[X86] Add basic support for -mtune command line option in clang

Building on the backend support from D85165. This parses the command line option in the driver, passes it on to CC1 and adds a function attribute.

-Still need to support tune on the target attribute.
-Need to use "generic" as the tuning by default. But need to change generic in the backend first.
-Need to set tune if march is specified and mtune isn't.
-May need to disable getHostCPUName's ability to guess CPU name from features when it doesn't have a family/model match for mtune=native. That's what gcc appears to do.

Differential Revision: https://reviews.llvm.org/D85384

Author: topperc

clang/include/clang/Basic/TargetOptions.h

@@ -35,6 +35,9 @@ class TargetOptions {
   /// If given, the name of the target CPU to generate code for.
   std::string CPU;
 
+  /// If given, the name of the target CPU to tune code for.
+  std::string TuneCPU;
+
   /// If given, the unit to use for floating point math.
   std::string FPMath;
 

clang/include/clang/Driver/Options.td

@@ -2715,7 +2715,7 @@ def module_file_info : Flag<["-"], "module-file-info">, Flags<[DriverOption,CC1O
   HelpText<"Provide information about a particular module file">;
 def mthumb : Flag<["-"], "mthumb">, Group<m_Group>;
 def mtune_EQ : Joined<["-"], "mtune=">, Group<m_Group>,
-  HelpText<"Accepted for compatibility with GCC. Currently has no effect.">;
+  HelpText<"Only supported on X86. Otherwise accepted for compatibility with GCC.">;
 def multi__module : Flag<["-"], "multi_module">;
 def multiply__defined__unused : Separate<["-"], "multiply_defined_unused">;
 def multiply__defined : Separate<["-"], "multiply_defined">;
@@ -3490,6 +3490,8 @@ let Flags = [CC1Option, CC1AsOption, NoDriverOption] in {
 
 def target_cpu : Separate<["-"], "target-cpu">,
   HelpText<"Target a specific cpu type">;
+def tune_cpu : Separate<["-"], "tune-cpu">,
+  HelpText<"Tune for a specific cpu type">;
 def target_feature : Separate<["-"], "target-feature">,
   HelpText<"Target specific attributes">;
 def triple : Separate<["-"], "triple">,

clang/lib/Basic/Targets.cpp

@@ -652,6 +652,16 @@ TargetInfo::CreateTargetInfo(DiagnosticsEngine &Diags,
     return nullptr;
   }
 
+  // Check the TuneCPU name if specified.
+  if (!Opts->TuneCPU.empty() && !Target->isValidCPUName(Opts->TuneCPU)) {
+    Diags.Report(diag::err_target_unknown_cpu) << Opts->TuneCPU;
+    SmallVector<StringRef, 32> ValidList;
+    Target->fillValidCPUList(ValidList);
+    if (!ValidList.empty())
+      Diags.Report(diag::note_valid_options) << llvm::join(ValidList, ", ");
+    return nullptr;
+  }
+
   // Set the target ABI if specified.
   if (!Opts->ABI.empty() && !Target->setABI(Opts->ABI)) {
     Diags.Report(diag::err_target_unknown_abi) << Opts->ABI;

clang/lib/CodeGen/CodeGenModule.cpp

@@ -1749,6 +1749,7 @@ bool CodeGenModule::GetCPUAndFeaturesAttributes(GlobalDecl GD,
   // we have a decl for the function and it has a target attribute then
   // parse that and add it to the feature set.
   StringRef TargetCPU = getTarget().getTargetOpts().CPU;
+  StringRef TuneCPU = getTarget().getTargetOpts().TuneCPU;
   std::vector<std::string> Features;
   const auto *FD = dyn_cast_or_null<FunctionDecl>(GD.getDecl());
   FD = FD ? FD->getMostRecentDecl() : FD;
@@ -1783,6 +1784,10 @@ bool CodeGenModule::GetCPUAndFeaturesAttributes(GlobalDecl GD,
     Attrs.addAttribute("target-cpu", TargetCPU);
     AddedAttr = true;
   }
+  if (TuneCPU != "") {
+    Attrs.addAttribute("tune-cpu", TuneCPU);
+    AddedAttr = true;
+  }
   if (!Features.empty()) {
     llvm::sort(Features);
     Attrs.addAttribute("target-features", llvm::join(Features, ","));

