[llvm] e02d081 - [X86] Support -march=sapphirerapids
Freddy Ye via llvm-commits
llvm-commits at lists.llvm.org
Mon Aug 24 23:22:01 PDT 2020
Author: Freddy Ye
Date: 2020-08-25T14:21:21+08:00
New Revision: e02d081f2b60b61eb60ef6a49b1a9f907e432d4c
URL: https://github.com/llvm/llvm-project/commit/e02d081f2b60b61eb60ef6a49b1a9f907e432d4c
DIFF: https://github.com/llvm/llvm-project/commit/e02d081f2b60b61eb60ef6a49b1a9f907e432d4c.diff
LOG: [X86] Support -march=sapphirerapids
Support -march=sapphirerapids for x86.
Compare with Icelake Server, it includes 14 more new features. They are
amxtile, amxint8, amxbf16, avx512bf16, avx512vp2intersect, cldemote,
enqcmd, movdir64b, movdiri, ptwrite, serialize, shstk, tsxldtrk, waitpkg.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D86503
Added:
Modified:
clang/lib/Basic/Targets/X86.cpp
clang/test/CodeGen/attr-target-mv.c
clang/test/CodeGen/target-builtin-noerror.c
clang/test/Driver/x86-march.c
clang/test/Misc/target-invalid-cpu-note.c
clang/test/Preprocessor/predefined-arch-macros.c
compiler-rt/lib/builtins/cpu_model.c
llvm/include/llvm/Support/X86TargetParser.def
llvm/include/llvm/Support/X86TargetParser.h
llvm/lib/Support/Host.cpp
llvm/lib/Support/X86TargetParser.cpp
llvm/lib/Target/X86/X86.td
llvm/test/CodeGen/X86/cpus-intel.ll
Removed:
################################################################################
diff --git a/clang/lib/Basic/Targets/X86.cpp b/clang/lib/Basic/Targets/X86.cpp
index 64c7ce9182c9..b829dfac74fb 100644
--- a/clang/lib/Basic/Targets/X86.cpp
+++ b/clang/lib/Basic/Targets/X86.cpp
@@ -459,6 +459,7 @@ void X86TargetInfo::getTargetDefines(const LangOptions &Opts,
case CK_IcelakeClient:
case CK_IcelakeServer:
case CK_Tigerlake:
+ case CK_SapphireRapids:
// 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.
@@ -1269,6 +1270,7 @@ Optional<unsigned> X86TargetInfo::getCPUCacheLineSize() const {
case CK_Cooperlake:
case CK_Cannonlake:
case CK_Tigerlake:
+ case CK_SapphireRapids:
case CK_IcelakeClient:
case CK_IcelakeServer:
case CK_KNL:
diff --git a/clang/test/CodeGen/attr-target-mv.c b/clang/test/CodeGen/attr-target-mv.c
index 89b219c25fec..57b266403dfc 100644
--- a/clang/test/CodeGen/attr-target-mv.c
+++ b/clang/test/CodeGen/attr-target-mv.c
@@ -11,6 +11,7 @@ int __attribute__((target("arch=icelake-client"))) foo(void) {return 6;}
int __attribute__((target("arch=icelake-server"))) foo(void) {return 7;}
int __attribute__((target("arch=cooperlake"))) foo(void) {return 8;}
int __attribute__((target("arch=tigerlake"))) foo(void) {return 9;}
+int __attribute__((target("arch=sapphirerapids"))) foo(void) {return 10;}
int __attribute__((target("default"))) foo(void) { return 2; }
int bar() {
@@ -91,6 +92,8 @@ __attribute__((target("avx,sse4.2"), used)) inline void foo_used2(int i, double
// LINUX: ret i32 8
// LINUX: define i32 @foo.arch_tigerlake()
// LINUX: ret i32 9
+// LINUX: define i32 @foo.arch_sapphirerapids()
+// LINUX: ret i32 10
// LINUX: define i32 @foo()
// LINUX: ret i32 2
// LINUX: define i32 @bar()
diff --git a/clang/test/CodeGen/target-builtin-noerror.c b/clang/test/CodeGen/target-builtin-noerror.c
index 339e5b15c88d..42164303e4a6 100644
--- a/clang/test/CodeGen/target-builtin-noerror.c
