[X86][AMX] Support AMX-AVX512 (#114070)

Author: phoebewang

clang/docs/ReleaseNotes.rst

@@ -723,6 +723,7 @@ X86 Support
   * Supported intrinsics of ``_mm(256|512)_(mask(z))_loadrs_epi(8|16|32|64)``.
 - Support ISA of ``AMX-FP8``.
 - Support ISA of ``AMX-TRANSPOSE``.
+- Support ISA of ``AMX-AVX512``.
 
 Arm and AArch64 Support
 ^^^^^^^^^^^^^^^^^^^^^^^

clang/include/clang/Basic/BuiltinsX86_64.def

@@ -133,6 +133,12 @@ TARGET_BUILTIN(__builtin_ia32_t2rpntlvwz0t1_internal, "vUsUsUsV256i*V256i*vC*z",
 TARGET_BUILTIN(__builtin_ia32_t2rpntlvwz1_internal, "vUsUsUsV256i*V256i*vC*z", "n", "amx-transpose")
 TARGET_BUILTIN(__builtin_ia32_t2rpntlvwz1t1_internal, "vUsUsUsV256i*V256i*vC*z", "n", "amx-transpose")
 TARGET_BUILTIN(__builtin_ia32_ttransposed_internal, "V256iUsUsV256i", "n", "amx-transpose")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowd2ps_internal, "V16fUsUsV256iUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2pbf16h_internal, "V32yUsUsV256iUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2pbf16l_internal, "V32yUsUsV256iUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2phh_internal, "V32xUsUsV256iUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2phl_internal, "V32xUsUsV256iUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tilemovrow_internal, "V16iUsUsV256iUi", "n", "amx-avx512,avx10.2-512")
 // AMX
 TARGET_BUILTIN(__builtin_ia32_tile_loadconfig, "vvC*", "n", "amx-tile")
 TARGET_BUILTIN(__builtin_ia32_tile_storeconfig, "vvC*", "n", "amx-tile")
@@ -159,6 +165,13 @@ TARGET_BUILTIN(__builtin_ia32_t2rpntlvwz1, "vIUcvC*z", "n", "amx-transpose")
 TARGET_BUILTIN(__builtin_ia32_t2rpntlvwz1t1, "vIUcvC*z", "n","amx-transpose")
 TARGET_BUILTIN(__builtin_ia32_ttransposed, "vIUcIUc", "n", "amx-transpose")
 
+TARGET_BUILTIN(__builtin_ia32_tcvtrowd2ps, "V16fIUcUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2pbf16h, "V32yIUcUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2pbf16l, "V32yIUcUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2phh, "V32xIUcUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tcvtrowps2phl, "V32xIUcUi", "n", "amx-avx512,avx10.2-512")
+TARGET_BUILTIN(__builtin_ia32_tilemovrow, "V16iIUcUi", "n", "amx-avx512,avx10.2-512")
+
 TARGET_BUILTIN(__builtin_ia32_prefetchi, "vvC*Ui", "nc", "prefetchi")
 TARGET_BUILTIN(__builtin_ia32_cmpccxadd32, "Siv*SiSiIi", "n", "cmpccxadd")
 TARGET_BUILTIN(__builtin_ia32_cmpccxadd64, "SLLiSLLi*SLLiSLLiIi", "n", "cmpccxadd")

clang/include/clang/Driver/Options.td

@@ -6285,6 +6285,8 @@ def mno_80387 : Flag<["-"], "mno-80387">, Alias<mno_x87>;
 def mno_fp_ret_in_387 : Flag<["-"], "mno-fp-ret-in-387">, Alias<mno_x87>;
 def mmmx : Flag<["-"], "mmmx">, Group<m_x86_Features_Group>;
 def mno_mmx : Flag<["-"], "mno-mmx">, Group<m_x86_Features_Group>;
+def mamx_avx512 : Flag<["-"], "mamx-avx512">, Group<m_x86_Features_Group>;
+def mno_amx_avx512 : Flag<["-"], "mno-amx-avx512">, Group<m_x86_Features_Group>;
 def mamx_bf16 : Flag<["-"], "mamx-bf16">, Group<m_x86_Features_Group>;
 def mno_amx_bf16 : Flag<["-"], "mno-amx-bf16">, Group<m_x86_Features_Group>;
 def mamx_complex : Flag<["-"], "mamx-complex">, Group<m_x86_Features_Group>;

clang/lib/Basic/Targets/X86.cpp

@@ -432,6 +432,8 @@ bool X86TargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
       HasAMXFP8 = true;
     } else if (Feature == "+amx-transpose") {
       HasAMXTRANSPOSE = true;
+    } else if (Feature == "+amx-avx512") {
+      HasAMXAVX512 = true;
     } else if (Feature == "+cmpccxadd") {
       HasCMPCCXADD = true;
     } else if (Feature == "+raoint") {
@@ -955,6 +957,8 @@ void X86TargetInfo::getTargetDefines(const LangOptions &Opts,
     Builder.defineMacro("__AMX_FP8__");
   if (HasAMXTRANSPOSE)
     Builder.defineMacro("__AMX_TRANSPOSE__");
+  if (HasAMXAVX512)
+    Builder.defineMacro("__AMX_AVX512__");
   if (HasCMPCCXADD)
     Builder.defineMacro("__CMPCCXADD__");
   if (HasRAOINT)
@@ -1080,6 +1084,7 @@ bool X86TargetInfo::isValidFeatureName(StringRef Name) const {
   return llvm::StringSwitch<bool>(Name)
       .Case("adx", true)
       .Case("aes", true)
+      .Case("amx-avx512", true)
       .Case("amx-bf16", true)
       .Case("amx-complex", true)
       .Case("amx-fp16", true)
@@ -1200,6 +1205,7 @@ bool X86TargetInfo::hasFeature(StringRef Feature) const {
   return llvm::StringSwitch<bool>(Feature)
       .Case("adx", HasADX)
       .Case("aes", HasAES)
+      .Case("amx-avx512", HasAMXAVX512)
       .Case("amx-bf16", HasAMXBF16)
       .Case("amx-complex", HasAMXCOMPLEX)
       .Case("amx-fp16", HasAMXFP16)

clang/lib/Basic/Targets/X86.h

@@ -159,6 +159,7 @@ class LLVM_LIBRARY_VISIBILITY X86TargetInfo : public TargetInfo {
   bool HasAMXCOMPLEX = false;
   bool HasAMXFP8 = false;
   bool HasAMXTRANSPOSE = false;
+  bool HasAMXAVX512 = false;
   bool HasSERIALIZE = false;
   bool HasTSXLDTRK = false;
   bool HasUSERMSR = false;

clang/lib/Headers/CMakeLists.txt

@@ -146,6 +146,7 @@ set(x86_files
   adcintrin.h
   adxintrin.h
   ammintrin.h
+  amxavx512intrin.h
   amxcomplexintrin.h
   amxfp16intrin.h
   amxfp8intrin.h