clang/lib/Driver/ToolChains/Clang.cpp

@@ -39,6 +39,7 @@
 #include "llvm/Support/Compiler.h"
 #include "llvm/Support/Compression.h"
 #include "llvm/Support/FileSystem.h"
+#include "llvm/Support/Host.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/Process.h"
 #include "llvm/Support/TargetParser.h"
@@ -2071,6 +2072,18 @@ void Clang::AddX86TargetArgs(const ArgList &Args,
     CmdArgs.push_back("soft");
     CmdArgs.push_back("-mstack-alignment=4");
   }
+
+  // Handle -mtune.
+  // FIXME: We should default to "generic" unless -march is set to match gcc.
+  if (const Arg *A = Args.getLastArg(clang::driver::options::OPT_mtune_EQ)) {
+    StringRef Name = A->getValue();
+
+    if (Name == "native")
+      Name = llvm::sys::getHostCPUName();
+
+    CmdArgs.push_back("-tune-cpu");
+    CmdArgs.push_back(Args.MakeArgString(Name));
+  }
 }
 
 void Clang::AddHexagonTargetArgs(const ArgList &Args,

clang/lib/Frontend/CompilerInvocation.cpp

@@ -3658,6 +3658,7 @@ static void ParseTargetArgs(TargetOptions &Opts, ArgList &Args,
       Opts.EABIVersion = EABIVersion;
   }
   Opts.CPU = std::string(Args.getLastArgValue(OPT_target_cpu));
+  Opts.TuneCPU = std::string(Args.getLastArgValue(OPT_tune_cpu));
   Opts.FPMath = std::string(Args.getLastArgValue(OPT_mfpmath));
   Opts.FeaturesAsWritten = Args.getAllArgValues(OPT_target_feature);
   Opts.LinkerVersion =

clang/test/Misc/target-invalid-cpu-note.c

@@ -8,6 +8,11 @@
 // AARCH64: note: valid target CPU values are:
 // AARCH64-SAME: cortex-a35,
 
+// RUN: not %clang_cc1 -triple arm64--- -tune-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix TUNE_AARCH64
+// TUNE_AARCH64: error: unknown target CPU 'not-a-cpu'
+// TUNE_AARCH64: note: valid target CPU values are:
+// TUNE_AARCH64-SAME: cortex-a35,
+
 // RUN: not %clang_cc1 -triple i386--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix X86
 // X86: error: unknown target CPU 'not-a-cpu'
 // X86: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3,
@@ -32,6 +37,30 @@
 // X86_64-SAME: athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1,
 // X86_64-SAME: btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, x86-64
 
+// RUN: not %clang_cc1 -triple i386--- -tune-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix TUNE_X86
+// TUNE_X86: error: unknown target CPU 'not-a-cpu'
+// TUNE_X86: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3,
+// TUNE_X86-SAME: i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3,
+// TUNE_X86-SAME: pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott,
+// TUNE_X86-SAME: nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont,
+// TUNE_X86-SAME: nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge,
+// TUNE_X86-SAME: core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512,
+// TUNE_X86-SAME: skx, cascadelake, cooperlake, cannonlake, icelake-client, icelake-server, tigerlake, knl, knm, lakemont, k6, k6-2, k6-3,
+// TUNE_X86-SAME: athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64,
+// TUNE_X86-SAME: athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10,
+// TUNE_X86-SAME: barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2,
+// TUNE_X86-SAME: x86-64, geode
+
+// RUN: not %clang_cc1 -triple x86_64--- -tune-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix TUNE_X86_64
+// TUNE_X86_64: error: unknown target CPU 'not-a-cpu'
+// TUNE_X86_64: note: valid target CPU values are: nocona, core2, penryn, bonnell,
+// TUNE_X86_64-SAME: atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere,
+// TUNE_X86_64-SAME: sandybridge, corei7-avx, ivybridge, core-avx-i, haswell,
+// TUNE_X86_64-SAME: core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake,
+// TUNE_X86_64-SAME: icelake-client, icelake-server, tigerlake, knl, knm, k8, athlon64, athlon-fx, opteron, k8-sse3,
+// TUNE_X86_64-SAME: athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1,
+// TUNE_X86_64-SAME: btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, x86-64
+
 // RUN: not %clang_cc1 -triple nvptx--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix NVPTX
 // NVPTX: error: unknown target CPU 'not-a-cpu'
 // NVPTX: note: valid target CPU values are: sm_20, sm_21, sm_30, sm_32, sm_35,