Win64: Put a REX_W prefix on all TAILJMP* instructions

MSDN's x64 software conventions page says that this is one of the fixed
list of legal epilogues:
https://msdn.microsoft.com/en-us/library/tawsa7cb.aspx

Presumably this is how the unwinder distinguishes epilogue jumps from
in-function control flow.

Also normalize the way we place "## TAILCALL" comments on such jumps.

llvm-svn: 227611

2 files changed, 38 insertions, 2 deletions

diff --git a/llvm/test/CodeGen/X86/tail-call-win64.ll b/llvm/test/CodeGen/X86/tail-call-win64.ll
new file mode 100644
index 00000000000..23e9280e772
--- /dev/null
+++ b/llvm/test/CodeGen/X86/tail-call-win64.ll
@@ -0,0 +1,36 @@
+; RUN: llc -mtriple=x86_64-windows -show-mc-encoding < %s | FileCheck %s
+
+; The Win64 ABI wants tail jmps to use a REX_W prefix so it can distinguish
+; in-function jumps from function exiting jumps.
+
+define void @tail_jmp_reg(i32, i32, void ()* %fptr) {
+  tail call void ()* %fptr()
+  ret void
+}
+
+; Check that we merge the REX prefixes into 0x49 instead of 0x48, 0x41.
+
+; CHECK-LABEL: tail_jmp_reg:
+; CHECK: rex64 jmpq *%r8
+; CHECK: 	encoding: [0x49,0xff,0xe0]
+
+declare void @tail_tgt()
+
+define void @tail_jmp_imm() {
+  tail call void @tail_tgt()
+  ret void
+}
+
+; CHECK-LABEL: tail_jmp_imm:
+; CHECK: rex64 jmp tail_tgt
+
+@g_fptr = global void ()* @tail_tgt
+
+define void @tail_jmp_mem() {
+  %fptr = load void ()** @g_fptr
+  tail call void ()* %fptr()
+  ret void
+}
+
+; CHECK-LABEL: tail_jmp_mem:
+; CHECK: rex64 jmpq *g_fptr(%rip)
diff --git a/llvm/test/CodeGen/X86/tailcall-64.ll b/llvm/test/CodeGen/X86/tailcall-64.ll
index deab1dcc7eb..25d3802ab61 100644
--- a/llvm/test/CodeGen/X86/tailcall-64.ll
+++ b/llvm/test/CodeGen/X86/tailcall-64.ll
@@ -182,7 +182,7 @@ define { i64, i64 } @crash(i8* %this) {
 ; Check that we can fold an indexed load into a tail call instruction.
 ; CHECK: fold_indexed_load
 ; CHECK: leaq (%rsi,%rsi,4), %[[RAX:r..]]
-; CHECK: jmpq *16(%{{r..}},%[[RAX]],8)  # TAILCALL
+; CHECK: jmpq *16(%{{r..}},%[[RAX]],8)  ## TAILCALL
 %struct.funcs = type { i32 (i8*, i32*, i32)*, i32 (i8*)*, i32 (i8*)*, i32 (i8*, i32)*, i32 }
 @func_table = external global [0 x %struct.funcs]
 define void @fold_indexed_load(i8* %mbstr, i64 %idxprom) nounwind uwtable ssp {
@@ -207,7 +207,7 @@ entry:
 ; }
 ;
 ; CHECK-LABEL: rdar12282281
-; CHECK: jmpq *%r11 # TAILCALL
+; CHECK: jmpq *%r11 ## TAILCALL
 @funcs = external constant [0 x i32 (i8*, ...)*]
 
 define i32 @rdar12282281(i32 %n) nounwind uwtable ssp {