X-Git-Url: http://plrg.eecs.uci.edu/git/?p=oota-llvm.git;a=blobdiff_plain;f=lib%2FTarget%2FARM%2FARMInstrVFP.td;h=63e7940bb14ee7cee5ed91182c0ab9ccb7271ac8;hp=1c07394f3983b57b848490c8fc68b5080fb0c694;hb=f1f1b483b6fd0324c325ef50cdc60d28bf38138c;hpb=4b4e62219be91839091f9e35d8accf877f925d81

diff --git a/lib/Target/ARM/ARMInstrVFP.td b/lib/Target/ARM/ARMInstrVFP.td
index 1c07394f398..63e7940bb14 100644
--- a/lib/Target/ARM/ARMInstrVFP.td
+++ b/lib/Target/ARM/ARMInstrVFP.td
@@ -1,4 +1,4 @@
-//===- ARMInstrVFP.td - VFP support for ARM ----------------*- tablegen -*-===//
+//===-- ARMInstrVFP.td - VFP support for ARM ---------------*- tablegen -*-===//
 //
 //                     The LLVM Compiler Infrastructure
 //
@@ -11,22 +11,15 @@
 //
 //===----------------------------------------------------------------------===//
 
-def SDT_FTOI    : SDTypeProfile<1, 1, [SDTCisVT<0, f32>, SDTCisFP<1>]>;
-def SDT_ITOF    : SDTypeProfile<1, 1, [SDTCisFP<0>, SDTCisVT<1, f32>]>;
 def SDT_CMPFP0  : SDTypeProfile<0, 1, [SDTCisFP<0>]>;
 def SDT_VMOVDRR : SDTypeProfile<1, 2, [SDTCisVT<0, f64>, SDTCisVT<1, i32>,
                                        SDTCisSameAs<1, 2>]>;
 
-def arm_ftoui  : SDNode<"ARMISD::FTOUI",   SDT_FTOI>;
-def arm_ftosi  : SDNode<"ARMISD::FTOSI",   SDT_FTOI>;
-def arm_sitof  : SDNode<"ARMISD::SITOF",   SDT_ITOF>;
-def arm_uitof  : SDNode<"ARMISD::UITOF",   SDT_ITOF>;
 def arm_fmstat : SDNode<"ARMISD::FMSTAT",  SDTNone, [SDNPInGlue, SDNPOutGlue]>;
 def arm_cmpfp  : SDNode<"ARMISD::CMPFP",   SDT_ARMCmp, [SDNPOutGlue]>;
 def arm_cmpfp0 : SDNode<"ARMISD::CMPFPw0", SDT_CMPFP0, [SDNPOutGlue]>;
 def arm_fmdrr  : SDNode<"ARMISD::VMOVDRR", SDT_VMOVDRR>;
 
-
 //===----------------------------------------------------------------------===//
 // Operand Definitions.
 //
@@ -43,7 +36,7 @@ def vfp_f32imm : Operand<f32>,
     }], SDNodeXForm<fpimm, [{
       APFloat InVal = N->getValueAPF();
       uint32_t enc = ARM_AM::getFP32Imm(InVal);
-      return CurDAG->getTargetConstant(enc, MVT::i32);
+      return CurDAG->getTargetConstant(enc, SDLoc(N), MVT::i32);
     }]>> {
   let PrintMethod = "printFPImmOperand";
   let ParserMatchClass = FPImmOperand;
@@ -55,12 +48,21 @@ def vfp_f64imm : Operand<f64>,
     }], SDNodeXForm<fpimm, [{
       APFloat InVal = N->getValueAPF();
       uint32_t enc = ARM_AM::getFP64Imm(InVal);
-      return CurDAG->getTargetConstant(enc, MVT::i32);
+      return CurDAG->getTargetConstant(enc, SDLoc(N), MVT::i32);
     }]>> {
   let PrintMethod = "printFPImmOperand";
   let ParserMatchClass = FPImmOperand;
 }
 
+def alignedload32 : PatFrag<(ops node:$ptr), (load node:$ptr), [{
+  return cast<LoadSDNode>(N)->getAlignment() >= 4;
+}]>;
+
+def alignedstore32 : PatFrag<(ops node:$val, node:$ptr),
+                             (store node:$val, node:$ptr), [{
+  return cast<StoreSDNode>(N)->getAlignment() >= 4;
+}]>;
+
 // The VCVT to/from fixed-point instructions encode the 'fbits' operand
 // (the number of fixed bits) differently than it appears in the assembly
 // source. It's encoded as "Size - fbits" where Size is the size of the
@@ -86,11 +88,11 @@ let canFoldAsLoad = 1, isReMaterializable = 1 in {
 
 def VLDRD : ADI5<0b1101, 0b01, (outs DPR:$Dd), (ins addrmode5:$addr),
                  IIC_fpLoad64, "vldr", "\t$Dd, $addr",
-                 [(set DPR:$Dd, (f64 (load addrmode5:$addr)))]>;
+                 [(set DPR:$Dd, (f64 (alignedload32 addrmode5:$addr)))]>;
 
 def VLDRS : ASI5<0b1101, 0b01, (outs SPR:$Sd), (ins addrmode5:$addr),
                  IIC_fpLoad32, "vldr", "\t$Sd, $addr",
-                 [(set SPR:$Sd, (load addrmode5:$addr))]> {
+                 [(set SPR:$Sd, (alignedload32 addrmode5:$addr))]> {
   // Some single precision VFP instructions may be executed on both NEON and VFP
   // pipelines.
   let D = VFPNeonDomain;
@@ -100,11 +102,11 @@ def VLDRS : ASI5<0b1101, 0b01, (outs SPR:$Sd), (ins addrmode5:$addr),
 
 def VSTRD : ADI5<0b1101, 0b00, (outs), (ins DPR:$Dd, addrmode5:$addr),
                  IIC_fpStore64, "vstr", "\t$Dd, $addr",
-                 [(store (f64 DPR:$Dd), addrmode5:$addr)]>;
+                 [(alignedstore32 (f64 DPR:$Dd), addrmode5:$addr)]>;
 
 def VSTRS : ASI5<0b1101, 0b00, (outs), (ins SPR:$Sd, addrmode5:$addr),
                  IIC_fpStore32, "vstr", "\t$Sd, $addr",
-                 [(store SPR:$Sd, addrmode5:$addr)]> {
+                 [(alignedstore32 SPR:$Sd, addrmode5:$addr)]> {
   // Some single precision VFP instructions may be executed on both NEON and VFP
   // pipelines.
   let D = VFPNeonDomain;
@@ -185,19 +187,40 @@ multiclass vfp_ldst_mult<string asm, bit L_bit,
   }
 }
 
-let neverHasSideEffects = 1 in {
+let hasSideEffects = 0 in {
 
 let mayLoad = 1, hasExtraDefRegAllocReq = 1 in
 defm VLDM : vfp_ldst_mult<"vldm", 1, IIC_fpLoad_m, IIC_fpLoad_mu>;
 
 let mayStore = 1, hasExtraSrcRegAllocReq = 1 in
-defm VSTM : vfp_ldst_mult<"vstm", 0, IIC_fpLoad_m, IIC_fpLoad_mu>;
+defm VSTM : vfp_ldst_mult<"vstm", 0, IIC_fpStore_m, IIC_fpStore_mu>;
 
-} // neverHasSideEffects
+} // hasSideEffects
 
 def : MnemonicAlias<"vldm", "vldmia">;
 def : MnemonicAlias<"vstm", "vstmia">;
 
+// FLDM/FSTM - Load / Store multiple single / double precision registers for
+// pre-ARMv6 cores.
+// These instructions are deprecated!
+def : VFP2MnemonicAlias<"fldmias", "vldmia">;
+def : VFP2MnemonicAlias<"fldmdbs", "vldmdb">;
+def : VFP2MnemonicAlias<"fldmeas", "vldmdb">;
+def : VFP2MnemonicAlias<"fldmfds", "vldmia">;
+def : VFP2MnemonicAlias<"fldmiad", "vldmia">;
+def : VFP2MnemonicAlias<"fldmdbd", "vldmdb">;
+def : VFP2MnemonicAlias<"fldmead", "vldmdb">;
+def : VFP2MnemonicAlias<"fldmfdd", "vldmia">;
+
+def : VFP2MnemonicAlias<"fstmias", "vstmia">;
+def : VFP2MnemonicAlias<"fstmdbs", "vstmdb">;
+def : VFP2MnemonicAlias<"fstmeas", "vstmia">;
+def : VFP2MnemonicAlias<"fstmfds", "vstmdb">;
+def : VFP2MnemonicAlias<"fstmiad", "vstmia">;
+def : VFP2MnemonicAlias<"fstmdbd", "vstmdb">;
+def : VFP2MnemonicAlias<"fstmead", "vstmia">;
+def : VFP2MnemonicAlias<"fstmfdd", "vstmdb">;
+
 def : InstAlias<"vpush${p} $r", (VSTMDDB_UPD SP, pred:$p, dpr_reglist:$r)>,
                 Requires<[HasVFP2]>;
 def : InstAlias<"vpush${p} $r", (VSTMSDB_UPD SP, pred:$p, spr_reglist:$r)>,
@@ -206,18 +229,63 @@ def : InstAlias<"vpop${p} $r",  (VLDMDIA_UPD SP, pred:$p, dpr_reglist:$r)>,
                 Requires<[HasVFP2]>;
 def : InstAlias<"vpop${p} $r",  (VLDMSIA_UPD SP, pred:$p, spr_reglist:$r)>,
                 Requires<[HasVFP2]>;
+defm : VFPDTAnyInstAlias<"vpush${p}", "$r",
+                         (VSTMSDB_UPD SP, pred:$p, spr_reglist:$r)>;
+defm : VFPDTAnyInstAlias<"vpush${p}", "$r",
+                         (VSTMDDB_UPD SP, pred:$p, dpr_reglist:$r)>;
+defm : VFPDTAnyInstAlias<"vpop${p}", "$r",
+                         (VLDMSIA_UPD SP, pred:$p, spr_reglist:$r)>;
+defm : VFPDTAnyInstAlias<"vpop${p}", "$r",
+                         (VLDMDIA_UPD SP, pred:$p, dpr_reglist:$r)>;
+
+// FLDMX, FSTMX - Load and store multiple unknown precision registers for
+// pre-armv6 cores.
+// These instruction are deprecated so we don't want them to get selected.
+multiclass vfp_ldstx_mult<string asm, bit L_bit> {
+  // Unknown precision
+  def XIA :
+    AXXI4<(outs), (ins GPR:$Rn, pred:$p, dpr_reglist:$regs, variable_ops),
+          IndexModeNone, !strconcat(asm, "iax${p}\t$Rn, $regs"), "", []> {
+    let Inst{24-23} = 0b01;       // Increment After
+    let Inst{21}    = 0;          // No writeback
+    let Inst{20}    = L_bit;
+  }
+  def XIA_UPD :
+    AXXI4<(outs GPR:$wb), (ins GPR:$Rn, pred:$p, dpr_reglist:$regs, variable_ops),
+          IndexModeUpd, !strconcat(asm, "iax${p}\t$Rn!, $regs"), "$Rn = $wb", []> {
+    let Inst{24-23} = 0b01;         // Increment After
+    let Inst{21}    = 1;            // Writeback
+    let Inst{20}    = L_bit;
+  }
+  def XDB_UPD :
+    AXXI4<(outs GPR:$wb), (ins GPR:$Rn, pred:$p, dpr_reglist:$regs, variable_ops),
+          IndexModeUpd, !strconcat(asm, "dbx${p}\t$Rn!, $regs"), "$Rn = $wb", []> {
+    let Inst{24-23} = 0b10;         // Decrement Before
+    let Inst{21}    = 1;            // Writeback
+    let Inst{20}    = L_bit;
+  }
+}
+
+defm FLDM : vfp_ldstx_mult<"fldm", 1>;
+defm FSTM : vfp_ldstx_mult<"fstm", 0>;
 
