/// to this register without worry about clobbering it.
static unsigned findDeadCallerSavedReg(MachineBasicBlock &MBB,
MachineBasicBlock::iterator &MBBI,
- const TargetRegisterInfo *TRI,
+ const X86RegisterInfo *TRI,
bool Is64Bit) {
const MachineFunction *MF = MBB.getParent();
const Function *F = MF->getFunction();
if (!F || MF->getMMI().callsEHReturn())
return 0;
- static const uint16_t CallerSavedRegs32Bit[] = {
- X86::EAX, X86::EDX, X86::ECX, 0
- };
-
- static const uint16_t CallerSavedRegs64Bit[] = {
- X86::RAX, X86::RDX, X86::RCX, X86::RSI, X86::RDI,
- X86::R8, X86::R9, X86::R10, X86::R11, 0
- };
+ const TargetRegisterClass &AvailableRegs = *TRI->getGPRsForTailCall(*MF);
unsigned Opc = MBBI->getOpcode();
switch (Opc) {
Uses.insert(*AI);
}
- const uint16_t *CS = Is64Bit ? CallerSavedRegs64Bit : CallerSavedRegs32Bit;
- for (; *CS; ++CS)
- if (!Uses.count(*CS))
- return *CS;
+ for (auto CS : AvailableRegs)
+ if (!Uses.count(CS) && CS != X86::RIP)
+ return CS;
}
}
return &X86::GR64_NOREX_NOSPRegClass;
return &X86::GR32_NOREX_NOSPRegClass;
case 4: // Available for tailcall (not callee-saved GPRs).
- const Function *F = MF.getFunction();
- if (IsWin64 || (F && F->getCallingConv() == CallingConv::X86_64_Win64))
- return &X86::GR64_TCW64RegClass;
- else if (Is64Bit)
- return &X86::GR64_TCRegClass;
-
- bool hasHipeCC = (F ? F->getCallingConv() == CallingConv::HiPE : false);
- if (hasHipeCC)
- return &X86::GR32RegClass;
- return &X86::GR32_TCRegClass;
+ return getGPRsForTailCall(MF);
}
}
+const TargetRegisterClass *
+X86RegisterInfo::getGPRsForTailCall(const MachineFunction &MF) const {
+ const Function *F = MF.getFunction();
+ if (IsWin64 || (F && F->getCallingConv() == CallingConv::X86_64_Win64))
+ return &X86::GR64_TCW64RegClass;
+ else if (Is64Bit)
+ return &X86::GR64_TCRegClass;
+
+ bool hasHipeCC = (F ? F->getCallingConv() == CallingConv::HiPE : false);
+ if (hasHipeCC)
+ return &X86::GR32RegClass;
+ return &X86::GR32_TCRegClass;
+}
+
const TargetRegisterClass *
X86RegisterInfo::getCrossCopyRegClass(const TargetRegisterClass *RC) const {
if (RC == &X86::CCRRegClass) {
const TargetRegisterClass *
getCrossCopyRegClass(const TargetRegisterClass *RC) const override;
+ /// getGPRsForTailCall - Returns a register class with registers that can be
+ /// used in forming tail calls.
+ const TargetRegisterClass *
+ getGPRsForTailCall(const MachineFunction &MF) const;
+
unsigned getRegPressureLimit(const TargetRegisterClass *RC,
MachineFunction &MF) const override;
def GR64_TC : RegisterClass<"X86", [i64], 64, (add RAX, RCX, RDX, RSI, RDI,
R8, R9, R11, RIP)>;
def GR64_TCW64 : RegisterClass<"X86", [i64], 64, (add RAX, RCX, RDX,
- R8, R9, R11)>;
+ R8, R9, R10, R11)>;
// GR8_NOREX - GR8 registers which do not require a REX prefix.
def GR8_NOREX : RegisterClass<"X86", [i8], 8,
; X64: movq my_emutls_v_xyz@GOTPCREL(%rip), %rdi
; X64-NEXT: callq my_emutls_get_address@PLT
; X64-NEXT: movl (%rax), %eax
-; X64-NEXT: popq %rdx
+; X64-NEXT: popq %rcx
; X64-NEXT: retq
entry:
; X64: movq __emutls_v.i@GOTPCREL(%rip), %rdi
; X64-NEXT: callq __emutls_get_address@PLT
; X64-NEXT: movl (%rax), %eax
-; X64-NEXT: popq %rdx
+; X64-NEXT: popq %rcx
; X64-NEXT: retq
entry:
; X64: movl $my_emutls_v_xyz, %edi
; X64-NEXT: callq my_emutls_get_address
; X64-NEXT: movl (%rax), %eax
-; X64-NEXT: popq %rdx
+; X64-NEXT: popq %rcx
; X64-NEXT: retq
entry:
; X64: movl $__emutls_v.i1, %edi
; X64-NEXT: callq __emutls_get_address
; X64-NEXT: movl (%rax), %eax
-; X64-NEXT: popq %rdx
+; X64-NEXT: popq %rcx
; X64-NEXT: retq
entry:
; X64-LABEL: f2:
; X64: movl $__emutls_v.i1, %edi
; X64-NEXT: callq __emutls_get_address
-; X64-NEXT: popq %rdx
+; X64-NEXT: popq %rcx
; X64-NEXT: retq
entry:
; CHECK-LIBCALL-NEXT: movzwl (%rdi), %edi
; CHECK-LIBCALL-NEXT: callq __gnu_h2f_ieee
; CHECK-LIBCALL-NEXT: cvttss2si %xmm0, %rax
-; CHECK-LIBCALL-NEXT: popq %rdx
+; CHECK-LIBCALL-NEXT: popq %rcx
; CHECK-LIBCALL-NEXT: retq
; CHECK-F16C-NEXT: movswl (%rdi), [[REG0:%[a-z0-9]+]]
; CHECK-LIBCALL-NEXT: cvttss2si %xmm0, [[REG5:%[a-z0-9]+]]
; CHECK-LIBCALL-NEXT: ucomiss [[REG1]], %xmm0
; CHECK-LIBCALL-NEXT: cmovaeq [[REG4]], [[REG5]]