+++ b/clang/test/CodeGen/target-builtin-noerror.c
@@ -122,6 +122,7 @@ void verifycpustrings() {
(void)__builtin_cpu_is("skylake-avx512");
(void)__builtin_cpu_is("slm");
(void)__builtin_cpu_is("tigerlake");
+ (void)__builtin_cpu_is("sapphirerapids");
(void)__builtin_cpu_is("tremont");
(void)__builtin_cpu_is("westmere");
(void)__builtin_cpu_is("znver1");
diff --git a/clang/test/Driver/x86-march.c b/clang/test/Driver/x86-march.c
index 87fdf624a969..57882a5bf637 100644
--- a/clang/test/Driver/x86-march.c
+++ b/clang/test/Driver/x86-march.c
@@ -104,6 +104,10 @@
// RUN: | FileCheck %s -check-prefix=tremont
// tremont: "-target-cpu" "tremont"
//
+// RUN: %clang -target x86_64-unknown-unknown -c -### %s -march=sapphirerapids 2>&1 \
+// RUN: | FileCheck %s -check-prefix=sapphirerapids
+// sapphirerapids: "-target-cpu" "sapphirerapids"
+//
// RUN: %clang -target x86_64-unknown-unknown -c -### %s -march=k8 2>&1 \
// RUN: | FileCheck %s -check-prefix=k8
// k8: "-target-cpu" "k8"
diff --git a/clang/test/Misc/target-invalid-cpu-note.c b/clang/test/Misc/target-invalid-cpu-note.c
index 803b784bc0ec..546ab6341f97 100644
--- a/clang/test/Misc/target-invalid-cpu-note.c
+++ b/clang/test/Misc/target-invalid-cpu-note.c
@@ -21,7 +21,7 @@
// X86-SAME: nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont,
// X86-SAME: nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge,
// X86-SAME: core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512,
-// X86-SAME: skx, cascadelake, cooperlake, cannonlake, icelake-client, icelake-server, tigerlake, knl, knm, lakemont, k6, k6-2, k6-3,
+// X86-SAME: skx, cascadelake, cooperlake, cannonlake, icelake-client, icelake-server, tigerlake, sapphirerapids, knl, knm, lakemont, k6, k6-2, k6-3,
// X86-SAME: athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64,
// X86-SAME: athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10,
// X86-SAME: barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2,
@@ -33,7 +33,7 @@
// X86_64-SAME: atom, silvermont, slm, goldmont, goldmont-plus, tremont, nehalem, corei7, westmere,
// X86_64-SAME: sandybridge, corei7-avx, ivybridge, core-avx-i, haswell,
// X86_64-SAME: core-avx2, broadwell, skylake, skylake-avx512, skx, cascadelake, cooperlake, cannonlake,
-// X86_64-SAME: icelake-client, icelake-server, tigerlake, knl, knm, k8, athlon64, athlon-fx, opteron, k8-sse3,
+// X86_64-SAME: icelake-client, icelake-server, tigerlake, sapphirerapids, knl, knm, k8, athlon64, athlon-fx, opteron, k8-sse3,
// X86_64-SAME: athlon64-sse3, opteron-sse3, amdfam10, barcelona, btver1,
// X86_64-SAME: btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2, x86-64
@@ -45,7 +45,7 @@
// 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: skx, cascadelake, cooperlake, cannonlake, icelake-client, icelake-server, tigerlake, sapphirerapids, 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,
@@ -59,7 +59,7 @@
// TUNE_X86_64-SAME: nocona, core2, penryn, bonnell, atom, silvermont, slm, goldmont, goldmont-plus, tremont,
// TUNE_X86_64-SAME: nehalem, corei7, westmere, sandybridge, corei7-avx, ivybridge,
// TUNE_X86_64-SAME: core-avx-i, haswell, core-avx2, broadwell, skylake, skylake-avx512,
