Automerge: [RISCV][Xqccmp] Correctly Parse/Disassemble pushfp (#133188)

In the `qc.cm.pushfp` instruction, it is like `cm.pushfp` except in one
important way - `qc.cm.pushfp {ra}, -N*16` is not a valid encoding,
because this would update `s0`/`fp`/`x8` without saving it.

This change now correctly rejects this variant of the instruction, both
during parsing and during disassembly. I also implemented validation for
immediates that represent register lists (both kinds), which may help to
catch bugs in the future.

Author: lenary

llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp

@@ -215,7 +215,14 @@ class RISCVAsmParser : public MCTargetAsmParser {
   ParseStatus parseGPRPair(OperandVector &Operands, bool IsRV64Inst);
   ParseStatus parseFRMArg(OperandVector &Operands);
   ParseStatus parseFenceArg(OperandVector &Operands);
-  ParseStatus parseReglist(OperandVector &Operands);
+  ParseStatus parseReglist(OperandVector &Operands) {
+    return parseRegListCommon(Operands, /*MustIncludeS0=*/false);
+  }
+  ParseStatus parseReglistS0(OperandVector &Operands) {
+    return parseRegListCommon(Operands, /*MustIncludeS0=*/true);
+  }
+  ParseStatus parseRegListCommon(OperandVector &Operands, bool MustIncludeS0);
+
   ParseStatus parseRegReg(OperandVector &Operands);
   ParseStatus parseRetval(OperandVector &Operands);
   ParseStatus parseZcmpStackAdj(OperandVector &Operands,
@@ -474,6 +481,9 @@ struct RISCVOperand final : public MCParsedAsmOperand {
   bool isSystemRegister() const { return Kind == KindTy::SystemRegister; }
   bool isRegReg() const { return Kind == KindTy::RegReg; }
   bool isRlist() const { return Kind == KindTy::Rlist; }
+  bool isRlistS0() const {
+    return Kind == KindTy::Rlist && Rlist.Val != RISCVZC::RA;
+  }
   bool isSpimm() const { return Kind == KindTy::Spimm; }
 
   bool isGPR() const {
@@ -2794,9 +2804,15 @@ ParseStatus RISCVAsmParser::parseRegReg(OperandVector &Operands) {
   return ParseStatus::Success;
 }
 
-ParseStatus RISCVAsmParser::parseReglist(OperandVector &Operands) {
+ParseStatus RISCVAsmParser::parseRegListCommon(OperandVector &Operands,
+                                               bool MustIncludeS0) {
   // Rlist: {ra [, s0[-sN]]}
   // XRlist: {x1 [, x8[-x9][, x18[-xN]]]}
+
+  // When MustIncludeS0 = true (not the default) (used for `qc.cm.pushfp`) which
+  // must include `fp`/`s0` in the list:
+  // Rlist: {ra, s0[-sN]}
+  // XRlist: {x1, x8[-x9][, x18[-xN]]}
   SMLoc S = getLoc();
 
   if (parseToken(AsmToken::LCurly, "register list must start with '{'"))
@@ -2814,8 +2830,20 @@ ParseStatus RISCVAsmParser::parseReglist(OperandVector &Operands) {
     return Error(getLoc(), "register list must start from 'ra' or 'x1'");
   getLexer().Lex();
 
-  // parse case like ,s0
-  if (parseOptionalToken(AsmToken::Comma)) {
+  bool SeenComma = parseOptionalToken(AsmToken::Comma);
+
+  // There are two choices here:
+  // - `s0` is not required (usual case), so only try to parse `s0` if there is
+  //   a comma
+  // - `s0` is required (qc.cm.pushfp), and so we must see the comma between
+  //   `ra` and `s0` and must always try to parse `s0`, below
+  if (MustIncludeS0 && !SeenComma) {
+    Error(getLoc(), "register list must include 's0' or 'x8'");
+    return ParseStatus::Failure;
+  }
+
+  // parse case like ,s0 (knowing the comma must be there if required)
+  if (SeenComma) {
     if (getLexer().isNot(AsmToken::Identifier))
       return Error(getLoc(), "invalid register");
     StringRef RegName = getLexer().getTok().getIdentifier();
@@ -2884,6 +2912,8 @@ ParseStatus RISCVAsmParser::parseReglist(OperandVector &Operands) {
 
   auto Encode = RISCVZC::encodeRlist(RegEnd, IsEABI);
   assert(Encode != RISCVZC::INVALID_RLIST);
+  if (MustIncludeS0)
+    assert(Encode != RISCVZC::RA);
   Operands.push_back(RISCVOperand::createRlist(Encode, S));
 
   return ParseStatus::Success;

llvm/lib/Target/RISCV/Disassembler/RISCVDisassembler.cpp

@@ -507,6 +507,9 @@ static DecodeStatus decodeXTHeadMemPair(MCInst &Inst, uint32_t Insn,
 static DecodeStatus decodeZcmpRlist(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder);
 