-; CHECK-LIBCALL-NEXT: popq %rdx
+; CHECK-LIBCALL-NEXT: popq %rcx
; CHECK-LIBCALL-NEXT: retq
; CHECK-F16C-NEXT: movswl (%rdi), [[REG0:%[a-z0-9]+]]
; CHECK: movq %rdi, (%rsp)
; CHECK: callq return_i1
; CHECK: movq (%rsp), %rax
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%alloca = alloca i32 addrspace(1)*, align 8
; CHECK: movq %rdi, (%rsp)
; CHECK: callq return_i1
; CHECK: xorl %eax, %eax
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%alloca = alloca i32 addrspace(1)*, align 8
; state arguments to the statepoint
; CHECK: pushq %rax
; CHECK: callq return_i1
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0)
; CHECK-LABEL: test_i32_return
; CHECK: pushq %rax
; CHECK: callq return_i32
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i32 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i32f(i64 0, i32 0, i32 ()* @return_i32, i32 0, i32 0, i32 0, i32 0)
; CHECK-LABEL: test_i32ptr_return
; CHECK: pushq %rax
; CHECK: callq return_i32ptr
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i32* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p0i32f(i64 0, i32 0, i32* ()* @return_i32ptr, i32 0, i32 0, i32 0, i32 0)
; CHECK: pushq %rax
; CHECK: callq return_i1
; CHECK-NEXT: .Ltmp11:
-; CHECK-NEXT: popq %rdx
+; CHECK-NEXT: popq %rcx
; CHECK-NEXT: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, i32 addrspace(1)* %a)
; A patchable variant of test_i1_return
; CHECK: pushq %rax
; CHECK: nopl
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 3, i1 ()*null, i32 0, i32 0, i32 0, i32 0)
; state arguments to the statepoint
; CHECK: pushq %rax
; CHECK: callq return_i1
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 1, i32 0, i32 0)
; CHECK-LABEL: test_i32_return
; CHECK: pushq %rax
; CHECK: callq return_i32
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i32 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i32f(i64 0, i32 0, i32 ()* @return_i32, i32 0, i32 1, i32 0, i32 0)
; CHECK-LABEL: test_i32ptr_return
; CHECK: pushq %rax
; CHECK: callq return_i32ptr
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i32* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p0i32f(i64 0, i32 0, i32* ()* @return_i32ptr, i32 0, i32 1, i32 0, i32 0)
; CHECK: pushq %rax
; CHECK: callq return_i1
; CHECK-NEXT: .Ltmp9:
-; CHECK-NEXT: popq %rdx
+; CHECK-NEXT: popq %rcx
; CHECK-NEXT: retq
entry:
%safepoint_token = tail call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 1, i32 0, i32 0, i32 addrspace(1)* %a)
; CHECK-LABEL: test_transition_args
; CHECK: pushq %rax
; CHECK: callq return_i32
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%val = alloca i32
; CHECK-LABEL: test_transition_args_2
; CHECK: pushq %rax
; CHECK: callq return_i32
-; CHECK: popq %rdx
+; CHECK: popq %rcx
; CHECK: retq
entry:
%val = alloca i32
; X32-SSE41-NEXT: .cfi_def_cfa_offset 8
; X32-SSE41-NEXT: pmovsxbw %xmm0, %xmm0
; X32-SSE41-NEXT: movd %xmm0, %eax
-; X32-SSE41-NEXT: popl %edx
+; X32-SSE41-NEXT: popl %ecx
; X32-SSE41-NEXT: retl
entry:
%Shuf = shufflevector <16 x i8> %A, <16 x i8> undef, <2 x i32> <i32 0, i32 1>
--- /dev/null
+; RUN: llc < %s -mtriple=x86_64-pc-win32-coreclr | FileCheck %s -check-prefix=WIN_X64
+; RUN: llc < %s -mtriple=x86_64-pc-linux | FileCheck %s -check-prefix=LINUX
+
+%Object = type <{ [0 x i64*]* }>
+
+define void @C1(%Object addrspace(1)* %param0) gc "coreclr" {
+entry:
+
+; WIN_X64: # BB#0:
+; WIN_X64: pushq %rax
+; LINUX: # BB#0: # %entry
+; LINUX: movq $0, -8(%rsp)
+
+ %this = alloca %Object addrspace(1)*
+ store %Object addrspace(1)* null, %Object addrspace(1)** %this
+ store %Object addrspace(1)* %param0, %Object addrspace(1)** %this
+ br label %0
+
+; <label>:0 ; preds = %entry
+ %1 = load %Object addrspace(1)*, %Object addrspace(1)** %this, align 8
+
+; WIN_X64: xorl %r8d, %r8d
+; WIN_X64: popq %rax
+; WIN_X64: rex64 jmp C2 # TAILCALL
+; LINUX: xorl %edx, %edx
+; LINUX: jmp C2 # TAILCALL
+
+ tail call void @C2(%Object addrspace(1)* %1, i32 0, %Object addrspace(1)* null)
+ ret void
+}
+
+declare void @C2(%Object addrspace(1)*, i32, %Object addrspace(1)*)
+
+; Function Attrs: nounwind
+declare void @llvm.localescape(...) #0
+
+attributes #0 = { nounwind }
+