ARM IAS: support .thumb_set

This performs the equivalent of a .set directive in that it creates a symbol
which is an alias for another symbol or value which may possibly be yet
undefined.  This directive also has the added property in that it marks the
aliased symbol as being a thumb function entry point, in the same way that the
.thumb_func directive does.

The current implementation fails one test due to an unrelated issue.  Functions
within .thumb sections are not marked as thumb_func.  The result is that
the aliasee function is not valued correctly.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204059 91177308-0d34-0410-b5e6-96231b3b80d8

4 files changed, 261 insertions, 2 deletions

diff --git a/lib/MC/ELFObjectWriter.cpp b/lib/MC/ELFObjectWriter.cpp
index 54703b9912..dd06304b71 100644
--- a/lib/MC/ELFObjectWriter.cpp
+++ b/lib/MC/ELFObjectWriter.cpp
@@ -470,6 +470,7 @@ uint64_t ELFObjectWriter::SymbolValue(MCSymbolData &OrigData,
     return Data->getCommonAlignment();
 
   const MCSymbol *Symbol = &Data->getSymbol();
+  const bool IsThumbFunc = OrigData.getFlags() & ELF_Other_ThumbFunc;
 
   uint64_t Res = 0;
   if (Symbol->isVariable()) {
@@ -479,7 +480,10 @@ uint64_t ELFObjectWriter::SymbolValue(MCSymbolData &OrigData,
       return 0;
     if (Value.getSymB())
       return 0;
+
     Res = Value.getConstant();
+    if (IsThumbFunc)
+      Res |= 1;
 
     const MCSymbolRefExpr *A = Value.getSymA();
     if (!A)
@@ -496,8 +500,8 @@ uint64_t ELFObjectWriter::SymbolValue(MCSymbolData &OrigData,
     return 0;
 
   Res += Layout.getSymbolOffset(Data);
-  if (Data->getFlags() & ELF_Other_ThumbFunc)
-    ++Res;
+  if (IsThumbFunc || Data->getFlags() & ELF_Other_ThumbFunc)
+    Res |= 1;
 
   return Res;
 }
@@ -590,6 +594,8 @@ void ELFObjectWriter::WriteSymbol(MCDataFragment *SymtabF,
   // Binding and Type share the same byte as upper and lower nibbles
   uint8_t Binding = MCELF::GetBinding(OrigData);
   uint8_t Type = mergeTypeForSet(MCELF::GetType(OrigData), MCELF::GetType(Data));
+  if (OrigData.getFlags() & ELF_Other_ThumbFunc)
+    Type = ELF::STT_FUNC;
   uint8_t Info = (Binding << ELF_STB_Shift) | (Type << ELF_STT_Shift);
 
   // Other and Visibility share the same byte with Visibility using the lower
diff --git a/lib/Target/ARM/AsmParser/ARMAsmParser.cpp b/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
index b361bb70fa..403e147a6a 100644
--- a/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
+++ b/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
@@ -30,6 +30,7 @@
 #include "llvm/MC/MCInst.h"
 #include "llvm/MC/MCInstrDesc.h"
 #include "llvm/MC/MCInstrInfo.h"
+#include "llvm/MC/MCObjectFileInfo.h"
 #include "llvm/MC/MCParser/MCAsmLexer.h"
 #include "llvm/MC/MCParser/MCAsmParser.h"
 #include "llvm/MC/MCParser/MCParsedAsmOperand.h"
@@ -41,6 +42,7 @@
 #include "llvm/MC/MCTargetAsmParser.h"
 #include "llvm/Support/ARMBuildAttributes.h"
 #include "llvm/Support/ARMEHABI.h"
+#include "llvm/Support/COFF.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/ELF.h"
 #include "llvm/Support/MathExtras.h"
@@ -229,6 +231,7 @@ class ARMAsmParser : public MCTargetAsmParser {
   bool parseDirectiveObjectArch(SMLoc L);
   bool parseDirectiveArchExtension(SMLoc L);
   bool parseDirectiveAlign(SMLoc L);
+  bool parseDirectiveThumbSet(SMLoc L);
 