-// TUNE_X86_64-SAME: skx, cascadelake, cooperlake, cannonlake, icelake-client, icelake-server, tigerlake, knl, knm, lakemont, k6, k6-2, k6-3,
+// TUNE_X86_64-SAME: skx, cascadelake, cooperlake, cannonlake, icelake-client, icelake-server, tigerlake, sapphirerapids, knl, knm, lakemont, k6, k6-2, k6-3,
// TUNE_X86_64-SAME: athlon, athlon-tbird, athlon-xp, athlon-mp, athlon-4, k8, athlon64,
// TUNE_X86_64-SAME: athlon-fx, opteron, k8-sse3, athlon64-sse3, opteron-sse3, amdfam10,
// TUNE_X86_64-SAME: barcelona, btver1, btver2, bdver1, bdver2, bdver3, bdver4, znver1, znver2,
diff --git a/clang/test/Preprocessor/predefined-arch-macros.c b/clang/test/Preprocessor/predefined-arch-macros.c
index abab9274ffbb..5326596fee93 100644
--- a/clang/test/Preprocessor/predefined-arch-macros.c
+++ b/clang/test/Preprocessor/predefined-arch-macros.c
@@ -1629,6 +1629,145 @@
// CHECK_TGL_M64: #define __x86_64 1
// CHECK_TGL_M64: #define __x86_64__ 1
+// 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
+// CHECK_SPR_M32: #define __AES__ 1
+// CHECK_SPR_M32: #define __AMXBF16__ 1
+// CHECK_SPR_M32: #define __AMXINT8__ 1
+// CHECK_SPR_M32: #define __AMXTILE__ 1
+// CHECK_SPR_M32: #define __AVX2__ 1
+// CHECK_SPR_M32: #define __AVX512BF16__ 1
+// CHECK_SPR_M32: #define __AVX512BITALG__ 1
+// CHECK_SPR_M32: #define __AVX512BW__ 1
+// CHECK_SPR_M32: #define __AVX512CD__ 1
+// CHECK_SPR_M32: #define __AVX512DQ__ 1
+// CHECK_SPR_M32: #define __AVX512F__ 1
+// CHECK_SPR_M32: #define __AVX512IFMA__ 1
+// CHECK_SPR_M32: #define __AVX512VBMI2__ 1
+// CHECK_SPR_M32: #define __AVX512VBMI__ 1
+// CHECK_SPR_M32: #define __AVX512VL__ 1
+// CHECK_SPR_M32: #define __AVX512VNNI__ 1
+// CHECK_SPR_M32: #define __AVX512VPOPCNTDQ__ 1
+// CHECK_SPR_M32: #define __AVX__ 1
+// CHECK_SPR_M32: #define __BMI2__ 1
+// CHECK_SPR_M32: #define __BMI__ 1
+// CHECK_SPR_M32: #define __CLDEMOTE__ 1
+// CHECK_SPR_M32: #define __CLFLUSHOPT__ 1
+// CHECK_SPR_M32: #define __CLWB__ 1
+// CHECK_SPR_M32: #define __ENQCMD__ 1
+// CHECK_SPR_M32: #define __F16C__ 1
+// CHECK_SPR_M32: #define __FMA__ 1
+// CHECK_SPR_M32: #define __GFNI__ 1
+// CHECK_SPR_M32: #define __INVPCID__ 1
+// CHECK_SPR_M32: #define __LZCNT__ 1
+// CHECK_SPR_M32: #define __MMX__ 1
+// CHECK_SPR_M32: #define __MOVBE__ 1
+// CHECK_SPR_M32: #define __PCLMUL__ 1
+// CHECK_SPR_M32: #define __PCONFIG__ 1
+// CHECK_SPR_M32: #define __PKU__ 1
+// CHECK_SPR_M32: #define __POPCNT__ 1
+// CHECK_SPR_M32: #define __PRFCHW__ 1
+// CHECK_SPR_M32: #define __PTWRITE__ 1
+// CHECK_SPR_M32: #define __RDPID__ 1
+// CHECK_SPR_M32: #define __RDRND__ 1
+// CHECK_SPR_M32: #define __RDSEED__ 1
+// CHECK_SPR_M32: #define __SERIALIZE__ 1
+// CHECK_SPR_M32: #define __SGX__ 1
+// CHECK_SPR_M32: #define __SHA__ 1
+// CHECK_SPR_M32: #define __SHSTK__ 1
+// CHECK_SPR_M32: #define __SSE2__ 1
+// CHECK_SPR_M32: #define __SSE3__ 1
+// CHECK_SPR_M32: #define __SSE4_1__ 1
+// CHECK_SPR_M32: #define __SSE4_2__ 1
+// CHECK_SPR_M32: #define __SSE__ 1
+// CHECK_SPR_M32: #define __SSSE3__ 1
+// CHECK_SPR_M32: #define __TSXLDTRK__ 1
+// CHECK_SPR_M32: #define __VAES__ 1
+// CHECK_SPR_M32: #define __VPCLMULQDQ__ 1
+// CHECK_SPR_M32: #define __WAITPKG__ 1
