[CUDA] Simplify GPU variant handling. NFC.

[tra]

Summary

Instead of hardcoding individual GPU mappings in multiple functions, keep them
all in one table and use it to look up the mappings.

We also don't care about 'virtual' architecture much, so the API is trimmed down
down to a simpler GPU->Virtual arch name lookup.

[tra]

Enumerate all known GPU variants during libdevice detection instead of
hardcoding them.

[yaxunl]

clang/lib/Basic/Cuda.cpp

61

Thanks for the efforts. Really appreciate this.

Can we separate this part to a def file and include it here and also in enum class CudaArch? Then we only need to maintain the def file.

Thanks.

[tra]

clang/lib/Basic/Cuda.cpp

61

If we had similar per-GPU code in many places, then it would make a lot of sense.
At the moment I'm not sure it would be worth it -- I don't see any other code that would benefit, other than this file.

[yaxunl]

LGTM. Thanks!

clang/include/clang/Basic/Cuda.h

enum class CudaArch {
   GFX906,
   GFX908,
   GFX909,
   GFX1010,
   GFX1011,
   GFX1012,
   LAST,
 };
+
+static inline bool IsNVIDIAGpuArch(CudaArch A) {
+  return A >= CudaArch::SM_20 && A < CudaArch::GFX600;
+}
+
+static inline bool IsAMDGpuArch(CudaArch A) {
+  return A >= CudaArch::GFX600 && A < CudaArch::LAST;
+}
+
 const char *CudaArchToString(CudaArch A);
+const char *CudaArchToVirtualArchString(CudaArch A);
 
 // The input should have the form "sm_20".
 CudaArch StringToCudaArch(llvm::StringRef S);
 
-enum class CudaVirtualArch {
-  UNKNOWN,
-  COMPUTE_20,
-  COMPUTE_30,
-  COMPUTE_32,
-  COMPUTE_35,
-  COMPUTE_37,
-  COMPUTE_50,
-  COMPUTE_52,
-  COMPUTE_53,
-  COMPUTE_60,
-  COMPUTE_61,
-  COMPUTE_62,
-  COMPUTE_70,
-  COMPUTE_72,
-  COMPUTE_75,
-  COMPUTE_AMDGCN,
-};
-const char *CudaVirtualArchToString(CudaVirtualArch A);
-
-// The input should have the form "compute_20".
-CudaVirtualArch StringToCudaVirtualArch(llvm::StringRef S);
-
-/// Get the compute_xx corresponding to an sm_yy.
-CudaVirtualArch VirtualArchForCudaArch(CudaArch A);
-
 /// Get the earliest CudaVersion that supports the given CudaArch.
 CudaVersion MinVersionForCudaArch(CudaArch A);
 
 /// Get the latest CudaVersion that supports the given CudaArch.
 CudaVersion MaxVersionForCudaArch(CudaArch A);
 
 //  Various SDK-dependent features that affect CUDA compilation
 enum class CudaFeature {

clang/lib/Basic/Cuda.cpp

return llvm::StringSwitch<CudaVersion>(S.str())
       .Case("9.0", CudaVersion::CUDA_90)
       .Case("9.1", CudaVersion::CUDA_91)
       .Case("9.2", CudaVersion::CUDA_92)
       .Case("10.0", CudaVersion::CUDA_100)
       .Case("10.1", CudaVersion::CUDA_101)
       .Default(CudaVersion::UNKNOWN);
 }
 