-// FLDMX, FSTMX - mixing S/D registers for pre-armv6 cores
+def : VFP2MnemonicAlias<"fldmeax", "fldmdbx">;
+def : VFP2MnemonicAlias<"fldmfdx", "fldmiax">;
+
+def : VFP2MnemonicAlias<"fstmeax", "fstmiax">;
+def : VFP2MnemonicAlias<"fstmfdx", "fstmdbx">;
 
 //===----------------------------------------------------------------------===//
 // FP Binary Operations.
 //
 
+let TwoOperandAliasConstraint = "$Dn = $Dd" in
 def VADDD  : ADbI<0b11100, 0b11, 0, 0,
                   (outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm),
                   IIC_fpALU64, "vadd", ".f64\t$Dd, $Dn, $Dm",
                   [(set DPR:$Dd, (fadd DPR:$Dn, (f64 DPR:$Dm)))]>;
 
+let TwoOperandAliasConstraint = "$Sn = $Sd" in
 def VADDS  : ASbIn<0b11100, 0b11, 0, 0,
                    (outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm),
                    IIC_fpALU32, "vadd", ".f32\t$Sd, $Sn, $Sm",
@@ -227,11 +295,13 @@ def VADDS  : ASbIn<0b11100, 0b11, 0, 0,
   let D = VFPNeonA8Domain;
 }
 
+let TwoOperandAliasConstraint = "$Dn = $Dd" in
 def VSUBD  : ADbI<0b11100, 0b11, 1, 0,
                   (outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm),
                   IIC_fpALU64, "vsub", ".f64\t$Dd, $Dn, $Dm",
                   [(set DPR:$Dd, (fsub DPR:$Dn, (f64 DPR:$Dm)))]>;
 
+let TwoOperandAliasConstraint = "$Sn = $Sd" in
 def VSUBS  : ASbIn<0b11100, 0b11, 1, 0,
                    (outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm),
                    IIC_fpALU32, "vsub", ".f32\t$Sd, $Sn, $Sm",
@@ -241,21 +311,25 @@ def VSUBS  : ASbIn<0b11100, 0b11, 1, 0,
   let D = VFPNeonA8Domain;
 }
 
+let TwoOperandAliasConstraint = "$Dn = $Dd" in
 def VDIVD  : ADbI<0b11101, 0b00, 0, 0,
                   (outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm),
                   IIC_fpDIV64, "vdiv", ".f64\t$Dd, $Dn, $Dm",
                   [(set DPR:$Dd, (fdiv DPR:$Dn, (f64 DPR:$Dm)))]>;
 
+let TwoOperandAliasConstraint = "$Sn = $Sd" in
 def VDIVS  : ASbI<0b11101, 0b00, 0, 0,
                   (outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm),
                   IIC_fpDIV32, "vdiv", ".f32\t$Sd, $Sn, $Sm",
                   [(set SPR:$Sd, (fdiv SPR:$Sn, SPR:$Sm))]>;
 
+let TwoOperandAliasConstraint = "$Dn = $Dd" in
 def VMULD  : ADbI<0b11100, 0b10, 0, 0,
                   (outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm),
                   IIC_fpMUL64, "vmul", ".f64\t$Dd, $Dn, $Dm",
                   [(set DPR:$Dd, (fmul DPR:$Dn, (f64 DPR:$Dm)))]>;
 
+let TwoOperandAliasConstraint = "$Sn = $Sd" in
 def VMULS  : ASbIn<0b11100, 0b10, 0, 0,
                    (outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm),
                    IIC_fpMUL32, "vmul", ".f32\t$Sd, $Sn, $Sm",
@@ -279,14 +353,57 @@ def VNMULS : ASbI<0b11100, 0b10, 1, 0,
   let D = VFPNeonA8Domain;
 }
 
+multiclass vsel_inst<string op, bits<2> opc, int CC> {
+  let DecoderNamespace = "VFPV8", PostEncoderMethod = "",
+      Uses = [CPSR], AddedComplexity = 4 in {
+    def S : ASbInp<0b11100, opc, 0,
+                   (outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm),
+                   NoItinerary, !strconcat("vsel", op, ".f32\t$Sd, $Sn, $Sm"),
+                   [(set SPR:$Sd, (ARMcmov SPR:$Sm, SPR:$Sn, CC))]>,
+                   Requires<[HasFPARMv8]>;
+
+    def D : ADbInp<0b11100, opc, 0,
+                   (outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm),
+                   NoItinerary, !strconcat("vsel", op, ".f64\t$Dd, $Dn, $Dm"),
+                   [(set DPR:$Dd, (ARMcmov (f64 DPR:$Dm), (f64 DPR:$Dn), CC))]>,
+                   Requires<[HasFPARMv8, HasDPVFP]>;
+  }
+}
+
+// The CC constants here match ARMCC::CondCodes.
+defm VSELGT : vsel_inst<"gt", 0b11, 12>;
+defm VSELGE : vsel_inst<"ge", 0b10, 10>;
+defm VSELEQ : vsel_inst<"eq", 0b00, 0>;
+defm VSELVS : vsel_inst<"vs", 0b01, 6>;
+
+multiclass vmaxmin_inst<string op, bit opc, SDNode SD> {
+  let DecoderNamespace = "VFPV8", PostEncoderMethod = "" in {
+    def S : ASbInp<0b11101, 0b00, opc,
+                   (outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm),
+                   NoItinerary, !strconcat(op, ".f32\t$Sd, $Sn, $Sm"),
+                   [(set SPR:$Sd, (SD SPR:$Sn, SPR:$Sm))]>,
+                   Requires<[HasFPARMv8]>;
+
+    def D : ADbInp<0b11101, 0b00, opc,
+                   (outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm),
+                   NoItinerary, !strconcat(op, ".f64\t$Dd, $Dn, $Dm"),
+                   [(set DPR:$Dd, (f64 (SD (f64 DPR:$Dn), (f64 DPR:$Dm))))]>,
+                   Requires<[HasFPARMv8, HasDPVFP]>;
+  }
+}
+
+defm VMAXNM : vmaxmin_inst<"vmaxnm", 0, fmaxnum>;
+defm VMINNM : vmaxmin_inst<"vminnm", 1, fminnum>;
+
 // Match reassociated forms only if not sign dependent rounding.
 def : Pat<(fmul (fneg DPR:$a), (f64 DPR:$b)),
-          (VNMULD DPR:$a, DPR:$b)>, Requires<[NoHonorSignDependentRounding]>;
+          (VNMULD DPR:$a, DPR:$b)>,
+          Requires<[NoHonorSignDependentRounding,HasDPVFP]>;
 def : Pat<(fmul (fneg SPR:$a), SPR:$b),
           (VNMULS SPR:$a, SPR:$b)>, Requires<[NoHonorSignDependentRounding]>;
 
 // These are encoded as unary instructions.
-let Defs = [FPSCR] in {
+let Defs = [FPSCR_NZCV] in {
 def VCMPED : ADuI<0b11101, 0b11, 0b0100, 0b11, 0,
                   (outs), (ins DPR:$Dd, DPR:$Dm),
                   IIC_fpCMP64, "vcmpe", ".f64\t$Dd, $Dm",
@@ -315,7 +432,7 @@ def VCMPS  : ASuI<0b11101, 0b11, 0b0100, 0b01, 0,
   // VFP pipelines on A8.
   let D = VFPNeonA8Domain;
 }
-} // Defs = [FPSCR]
+} // Defs = [FPSCR_NZCV]
 
 //===----------------------------------------------------------------------===//
 // FP Unary Operations.
@@ -335,7 +452,7 @@ def VABSS  : ASuIn<0b11101, 0b11, 0b0000, 0b11, 0,
   let D = VFPNeonA8Domain;
 }
 
-let Defs = [FPSCR] in {
+let Defs = [FPSCR_NZCV] in {
 def VCMPEZD : ADuI<0b11101, 0b11, 0b0101, 0b11, 0,
                    (outs), (ins DPR:$Dd),
                    IIC_fpCMP64, "vcmpe", ".f64\t$Dd, #0",
@@ -376,7 +493,7 @@ def VCMPZS  : ASuI<0b11101, 0b11, 0b0101, 0b01, 0,
   // VFP pipelines on A8.
   let D = VFPNeonA8Domain;
 }
-} // Defs = [FPSCR]
+} // Defs = [FPSCR_NZCV]
 
 def VCVTDS  : ASuI<0b11101, 0b11, 0b0111, 0b11, 0,
                    (outs DPR:$Dd), (ins SPR:$Sm),
@@ -391,6 +508,8 @@ def VCVTDS  : ASuI<0b11101, 0b11, 0b0111, 0b11, 0,
   let Inst{5}     = Sm{0};
   let Inst{15-12} = Dd{3-0};
   let Inst{22}    = Dd{4};
+
+  let Predicates = [HasVFP2, HasDPVFP];
 }
 
 // Special case encoding: bits 11-8 is 0b1011.
@@ -412,32 +531,175 @@ def VCVTSD  : VFPAI<(outs SPR:$Sd), (ins DPR:$Dm), VFPUnaryFrm,
   let Inst{11-8}  = 0b1011;
   let Inst{7-6}   = 0b11;
   let Inst{4}     = 0;
+
+  let Predicates = [HasVFP2, HasDPVFP];
 }
 
-// Between half-precision and single-precision.  For disassembly only.
+// Between half, single and double-precision.  For disassembly only.
 
