[ARM] VFPv2 only supports 16 D registers.

[efriedma]

Summary

r361845 changed the way we handle "D16" vs. "D32" targets; there used to be a negative "d16" which removed instructions from the instruction set, and now there's a "d32" feature which adds instructions to the instruction set. This is good, but there was an oversight in the implementation: the behavior of VFPv2 was changed. In particular, the "vfp2" feature was changed to imply "d32". This is wrong: VFPv2 only supports 16 D registers.

In practice, this means if you specify -mfpu=vfpv2, the compiler will generate illegal instructions.

This patch gets rid of "vfp2d16" and "vfp2d16sp", and fixes "vfp2" and "vfp2sp" so they don't imply "d32".

(There are a couple minor corresponding changes to clang; I'll post a corresponding patch soon.)

[ostannard]

This patch looks good, but presumably needs the clang half to be committed at the same time?

I considered suggesting using vfp2d16 and vfp2d16sp instead of vfp2 and vfp2sp to be consistent with the vfp3 feature names, but that would then be inconsistent with the user-facing name, so this way is fine.

[efriedma]

Minor fix for ARMTargetParser: vfp2sp should be FPURestriction::SP_D16. I don't think this has any practical effect beyond making clang emit the "+vfp2sp" attribute in all the cases where it's legal.

Posted https://reviews.llvm.org/D67467 with the clang fix.

[ostannard]

LGTM

llvm/trunk/lib/Object/ELFObjectFile.cpp

if (Attributes.hasAttribute(ARMBuildAttrs::THUMB_ISA_use)) {
     }
   }
 
   if (Attributes.hasAttribute(ARMBuildAttrs::FP_arch)) {
     switch(Attributes.getAttributeValue(ARMBuildAttrs::FP_arch)) {
     default:
       break;
     case ARMBuildAttrs::Not_Allowed:
-      Features.AddFeature("vfp2d16sp", false);
+      Features.AddFeature("vfp2sp", false);
       Features.AddFeature("vfp3d16sp", false);
       Features.AddFeature("vfp4d16sp", false);
       break;
     case ARMBuildAttrs::AllowFPv2:
       Features.AddFeature("vfp2");
       break;
     case ARMBuildAttrs::AllowFPv3A:
     case ARMBuildAttrs::AllowFPv3B:

llvm/trunk/lib/Support/ARMTargetParser.cpp

static const struct FPUFeatureNameInfo {
     // that we can return them as static StringRefs.
     //
     // Also, the SubtargetFeatures ending in just "sp" are listed here
     // under FPURestriction::None, which is the only FPURestriction in
     // which they would be valid (since FPURestriction::SP doesn't
     // exist).
 
     {"+fpregs", "-fpregs", FPUVersion::VFPV2, FPURestriction::SP_D16},
-    {"+vfp2", "-vfp2", FPUVersion::VFPV2, FPURestriction::None},
-    {"+vfp2d16", "-vfp2d16", FPUVersion::VFPV2, FPURestriction::D16},
-    {"+vfp2d16sp", "-vfp2d16sp", FPUVersion::VFPV2, FPURestriction::SP_D16},
-    {"+vfp2sp", "-vfp2sp", FPUVersion::VFPV2, FPURestriction::None},
+    {"+vfp2", "-vfp2", FPUVersion::VFPV2, FPURestriction::D16},
+    {"+vfp2sp", "-vfp2sp", FPUVersion::VFPV2, FPURestriction::SP_D16},
     {"+vfp3", "-vfp3", FPUVersion::VFPV3, FPURestriction::None},
     {"+vfp3d16", "-vfp3d16", FPUVersion::VFPV3, FPURestriction::D16},
     {"+vfp3d16sp", "-vfp3d16sp", FPUVersion::VFPV3, FPURestriction::SP_D16},
     {"+vfp3sp", "-vfp3sp", FPUVersion::VFPV3, FPURestriction::None},
     {"+fp16", "-fp16", FPUVersion::VFPV3_FP16, FPURestriction::SP_D16},
     {"+vfp4", "-vfp4", FPUVersion::VFPV4, FPURestriction::None},
     {"+vfp4d16", "-vfp4d16", FPUVersion::VFPV4, FPURestriction::D16},
     {"+vfp4d16sp", "-vfp4d16sp", FPUVersion::VFPV4, FPURestriction::SP_D16},
     {"+vfp4sp", "-vfp4sp", FPUVersion::VFPV4, FPURestriction::None},
     {"+fp-armv8", "-fp-armv8", FPUVersion::VFPV5, FPURestriction::None},
     {"+fp-armv8d16", "-fp-armv8d16", FPUVersion::VFPV5, FPURestriction::D16},
     {"+fp-armv8d16sp", "-fp-armv8d16sp", FPUVersion::VFPV5, FPURestriction::SP_D16},
     {"+fp-armv8sp", "-fp-armv8sp", FPUVersion::VFPV5, FPURestriction::None},
     {"+fullfp16", "-fullfp16", FPUVersion::VFPV5_FULLFP16, FPURestriction::SP_D16},
     {"+fp64", "-fp64", FPUVersion::VFPV2, FPURestriction::D16},
-    {"+d32", "-d32", FPUVersion::VFPV2, FPURestriction::None},
+    {"+d32", "-d32", FPUVersion::VFPV3, FPURestriction::None},
   };
 
