[X86] Support -march=emeraldrapids

Reviewed By: pengfei, skan Differential Revision: https://reviews.llvm.org/D140950
2026-01-24 17:01:00 +08:00 · 2023-01-05 20:27:13 +08:00
parent 60442f0d44
commit 27b8f54f51
15 changed files with 37 additions and 8 deletions
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -798,7 +798,7 @@ X86 Support in Clang
  * Support intrinsic of ``_mm(256)_cvtneobf16_ps``.
  * Support intrinsic of ``_mm(256)_cvtneoph_ps``.
  * Support intrinsic of ``_mm(256)_cvtneps_avx_pbh``.
- ``-march=raptorlake`` and ``-march=meteorlake`` are now supported.
+- ``-march=raptorlake``, ``-march=meteorlake`` and ``-march=emeraldrapids`` are now supported.
 - ``-march=sierraforest``, ``-march=graniterapids`` and ``-march=grandridge`` are now supported.
 - Lift _BitInt() supported max width from 128 to 8388608.
 - Support intrinsics of ``_mm(256)_reduce_(add|mul|or|and)_epi8/16``.
--- a/clang/lib/Basic/Targets/X86.cpp
+++ b/clang/lib/Basic/Targets/X86.cpp
@@ -529,6 +529,7 @@ void X86TargetInfo::getTargetDefines(const LangOptions &Opts,
  case CK_Sierraforest:
  case CK_Grandridge:
  case CK_Graniterapids:
+  case CK_Emeraldrapids:
    // FIXME: Historically, we defined this legacy name, it would be nice to
    // remove it at some point. We've never exposed fine-grained names for
    // recent primary x86 CPUs, and we should keep it that way.
@@ -1422,6 +1423,7 @@ Optional<unsigned> X86TargetInfo::getCPUCacheLineSize() const {
    case CK_Sierraforest:
    case CK_Grandridge:
    case CK_Graniterapids:
+    case CK_Emeraldrapids:
    case CK_KNL:
    case CK_KNM:
    // K7
--- a/clang/test/CodeGen/attr-target-mv.c
+++ b/clang/test/CodeGen/attr-target-mv.c
@@ -20,6 +20,7 @@ int __attribute__((target("arch=meteorlake"))) foo(void) {return 15;}
 int __attribute__((target("arch=sierraforest"))) foo(void) {return 16;}
 int __attribute__((target("arch=grandridge"))) foo(void) {return 17;}
 int __attribute__((target("arch=graniterapids"))) foo(void) {return 18;}
+int __attribute__((target("arch=emeraldrapids"))) foo(void) {return 19;}
 int __attribute__((target("default"))) foo(void) { return 2; }

 int bar(void) {
@@ -164,6 +165,8 @@ void calls_pr50025c(void) { pr50025c(); }
 // LINUX: ret i32 17
 // LINUX: define{{.*}} i32 @foo.arch_graniterapids()
 // LINUX: ret i32 18
+// LINUX: define{{.*}} i32 @foo.arch_emeraldrapids()
+// LINUX: ret i32 19
 // LINUX: define{{.*}} i32 @foo()
 // LINUX: ret i32 2
 // LINUX: define{{.*}} i32 @bar()
@@ -205,6 +208,8 @@ void calls_pr50025c(void) { pr50025c(); }
 // WINDOWS: ret i32 17
 // WINDOWS: define{{.*}} i32 @foo.arch_graniterapids()
 // WINDOWS: ret i32 18
+// WINDOWS: define dso_local i32 @foo.arch_emeraldrapids()
+// WINDOWS: ret i32 19
 // WINDOWS: define dso_local i32 @foo()
 // WINDOWS: ret i32 2
 // WINDOWS: define dso_local i32 @bar()
--- a/clang/test/CodeGen/target-builtin-noerror.c
+++ b/clang/test/CodeGen/target-builtin-noerror.c
@@ -109,6 +109,7 @@ void verifycpustrings(void) {
  (void)__builtin_cpu_is("goldmont-plus");
  (void)__builtin_cpu_is("grandridge");
  (void)__builtin_cpu_is("graniterapids");
+  (void)__builtin_cpu_is("emeraldrapids");
  (void)__builtin_cpu_is("haswell");
  (void)__builtin_cpu_is("icelake-client");
  (void)__builtin_cpu_is("icelake-server");
--- a/clang/test/Driver/x86-march.c
+++ b/clang/test/Driver/x86-march.c
@@ -131,6 +131,10 @@
 // RUN:   | FileCheck %s -check-prefix=graniterapids
 // graniterapids: "-target-cpu" "graniterapids"
 //
+// RUN: %clang -target x86_64-unknown-unknown -c -### %s -march=emeraldrapids 2>&1 \
+// RUN:   | FileCheck %s -check-prefix=emeraldrapids
+// emeraldrapids: "-target-cpu" "emeraldrapids"
+//
 // RUN: %clang -target x86_64-unknown-unknown -c -### %s -march=sapphirerapids 2>&1 \
 // RUN:   | FileCheck %s -check-prefix=sapphirerapids
 // sapphirerapids: "-target-cpu" "sapphirerapids"
--- a/clang/test/Misc/target-invalid-cpu-note.c
+++ b/clang/test/Misc/target-invalid-cpu-note.c
@@ -13,19 +13,19 @@

