diff --git a/clang/include/clang/Basic/Cuda.h b/clang/include/clang/Basic/Cuda.h --- a/clang/include/clang/Basic/Cuda.h +++ b/clang/include/clang/Basic/Cuda.h @@ -111,6 +111,8 @@ GFX1101, GFX1102, GFX1103, + GFX1150, + GFX1151, Generic, // A processor model named 'generic' if the target backend defines a // public one. LAST, diff --git a/clang/lib/Basic/Cuda.cpp b/clang/lib/Basic/Cuda.cpp --- a/clang/lib/Basic/Cuda.cpp +++ b/clang/lib/Basic/Cuda.cpp @@ -133,6 +133,8 @@ GFX(1101), // gfx1101 GFX(1102), // gfx1102 GFX(1103), // gfx1103 + GFX(1150), // gfx1150 + GFX(1151), // gfx1151 {CudaArch::Generic, "generic", ""}, // clang-format on }; diff --git a/clang/lib/Basic/Targets/NVPTX.cpp b/clang/lib/Basic/Targets/NVPTX.cpp --- a/clang/lib/Basic/Targets/NVPTX.cpp +++ b/clang/lib/Basic/Targets/NVPTX.cpp @@ -212,6 +212,8 @@ case CudaArch::GFX1101: case CudaArch::GFX1102: case CudaArch::GFX1103: + case CudaArch::GFX1150: + case CudaArch::GFX1151: case CudaArch::Generic: case CudaArch::LAST: break; diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp --- a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp +++ b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp @@ -3631,6 +3631,8 @@ case CudaArch::GFX1101: case CudaArch::GFX1102: case CudaArch::GFX1103: + case CudaArch::GFX1150: + case CudaArch::GFX1151: case CudaArch::Generic: case CudaArch::UNUSED: case CudaArch::UNKNOWN: diff --git a/clang/test/CodeGenOpenCL/amdgpu-features.cl b/clang/test/CodeGenOpenCL/amdgpu-features.cl --- a/clang/test/CodeGenOpenCL/amdgpu-features.cl +++ b/clang/test/CodeGenOpenCL/amdgpu-features.cl @@ -47,6 +47,8 @@ // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1101 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1101 %s // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1102 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1102 %s // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1103 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1103 %s +// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1150 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1150 %s +// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1151 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1151 %s // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1103 -target-feature +wavefrontsize64 -S -emit-llvm -o - %s | FileCheck --check-prefix=GFX1103-W64 %s @@ -94,6 +96,9 @@ // GFX1101: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32" // GFX1102: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32" // GFX1103: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32" +// GFX1150: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32" +// GFX1151: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize32" + // GFX1103-W64: "target-features"="+16-bit-insts,+atomic-fadd-rtn-insts,+ci-insts,+dl-insts,+dot10-insts,+dot5-insts,+dot7-insts,+dot8-insts,+dot9-insts,+dpp,+gfx10-3-insts,+gfx10-insts,+gfx11-insts,+gfx8-insts,+gfx9-insts,+wavefrontsize64" kernel void test() {} diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx11.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx11.cl --- a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx11.cl +++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx11.cl @@ -3,6 +3,8 @@ // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1101 -S -emit-llvm -o - %s | FileCheck %s // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1102 -S -emit-llvm -o - %s | FileCheck %s // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1103 -S -emit-llvm -o - %s | FileCheck %s +// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1150 -S -emit-llvm -o - %s | FileCheck %s +// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1151 -S -emit-llvm -o - %s | FileCheck %s typedef unsigned int uint; typedef unsigned long ulong; diff --git a/clang/test/Driver/amdgpu-macros.cl b/clang/test/Driver/amdgpu-macros.cl --- a/clang/test/Driver/amdgpu-macros.cl +++ b/clang/test/Driver/amdgpu-macros.cl @@ -126,6 +126,8 @@ // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1101 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1101 -DFAMILY=GFX11 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1102 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1102 -DFAMILY=GFX11 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1103 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1103 -DFAMILY=GFX11 +// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1150 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1150 -DFAMILY=GFX11 +// RUN: %clang -E -dM -target amdgcn -mcpu=gfx1151 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1151 -DFAMILY=GFX11 // ARCH-GCN-DAG: #define FP_FAST_FMA 1 diff --git a/clang/test/Driver/amdgpu-mcpu.cl b/clang/test/Driver/amdgpu-mcpu.cl --- a/clang/test/Driver/amdgpu-mcpu.cl +++ b/clang/test/Driver/amdgpu-mcpu.cl @@ -110,6 +110,8 @@ // RUN: %clang -### -target amdgcn -mcpu=gfx1101 %s 2>&1 | FileCheck --check-prefix=GFX1101 %s // RUN: %clang -### -target amdgcn -mcpu=gfx1102 %s 2>&1 | FileCheck --check-prefix=GFX1102 %s // RUN: %clang -### -target amdgcn -mcpu=gfx1103 %s 2>&1 | FileCheck --check-prefix=GFX1103 %s +// RUN: %clang -### -target amdgcn -mcpu=gfx1150 %s 2>&1 | FileCheck --check-prefix=GFX1150 %s +// RUN: %clang -### -target amdgcn -mcpu=gfx1151 %s 2>&1 | FileCheck --check-prefix=GFX1151 %s // GCNDEFAULT-NOT: -target-cpu // GFX600: "-target-cpu" "gfx600" @@ -152,3 +154,5 @@ // GFX1101: "-target-cpu" "gfx1101" // GFX1102: "-target-cpu" "gfx1102" // GFX1103: "-target-cpu" "gfx1103" +// GFX1150: "-target-cpu" "gfx1150" +// GFX1151: "-target-cpu" "gfx1151" diff --git a/clang/test/Misc/target-invalid-cpu-note.c b/clang/test/Misc/target-invalid-cpu-note.c --- a/clang/test/Misc/target-invalid-cpu-note.c +++ b/clang/test/Misc/target-invalid-cpu-note.c @@ -29,7 +29,7 @@ // 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-NEXT: note: valid target CPU values are: sm_20, sm_21, sm_30, sm_32, sm_35, sm_37, sm_50, sm_52, sm_53, sm_60, sm_61, sm_62, sm_70, sm_72, sm_75, sm_80, sm_86, sm_87, sm_89, sm_90, gfx600, gfx601, gfx602, gfx700, gfx701, gfx702, gfx703, gfx704, gfx705, gfx801, gfx802, gfx803, gfx805, gfx810, gfx900, gfx902, gfx904, gfx906, gfx908, gfx909, gfx90a, gfx90c, gfx940, gfx941, gfx942, gfx1010, gfx1011, gfx1012, gfx1013, gfx1030, gfx1031, gfx1032, gfx1033, gfx1034, gfx1035, gfx1036, gfx1100, gfx1101, gfx1102, gfx1103{{$}} +// NVPTX-NEXT: note: valid target CPU values are: sm_20, sm_21, sm_30, sm_32, sm_35, sm_37, sm_50, sm_52, sm_53, sm_60, sm_61, sm_62, sm_70, sm_72, sm_75, sm_80, sm_86, sm_87, sm_89, sm_90, gfx600, gfx601, gfx602, gfx700, gfx701, gfx702, gfx703, gfx704, gfx705, gfx801, gfx802, gfx803, gfx805, gfx810, gfx900, gfx902, gfx904, gfx906, gfx908, gfx909, gfx90a, gfx90c, gfx940, gfx941, gfx942, gfx1010, gfx1011, gfx1012, gfx1013, gfx1030, gfx1031, gfx1032, gfx1033, gfx1034, gfx1035, gfx1036, gfx1100, gfx1101, gfx1102, gfx1103, gfx1150, gfx1151{{$}} // RUN: not %clang_cc1 -triple r600--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix R600 // R600: error: unknown target CPU 'not-a-cpu' @@ -37,7 +37,7 @@ // RUN: not %clang_cc1 -triple amdgcn--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix AMDGCN // AMDGCN: error: unknown target CPU 'not-a-cpu' -// AMDGCN-NEXT: note: valid target CPU values are: gfx600, tahiti, gfx601, pitcairn, verde, gfx602, hainan, oland, gfx700, kaveri, gfx701, hawaii, gfx702, gfx703, kabini, mullins, gfx704, bonaire, gfx705, gfx801, carrizo, gfx802, iceland, tonga, gfx803, fiji, polaris10, polaris11, gfx805, tongapro, gfx810, stoney, gfx900, gfx902, gfx904, gfx906, gfx908, gfx909, gfx90a, gfx90c, gfx940, gfx941, gfx942, gfx1010, gfx1011, gfx1012, gfx1013, gfx1030, gfx1031, gfx1032, gfx1033, gfx1034, gfx1035, gfx1036, gfx1100, gfx1101, gfx1102, gfx1103{{$}} +// AMDGCN-NEXT: note: valid target CPU values are: gfx600, tahiti, gfx601, pitcairn, verde, gfx602, hainan, oland, gfx700, kaveri, gfx701, hawaii, gfx702, gfx703, kabini, mullins, gfx704, bonaire, gfx705, gfx801, carrizo, gfx802, iceland, tonga, gfx803, fiji, polaris10, polaris11, gfx805, tongapro, gfx810, stoney, gfx900, gfx902, gfx904, gfx906, gfx908, gfx909, gfx90a, gfx90c, gfx940, gfx941, gfx942, gfx1010, gfx1011, gfx1012, gfx1013, gfx1030, gfx1031, gfx1032, gfx1033, gfx1034, gfx1035, gfx1036, gfx1100, gfx1101, gfx1102, gfx1103, gfx1150, gfx1151{{$}} // RUN: not %clang_cc1 -triple wasm64--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix WEBASM // WEBASM: error: unknown target CPU 'not-a-cpu' diff --git a/flang/runtime/CMakeLists.txt b/flang/runtime/CMakeLists.txt --- a/flang/runtime/CMakeLists.txt +++ b/flang/runtime/CMakeLists.txt @@ -217,7 +217,7 @@ "gfx700;gfx701;gfx801;gfx803;gfx900;gfx902;gfx906" "gfx908;gfx90a;gfx90c;gfx940;gfx1010;gfx1030" "gfx1031;gfx1032;gfx1033;gfx1034;gfx1035;gfx1036" - "gfx1100;gfx1101;gfx1102;gfx1103" + "gfx1100;gfx1101;gfx1102;gfx1103;gfx1150;gfx1151" ) set(all_nvptx_architectures "sm_35;sm_37;sm_50;sm_52;sm_53;sm_60;sm_61;sm_62" diff --git a/libc/cmake/modules/prepare_libc_gpu_build.cmake b/libc/cmake/modules/prepare_libc_gpu_build.cmake --- a/libc/cmake/modules/prepare_libc_gpu_build.cmake +++ b/libc/cmake/modules/prepare_libc_gpu_build.cmake @@ -7,7 +7,7 @@ set(all_amdgpu_architectures "gfx700;gfx701;gfx801;gfx803;gfx900;gfx902;gfx906" "gfx908;gfx90a;gfx90c;gfx940;gfx1010;gfx1030" "gfx1031;gfx1032;gfx1033;gfx1034;gfx1035;gfx1036" - "gfx1100;gfx1101;gfx1102;gfx1103") + "gfx1100;gfx1101;gfx1102;gfx1103;gfx1150;gfx1151") set(all_nvptx_architectures "sm_35;sm_37;sm_50;sm_52;sm_53;sm_60;sm_61;sm_62" "sm_70;sm_72;sm_75;sm_80;sm_86") set(all_gpu_architectures diff --git a/libc/src/math/gpu/vendor/amdgpu/platform.h b/libc/src/math/gpu/vendor/amdgpu/platform.h --- a/libc/src/math/gpu/vendor/amdgpu/platform.h +++ b/libc/src/math/gpu/vendor/amdgpu/platform.h @@ -100,6 +100,10 @@ extern const uint32_t __oclc_ISA_version = 11002; #elif defined(__gfx1103__) extern const uint32_t __oclc_ISA_version = 11003; +#elif defined(__gfx1150__) +extern const uint32_t __oclc_ISA_version = 11500; +#elif defined(__gfx1151__) +extern const uint32_t __oclc_ISA_version = 11501; #else #error "Unknown AMDGPU architecture" #endif diff --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst --- a/llvm/docs/AMDGPUUsage.rst +++ b/llvm/docs/AMDGPUUsage.rst @@ -490,6 +490,20 @@ work-item Add product IDs names. + ``gfx1150`` ``amdgcn`` APU - cumode - Architected *TBA* + - wavefrontsize64 flat + scratch .. TODO:: + - Packed + work-item Add product + IDs names. + + ``gfx1151`` ``amdgcn`` APU - cumode - Architected *TBA* + - wavefrontsize64 flat + scratch .. TODO:: + - Packed + work-item Add product + IDs names. + =========== =============== ============ ===== ================= =============== =============== ====================== .. _amdgpu-target-features: @@ -1489,14 +1503,14 @@ ``EF_AMDGPU_MACH_AMDGCN_GFX940`` 0x040 ``gfx940`` ``EF_AMDGPU_MACH_AMDGCN_GFX1100`` 0x041 ``gfx1100`` ``EF_AMDGPU_MACH_AMDGCN_GFX1013`` 0x042 ``gfx1013`` - *reserved* 0x043 Reserved. + ``EF_AMDGPU_MACH_AMDGCN_GFX1150`` 0x043 ``gfx1150`` ``EF_AMDGPU_MACH_AMDGCN_GFX1103`` 0x044 ``gfx1103`` ``EF_AMDGPU_MACH_AMDGCN_GFX1036`` 0x045 ``gfx1036`` ``EF_AMDGPU_MACH_AMDGCN_GFX1101`` 0x046 ``gfx1101`` ``EF_AMDGPU_MACH_AMDGCN_GFX1102`` 0x047 ``gfx1102`` *reserved* 0x048 Reserved. *reserved* 0x049 Reserved. - *reserved* 0x04a Reserved. + ``EF_AMDGPU_MACH_AMDGCN_GFX1151`` 0x04a ``gfx1151`` ``EF_AMDGPU_MACH_AMDGCN_GFX941`` 0x04b ``gfx941`` ``EF_AMDGPU_MACH_AMDGCN_GFX942`` 0x04c ``gfx942`` ==================================== ========== ============================= diff --git a/llvm/include/llvm/BinaryFormat/ELF.h b/llvm/include/llvm/BinaryFormat/ELF.h --- a/llvm/include/llvm/BinaryFormat/ELF.h +++ b/llvm/include/llvm/BinaryFormat/ELF.h @@ -773,14 +773,14 @@ EF_AMDGPU_MACH_AMDGCN_GFX940 = 0x040, EF_AMDGPU_MACH_AMDGCN_GFX1100 = 0x041, EF_AMDGPU_MACH_AMDGCN_GFX1013 = 0x042, - EF_AMDGPU_MACH_AMDGCN_RESERVED_0X43 = 0x043, + EF_AMDGPU_MACH_AMDGCN_GFX1150 = 0x043, EF_AMDGPU_MACH_AMDGCN_GFX1103 = 0x044, EF_AMDGPU_MACH_AMDGCN_GFX1036 = 0x045, EF_AMDGPU_MACH_AMDGCN_GFX1101 = 0x046, EF_AMDGPU_MACH_AMDGCN_GFX1102 = 0x047, EF_AMDGPU_MACH_AMDGCN_RESERVED_0X48 = 0x048, EF_AMDGPU_MACH_AMDGCN_RESERVED_0X49 = 0x049, - EF_AMDGPU_MACH_AMDGCN_RESERVED_0X4A = 0x04a, + EF_AMDGPU_MACH_AMDGCN_GFX1151 = 0x04a, EF_AMDGPU_MACH_AMDGCN_GFX941 = 0x04b, EF_AMDGPU_MACH_AMDGCN_GFX942 = 0x04c, diff --git a/llvm/include/llvm/TargetParser/TargetParser.h b/llvm/include/llvm/TargetParser/TargetParser.h --- a/llvm/include/llvm/TargetParser/TargetParser.h +++ b/llvm/include/llvm/TargetParser/TargetParser.h @@ -103,9 +103,11 @@ GK_GFX1101 = 91, GK_GFX1102 = 92, GK_GFX1103 = 93, + GK_GFX1150 = 94, + GK_GFX1151 = 95, GK_AMDGCN_FIRST = GK_GFX600, - GK_AMDGCN_LAST = GK_GFX1103, + GK_AMDGCN_LAST = GK_GFX1151, }; /// Instruction set architecture version. diff --git a/llvm/lib/Object/ELFObjectFile.cpp b/llvm/lib/Object/ELFObjectFile.cpp --- a/llvm/lib/Object/ELFObjectFile.cpp +++ b/llvm/lib/Object/ELFObjectFile.cpp @@ -502,6 +502,10 @@ return "gfx1102"; case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1103: return "gfx1103"; + case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1150: + return "gfx1150"; + case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1151: + return "gfx1151"; default: llvm_unreachable("Unknown EF_AMDGPU_MACH value"); } diff --git a/llvm/lib/ObjectYAML/ELFYAML.cpp b/llvm/lib/ObjectYAML/ELFYAML.cpp --- a/llvm/lib/ObjectYAML/ELFYAML.cpp +++ b/llvm/lib/ObjectYAML/ELFYAML.cpp @@ -607,6 +607,8 @@ BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1101, EF_AMDGPU_MACH); BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1102, EF_AMDGPU_MACH); BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1103, EF_AMDGPU_MACH); + BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1150, EF_AMDGPU_MACH); + BCaseMask(EF_AMDGPU_MACH_AMDGCN_GFX1151, EF_AMDGPU_MACH); switch (Object->Header.ABIVersion) { default: // ELFOSABI_AMDGPU_PAL, ELFOSABI_AMDGPU_MESA3D support *_V3 flags. diff --git a/llvm/lib/Target/AMDGPU/AMDGPU.td b/llvm/lib/Target/AMDGPU/AMDGPU.td --- a/llvm/lib/Target/AMDGPU/AMDGPU.td +++ b/llvm/lib/Target/AMDGPU/AMDGPU.td @@ -1309,26 +1309,37 @@ FeatureImageInsts, FeaturePackedTID, FeatureVcmpxPermlaneHazard, - FeatureVALUTransUseHazard, FeatureMADIntraFwdBug]>; -def FeatureISAVersion11_0_0 : FeatureSet< +def FeatureISAVersion11_0_Common : FeatureSet< !listconcat(FeatureISAVersion11_Common.Features, + [FeatureVALUTransUseHazard])>; + +def FeatureISAVersion11_0_0 : FeatureSet< + !listconcat(FeatureISAVersion11_0_Common.Features, [FeatureGFX11FullVGPRs, FeatureUserSGPRInit16Bug])>; def FeatureISAVersion11_0_1 : FeatureSet< - !listconcat(FeatureISAVersion11_Common.Features, + !listconcat(FeatureISAVersion11_0_Common.Features, [FeatureGFX11FullVGPRs])>; def FeatureISAVersion11_0_2 : FeatureSet< - !listconcat(FeatureISAVersion11_Common.Features, + !listconcat(FeatureISAVersion11_0_Common.Features, [FeatureUserSGPRInit16Bug])>; def FeatureISAVersion11_0_3 : FeatureSet< + !listconcat(FeatureISAVersion11_0_Common.Features, + [])>; + +def FeatureISAVersion11_5_0 : FeatureSet< !listconcat(FeatureISAVersion11_Common.Features, [])>; +def FeatureISAVersion11_5_1 : FeatureSet< + !listconcat(FeatureISAVersion11_Common.Features, + [FeatureGFX11FullVGPRs])>; + //===----------------------------------------------------------------------===// def AMDGPUInstrInfo : InstrInfo { diff --git a/llvm/lib/Target/AMDGPU/GCNProcessors.td b/llvm/lib/Target/AMDGPU/GCNProcessors.td --- a/llvm/lib/Target/AMDGPU/GCNProcessors.td +++ b/llvm/lib/Target/AMDGPU/GCNProcessors.td @@ -271,3 +271,11 @@ def : ProcessorModel<"gfx1103", GFX11SpeedModel, FeatureISAVersion11_0_3.Features >; + +def : ProcessorModel<"gfx1150", GFX11SpeedModel, + FeatureISAVersion11_5_0.Features +>; + +def : ProcessorModel<"gfx1151", GFX11SpeedModel, + FeatureISAVersion11_5_1.Features +>; diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp --- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp +++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp @@ -124,6 +124,8 @@ case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1101: AK = GK_GFX1101; break; case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1102: AK = GK_GFX1102; break; case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1103: AK = GK_GFX1103; break; + case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1150: AK = GK_GFX1150; break; + case ELF::EF_AMDGPU_MACH_AMDGCN_GFX1151: AK = GK_GFX1151; break; case ELF::EF_AMDGPU_MACH_NONE: AK = GK_NONE; break; } @@ -195,6 +197,8 @@ case GK_GFX1101: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1101; case GK_GFX1102: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1102; case GK_GFX1103: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1103; + case GK_GFX1150: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1150; + case GK_GFX1151: return ELF::EF_AMDGPU_MACH_AMDGCN_GFX1151; case GK_NONE: return ELF::EF_AMDGPU_MACH_NONE; } diff --git a/llvm/lib/TargetParser/TargetParser.cpp b/llvm/lib/TargetParser/TargetParser.cpp --- a/llvm/lib/TargetParser/TargetParser.cpp +++ b/llvm/lib/TargetParser/TargetParser.cpp @@ -121,6 +121,8 @@ {{"gfx1101"}, {"gfx1101"}, GK_GFX1101, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP}, {{"gfx1102"}, {"gfx1102"}, GK_GFX1102, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP}, {{"gfx1103"}, {"gfx1103"}, GK_GFX1103, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP}, + {{"gfx1150"}, {"gfx1150"}, GK_GFX1150, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP}, + {{"gfx1151"}, {"gfx1151"}, GK_GFX1151, FEATURE_FAST_FMA_F32|FEATURE_FAST_DENORMAL_F32|FEATURE_WAVE32|FEATURE_WGP}, }; const GPUInfo *getArchEntry(AMDGPU::GPUKind AK, ArrayRef Table) { @@ -242,6 +244,8 @@ case GK_GFX1101: return {11, 0, 1}; case GK_GFX1102: return {11, 0, 2}; case GK_GFX1103: return {11, 0, 3}; + case GK_GFX1150: return {11, 5, 0}; + case GK_GFX1151: return {11, 5, 1}; default: return {0, 0, 0}; } } @@ -260,6 +264,8 @@ // XXX - What does the member GPU mean if device name string passed here? if (T.isAMDGCN()) { switch (parseArchAMDGCN(GPU)) { + case GK_GFX1151: + case GK_GFX1150: case GK_GFX1103: case GK_GFX1102: case GK_GFX1101: @@ -447,6 +453,8 @@ // XXX - What does the member GPU mean if device name string passed here? if (T.isAMDGCN()) { switch (parseArchAMDGCN(GPU)) { + case GK_GFX1151: + case GK_GFX1150: case GK_GFX1103: case GK_GFX1102: case GK_GFX1101: diff --git a/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target-v3.ll b/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target-v3.ll --- a/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target-v3.ll +++ b/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target-v3.ll @@ -97,6 +97,8 @@ ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1101 < %s | FileCheck --check-prefixes=V3-GFX1101 %s ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1102 < %s | FileCheck --check-prefixes=V3-GFX1102 %s ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1103 < %s | FileCheck --check-prefixes=V3-GFX1103 %s +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1150 < %s | FileCheck --check-prefixes=V3-GFX1150 %s +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1151 < %s | FileCheck --check-prefixes=V3-GFX1151 %s ; V3-GFX600: .amdgcn_target "amdgcn-amd-amdhsa--gfx600" ; V3-GFX601: .amdgcn_target "amdgcn-amd-amdhsa--gfx601" @@ -153,6 +155,8 @@ ; V3-GFX1101: .amdgcn_target "amdgcn-amd-amdhsa--gfx1101" ; V3-GFX1102: .amdgcn_target "amdgcn-amd-amdhsa--gfx1102" ; V3-GFX1103: .amdgcn_target "amdgcn-amd-amdhsa--gfx1103" +; V3-GFX1150: .amdgcn_target "amdgcn-amd-amdhsa--gfx1150" +; V3-GFX1151: .amdgcn_target "amdgcn-amd-amdhsa--gfx1151" diff --git a/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target.ll b/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target.ll --- a/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target.ll +++ b/llvm/test/CodeGen/AMDGPU/directive-amdgcn-target.ll @@ -103,6 +103,8 @@ ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1101 < %s | FileCheck --check-prefixes=GFX1101 %s ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1102 < %s | FileCheck --check-prefixes=GFX1102 %s ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1103 < %s | FileCheck --check-prefixes=GFX1103 %s +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1150 < %s | FileCheck --check-prefixes=GFX1150 %s +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1151 < %s | FileCheck --check-prefixes=GFX1151 %s ; GFX600: .amdgcn_target "amdgcn-amd-amdhsa--gfx600" ; GFX601: .amdgcn_target "amdgcn-amd-amdhsa--gfx601" @@ -187,6 +189,8 @@ ; GFX1101: .amdgcn_target "amdgcn-amd-amdhsa--gfx1101" ; GFX1102: .amdgcn_target "amdgcn-amd-amdhsa--gfx1102" ; GFX1103: .amdgcn_target "amdgcn-amd-amdhsa--gfx1103" +; GFX1150: .amdgcn_target "amdgcn-amd-amdhsa--gfx1150" +; GFX1151: .amdgcn_target "amdgcn-amd-amdhsa--gfx1151" define amdgpu_kernel void @directive_amdgcn_target() { ret void diff --git a/llvm/test/CodeGen/AMDGPU/elf-header-flags-mach.ll b/llvm/test/CodeGen/AMDGPU/elf-header-flags-mach.ll --- a/llvm/test/CodeGen/AMDGPU/elf-header-flags-mach.ll +++ b/llvm/test/CodeGen/AMDGPU/elf-header-flags-mach.ll @@ -72,6 +72,8 @@ ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1101 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1101 %s ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1102 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1102 %s ; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1103 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1103 %s +; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1150 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1150 %s +; RUN: llc -filetype=obj -march=amdgcn -mcpu=gfx1151 < %s | llvm-readobj --file-header - | FileCheck --check-prefixes=ALL,ARCH-GCN,GFX1151 %s ; FIXME: With the default attributes the eflags are not accurate for ; xnack and sramecc. Subsequent Target-ID patches will address this. @@ -141,6 +143,8 @@ ; GFX1101: EF_AMDGPU_MACH_AMDGCN_GFX1101 (0x46) ; GFX1102: EF_AMDGPU_MACH_AMDGCN_GFX1102 (0x47) ; GFX1103: EF_AMDGPU_MACH_AMDGCN_GFX1103 (0x44) +; GFX1150: EF_AMDGPU_MACH_AMDGCN_GFX1150 (0x43) +; GFX1151: EF_AMDGPU_MACH_AMDGCN_GFX1151 (0x4A) ; ALL: ] define amdgpu_kernel void @elf_header() { diff --git a/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll b/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll --- a/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll +++ b/llvm/test/CodeGen/AMDGPU/occupancy-levels.ll @@ -9,6 +9,10 @@ ; RUN: llc -march=amdgcn -mcpu=gfx1101 -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX1100,GFX1100W64 %s ; RUN: llc -march=amdgcn -mcpu=gfx1102 < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W32,GFX1030,GFX1030W32 %s ; RUN: llc -march=amdgcn -mcpu=gfx1102 -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W64,GFX1030,GFX1030W64 %s +; RUN: llc -march=amdgcn -mcpu=gfx1150 < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W32,GFX1030,GFX1030W32 %s +; RUN: llc -march=amdgcn -mcpu=gfx1150 -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX10,GFX10W64,GFX1030,GFX1030W64 %s +; RUN: llc -march=amdgcn -mcpu=gfx1151 < %s | FileCheck --check-prefixes=GCN,GFX1100,GFX1100W32 %s +; RUN: llc -march=amdgcn -mcpu=gfx1151 -mattr=+wavefrontsize64 < %s | FileCheck --check-prefixes=GCN,GFX1100,GFX1100W64 %s ; GCN-LABEL: {{^}}max_occupancy: ; GFX9: ; Occupancy: 10 diff --git a/llvm/test/Object/AMDGPU/elf-header-flags-mach.yaml b/llvm/test/Object/AMDGPU/elf-header-flags-mach.yaml --- a/llvm/test/Object/AMDGPU/elf-header-flags-mach.yaml +++ b/llvm/test/Object/AMDGPU/elf-header-flags-mach.yaml @@ -222,6 +222,14 @@ # RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1103 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1103 %s # RUN: obj2yaml %t.o.AMDGCN_GFX1103 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1103 %s +# RUN: sed -e 's//64/' -e 's//AMDGCN_GFX1150/' %s | yaml2obj -o %t.o.AMDGCN_GFX1150 +# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1150 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1150 %s +# RUN: obj2yaml %t.o.AMDGCN_GFX1150 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1150 %s + +# RUN: sed -e 's//64/' -e 's//AMDGCN_GFX1151/' %s | yaml2obj -o %t.o.AMDGCN_GFX1151 +# RUN: llvm-readobj -S --file-headers %t.o.AMDGCN_GFX1151 | FileCheck --check-prefixes=ELF-AMDGCN-ALL,ELF-AMDGCN-GFX1151 %s +# RUN: obj2yaml %t.o.AMDGCN_GFX1151 | FileCheck --check-prefixes=YAML-AMDGCN-ALL,YAML-AMDGCN-GFX1151 %s + # ELF-R600-ALL: Format: elf32-amdgpu # ELF-R600-ALL: Arch: r600 # ELF-R600-ALL: AddressSize: 32bit @@ -407,6 +415,12 @@ # ELF-AMDGCN-GFX1103: EF_AMDGPU_MACH_AMDGCN_GFX1103 (0x44) # YAML-AMDGCN-GFX1103: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1103 ] +# ELF-AMDGCN-GFX1150: EF_AMDGPU_MACH_AMDGCN_GFX1150 (0x43) +# YAML-AMDGCN-GFX1150: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1150 ] + +# ELF-AMDGCN-GFX1151: EF_AMDGPU_MACH_AMDGCN_GFX1151 (0x4A) +# YAML-AMDGCN-GFX1151: Flags: [ EF_AMDGPU_MACH_AMDGCN_GFX1151 ] + # ELF-AMDGCN-ALL: ] diff --git a/llvm/test/tools/llvm-objdump/ELF/AMDGPU/subtarget.ll b/llvm/test/tools/llvm-objdump/ELF/AMDGPU/subtarget.ll --- a/llvm/test/tools/llvm-objdump/ELF/AMDGPU/subtarget.ll +++ b/llvm/test/tools/llvm-objdump/ELF/AMDGPU/subtarget.ll @@ -6,6 +6,16 @@ ; ; ----------------------------------GFX11-------------------------------------- ; +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1151 -filetype=obj -O0 -o %t.o %s +; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1151 %t.o > %t-specify.txt +; RUN: llvm-objdump -D %t.o > %t-detect.txt +; RUN: diff %t-specify.txt %t-detect.txt + +; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1150 -filetype=obj -O0 -o %t.o %s +; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1150 %t.o > %t-specify.txt +; RUN: llvm-objdump -D %t.o > %t-detect.txt +; RUN: diff %t-specify.txt %t-detect.txt + ; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1103 -filetype=obj -O0 -o %t.o %s ; RUN: llvm-objdump -D --arch-name=amdgcn --mcpu=gfx1103 %t.o > %t-specify.txt ; RUN: llvm-objdump -D %t.o > %t-detect.txt diff --git a/llvm/test/tools/llvm-readobj/ELF/amdgpu-elf-headers.test b/llvm/test/tools/llvm-readobj/ELF/amdgpu-elf-headers.test --- a/llvm/test/tools/llvm-readobj/ELF/amdgpu-elf-headers.test +++ b/llvm/test/tools/llvm-readobj/ELF/amdgpu-elf-headers.test @@ -391,6 +391,24 @@ # RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1103 # RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=2 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1103 -DFLAG_VALUE=0x44 +# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1150 +# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=0 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1150 -DFLAG_VALUE=0x43 + +# RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1150 +# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=1 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1150 -DFLAG_VALUE=0x43 + +# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1150 +# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=2 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1150 -DFLAG_VALUE=0x43 + +# RUN: yaml2obj %s -o %t -DABI_VERSION=0 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1151 +# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=0 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1151 -DFLAG_VALUE=0x4A + +# RUN: yaml2obj %s -o %t -DABI_VERSION=1 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1151 +# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=1 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1151 -DFLAG_VALUE=0x4A + +# RUN: yaml2obj %s -o %t -DABI_VERSION=2 -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1151 +# RUN: llvm-readobj -h %t | FileCheck %s --check-prefixes=ALL,KNOWN-ABI-VERSION,SINGLE-FLAG --match-full-lines -DABI_VERSION=2 -DFILE=%t -DFLAG_NAME=EF_AMDGPU_MACH_AMDGCN_GFX1151 -DFLAG_VALUE=0x4A + --- !ELF FileHeader: Class: ELFCLASS64 diff --git a/llvm/tools/llvm-readobj/ELFDumper.cpp b/llvm/tools/llvm-readobj/ELFDumper.cpp --- a/llvm/tools/llvm-readobj/ELFDumper.cpp +++ b/llvm/tools/llvm-readobj/ELFDumper.cpp @@ -1605,6 +1605,8 @@ LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1101), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1102), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1103), + LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1150), + LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1151), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_V3), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_SRAMECC_V3) }; @@ -1667,6 +1669,8 @@ LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1101), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1102), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1103), + LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1150), + LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_MACH_AMDGCN_GFX1151), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_ANY_V4), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_OFF_V4), LLVM_READOBJ_ENUM_ENT(ELF, EF_AMDGPU_FEATURE_XNACK_ON_V4), diff --git a/openmp/libomptarget/DeviceRTL/CMakeLists.txt b/openmp/libomptarget/DeviceRTL/CMakeLists.txt --- a/openmp/libomptarget/DeviceRTL/CMakeLists.txt +++ b/openmp/libomptarget/DeviceRTL/CMakeLists.txt @@ -57,7 +57,7 @@ set(all_amdgpu_architectures "gfx700;gfx701;gfx801;gfx803;gfx900;gfx902;gfx906" "gfx908;gfx90a;gfx90c;gfx940;gfx1010;gfx1030" "gfx1031;gfx1032;gfx1033;gfx1034;gfx1035;gfx1036" - "gfx1100;gfx1101;gfx1102;gfx1103") + "gfx1100;gfx1101;gfx1102;gfx1103;gfx1150;gfx1151") set(all_nvptx_architectures "sm_35;sm_37;sm_50;sm_52;sm_53;sm_60;sm_61;sm_62" "sm_70;sm_72;sm_75;sm_80;sm_86;sm_87;sm_89;sm_90") set(all_gpu_architectures