https://github.com/ganeshgit updated https://github.com/llvm/llvm-project/pull/179150
>From 019e139b587fe0980e4c1fa64b0c3e98806bef3f Mon Sep 17 00:00:00 2001 From: Ganesh Gopalasubramanian <[email protected]> Date: Mon, 2 Feb 2026 12:36:42 +0000 Subject: [PATCH] [X86] Add basic AMD Zen 6 (znver6) CPU support This patch adds initial support for AMD Zen 6 architecture (znver6): - Added znver6 CPU target recognition in Clang and LLVM - Added znver6 to target parser and host CPU detection - Updated scheduling models to inherit from znver4 - Added znver6 to various optimizer tests - Updated compiler-rt CPU model detection for znver6 znver6 features: FP16, AVXVNNIINT8, AVXNECONVERT, AVXIFMA (without BMM). Change-Id: I06958a56cacc7538a42045bd4ff2df81ae966ab1 --- clang/docs/ReleaseNotes.rst | 1 + clang/lib/Basic/Targets/X86.cpp | 4 + clang/test/CodeGen/target-builtin-noerror.c | 1 + clang/test/Driver/x86-march.c | 4 + clang/test/Frontend/x86-target-cpu.c | 1 + clang/test/Misc/target-invalid-cpu-note/x86.c | 4 + .../Preprocessor/predefined-arch-macros.c | 151 ++++++++++++++++++ compiler-rt/lib/builtins/cpu_model/x86.c | 19 ++- .../llvm/TargetParser/X86TargetParser.def | 1 + .../llvm/TargetParser/X86TargetParser.h | 1 + llvm/lib/Target/X86/X86.td | 11 ++ llvm/lib/Target/X86/X86PfmCounters.td | 1 + llvm/lib/Target/X86/X86ScheduleZnver4.td | 32 ++-- llvm/lib/TargetParser/Host.cpp | 6 + llvm/lib/TargetParser/X86TargetParser.cpp | 5 + .../CodeGen/X86/bypass-slow-division-64.ll | 1 + llvm/test/CodeGen/X86/cmp16.ll | 1 + llvm/test/CodeGen/X86/cpus-amd.ll | 1 + llvm/test/CodeGen/X86/rdpru.ll | 1 + llvm/test/CodeGen/X86/shuffle-as-shifts.ll | 1 + llvm/test/CodeGen/X86/slow-unaligned-mem.ll | 1 + llvm/test/CodeGen/X86/sqrt-fastmath-tune.ll | 1 + .../X86/tuning-shuffle-permilpd-avx512.ll | 1 + .../X86/tuning-shuffle-permilps-avx512.ll | 1 + .../X86/tuning-shuffle-unpckpd-avx512.ll | 1 + .../X86/tuning-shuffle-unpckps-avx512.ll | 1 + .../X86/vector-shuffle-fast-per-lane.ll | 1 + llvm/test/CodeGen/X86/vpdpwssd.ll | 1 + .../CodeGen/X86/x86-64-double-shifts-var.ll | 1 + llvm/test/MC/X86/x86_long_nop.s | 2 + .../Transforms/LoopUnroll/X86/call-remark.ll | 1 + .../Transforms/SLPVectorizer/X86/pr63668.ll | 1 + 32 files changed, 234 insertions(+), 26 deletions(-) diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst index 3a3d76112a02b..a510610741c2a 100644 --- a/clang/docs/ReleaseNotes.rst +++ b/clang/docs/ReleaseNotes.rst @@ -265,6 +265,7 @@ NVPTX Support X86 Support ^^^^^^^^^^^ +- ``march=znver6`` is now supported. Arm and AArch64 Support ^^^^^^^^^^^^^^^^^^^^^^^ diff --git a/clang/lib/Basic/Targets/X86.cpp b/clang/lib/Basic/Targets/X86.cpp index f00d435937b92..881a06adf5ea5 100644 --- a/clang/lib/Basic/Targets/X86.cpp +++ b/clang/lib/Basic/Targets/X86.cpp @@ -716,6 +716,9 @@ void X86TargetInfo::getTargetDefines(const LangOptions &Opts, case CK_ZNVER5: defineCPUMacros(Builder, "znver5"); break; + case CK_ZNVER6: + defineCPUMacros(Builder, "znver6"); + break; case CK_Geode: defineCPUMacros(Builder, "geode"); break; @@ -1641,6 +1644,7 @@ std::optional<unsigned> X86TargetInfo::getCPUCacheLineSize() const { case CK_ZNVER3: case CK_ZNVER4: case CK_ZNVER5: + case CK_ZNVER6: // Deprecated case CK_x86_64: case CK_x86_64_v2: diff --git a/clang/test/CodeGen/target-builtin-noerror.c b/clang/test/CodeGen/target-builtin-noerror.c index 47d5ae51d643a..a65a07d81b8c0 100644 --- a/clang/test/CodeGen/target-builtin-noerror.c +++ b/clang/test/CodeGen/target-builtin-noerror.c @@ -209,5 +209,6 @@ void verifycpustrings(void) { (void)__builtin_cpu_is("znver3"); (void)__builtin_cpu_is("znver4"); (void)__builtin_cpu_is("znver5"); + (void)__builtin_cpu_is("znver6"); (void)__builtin_cpu_is("diamondrapids"); } diff --git a/clang/test/Driver/x86-march.c b/clang/test/Driver/x86-march.c index 15f82547892c2..6a3ef5be67d8a 100644 --- a/clang/test/Driver/x86-march.c +++ b/clang/test/Driver/x86-march.c @@ -258,6 +258,10 @@ // RUN: %clang -target x86_64-unknown-unknown -c -### %s -march=znver5 2>&1 \ // RUN: | FileCheck %s -check-prefix=znver5 // znver5: "-target-cpu" "znver5" +// +// RUN: %clang -target x86_64-unknown-unknown -c -### %s -march=znver6 2>&1 \ +// RUN: | FileCheck %s -check-prefix=znver6 +// znver6: "-target-cpu" "znver6" // RUN: %clang -target x86_64 -c -### %s -march=x86-64 2>&1 | FileCheck %s --check-prefix=x86-64 // x86-64: "-target-cpu" "x86-64" diff --git a/clang/test/Frontend/x86-target-cpu.c b/clang/test/Frontend/x86-target-cpu.c index f2885a040c370..7dc7f5474687e 100644 --- a/clang/test/Frontend/x86-target-cpu.c +++ b/clang/test/Frontend/x86-target-cpu.c @@ -39,5 +39,6 @@ // RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-cpu znver3 -verify %s // RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-cpu znver4 -verify %s // RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-cpu znver5 -verify %s +// RUN: %clang_cc1 -triple x86_64-unknown-unknown -target-cpu znver6 -verify %s // // expected-no-diagnostics diff --git a/clang/test/Misc/target-invalid-cpu-note/x86.c b/clang/test/Misc/target-invalid-cpu-note/x86.c index 4a70e9bff3fef..766bd679796f5 100644 --- a/clang/test/Misc/target-invalid-cpu-note/x86.c +++ b/clang/test/Misc/target-invalid-cpu-note/x86.c @@ -103,6 +103,7 @@ // X86-SAME: {{^}}, znver3 // X86-SAME: {{^}}, znver4 // X86-SAME: {{^}}, znver5 +// X86-SAME: {{^}}, znver6 // X86-SAME: {{^}}, x86-64 // X86-SAME: {{^}}, x86-64-v2 // X86-SAME: {{^}}, x86-64-v3 @@ -183,6 +184,7 @@ // X86_64-SAME: {{^}}, znver3 // X86_64-SAME: {{^}}, znver4 // X86_64-SAME: {{^}}, znver5 +// X86_64-SAME: {{^}}, znver6 // X86_64-SAME: {{^}}, x86-64 // X86_64-SAME: {{^}}, x86-64-v2 // X86_64-SAME: {{^}}, x86-64-v3 @@ -290,6 +292,7 @@ // TUNE_X86-SAME: {{^}}, znver3 // TUNE_X86-SAME: {{^}}, znver4 // TUNE_X86-SAME: {{^}}, znver5 +// TUNE_X86-SAME: {{^}}, znver6 // TUNE_X86-SAME: {{^}}, x86-64 // TUNE_X86-SAME: {{^}}, geode // TUNE_X86-SAME: {{$}} @@ -395,6 +398,7 @@ // TUNE_X86_64-SAME: {{^}}, znver3 // TUNE_X86_64-SAME: {{^}}, znver4 // TUNE_X86_64-SAME: {{^}}, znver5 +// TUNE_X86_64-SAME: {{^}}, znver6 // TUNE_X86_64-SAME: {{^}}, x86-64 // TUNE_X86_64-SAME: {{^}}, geode // TUNE_X86_64-SAME: {{$}} diff --git a/clang/test/Preprocessor/predefined-arch-macros.c b/clang/test/Preprocessor/predefined-arch-macros.c index 1e38b4d3ba350..be94eb064cf91 100644 --- a/clang/test/Preprocessor/predefined-arch-macros.c +++ b/clang/test/Preprocessor/predefined-arch-macros.c @@ -4133,6 +4133,157 @@ // CHECK_ZNVER5_M64: #define __znver5 1 // CHECK_ZNVER5_M64: #define __znver5__ 1 +// RUN: %clang -march=znver6 -m32 -E -dM %s -o - 2>&1 \ +// RUN: -target i386-unknown-linux \ +// RUN: | FileCheck -match-full-lines %s -check-prefix=CHECK_ZNVER6_M32 +// CHECK_ZNVER6_M32-NOT: #define __3dNOW_A__ 1 +// CHECK_ZNVER6_M32-NOT: #define __3dNOW__ 1 +// CHECK_ZNVER6_M32: #define __ADX__ 1 +// CHECK_ZNVER6_M32: #define __AES__ 1 +// CHECK_ZNVER6_M32: #define __AVX2__ 1 +// CHECK_ZNVER6_M32: #define __AVX512BF16__ 1 +// CHECK_ZNVER6_M32: #define __AVX512BITALG__ 1 +// CHECK_ZNVER6_M32: #define __AVX512BW__ 1 +// CHECK_ZNVER6_M32: #define __AVX512CD__ 1 +// CHECK_ZNVER6_M32: #define __AVX512DQ__ 1 +// CHECK_ZNVER6_M32: #define __AVX512FP16__ 1 +// CHECK_ZNVER6_M32: #define __AVX512F__ 1 +// CHECK_ZNVER6_M32: #define __AVX512IFMA__ 1 +// CHECK_ZNVER6_M32: #define __AVX512VBMI2__ 1 +// CHECK_ZNVER6_M32: #define __AVX512VBMI__ 1 +// CHECK_ZNVER6_M32: #define __AVX512VL__ 1 +// CHECK_ZNVER6_M32: #define __AVX512VNNI__ 1 +// CHECK_ZNVER6_M32: #define __AVX512VP2INTERSECT__ 1 +// CHECK_ZNVER6_M32: #define __AVX512VPOPCNTDQ__ 1 +// CHECK_ZNVER6_M32: #define __AVXIFMA__ 1 +// CHECK_ZNVER6_M32: #define __AVXNECONVERT__ 1 +// CHECK_ZNVER6_M32: #define __AVXVNNIINT8__ 1 +// CHECK_ZNVER6_M32: #define __AVXVNNI__ 1 +// CHECK_ZNVER6_M32: #define __AVX__ 1 +// CHECK_ZNVER6_M32: #define __BMI2__ 1 +// CHECK_ZNVER6_M32: #define __BMI__ 1 +// CHECK_ZNVER6_M32: #define __CLFLUSHOPT__ 1 +// CHECK_ZNVER6_M32: #define __CLWB__ 1 +// CHECK_ZNVER6_M32: #define __CLZERO__ 1 +// CHECK_ZNVER6_M32: #define __F16C__ 1 +// CHECK_ZNVER6_M32-NOT: #define __FMA4__ 1 +// CHECK_ZNVER6_M32: #define __FMA__ 1 +// CHECK_ZNVER6_M32: #define __FSGSBASE__ 1 +// CHECK_ZNVER6_M32: #define __GFNI__ 1 +// CHECK_ZNVER6_M32: #define __LZCNT__ 1 +// CHECK_ZNVER6_M32: #define __MMX__ 1 +// CHECK_ZNVER6_M32: #define __MOVDIR64B__ 1 +// CHECK_ZNVER6_M32: #define __MOVDIRI__ 1 +// CHECK_ZNVER6_M32: #define __PCLMUL__ 1 +// CHECK_ZNVER6_M32: #define __PKU__ 1 +// CHECK_ZNVER6_M32: #define __POPCNT__ 1 +// CHECK_ZNVER6_M32: #define __PREFETCHI__ 1 +// CHECK_ZNVER6_M32: #define __PRFCHW__ 1 +// CHECK_ZNVER6_M32: #define __RDPID__ 1 +// CHECK_ZNVER6_M32: #define __RDPRU__ 1 +// CHECK_ZNVER6_M32: #define __RDRND__ 1 +// CHECK_ZNVER6_M32: #define __RDSEED__ 1 +// CHECK_ZNVER6_M32: #define __SHA__ 1 +// CHECK_ZNVER6_M32: #define __SSE2_MATH__ 1 +// CHECK_ZNVER6_M32: #define __SSE2__ 1 +// CHECK_ZNVER6_M32: #define __SSE3__ 1 +// CHECK_ZNVER6_M32: #define __SSE4A__ 1 +// CHECK_ZNVER6_M32: #define __SSE4_1__ 1 +// CHECK_ZNVER6_M32: #define __SSE4_2__ 1 +// CHECK_ZNVER6_M32: #define __SSE_MATH__ 1 +// CHECK_ZNVER6_M32: #define __SSE__ 1 +// CHECK_ZNVER6_M32: #define __SSSE3__ 1 +// CHECK_ZNVER6_M32-NOT: #define __TBM__ 1 +// CHECK_ZNVER6_M32: #define __WBNOINVD__ 1 +// CHECK_ZNVER6_M32-NOT: #define __XOP__ 1 +// CHECK_ZNVER6_M32: #define __XSAVEC__ 1 +// CHECK_ZNVER6_M32: #define __XSAVEOPT__ 1 +// CHECK_ZNVER6_M32: #define __XSAVES__ 1 +// CHECK_ZNVER6_M32: #define __XSAVE__ 1 +// CHECK_ZNVER6_M32: #define __i386 1 +// CHECK_ZNVER6_M32: #define __i386__ 1 +// CHECK_ZNVER6_M32: #define __tune_znver6__ 1 +// CHECK_ZNVER6_M32: #define __znver6 1 +// CHECK_ZNVER6_M32: #define __znver6__ 1 + +// RUN: %clang -march=znver6 -m64 -E -dM %s -o - 2>&1 \ +// RUN: -target i386-unknown-linux \ +// RUN: | FileCheck -match-full-lines %s -check-prefix=CHECK_ZNVER6_M64 +// CHECK_ZNVER6_M64-NOT: #define __3dNOW_A__ 1 +// CHECK_ZNVER6_M64-NOT: #define __3dNOW__ 1 +// CHECK_ZNVER6_M64: #define __ADX__ 1 +// CHECK_ZNVER6_M64: #define __AES__ 1 +// CHECK_ZNVER6_M64: #define __AVX2__ 1 +// CHECK_ZNVER6_M64: #define __AVX512BF16__ 1 +// CHECK_ZNVER6_M64: #define __AVX512BITALG__ 1 +// CHECK_ZNVER6_M64: #define __AVX512BW__ 1 +// CHECK_ZNVER6_M64: #define __AVX512CD__ 1 +// CHECK_ZNVER6_M64: #define __AVX512DQ__ 1 +// CHECK_ZNVER6_M64: #define __AVX512FP16__ 1 +// CHECK_ZNVER6_M64: #define __AVX512F__ 1 +// CHECK_ZNVER6_M64: #define __AVX512IFMA__ 1 +// CHECK_ZNVER6_M64: #define __AVX512VBMI2__ 1 +// CHECK_ZNVER6_M64: #define __AVX512VBMI__ 1 +// CHECK_ZNVER6_M64: #define __AVX512VL__ 1 +// CHECK_ZNVER6_M64: #define __AVX512VNNI__ 1 +// CHECK_ZNVER6_M64: #define __AVX512VP2INTERSECT__ 1 +// CHECK_ZNVER6_M64: #define __AVX512VPOPCNTDQ__ 1 +// CHECK_ZNVER6_M64: #define __AVXIFMA__ 1 +// CHECK_ZNVER6_M64: #define __AVXNECONVERT__ 1 +// CHECK_ZNVER6_M64: #define __AVXVNNIINT8__ 1 +// CHECK_ZNVER6_M64: #define __AVXVNNI__ 1 +// CHECK_ZNVER6_M64: #define __AVX__ 1 +// CHECK_ZNVER6_M64: #define __BMI2__ 1 +// CHECK_ZNVER6_M64: #define __BMI__ 1 +// CHECK_ZNVER6_M64: #define __CLFLUSHOPT__ 1 +// CHECK_ZNVER6_M64: #define __CLWB__ 1 +// CHECK_ZNVER6_M64: #define __CLZERO__ 1 +// CHECK_ZNVER6_M64: #define __F16C__ 1 +// CHECK_ZNVER6_M64-NOT: #define __FMA4__ 1 +// CHECK_ZNVER6_M64: #define __FMA__ 1 +// CHECK_ZNVER6_M64: #define __FSGSBASE__ 1 +// CHECK_ZNVER6_M64: #define __GFNI__ 1 +// CHECK_ZNVER6_M64: #define __LZCNT__ 1 +// CHECK_ZNVER6_M64: #define __MMX__ 1 +// CHECK_ZNVER6_M64: #define __MOVDIR64B__ 1 +// CHECK_ZNVER6_M64: #define __MOVDIRI__ 1 +// CHECK_ZNVER6_M64: #define __PCLMUL__ 1 +// CHECK_ZNVER6_M64: #define __PKU__ 1 +// CHECK_ZNVER6_M64: #define __POPCNT__ 1 +// CHECK_ZNVER6_M64: #define __PREFETCHI__ 1 +// CHECK_ZNVER6_M64: #define __PRFCHW__ 1 +// CHECK_ZNVER6_M64: #define __RDPID__ 1 +// CHECK_ZNVER6_M64: #define __RDPRU__ 1 +// CHECK_ZNVER6_M64: #define __RDRND__ 1 +// CHECK_ZNVER6_M64: #define __RDSEED__ 1 +// CHECK_ZNVER6_M64: #define __SHA__ 1 +// CHECK_ZNVER6_M64: #define __SSE2_MATH__ 1 +// CHECK_ZNVER6_M64: #define __SSE2__ 1 +// CHECK_ZNVER6_M64: #define __SSE3__ 1 +// CHECK_ZNVER6_M64: #define __SSE4A__ 1 +// CHECK_ZNVER6_M64: #define __SSE4_1__ 1 +// CHECK_ZNVER6_M64: #define __SSE4_2__ 1 +// CHECK_ZNVER6_M64: #define __SSE_MATH__ 1 +// CHECK_ZNVER6_M64: #define __SSE__ 1 +// CHECK_ZNVER6_M64: #define __SSSE3__ 1 +// CHECK_ZNVER6_M64-NOT: #define __TBM__ 1 +// CHECK_ZNVER6_M64: #define __VAES__ 1 +// CHECK_ZNVER6_M64: #define __VPCLMULQDQ__ 1 +// CHECK_ZNVER6_M64: #define __WBNOINVD__ 1 +// CHECK_ZNVER6_M64-NOT: #define __XOP__ 1 +// CHECK_ZNVER6_M64: #define __XSAVEC__ 1 +// CHECK_ZNVER6_M64: #define __XSAVEOPT__ 1 +// CHECK_ZNVER6_M64: #define __XSAVES__ 1 +// CHECK_ZNVER6_M64: #define __XSAVE__ 1 +// CHECK_ZNVER6_M64: #define __amd64 1 +// CHECK_ZNVER6_M64: #define __amd64__ 1 +// CHECK_ZNVER6_M64: #define __tune_znver6__ 1 +// CHECK_ZNVER6_M64: #define __x86_64 1 +// CHECK_ZNVER6_M64: #define __x86_64__ 1 +// CHECK_ZNVER6_M64: #define __znver6 1 +// CHECK_ZNVER6_M64: #define __znver6__ 1 + + // End X86/GCC/Linux tests ------------------ // Begin PPC/GCC/Linux tests ---------------- diff --git a/compiler-rt/lib/builtins/cpu_model/x86.c b/compiler-rt/lib/builtins/cpu_model/x86.c index 55eb2b0958450..f4b95fa0b5a07 100644 --- a/compiler-rt/lib/builtins/cpu_model/x86.c +++ b/compiler-rt/lib/builtins/cpu_model/x86.c @@ -105,6 +105,7 @@ enum ProcessorSubtypes { INTEL_COREI7_ARROWLAKE_S, INTEL_COREI7_PANTHERLAKE, AMDFAM1AH_ZNVER5, + AMDFAM1AH_ZNVER6, INTEL_COREI7_DIAMONDRAPIDS, INTEL_COREI7_NOVALAKE, CPU_SUBTYPE_MAX @@ -837,20 +838,18 @@ getAMDProcessorTypeAndSubtype(unsigned Family, unsigned Model, case 26: CPU = "znver5"; Type = AMDFAM1AH; - if (Model <= 0x77) { - // Models 00h-0Fh (Breithorn). - // Models 10h-1Fh (Breithorn-Dense). - // Models 20h-2Fh (Strix 1). - // Models 30h-37h (Strix 2). - // Models 38h-3Fh (Strix 3). - // Models 40h-4Fh (Granite Ridge). - // Models 50h-5Fh (Weisshorn). - // Models 60h-6Fh (Krackan1). - // Models 70h-77h (Sarlak). + if (Model <= 0x4f || (Model >= 0x60 && Model <= 0x77) || + (Model >= 0xd0 && Model <= 0xd7)) { CPU = "znver5"; Subtype = AMDFAM1AH_ZNVER5; break; // "znver5" } + if ((Model >= 0x50 && Model <= 0x5f) || (Model >= 0x80 && Model <= 0xcf) || + (Model >= 0xd8 && Model <= 0xe7)) { + CPU = "znver6"; + Subtype = AMDFAM1AH_ZNVER6; + break; // "znver6" + } break; default: break; // Unknown AMD CPU. diff --git a/llvm/include/llvm/TargetParser/X86TargetParser.def b/llvm/include/llvm/TargetParser/X86TargetParser.def index 09592bcea27f4..bc05452400458 100644 --- a/llvm/include/llvm/TargetParser/X86TargetParser.def +++ b/llvm/include/llvm/TargetParser/X86TargetParser.def @@ -107,6 +107,7 @@ X86_CPU_SUBTYPE(INTEL_COREI7_ARROWLAKE, "arrowlake") X86_CPU_SUBTYPE(INTEL_COREI7_ARROWLAKE_S, "arrowlake-s") X86_CPU_SUBTYPE(INTEL_COREI7_PANTHERLAKE, "pantherlake") X86_CPU_SUBTYPE(AMDFAM1AH_ZNVER5, "znver5") +X86_CPU_SUBTYPE(AMDFAM1AH_ZNVER6, "znver6") X86_CPU_SUBTYPE(INTEL_COREI7_DIAMONDRAPIDS, "diamondrapids") X86_CPU_SUBTYPE(INTEL_COREI7_NOVALAKE, "novalake") diff --git a/llvm/include/llvm/TargetParser/X86TargetParser.h b/llvm/include/llvm/TargetParser/X86TargetParser.h index 46061f9d1fc7d..31d13ce29f7fc 100644 --- a/llvm/include/llvm/TargetParser/X86TargetParser.h +++ b/llvm/include/llvm/TargetParser/X86TargetParser.h @@ -146,6 +146,7 @@ enum CPUKind { CK_ZNVER3, CK_ZNVER4, CK_ZNVER5, + CK_ZNVER6, CK_x86_64, CK_x86_64_v2, CK_x86_64_v3, diff --git a/llvm/lib/Target/X86/X86.td b/llvm/lib/Target/X86/X86.td index 1b9a6ee2b4ef4..fa41d7a7a7c5e 100644 --- a/llvm/lib/Target/X86/X86.td +++ b/llvm/lib/Target/X86/X86.td @@ -1631,6 +1631,15 @@ def ProcessorFeatures { ]; list<SubtargetFeature> ZN5Features = !listconcat(ZN4Features, ZN5AdditionalFeatures); + + list<SubtargetFeature> ZN6Tuning = ZN5Tuning; + list<SubtargetFeature> ZN6AdditionalFeatures = [FeatureFP16, + FeatureAVXVNNIINT8, + FeatureAVXNECONVERT, + FeatureAVXIFMA + ]; + list<SubtargetFeature> ZN6Features = + !listconcat(ZN5Features, ZN6AdditionalFeatures); } //===----------------------------------------------------------------------===// @@ -1993,6 +2002,8 @@ def : ProcModel<"znver4", Znver4Model, ProcessorFeatures.ZN4Features, ProcessorFeatures.ZN4Tuning>; def : ProcModel<"znver5", Znver4Model, ProcessorFeatures.ZN5Features, ProcessorFeatures.ZN5Tuning>; +def : ProcModel<"znver6", Znver4Model, ProcessorFeatures.ZN6Features, + ProcessorFeatures.ZN6Tuning>; def : Proc<"geode", [FeatureX87, FeatureCX8, FeatureMMX, FeaturePRFCHW], [TuningSlowUAMem16, TuningInsertVZEROUPPER]>; diff --git a/llvm/lib/Target/X86/X86PfmCounters.td b/llvm/lib/Target/X86/X86PfmCounters.td index b299633446410..9d2a1ce01c273 100644 --- a/llvm/lib/Target/X86/X86PfmCounters.td +++ b/llvm/lib/Target/X86/X86PfmCounters.td @@ -387,3 +387,4 @@ def ZnVer4PfmCounters : ProcPfmCounters { } def : PfmCountersBinding<"znver4", ZnVer4PfmCounters>; def : PfmCountersBinding<"znver5", ZnVer4PfmCounters>; +def : PfmCountersBinding<"znver6", ZnVer4PfmCounters>; diff --git a/llvm/lib/Target/X86/X86ScheduleZnver4.td b/llvm/lib/Target/X86/X86ScheduleZnver4.td index ac4d31de8dbfe..bffcf35d6b65d 100644 --- a/llvm/lib/Target/X86/X86ScheduleZnver4.td +++ b/llvm/lib/Target/X86/X86ScheduleZnver4.td @@ -9,7 +9,7 @@ // This file defines the machine model for Znver4 to support instruction // scheduling and other instruction cost heuristics. // Based on: -// * AMD Software Optimization Guide for the AMD Family 19h (Zen4) +// * AMD Software Optimization Guide for the AMD Family 19h (Zen4) // Microarchitecture // https://www.amd.com/system/files/TechDocs/57647.zip //===----------------------------------------------------------------------===// @@ -1550,7 +1550,7 @@ def Zn4WriteVFIXUPIMMPDZrr_VRANGESDrr : SchedWriteRes<[Zn4FPFMisc01]> { let NumMicroOps = 1; } def : InstRW<[Zn4WriteVFIXUPIMMPDZrr_VRANGESDrr], (instregex - "VFIXUPIMM(S|P)(S|D)(Z|Z128|Z256?)rrik", "VFIXUPIMM(S|P)(S|D)(Z?|Z128?|Z256?)rrikz", + "VFIXUPIMM(S|P)(S|D)(Z|Z128|Z256?)rrik", "VFIXUPIMM(S|P)(S|D)(Z?|Z128?|Z256?)rrikz", "VFIXUPIMM(S|P)(S|D)(Z128|Z256?)rri", "VRANGE(S|P)(S|D)(Z?|Z128?|Z256?)rri(b?)", "VRANGE(S|P)(S|D)(Z|Z128|Z256?)rri(b?)k","VRANGE(S|P)(S|D)(Z?|Z128?|Z256?)rri(b?)kz" )>; @@ -1824,20 +1824,20 @@ def Zn4VecALUZSlow: SchedWriteRes<[Zn4FPFMisc01]> { let ReleaseAtCycles = [2]; let NumMicroOps = 1; } -def : InstRW<[Zn4VecALUZSlow], (instrs - VPABSBZ128rr, VPABSBZ128rrk, VPABSBZ128rrkz, VPABSDZ128rr, - VPABSDZ128rrk, VPABSDZ128rrkz, VPABSQZ128rr, VPABSQZ128rrk, - VPABSQZ128rrkz, VPABSWZ128rr, VPABSWZ128rrk, VPABSWZ128rrkz, - VPADDSBZ128rr, VPADDSBZ128rrk, VPADDSBZ128rrkz, VPADDSWZ128rr, - VPADDSWZ128rrk, VPADDSWZ128rrkz,VPADDUSBZ128rr, VPADDUSBZ128rrk, - VPADDUSBZ128rrkz, VPADDUSWZ128rr, VPADDUSWZ128rrk, VPADDUSWZ128rrkz, - VPAVGBZ128rr, VPAVGBZ128rrk, VPAVGBZ128rrkz, VPAVGWZ128rr, - VPAVGWZ128rrk, VPAVGWZ128rrkz, VPOPCNTBZ128rr, VPOPCNTBZ128rrk, - VPOPCNTBZ128rrkz, VPOPCNTDZ128rr, VPOPCNTDZ128rrk, VPOPCNTDZ128rrkz, - VPOPCNTQZ128rr, VPOPCNTQZ128rrk,VPOPCNTQZ128rrkz, VPOPCNTWZ128rr, - VPOPCNTWZ128rrk, VPOPCNTWZ128rrkz,VPSUBSBZ128rr, VPSUBSBZ128rrk, - VPSUBSBZ128rrkz, VPSUBSWZ128rr, VPSUBSWZ128rrk, VPSUBSWZ128rrkz, - VPSUBUSBZ128rr, VPSUBUSBZ128rrk, VPSUBUSBZ128rrkz,VPSUBUSWZ128rr, +def : InstRW<[Zn4VecALUZSlow], (instrs + VPABSBZ128rr, VPABSBZ128rrk, VPABSBZ128rrkz, VPABSDZ128rr, + VPABSDZ128rrk, VPABSDZ128rrkz, VPABSQZ128rr, VPABSQZ128rrk, + VPABSQZ128rrkz, VPABSWZ128rr, VPABSWZ128rrk, VPABSWZ128rrkz, + VPADDSBZ128rr, VPADDSBZ128rrk, VPADDSBZ128rrkz, VPADDSWZ128rr, + VPADDSWZ128rrk, VPADDSWZ128rrkz,VPADDUSBZ128rr, VPADDUSBZ128rrk, + VPADDUSBZ128rrkz, VPADDUSWZ128rr, VPADDUSWZ128rrk, VPADDUSWZ128rrkz, + VPAVGBZ128rr, VPAVGBZ128rrk, VPAVGBZ128rrkz, VPAVGWZ128rr, + VPAVGWZ128rrk, VPAVGWZ128rrkz, VPOPCNTBZ128rr, VPOPCNTBZ128rrk, + VPOPCNTBZ128rrkz, VPOPCNTDZ128rr, VPOPCNTDZ128rrk, VPOPCNTDZ128rrkz, + VPOPCNTQZ128rr, VPOPCNTQZ128rrk,VPOPCNTQZ128rrkz, VPOPCNTWZ128rr, + VPOPCNTWZ128rrk, VPOPCNTWZ128rrkz,VPSUBSBZ128rr, VPSUBSBZ128rrk, + VPSUBSBZ128rrkz, VPSUBSWZ128rr, VPSUBSWZ128rrk, VPSUBSWZ128rrkz, + VPSUBUSBZ128rr, VPSUBUSBZ128rrk, VPSUBUSBZ128rrkz,VPSUBUSWZ128rr, VPSUBUSWZ128rrk, VPSUBUSWZ128rrkz )>; diff --git a/llvm/lib/TargetParser/Host.cpp b/llvm/lib/TargetParser/Host.cpp index 5d4fa2c88153c..362c151ca5bad 100644 --- a/llvm/lib/TargetParser/Host.cpp +++ b/llvm/lib/TargetParser/Host.cpp @@ -1340,6 +1340,12 @@ static const char *getAMDProcessorTypeAndSubtype(unsigned Family, *Subtype = X86::AMDFAM1AH_ZNVER5; break; // "znver5" } + if ((Model >= 0x50 && Model <= 0x5f) || (Model >= 0x80 && Model <= 0xcf) || + (Model >= 0xd8 && Model <= 0xe7)) { + CPU = "znver6"; + *Subtype = X86::AMDFAM1AH_ZNVER6; + break; // "znver6" + } break; default: diff --git a/llvm/lib/TargetParser/X86TargetParser.cpp b/llvm/lib/TargetParser/X86TargetParser.cpp index 2810849e4af9e..f848b1ac08607 100644 --- a/llvm/lib/TargetParser/X86TargetParser.cpp +++ b/llvm/lib/TargetParser/X86TargetParser.cpp @@ -255,6 +255,10 @@ static constexpr FeatureBitset FeaturesZNVER5 = FeaturesZNVER4 | FeatureAVXVNNI | FeatureMOVDIRI | FeatureMOVDIR64B | FeatureAVX512VP2INTERSECT | FeaturePREFETCHI | FeatureAVXVNNI; +static constexpr FeatureBitset FeaturesZNVER6 = + FeaturesZNVER5 | FeatureAVXVNNIINT8 | FeatureAVX512FP16 | FeatureAVXIFMA | + FeatureAVXNECONVERT; + // D151696 tranplanted Mangling and OnlyForCPUDispatchSpecific from // X86TargetParser.def to here. They are assigned by following ways: // 1. Copy the mangling from the original CPU_SPEICIFC MACROs. If no, assign @@ -440,6 +444,7 @@ constexpr ProcInfo Processors[] = { { {"znver3"}, CK_ZNVER3, FEATURE_AVX2, FeaturesZNVER3, '\0', false }, { {"znver4"}, CK_ZNVER4, FEATURE_AVX512VBMI2, FeaturesZNVER4, '\0', false }, { {"znver5"}, CK_ZNVER5, FEATURE_AVX512VP2INTERSECT, FeaturesZNVER5, '\0', false }, + { {"znver6"}, CK_ZNVER6, FEATURE_AVX512FP16, FeaturesZNVER6, '\0', false }, // Generic 64-bit processor. { {"x86-64"}, CK_x86_64, FEATURE_SSE2 , FeaturesX86_64, '\0', false }, { {"x86-64-v2"}, CK_x86_64_v2, FEATURE_SSE4_2 , FeaturesX86_64_V2, '\0', false }, diff --git a/llvm/test/CodeGen/X86/bypass-slow-division-64.ll b/llvm/test/CodeGen/X86/bypass-slow-division-64.ll index b0ca0069a526b..821b7b8e4144f 100644 --- a/llvm/test/CodeGen/X86/bypass-slow-division-64.ll +++ b/llvm/test/CodeGen/X86/bypass-slow-division-64.ll @@ -24,6 +24,7 @@ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver3 | FileCheck %s --check-prefixes=CHECK,SLOW-DIVQ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,SLOW-DIVQ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,SLOW-DIVQ +; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,SLOW-DIVQ ; Additional tests for 64-bit divide bypass diff --git a/llvm/test/CodeGen/X86/cmp16.ll b/llvm/test/CodeGen/X86/cmp16.ll index 8c14a78d9e113..ff6ee68074088 100644 --- a/llvm/test/CodeGen/X86/cmp16.ll +++ b/llvm/test/CodeGen/X86/cmp16.ll @@ -14,6 +14,7 @@ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver3 | FileCheck %s --check-prefixes=X64,X64-FAST ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver4 | FileCheck %s --check-prefixes=X64,X64-FAST ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver5 | FileCheck %s --check-prefixes=X64,X64-FAST +; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver6 | FileCheck %s --check-prefixes=X64,X64-FAST define i1 @cmp16_reg_eq_reg(i16 %a0, i16 %a1) { ; X86-GENERIC-LABEL: cmp16_reg_eq_reg: diff --git a/llvm/test/CodeGen/X86/cpus-amd.ll b/llvm/test/CodeGen/X86/cpus-amd.ll index 33b2cf3731478..33cbc71b41ecd 100644 --- a/llvm/test/CodeGen/X86/cpus-amd.ll +++ b/llvm/test/CodeGen/X86/cpus-amd.ll @@ -30,6 +30,7 @@ ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=znver3 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=znver4 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=znver5 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=znver6 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty define void @foo() { ret void diff --git a/llvm/test/CodeGen/X86/rdpru.ll b/llvm/test/CodeGen/X86/rdpru.ll index be79a4499a338..067ae31142c39 100644 --- a/llvm/test/CodeGen/X86/rdpru.ll +++ b/llvm/test/CodeGen/X86/rdpru.ll @@ -7,6 +7,7 @@ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver3 -fast-isel | FileCheck %s --check-prefix=X64 ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver4 -fast-isel | FileCheck %s --check-prefix=X64 ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver5 -fast-isel | FileCheck %s --check-prefix=X64 +; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver6 -fast-isel | FileCheck %s --check-prefix=X64 define void @rdpru_asm() { ; X86-LABEL: rdpru_asm: diff --git a/llvm/test/CodeGen/X86/shuffle-as-shifts.ll b/llvm/test/CodeGen/X86/shuffle-as-shifts.ll index 4b8f78d36c3f5..021f8d6fb971d 100644 --- a/llvm/test/CodeGen/X86/shuffle-as-shifts.ll +++ b/llvm/test/CodeGen/X86/shuffle-as-shifts.ll @@ -4,6 +4,7 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64-v4 | FileCheck %s --check-prefixes=CHECK,CHECK-V4 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 define <4 x i32> @shuf_rot_v4i32_1032(<4 x i32> %x) { diff --git a/llvm/test/CodeGen/X86/slow-unaligned-mem.ll b/llvm/test/CodeGen/X86/slow-unaligned-mem.ll index ceef3fb4bb188..a215b60055dd5 100644 --- a/llvm/test/CodeGen/X86/slow-unaligned-mem.ll +++ b/llvm/test/CodeGen/X86/slow-unaligned-mem.ll @@ -51,6 +51,7 @@ ; RUN: llc < %s -mtriple=i386-unknown-unknown -mcpu=znver3 2>&1 | FileCheck %s --check-prefixes=FAST,FAST-AVX256 ; RUN: llc < %s -mtriple=i386-unknown-unknown -mcpu=znver4 2>&1 | FileCheck %s --check-prefixes=FAST,FAST-AVX512 ; RUN: llc < %s -mtriple=i386-unknown-unknown -mcpu=znver5 2>&1 | FileCheck %s --check-prefixes=FAST,FAST-AVX512 +; RUN: llc < %s -mtriple=i386-unknown-unknown -mcpu=znver6 2>&1 | FileCheck %s --check-prefixes=FAST,FAST-AVX512 ; Other chips with slow unaligned memory accesses