   for (const auto &Info: FPUFeatureInfoList) {
     if (FPUNames[FPUKind].FPUVer >= Info.MinVersion &&
         FPUNames[FPUKind].Restriction <= Info.MaxRestriction)
       Features.push_back(Info.PlusName);
     else
       Features.push_back(Info.MinusName);

llvm/trunk/lib/Target/ARM/ARM.td

def FeatureFP64           : SubtargetFeature<"fp64", "HasFP64", "true",
                                              "Floating point unit supports "
                                              "double precision",
                                              [FeatureFPRegs64]>;
 
 def FeatureD32            : SubtargetFeature<"d32", "HasD32", "true",
                                              "Extend FP to 32 double registers">;
 
 multiclass VFPver<string name, string query, string description,
-                  list<SubtargetFeature> prev = [],
-                  list<SubtargetFeature> otherimplies = []> {
+                  list<SubtargetFeature> prev,
+                  list<SubtargetFeature> otherimplies,
+                  list<SubtargetFeature> vfp2prev = []> {
   def _D16_SP: SubtargetFeature<
     name#"d16sp", query#"D16SP", "true",
     description#" with only 16 d-registers and no double precision",
-    !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16_SP")) # otherimplies>;
+    !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16_SP")) #
+      !foreach(v, vfp2prev, !cast<SubtargetFeature>(v # "_SP")) #
+      otherimplies>;
   def _SP: SubtargetFeature<
     name#"sp", query#"SP", "true",
     description#" with no double precision",
     !foreach(v, prev, !cast<SubtargetFeature>(v # "_SP")) #
       otherimplies # [FeatureD32, !cast<SubtargetFeature>(NAME # "_D16_SP")]>;
   def _D16: SubtargetFeature<
     name#"d16", query#"D16", "true",
     description#" with only 16 d-registers",
     !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16")) #
+      vfp2prev #
       otherimplies # [FeatureFP64, !cast<SubtargetFeature>(NAME # "_D16_SP")]>;
   def "": SubtargetFeature<
     name, query, "true", description,
     prev # otherimplies # [
         !cast<SubtargetFeature>(NAME # "_D16"),
         !cast<SubtargetFeature>(NAME # "_SP")]>;
 }
 
-defm FeatureVFP2: VFPver<"vfp2", "HasVFPv2", "Enable VFP2 instructions",
-                         [], [FeatureFPRegs]>;
+def FeatureVFP2_SP        : SubtargetFeature<"vfp2sp", "HasVFPv2SP", "true",
+                                             "Enable VFP2 instructions with "
+                                             "no double precision",
+                                             [FeatureFPRegs]>;
+
+def FeatureVFP2           : SubtargetFeature<"vfp2", "HasVFPv2", "true",
+                                             "Enable VFP2 instructions",
+                                             [FeatureFP64, FeatureVFP2_SP]>;
 
 defm FeatureVFP3: VFPver<"vfp3", "HasVFPv3", "Enable VFP3 instructions",
-                         [FeatureVFP2]>;
+                         [], [], [FeatureVFP2]>;
 
 def FeatureNEON           : SubtargetFeature<"neon", "HasNEON", "true",
                                              "Enable NEON instructions",
                                              [FeatureVFP3]>;
 
