x86 interrupt calling convention: only save xmm registers if the target supports SSE

The existing code always saves the xmm registers for 64-bit targets even if the
target doesn't support SSE (which is common for kernels). Thus, the compiler
inserts movaps instructions which lead to CPU exceptions when an interrupt
handler is invoked.

This commit fixes this bug by returning a register set without xmm registers
from getCalleeSavedRegs and getCallPreservedMask for such targets.

Patch by Philipp Oppermann.

Differential Revision: https://reviews.llvm.org/D29959

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@295347 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86CallingConv.td b/lib/Target/X86/X86CallingConv.td
index cf7bc981b8a587ed771f8537cd58709391363340..6781d761a1c4fca81d0073959095900996aa5846 100644 (file)
--- a/lib/Target/X86/X86CallingConv.td
+++ b/lib/Target/X86/X86CallingConv.td
@@ -1074,6 +1074,8 @@ def CSR_32_AllRegs_AVX512 : CalleeSavedRegs<(add CSR_32_AllRegs,
                                                  (sequence "K%u", 0, 7))>;
 
 def CSR_64_AllRegs     : CalleeSavedRegs<(add CSR_64_MostRegs, RAX)>;
+def CSR_64_AllRegs_NoSSE : CalleeSavedRegs<(add RAX, RBX, RCX, RDX, RSI, RDI, R8, R9,
+                                                R10, R11, R12, R13, R14, R15, RBP)>;
 def CSR_64_AllRegs_AVX : CalleeSavedRegs<(sub (add CSR_64_MostRegs, RAX,
                                                    (sequence "YMM%u", 0, 15)),
                                               (sequence "XMM%u", 0, 15))>;

diff --git a/lib/Target/X86/X86RegisterInfo.cpp b/lib/Target/X86/X86RegisterInfo.cpp
index 65f438f94b042804d078542858a6ac67ec077ae5..9e886e13ac599af6606a677cfbf1626949efb22c 100644 (file)
--- a/lib/Target/X86/X86RegisterInfo.cpp
+++ b/lib/Target/X86/X86RegisterInfo.cpp
@@ -337,7 +337,9 @@ X86RegisterInfo::getCalleeSavedRegs(const MachineFunction *MF) const {
         return CSR_64_AllRegs_AVX512_SaveList;
       if (HasAVX)
         return CSR_64_AllRegs_AVX_SaveList;
-      return CSR_64_AllRegs_SaveList;
+      if (HasSSE)
+        return CSR_64_AllRegs_SaveList;
+      return CSR_64_AllRegs_NoSSE_SaveList;
     } else {
       if (HasAVX512)
         return CSR_32_AllRegs_AVX512_SaveList;
@@ -447,7 +449,9 @@ X86RegisterInfo::getCallPreservedMask(const MachineFunction &MF,
         return CSR_64_AllRegs_AVX512_RegMask;
       if (HasAVX)
         return CSR_64_AllRegs_AVX_RegMask;
-      return CSR_64_AllRegs_RegMask;
+      if (HasSSE)
+        return CSR_64_AllRegs_RegMask;
+      return CSR_64_AllRegs_NoSSE_RegMask;
     } else {
       if (HasAVX512)
         return CSR_32_AllRegs_AVX512_RegMask;

diff --git a/test/CodeGen/X86/x86-64-intrcc-nosse.ll b/test/CodeGen/X86/x86-64-intrcc-nosse.ll
new file mode 100644 (file)
index 0000000..0bb4e47
--- /dev/null
+++ b/test/CodeGen/X86/x86-64-intrcc-nosse.ll
@@ -0,0 +1,19 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=x86_64-unknown-unknown -mattr=-sse < %s | FileCheck %s
+
+%struct.interrupt_frame = type { i64, i64, i64, i64, i64 }
+
+@llvm.used = appending global [1 x i8*] [i8* bitcast (void (%struct.interrupt_frame*, i64)* @test_isr_sse_clobbers to i8*)], section "llvm.metadata"
+
+; Clobbered SSE must not be saved when the target doesn't support SSE
+define x86_intrcc void @test_isr_sse_clobbers(%struct.interrupt_frame* %frame, i64 %ecode) {
+  ; CHECK-LABEL: test_isr_sse_clobbers:
+  ; CHECK:       # BB#0:
+  ; CHECK-NEXT:    cld
+  ; CHECK-NEXT:    #APP
+  ; CHECK-NEXT:    #NO_APP
+  ; CHECK-NEXT:    addq $8, %rsp
+  ; CHECK-NEXT:    iretq
+  call void asm sideeffect "", "~{xmm0},~{xmm6}"()
+  ret void
+}