+struct CudaArchToStringMap {
+  CudaArch arch;
+  const char *arch_name;
+  const char *virtual_arch_name;
+};
+
+#define SM2(sm, ca)                                                            \
+  { CudaArch::SM_##sm, "sm_" #sm, ca }
+#define SM(sm) SM2(sm, "compute_" #sm)
+#define GFX(gpu)                                                               \
+  { CudaArch::GFX##gpu, "gfx" #gpu, "compute_amdgcn" }
+CudaArchToStringMap arch_names[] = {
+    // clang-format off
+    SM2(20, "compute_20"), SM2(21, "compute_20"), // Fermi
+    SM(30), SM(32), SM(35), SM(37),  // Kepler
+    SM(50), SM(52), SM(53),          // Maxwell
+    SM(60), SM(61), SM(62),          // Pascal
+    SM(70), SM(72),                  // Volta
+    SM(75),                          // Turing
+    GFX(600), // tahiti
+    GFX(601), // pitcairn, verde, oland,hainan
+    GFX(700), // kaveri
+    GFX(701), // hawaii
+    GFX(702), // 290,290x,R390,R390x
+    GFX(703), // kabini mullins
+    GFX(704), // bonaire
+    GFX(801), // carrizo
+    GFX(802), // tonga,iceland
+    GFX(803), // fiji,polaris10
+    GFX(810), // stoney
+    GFX(900), // vega, instinct
+    GFX(902), GFX(904), GFX(906), GFX(908), GFX(909),
+    GFX(1010), GFX(1011), GFX(1012),
+    // clang-format on
+};
+#undef SM
+#undef SM2
+#undef GFX
+
 const char *CudaArchToString(CudaArch A) {
-  switch (A) {
-  case CudaArch::LAST:
-    break;
-  case CudaArch::UNKNOWN:
+  auto result = std::find_if(
+      std::begin(arch_names), std::end(arch_names),
+      [A](const CudaArchToStringMap &map) { return A == map.arch; });
+  if (result == std::end(arch_names))
     return "unknown";
-  case CudaArch::SM_20:
-    return "sm_20";
-  case CudaArch::SM_21:
-    return "sm_21";
-  case CudaArch::SM_30:
-    return "sm_30";
-  case CudaArch::SM_32:
-    return "sm_32";
-  case CudaArch::SM_35:
-    return "sm_35";
-  case CudaArch::SM_37:
-    return "sm_37";
-  case CudaArch::SM_50:
-    return "sm_50";
-  case CudaArch::SM_52:
-    return "sm_52";
-  case CudaArch::SM_53:
-    return "sm_53";
-  case CudaArch::SM_60:
-    return "sm_60";
-  case CudaArch::SM_61:
-    return "sm_61";
-  case CudaArch::SM_62:
-    return "sm_62";
-  case CudaArch::SM_70:
-    return "sm_70";
-  case CudaArch::SM_72:
-    return "sm_72";
-  case CudaArch::SM_75:
-    return "sm_75";
-  case CudaArch::GFX600: // tahiti
-    return "gfx600";
-  case CudaArch::GFX601: // pitcairn, verde, oland,hainan
-    return "gfx601";
-  case CudaArch::GFX700: // kaveri
-    return "gfx700";
-  case CudaArch::GFX701: // hawaii
-    return "gfx701";
-  case CudaArch::GFX702: // 290,290x,R390,R390x
-    return "gfx702";
-  case CudaArch::GFX703: // kabini mullins
-    return "gfx703";
-  case CudaArch::GFX704: // bonaire
-    return "gfx704";
-  case CudaArch::GFX801: // carrizo
-    return "gfx801";
-  case CudaArch::GFX802: // tonga,iceland
-    return "gfx802";
-  case CudaArch::GFX803: // fiji,polaris10
-    return "gfx803";
-  case CudaArch::GFX810: // stoney
-    return "gfx810";
-  case CudaArch::GFX900: // vega, instinct
-    return "gfx900";
-  case CudaArch::GFX902: // TBA
-    return "gfx902";
-  case CudaArch::GFX904: // TBA
-    return "gfx904";
-  case CudaArch::GFX906: // TBA
-    return "gfx906";
-  case CudaArch::GFX908: // TBA
-    return "gfx908";
-  case CudaArch::GFX909: // TBA
-    return "gfx909";
-  case CudaArch::GFX1010: // TBA
-    return "gfx1010";
-  case CudaArch::GFX1011: // TBA
-    return "gfx1011";
-  case CudaArch::GFX1012: // TBA
-    return "gfx1012";
-  }
-  llvm_unreachable("invalid enum");
-}
-
-CudaArch StringToCudaArch(llvm::StringRef S) {
-  return llvm::StringSwitch<CudaArch>(S)
-      .Case("sm_20", CudaArch::SM_20)
-      .Case("sm_21", CudaArch::SM_21)
-      .Case("sm_30", CudaArch::SM_30)
-      .Case("sm_32", CudaArch::SM_32)
-      .Case("sm_35", CudaArch::SM_35)
-      .Case("sm_37", CudaArch::SM_37)
-      .Case("sm_50", CudaArch::SM_50)
-      .Case("sm_52", CudaArch::SM_52)
-      .Case("sm_53", CudaArch::SM_53)
-      .Case("sm_60", CudaArch::SM_60)
-      .Case("sm_61", CudaArch::SM_61)
-      .Case("sm_62", CudaArch::SM_62)
-      .Case("sm_70", CudaArch::SM_70)
-      .Case("sm_72", CudaArch::SM_72)
-      .Case("sm_75", CudaArch::SM_75)
-      .Case("gfx600", CudaArch::GFX600)
-      .Case("gfx601", CudaArch::GFX601)
-      .Case("gfx700", CudaArch::GFX700)
-      .Case("gfx701", CudaArch::GFX701)
-      .Case("gfx702", CudaArch::GFX702)
-      .Case("gfx703", CudaArch::GFX703)
-      .Case("gfx704", CudaArch::GFX704)
-      .Case("gfx801", CudaArch::GFX801)
-      .Case("gfx802", CudaArch::GFX802)
-      .Case("gfx803", CudaArch::GFX803)
-      .Case("gfx810", CudaArch::GFX810)
-      .Case("gfx900", CudaArch::GFX900)
-      .Case("gfx902", CudaArch::GFX902)
-      .Case("gfx904", CudaArch::GFX904)
-      .Case("gfx906", CudaArch::GFX906)
-      .Case("gfx908", CudaArch::GFX908)
-      .Case("gfx909", CudaArch::GFX909)
-      .Case("gfx1010", CudaArch::GFX1010)
-      .Case("gfx1011", CudaArch::GFX1011)
-      .Case("gfx1012", CudaArch::GFX1012)
-      .Default(CudaArch::UNKNOWN);
+  return result->arch_name;
 }
 
-const char *CudaVirtualArchToString(CudaVirtualArch A) {
-  switch (A) {
-  case CudaVirtualArch::UNKNOWN:
+const char *CudaArchToVirtualArchString(CudaArch A) {
+  auto result = std::find_if(
+      std::begin(arch_names), std::end(arch_names),
+      [A](const CudaArchToStringMap &map) { return A == map.arch; });
+  if (result == std::end(arch_names))
     return "unknown";
-  case CudaVirtualArch::COMPUTE_20:
-    return "compute_20";
-  case CudaVirtualArch::COMPUTE_30:
-    return "compute_30";
-  case CudaVirtualArch::COMPUTE_32:
-    return "compute_32";
-  case CudaVirtualArch::COMPUTE_35:
-    return "compute_35";
-  case CudaVirtualArch::COMPUTE_37:
-    return "compute_37";
-  case CudaVirtualArch::COMPUTE_50:
-    return "compute_50";
-  case CudaVirtualArch::COMPUTE_52:
-    return "compute_52";
-  case CudaVirtualArch::COMPUTE_53:
-    return "compute_53";
-  case CudaVirtualArch::COMPUTE_60:
-    return "compute_60";
-  case CudaVirtualArch::COMPUTE_61:
-    return "compute_61";
-  case CudaVirtualArch::COMPUTE_62:
-    return "compute_62";
-  case CudaVirtualArch::COMPUTE_70:
-    return "compute_70";
-  case CudaVirtualArch::COMPUTE_72:
-    return "compute_72";
-  case CudaVirtualArch::COMPUTE_75:
-    return "compute_75";
-  case CudaVirtualArch::COMPUTE_AMDGCN:
-    return "compute_amdgcn";
-  }
-  llvm_unreachable("invalid enum");
+  return result->virtual_arch_name;
 }
 
-CudaVirtualArch StringToCudaVirtualArch(llvm::StringRef S) {
-  return llvm::StringSwitch<CudaVirtualArch>(S)
-      .Case("compute_20", CudaVirtualArch::COMPUTE_20)
-      .Case("compute_30", CudaVirtualArch::COMPUTE_30)
-      .Case("compute_32", CudaVirtualArch::COMPUTE_32)
-      .Case("compute_35", CudaVirtualArch::COMPUTE_35)
-      .Case("compute_37", CudaVirtualArch::COMPUTE_37)
-      .Case("compute_50", CudaVirtualArch::COMPUTE_50)
-      .Case("compute_52", CudaVirtualArch::COMPUTE_52)
-      .Case("compute_53", CudaVirtualArch::COMPUTE_53)
-      .Case("compute_60", CudaVirtualArch::COMPUTE_60)
-      .Case("compute_61", CudaVirtualArch::COMPUTE_61)
-      .Case("compute_62", CudaVirtualArch::COMPUTE_62)
-      .Case("compute_70", CudaVirtualArch::COMPUTE_70)
-      .Case("compute_72", CudaVirtualArch::COMPUTE_72)
-      .Case("compute_75", CudaVirtualArch::COMPUTE_75)
-      .Case("compute_amdgcn", CudaVirtualArch::COMPUTE_AMDGCN)
-      .Default(CudaVirtualArch::UNKNOWN);
-}
-
-CudaVirtualArch VirtualArchForCudaArch(CudaArch A) {
-  switch (A) {
-  case CudaArch::LAST:
-    break;
-  case CudaArch::UNKNOWN:
-    return CudaVirtualArch::UNKNOWN;
-  case CudaArch::SM_20:
-  case CudaArch::SM_21:
-    return CudaVirtualArch::COMPUTE_20;
-  case CudaArch::SM_30:
-    return CudaVirtualArch::COMPUTE_30;
-  case CudaArch::SM_32:
-    return CudaVirtualArch::COMPUTE_32;
-  case CudaArch::SM_35:
-    return CudaVirtualArch::COMPUTE_35;
-  case CudaArch::SM_37:
-    return CudaVirtualArch::COMPUTE_37;
-  case CudaArch::SM_50:
-    return CudaVirtualArch::COMPUTE_50;
-  case CudaArch::SM_52:
-    return CudaVirtualArch::COMPUTE_52;
-  case CudaArch::SM_53:
-    return CudaVirtualArch::COMPUTE_53;
-  case CudaArch::SM_60:
-    return CudaVirtualArch::COMPUTE_60;
-  case CudaArch::SM_61:
-    return CudaVirtualArch::COMPUTE_61;
-  case CudaArch::SM_62:
-    return CudaVirtualArch::COMPUTE_62;
-  case CudaArch::SM_70:
-    return CudaVirtualArch::COMPUTE_70;
-  case CudaArch::SM_72:
-    return CudaVirtualArch::COMPUTE_72;
-  case CudaArch::SM_75:
-    return CudaVirtualArch::COMPUTE_75;
-  case CudaArch::GFX600:
-  case CudaArch::GFX601:
-  case CudaArch::GFX700:
-  case CudaArch::GFX701:
-  case CudaArch::GFX702:
-  case CudaArch::GFX703:
-  case CudaArch::GFX704:
-  case CudaArch::GFX801:
-  case CudaArch::GFX802:
-  case CudaArch::GFX803:
-  case CudaArch::GFX810:
-  case CudaArch::GFX900:
-  case CudaArch::GFX902:
-  case CudaArch::GFX904:
-  case CudaArch::GFX906:
-  case CudaArch::GFX908:
-  case CudaArch::GFX909:
-  case CudaArch::GFX1010:
-  case CudaArch::GFX1011:
-  case CudaArch::GFX1012:
-    return CudaVirtualArch::COMPUTE_AMDGCN;
-  }
-  llvm_unreachable("invalid enum");
+CudaArch StringToCudaArch(llvm::StringRef S) {
+  auto result = std::find_if(
+      std::begin(arch_names), std::end(arch_names),
+      [S](const CudaArchToStringMap &map) { return S == map.arch_name; });
+  if (result == std::end(arch_names))
+    return CudaArch::UNKNOWN;
+  return result->arch;
 }
 
 CudaVersion MinVersionForCudaArch(CudaArch A) {
-  switch (A) {
-  case CudaArch::LAST:
-    break;
-  case CudaArch::UNKNOWN:
+  if (A == CudaArch::UNKNOWN)
     return CudaVersion::UNKNOWN;
+
+  // AMD GPUs do not depend on CUDA versions.
+  if (IsAMDGpuArch(A))
+    return CudaVersion::CUDA_70;
+
+  switch (A) {
   case CudaArch::SM_20:
   case CudaArch::SM_21:
   case CudaArch::SM_30:
   case CudaArch::SM_32:
   case CudaArch::SM_35:
   case CudaArch::SM_37:
   case CudaArch::SM_50:
   case CudaArch::SM_52:
   case CudaArch::SM_53:
     return CudaVersion::CUDA_70;
   case CudaArch::SM_60:
   case CudaArch::SM_61:
   case CudaArch::SM_62:
     return CudaVersion::CUDA_80;
   case CudaArch::SM_70:
     return CudaVersion::CUDA_90;
   case CudaArch::SM_72:
     return CudaVersion::CUDA_91;
   case CudaArch::SM_75:
     return CudaVersion::CUDA_100;
-  case CudaArch::GFX600:
-  case CudaArch::GFX601:
-  case CudaArch::GFX700:
-  case CudaArch::GFX701:
-  case CudaArch::GFX702:
-  case CudaArch::GFX703:
-  case CudaArch::GFX704:
-  case CudaArch::GFX801:
-  case CudaArch::GFX802:
-  case CudaArch::GFX803:
-  case CudaArch::GFX810:
-  case CudaArch::GFX900:
-  case CudaArch::GFX902:
-  case CudaArch::GFX904:
-  case CudaArch::GFX906:
-  case CudaArch::GFX908:
-  case CudaArch::GFX909:
-  case CudaArch::GFX1010:
-  case CudaArch::GFX1011:
-  case CudaArch::GFX1012:
-    return CudaVersion::CUDA_70;
-  }
+  default:
     llvm_unreachable("invalid enum");
   }
+}
 
 CudaVersion MaxVersionForCudaArch(CudaArch A) {
+  // AMD GPUs do not depend on CUDA versions.
+  if (IsAMDGpuArch(A))
+    return CudaVersion::LATEST;
+
   switch (A) {
   case CudaArch::UNKNOWN:
     return CudaVersion::UNKNOWN;
   case CudaArch::SM_20:
   case CudaArch::SM_21:
-  case CudaArch::GFX600:
-  case CudaArch::GFX601:
-  case CudaArch::GFX700:
-  case CudaArch::GFX701:
-  case CudaArch::GFX702:
-  case CudaArch::GFX703:
-  case CudaArch::GFX704:
-  case CudaArch::GFX801:
-  case CudaArch::GFX802:
-  case CudaArch::GFX803:
-  case CudaArch::GFX810:
-  case CudaArch::GFX900:
-  case CudaArch::GFX902:
-  case CudaArch::GFX1010:
-  case CudaArch::GFX1011:
-  case CudaArch::GFX1012:
     return CudaVersion::CUDA_80;
   default:
     return CudaVersion::LATEST;
   }
 }
 
 CudaVersion ToCudaVersion(llvm::VersionTuple Version) {
   int IVer =

clang/lib/Driver/ToolChains/Cuda.cpp

for (const auto &Candidate : Candidates) {
     } else {
       ParseCudaVersionFile((*VersionFile)->getBuffer());
     }
 
     if (Version >= CudaVersion::CUDA_90) {
       // CUDA-9+ uses single libdevice file for all GPU variants.
       std::string FilePath = LibDevicePath + "/libdevice.10.bc";
       if (FS.exists(FilePath)) {
-        for (const char *GpuArchName :
-             {"sm_30", "sm_32", "sm_35", "sm_37", "sm_50", "sm_52", "sm_53",
-              "sm_60", "sm_61", "sm_62", "sm_70", "sm_72", "sm_75"}) {
-          const CudaArch GpuArch = StringToCudaArch(GpuArchName);
-          if (Version >= MinVersionForCudaArch(GpuArch) &&
-              Version <= MaxVersionForCudaArch(GpuArch))
-            LibDeviceMap[GpuArchName] = FilePath;
+        for (int Arch = (int)CudaArch::SM_30, E = (int)CudaArch::LAST; Arch < E;
+             ++Arch) {
+          CudaArch GpuArch = static_cast<CudaArch>(Arch);
+          if (!IsNVIDIAGpuArch(GpuArch))
+            continue;
+          std::string GpuArchName(CudaArchToString(GpuArch));
         }
       }
     } else {
       std::error_code EC;
       for (llvm::sys::fs::directory_iterator LI(LibDevicePath, EC), LE;
            !EC && LI != LE; LI = LI.increment(EC)) {
         StringRef FilePath = LI->path();
         StringRef FileName = llvm::sys::path::filename(FilePath);
assert(gpu_arch_str &&
            "Device action expected to have associated a GPU architecture!");
     CudaArch gpu_arch = StringToCudaArch(gpu_arch_str);
 
     if (II.getType() == types::TY_PP_Asm &&
         !shouldIncludePTX(Args, gpu_arch_str))
       continue;
     // We need to pass an Arch of the form "sm_XX" for cubin files and
     // "compute_XX" for ptx.
-    const char *Arch =
-        (II.getType() == types::TY_PP_Asm)
-            ? CudaVirtualArchToString(VirtualArchForCudaArch(gpu_arch))
+    const char *Arch = (II.getType() == types::TY_PP_Asm)
+                           ? CudaArchToVirtualArchString(gpu_arch)
                            : gpu_arch_str;
     CmdArgs.push_back(Args.MakeArgString(llvm::Twine("--image=profile=") +
                                          Arch + ",file=" + II.getFilename()));
   }
 
   for (const auto& A : Args.getAllArgValues(options::OPT_Xcuda_fatbinary))
     CmdArgs.push_back(Args.MakeArgString(A));
 
   const char *Exec = Args.MakeArgString(TC.GetProgramPath("fatbinary"));