 def FeatureFP16           : SubtargetFeature<"fp16", "HasFP16", "true",
                                              "Enable half-precision "
                                              "floating point">;
 
 defm FeatureVFP4: VFPver<"vfp4", "HasVFPv4", "Enable VFP4 instructions",
                          [FeatureVFP3], [FeatureFP16]>;
 
 defm FeatureFPARMv8: VFPver<"fp-armv8", "HasFPARMv8", "Enable ARMv8 FP",
-                         [FeatureVFP4]>;
+                         [FeatureVFP4], []>;
 
 def FeatureFullFP16       : SubtargetFeature<"fullfp16", "HasFullFP16", "true",
                                              "Enable full half-precision "
                                              "floating point",
                                              [FeatureFPARMv8_D16_SP, FeatureFPRegs16]>;
 
 def FeatureFP16FML        : SubtargetFeature<"fp16fml", "HasFP16FML", "true",
                                              "Enable full half-precision "

llvm/trunk/lib/Target/ARM/ARMPredicates.td

 def HasV8_3a         : Predicate<"Subtarget->hasV8_3aOps()">,
                                  AssemblerPredicate<"HasV8_3aOps", "armv8.3a">;
 def HasV8_4a         : Predicate<"Subtarget->hasV8_4aOps()">,
                                  AssemblerPredicate<"HasV8_4aOps", "armv8.4a">;
 def HasV8_5a         : Predicate<"Subtarget->hasV8_5aOps()">,
                                  AssemblerPredicate<"HasV8_5aOps", "armv8.5a">;
 def NoVFP            : Predicate<"!Subtarget->hasVFP2Base()">;
 def HasVFP2          : Predicate<"Subtarget->hasVFP2Base()">,
-                                 AssemblerPredicate<"FeatureVFP2_D16_SP", "VFP2">;
+                                 AssemblerPredicate<"FeatureVFP2_SP", "VFP2">;
 def HasVFP3          : Predicate<"Subtarget->hasVFP3Base()">,
                                  AssemblerPredicate<"FeatureVFP3_D16_SP", "VFP3">;
 def HasVFP4          : Predicate<"Subtarget->hasVFP4Base()">,
                                  AssemblerPredicate<"FeatureVFP4_D16_SP", "VFP4">;
 def HasDPVFP         : Predicate<"Subtarget->hasFP64()">,
                                  AssemblerPredicate<"FeatureFP64",
                                                     "double precision VFP">;
 def HasFPARMv8       : Predicate<"Subtarget->hasFPARMv8Base()">,

llvm/trunk/lib/Target/ARM/ARMSubtarget.h

protected:
   bool HasFPRegs64 = false;
 
   /// Versions of the VFP flags restricted to single precision, or to
   /// 16 d-registers, or both.
   bool HasVFPv2SP = false;
   bool HasVFPv3SP = false;
   bool HasVFPv4SP = false;
   bool HasFPARMv8SP = false;
-  bool HasVFPv2D16 = false;
   bool HasVFPv3D16 = false;
   bool HasVFPv4D16 = false;
   bool HasFPARMv8D16 = false;
-  bool HasVFPv2D16SP = false;
   bool HasVFPv3D16SP = false;
   bool HasVFPv4D16SP = false;
   bool HasFPARMv8D16SP = false;
 
   /// HasDotProd - True if the ARMv8.2A dot product instructions are supported.
   bool HasDotProd = false;
 
   /// UseNEONForSinglePrecisionFP - if the NEONFP attribute has been
public:
   bool isCortexM3() const { return ARMProcFamily == CortexM3; }
   bool isLikeA9() const { return isCortexA9() || isCortexA15() || isKrait(); }
   bool isCortexR5() const { return ARMProcFamily == CortexR5; }
   bool isKrait() const { return ARMProcFamily == Krait; }
   /// @}
 
   bool hasARMOps() const { return !NoARM; }
 
-  bool hasVFP2Base() const { return HasVFPv2D16SP; }
+  bool hasVFP2Base() const { return HasVFPv2SP; }
   bool hasVFP3Base() const { return HasVFPv3D16SP; }
   bool hasVFP4Base() const { return HasVFPv4D16SP; }
   bool hasFPARMv8Base() const { return HasFPARMv8D16SP; }
   bool hasNEON() const { return HasNEON;  }
   bool hasSHA2() const { return HasSHA2; }
   bool hasAES() const { return HasAES; }
   bool hasCrypto() const { return HasCrypto; }
   bool hasDotProd() const { return HasDotProd; }