   StringRef splitMnemonic(StringRef Mnemonic, unsigned &PredicationCode,
                           bool &CarrySetting, unsigned &ProcessorIMod,
@@ -8030,6 +8033,8 @@ bool ARMAsmParser::ParseDirective(AsmToken DirectiveID) {
     return parseDirectiveArchExtension(DirectiveID.getLoc());
   else if (IDVal == ".align")
     return parseDirectiveAlign(DirectiveID.getLoc());
+  else if (IDVal == ".thumb_set")
+    return parseDirectiveThumbSet(DirectiveID.getLoc());
   return true;
 }
 
@@ -9086,6 +9091,71 @@ bool ARMAsmParser::parseDirectiveAlign(SMLoc L) {
   return false;
 }
 
+/// parseDirectiveThumbSet
+///  ::= .thumb_set name, value
+bool ARMAsmParser::parseDirectiveThumbSet(SMLoc L) {
+  StringRef Name;
+  if (Parser.parseIdentifier(Name)) {
+    TokError("expected identifier after '.thumb_set'");
+    Parser.eatToEndOfStatement();
+    return false;
+  }
+
+  if (getLexer().isNot(AsmToken::Comma)) {
+    TokError("expected comma after name '" + Name + "'");
+    Parser.eatToEndOfStatement();
+    return false;
+  }
+  Lex();
+
+  const MCExpr *Value;
+  if (Parser.parseExpression(Value)) {
+    TokError("missing expression");
+    Parser.eatToEndOfStatement();
+    return false;
+  }
+
+  if (getLexer().isNot(AsmToken::EndOfStatement)) {
+    TokError("unexpected token");
+    Parser.eatToEndOfStatement();
+    return false;
+  }
+  Lex();
+
+  MCSymbol *Alias = getContext().GetOrCreateSymbol(Name);
+  if (const MCSymbolRefExpr *SRE = dyn_cast<MCSymbolRefExpr>(Value)) {
+    MCSymbol *Sym = getContext().LookupSymbol(SRE->getSymbol().getName());
+    if (!Sym->isDefined()) {
+      getStreamer().EmitSymbolAttribute(Sym, MCSA_Global);
+      getStreamer().EmitAssignment(Alias, Value);
+      return false;
+    }
+
+    const MCObjectFileInfo::Environment Format =
+      getContext().getObjectFileInfo()->getObjectFileType();
+    switch (Format) {
+    case MCObjectFileInfo::IsCOFF: {
+      char Type = COFF::IMAGE_SYM_DTYPE_FUNCTION << COFF::SCT_COMPLEX_TYPE_SHIFT;
+      getStreamer().EmitCOFFSymbolType(Type);
+      // .set values are always local in COFF
+      getStreamer().EmitSymbolAttribute(Alias, MCSA_Local);
+      break;
+    }
+    case MCObjectFileInfo::IsELF:
+      getStreamer().EmitSymbolAttribute(Alias, MCSA_ELF_TypeFunction);
+      break;
+    case MCObjectFileInfo::IsMachO:
+      break;
+    }
+  }
+
+  // FIXME: set the function as being a thumb function via the assembler
+  getStreamer().EmitThumbFunc(Alias);
+  getStreamer().EmitAssignment(Alias, Value);
+
+  return false;
+}
+
 /// Force static initialization.
 extern "C" void LLVMInitializeARMAsmParser() {
   RegisterMCAsmParser<ARMAsmParser> X(TheARMTarget);
diff --git a/test/MC/ARM/thumb_set-diagnostics.s b/test/MC/ARM/thumb_set-diagnostics.s
new file mode 100644
index 0000000000..5f1844de01
--- /dev/null
+++ b/test/MC/ARM/thumb_set-diagnostics.s
@@ -0,0 +1,43 @@
+@ RUN: not llvm-mc -triple armv7-eabi -o /dev/null 2>&1 %s | FileCheck %s
+
+	.syntax unified
+
+	.thumb
+
+	.thumb_set
+
+@ CHECK: error: expected identifier after '.thumb_set'
+@ CHECK: 	.thumb_set
+@ CHECL:                  ^
+
+	.thumb_set ., 0x0b5e55ed
+
+@ CHECK: error: expected identifier after '.thumb_set'