 // FIXME: Verify encoding after integrated assembler is working.
-def VCVTBSH: ASuI<0b11101, 0b11, 0b0010, 0b01, 0, (outs SPR:$Sd), (ins SPR:$Sm),
+def VCVTBHS: ASuI<0b11101, 0b11, 0b0010, 0b01, 0, (outs SPR:$Sd), (ins SPR:$Sm),
                  /* FIXME */ IIC_fpCVTSH, "vcvtb", ".f32.f16\t$Sd, $Sm",
-                 [/* For disassembly only; pattern left blank */]>;
-
-def : ARMPat<(f32_to_f16 SPR:$a),
-             (i32 (COPY_TO_REGCLASS (VCVTBSH SPR:$a), GPR))>;
+                 [/* For disassembly only; pattern left blank */]>,
+                 Requires<[HasFP16]>;
 
-def VCVTBHS: ASuI<0b11101, 0b11, 0b0011, 0b01, 0, (outs SPR:$Sd), (ins SPR:$Sm),
+def VCVTBSH: ASuI<0b11101, 0b11, 0b0011, 0b01, 0, (outs SPR:$Sd), (ins SPR:$Sm),
                  /* FIXME */ IIC_fpCVTHS, "vcvtb", ".f16.f32\t$Sd, $Sm",
-                 [/* For disassembly only; pattern left blank */]>;
+                 [/* For disassembly only; pattern left blank */]>,
+                 Requires<[HasFP16]>;
 
-def : ARMPat<(f16_to_f32 GPR:$a),
-             (VCVTBHS (COPY_TO_REGCLASS GPR:$a, SPR))>;
-
-def VCVTTSH: ASuI<0b11101, 0b11, 0b0010, 0b11, 0, (outs SPR:$Sd), (ins SPR:$Sm),
+def VCVTTHS: ASuI<0b11101, 0b11, 0b0010, 0b11, 0, (outs SPR:$Sd), (ins SPR:$Sm),
                  /* FIXME */ IIC_fpCVTSH, "vcvtt", ".f32.f16\t$Sd, $Sm",
-                 [/* For disassembly only; pattern left blank */]>;
+                 [/* For disassembly only; pattern left blank */]>,
+                 Requires<[HasFP16]>;
 
-def VCVTTHS: ASuI<0b11101, 0b11, 0b0011, 0b11, 0, (outs SPR:$Sd), (ins SPR:$Sm),
+def VCVTTSH: ASuI<0b11101, 0b11, 0b0011, 0b11, 0, (outs SPR:$Sd), (ins SPR:$Sm),
                  /* FIXME */ IIC_fpCVTHS, "vcvtt", ".f16.f32\t$Sd, $Sm",
-                 [/* For disassembly only; pattern left blank */]>;
+                 [/* For disassembly only; pattern left blank */]>,
+                 Requires<[HasFP16]>;
+
+def VCVTBHD : ADuI<0b11101, 0b11, 0b0010, 0b01, 0,
+                   (outs DPR:$Dd), (ins SPR:$Sm),
+                   NoItinerary, "vcvtb", ".f64.f16\t$Dd, $Sm",
+                   []>, Requires<[HasFPARMv8, HasDPVFP]> {
+  // Instruction operands.
+  bits<5> Sm;
+
+  // Encode instruction operands.
+  let Inst{3-0} = Sm{4-1};
+  let Inst{5}   = Sm{0};
+}
+
+def VCVTBDH : ADuI<0b11101, 0b11, 0b0011, 0b01, 0,
+                   (outs SPR:$Sd), (ins DPR:$Dm),
+                   NoItinerary, "vcvtb", ".f16.f64\t$Sd, $Dm",
+                   []>, Requires<[HasFPARMv8, HasDPVFP]> {
+  // Instruction operands.
+  bits<5> Sd;
+  bits<5> Dm;
+
+  // Encode instruction operands.
+  let Inst{3-0}     = Dm{3-0};
+  let Inst{5}       = Dm{4};
+  let Inst{15-12}   = Sd{4-1};
+  let Inst{22}      = Sd{0};
+}
+
+def VCVTTHD : ADuI<0b11101, 0b11, 0b0010, 0b11, 0,
+                   (outs DPR:$Dd), (ins SPR:$Sm),
+                   NoItinerary, "vcvtt", ".f64.f16\t$Dd, $Sm",
+                   []>, Requires<[HasFPARMv8, HasDPVFP]> {
+  // Instruction operands.
+  bits<5> Sm;
+
+  // Encode instruction operands.
+  let Inst{3-0} = Sm{4-1};
+  let Inst{5}   = Sm{0};
+}
+
+def VCVTTDH : ADuI<0b11101, 0b11, 0b0011, 0b11, 0,
+                   (outs SPR:$Sd), (ins DPR:$Dm),
+                   NoItinerary, "vcvtt", ".f16.f64\t$Sd, $Dm",
+                   []>, Requires<[HasFPARMv8, HasDPVFP]> {
+  // Instruction operands.
+  bits<5> Sd;
+  bits<5> Dm;
+
+  // Encode instruction operands.
+  let Inst{15-12} = Sd{4-1};
+  let Inst{22}    = Sd{0};
+  let Inst{3-0}   = Dm{3-0};
+  let Inst{5}     = Dm{4};
+}
+
+def : Pat<(fp_to_f16 SPR:$a),
+          (i32 (COPY_TO_REGCLASS (VCVTBSH SPR:$a), GPR))>;
+
+def : Pat<(fp_to_f16 (f64 DPR:$a)),
+          (i32 (COPY_TO_REGCLASS (VCVTBDH DPR:$a), GPR))>;
+
+def : Pat<(f16_to_fp GPR:$a),
+          (VCVTBHS (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+def : Pat<(f64 (f16_to_fp GPR:$a)),
+          (VCVTBHD (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+multiclass vcvt_inst<string opc, bits<2> rm,
+                     SDPatternOperator node = null_frag> {
+  let PostEncoderMethod = "", DecoderNamespace = "VFPV8" in {
+    def SS : ASuInp<0b11101, 0b11, 0b1100, 0b11, 0,
+                    (outs SPR:$Sd), (ins SPR:$Sm),
+                    NoItinerary, !strconcat("vcvt", opc, ".s32.f32\t$Sd, $Sm"),
+                    []>,
+                    Requires<[HasFPARMv8]> {
+      let Inst{17-16} = rm;
+    }
+
+    def US : ASuInp<0b11101, 0b11, 0b1100, 0b01, 0,
+                    (outs SPR:$Sd), (ins SPR:$Sm),
+                    NoItinerary, !strconcat("vcvt", opc, ".u32.f32\t$Sd, $Sm"),
+                    []>,
+                    Requires<[HasFPARMv8]> {
+      let Inst{17-16} = rm;
+    }
+
+    def SD : ASuInp<0b11101, 0b11, 0b1100, 0b11, 0,
+                    (outs SPR:$Sd), (ins DPR:$Dm),
+                    NoItinerary, !strconcat("vcvt", opc, ".s32.f64\t$Sd, $Dm"),
+                    []>,
+                    Requires<[HasFPARMv8, HasDPVFP]> {
+      bits<5> Dm;
+
+      let Inst{17-16} = rm;
+
+      // Encode instruction operands
+      let Inst{3-0} = Dm{3-0};
+      let Inst{5}   = Dm{4};
+      let Inst{8} = 1;
+    }
+
+    def UD : ASuInp<0b11101, 0b11, 0b1100, 0b01, 0,
+                    (outs SPR:$Sd), (ins DPR:$Dm),
+                    NoItinerary, !strconcat("vcvt", opc, ".u32.f64\t$Sd, $Dm"),
+                    []>,
+                    Requires<[HasFPARMv8, HasDPVFP]> {
+      bits<5> Dm;
+
+      let Inst{17-16} = rm;
+
+      // Encode instruction operands
+      let Inst{3-0}  = Dm{3-0};
+      let Inst{5}    = Dm{4};
+      let Inst{8} = 1;
+    }
+  }
+
+  let Predicates = [HasFPARMv8] in {
+    def : Pat<(i32 (fp_to_sint (node SPR:$a))),
+              (COPY_TO_REGCLASS
+                (!cast<Instruction>(NAME#"SS") SPR:$a),
+                GPR)>;
+    def : Pat<(i32 (fp_to_uint (node SPR:$a))),
+              (COPY_TO_REGCLASS
+                (!cast<Instruction>(NAME#"US") SPR:$a),
+                GPR)>;
+  }
+  let Predicates = [HasFPARMv8, HasDPVFP] in {
+    def : Pat<(i32 (fp_to_sint (node (f64 DPR:$a)))),
+              (COPY_TO_REGCLASS
+                (!cast<Instruction>(NAME#"SD") DPR:$a),
+                GPR)>;
+    def : Pat<(i32 (fp_to_uint (node (f64 DPR:$a)))),
+              (COPY_TO_REGCLASS
+                (!cast<Instruction>(NAME#"UD") DPR:$a),
+                GPR)>;
+  }
+}
+
+defm VCVTA : vcvt_inst<"a", 0b00, frnd>;
+defm VCVTN : vcvt_inst<"n", 0b01>;
+defm VCVTP : vcvt_inst<"p", 0b10, fceil>;
+defm VCVTM : vcvt_inst<"m", 0b11, ffloor>;
 
 def VNEGD  : ADuI<0b11101, 0b11, 0b0001, 0b01, 0,
                   (outs DPR:$Dd), (ins DPR:$Dm),
@@ -453,6 +715,68 @@ def VNEGS  : ASuIn<0b11101, 0b11, 0b0001, 0b01, 0,
   let D = VFPNeonA8Domain;
 }
 