+// CHECK_SPR_M32: #define __WBNOINVD__ 1
+// CHECK_SPR_M32: #define __XSAVEC__ 1
+// CHECK_SPR_M32: #define __XSAVEOPT__ 1
+// CHECK_SPR_M32: #define __XSAVES__ 1
+// CHECK_SPR_M32: #define __XSAVE__ 1
+// CHECK_SPR_M32: #define __corei7 1
+// CHECK_SPR_M32: #define __corei7__ 1
+// CHECK_SPR_M32: #define __i386 1
+// CHECK_SPR_M32: #define __i386__ 1
+// CHECK_SPR_M32: #define __tune_corei7__ 1
+// CHECK_SPR_M32: #define i386 1
+
+// 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
+// CHECK_SPR_M64: #define __AES__ 1
+// CHECK_SPR_M64: #define __AMXBF16__ 1
+// CHECK_SPR_M64: #define __AMXINT8__ 1
+// CHECK_SPR_M64: #define __AMXTILE__ 1
+// CHECK_SPR_M64: #define __AVX2__ 1
+// CHECK_SPR_M64: #define __AVX512BF16__ 1
+// CHECK_SPR_M64: #define __AVX512BITALG__ 1
+// CHECK_SPR_M64: #define __AVX512BW__ 1
+// CHECK_SPR_M64: #define __AVX512CD__ 1
+// CHECK_SPR_M64: #define __AVX512DQ__ 1
+// CHECK_SPR_M64: #define __AVX512F__ 1
+// CHECK_SPR_M64: #define __AVX512IFMA__ 1
+// CHECK_SPR_M64: #define __AVX512VBMI2__ 1
+// CHECK_SPR_M64: #define __AVX512VBMI__ 1
+// CHECK_SPR_M64: #define __AVX512VL__ 1
+// CHECK_SPR_M64: #define __AVX512VNNI__ 1
+// CHECK_SPR_M64: #define __AVX512VPOPCNTDQ__ 1
+// CHECK_SPR_M64: #define __AVX__ 1
+// CHECK_SPR_M64: #define __BMI2__ 1
+// CHECK_SPR_M64: #define __BMI__ 1
+// CHECK_SPR_M64: #define __CLDEMOTE__ 1
+// CHECK_SPR_M64: #define __CLFLUSHOPT__ 1
+// CHECK_SPR_M64: #define __CLWB__ 1
+// CHECK_SPR_M64: #define __ENQCMD__ 1
+// CHECK_SPR_M64: #define __F16C__ 1
+// CHECK_SPR_M64: #define __FMA__ 1
+// CHECK_SPR_M64: #define __GFNI__ 1
+// CHECK_SPR_M64: #define __INVPCID__ 1
+// CHECK_SPR_M64: #define __LZCNT__ 1
+// CHECK_SPR_M64: #define __MMX__ 1
+// CHECK_SPR_M64: #define __MOVBE__ 1
+// CHECK_SPR_M64: #define __PCLMUL__ 1
+// CHECK_SPR_M64: #define __PCONFIG__ 1
+// CHECK_SPR_M64: #define __PKU__ 1
+// CHECK_SPR_M64: #define __POPCNT__ 1
+// CHECK_SPR_M64: #define __PRFCHW__ 1
+// CHECK_SPR_M64: #define __PTWRITE__ 1
+// CHECK_SPR_M64: #define __RDPID__ 1
+// CHECK_SPR_M64: #define __RDRND__ 1
+// CHECK_SPR_M64: #define __RDSEED__ 1
+// CHECK_SPR_M64: #define __SERIALIZE__ 1
+// CHECK_SPR_M64: #define __SGX__ 1
+// CHECK_SPR_M64: #define __SHA__ 1
+// CHECK_SPR_M64: #define __SHSTK__ 1
+// CHECK_SPR_M64: #define __SSE2__ 1
+// CHECK_SPR_M64: #define __SSE3__ 1
+// CHECK_SPR_M64: #define __SSE4_1__ 1
+// CHECK_SPR_M64: #define __SSE4_2__ 1
+// CHECK_SPR_M64: #define __SSE__ 1
+// CHECK_SPR_M64: #define __SSSE3__ 1
+// CHECK_SPR_M64: #define __TSXLDTRK__ 1
+// CHECK_SPR_M64: #define __VAES__ 1
+// CHECK_SPR_M64: #define __VPCLMULQDQ__ 1
+// CHECK_SPR_M64: #define __WAITPKG__ 1
+// CHECK_SPR_M64: #define __WBNOINVD__ 1
+// CHECK_SPR_M64: #define __XSAVEC__ 1
+// CHECK_SPR_M64: #define __XSAVEOPT__ 1
+// CHECK_SPR_M64: #define __XSAVES__ 1
+// CHECK_SPR_M64: #define __XSAVE__ 1
+// CHECK_SPR_M64: #define __amd64 1
+// CHECK_SPR_M64: #define __amd64__ 1
+// CHECK_SPR_M64: #define __corei7 1
+// CHECK_SPR_M64: #define __corei7__ 1
+// CHECK_SPR_M64: #define __tune_corei7__ 1
+// CHECK_SPR_M64: #define __x86_64 1
+// CHECK_SPR_M64: #define __x86_64__ 1
+
// RUN: %clang -march=atom -m32 -E -dM %s -o - 2>&1 \