+static DecodeStatus decodeXqccmpRlistS0(MCInst &Inst, uint32_t Imm,
+                                        uint64_t Address, const void *Decoder);
+
 static DecodeStatus decodeZcmpSpimm(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder);
 
@@ -612,7 +615,15 @@ static DecodeStatus decodeXTHeadMemPair(MCInst &Inst, uint32_t Insn,
 
 static DecodeStatus decodeZcmpRlist(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder) {
-  if (Imm <= 3)
+  if (Imm < RISCVZC::RA)
+    return MCDisassembler::Fail;
+  Inst.addOperand(MCOperand::createImm(Imm));
+  return MCDisassembler::Success;
+}
+
+static DecodeStatus decodeXqccmpRlistS0(MCInst &Inst, uint32_t Imm,
+                                        uint64_t Address, const void *Decoder) {
+  if (Imm < RISCVZC::RA_S0)
     return MCDisassembler::Fail;
   Inst.addOperand(MCOperand::createImm(Imm));
   return MCDisassembler::Success;

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.cpp

@@ -242,12 +242,12 @@ float RISCVLoadFPImm::getFPImm(unsigned Imm) {
 
 void RISCVZC::printRlist(unsigned SlistEncode, raw_ostream &OS) {
   OS << "{ra";
-  if (SlistEncode > 4) {
+  if (SlistEncode > RISCVZC::RA) {
     OS << ", s0";
-    if (SlistEncode == 15)
+    if (SlistEncode == RISCVZC::RA_S0_S11)
       OS << "-s11";
-    else if (SlistEncode > 5 && SlistEncode <= 14)
-      OS << "-s" << (SlistEncode - 5);
+    else if (SlistEncode > RISCVZC::RA_S0 && SlistEncode <= RISCVZC::RA_S0_S11)
+      OS << "-s" << (SlistEncode - RISCVZC::RA_S0);
   }
   OS << "}";
 }

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

@@ -343,6 +343,8 @@ enum OperandType : unsigned {
   OPERAND_RVKRNUM_0_7,
   OPERAND_RVKRNUM_1_10,
   OPERAND_RVKRNUM_2_14,
+  OPERAND_RLIST,
+  OPERAND_RLIST_S0,
   OPERAND_SPIMM,
   // Operand is a 3-bit rounding mode, '111' indicates FRM register.
   // Represents 'frm' argument passing to floating-point operations.

llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCCodeEmitter.cpp

@@ -103,6 +103,10 @@ class RISCVMCCodeEmitter : public MCCodeEmitter {
   unsigned getRlistOpValue(const MCInst &MI, unsigned OpNo,
                            SmallVectorImpl<MCFixup> &Fixups,
                            const MCSubtargetInfo &STI) const;
+
+  unsigned getRlistS0OpValue(const MCInst &MI, unsigned OpNo,
+                             SmallVectorImpl<MCFixup> &Fixups,
+                             const MCSubtargetInfo &STI) const;
 };
 } // end anonymous namespace
 
@@ -616,5 +620,16 @@ unsigned RISCVMCCodeEmitter::getRlistOpValue(const MCInst &MI, unsigned OpNo,
   assert(Imm >= 4 && "EABI is currently not implemented");
   return Imm;
 }
+unsigned
+RISCVMCCodeEmitter::getRlistS0OpValue(const MCInst &MI, unsigned OpNo,
+                                      SmallVectorImpl<MCFixup> &Fixups,
+                                      const MCSubtargetInfo &STI) const {
+  const MCOperand &MO = MI.getOperand(OpNo);
+  assert(MO.isImm() && "Rlist operand must be immediate");
+  auto Imm = MO.getImm();
+  assert(Imm >= 4 && "EABI is currently not implemented");
+  assert(Imm != RISCVZC::RA && "Rlist operand must include s0");
+  return Imm;
+}
 
 #include "RISCVGenMCCodeEmitter.inc"

llvm/lib/Target/RISCV/RISCVInstrInfo.cpp

@@ -2679,6 +2679,12 @@ bool RISCVInstrInfo::verifyInstruction(const MachineInstr &MI,
         case RISCVOp::OPERAND_RVKRNUM_2_14:
           Ok = Imm >= 2 && Imm <= 14;
           break;
+        case RISCVOp::OPERAND_RLIST:
+          Ok = Imm >= RISCVZC::RA && Imm <= RISCVZC::RA_S0_S11;
+          break;
+        case RISCVOp::OPERAND_RLIST_S0:
+          Ok = Imm >= RISCVZC::RA_S0 && Imm <= RISCVZC::RA_S0_S11;
+          break;
         case RISCVOp::OPERAND_SPIMM:
           Ok = (Imm & 0xf) == 0;
           break;

llvm/lib/Target/RISCV/RISCVInstrInfoXqccmp.td

@@ -25,6 +25,30 @@
 // Operand and SDNode transformation definitions.
 //===----------------------------------------------------------------------===//
 
