[X86] Support -fno-plt __tls_get_addr calls

In general dynamic/local dynamic TLS models, with -fno-plt,

* x86: emit `calll *___tls_get_addr@GOT(%ebx)` instead of `calll ___tls_get_addr@PLT`
  Note, on x86, if we can get rid of %ebx as the PIC register,
  it may be better to use a register not preserved across function calls.
* x86_64: emit `callq *__tls_get_addr@GOTPCREL(%rip)` instead of `callq __tls_get_addr@PLT`

Reorganize the code by separating 32-bit and 64-bit.

Reviewed By: rnk

Differential Revision: https://reviews.llvm.org/D62106

llvm-svn: 361453

llvm/lib/Target/X86/X86MCInstLower.cpp

@@ -683,16 +683,9 @@ void X86MCInstLower::Lower(const MachineInstr *MI, MCInst &OutMI) const {
 
 void X86AsmPrinter::LowerTlsAddr(X86MCInstLower &MCInstLowering,
                                  const MachineInstr &MI) {
-
+  bool Is64Bits = MI.getOpcode() == X86::TLS_addr64 ||
-  bool is64Bits = MI.getOpcode() == X86::TLS_addr64 ||
                   MI.getOpcode() == X86::TLS_base_addr64;
-
+  MCContext &Ctx = OutStreamer->getContext();
-  bool needsPadding = MI.getOpcode() == X86::TLS_addr64;
-
-  MCContext &context = OutStreamer->getContext();
-
-  if (needsPadding)
-    EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX));
 
   MCSymbolRefExpr::VariantKind SRVK;
   switch (MI.getOpcode()) {
@@ -710,51 +703,79 @@ void X86AsmPrinter::LowerTlsAddr(X86MCInstLower &MCInstLowering,
     llvm_unreachable("unexpected opcode");
   }
 
-  MCSymbol *sym = MCInstLowering.GetSymbolFromOperand(MI.getOperand(3));
+  const MCSymbolRefExpr *Sym = MCSymbolRefExpr::create(
-  const MCSymbolRefExpr *symRef = MCSymbolRefExpr::create(sym, SRVK, context);
+      MCInstLowering.GetSymbolFromOperand(MI.getOperand(3)), SRVK, Ctx);
+  bool UseGot = MMI->getModule()->getRtLibUseGOT();
 