// RUN: -target i386-unknown-linux \
// RUN: | FileCheck -match-full-lines %s -check-prefix=CHECK_ATOM_M32
diff --git a/compiler-rt/lib/builtins/cpu_model.c b/compiler-rt/lib/builtins/cpu_model.c
index 468bcc84cbcb..d6dc73b88bc2 100644
--- a/compiler-rt/lib/builtins/cpu_model.c
+++ b/compiler-rt/lib/builtins/cpu_model.c
@@ -84,6 +84,7 @@ enum ProcessorSubtypes {
INTEL_COREI7_CASCADELAKE,
INTEL_COREI7_TIGERLAKE,
INTEL_COREI7_COOPERLAKE,
+ INTEL_COREI7_SAPPHIRERAPIDS,
CPU_SUBTYPE_MAX
};
@@ -407,6 +408,13 @@ getIntelProcessorTypeAndSubtype(unsigned Family, unsigned Model,
*Subtype = INTEL_COREI7_ICELAKE_SERVER;
break;
+ // Sapphire Rapids:
+ case 0x8f:
+ CPU = "sapphirerapids";
+ *Type = INTEL_COREI7;
+ *Subtype = INTEL_COREI7_SAPPHIRERAPIDS;
+ break;
+
case 0x1c: // Most 45 nm Intel Atom processors
case 0x26: // 45 nm Atom Lincroft
case 0x27: // 32 nm Atom Medfield
diff --git a/llvm/include/llvm/Support/X86TargetParser.def b/llvm/include/llvm/Support/X86TargetParser.def
index 697f8c70f962..e3998c99a50a 100644
--- a/llvm/include/llvm/Support/X86TargetParser.def
+++ b/llvm/include/llvm/Support/X86TargetParser.def
@@ -84,6 +84,7 @@ X86_CPU_SUBTYPE(AMDFAM17H_ZNVER2, "znver2")
X86_CPU_SUBTYPE(INTEL_COREI7_CASCADELAKE, "cascadelake")
X86_CPU_SUBTYPE(INTEL_COREI7_TIGERLAKE, "tigerlake")
X86_CPU_SUBTYPE(INTEL_COREI7_COOPERLAKE, "cooperlake")
+X86_CPU_SUBTYPE(INTEL_COREI7_SAPPHIRERAPIDS, "sapphirerapids")
#undef X86_CPU_SUBTYPE
diff --git a/llvm/include/llvm/Support/X86TargetParser.h b/llvm/include/llvm/Support/X86TargetParser.h
index 36d8f41df6ec..d97f620419e1 100644
--- a/llvm/include/llvm/Support/X86TargetParser.h
+++ b/llvm/include/llvm/Support/X86TargetParser.h
@@ -100,6 +100,7 @@ enum CPUKind {
CK_IcelakeClient,
CK_IcelakeServer,
CK_Tigerlake,
+ CK_SapphireRapids,
CK_KNL,
CK_KNM,
CK_Lakemont,
diff --git a/llvm/lib/Support/Host.cpp b/llvm/lib/Support/Host.cpp
index bc06a7858f5b..26534580d02d 100644
--- a/llvm/lib/Support/Host.cpp
+++ b/llvm/lib/Support/Host.cpp
@@ -730,6 +730,13 @@ getIntelProcessorTypeAndSubtype(unsigned Family, unsigned Model,
*Subtype = X86::INTEL_COREI7_ICELAKE_SERVER;
break;
+ // Sapphire Rapids:
+ case 0x8f:
+ CPU = "sapphirerapids";
+ *Type = X86::INTEL_COREI7;
+ *Subtype = X86::INTEL_COREI7_SAPPHIRERAPIDS;
+ break;
+
case 0x1c: // Most 45 nm Intel Atom processors
case 0x26: // 45 nm Atom Lincroft
case 0x27: // 32 nm Atom Medfield
diff --git a/llvm/lib/Support/X86TargetParser.cpp b/llvm/lib/Support/X86TargetParser.cpp
index 680ec91dc8ef..a5af98582452 100644
--- a/llvm/lib/Support/X86TargetParser.cpp
+++ b/llvm/lib/Support/X86TargetParser.cpp
@@ -195,6 +195,11 @@ static constexpr FeatureBitset FeaturesICLServer =
static constexpr FeatureBitset FeaturesTigerlake =
FeaturesICLClient | FeatureAVX512VP2INTERSECT | FeatureMOVDIR64B |
FeatureMOVDIRI | FeatureSHSTK;
+static constexpr FeatureBitset FeaturesSapphireRapids =
+ FeaturesICLServer | FeatureAMX_TILE | FeatureAMX_INT8 | FeatureAMX_BF16 |
+ FeatureAVX512BF16 | FeatureAVX512VP2INTERSECT | FeatureCLDEMOTE | FeatureENQCMD |
+ FeatureMOVDIR64B | FeatureMOVDIRI | FeaturePTWRITE | FeatureSERIALIZE |
+ FeatureSHSTK | FeatureTSXLDTRK | FeatureWAITPKG;
// Intel Atom processors.