+multiclass vrint_inst_zrx<string opc, bit op, bit op2, SDPatternOperator node> {
+  def S : ASuI<0b11101, 0b11, 0b0110, 0b11, 0,
+               (outs SPR:$Sd), (ins SPR:$Sm),
+               NoItinerary, !strconcat("vrint", opc), ".f32\t$Sd, $Sm",
+               [(set (f32 SPR:$Sd), (node (f32 SPR:$Sm)))]>,
+               Requires<[HasFPARMv8]> {
+    let Inst{7} = op2;
+    let Inst{16} = op;
+  }
+  def D : ADuI<0b11101, 0b11, 0b0110, 0b11, 0,
+                (outs DPR:$Dd), (ins DPR:$Dm),
+                NoItinerary, !strconcat("vrint", opc), ".f64\t$Dd, $Dm",
+                [(set (f64 DPR:$Dd), (node (f64 DPR:$Dm)))]>,
+                Requires<[HasFPARMv8, HasDPVFP]> {
+    let Inst{7} = op2;
+    let Inst{16} = op;
+  }
+
+  def : InstAlias<!strconcat("vrint", opc, "$p.f32.f32\t$Sd, $Sm"),
+                  (!cast<Instruction>(NAME#"S") SPR:$Sd, SPR:$Sm, pred:$p)>,
+        Requires<[HasFPARMv8]>;
+  def : InstAlias<!strconcat("vrint", opc, "$p.f64.f64\t$Dd, $Dm"),
+                  (!cast<Instruction>(NAME#"D") DPR:$Dd, DPR:$Dm, pred:$p)>,
+        Requires<[HasFPARMv8,HasDPVFP]>;
+}
+
+defm VRINTZ : vrint_inst_zrx<"z", 0, 1, ftrunc>;
+defm VRINTR : vrint_inst_zrx<"r", 0, 0, fnearbyint>;
+defm VRINTX : vrint_inst_zrx<"x", 1, 0, frint>;
+
+multiclass vrint_inst_anpm<string opc, bits<2> rm,
+                           SDPatternOperator node = null_frag> {
+  let PostEncoderMethod = "", DecoderNamespace = "VFPV8" in {
+    def S : ASuInp<0b11101, 0b11, 0b1000, 0b01, 0,
+                   (outs SPR:$Sd), (ins SPR:$Sm),
+                   NoItinerary, !strconcat("vrint", opc, ".f32\t$Sd, $Sm"),
+                   [(set (f32 SPR:$Sd), (node (f32 SPR:$Sm)))]>,
+                   Requires<[HasFPARMv8]> {
+      let Inst{17-16} = rm;
+    }
+    def D : ADuInp<0b11101, 0b11, 0b1000, 0b01, 0,
+                   (outs DPR:$Dd), (ins DPR:$Dm),
+                   NoItinerary, !strconcat("vrint", opc, ".f64\t$Dd, $Dm"),
+                   [(set (f64 DPR:$Dd), (node (f64 DPR:$Dm)))]>,
+                   Requires<[HasFPARMv8, HasDPVFP]> {
+      let Inst{17-16} = rm;
+    }
+  }
+
+  def : InstAlias<!strconcat("vrint", opc, ".f32.f32\t$Sd, $Sm"),
+                  (!cast<Instruction>(NAME#"S") SPR:$Sd, SPR:$Sm)>,
+        Requires<[HasFPARMv8]>;
+  def : InstAlias<!strconcat("vrint", opc, ".f64.f64\t$Dd, $Dm"),
+                  (!cast<Instruction>(NAME#"D") DPR:$Dd, DPR:$Dm)>,
+        Requires<[HasFPARMv8,HasDPVFP]>;
+}
+
+defm VRINTA : vrint_inst_anpm<"a", 0b00, frnd>;
+defm VRINTN : vrint_inst_anpm<"n", 0b01>;
+defm VRINTP : vrint_inst_anpm<"p", 0b10, fceil>;
+defm VRINTM : vrint_inst_anpm<"m", 0b11, ffloor>;
+
 def VSQRTD : ADuI<0b11101, 0b11, 0b0001, 0b11, 0,
                   (outs DPR:$Dd), (ins DPR:$Dm),
                   IIC_fpSQRT64, "vsqrt", ".f64\t$Dd, $Dm",
@@ -463,7 +787,7 @@ def VSQRTS : ASuI<0b11101, 0b11, 0b0001, 0b11, 0,
                   IIC_fpSQRT32, "vsqrt", ".f32\t$Sd, $Sm",
                   [(set SPR:$Sd, (fsqrt SPR:$Sm))]>;
 
-let neverHasSideEffects = 1 in {
+let hasSideEffects = 0 in {
 def VMOVD  : ADuI<0b11101, 0b11, 0b0000, 0b01, 0,
                   (outs DPR:$Dd), (ins DPR:$Dm),
                   IIC_fpUNA64, "vmov", ".f64\t$Dd, $Dm", []>;
@@ -471,7 +795,7 @@ def VMOVD  : ADuI<0b11101, 0b11, 0b0000, 0b01, 0,
 def VMOVS  : ASuI<0b11101, 0b11, 0b0000, 0b01, 0,
                   (outs SPR:$Sd), (ins SPR:$Sm),
                   IIC_fpUNA32, "vmov", ".f32\t$Sd, $Sm", []>;
-} // neverHasSideEffects
+} // hasSideEffects
 
 //===----------------------------------------------------------------------===//
 // FP <-> GPR Copies.  Int <-> FP Conversions.
@@ -498,10 +822,12 @@ def VMOVRS : AVConv2I<0b11100001, 0b1010,
   let D = VFPNeonDomain;
 }
 
+// Bitcast i32 -> f32.  NEON prefers to use VMOVDRR.
 def VMOVSR : AVConv4I<0b11100000, 0b1010,
                       (outs SPR:$Sn), (ins GPR:$Rt),
                       IIC_fpMOVIS, "vmov", "\t$Sn, $Rt",
-                      [(set SPR:$Sn, (bitconvert GPR:$Rt))]> {
+                      [(set SPR:$Sn, (bitconvert GPR:$Rt))]>,
+             Requires<[HasVFP2, UseVMOVSR]> {
   // Instruction operands.
   bits<5> Sn;
   bits<4> Rt;
@@ -519,7 +845,7 @@ def VMOVSR : AVConv4I<0b11100000, 0b1010,
   let D = VFPNeonDomain;
 }
 
-let neverHasSideEffects = 1 in {
+let hasSideEffects = 0 in {
 def VMOVRRD  : AVConv3I<0b11000101, 0b1011,
                         (outs GPR:$Rt, GPR:$Rt2), (ins DPR:$Dm),
                         IIC_fpMOVDI, "vmov", "\t$Rt, $Rt2, $Dm",
@@ -540,6 +866,11 @@ def VMOVRRD  : AVConv3I<0b11000101, 0b1011,
   // Some single precision VFP instructions may be executed on both NEON and VFP
   // pipelines.
   let D = VFPNeonDomain;
+
+  // This instruction is equivalent to
+  // $Rt = EXTRACT_SUBREG $Dm, ssub_0
+  // $Rt2 = EXTRACT_SUBREG $Dm, ssub_1
+  let isExtractSubreg = 1;
 }
 
 def VMOVRRS  : AVConv3I<0b11000101, 0b1010,
@@ -551,8 +882,8 @@ def VMOVRRS  : AVConv3I<0b11000101, 0b1010,
   bits<4> Rt2;
 
   // Encode instruction operands.
-  let Inst{3-0}   = src1{3-0};
-  let Inst{5}     = src1{4};
+  let Inst{3-0}   = src1{4-1};
+  let Inst{5}     = src1{0};
   let Inst{15-12} = Rt;
   let Inst{19-16} = Rt2;
 
@@ -563,7 +894,7 @@ def VMOVRRS  : AVConv3I<0b11000101, 0b1010,
   let D = VFPNeonDomain;
   let DecoderMethod = "DecodeVMOVRRS";
 }
-} // neverHasSideEffects
+} // hasSideEffects
 
 // FMDHR: GPR -> SPR
 // FMDLR: GPR -> SPR
@@ -588,9 +919,29 @@ def VMOVDRR : AVConv5I<0b11000100, 0b1011,
   // Some single precision VFP instructions may be executed on both NEON and VFP
   // pipelines.
   let D = VFPNeonDomain;
-}
 
-let neverHasSideEffects = 1 in
+  // This instruction is equivalent to
+  // $Dm = REG_SEQUENCE $Rt, ssub_0, $Rt2, ssub_1
+  let isRegSequence = 1;
+}
+
+// Hoist an fabs or a fneg of a value coming from integer registers
+// and do the fabs/fneg on the integer value. This is never a lose
+// and could enable the conversion to float to be removed completely.
+def : Pat<(fabs (arm_fmdrr GPR:$Rl, GPR:$Rh)),
+          (VMOVDRR GPR:$Rl, (BFC GPR:$Rh, (i32 0x7FFFFFFF)))>,
+      Requires<[IsARM, HasV6T2]>;
+def : Pat<(fabs (arm_fmdrr GPR:$Rl, GPR:$Rh)),
+          (VMOVDRR GPR:$Rl, (t2BFC GPR:$Rh, (i32 0x7FFFFFFF)))>,
+      Requires<[IsThumb2, HasV6T2]>;
+def : Pat<(fneg (arm_fmdrr GPR:$Rl, GPR:$Rh)),
+          (VMOVDRR GPR:$Rl, (EORri GPR:$Rh, (i32 0x80000000)))>,
+      Requires<[IsARM]>;
+def : Pat<(fneg (arm_fmdrr GPR:$Rl, GPR:$Rh)),
+          (VMOVDRR GPR:$Rl, (t2EORri GPR:$Rh, (i32 0x80000000)))>,
+      Requires<[IsThumb2]>;
+
+let hasSideEffects = 0 in
 def VMOVSRR : AVConv5I<0b11000100, 0b1010,
                      (outs SPR:$dst1, SPR:$dst2), (ins GPR:$src1, GPR:$src2),
                 IIC_fpMOVID, "vmov", "\t$dst1, $dst2, $src1, $src2",
@@ -601,8 +952,8 @@ def VMOVSRR : AVConv5I<0b11000100, 0b1010,
   bits<4> src2;
 
   // Encode instruction operands.
-  let Inst{3-0}   = dst1{3-0};
-  let Inst{5}     = dst1{4};
+  let Inst{3-0}   = dst1{4-1};
+  let Inst{5}     = dst1{0};
   let Inst{15-12} = src1;
   let Inst{19-16} = src2;
 
@@ -640,6 +991,8 @@ class AVConv1IDs_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
   let Inst{5}     = Sm{0};
   let Inst{15-12} = Dd{3-0};
   let Inst{22}    = Dd{4};
+
+  let Predicates = [HasVFP2, HasDPVFP];
 }
 
 class AVConv1InSs_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
@@ -661,14 +1014,22 @@ class AVConv1InSs_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
 def VSITOD : AVConv1IDs_Encode<0b11101, 0b11, 0b1000, 0b1011,
                                (outs DPR:$Dd), (ins SPR:$Sm),
                                IIC_fpCVTID, "vcvt", ".f64.s32\t$Dd, $Sm",
-                               [(set DPR:$Dd, (f64 (arm_sitof SPR:$Sm)))]> {
+                               []> {
   let Inst{7} = 1; // s32
 }
 
+let Predicates=[HasVFP2, HasDPVFP] in {
+  def : VFPPat<(f64 (sint_to_fp GPR:$a)),
+               (VSITOD (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+  def : VFPPat<(f64 (sint_to_fp (i32 (alignedload32 addrmode5:$a)))),
+               (VSITOD (VLDRS addrmode5:$a))>;
+}
+
 def VSITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
                                 (outs SPR:$Sd),(ins SPR:$Sm),
                                 IIC_fpCVTIS, "vcvt", ".f32.s32\t$Sd, $Sm",
-                                [(set SPR:$Sd, (arm_sitof SPR:$Sm))]> {
+                                []> {
   let Inst{7} = 1; // s32
 
