[XRay][CodeGen] Use PIC-friendly code in XRay sleds and remove synthetic references in .text
Summary:
This is a re-roll of D36615 which uses PLT relocations in the back-end
to the call to __xray_CustomEvent() when building in -fPIC and
-fxray-instrument mode.
Reviewers: pcc, djasper, bkramer
Subscribers: sdardis, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D37373
llvm-svn: 312466
diff --git a/llvm/lib/Target/X86/X86MCInstLower.cpp b/llvm/lib/Target/X86/X86MCInstLower.cpp
index fd2837b..36d8112 100644
--- a/llvm/lib/Target/X86/X86MCInstLower.cpp
+++ b/llvm/lib/Target/X86/X86MCInstLower.cpp
@@ -1047,20 +1047,20 @@
// We want to emit the following pattern, which follows the x86 calling
// convention to prepare for the trampoline call to be patched in.
//
- // <args placement according SysV64 calling convention>
// .p2align 1, ...
// .Lxray_event_sled_N:
- // jmp +N // jump across the call instruction
- // callq __xray_CustomEvent // force relocation to symbol
- // <args cleanup, jump to here>
- //
- // The relative jump needs to jump forward 24 bytes:
- // 10 (args) + 5 (nops) + 9 (cleanup)
+ // jmp +N // jump across the instrumentation sled
+ // ... // set up arguments in register
+ // callq __xray_CustomEvent@plt // force dependency to symbol
+ // ...
+ // <jump here>
//
// After patching, it would look something like:
//
// nopw (2-byte nop)
+ // ...
// callq __xrayCustomEvent // already lowered
+ // ...
//
// ---
// First we emit the label and the jump.
@@ -1072,49 +1072,57 @@
// Use a two-byte `jmp`. This version of JMP takes an 8-bit relative offset as
// an operand (computed as an offset from the jmp instruction).
// FIXME: Find another less hacky way do force the relative jump.
- OutStreamer->EmitBytes("\xeb\x14");
+ OutStreamer->EmitBinaryData("\xeb\x0f");
// The default C calling convention will place two arguments into %rcx and
// %rdx -- so we only work with those.
- unsigned UsedRegs[] = {X86::RDI, X86::RSI, X86::RAX};
+ unsigned UsedRegs[] = {X86::RDI, X86::RSI};
+ bool UsedMask[] = {false, false};
- // Because we will use %rax, we preserve that across the call.
- EmitAndCountInstruction(MCInstBuilder(X86::PUSH64r).addReg(X86::RAX));
-
- // Then we put the operands in the %rdi and %rsi registers.
+ // Then we put the operands in the %rdi and %rsi registers. We spill the
+ // values in the register before we clobber them, and mark them as used in
+ // UsedMask. In case the arguments are already in the correct register, we use
+ // emit nops appropriately sized to keep the sled the same size in every
+ // situation.
for (unsigned I = 0; I < MI.getNumOperands(); ++I)
if (auto Op = MCIL.LowerMachineOperand(&MI, MI.getOperand(I))) {
- if (Op->isImm())
- EmitAndCountInstruction(MCInstBuilder(X86::MOV64ri)
+ assert(Op->isReg() && "Only support arguments in registers");
+ if (Op->getReg() != UsedRegs[I]) {
+ UsedMask[I] = true;
+ EmitAndCountInstruction(
+ MCInstBuilder(X86::PUSH64r).addReg(UsedRegs[I]));
+ EmitAndCountInstruction(MCInstBuilder(X86::MOV64rr)
.addReg(UsedRegs[I])
- .addImm(Op->getImm()));
- else if (Op->isReg()) {
- if (Op->getReg() != UsedRegs[I])
- EmitAndCountInstruction(MCInstBuilder(X86::MOV64rr)
- .addReg(UsedRegs[I])
- .addReg(Op->getReg()));
- else
- EmitNops(*OutStreamer, 3, Subtarget->is64Bit(), getSubtargetInfo());
+ .addReg(Op->getReg()));
+ } else {
+ EmitNops(*OutStreamer, 4, Subtarget->is64Bit(), getSubtargetInfo());
}
}
// We emit a hard dependency on the __xray_CustomEvent symbol, which is the
- // name of the trampoline to be implemented by the XRay runtime. We put this
- // explicitly in the %rax register.
+ // name of the trampoline to be implemented by the XRay runtime.
auto TSym = OutContext.getOrCreateSymbol("__xray_CustomEvent");
MachineOperand TOp = MachineOperand::CreateMCSymbol(TSym);
- EmitAndCountInstruction(MCInstBuilder(X86::MOV64ri)
- .addReg(X86::RAX)
- .addOperand(MCIL.LowerSymbolOperand(TOp, TSym)));
+ if (isPositionIndependent())
+ TOp.setTargetFlags(X86II::MO_PLT);
// Emit the call instruction.
- EmitAndCountInstruction(MCInstBuilder(X86::CALL64r).addReg(X86::RAX));
+ EmitAndCountInstruction(MCInstBuilder(X86::CALL64pcrel32)
+ .addOperand(MCIL.LowerSymbolOperand(TOp, TSym)));
// Restore caller-saved and used registers.
- OutStreamer->AddComment("xray custom event end.");
- EmitAndCountInstruction(MCInstBuilder(X86::POP64r).addReg(X86::RAX));
+ for (unsigned I = sizeof UsedMask; I-- > 0;)
+ if (UsedMask[I])
+ EmitAndCountInstruction(MCInstBuilder(X86::POP64r).addReg(UsedRegs[I]));
+ else
+ EmitNops(*OutStreamer, 1, Subtarget->is64Bit(), getSubtargetInfo());
- recordSled(CurSled, MI, SledKind::CUSTOM_EVENT);
+ OutStreamer->AddComment("xray custom event end.");
+
+ // Record the sled version. Older versions of this sled were spelled
+ // differently, so we let the runtime handle the different offsets we're
+ // using.
+ recordSled(CurSled, MI, SledKind::CUSTOM_EVENT, 1);
}
void X86AsmPrinter::LowerPATCHABLE_FUNCTION_ENTER(const MachineInstr &MI,
@@ -1125,7 +1133,6 @@
// .Lxray_sled_N:
// jmp .tmpN
// # 9 bytes worth of noops
- // .tmpN
//
// We need the 9 bytes because at runtime, we'd be patching over the full 11
// bytes with the following pattern:
@@ -1136,14 +1143,12 @@
auto CurSled = OutContext.createTempSymbol("xray_sled_", true);
OutStreamer->EmitCodeAlignment(2);
OutStreamer->EmitLabel(CurSled);
- auto Target = OutContext.createTempSymbol();
// Use a two-byte `jmp`. This version of JMP takes an 8-bit relative offset as
// an operand (computed as an offset from the jmp instruction).
// FIXME: Find another less hacky way do force the relative jump.
OutStreamer->EmitBytes("\xeb\x09");
EmitNops(*OutStreamer, 9, Subtarget->is64Bit(), getSubtargetInfo());
- OutStreamer->EmitLabel(Target);
recordSled(CurSled, MI, SledKind::FUNCTION_ENTER);
}