From 02265382929b0275d7b7b334eab5e2fd34e1b9fe Mon Sep 17 00:00:00 2001
From: Mihai Popa <mihail.popa@gmail.com>
Date: Mon, 22 Jul 2013 15:49:36 +0000
Subject: [PATCH] This adds range checking for "ldr Rn, [pc, #imm]" Thumb
 instructions. With this patch:

1. ldr.n is recognized as mnemonic for the short encoding
2. ldr.w is recognized as menmonic for the long encoding
3. ldr will map to either short or long encodings depending on the size of the offset

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186831 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/ARM/ARMInstrThumb.td               | 43 +++++++++----------
 lib/Target/ARM/ARMInstrThumb2.td              |  2 +-
 lib/Target/ARM/AsmParser/ARMAsmParser.cpp     | 40 +++++++++++++++++
 lib/Target/ARM/MCTargetDesc/ARMAsmBackend.cpp |  2 +-
 test/CodeGen/ARM/fast-isel-pic.ll             |  8 ++--
 test/CodeGen/ARM/indirectbr.ll                |  6 +--
 test/CodeGen/ARM/load-global.ll               |  2 +-
 test/CodeGen/ARM/machine-licm.ll              |  2 +-
 test/CodeGen/Thumb/large-stack.ll             |  6 +--
 test/MC/ARM/basic-thumb2-instructions.s       | 21 ++++++++-
 10 files changed, 93 insertions(+), 39 deletions(-)

diff --git a/lib/Target/ARM/ARMInstrThumb.td b/lib/Target/ARM/ARMInstrThumb.td
index a0edaba04e8..e7218c66586 100644
--- a/lib/Target/ARM/ARMInstrThumb.td
+++ b/lib/Target/ARM/ARMInstrThumb.td
@@ -100,6 +100,13 @@ class OperandUnsignedOffset_b8s2 : AsmOperandClass {
 
 def UnsignedOffset_b8s2 : OperandUnsignedOffset_b8s2;
 
+// thumb style PC relative operand. signed, 8 bits magnitude,
+// two bits shift. can be represented as either [pc, #imm], #imm,
+// or relocatable expression...
+def ThumbMemPC : AsmOperandClass {
+  let Name = "ThumbMemPC";
+}
+
 let OperandType = "OPERAND_PCREL" in {
 def t_brtarget : Operand<OtherVT> {
   let EncoderMethod = "getThumbBRTargetOpValue";
@@ -132,6 +139,15 @@ def t_blxtarget : Operand<i32> {
   let EncoderMethod = "getThumbBLXTargetOpValue";
   let DecoderMethod = "DecodeThumbBLXOffset";
 }
+
+// t_addrmode_pc := <label> => pc + imm8 * 4
+//
+def t_addrmode_pc : Operand<i32> {
+  let EncoderMethod = "getAddrModePCOpValue";
+  let DecoderMethod = "DecodeThumbAddrModePC";
+  let PrintMethod = "printThumbLdrLabelOperand";
+  let ParserMatchClass = ThumbMemPC;
+}
 }
 
 // t_addrmode_rr := reg + reg
@@ -228,14 +244,6 @@ def t_addrmode_sp : Operand<i32>,
   let MIOperandInfo = (ops GPR:$base, i32imm:$offsimm);
 }
 
-// t_addrmode_pc := <label> => pc + imm8 * 4
-//
-def t_addrmode_pc : Operand<i32> {
-  let EncoderMethod = "getAddrModePCOpValue";
-  let DecoderMethod = "DecodeThumbAddrModePC";
-  let PrintMethod = "printThumbLdrLabelOperand";
-}
-
 //===----------------------------------------------------------------------===//
 //  Miscellaneous Instructions.
 //
@@ -640,11 +648,9 @@ def tLDRspi : T1pIs<(outs tGPR:$Rt), (ins t_addrmode_sp:$addr), IIC_iLoad_i,
   let Inst{7-0} = addr;
 }
 
-// Load tconstpool
-// FIXME: Use ldr.n to work around a darwin assembler bug.
-let canFoldAsLoad = 1, isReMaterializable = 1, isCodeGenOnly = 1 in
+let canFoldAsLoad = 1, isReMaterializable = 1 in
 def tLDRpci : T1pIs<(outs tGPR:$Rt), (ins t_addrmode_pc:$addr), IIC_iLoad_i,
-                  "ldr", ".n\t$Rt, $addr",
+                  "ldr", "\t$Rt, $addr",
                   [(set tGPR:$Rt, (load (ARMWrapper tconstpool:$addr)))]>,
               T1Encoding<{0,1,0,0,1,?}> {
   // A6.2 & A8.6.59
@@ -654,17 +660,8 @@ def tLDRpci : T1pIs<(outs tGPR:$Rt), (ins t_addrmode_pc:$addr), IIC_iLoad_i,
   let Inst{7-0}  = addr;
 }
 