   // Some single precision VFP instructions may be executed on both NEON and
@@ -676,17 +1037,31 @@ def VSITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
   let D = VFPNeonA8Domain;
 }
 
+def : VFPNoNEONPat<(f32 (sint_to_fp GPR:$a)),
+                   (VSITOS (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+def : VFPNoNEONPat<(f32 (sint_to_fp (i32 (alignedload32 addrmode5:$a)))),
+                   (VSITOS (VLDRS addrmode5:$a))>;
+
 def VUITOD : AVConv1IDs_Encode<0b11101, 0b11, 0b1000, 0b1011,
                                (outs DPR:$Dd), (ins SPR:$Sm),
                                IIC_fpCVTID, "vcvt", ".f64.u32\t$Dd, $Sm",
-                               [(set DPR:$Dd, (f64 (arm_uitof SPR:$Sm)))]> {
+                               []> {
   let Inst{7} = 0; // u32
 }
 
+let Predicates=[HasVFP2, HasDPVFP] in {
+  def : VFPPat<(f64 (uint_to_fp GPR:$a)),
+               (VUITOD (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+  def : VFPPat<(f64 (uint_to_fp (i32 (alignedload32 addrmode5:$a)))),
+               (VUITOD (VLDRS addrmode5:$a))>;
+}
+
 def VUITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
                                 (outs SPR:$Sd), (ins SPR:$Sm),
                                 IIC_fpCVTIS, "vcvt", ".f32.u32\t$Sd, $Sm",
-                                [(set SPR:$Sd, (arm_uitof SPR:$Sm))]> {
+                                []> {
   let Inst{7} = 0; // u32
 
   // Some single precision VFP instructions may be executed on both NEON and
@@ -694,6 +1069,12 @@ def VUITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
   let D = VFPNeonA8Domain;
 }
 
+def : VFPNoNEONPat<(f32 (uint_to_fp GPR:$a)),
+                   (VUITOS (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+def : VFPNoNEONPat<(f32 (uint_to_fp (i32 (alignedload32 addrmode5:$a)))),
+                   (VUITOS (VLDRS addrmode5:$a))>;
+
 // FP -> Int:
 
 class AVConv1IsD_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
@@ -711,6 +1092,8 @@ class AVConv1IsD_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
   let Inst{5}     = Dm{4};
   let Inst{15-12} = Sd{4-1};
   let Inst{22}    = Sd{0};
+
+  let Predicates = [HasVFP2, HasDPVFP];
 }
 
 class AVConv1InsS_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
@@ -734,14 +1117,22 @@ class AVConv1InsS_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
 def VTOSIZD : AVConv1IsD_Encode<0b11101, 0b11, 0b1101, 0b1011,
                                 (outs SPR:$Sd), (ins DPR:$Dm),
                                 IIC_fpCVTDI, "vcvt", ".s32.f64\t$Sd, $Dm",
-                                [(set SPR:$Sd, (arm_ftosi (f64 DPR:$Dm)))]> {
+                                []> {
   let Inst{7} = 1; // Z bit
 }
 
+let Predicates=[HasVFP2, HasDPVFP] in {
+  def : VFPPat<(i32 (fp_to_sint (f64 DPR:$a))),
+               (COPY_TO_REGCLASS (VTOSIZD DPR:$a), GPR)>;
+
+  def : VFPPat<(alignedstore32 (i32 (fp_to_sint (f64 DPR:$a))), addrmode5:$ptr),
+               (VSTRS (VTOSIZD DPR:$a), addrmode5:$ptr)>;
+}
+
 def VTOSIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1101, 0b1010,
                                  (outs SPR:$Sd), (ins SPR:$Sm),
                                  IIC_fpCVTSI, "vcvt", ".s32.f32\t$Sd, $Sm",
-                                 [(set SPR:$Sd, (arm_ftosi SPR:$Sm))]> {
+                                 []> {
   let Inst{7} = 1; // Z bit
 
   // Some single precision VFP instructions may be executed on both NEON and
@@ -749,17 +1140,32 @@ def VTOSIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1101, 0b1010,
   let D = VFPNeonA8Domain;
 }
 
+def : VFPNoNEONPat<(i32 (fp_to_sint SPR:$a)),
+                   (COPY_TO_REGCLASS (VTOSIZS SPR:$a), GPR)>;
+
+def : VFPNoNEONPat<(alignedstore32 (i32 (fp_to_sint (f32 SPR:$a))),
+                                   addrmode5:$ptr),
+                   (VSTRS (VTOSIZS SPR:$a), addrmode5:$ptr)>;
+
 def VTOUIZD : AVConv1IsD_Encode<0b11101, 0b11, 0b1100, 0b1011,
                                (outs SPR:$Sd), (ins DPR:$Dm),
                                IIC_fpCVTDI, "vcvt", ".u32.f64\t$Sd, $Dm",
-                               [(set SPR:$Sd, (arm_ftoui (f64 DPR:$Dm)))]> {
+                               []> {
   let Inst{7} = 1; // Z bit
 }
 
+let Predicates=[HasVFP2, HasDPVFP] in {
+  def : VFPPat<(i32 (fp_to_uint (f64 DPR:$a))),
+               (COPY_TO_REGCLASS (VTOUIZD DPR:$a), GPR)>;
+
+  def : VFPPat<(alignedstore32 (i32 (fp_to_uint (f64 DPR:$a))), addrmode5:$ptr),
+               (VSTRS (VTOUIZD DPR:$a), addrmode5:$ptr)>;
+}
+
 def VTOUIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1100, 0b1010,
                                  (outs SPR:$Sd), (ins SPR:$Sm),
                                  IIC_fpCVTSI, "vcvt", ".u32.f32\t$Sd, $Sm",
-                                 [(set SPR:$Sd, (arm_ftoui SPR:$Sm))]> {
+                                 []> {
   let Inst{7} = 1; // Z bit
 
   // Some single precision VFP instructions may be executed on both NEON and
@@ -767,6 +1173,13 @@ def VTOUIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1100, 0b1010,
   let D = VFPNeonA8Domain;
 }
 
+def : VFPNoNEONPat<(i32 (fp_to_uint SPR:$a)),
+                   (COPY_TO_REGCLASS (VTOUIZS SPR:$a), GPR)>;
+
+def : VFPNoNEONPat<(alignedstore32 (i32 (fp_to_uint (f32 SPR:$a))),
+                                   addrmode5:$ptr),
+                  (VSTRS (VTOUIZS SPR:$a), addrmode5:$ptr)>;
+
 // And the Z bit '0' variants, i.e. use the rounding mode specified by FPSCR.
 let Uses = [FPSCR] in {
 // FIXME: Verify encoding after integrated assembler is working.
@@ -810,7 +1223,33 @@ let Constraints = "$a = $dst" in {
 
 // FP to Fixed-Point:
 
-def VTOSHS : AVConv1XI<0b11101, 0b11, 0b1110, 0b1010, 0,
+// Single Precision register
+class AVConv1XInsS_Encode<bits<5> op1, bits<2> op2, bits<4> op3, bits<4> op4,
+                          bit op5, dag oops, dag iops, InstrItinClass itin,
+                          string opc, string asm, list<dag> pattern>
+  : AVConv1XI<op1, op2, op3, op4, op5, oops, iops, itin, opc, asm, pattern>,
+  Sched<[WriteCvtFP]> {
+  bits<5> dst;
+  // if dp_operation then UInt(D:Vd) else UInt(Vd:D);
+  let Inst{22} = dst{0};
+  let Inst{15-12} = dst{4-1};
+}
+
+// Double Precision register
+class AVConv1XInsD_Encode<bits<5> op1, bits<2> op2, bits<4> op3, bits<4> op4,
+                          bit op5, dag oops, dag iops, InstrItinClass itin,
+                          string opc, string asm, list<dag> pattern>
+  : AVConv1XI<op1, op2, op3, op4, op5, oops, iops, itin, opc, asm, pattern>,
+    Sched<[WriteCvtFP]> {
+  bits<5> dst;
+  // if dp_operation then UInt(D:Vd) else UInt(Vd:D);
+  let Inst{22} = dst{4};
+  let Inst{15-12} = dst{3-0};
+
+  let Predicates = [HasVFP2, HasDPVFP];
+}
+
+def VTOSHS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1110, 0b1010, 0,
                        (outs SPR:$dst), (ins SPR:$a, fbits16:$fbits),
                  IIC_fpCVTSI, "vcvt", ".s16.f32\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -818,7 +1257,7 @@ def VTOSHS : AVConv1XI<0b11101, 0b11, 0b1110, 0b1010, 0,
   let D = VFPNeonA8Domain;
 }
 
-def VTOUHS : AVConv1XI<0b11101, 0b11, 0b1111, 0b1010, 0,
+def VTOUHS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1111, 0b1010, 0,
                        (outs SPR:$dst), (ins SPR:$a, fbits16:$fbits),
                  IIC_fpCVTSI, "vcvt", ".u16.f32\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -826,7 +1265,7 @@ def VTOUHS : AVConv1XI<0b11101, 0b11, 0b1111, 0b1010, 0,
   let D = VFPNeonA8Domain;
 }
 
-def VTOSLS : AVConv1XI<0b11101, 0b11, 0b1110, 0b1010, 1,
+def VTOSLS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1110, 0b1010, 1,
                        (outs SPR:$dst), (ins SPR:$a, fbits32:$fbits),
                  IIC_fpCVTSI, "vcvt", ".s32.f32\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -834,7 +1273,7 @@ def VTOSLS : AVConv1XI<0b11101, 0b11, 0b1110, 0b1010, 1,
   let D = VFPNeonA8Domain;
 }
 
-def VTOULS : AVConv1XI<0b11101, 0b11, 0b1111, 0b1010, 1,
+def VTOULS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1111, 0b1010, 1,
                        (outs SPR:$dst), (ins SPR:$a, fbits32:$fbits),
                  IIC_fpCVTSI, "vcvt", ".u32.f32\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -842,25 +1281,25 @@ def VTOULS : AVConv1XI<0b11101, 0b11, 0b1111, 0b1010, 1,
   let D = VFPNeonA8Domain;
 }
 
-def VTOSHD : AVConv1XI<0b11101, 0b11, 0b1110, 0b1011, 0,
+def VTOSHD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1110, 0b1011, 0,
                        (outs DPR:$dst), (ins DPR:$a, fbits16:$fbits),
                  IIC_fpCVTDI, "vcvt", ".s16.f64\t$dst, $a, $fbits", []>;
 
-def VTOUHD : AVConv1XI<0b11101, 0b11, 0b1111, 0b1011, 0,
+def VTOUHD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1111, 0b1011, 0,
                        (outs DPR:$dst), (ins DPR:$a, fbits16:$fbits),
                  IIC_fpCVTDI, "vcvt", ".u16.f64\t$dst, $a, $fbits", []>;
 
-def VTOSLD : AVConv1XI<0b11101, 0b11, 0b1110, 0b1011, 1,
+def VTOSLD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1110, 0b1011, 1,
                        (outs DPR:$dst), (ins DPR:$a, fbits32:$fbits),
                  IIC_fpCVTDI, "vcvt", ".s32.f64\t$dst, $a, $fbits", []>;
 
-def VTOULD : AVConv1XI<0b11101, 0b11, 0b1111, 0b1011, 1,
+def VTOULD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1111, 0b1011, 1,
                        (outs DPR:$dst), (ins DPR:$a, fbits32:$fbits),
                  IIC_fpCVTDI, "vcvt", ".u32.f64\t$dst, $a, $fbits", []>;
 
 // Fixed-Point to FP:
 
-def VSHTOS : AVConv1XI<0b11101, 0b11, 0b1010, 0b1010, 0,
+def VSHTOS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1010, 0b1010, 0,
                        (outs SPR:$dst), (ins SPR:$a, fbits16:$fbits),
                  IIC_fpCVTIS, "vcvt", ".f32.s16\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -868,7 +1307,7 @@ def VSHTOS : AVConv1XI<0b11101, 0b11, 0b1010, 0b1010, 0,
   let D = VFPNeonA8Domain;
 }
 
-def VUHTOS : AVConv1XI<0b11101, 0b11, 0b1011, 0b1010, 0,
+def VUHTOS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1011, 0b1010, 0,
                        (outs SPR:$dst), (ins SPR:$a, fbits16:$fbits),
                  IIC_fpCVTIS, "vcvt", ".f32.u16\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -876,7 +1315,7 @@ def VUHTOS : AVConv1XI<0b11101, 0b11, 0b1011, 0b1010, 0,
   let D = VFPNeonA8Domain;
 }
 
-def VSLTOS : AVConv1XI<0b11101, 0b11, 0b1010, 0b1010, 1,
+def VSLTOS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1010, 0b1010, 1,
                        (outs SPR:$dst), (ins SPR:$a, fbits32:$fbits),
                  IIC_fpCVTIS, "vcvt", ".f32.s32\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -884,7 +1323,7 @@ def VSLTOS : AVConv1XI<0b11101, 0b11, 0b1010, 0b1010, 1,
   let D = VFPNeonA8Domain;
 }
 
-def VULTOS : AVConv1XI<0b11101, 0b11, 0b1011, 0b1010, 1,
+def VULTOS : AVConv1XInsS_Encode<0b11101, 0b11, 0b1011, 0b1010, 1,
                        (outs SPR:$dst), (ins SPR:$a, fbits32:$fbits),
                  IIC_fpCVTIS, "vcvt", ".f32.u32\t$dst, $a, $fbits", []> {
   // Some single precision VFP instructions may be executed on both NEON and
@@ -892,19 +1331,19 @@ def VULTOS : AVConv1XI<0b11101, 0b11, 0b1011, 0b1010, 1,
   let D = VFPNeonA8Domain;
 }
 
-def VSHTOD : AVConv1XI<0b11101, 0b11, 0b1010, 0b1011, 0,
+def VSHTOD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1010, 0b1011, 0,
                        (outs DPR:$dst), (ins DPR:$a, fbits16:$fbits),
                  IIC_fpCVTID, "vcvt", ".f64.s16\t$dst, $a, $fbits", []>;
 
-def VUHTOD : AVConv1XI<0b11101, 0b11, 0b1011, 0b1011, 0,
+def VUHTOD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1011, 0b1011, 0,
                        (outs DPR:$dst), (ins DPR:$a, fbits16:$fbits),
                  IIC_fpCVTID, "vcvt", ".f64.u16\t$dst, $a, $fbits", []>;
 
-def VSLTOD : AVConv1XI<0b11101, 0b11, 0b1010, 0b1011, 1,
+def VSLTOD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1010, 0b1011, 1,
                        (outs DPR:$dst), (ins DPR:$a, fbits32:$fbits),
                  IIC_fpCVTID, "vcvt", ".f64.s32\t$dst, $a, $fbits", []>;
 
-def VULTOD : AVConv1XI<0b11101, 0b11, 0b1011, 0b1011, 1,
+def VULTOD : AVConv1XInsD_Encode<0b11101, 0b11, 0b1011, 0b1011, 1,
                        (outs DPR:$dst), (ins DPR:$a, fbits32:$fbits),
                  IIC_fpCVTID, "vcvt", ".f64.u32\t$dst, $a, $fbits", []>;
 
@@ -920,7 +1359,7 @@ def VMLAD : ADbI<0b11100, 0b00, 0, 0,
                  [(set DPR:$Dd, (fadd_mlx (fmul_su DPR:$Dn, DPR:$Dm),
                                           (f64 DPR:$Ddin)))]>,
               RegConstraint<"$Ddin = $Dd">,
-              Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+              Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 
 def VMLAS : ASbIn<0b11100, 0b00, 0, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
@@ -928,7 +1367,7 @@ def VMLAS : ASbIn<0b11100, 0b00, 0, 0,
                   [(set SPR:$Sd, (fadd_mlx (fmul_su SPR:$Sn, SPR:$Sm),
                                            SPR:$Sdin))]>,
               RegConstraint<"$Sdin = $Sd">,
-              Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]> {
+              Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines on A8.
   let D = VFPNeonA8Domain;
@@ -936,10 +1375,10 @@ def VMLAS : ASbIn<0b11100, 0b00, 0, 0,
 
 def : Pat<(fadd_mlx DPR:$dstin, (fmul_su DPR:$a, (f64 DPR:$b))),
           (VMLAD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 def : Pat<(fadd_mlx SPR:$dstin, (fmul_su SPR:$a, SPR:$b)),
           (VMLAS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP2,DontUseNEONForFP, UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,DontUseNEONForFP, UseFPVMLx,DontUseFusedMAC]>;
 
 def VMLSD : ADbI<0b11100, 0b00, 1, 0,
                  (outs DPR:$Dd), (ins DPR:$Ddin, DPR:$Dn, DPR:$Dm),
@@ -947,7 +1386,7 @@ def VMLSD : ADbI<0b11100, 0b00, 1, 0,
                  [(set DPR:$Dd, (fadd_mlx (fneg (fmul_su DPR:$Dn,DPR:$Dm)),
                                           (f64 DPR:$Ddin)))]>,
               RegConstraint<"$Ddin = $Dd">,
-              Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+              Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 
 def VMLSS : ASbIn<0b11100, 0b00, 1, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
@@ -955,7 +1394,7 @@ def VMLSS : ASbIn<0b11100, 0b00, 1, 0,
                   [(set SPR:$Sd, (fadd_mlx (fneg (fmul_su SPR:$Sn, SPR:$Sm)),
                                            SPR:$Sdin))]>,
               RegConstraint<"$Sdin = $Sd">,
-              Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]> {
+              Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines on A8.
   let D = VFPNeonA8Domain;
@@ -963,10 +1402,10 @@ def VMLSS : ASbIn<0b11100, 0b00, 1, 0,
 
 def : Pat<(fsub_mlx DPR:$dstin, (fmul_su DPR:$a, (f64 DPR:$b))),
           (VMLSD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 def : Pat<(fsub_mlx SPR:$dstin, (fmul_su SPR:$a, SPR:$b)),
           (VMLSS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]>;
 
 def VNMLAD : ADbI<0b11100, 0b01, 1, 0,
                   (outs DPR:$Dd), (ins DPR:$Ddin, DPR:$Dn, DPR:$Dm),
@@ -974,7 +1413,7 @@ def VNMLAD : ADbI<0b11100, 0b01, 1, 0,
                   [(set DPR:$Dd,(fsub_mlx (fneg (fmul_su DPR:$Dn,DPR:$Dm)),
                                           (f64 DPR:$Ddin)))]>,
                 RegConstraint<"$Ddin = $Dd">,
-                Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+                Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 
 def VNMLAS : ASbI<0b11100, 0b01, 1, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
@@ -982,7 +1421,7 @@ def VNMLAS : ASbI<0b11100, 0b01, 1, 0,
                   [(set SPR:$Sd, (fsub_mlx (fneg (fmul_su SPR:$Sn, SPR:$Sm)),
                                            SPR:$Sdin))]>,
                 RegConstraint<"$Sdin = $Sd">,
-                Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]> {
+                Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines on A8.
   let D = VFPNeonA8Domain;
@@ -990,10 +1429,10 @@ def VNMLAS : ASbI<0b11100, 0b01, 1, 0,
 
 def : Pat<(fsub_mlx (fneg (fmul_su DPR:$a, (f64 DPR:$b))), DPR:$dstin),
           (VNMLAD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 def : Pat<(fsub_mlx (fneg (fmul_su SPR:$a, SPR:$b)), SPR:$dstin),
           (VNMLAS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]>;
 
 def VNMLSD : ADbI<0b11100, 0b01, 0, 0,
                   (outs DPR:$Dd), (ins DPR:$Ddin, DPR:$Dn, DPR:$Dm),
@@ -1001,14 +1440,14 @@ def VNMLSD : ADbI<0b11100, 0b01, 0, 0,
                   [(set DPR:$Dd, (fsub_mlx (fmul_su DPR:$Dn, DPR:$Dm),
                                            (f64 DPR:$Ddin)))]>,
                RegConstraint<"$Ddin = $Dd">,
-               Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+               Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 
 def VNMLSS : ASbI<0b11100, 0b01, 0, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
                   IIC_fpMAC32, "vnmls", ".f32\t$Sd, $Sn, $Sm",
              [(set SPR:$Sd, (fsub_mlx (fmul_su SPR:$Sn, SPR:$Sm), SPR:$Sdin))]>,
                          RegConstraint<"$Sdin = $Sd">,
-                  Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]> {
+                Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines on A8.
   let D = VFPNeonA8Domain;
@@ -1016,10 +1455,10 @@ def VNMLSS : ASbI<0b11100, 0b01, 0, 0,
 
 def : Pat<(fsub_mlx (fmul_su DPR:$a, (f64 DPR:$b)), DPR:$dstin),
           (VNMLSD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP2,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,HasDPVFP,UseFPVMLx,DontUseFusedMAC]>;
 def : Pat<(fsub_mlx (fmul_su SPR:$a, SPR:$b), SPR:$dstin),
           (VNMLSS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,NoVFP4]>;
+          Requires<[HasVFP2,DontUseNEONForFP,UseFPVMLx,DontUseFusedMAC]>;
 