 // 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-NEXT: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3, i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3, pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott, nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, knl, knm, lakemont, k6, k6-2, k6-3, athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, x86-64-v2, x86-64-v3, x86-64-v4, geode{{$}}
+// X86-NEXT: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3, i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3, pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott, nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, emeraldrapids, knl, knm, lakemont, k6, k6-2, k6-3, athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, x86-64-v2, x86-64-v3, x86-64-v4, geode{{$}}

 // RUN: not %clang_cc1 -triple x86_64--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix X86_64
 // X86_64: error: unknown target CPU 'not-a-cpu'
-// X86_64-NEXT: note: valid target CPU values are: nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, knl, knm, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, x86-64-v2, x86-64-v3, x86-64-v4{{$}}
+// X86_64-NEXT: note: valid target CPU values are: nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, emeraldrapids, knl, knm, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, x86-64-v2, x86-64-v3, x86-64-v4{{$}}

 // 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-NEXT: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3, i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3, pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott, nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, knl, knm, lakemont, k6, k6-2, k6-3, athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, geode{{$}}
+// TUNE_X86-NEXT: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3, i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3, pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott, nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, emeraldrapids, knl, knm, lakemont, k6, k6-2, k6-3, athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, geode{{$}}

 // 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-NEXT: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3, i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3, pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott, nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, knl, knm, lakemont, k6, k6-2, k6-3, athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, geode{{$}}
+// TUNE_X86_64-NEXT: note: valid target CPU values are: i386, i486, winchip-c6, winchip2, c3, i586, pentium, pentium-mmx, pentiumpro, i686, pentium2, pentium3, pentium3m, pentium-m, c3-2, yonah, pentium4, pentium4m, prescott, nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge, core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake, icelake-client, rocketlake, icelake-server, tigerlake, sapphirerapids, alderlake, raptorlake, meteorlake, sierraforest, grandridge, graniterapids, emeraldrapids, knl, knm, lakemont, k6, k6-2, k6-3, athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64, athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, znver3, znver4, x86-64, geode{{$}}

 // 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'
--- a/clang/test/Preprocessor/predefined-arch-macros.c
+++ b/clang/test/Preprocessor/predefined-arch-macros.c
@@ -1646,6 +1646,9 @@
 // RUN: %clang -march=sapphirerapids -m32 -E -dM %s -o - 2>&1 \
 // RUN:     -target i386-unknown-linux \
 // RUN:   | FileCheck -match-full-lines %s -check-prefix=CHECK_SPR_M32
+// RUN: %clang -march=emeraldrapids -m32 -E -dM %s -o - 2>&1 \
+// RUN:     -target i386-unknown-linux \
+// RUN:   | FileCheck -match-full-lines %s -check-prefix=CHECK_SPR_M32
 // CHECK_SPR_M32: #define __AES__ 1
 // CHECK_SPR_M32: #define __AMXBF16__ 1
 // CHECK_SPR_M32: #define __AMXINT8__ 1
@@ -1718,6 +1721,9 @@
 // RUN: %clang -march=sapphirerapids -m64 -E -dM %s -o - 2>&1 \
 // RUN:     -target i386-unknown-linux \
 // RUN:   | FileCheck -match-full-lines %s -check-prefix=CHECK_SPR_M64
+// RUN: %clang -march=emeraldrapids -m64 -E -dM %s -o - 2>&1 \
+// RUN:     -target i386-unknown-linux \
+// RUN:   | FileCheck -match-full-lines %s -check-prefix=CHECK_SPR_M64
 // CHECK_SPR_M64: #define __AES__ 1
 // CHECK_SPR_M64: #define __AMXBF16__ 1
 // CHECK_SPR_M64: #define __AMXINT8__ 1
--- a/compiler-rt/lib/builtins/cpu_model.c
+++ b/compiler-rt/lib/builtins/cpu_model.c
@@ -464,6 +464,8 @@ getIntelProcessorTypeAndSubtype(unsigned Family, unsigned Model,
      *Subtype = INTEL_COREI7_ICELAKE_SERVER;
      break;

+    // Emerald Rapids:
+    case 0xcf:
    // Sapphire Rapids:
    case 0x8f:
      CPU = "sapphirerapids";
@@ -471,7 +473,7 @@ getIntelProcessorTypeAndSubtype(unsigned Family, unsigned Model,
      *Subtype = INTEL_COREI7_SAPPHIRERAPIDS;
      break;

