[x86][MC]: Fix data16.

Author: daym

llvm/lib/Target/X86/AsmParser/X86AsmParser.cpp

@@ -3523,8 +3523,35 @@ bool X86AsmParser::parseInstruction(ParseInstructionInfo &Info, StringRef Name,
     PatchedName = Name;
 
   // Hacks to handle 'data16' and 'data32'
-  if (PatchedName == "data16" && is16BitMode()) {
-    return Error(NameLoc, "redundant data16 prefix");
+  if (PatchedName == "data16") {
+    if (is16BitMode())
+      return Error(NameLoc, "redundant data16 prefix");
+    if (getLexer().isNot(AsmToken::EndOfStatement)) {
+      StringRef Next = Parser.getTok().getString();
+      getLexer().Lex();
+      // data16 effectively changes the instruction suffix.
+      // TODO Generalize.
+      if (Next == "call")
+        Next = "callw";
+      if (Next == "ljmp")
+        Next = "ljmpw";
+      if (Next == "retf")
+        Next = "retfw";
+      if (Next == "lgdt") {
+        if (is64BitMode()) {
+          // Use the special lgdtq variant with OpSize16 flag.
+          // I think the CPU ignores the prefix anyway--but hey.
+          Next = "lgdtq16";
+        } else {
+          Next = "lgdtw";
+        }
+      }
+
+      Name = Next;
+      PatchedName = Name;
+      ForcedDataPrefix = X86::Is16Bit;
+      IsPrefix = false;
+    }
   }
   if (PatchedName == "data32") {
     if (is32BitMode())

llvm/lib/Target/X86/X86InstrAsmAlias.td

@@ -260,6 +260,9 @@ def : MnemonicAlias<"cqo",  "cqto", "att">;
 // In 64-bit mode lret maps to lretl; it is not ambiguous with lretq.
 def : MnemonicAlias<"lret", "lretw", "att">, Requires<[In16BitMode]>;
 def : MnemonicAlias<"lret", "lretl", "att">, Requires<[Not16BitMode]>;
+def : InstAlias<"lgdtw\t$src", (LGDT16m opaquemem:$src), 0>, Requires<[Not64BitMode]>;
+def : InstAlias<"lgdtq16\t$src", (LGDT64m_16 opaquemem:$src), 0>, Requires<[In64BitMode]>;
+def : InstAlias<"retfw", (LRET16), 0>;
 
 def : MnemonicAlias<"leavel", "leave", "att">, Requires<[Not64BitMode]>;
 def : MnemonicAlias<"leaveq", "leave", "att">, Requires<[In64BitMode]>;
@@ -723,6 +726,7 @@ def : InstAlias<"shrd{l}\t{$reg, $mem|$mem, $reg}", (SHRD32mrCL i32mem:$mem, GR3
 def : InstAlias<"shrd{q}\t{$reg, $mem|$mem, $reg}", (SHRD64mrCL i64mem:$mem, GR64:$reg), 0>;
 
 // test: We accept "testX <reg>, <mem>" and "testX <mem>, <reg>" as synonyms.
+def : InstAlias<"ljmpw\t$seg, $off", (FARJMP16i i16imm:$off, i16imm:$seg)>, Requires<[Not64BitMode]>;
 def : InstAlias<"test{b}\t{$mem, $val|$val, $mem}",
                 (TEST8mr  i8mem :$mem, GR8 :$val), 0>;
 def : InstAlias<"test{w}\t{$mem, $val|$val, $mem}",

llvm/lib/Target/X86/X86InstrSystem.td

@@ -427,6 +427,9 @@ def LGDT32m : I<0x01, MRM2m, (outs), (ins opaquemem:$src),
                 "lgdt{l|d}\t$src", []>, OpSize32, TB, Requires<[Not64BitMode]>;
 def LGDT64m : I<0x01, MRM2m, (outs), (ins opaquemem:$src),
                 "lgdt{q}\t$src", []>, TB, Requires<[In64BitMode]>;
+// Special variant with data16 prefix (OpSize16) for 64-bit mode
+def LGDT64m_16 : I<0x01, MRM2m, (outs), (ins opaquemem:$src),
+                 "lgdt{q}\t$src", []>, TB, OpSize16, Requires<[In64BitMode]>;
 def LIDT16m : I<0x01, MRM3m, (outs), (ins opaquemem:$src),
                 "lidtw\t$src", []>, TB, OpSize16, Requires<[Not64BitMode]>;
 def LIDT32m : I<0x01, MRM3m, (outs), (ins opaquemem:$src),

llvm/test/MC/X86/intel-syntax-32.s

@@ -1,6 +1,35 @@
-// RUN: llvm-mc -triple i686-unknown-unknown -x86-asm-syntax=intel %s | FileCheck %s
+// RUN: llvm-mc -triple i686-unknown-unknown -x86-asm-syntax=intel --show-encoding %s | FileCheck %s
 
 // CHECK: leaw	(%bp,%si), %ax
+// CHECK: encoding: [0x67,0x66,0x8d,0x02]
 lea ax, [bp+si]
 // CHECK: leaw	(%bp,%si), %ax
+// CHECK: encoding: [0x67,0x66,0x8d,0x02]
 lea ax, [si+bp]
+
+// CHECK: encoding: [0x66,0x6a,0x08]
+          data16 push 8
+
+// CHECK: encoding: [0x6a,0x08]
+          push 8
+
+// CHECK: encoding: [0x66,0xcb]
+          data16 retf
+
+// CHECK: encoding: [0xcb]
+          retf
+
+// CHECK: encoding: [0x66,0x9a,0xcd,0xab,0xce,0x7a]
+          data16 call 0x7ace, 0xabcd
+
+// CHECK: encoding: [0x9a,0xcd,0xab,0x00,0x00,0xce,0x7a]
+          call 0x7ace, 0xabcd
+
+// CHECK: encoding: [0xe8,A,A,A,A]
+          call a
+
+// CHECK: encoding: [0x66,0xea,0xcd,0xab,0xce,0x7a]
+          data16 ljmp 0x7ace, 0xabcd
+
+// CHECK: encoding: [0xea,0xcd,0xab,0x00,0x00,0xce,0x7a]
+          ljmp 0x7ace, 0xabcd