 //===----------------------------------------------------------------------===//
 // Fused FP Multiply-Accumulate Operations.
@@ -1030,7 +1469,7 @@ def VFMAD : ADbI<0b11101, 0b10, 0, 0,
                  [(set DPR:$Dd, (fadd_mlx (fmul_su DPR:$Dn, DPR:$Dm),
                                           (f64 DPR:$Ddin)))]>,
               RegConstraint<"$Ddin = $Dd">,
-              Requires<[HasVFP4]>;
+              Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 
 def VFMAS : ASbIn<0b11101, 0b10, 0, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
@@ -1038,17 +1477,26 @@ def VFMAS : ASbIn<0b11101, 0b10, 0, 0,
                   [(set SPR:$Sd, (fadd_mlx (fmul_su SPR:$Sn, SPR:$Sm),
                                            SPR:$Sdin))]>,
               RegConstraint<"$Sdin = $Sd">,
-              Requires<[HasVFP4,DontUseNEONForFP]> {
+              Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines.
 }
 
 def : Pat<(fadd_mlx DPR:$dstin, (fmul_su DPR:$a, (f64 DPR:$b))),
           (VFMAD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP4]>;
+          Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 def : Pat<(fadd_mlx SPR:$dstin, (fmul_su SPR:$a, SPR:$b)),
           (VFMAS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP4,DontUseNEONForFP]>;
+          Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]>;
+
+// Match @llvm.fma.* intrinsics
+// (fma x, y, z) -> (vfms z, x, y)
+def : Pat<(f64 (fma DPR:$Dn, DPR:$Dm, DPR:$Ddin)),
+          (VFMAD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(f32 (fma SPR:$Sn, SPR:$Sm, SPR:$Sdin)),
+          (VFMAS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
 
 def VFMSD : ADbI<0b11101, 0b10, 1, 0,
                  (outs DPR:$Dd), (ins DPR:$Ddin, DPR:$Dn, DPR:$Dm),
@@ -1056,7 +1504,7 @@ def VFMSD : ADbI<0b11101, 0b10, 1, 0,
                  [(set DPR:$Dd, (fadd_mlx (fneg (fmul_su DPR:$Dn,DPR:$Dm)),
                                           (f64 DPR:$Ddin)))]>,
               RegConstraint<"$Ddin = $Dd">,
-              Requires<[HasVFP4]>;
+              Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 
 def VFMSS : ASbIn<0b11101, 0b10, 1, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
@@ -1064,17 +1512,33 @@ def VFMSS : ASbIn<0b11101, 0b10, 1, 0,
                   [(set SPR:$Sd, (fadd_mlx (fneg (fmul_su SPR:$Sn, SPR:$Sm)),
                                            SPR:$Sdin))]>,
               RegConstraint<"$Sdin = $Sd">,
-              Requires<[HasVFP4,DontUseNEONForFP]> {
+              Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines.
 }
 
 def : Pat<(fsub_mlx DPR:$dstin, (fmul_su DPR:$a, (f64 DPR:$b))),
           (VFMSD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP4]>;
+          Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 def : Pat<(fsub_mlx SPR:$dstin, (fmul_su SPR:$a, SPR:$b)),
           (VFMSS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP4,DontUseNEONForFP]>;
+          Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]>;
+
+// Match @llvm.fma.* intrinsics
+// (fma (fneg x), y, z) -> (vfms z, x, y)
+def : Pat<(f64 (fma (fneg DPR:$Dn), DPR:$Dm, DPR:$Ddin)),
+          (VFMSD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(f32 (fma (fneg SPR:$Sn), SPR:$Sm, SPR:$Sdin)),
+          (VFMSS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
+// (fma x, (fneg y), z) -> (vfms z, x, y)
+def : Pat<(f64 (fma DPR:$Dn, (fneg DPR:$Dm), DPR:$Ddin)),
+          (VFMSD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(f32 (fma SPR:$Sn, (fneg SPR:$Sm), SPR:$Sdin)),
+          (VFMSS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
 
 def VFNMAD : ADbI<0b11101, 0b01, 1, 0,
                   (outs DPR:$Dd), (ins DPR:$Ddin, DPR:$Dn, DPR:$Dm),
@@ -1082,7 +1546,7 @@ def VFNMAD : ADbI<0b11101, 0b01, 1, 0,
                   [(set DPR:$Dd,(fsub_mlx (fneg (fmul_su DPR:$Dn,DPR:$Dm)),
                                           (f64 DPR:$Ddin)))]>,
                 RegConstraint<"$Ddin = $Dd">,
-                Requires<[HasVFP4]>;
+                Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 
 def VFNMAS : ASbI<0b11101, 0b01, 1, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
@@ -1090,17 +1554,33 @@ def VFNMAS : ASbI<0b11101, 0b01, 1, 0,
                   [(set SPR:$Sd, (fsub_mlx (fneg (fmul_su SPR:$Sn, SPR:$Sm)),
                                            SPR:$Sdin))]>,
                 RegConstraint<"$Sdin = $Sd">,
-                Requires<[HasVFP4,DontUseNEONForFP]> {
+                Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines.
 }
 
 def : Pat<(fsub_mlx (fneg (fmul_su DPR:$a, (f64 DPR:$b))), DPR:$dstin),
           (VFNMAD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP4]>;
+          Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 def : Pat<(fsub_mlx (fneg (fmul_su SPR:$a, SPR:$b)), SPR:$dstin),
           (VFNMAS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP4,DontUseNEONForFP]>;
+          Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]>;
+
+// Match @llvm.fma.* intrinsics
+// (fneg (fma x, y, z)) -> (vfnma z, x, y)
+def : Pat<(fneg (fma (f64 DPR:$Dn), (f64 DPR:$Dm), (f64 DPR:$Ddin))),
+          (VFNMAD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(fneg (fma (f32 SPR:$Sn), (f32 SPR:$Sm), (f32 SPR:$Sdin))),
+          (VFNMAS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
+// (fma (fneg x), y, (fneg z)) -> (vfnma z, x, y)
+def : Pat<(f64 (fma (fneg DPR:$Dn), DPR:$Dm, (fneg DPR:$Ddin))),
+          (VFNMAD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(f32 (fma (fneg SPR:$Sn), SPR:$Sm, (fneg SPR:$Sdin))),
+          (VFNMAS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
 
 def VFNMSD : ADbI<0b11101, 0b01, 0, 0,
                   (outs DPR:$Dd), (ins DPR:$Ddin, DPR:$Dn, DPR:$Dm),
@@ -1108,40 +1588,66 @@ def VFNMSD : ADbI<0b11101, 0b01, 0, 0,
                   [(set DPR:$Dd, (fsub_mlx (fmul_su DPR:$Dn, DPR:$Dm),
                                            (f64 DPR:$Ddin)))]>,
                RegConstraint<"$Ddin = $Dd">,
-               Requires<[HasVFP4]>;
+               Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 
 def VFNMSS : ASbI<0b11101, 0b01, 0, 0,
                   (outs SPR:$Sd), (ins SPR:$Sdin, SPR:$Sn, SPR:$Sm),
                   IIC_fpFMAC32, "vfnms", ".f32\t$Sd, $Sn, $Sm",
              [(set SPR:$Sd, (fsub_mlx (fmul_su SPR:$Sn, SPR:$Sm), SPR:$Sdin))]>,
                          RegConstraint<"$Sdin = $Sd">,
-                  Requires<[HasVFP4,DontUseNEONForFP]> {
+                  Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]> {
   // Some single precision VFP instructions may be executed on both NEON and
   // VFP pipelines.
 }
 
 def : Pat<(fsub_mlx (fmul_su DPR:$a, (f64 DPR:$b)), DPR:$dstin),
           (VFNMSD DPR:$dstin, DPR:$a, DPR:$b)>,
-          Requires<[HasVFP4]>;
+          Requires<[HasVFP4,HasDPVFP,UseFusedMAC]>;
 def : Pat<(fsub_mlx (fmul_su SPR:$a, SPR:$b), SPR:$dstin),
           (VFNMSS SPR:$dstin, SPR:$a, SPR:$b)>,
-          Requires<[HasVFP4,DontUseNEONForFP]>;
+          Requires<[HasVFP4,DontUseNEONForFP,UseFusedMAC]>;
+
+// Match @llvm.fma.* intrinsics
+
+// (fma x, y, (fneg z)) -> (vfnms z, x, y))
+def : Pat<(f64 (fma DPR:$Dn, DPR:$Dm, (fneg DPR:$Ddin))),
+          (VFNMSD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(f32 (fma SPR:$Sn, SPR:$Sm, (fneg SPR:$Sdin))),
+          (VFNMSS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
+// (fneg (fma (fneg x), y, z)) -> (vfnms z, x, y)
+def : Pat<(fneg (f64 (fma (fneg DPR:$Dn), DPR:$Dm, DPR:$Ddin))),
+          (VFNMSD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(fneg (f32 (fma (fneg SPR:$Sn), SPR:$Sm, SPR:$Sdin))),
+          (VFNMSS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
+// (fneg (fma x, (fneg y), z) -> (vfnms z, x, y)
+def : Pat<(fneg (f64 (fma DPR:$Dn, (fneg DPR:$Dm), DPR:$Ddin))),
+          (VFNMSD DPR:$Ddin, DPR:$Dn, DPR:$Dm)>,
+      Requires<[HasVFP4,HasDPVFP]>;
+def : Pat<(fneg (f32 (fma SPR:$Sn, (fneg SPR:$Sm), SPR:$Sdin))),
+          (VFNMSS SPR:$Sdin, SPR:$Sn, SPR:$Sm)>,
+      Requires<[HasVFP4]>;
 
 //===----------------------------------------------------------------------===//
 // FP Conditional moves.
 //
 
-let neverHasSideEffects = 1 in {
-def VMOVDcc  : ARMPseudoInst<(outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm, pred:$p),
-                    4, IIC_fpUNA64,
-                    [/*(set DPR:$Dd, (ARMcmov DPR:$Dn, DPR:$Dm, imm:$cc))*/]>,
-                 RegConstraint<"$Dn = $Dd">;
+let hasSideEffects = 0 in {
+def VMOVDcc  : PseudoInst<(outs DPR:$Dd), (ins DPR:$Dn, DPR:$Dm, cmovpred:$p),
+                    IIC_fpUNA64,
+                    [(set (f64 DPR:$Dd),
+                          (ARMcmov DPR:$Dn, DPR:$Dm, cmovpred:$p))]>,
+               RegConstraint<"$Dn = $Dd">, Requires<[HasVFP2,HasDPVFP]>;
 