-    // Graniterapids:
+    // Granite Rapids:
    case 0xae:
    case 0xad:
      CPU = "graniterapids";
--- a/llvm/docs/ReleaseNotes.rst
+++ b/llvm/docs/ReleaseNotes.rst
@@ -186,7 +186,7 @@ Changes to the X86 Backend
 * Support ISA of ``AVX-IFMA``.
 * Support ISA of ``AVX-VNNI-INT8``.
 * Support ISA of ``AVX-NE-CONVERT``.
-* ``-mcpu=raptorlake`` and ``-mcpu=meteorlake`` are now supported.
+* ``-mcpu=raptorlake``, ``-mcpu=meteorlake`` and ``-mcpu=emeraldrapids`` are now supported.
 * ``-mcpu=sierraforest``, ``-mcpu=graniterapids`` and ``-mcpu=grandridge`` are now supported.

 Changes to the OCaml bindings
--- a/llvm/include/llvm/TargetParser/X86TargetParser.def
+++ b/llvm/include/llvm/TargetParser/X86TargetParser.def
@@ -103,6 +103,7 @@ X86_CPU_SUBTYPE(INTEL_COREI7_GRANITERAPIDS,  "graniterapids")
 // Alternate names supported by __builtin_cpu_is and target multiversioning.
 X86_CPU_SUBTYPE_ALIAS(INTEL_COREI7_ALDERLAKE, "raptorlake")
 X86_CPU_SUBTYPE_ALIAS(INTEL_COREI7_ALDERLAKE, "meteorlake")
+X86_CPU_SUBTYPE_ALIAS(INTEL_COREI7_SAPPHIRERAPIDS, "emeraldrapids")

 #undef X86_CPU_SUBTYPE_ALIAS
 #undef X86_CPU_SUBTYPE
--- a/llvm/include/llvm/TargetParser/X86TargetParser.h
+++ b/llvm/include/llvm/TargetParser/X86TargetParser.h
@@ -109,6 +109,7 @@ enum CPUKind {
  CK_Sierraforest,
  CK_Grandridge,
  CK_Graniterapids,
+  CK_Emeraldrapids,
  CK_KNL,
  CK_KNM,
  CK_Lakemont,
--- a/llvm/lib/Target/X86/X86.td
+++ b/llvm/lib/Target/X86/X86.td
@@ -1548,6 +1548,8 @@ def : ProcModel<"meteorlake", AlderlakePModel,
                ProcessorFeatures.ADLFeatures, ProcessorFeatures.ADLTuning>;
 def : ProcModel<"graniterapids", SkylakeServerModel,
                ProcessorFeatures.GNRFeatures, ProcessorFeatures.SPRTuning>;
+def : ProcModel<"emeraldrapids", SkylakeServerModel,
+                ProcessorFeatures.SPRFeatures, ProcessorFeatures.SPRTuning>;

 // AMD CPUs.

--- a/llvm/lib/TargetParser/Host.cpp
+++ b/llvm/lib/TargetParser/Host.cpp
@@ -847,6 +847,8 @@ getIntelProcessorTypeAndSubtype(unsigned Family, unsigned Model,
      *Subtype = X86::INTEL_COREI7_ICELAKE_SERVER;
      break;

+    // Emerald Rapids:
+    case 0xcf:
    // Sapphire Rapids:
    case 0x8f:
      CPU = "sapphirerapids";
--- a/llvm/lib/TargetParser/X86TargetParser.cpp
+++ b/llvm/lib/TargetParser/X86TargetParser.cpp
@@ -391,8 +391,10 @@ constexpr ProcInfo Processors[] = {
  { {"sierraforest"}, CK_Sierraforest, FEATURE_AVX2, FeaturesSierraforest },
  // Grandridge microarchitecture based processors.
  { {"grandridge"}, CK_Grandridge, FEATURE_AVX2, FeaturesGrandridge },
-  // Graniterapids microarchitecture based processors.
+  // Granite Rapids microarchitecture based processors.
  { {"graniterapids"}, CK_Graniterapids, FEATURE_AVX512BF16, FeaturesGraniteRapids },
+  // Emerald Rapids microarchitecture based processors.
+  { {"emeraldrapids"}, CK_Emeraldrapids, FEATURE_AVX512BF16, FeaturesSapphireRapids },
  // Knights Landing processor.
  { {"knl"}, CK_KNL, FEATURE_AVX512F, FeaturesKNL },
  // Knights Mill processor.
--- a/llvm/test/CodeGen/X86/cpus-intel.ll
+++ b/llvm/test/CodeGen/X86/cpus-intel.ll
@@ -22,6 +22,7 @@
 ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=sierraforest 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
 ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=grandridge 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
 ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=graniterapids 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
+; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=emeraldrapids 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty

 ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=nocona 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
 ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=core2 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty