(i8* %dest, i8* %src) {
 entry:
   ; CHECK-LABEL: f1
 
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 0, i1 false)
 
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 0, i1 false)
 
   ; EABI memset swaps arguments
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   call void @llvm.memset.p0i8.i32(i8* %dest, i8 1, i32 500, i32 0, i1 false)
 
   ; EABI uses memclr if value set to 0
   ; CHECK-IOS: mov r1, #0
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #0
-  ; CHECK-DARWIN: memset
-  ; CHECK-EABI: __aeabi_memclr
+  ; CHECK-DARWIN: bl _memset
+  ; CHECK-EABI: bl __aeabi_memclr
+  ; CHECK-GNUEABI: bl memset
   call void @llvm.memset.p0i8.i32(i8* %dest, i8 0, i32 500, i32 0, i1 false)
-  
+
   ; EABI uses aligned function variants if possible
 
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove4
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove4
+  ; CHECK-GNUEABI: bl memmove
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 4, i1 false)
 
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy4
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy4
+  ; CHECK-GNUEABI: bl memcpy
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 4, i1 false)
 
-  ; CHECK-IOS: memset
-  ; CHECK-DARWIN: memset
-  ; CHECK-EABI: __aeabi_memset4
+  ; CHECK-IOS: bl _memset
+  ; CHECK-DARWIN: bl _memset
+  ; CHECK-EABI: bl __aeabi_memset4
+  ; CHECK-GNUEABI: bl memset
   call void @llvm.memset.p0i8.i32(i8* %dest, i8 1, i32 500, i32 4, i1 false)
 
-  ; CHECK-IOS: memset
-  ; CHECK-DARWIN: memset
-  ; CHECK-EABI: __aeabi_memclr4
+  ; CHECK-IOS: bl _memset
+  ; CHECK-DARWIN: bl _memset
+  ; CHECK-EABI: bl __aeabi_memclr4
+  ; CHECK-GNUEABI: bl memset
   call void @llvm.memset.p0i8.i32(i8* %dest, i8 0, i32 500, i32 4, i1 false)
 
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove8
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove8
+  ; CHECK-GNUEABI: bl memmove
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 8, i1 false)
 
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy8
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy8
+  ; CHECK-GNUEABI: bl memcpy
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 8, i1 false)
 
-  ; CHECK-IOS: memset
-  ; CHECK-DARWIN: memset
-  ; CHECK-EABI: __aeabi_memset8
+  ; CHECK-IOS: bl _memset
+  ; CHECK-DARWIN: bl _memset
+  ; CHECK-EABI: bl __aeabi_memset8
+  ; CHECK-GNUEABI: bl memset
   call void @llvm.memset.p0i8.i32(i8* %dest, i8 1, i32 500, i32 8, i1 false)
 
-  ; CHECK-IOS: memset
-  ; CHECK-DARWIN: memset
-  ; CHECK-EABI: __aeabi_memclr8
+  ; CHECK-IOS: bl _memset
+  ; CHECK-DARWIN: bl _memset
+  ; CHECK-EABI: bl __aeabi_memclr8
+  ; CHECK-GNUEABI: bl memset
   call void @llvm.memset.p0i8.i32(i8* %dest, i8 0, i32 500, i32 8, i1 false)
 
   unreachable
@@ -86,32 +102,38 @@ entry:
 
   ; IOS (ARMv7) should 8-byte align, others should 4-byte align
   ; CHECK-IOS: add r1, sp, #32
-  ; CHECK-IOS: memmove
+  ; CHECK-IOS: bl _memmove
   ; CHECK-DARWIN: add r1, sp, #28
-  ; CHECK-DARWIN: memmove
+  ; CHECK-DARWIN: bl _memmove
   ; CHECK-EABI: add r1, sp, #28
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: add r1, sp, #28
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [9 x i8], align 1
   %0 = bitcast [9 x i8]* %arr0 to i8*
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: add r1, sp, #16
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [9 x i8], align 1
   %1 = bitcast [9 x i8]* %arr1 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK-IOS: mov r0, sp
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: add r0, sp, #4
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: add r0, sp, #4
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: add r0, sp, #4
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [9 x i8], align 1
   %2 = bitcast [9 x i8]* %arr2 to i8*
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -125,28 +147,32 @@ entry:
   ; CHECK-LABEL: f3
 
   ; CHECK: {{add(.w)? r1, sp, #17|sub(.w)? r1, r7, #15}}
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [7 x i8], align 1
   %0 = bitcast [7 x i8]* %arr0 to i8*
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r1, sp, #10}}
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [7 x i8], align 1
   %1 = bitcast [7 x i8]* %arr1 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r0, sp, #3}}
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [7 x i8], align 1
   %2 = bitcast [7 x i8]* %arr2 to i8*
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -160,28 +186,32 @@ entry:
   ; CHECK-LABEL: f4
 