llvm/trunk/lib/Target/ARM/AsmParser/ARMAsmParser.cpp

static const struct {
     const unsigned Kind;
     const FeatureBitset ArchCheck;
     const FeatureBitset Features;
   } Extensions[] = {
     { ARM::AEK_CRC, {Feature_HasV8Bit}, {ARM::FeatureCRC} },
     { ARM::AEK_CRYPTO,  {Feature_HasV8Bit},
       {ARM::FeatureCrypto, ARM::FeatureNEON, ARM::FeatureFPARMv8} },
     { ARM::AEK_FP, {Feature_HasV8Bit},
-      {ARM::FeatureVFP2_D16_SP, ARM::FeatureFPARMv8} },
+      {ARM::FeatureVFP2_SP, ARM::FeatureFPARMv8} },
     { (ARM::AEK_HWDIVTHUMB | ARM::AEK_HWDIVARM),
       {Feature_HasV7Bit, Feature_IsNotMClassBit},
       {ARM::FeatureHWDivThumb, ARM::FeatureHWDivARM} },
     { ARM::AEK_MP, {Feature_HasV7Bit, Feature_IsNotMClassBit},
       {ARM::FeatureMP} },
     { ARM::AEK_SIMD, {Feature_HasV8Bit},
-      {ARM::FeatureNEON, ARM::FeatureVFP2_D16_SP, ARM::FeatureFPARMv8} },
+      {ARM::FeatureNEON, ARM::FeatureVFP2_SP, ARM::FeatureFPARMv8} },
     { ARM::AEK_SEC, {Feature_HasV6KBit}, {ARM::FeatureTrustZone} },
     // FIXME: Only available in A-class, isel not predicated
     { ARM::AEK_VIRT, {Feature_HasV7Bit}, {ARM::FeatureVirtualization} },
     { ARM::AEK_FP16, {Feature_HasV8_2aBit},
       {ARM::FeatureFPARMv8, ARM::FeatureFullFP16} },
     { ARM::AEK_RAS, {Feature_HasV8Bit}, {ARM::FeatureRAS} },
     { ARM::AEK_LOB, {Feature_HasV8_1MMainlineBit}, {ARM::FeatureLOB} },
     // FIXME: Unsupported extensions.

llvm/trunk/lib/Target/ARM/MCTargetDesc/ARMTargetStreamer.cpp

else if (STI.hasFeature(ARM::FeatureVFP3_D16_SP))
                                                   : ARM::FK_VFPV3)
               // -d32
               : (STI.hasFeature(ARM::FeatureFP64)
                      ? (STI.hasFeature(ARM::FeatureFP16)
                             ? ARM::FK_VFPV3_D16_FP16
                             : ARM::FK_VFPV3_D16)
                      : (STI.hasFeature(ARM::FeatureFP16) ? ARM::FK_VFPV3XD_FP16
                                                          : ARM::FK_VFPV3XD)));
-    else if (STI.hasFeature(ARM::FeatureVFP2_D16_SP))
+    else if (STI.hasFeature(ARM::FeatureVFP2_SP))
       emitFPU(ARM::FK_VFPV2);
   }
 
   // ABI_HardFP_use attribute to indicate single precision FP.
-  if (STI.hasFeature(ARM::FeatureVFP2_D16_SP) && !STI.hasFeature(ARM::FeatureFP64))
+  if (STI.hasFeature(ARM::FeatureVFP2_SP) && !STI.hasFeature(ARM::FeatureFP64))
     emitAttribute(ARMBuildAttrs::ABI_HardFP_use,
                   ARMBuildAttrs::HardFPSinglePrecision);
 
   if (STI.hasFeature(ARM::FeatureFP16))
     emitAttribute(ARMBuildAttrs::FP_HP_extension, ARMBuildAttrs::AllowHPFP);
 
   if (STI.hasFeature(ARM::FeatureMP))
     emitAttribute(ARMBuildAttrs::MPextension_use, ARMBuildAttrs::AllowMP);