// Bonnell has feature parity with Core2 and adds MOVBE.
@@ -342,6 +347,8 @@ static constexpr ProcInfo Processors[] = {
{ {"icelake-server"}, CK_IcelakeServer, FEATURE_AVX512VBMI2, FeaturesICLServer },
// Tigerlake microarchitecture based processors.
{ {"tigerlake"}, CK_Tigerlake, FEATURE_AVX512VP2INTERSECT, FeaturesTigerlake },
+ // Sapphire Rapids microarchitecture based processors.
+ { {"sapphirerapids"}, CK_SapphireRapids, FEATURE_AVX512VP2INTERSECT, FeaturesSapphireRapids },
// Knights Landing processor.
{ {"knl"}, CK_KNL, FEATURE_AVX512F, FeaturesKNL },
// Knights Mill processor.
diff --git a/llvm/lib/Target/X86/X86.td b/llvm/lib/Target/X86/X86.td
index 74e6c89ca9b4..99eb8399d24f 100644
--- a/llvm/lib/Target/X86/X86.td
+++ b/llvm/lib/Target/X86/X86.td
@@ -741,6 +741,25 @@ def ProcessorFeatures {
list<SubtargetFeature> TGLFeatures =
!listconcat(ICLFeatures, TGLAdditionalFeatures );
+ //Sapphirerapids
+ list<SubtargetFeature> SPRAdditionalFeatures = [FeatureAMXTILE,
+ FeatureAMXINT8,
+ FeatureAMXBF16,
+ FeatureBF16,
+ FeatureSERIALIZE,
+ FeatureCLDEMOTE,
+ FeatureWAITPKG,
+ FeaturePTWRITE,
+ FeatureTSXLDTRK,
+ FeatureENQCMD,
+ FeatureSHSTK,
+ FeatureVP2INTERSECT,
+ FeatureMOVDIRI,
+ FeatureMOVDIR64B];
+ list<SubtargetFeature> SPRTuning = ICXTuning;
+ list<SubtargetFeature> SPRFeatures =
+ !listconcat(ICXFeatures, SPRAdditionalFeatures);
+
// Atom
list<SubtargetFeature> AtomFeatures = [FeatureX87,
FeatureCMPXCHG8B,
@@ -1243,6 +1262,8 @@ def : ProcModel<"icelake-server", SkylakeServerModel,
ProcessorFeatures.ICXFeatures, ProcessorFeatures.ICXTuning>;
def : ProcModel<"tigerlake", SkylakeServerModel,
ProcessorFeatures.TGLFeatures, ProcessorFeatures.TGLTuning>;
+def : ProcModel<"sapphirerapids", SkylakeServerModel,
+ ProcessorFeatures.SPRFeatures, ProcessorFeatures.SPRTuning>;
// AMD CPUs.
diff --git a/llvm/test/CodeGen/X86/cpus-intel.ll b/llvm/test/CodeGen/X86/cpus-intel.ll
index e0dd647409f9..b065c6d34f80 100644
--- a/llvm/test/CodeGen/X86/cpus-intel.ll
+++ b/llvm/test/CodeGen/X86/cpus-intel.ll
@@ -40,6 +40,7 @@
; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=icelake-client 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=icelake-server 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=tigerlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
+; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=sapphirerapids 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=atom 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=bonnell 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=silvermont 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty
More information about the llvm-commits
mailing list