   ; CHECK: {{add(.w)? r., sp, #23|sub(.w)? r., r7, #17}}
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [9 x i8], align 1
   %0 = getelementptr inbounds [9 x i8], [9 x i8]* %arr0, i32 0, i32 4
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(10|14)}}
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [9 x i8], align 1
   %1 = getelementptr inbounds [9 x i8], [9 x i8]* %arr1, i32 0, i32 4
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(1|5)}}
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [9 x i8], align 1
   %2 = getelementptr inbounds [9 x i8], [9 x i8]* %arr2, i32 0, i32 4
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -195,28 +225,32 @@ entry:
   ; CHECK-LABEL: f5
 
   ; CHECK: {{add(.w)? r., sp, #27|sub(.w)? r., r7, #21}}
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [13 x i8], align 1
   %0 = getelementptr inbounds [13 x i8], [13 x i8]* %arr0, i32 0, i32 1
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(10|14)}}
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [13 x i8], align 1
   %1 = getelementptr inbounds [13 x i8], [13 x i8]* %arr1, i32 0, i32 1
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(1|5)}}
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [13 x i8], align 1
   %2 = getelementptr inbounds [13 x i8], [13 x i8]* %arr2, i32 0, i32 1
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -230,28 +264,32 @@ entry:
   ; CHECK-LABEL: f6
 
   ; CHECK: {{add(.w)? r., sp, #27|sub(.w)? r., r7, #25}}
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [13 x i8], align 1
   %0 = getelementptr inbounds [13 x i8], [13 x i8]* %arr0, i32 0, i32 %i
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(10|14)}}
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [13 x i8], align 1
   %1 = getelementptr inbounds [13 x i8], [13 x i8]* %arr1, i32 0, i32 %i
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(1|5)}}
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [13 x i8], align 1
   %2 = getelementptr inbounds [13 x i8], [13 x i8]* %arr2, i32 0, i32 %i
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -265,28 +303,32 @@ entry:
   ; CHECK-LABEL: f7
 
   ; CHECK: {{add(.w)? r., sp, #27|sub(.w)? r., r7, #21}}
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [13 x i8], align 1
   %0 = getelementptr [13 x i8], [13 x i8]* %arr0, i32 0, i32 4
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(10|14)}}
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [13 x i8], align 1
   %1 = getelementptr [13 x i8], [13 x i8]* %arr1, i32 0, i32 4
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(1|5)}}
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [13 x i8], align 1
   %2 = getelementptr [13 x i8], [13 x i8]* %arr2, i32 0, i32 4
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -300,28 +342,32 @@ entry:
   ; CHECK-LABEL: f8
 
   ; CHECK: {{add(.w)? r., sp, #27|sub(.w)? r., r7, #21}}
-  ; CHECK-IOS: memmove
-  ; CHECK-DARWIN: memmove
-  ; CHECK-EABI: __aeabi_memmove
+  ; CHECK-IOS: bl _memmove
+  ; CHECK-DARWIN: bl _memmove
+  ; CHECK-EABI: bl __aeabi_memmove
+  ; CHECK-GNUEABI: bl memmove
   %arr0 = alloca [13 x i8], align 1
   %0 = getelementptr inbounds [13 x i8], [13 x i8]* %arr0, i32 0, i32 16
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %dest, i8* %0, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(10|14)}}
-  ; CHECK-IOS: memcpy
-  ; CHECK-DARWIN: memcpy
-  ; CHECK-EABI: __aeabi_memcpy
+  ; CHECK-IOS: bl _memcpy
+  ; CHECK-DARWIN: bl _memcpy
+  ; CHECK-EABI: bl __aeabi_memcpy
+  ; CHECK-GNUEABI: bl memcpy
   %arr1 = alloca [13 x i8], align 1
   %1 = getelementptr inbounds [13 x i8], [13 x i8]* %arr1, i32 0, i32 16
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %1, i32 %n, i32 0, i1 false)
 
   ; CHECK: {{add(.w)? r., sp, #(1|5)}}
   ; CHECK-IOS: mov r1, #1
-  ; CHECK-IOS: memset
+  ; CHECK-IOS: bl _memset
   ; CHECK-DARWIN: movs r1, #1
-  ; CHECK-DARWIN: memset
+  ; CHECK-DARWIN: bl _memset
   ; CHECK-EABI: mov r2, #1
-  ; CHECK-EABI: __aeabi_memset
+  ; CHECK-EABI: bl __aeabi_memset
+  ; CHECK-GNUEABI: mov r1, #1
+  ; CHECK-GNUEABI: bl memset
   %arr2 = alloca [13 x i8], align 1
   %2 = getelementptr inbounds [13 x i8], [13 x i8]* %arr2, i32 0, i32 16
   call void @llvm.memset.p0i8.i32(i8* %2, i8 1, i32 %n, i32 0, i1 false)
@@ -357,6 +403,7 @@ entry:
 ; CHECK-IOS: .align 3
 ; CHECK-DARWIN: .align 2
 ; CHECK-EABI: .align 2
+; CHECK-GNUEABI: .align 2
 ; CHECK: arr2:
 ; CHECK: {{\.section.+foo,bar}}
 ; CHECK-NOT: .align