llvm/trunk/test/CodeGen/ARM/GlobalISel/arm-legalize-load-store.mir

 # RUN: llc -mtriple arm-- -run-pass=legalizer %s -o - | FileCheck %s
 # RUN: llc -mtriple thumbv7-- -run-pass=legalizer %s -o - | FileCheck %s
 --- |
   define void @test_legal_loads_stores() { ret void }
   define void @test_load_from_stack() { ret void }
 
   define void @test_load_store_64_vfp() #0 { ret void }
   define void @test_load_store_64_novfp() #1 { ret void }
 
   define void @test_gep_s32() { ret void }
   define void @test_gep_s16() { ret void }
 
   attributes #0 = { "target-features"="+vfp2" }
-  attributes #1 = { "target-features"="-vfp2d16sp" }
+  attributes #1 = { "target-features"="-vfp2sp" }
 ...
 ---
 name:            test_legal_loads_stores
 # CHECK-LABEL: name: test_legal_loads_stores
 legalized:       false
 # CHECK: legalized: true
 regBankSelected: false
 selected:        false

llvm/trunk/test/CodeGen/ARM/build-attributes.ll

 ; RUN: llc < %s -mtriple=armv8-linux-gnueabi | FileCheck %s --check-prefix=V8-FPARMv8-NEON-CRYPTO
 ; RUN: llc < %s -mtriple=thumbv8m.base-linux-gnueabi | FileCheck %s --check-prefix=V8MBASELINE
 ; RUN: llc < %s -mtriple=thumbv8m.main-linux-gnueabi | FileCheck %s --check-prefix=V8MMAINLINE
 ; RUN: llc < %s -mtriple=thumbv8m.main-linux-gnueabi -mattr=+dsp | FileCheck %s --check-prefix=V8MMAINLINE_DSP
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 | FileCheck %s --check-prefix=CORTEX-A5-DEFAULT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A5-DEFAULT-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 -mattr=-neon,-d32 | FileCheck %s --check-prefix=CORTEX-A5-NONEON
-; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 -mattr=-vfp2d16sp | FileCheck %s --check-prefix=CORTEX-A5-NOFPU
-; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 -mattr=-vfp2d16sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A5-NOFPU-FAST
+; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 -mattr=-vfp2sp | FileCheck %s --check-prefix=CORTEX-A5-NOFPU
+; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a5 -mattr=-vfp2sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A5-NOFPU-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a8 -float-abi=soft | FileCheck %s --check-prefix=CORTEX-A8-SOFT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a8 -float-abi=soft  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A8-SOFT-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a8 -float-abi=hard | FileCheck %s --check-prefix=CORTEX-A8-HARD
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a8 -float-abi=hard  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A8-HARD-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a8 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a8 -float-abi=soft | FileCheck %s --check-prefix=CORTEX-A8-SOFT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a9 -float-abi=soft | FileCheck %s --check-prefix=CORTEX-A9-SOFT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a9 -float-abi=soft  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A9-SOFT-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a9 -float-abi=hard | FileCheck %s --check-prefix=CORTEX-A9-HARD
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a9 -float-abi=hard  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A9-HARD-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a9 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12 | FileCheck %s --check-prefix=CORTEX-A12-DEFAULT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a9 -float-abi=soft | FileCheck %s --check-prefix=CORTEX-A9-SOFT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A12-DEFAULT-FAST
-; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12 -mattr=-vfp2d16sp | FileCheck %s --check-prefix=CORTEX-A12-NOFPU
-; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12 -mattr=-vfp2d16sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A12-NOFPU-FAST
+; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12 -mattr=-vfp2sp | FileCheck %s --check-prefix=CORTEX-A12-NOFPU
+; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12 -mattr=-vfp2sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A12-NOFPU-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a12 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15 | FileCheck %s --check-prefix=CORTEX-A15
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A15-FAST
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a17 | FileCheck %s --check-prefix=CORTEX-A17-DEFAULT
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a17  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A17-FAST
-; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a17 -mattr=-vfp2d16sp | FileCheck %s --check-prefix=CORTEX-A17-NOFPU
-; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a17 -mattr=-vfp2d16sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A17-NOFPU-FAST
+; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a17 -mattr=-vfp2sp | FileCheck %s --check-prefix=CORTEX-A17-NOFPU
+; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a17 -mattr=-vfp2sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A17-NOFPU-FAST
 