+@ CHECK: 	.thumb_set ., 0x0b5e55ed
+@ CHECK:                   ^
+
+	.thumb_set labelled, 0x1abe11ed
+	.thumb_set invalid, :lower16:labelled
+
+@ CHECK: error: unknown token in expression
+@ CHECK: 	.thumb_set invalid, :lower16:labelled
+@ CHECK:                            ^
+
+	.thumb_set missing_comma
+
+@ CHECK: error: expected comma after name 'missing_comma'
+@ CHECK: 	.thumb_set missing_comma
+@ CHECK:                                ^
+
+	.thumb_set missing_expression,
+
+@ CHECK: error: missing expression
+@ CHECK: 	.thumb_set missing_expression,
+@ CHECK:                                      ^
+
+	.thumb_set trailer_trash, 0x11fe1e55,
+
+@ CHECK: error: unexpected token
+@ CHECK: 	.thumb_set trailer_trash, 0x11fe1e55,
+@ CHECK:                                            ^
+
diff --git a/test/MC/ARM/thumb_set.s b/test/MC/ARM/thumb_set.s
new file mode 100644
index 0000000000..7381a98134
--- /dev/null
+++ b/test/MC/ARM/thumb_set.s
@@ -0,0 +1,140 @@
+@ RUN: llvm-mc -triple armv7-eabi -filetype obj -o - %s | llvm-readobj -t \
+@ RUN:   | FileCheck %s
+
+	.syntax unified
+
+	.arm
+
+	.type arm_func,%function
+arm_func:
+	nop
+
+	.thumb_set alias_arm_func, arm_func
+
+	.thumb
+
+	.type thumb_func,%function
+	.thumb_func
+thumb_func:
+	nop
+
+	.thumb_set alias_thumb_func, thumb_func
+
+	.thumb_set seedless, 0x5eed1e55
+	.thumb_set eggsalad, seedless + 0x87788358
+	.thumb_set faceless, ~eggsalad + 0xe133c002
+
+	.thumb_set alias_undefined_data, badblood
+
+	.data
+
+	.type badblood,%object
+badblood:
+	.long 0xbadb100d
+
+	.type bedazzle,%object
+bedazzle:
+	.long 0xbeda221e
+
+	.text
+	.thumb
+
+	.thumb_set alias_defined_data, bedazzle
+
+	.type alpha,%function
+alpha:
+	nop
+
+        .type beta,%function
+beta:
+	bkpt
+
+	.thumb_set beta, alpha
+
+	.thumb_set alias_undefined, undefined
+
+@ CHECK: Symbol {
+@ CHECK:   Name: alias_arm_func
+@ CHECK:   Value: 0x1
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: alias_defined_data
+@ CHECK:   Value: 0x5
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: alias_thumb_func
+@ CHECK:   Value: 0x5
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: alias_undefined_data
+@ CHECK:   Value: 0x0
+@ CHECK:   Type: Object
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: alpha
+@ CHECK:   Value: 0x6
+@ XFAIL-CHECK:   Value: 0x7
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: arm_func
+@ CHECK:   Value: 0x0
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: bedazzle
+@ CHECK:   Value: 0x4
+@ CHECK:   Type: Object
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: beta
+@ CHECK:   Value: 0x7
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: eggsalad
+@ CHECK:   Value: 0xE665A1AD
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: faceless
+@ CHECK:   Value: 0xFACE1E55
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: seedless
+@ CHECK:   Value: 0x5EED1E55
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: thumb_func
+@ CHECK:   Value: 0x5
+@ CHECK:   Type: Function
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: badblood
+@ CHECK:   Value: 0x0
+@ CHECK:   Type: Object
+@ CHECK: }
+
+@ CHECK: Symbol {
+@ CHECK:   Name: undefined
+@ CHECK:   Value: 0x0
+@ CHECK:   Type: None
+@ CHECK: }
+