+def RlistS0AsmOperand : AsmOperandClass {
+  let Name = "RlistS0";
+  let ParserMethod = "parseReglistS0";
+  let RenderMethod = "addRlistOperands";
+  let DiagnosticType = "InvalidRlistS0";
+  let DiagnosticString = "operand must be {ra, s0[-sN]} or {x1, x8[-x9][, x18[-xN]]}";
+}
+
+def rlist_s0 : RISCVOp<OtherVT> {
+   let ParserMatchClass = RlistS0AsmOperand;
+   let PrintMethod = "printRlist";
+   let DecoderMethod = "decodeXqccmpRlistS0";
+   let EncoderMethod = "getRlistS0OpValue";
+   let MCOperandPredicate = [{
+    int64_t Imm;
+    if (!MCOp.evaluateAsConstantImm(Imm))
+      return false;
+    // 0~4 invalid for `qc.cm.pushfp`
+    return isUInt<4>(Imm) && Imm >= RISCVZC::RA_S0;
+  }];
+
+  string OperandType = "OPERAND_RLIST_S0";
+}
+
 //===----------------------------------------------------------------------===//
 // Instruction Formats
 //===----------------------------------------------------------------------===//
@@ -33,6 +57,20 @@
 // Instruction Class Templates
 //===----------------------------------------------------------------------===//
 
+class RVInstXqccmpCPPPFP<bits<5> funct5, string opcodestr,
+                         DAGOperand immtype = stackadj>
+    : RVInst16<(outs), (ins rlist_s0:$rlist, immtype:$stackadj),
+               opcodestr, "$rlist, $stackadj", [], InstFormatOther> {
+  bits<4> rlist;
+  bits<16> stackadj;
+
+  let Inst{1-0} = 0b10;
+  let Inst{3-2} = stackadj{5-4};
+  let Inst{7-4} = rlist;
+  let Inst{12-8} = funct5;
+  let Inst{15-13} = 0b101;
+}
+
 //===----------------------------------------------------------------------===//
 // Instructions
 //===----------------------------------------------------------------------===//
@@ -59,7 +97,7 @@ def QC_CM_PUSH : RVInstZcCPPP<0b11000, "qc.cm.push", negstackadj>,
                         ReadStoreData, ReadStoreData, ReadStoreData]>;
 
 let hasSideEffects = 0, mayLoad = 0, mayStore = 1, Uses = [X2], Defs = [X2, X8] in
-def QC_CM_PUSHFP : RVInstZcCPPP<0b11001, "qc.cm.pushfp", negstackadj>,
+def QC_CM_PUSHFP : RVInstXqccmpCPPPFP<0b11001, "qc.cm.pushfp", negstackadj>,
                    Sched<[WriteIALU, WriteIALU, ReadIALU, ReadStoreData, ReadStoreData,
                           ReadStoreData, ReadStoreData, ReadStoreData, ReadStoreData,
                           ReadStoreData, ReadStoreData, ReadStoreData, ReadStoreData,

llvm/lib/Target/RISCV/RISCVInstrInfoZc.td

@@ -58,7 +58,7 @@ def NegStackAdjAsmOperand : AsmOperandClass {
   let RenderMethod = "addSpimmOperands";
 }
 
-def rlist : Operand<OtherVT> {
+def rlist : RISCVOp<OtherVT> {
    let ParserMatchClass = RlistAsmOperand;
    let PrintMethod = "printRlist";
    let DecoderMethod = "decodeZcmpRlist";
@@ -68,8 +68,10 @@ def rlist : Operand<OtherVT> {
     if (!MCOp.evaluateAsConstantImm(Imm))
       return false;
     // 0~3 Reserved for EABI
-    return isUInt<4>(Imm) && Imm >= 4;
+    return isUInt<4>(Imm) && Imm >= RISCVZC::RA;
   }];
+
+  let OperandType = "OPERAND_RLIST";
 }
 
 def stackadj : RISCVOp<OtherVT> {

llvm/test/MC/Disassembler/RISCV/xqccmp-invalid-rlist.txt

@@ -0,0 +1,11 @@
+# RUN: not llvm-mc -disassemble -triple=riscv32 -mattr=+experimental-xqccmp %s \
+# RUN:   | FileCheck -check-prefixes=CHECK,CHECK-XQCCMP %s
+
+[0x00,0x00]
+# CHECK: unimp
+
+[0x42,0xb9]
+# CHECK-XQCCMP-NOT: qc.cm.pushfp {ra}, -{{[0-9]+}}
+
+[0x00,0x00]
+# CHECK: unimp

llvm/test/MC/RISCV/rv32xqccmp-invalid.s

@@ -33,3 +33,7 @@ qc.cm.pushfp {ra, s0}, -12
 
 # CHECK-ERROR: error: stack adjustment for register list must be a multiple of 16 bytes in the range [16, 64]
 qc.cm.pop {ra, s0-s1}, -40
+
+# CHECK-ERROR: error: register list must include 's0' or 'x8'
+qc.cm.pushfp {ra}, -16
+