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15 -enable-no-trapping-fp-math | FileCheck %s --check-prefix=NO-TRAPPING-MATH
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15 -denormal-fp-math=ieee | FileCheck %s --check-prefix=DENORMAL-IEEE
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15 -denormal-fp-math=preserve-sign | FileCheck %s --check-prefix=DENORMAL-PRESERVE-SIGN
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mcpu=cortex-a15 -denormal-fp-math=positive-zero | FileCheck %s --check-prefix=DENORMAL-POSITIVE-ZERO
 
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mattr=-neon,+vfp3,+fp16 | FileCheck %s --check-prefix=GENERIC-FPU-VFPV3-FP16
 ; RUN: llc < %s -mtriple=armv7-linux-gnueabi -mattr=-neon,+vfp3,-d32,+fp16 | FileCheck %s --check-prefix=GENERIC-FPU-VFPV3-D16-FP16

 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=sc300 | FileCheck %s --check-prefix=SC300
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=sc300  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=SC300-FAST
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=sc300 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=cortex-m4 -float-abi=soft | FileCheck %s --check-prefix=CORTEX-M4-SOFT
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=cortex-m4 -float-abi=soft  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-M4-SOFT-FAST
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=cortex-m4 -float-abi=hard | FileCheck %s --check-prefix=CORTEX-M4-HARD
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=cortex-m4 -float-abi=hard  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-M4-HARD-FAST
 ; RUN: llc < %s -mtriple=thumbv7m-linux-gnueabi -mcpu=cortex-m4 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
-; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -mattr=-vfp2d16sp | FileCheck %s --check-prefix=CORTEX-M7 --check-prefix=CORTEX-M7-SOFT
-; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -mattr=-vfp2d16sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-M7-NOFPU-FAST
+; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -mattr=-vfp2sp | FileCheck %s --check-prefix=CORTEX-M7 --check-prefix=CORTEX-M7-SOFT
+; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -mattr=-vfp2sp  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-M7-NOFPU-FAST
 ; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -mattr=-fp64 | FileCheck %s --check-prefix=CORTEX-M7 --check-prefix=CORTEX-M7-SINGLE
 ; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -mattr=-fp64  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-M7-FAST
 ; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 | FileCheck %s --check-prefix=CORTEX-M7-DOUBLE
 ; RUN: llc < %s -mtriple=thumbv7em-linux-gnueabi -mcpu=cortex-m7 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=thumbv8-linux-gnueabi -mcpu=cortex-m23 | FileCheck %s --check-prefix=CORTEX-M23
 ; RUN: llc < %s -mtriple=thumbv8-linux-gnueabi -mcpu=cortex-m33 | FileCheck %s --check-prefix=CORTEX-M33
 ; RUN: llc < %s -mtriple=thumbv8-linux-gnueabi -mcpu=cortex-m33 -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-M33-FAST
 ; RUN: llc < %s -mtriple=thumbv8-linux-gnueabi -mcpu=cortex-m33 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING

 ; RUN: llc < %s -mtriple=armv8-linux-gnueabi -mcpu=exynos-m4 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv8-linux-gnueabi -mcpu=exynos-m5 | FileCheck %s --check-prefix=EXYNOS-M5
 ; RUN: llc < %s -mtriple=armv8-linux-gnueabi -mcpu=exynos-m5  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=EXYNOS-M1-FAST
 ; RUN: llc < %s -mtriple=armv8-linux-gnueabi -mcpu=exynos-m5 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv8.1a-linux-gnueabi  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=GENERIC-ARMV8_1-A-FAST
 ; RUN: llc < %s -mtriple=armv8.1a-linux-gnueabi -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 | FileCheck %s  --check-prefix=CORTEX-A7-CHECK
 ; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s  --check-prefix=CORTEX-A7-CHECK-FAST
-; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=-vfp2d16sp,-vfp3,-vfp4,-neon,-fp16 | FileCheck %s --check-prefix=CORTEX-A7-NOFPU
-; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=-vfp2d16sp,-vfp3,-vfp4,-neon,-fp16  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A7-NOFPU-FAST
+; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=-vfp2sp,-vfp3,-vfp4,-neon,-fp16 | FileCheck %s --check-prefix=CORTEX-A7-NOFPU
+; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=-vfp2sp,-vfp3,-vfp4,-neon,-fp16  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A7-NOFPU-FAST
 ; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=+vfp4,-neon | FileCheck %s --check-prefix=CORTEX-A7-FPUV4
 ; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -enable-sign-dependent-rounding-fp-math | FileCheck %s --check-prefix=DYN-ROUNDING
 ; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=+vfp4,-neon  -enable-unsafe-fp-math -frame-pointer=all -enable-no-infs-fp-math -enable-no-nans-fp-math -fp-contract=fast | FileCheck %s --check-prefix=CORTEX-A7-FPUV4-FAST
 ; RUN: llc < %s -mtriple=armv7-none-linux-gnueabi -mcpu=cortex-a7 -mattr=+vfp4,,-d32,-neon | FileCheck %s --check-prefix=CORTEX-A7-FPUV4
 ; RUN: llc < %s -mtriple=arm-none-linux-gnueabi -mattr=+strict-align -relocation-model=pic | FileCheck %s --check-prefix=RELOC-PIC
 ; RUN: llc < %s -mtriple=arm-none-linux-gnueabi -mattr=+strict-align -relocation-model=static | FileCheck %s --check-prefix=RELOC-OTHER
 ; RUN: llc < %s -mtriple=arm-none-linux-gnueabi -mattr=+strict-align -relocation-model=dynamic-no-pic | FileCheck %s --check-prefix=RELOC-OTHER
 ; RUN: llc < %s -mtriple=arm-none-linux-gnueabi -mattr=+strict-align | FileCheck %s --check-prefix=RELOC-OTHER

 ; RUN: llc < %s -mtriple=thumb-none-linux-gnueabi -mattr=+strict-align -mcpu=cortex-m0 | FileCheck %s --check-prefix=STRICT-ALIGN
 ; RUN: llc < %s -mtriple=thumbv6m-none-linux-gnueabi -mattr=+strict-align | FileCheck %s --check-prefix=STRICT-ALIGN
 ; RUN: llc < %s -mtriple=thumb-none-linux-gnueabi -mcpu=cortex-m0 -mattr=+strict-align | FileCheck %s --check-prefix=STRICT-ALIGN
 ; ARMv5
 ; RUN: llc < %s -mtriple=armv5-none-linux-gnueabi -mcpu=arm1022e | FileCheck %s --check-prefix=NO-STRICT-ALIGN
 ; RUN: llc < %s -mtriple=armv5-none-linux-gnueabi -mcpu=arm1022e -mattr=+strict-align | FileCheck %s --check-prefix=STRICT-ALIGN
 
 ; ARMv8-R
-; RUN: llc < %s -mtriple=arm-none-none-eabi -mcpu=cortex-r52 -mattr=-vfp2d16sp,-fp16 | FileCheck %s --check-prefix=ARMv8R --check-prefix=ARMv8R-NOFPU
+; RUN: llc < %s -mtriple=arm-none-none-eabi -mcpu=cortex-r52 -mattr=-vfp2sp,-fp16 | FileCheck %s --check-prefix=ARMv8R --check-prefix=ARMv8R-NOFPU
 ; RUN: llc < %s -mtriple=arm-none-none-eabi -mcpu=cortex-r52 -mattr=-neon,-fp64,-d32 | FileCheck %s --check-prefix=ARMv8R --check-prefix=ARMv8R-SP
 ; RUN: llc < %s -mtriple=arm-none-none-eabi -mcpu=cortex-r52 | FileCheck %s --check-prefix=ARMv8R --check-prefix=ARMv8R-NEON
 
 ; ARMv8-M
 ; RUN: llc < %s -mtriple=thumbv8-none-none-eabi -mcpu=cortex-m23 | FileCheck %s --check-prefix=STRICT-ALIGN
 ; RUN: llc < %s -mtriple=thumbv8-none-none-eabi -mcpu=cortex-m33 | FileCheck %s --check-prefix=NO-STRICT-ALIGN
 ; RUN: llc < %s -mtriple=thumbv8-none-none-eabi -mcpu=cortex-m33 -mattr=+strict-align | FileCheck %s --check-prefix=STRICT-ALIGN
 ; RUN: llc < %s -mtriple=thumbv8-none-none-eabi -mcpu=cortex-m35p | FileCheck %s --check-prefix=NO-STRICT-ALIGN