-def VMOVScc  : ARMPseudoInst<(outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm, pred:$p),
-                    4, IIC_fpUNA32,
-                    [/*(set SPR:$Sd, (ARMcmov SPR:$Sn, SPR:$Sm, imm:$cc))*/]>,
-                 RegConstraint<"$Sn = $Sd">;
-} // neverHasSideEffects
+def VMOVScc  : PseudoInst<(outs SPR:$Sd), (ins SPR:$Sn, SPR:$Sm, cmovpred:$p),
+                    IIC_fpUNA32,
+                    [(set (f32 SPR:$Sd),
+                          (ARMcmov SPR:$Sn, SPR:$Sm, cmovpred:$p))]>,
+               RegConstraint<"$Sn = $Sd">, Requires<[HasVFP2]>;
+} // hasSideEffects
 
 //===----------------------------------------------------------------------===//
 // Move from VFP System Register to ARM core register.
@@ -1166,9 +1672,9 @@ class MovFromVFP<bits<4> opc19_16, dag oops, dag iops, string opc, string asm,
 
 // APSR is the application level alias of CPSR. This FPSCR N, Z, C, V flags
 // to APSR.
-let Defs = [CPSR], Uses = [FPSCR], Rt = 0b1111 /* apsr_nzcv */ in
+let Defs = [CPSR], Uses = [FPSCR_NZCV], Rt = 0b1111 /* apsr_nzcv */ in
 def FMSTAT : MovFromVFP<0b0001 /* fpscr */, (outs), (ins),
-                        "vmrs", "\tapsr_nzcv, fpscr", [(arm_fmstat)]>;
+                        "vmrs", "\tAPSR_nzcv, fpscr", [(arm_fmstat)]>;
 
 // Application level FPSCR -> GPR
 let hasSideEffects = 1, Uses = [FPSCR] in
@@ -1182,6 +1688,16 @@ let Uses = [FPSCR] in {
                               "vmrs", "\t$Rt, fpexc", []>;
   def VMRS_FPSID : MovFromVFP<0b0000 /* fpsid */, (outs GPR:$Rt), (ins),
                               "vmrs", "\t$Rt, fpsid", []>;
+  def VMRS_MVFR0 : MovFromVFP<0b0111 /* mvfr0 */, (outs GPR:$Rt), (ins),
+                              "vmrs", "\t$Rt, mvfr0", []>;
+  def VMRS_MVFR1 : MovFromVFP<0b0110 /* mvfr1 */, (outs GPR:$Rt), (ins),
+                              "vmrs", "\t$Rt, mvfr1", []>;
+  def VMRS_MVFR2 : MovFromVFP<0b0101 /* mvfr2 */, (outs GPR:$Rt), (ins),
+                              "vmrs", "\t$Rt, mvfr2", []>, Requires<[HasFPARMv8]>;
+  def VMRS_FPINST : MovFromVFP<0b1001 /* fpinst */, (outs GPR:$Rt), (ins),
+                              "vmrs", "\t$Rt, fpinst", []>;
+  def VMRS_FPINST2 : MovFromVFP<0b1010 /* fpinst2 */, (outs GPR:$Rt), (ins),
+                                "vmrs", "\t$Rt, fpinst2", []>;
 }
 
 //===----------------------------------------------------------------------===//
@@ -1215,6 +1731,11 @@ let Defs = [FPSCR] in {
   // System level GPR -> FPSID
   def VMSR_FPSID : MovToVFP<0b0000 /* fpsid */, (outs), (ins GPR:$src),
                       "vmsr", "\tfpsid, $src", []>;
+
+  def VMSR_FPINST : MovToVFP<0b1001 /* fpinst */, (outs), (ins GPR:$src),
+                              "vmsr", "\tfpinst, $src", []>;
+  def VMSR_FPINST2 : MovToVFP<0b1010 /* fpinst2 */, (outs), (ins GPR:$src),
+                                "vmsr", "\tfpinst2, $src", []>;
 }
 
 //===----------------------------------------------------------------------===//
@@ -1226,7 +1747,8 @@ let isReMaterializable = 1 in {
 def FCONSTD : VFPAI<(outs DPR:$Dd), (ins vfp_f64imm:$imm),
                     VFPMiscFrm, IIC_fpUNA64,
                     "vmov", ".f64\t$Dd, $imm",
-                    [(set DPR:$Dd, vfp_f64imm:$imm)]>, Requires<[HasVFP3]> {
+                    [(set DPR:$Dd, vfp_f64imm:$imm)]>,
+              Requires<[HasVFP3,HasDPVFP]> {
   bits<5> Dd;
   bits<8> imm;
 
@@ -1263,7 +1785,7 @@ def FCONSTS : VFPAI<(outs SPR:$Sd), (ins vfp_f32imm:$imm),
 //===----------------------------------------------------------------------===//
 // Assembler aliases.
 //
-// A few mnemnoic aliases for pre-unifixed syntax. We don't guarantee to
+// A few mnemonic aliases for pre-unifixed syntax. We don't guarantee to
 // support them all, but supporting at least some of the basics is
 // good to be friendly.
 def : VFP2MnemonicAlias<"flds", "vldr">;
@@ -1304,20 +1826,27 @@ def : VFP2MnemonicAlias<"fcmps", "vcmp.f32">;
 def : VFP2MnemonicAlias<"fcmpd", "vcmp.f64">;
 def : VFP2MnemonicAlias<"fdivs", "vdiv.f32">;
 def : VFP2MnemonicAlias<"fdivd", "vdiv.f64">;
+def : VFP2MnemonicAlias<"fmrx", "vmrs">;
+def : VFP2MnemonicAlias<"fmxr", "vmsr">;
+
+// Be friendly and accept the old form of zero-compare
+def : VFP2DPInstAlias<"fcmpzd${p} $val", (VCMPZD DPR:$val, pred:$p)>;
+def : VFP2InstAlias<"fcmpzs${p} $val", (VCMPZS SPR:$val, pred:$p)>;
+
 
 def : VFP2InstAlias<"fmstat${p}", (FMSTAT pred:$p)>;
 def : VFP2InstAlias<"fadds${p} $Sd, $Sn, $Sm",
                     (VADDS SPR:$Sd, SPR:$Sn, SPR:$Sm, pred:$p)>;
-def : VFP2InstAlias<"faddd${p} $Dd, $Dn, $Dm",
-                    (VADDD DPR:$Dd, DPR:$Dn, DPR:$Dm, pred:$p)>;
+def : VFP2DPInstAlias<"faddd${p} $Dd, $Dn, $Dm",
+                      (VADDD DPR:$Dd, DPR:$Dn, DPR:$Dm, pred:$p)>;
 def : VFP2InstAlias<"fsubs${p} $Sd, $Sn, $Sm",
                     (VSUBS SPR:$Sd, SPR:$Sn, SPR:$Sm, pred:$p)>;
-def : VFP2InstAlias<"fsubd${p} $Dd, $Dn, $Dm",
-                    (VSUBD DPR:$Dd, DPR:$Dn, DPR:$Dm, pred:$p)>;
+def : VFP2DPInstAlias<"fsubd${p} $Dd, $Dn, $Dm",
+                      (VSUBD DPR:$Dd, DPR:$Dn, DPR:$Dm, pred:$p)>;
 
 // No need for the size suffix on VSQRT. It's implied by the register classes.
 def : VFP2InstAlias<"vsqrt${p} $Sd, $Sm", (VSQRTS SPR:$Sd, SPR:$Sm, pred:$p)>;
-def : VFP2InstAlias<"vsqrt${p} $Dd, $Dm", (VSQRTD DPR:$Dd, DPR:$Dm, pred:$p)>;
+def : VFP2DPInstAlias<"vsqrt${p} $Dd, $Dm", (VSQRTD DPR:$Dd, DPR:$Dm, pred:$p)>;
 
 // VLDR/VSTR accept an optional type suffix.
 def : VFP2InstAlias<"vldr${p}.32 $Sd, $addr",
@@ -1329,22 +1858,6 @@ def : VFP2InstAlias<"vldr${p}.64 $Dd, $addr",
 def : VFP2InstAlias<"vstr${p}.64 $Dd, $addr",
                     (VSTRD DPR:$Dd, addrmode5:$addr, pred:$p)>;
 
-// VMUL has a two-operand form (implied destination operand)
-def : VFP2InstAlias<"vmul${p}.f64 $Dn, $Dm",
-                    (VMULD DPR:$Dn, DPR:$Dn, DPR:$Dm, pred:$p)>;
-def : VFP2InstAlias<"vmul${p}.f32 $Sn, $Sm",
-                    (VMULS SPR:$Sn, SPR:$Sn, SPR:$Sm, pred:$p)>;
-// VADD has a two-operand form (implied destination operand)
-def : VFP2InstAlias<"vadd${p}.f64 $Dn, $Dm",
-                    (VADDD DPR:$Dn, DPR:$Dn, DPR:$Dm, pred:$p)>;
-def : VFP2InstAlias<"vadd${p}.f32 $Sn, $Sm",
-                    (VADDS SPR:$Sn, SPR:$Sn, SPR:$Sm, pred:$p)>;
-// VSUB has a two-operand form (implied destination operand)
-def : VFP2InstAlias<"vsub${p}.f64 $Dn, $Dm",
-                    (VSUBD DPR:$Dn, DPR:$Dn, DPR:$Dm, pred:$p)>;
-def : VFP2InstAlias<"vsub${p}.f32 $Sn, $Sm",
-                    (VSUBS SPR:$Sn, SPR:$Sn, SPR:$Sm, pred:$p)>;
-
 // VMOV can accept optional 32-bit or less data type suffix suffix.
 def : VFP2InstAlias<"vmov${p}.8 $Rt, $Sn",
                     (VMOVRS GPR:$Rt, SPR:$Sn, pred:$p)>;
@@ -1368,3 +1881,14 @@ def : VFP2InstAlias<"vmov${p}.f64 $Dn, $Rt, $Rt2",
 // VMOVD does.
 def : VFP2InstAlias<"vmov${p} $Sd, $Sm",
                     (VMOVS SPR:$Sd, SPR:$Sm, pred:$p)>;
+
+// FCONSTD/FCONSTS alias for vmov.f64/vmov.f32
+// These aliases provide added functionality over vmov.f instructions by
+// allowing users to write assembly containing encoded floating point constants
+// (e.g. #0x70 vs #1.0).  Without these alises there is no way for the
+// assembler to accept encoded fp constants (but the equivalent fp-literal is
+// accepted directly by vmovf).
+def : VFP3InstAlias<"fconstd${p} $Dd, $val",
+                    (FCONSTD DPR:$Dd, vfp_f64imm:$val, pred:$p)>;
+def : VFP3InstAlias<"fconsts${p} $Sd, $val",
+                    (FCONSTS SPR:$Sd, vfp_f32imm:$val, pred:$p)>;