-  MCInst LEA;
+  if (Is64Bits) {
-  if (is64Bits) {
+    bool NeedsPadding = SRVK == MCSymbolRefExpr::VK_TLSGD;
-    LEA.setOpcode(X86::LEA64r);
+    if (NeedsPadding)
-    LEA.addOperand(MCOperand::createReg(X86::RDI)); // dest
+      EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX));
-    LEA.addOperand(MCOperand::createReg(X86::RIP)); // base
+    EmitAndCountInstruction(MCInstBuilder(X86::LEA64r)
-    LEA.addOperand(MCOperand::createImm(1));        // scale
+                                .addReg(X86::RDI)
-    LEA.addOperand(MCOperand::createReg(0));        // index
+                                .addReg(X86::RIP)
-    LEA.addOperand(MCOperand::createExpr(symRef));  // disp
+                                .addImm(1)
-    LEA.addOperand(MCOperand::createReg(0));        // seg
+                                .addReg(0)
-  } else if (SRVK == MCSymbolRefExpr::VK_TLSLDM) {
+                                .addExpr(Sym)
-    LEA.setOpcode(X86::LEA32r);
+                                .addReg(0));
-    LEA.addOperand(MCOperand::createReg(X86::EAX)); // dest
+    const MCSymbol *TlsGetAddr = Ctx.getOrCreateSymbol("__tls_get_addr");
-    LEA.addOperand(MCOperand::createReg(X86::EBX)); // base
+    if (NeedsPadding) {
-    LEA.addOperand(MCOperand::createImm(1));        // scale
+      if (!UseGot)
-    LEA.addOperand(MCOperand::createReg(0));        // index
+        EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX));
-    LEA.addOperand(MCOperand::createExpr(symRef));  // disp
+      EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX));
-    LEA.addOperand(MCOperand::createReg(0));        // seg
+      EmitAndCountInstruction(MCInstBuilder(X86::REX64_PREFIX));
+    }
+    if (UseGot) {
+      const MCExpr *Expr = MCSymbolRefExpr::create(
+          TlsGetAddr, MCSymbolRefExpr::VK_GOTPCREL, Ctx);
+      EmitAndCountInstruction(MCInstBuilder(X86::CALL64m)
+                                  .addReg(X86::RIP)
+                                  .addImm(1)
+                                  .addReg(0)
+                                  .addExpr(Expr)
+                                  .addReg(0));
+    } else {
+      EmitAndCountInstruction(
+          MCInstBuilder(X86::CALL64pcrel32)
+              .addExpr(MCSymbolRefExpr::create(TlsGetAddr,
+                                               MCSymbolRefExpr::VK_PLT, Ctx)));
+    }
   } else {
-    LEA.setOpcode(X86::LEA32r);
+    if (SRVK == MCSymbolRefExpr::VK_TLSGD && !UseGot) {
-    LEA.addOperand(MCOperand::createReg(X86::EAX)); // dest
+      EmitAndCountInstruction(MCInstBuilder(X86::LEA32r)
-    LEA.addOperand(MCOperand::createReg(0));        // base
+                                  .addReg(X86::EAX)
-    LEA.addOperand(MCOperand::createImm(1));        // scale
+                                  .addReg(0)
-    LEA.addOperand(MCOperand::createReg(X86::EBX)); // index
+                                  .addImm(1)
-    LEA.addOperand(MCOperand::createExpr(symRef));  // disp
+                                  .addReg(X86::EBX)
-    LEA.addOperand(MCOperand::createReg(0));        // seg
+                                  .addExpr(Sym)
+                                  .addReg(0));
+    } else {
+      EmitAndCountInstruction(MCInstBuilder(X86::LEA32r)
+                                  .addReg(X86::EAX)
+                                  .addReg(X86::EBX)
+                                  .addImm(1)
+                                  .addReg(0)
+                                  .addExpr(Sym)
+                                  .addReg(0));
+    }
+
+    const MCSymbol *TlsGetAddr = Ctx.getOrCreateSymbol("___tls_get_addr");
+    if (UseGot) {
+      const MCExpr *Expr =
+          MCSymbolRefExpr::create(TlsGetAddr, MCSymbolRefExpr::VK_GOT, Ctx);
+      EmitAndCountInstruction(MCInstBuilder(X86::CALL32m)
+                                  .addReg(X86::EBX)
+                                  .addImm(1)
+                                  .addReg(0)
+                                  .addExpr(Expr)
+                                  .addReg(0));
+    } else {
+      EmitAndCountInstruction(
+          MCInstBuilder(X86::CALLpcrel32)
+              .addExpr(MCSymbolRefExpr::create(TlsGetAddr,
+                                               MCSymbolRefExpr::VK_PLT, Ctx)));
+    }
   }
-  EmitAndCountInstruction(LEA);
-
-  if (needsPadding) {
-    EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX));
-    EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX));
-    EmitAndCountInstruction(MCInstBuilder(X86::REX64_PREFIX));
-  }
-
-  StringRef name = is64Bits ? "__tls_get_addr" : "___tls_get_addr";
-  MCSymbol *tlsGetAddr = context.getOrCreateSymbol(name);
-  const MCSymbolRefExpr *tlsRef =
-      MCSymbolRefExpr::create(tlsGetAddr, MCSymbolRefExpr::VK_PLT, context);
-
-  EmitAndCountInstruction(
-      MCInstBuilder(is64Bits ? X86::CALL64pcrel32 : X86::CALLpcrel32)
-          .addExpr(tlsRef));
 }
 
 /// Emit the largest nop instruction smaller than or equal to \p NumBytes

llvm/test/CodeGen/X86/tls-no-plt.ll

@@ -0,0 +1,30 @@
+; RUN: llc < %s -mtriple=i386-linux-musl -relocation-model=pic | FileCheck --check-prefixes=CHECK,X86 %s
+; RUN: llc < %s -mtriple=x86_64-linux-musl -relocation-model=pic | FileCheck --check-prefixes=CHECK,X64 %s
+
+@gd = thread_local global i32 0
+@ld = internal thread_local global i32 0
+
+define i32* @get_gd() {
+entry:
+; CHECK-LABEL: get_gd:
+; X86: leal gd@TLSGD(%ebx), %eax
+; X86: calll *___tls_get_addr@GOT(%ebx)
+
+; X64: leaq gd@TLSGD(%rip), %rdi
+; X64: callq *__tls_get_addr@GOTPCREL(%rip)
+  ret i32* @gd
+}
+
+define i32* @get_ld() {
+; FIXME: This function uses a single thread-local variable, we might want to fall back to general-dynamic.
+; CHECK-LABEL: get_ld:
+; X86: leal ld@TLSLDM(%ebx), %eax
+; X86: calll *___tls_get_addr@GOT(%ebx)
+
+; X64: leaq ld@TLSLD(%rip), %rdi
+; X64: callq *__tls_get_addr@GOTPCREL(%rip)
+  ret i32* @ld
+}
+
+!llvm.module.flags = !{!1}
+!1 = !{i32 7, !"RtLibUseGOT", i32 1}