-// FIXME: Remove this entry when the above ldr.n workaround is fixed.
-// For assembly/disassembly use only.
-def tLDRpciASM : T1pIs<(outs tGPR:$Rt), (ins t_addrmode_pc:$addr), IIC_iLoad_i,
-                       "ldr", "\t$Rt, $addr", []>,
-                 T1Encoding<{0,1,0,0,1,?}> {
-  // A6.2 & A8.6.59
-  bits<3> Rt;
-  bits<8> addr;
-  let Inst{10-8} = Rt;
-  let Inst{7-0}  = addr;
-}
+def : tInstAlias<"ldr${p}.n $Rt, $addr", 
+                 (tLDRpci tGPR:$Rt, t_addrmode_pc:$addr, pred:$p), 0>;
 
 // A8.6.194 & A8.6.192
 defm tSTR  : thumb_st_rr_ri_enc<0b000, 0b0110, t_addrmode_rrs4,
diff --git a/lib/Target/ARM/ARMInstrThumb2.td b/lib/Target/ARM/ARMInstrThumb2.td
index 5a425c69228..8e5e8c1735a 100644
--- a/lib/Target/ARM/ARMInstrThumb2.td
+++ b/lib/Target/ARM/ARMInstrThumb2.td
@@ -4399,7 +4399,7 @@ def t2LDRSHpcrel  : t2AsmPseudo<"ldrsh${p} $Rt, $addr",
                          (ins GPRnopc:$Rt, t2ldr_pcrel_imm12:$addr, pred:$p)>;
     // Version w/ the .w suffix.
 def : t2InstAlias<"ldr${p}.w $Rt, $addr",
-                  (t2LDRpcrel GPRnopc:$Rt, t2ldr_pcrel_imm12:$addr, pred:$p)>;
+                  (t2LDRpcrel GPRnopc:$Rt, t2ldr_pcrel_imm12:$addr, pred:$p), 0>;
 def : t2InstAlias<"ldrb${p}.w $Rt, $addr",
                   (t2LDRBpcrel GPRnopc:$Rt, t2ldr_pcrel_imm12:$addr, pred:$p)>;
 def : t2InstAlias<"ldrh${p}.w $Rt, $addr",
diff --git a/lib/Target/ARM/AsmParser/ARMAsmParser.cpp b/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
index fc1faffd0bb..6d885a0d918 100644
--- a/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
+++ b/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
@@ -609,6 +609,26 @@ public:
     }
     return false;
   }
+  // checks whether this operand is a memory operand computed as an offset
+  // applied to PC. the offset may have 8 bits of magnitude and is represented
+  // with two bits of shift. textually it may be either [pc, #imm], #imm or 
+  // relocable expression...
+  bool isThumbMemPC() const {
+    int64_t Val = 0;
+    if (isImm()) {
+      if (isa<MCSymbolRefExpr>(Imm.Val)) return true;
+      const MCConstantExpr *CE = dyn_cast<MCConstantExpr>(Imm.Val);
+      if (!CE) return false;
+      Val = CE->getValue();
+    }
+    else if (isMem()) {
+      if(!Memory.OffsetImm || Memory.OffsetRegNum) return false;
+      if(Memory.BaseRegNum != ARM::PC) return false;
+      Val = Memory.OffsetImm->getValue();
+    }
+    else return false;
+    return ((Val % 4) == 0) && (Val >= -1020) && (Val <= 1020);
+  }
   bool isFPImm() const {
     if (!isImm()) return false;
     const MCConstantExpr *CE = dyn_cast<MCConstantExpr>(getImm());
@@ -1698,6 +1718,26 @@ public:
     Inst.addOperand(MCOperand::CreateExpr(SR));
   }
 