diff --git a/llvm/test/CodeGen/X86/sqrt-fastmath-tune.ll b/llvm/test/CodeGen/X86/sqrt-fastmath-tune.ll index 74b51ac21dc1f..9d2708674c3ff 100644 --- a/llvm/test/CodeGen/X86/sqrt-fastmath-tune.ll +++ b/llvm/test/CodeGen/X86/sqrt-fastmath-tune.ll @@ -7,6 +7,7 @@ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver3 | FileCheck %s --check-prefixes=FAST-SCALAR,FAST-VECTOR ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver4 | FileCheck %s --check-prefixes=FAST-SCALAR,FAST-VECTOR ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver5 | FileCheck %s --check-prefixes=FAST-SCALAR,FAST-VECTOR +; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver6 | FileCheck %s --check-prefixes=FAST-SCALAR,FAST-VECTOR ; RUN: llc < %s -mtriple=x86_64-- -mcpu=x86-64 | FileCheck %s --check-prefixes=X86-64 define float @f32_no_daz(float %f) #0 { diff --git a/llvm/test/CodeGen/X86/tuning-shuffle-permilpd-avx512.ll b/llvm/test/CodeGen/X86/tuning-shuffle-permilpd-avx512.ll index 162ab71fc00d4..e2c8b6df6e744 100644 --- a/llvm/test/CodeGen/X86/tuning-shuffle-permilpd-avx512.ll +++ b/llvm/test/CodeGen/X86/tuning-shuffle-permilpd-avx512.ll @@ -5,6 +5,7 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl,+avx512bw,+avx512dq | FileCheck %s --check-prefixes=CHECK,CHECK-AVX512 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 define <8 x double> @transform_VPERMILPSZrr(<8 x double> %a) nounwind { ; CHECK-LABEL: transform_VPERMILPSZrr: diff --git a/llvm/test/CodeGen/X86/tuning-shuffle-permilps-avx512.ll b/llvm/test/CodeGen/X86/tuning-shuffle-permilps-avx512.ll index cd97946da248f..53bad74552f8a 100644 --- a/llvm/test/CodeGen/X86/tuning-shuffle-permilps-avx512.ll +++ b/llvm/test/CodeGen/X86/tuning-shuffle-permilps-avx512.ll @@ -5,6 +5,7 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl,+avx512bw,+avx512dq | FileCheck %s --check-prefixes=CHECK,CHECK-AVX512 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 define <16 x float> @transform_VPERMILPSZrr(<16 x float> %a) nounwind { ; CHECK-LABEL: transform_VPERMILPSZrr: diff --git a/llvm/test/CodeGen/X86/tuning-shuffle-unpckpd-avx512.ll b/llvm/test/CodeGen/X86/tuning-shuffle-unpckpd-avx512.ll index 5ea991f85523e..39a072eeeea4c 100644 --- a/llvm/test/CodeGen/X86/tuning-shuffle-unpckpd-avx512.ll +++ b/llvm/test/CodeGen/X86/tuning-shuffle-unpckpd-avx512.ll @@ -6,6 +6,7 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl,+avx512bw,+avx512dq | FileCheck %s --check-prefixes=CHECK,CHECK-AVX512 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 define <16 x float> @transform_VUNPCKLPDZrr(<16 x float> %a, <16 x float> %b) nounwind { diff --git a/llvm/test/CodeGen/X86/tuning-shuffle-unpckps-avx512.ll b/llvm/test/CodeGen/X86/tuning-shuffle-unpckps-avx512.ll index 96155f0300d2d..f8b9dac4c7ba8 100644 --- a/llvm/test/CodeGen/X86/tuning-shuffle-unpckps-avx512.ll +++ b/llvm/test/CodeGen/X86/tuning-shuffle-unpckps-avx512.ll @@ -6,6 +6,7 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl,+avx512bw,+avx512dq | FileCheck %s --check-prefixes=CHECK,CHECK-AVX512 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,CHECK-ZNVER4 define <16 x float> @transform_VUNPCKLPSZrr(<16 x float> %a, <16 x float> %b) nounwind { ; CHECK-LABEL: transform_VUNPCKLPSZrr: diff --git a/llvm/test/CodeGen/X86/vector-shuffle-fast-per-lane.ll b/llvm/test/CodeGen/X86/vector-shuffle-fast-per-lane.ll index 4021b1bf292bb..5bf936c6e5cec 100644 --- a/llvm/test/CodeGen/X86/vector-shuffle-fast-per-lane.ll +++ b/llvm/test/CodeGen/X86/vector-shuffle-fast-per-lane.ll @@ -9,6 +9,7 @@ ; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=znver3 | FileCheck %s --check-prefixes=FAST ; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=FAST ; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=FAST +; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=FAST ; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=haswell | FileCheck %s --check-prefixes=FAST ; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=skx | FileCheck %s --check-prefixes=FAST diff --git a/llvm/test/CodeGen/X86/vpdpwssd.ll b/llvm/test/CodeGen/X86/vpdpwssd.ll index 2ac2b48af4ce7..ea97800505bc2 100644 --- a/llvm/test/CodeGen/X86/vpdpwssd.ll +++ b/llvm/test/CodeGen/X86/vpdpwssd.ll @@ -1,6 +1,7 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver4 | FileCheck %s --check-prefixes=CHECK,AVX512VL-VNNI ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver5 | FileCheck %s --check-prefixes=CHECK,AVX-VNNI +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=znver6 | FileCheck %s --check-prefixes=CHECK,AVX-VNNI ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vnni,+fast-dpwssd | FileCheck %s --check-prefixes=CHECK,AVX512-VNNI ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vnni,+avx512vl,+fast-dpwssd | FileCheck %s --check-prefixes=CHECK,AVX512VL-VNNI diff --git a/llvm/test/CodeGen/X86/x86-64-double-shifts-var.ll b/llvm/test/CodeGen/X86/x86-64-double-shifts-var.ll index c5e879c0135f4..bb1a4e5fcb75b 100644 --- a/llvm/test/CodeGen/X86/x86-64-double-shifts-var.ll +++ b/llvm/test/CodeGen/X86/x86-64-double-shifts-var.ll @@ -18,6 +18,7 @@ ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver3 | FileCheck %s --check-prefixes=BMI2-FAST ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver4 | FileCheck %s --check-prefixes=BMI2-FAST ; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver5 | FileCheck %s --check-prefixes=BMI2-FAST +; RUN: llc < %s -mtriple=x86_64-- -mcpu=znver6 | FileCheck %s --check-prefixes=BMI2-FAST ; Verify that for the X86_64 processors that are known to have poor latency ; double precision shift instructions we do not generate 'shld' or 'shrd' diff --git a/llvm/test/MC/X86/x86_long_nop.s b/llvm/test/MC/X86/x86_long_nop.s index b79403bb5f1ec..2c5fe3acde26c 100644 --- a/llvm/test/MC/X86/x86_long_nop.s +++ b/llvm/test/MC/X86/x86_long_nop.s @@ -21,6 +21,8 @@ # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=znver4 | llvm-objdump -d --no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=znver5 %s | llvm-objdump -d --no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=znver5 | llvm-objdump -d --no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15 +# RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=znver6 %s | llvm-objdump -d --no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15 +# RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=znver6 | llvm-objdump -d --no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu -mcpu=nehalem %s | llvm-objdump -d --no-show-raw-insn - | FileCheck --check-prefix=LNOP10 %s # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu -mcpu=westmere %s | llvm-objdump -d --no-show-raw-insn - | FileCheck --check-prefix=LNOP10 %s # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu -mcpu=sandybridge %s | llvm-objdump -d --no-show-raw-insn - | FileCheck --check-prefix=LNOP15 %s diff --git a/llvm/test/Transforms/LoopUnroll/X86/call-remark.ll b/llvm/test/Transforms/LoopUnroll/X86/call-remark.ll index b0f4385b7913d..f9768141e5d9c 100644 --- a/llvm/test/Transforms/LoopUnroll/X86/call-remark.ll +++ b/llvm/test/Transforms/LoopUnroll/X86/call-remark.ll @@ -2,6 +2,7 @@ ; RUN: opt -passes=debugify,loop-unroll -mcpu=znver3 -pass-remarks=TTI -pass-remarks-analysis=TTI < %s -S 2>&1 | FileCheck --check-prefixes=ALL,TTI %s ; RUN: opt -passes=debugify,loop-unroll -mcpu=znver4 -pass-remarks=loop-unroll -pass-remarks-analysis=loop-unroll < %s -S 2>&1 | FileCheck --check-prefixes=ALL,UNROLL %s ; RUN: opt -passes=debugify,loop-unroll -mcpu=znver5 -pass-remarks=loop-unroll -pass-remarks-analysis=loop-unroll < %s -S 2>&1 | FileCheck --check-prefixes=ALL,UNROLL %s +; RUN: opt -passes=debugify,loop-unroll -mcpu=znver6 -pass-remarks=loop-unroll -pass-remarks-analysis=loop-unroll < %s -S 2>&1 | FileCheck --check-prefixes=ALL,UNROLL %s target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128" diff --git a/llvm/test/Transforms/SLPVectorizer/X86/pr63668.ll b/llvm/test/Transforms/SLPVectorizer/X86/pr63668.ll index 037e073de9d59..a1ee268392d0e 100644 --- a/llvm/test/Transforms/SLPVectorizer/X86/pr63668.ll +++ b/llvm/test/Transforms/SLPVectorizer/X86/pr63668.ll @@ -1,6 +1,7 @@ ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 3 ; RUN: opt -passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -mcpu=znver4 -S < %s | FileCheck %s ; RUN: opt -passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -mcpu=znver5 -S < %s | FileCheck %s +; RUN: opt -passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -mcpu=znver6 -S < %s | FileCheck %s define internal i32 @testfunc() { ; CHECK-LABEL: define internal i32 @testfunc _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