llvm/trunk/test/CodeGen/ARM/inlineasm-X-allocation.ll

-; RUN: llc -mtriple=armv7-none-eabi -mattr=-neon,-vfp2d16sp %s -o - | FileCheck %s  -check-prefixes=COMMON,NOVFP
+; RUN: llc -mtriple=armv7-none-eabi -mattr=-neon,-vfp2sp %s -o - | FileCheck %s  -check-prefixes=COMMON,NOVFP
 ; RUN: llc -mtriple=armv7-none-eabi -mattr=+neon %s -float-abi=hard -o - | FileCheck %s -check-prefixes=COMMON,VFP
 
 ; The intent here is to test "X", which says that any operand whatsoever is allowed.
 ; Using this mechanism, we want to test toggling allocating GPR or SPR registers
 ; depending on whether the float registers are available. Thus, the mnemonic is
 ; totally irrelevant here, which is why we use FOO and also comment it out using "@"
 ; to avoid assembler errors.
 

llvm/trunk/test/CodeGen/ARM/softfp-constant-comparison.ll

land.rhs:                                         ; preds = %entry
   br label %land.end
 
 land.end:                                         ; preds = %land.rhs, %entry
   %1 = phi i1 [ false, %entry ], [ false, %land.rhs ]
   %land.ext = zext i1 %1 to i32
   ret void
 }
 
-attributes #0 = { noinline nounwind optnone "correctly-rounded-divide-sqrt-fp-math"="false" "denormal-fp-math"="preserve-sign" "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="true" "no-jump-tables"="false" "no-nans-fp-math"="true" "no-signed-zeros-fp-math"="true" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="cortex-m4" "target-features"="+armv7e-m,+dsp,+fp16,+fpregs,+hwdiv,+thumb-mode,+vfp2d16sp,+vfp3d16sp,+vfp4d16sp,-aes,-crc,-crypto,-dotprod,-fp16fml,-fullfp16,-hwdiv-arm,-lob,-mve,-mve.fp,-ras,-sb,-sha2" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #0 = { noinline nounwind optnone "correctly-rounded-divide-sqrt-fp-math"="false" "denormal-fp-math"="preserve-sign" "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="true" "no-jump-tables"="false" "no-nans-fp-math"="true" "no-signed-zeros-fp-math"="true" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="cortex-m4" "target-features"="+armv7e-m,+dsp,+fp16,+fpregs,+hwdiv,+thumb-mode,+vfp2sp,+vfp3d16sp,+vfp4d16sp,-aes,-crc,-crypto,-dotprod,-fp16fml,-fullfp16,-hwdiv-arm,-lob,-mve,-mve.fp,-ras,-sb,-sha2" "unsafe-fp-math"="false" "use-soft-float"="false" }

llvm/trunk/test/MC/ARM/vfp-aliases-diagnostics.s

 @ RUN: not llvm-mc -triple armv7-eabi -filetype asm -o /dev/null %s 2>&1 \
 @ RUN:   | FileCheck %s
 
   .syntax unified
   .fpu vfp
 
   .type aliases,%function
 aliases:
   fstmeax sp!, {s0}
   fldmfdx sp!, {s0}
 
   fstmfdx sp!, {s0}
   fldmeax sp!, {s0}
 
 @ CHECK-LABEL: aliases
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fstmeax sp!, {s0}
 @ CHECK:                     ^
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fldmfdx sp!, {s0}
 @ CHECK:                     ^
 
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fstmfdx sp!, {s0}
 @ CHECK:                     ^
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fldmeax sp!, {s0}
 @ CHECK:                     ^
 
   fstmiaxcs r0, {s0}
   fstmiaxhs r0, {s0}
   fstmiaxls r0, {s0}
   fstmiaxvs r0, {s0}
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fstmiaxcs r0, {s0}
 @ CHECK:                      ^
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fstmiaxhs r0, {s0}
 @ CHECK:                      ^
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fstmiaxls r0, {s0}
 @ CHECK:                      ^
-@ CHECK: error: operand must be a list of registers in range [d0, d31]
+@ CHECK: error: operand must be a list of registers in range [d0, d15]
 @ CHECK:  fstmiaxvs r0, {s0}
 @ CHECK:                      ^