+  void addThumbMemPCOperands(MCInst &Inst, unsigned N) const {
+    assert(N == 1 && "Invalid number of operands!");
+    if (isImm()) {
+      const MCConstantExpr *CE = dyn_cast<MCConstantExpr>(getImm());
+      if (CE) {
+        Inst.addOperand(MCOperand::CreateImm(CE->getValue()));
+        return;
+      }
+
+      const MCSymbolRefExpr *SR = dyn_cast<MCSymbolRefExpr>(Imm.Val);
+      assert(SR && "Unknown value type!");
+      Inst.addOperand(MCOperand::CreateExpr(SR));
+      return;
+    }
+
+    assert(isMem()  && "Unknown value type!");
+    assert(isa<MCConstantExpr>(Memory.OffsetImm) && "Unknown value type!");
+    Inst.addOperand(MCOperand::CreateImm(Memory.OffsetImm->getValue()));
+  }
+
   void addARMSOImmNotOperands(MCInst &Inst, unsigned N) const {
     assert(N == 1 && "Invalid number of operands!");
     // The operand is actually a so_imm, but we have its bitwise
diff --git a/lib/Target/ARM/MCTargetDesc/ARMAsmBackend.cpp b/lib/Target/ARM/MCTargetDesc/ARMAsmBackend.cpp
index 8baa3a6ce6a..b1e25d8b89e 100644
--- a/lib/Target/ARM/MCTargetDesc/ARMAsmBackend.cpp
+++ b/lib/Target/ARM/MCTargetDesc/ARMAsmBackend.cpp
@@ -152,7 +152,7 @@ static unsigned getRelaxedOpcode(unsigned Op) {
   switch (Op) {
   default: return Op;
   case ARM::tBcc:       return ARM::t2Bcc;
-  case ARM::tLDRpciASM: return ARM::t2LDRpci;
+  case ARM::tLDRpci:    return ARM::t2LDRpci;
   case ARM::tADR:       return ARM::t2ADR;
   case ARM::tB:         return ARM::t2B;
   }
diff --git a/test/CodeGen/ARM/fast-isel-pic.ll b/test/CodeGen/ARM/fast-isel-pic.ll
index 0589f9ada8f..ad0f15966f7 100644
--- a/test/CodeGen/ARM/fast-isel-pic.ll
+++ b/test/CodeGen/ARM/fast-isel-pic.ll
@@ -13,8 +13,8 @@ entry:
 ; THUMB: movt [[reg0]],
 ; THUMB: add  [[reg0]], pc
 ; THUMB-ELF: LoadGV
-; THUMB-ELF: ldr.n r[[reg0:[0-9]+]],
-; THUMB-ELF: ldr.n r[[reg1:[0-9]+]],
+; THUMB-ELF: ldr r[[reg0:[0-9]+]],
+; THUMB-ELF: ldr r[[reg1:[0-9]+]],
 ; THUMB-ELF: ldr r[[reg0]], [r[[reg0]], r[[reg1]]]
 ; ARM: LoadGV
 ; ARM: ldr [[reg1:r[0-9]+]],
@@ -41,8 +41,8 @@ entry:
 ; THUMB: add  r[[reg3]], pc
 ; THUMB: ldr  r[[reg3]], [r[[reg3]]]
 ; THUMB-ELF: LoadIndirectSymbol
-; THUMB-ELF: ldr.n r[[reg3:[0-9]+]],
-; THUMB-ELF: ldr.n r[[reg4:[0-9]+]],
+; THUMB-ELF: ldr r[[reg3:[0-9]+]],
+; THUMB-ELF: ldr r[[reg4:[0-9]+]],
 ; THUMB-ELF: ldr r[[reg3]], [r[[reg3]], r[[reg4]]]
 ; ARM: LoadIndirectSymbol
 ; ARM: ldr [[reg4:r[0-9]+]],
diff --git a/test/CodeGen/ARM/indirectbr.ll b/test/CodeGen/ARM/indirectbr.ll
index 6fed8c492f9..99e84a6a625 100644
--- a/test/CodeGen/ARM/indirectbr.ll
+++ b/test/CodeGen/ARM/indirectbr.ll
@@ -51,12 +51,12 @@ L1:                                               ; preds = %L2, %bb2
 ; ARM: ldr [[R1:r[0-9]+]], LCPI
 ; ARM: add [[R1b:r[0-9]+]], pc, [[R1]]
 ; ARM: str [[R1b]]
-; THUMB: ldr.n
+; THUMB: ldr
 ; THUMB: add
-; THUMB: ldr.n [[R2:r[0-9]+]], LCPI
+; THUMB: ldr [[R2:r[0-9]+]], LCPI
 ; THUMB: add [[R2]], pc
 ; THUMB: str [[R2]]
-; THUMB2: ldr.n [[R2:r[0-9]+]], LCPI
+; THUMB2: ldr [[R2:r[0-9]+]], LCPI
 ; THUMB2-NEXT: str{{(.w)?}} [[R2]]
   store i8* blockaddress(@foo, %L5), i8** @nextaddr, align 4
   ret i32 %res.3
diff --git a/test/CodeGen/ARM/load-global.ll b/test/CodeGen/ARM/load-global.ll
index 15a415df731..00ca2e8b1b7 100644
--- a/test/CodeGen/ARM/load-global.ll
+++ b/test/CodeGen/ARM/load-global.ll
@@ -26,7 +26,7 @@ define i32 @test1() {
 ; PIC: .long L_G$non_lazy_ptr-(LPC0_0+8)
 
 ; PIC_T: _test1
-; PIC_T: ldr.n r0, LCPI0_0
+; PIC_T: ldr r0, LCPI0_0
 ; PIC_T: add r0, pc
 ; PIC_T: ldr r0, [r0]
 ; PIC_T: ldr r0, [r0]
diff --git a/test/CodeGen/ARM/machine-licm.ll b/test/CodeGen/ARM/machine-licm.ll
index 87aaacfb55a..fc9b22614d6 100644
--- a/test/CodeGen/ARM/machine-licm.ll
+++ b/test/CodeGen/ARM/machine-licm.ll
@@ -40,7 +40,7 @@ bb.nph:                                           ; preds = %entry
 ; ARM: .section
 
 ; THUMB: BB#1
-; THUMB: ldr.n r2, LCPI0_0
+; THUMB: ldr r2, LCPI0_0
 ; THUMB: add r2, pc
 ; THUMB: ldr r{{[0-9]+}}, [r2]
 ; THUMB: LBB0_2
diff --git a/test/CodeGen/Thumb/large-stack.ll b/test/CodeGen/Thumb/large-stack.ll
index 6fa623123ad..fb6daa47865 100644
--- a/test/CodeGen/Thumb/large-stack.ll
+++ b/test/CodeGen/Thumb/large-stack.ll
@@ -10,7 +10,7 @@ define void @test1() {
 
 define void @test2() {
 ; CHECK-LABEL: test2:
-; CHECK: ldr.n r0, LCPI
+; CHECK: ldr r0, LCPI
 ; CHECK: add sp, r0
 ; CHECK: subs r4, r7, #4
 ; CHECK: mov sp, r4
@@ -20,9 +20,9 @@ define void @test2() {
 
 define i32 @test3() {
 ; CHECK-LABEL: test3:
-; CHECK: ldr.n r1, LCPI
+; CHECK: ldr r1, LCPI
 ; CHECK: add sp, r1
-; CHECK: ldr.n r1, LCPI
+; CHECK: ldr r1, LCPI
 ; CHECK: add r1, sp
 ; CHECK: subs r4, r7, #4
 ; CHECK: mov sp, r4
diff --git a/test/MC/ARM/basic-thumb2-instructions.s b/test/MC/ARM/basic-thumb2-instructions.s
index 2bd9a133384..43e03c9833a 100644
--- a/test/MC/ARM/basic-thumb2-instructions.s
+++ b/test/MC/ARM/basic-thumb2-instructions.s
@@ -772,6 +772,23 @@ _func:
 @ CHECK: ldr.w	lr, _strcmp-4           @ encoding: [0x5f'A',0xf8'A',A,0xe0'A']
 @ CHECK: @   fixup A - offset: 0, value: _strcmp-4, kind: fixup_t2_ldst_pcrel_12
 
+        ldr r4, [pc, #1020]
+        ldr r3, [pc, #-1020]
+        ldr r6, [pc, #1024]
+        ldr r0, [pc, #-1024]
+        ldr r2, [pc, #4095]
+        ldr r1, [pc, #-4095]
+        ldr.n r8, [pc, #132]
+        ldr.w r8, [pc, #132]
+
+@ CHECK: ldr	r4, [pc, #1020]       @ encoding: [0xff,0x4c]
+@ CHECK: ldr	r3, [pc, #-1020]      @ encoding: [0x01,0x4b]
+@ CHECK: ldr.w	r6, [pc, #1024]       @ encoding: [0xdf,0xf8,0x00,0x64]
+@ CHECK: ldr.w	r0, [pc, #-1024]      @ encoding: [0x5f,0xf8,0x00,0x04]
+@ CHECK: ldr.w	r2, [pc, #4095]       @ encoding: [0xdf,0xf8,0xff,0x2f]
+@ CHECK: ldr.w	r1, [pc, #-4095]      @ encoding: [0x5f,0xf8,0xff,0x1f]
+@ CHECK: ldr	r8, [pc, #132]        @ encoding: [0x21,0x48]
+@ CHECK: ldr.w	r8, [pc, #132]        @ encoding: [0xdf,0xf8,0x84,0x80]
 
 @------------------------------------------------------------------------------
 @ LDR(register)
@@ -3569,5 +3586,5 @@ _func:
 @ CHECK: ldrsh.w r11, [pc, #-22]        @ encoding: [0x3f,0xf9,0x16,0xb0]
 
 @ rdar://12596361
-        ldr r1, [pc, #12]
-@ CHECK: ldr.n r1, [pc, #12]        @ encoding: [0x03,0x49]
+         ldr r1, [pc, #12]
+@ CHECK: ldr r1, [pc, #12]              @ encoding: [0x03,0x49]
-- 
2.34.1