| /*------------------------------------------------------------------------ |
| * Vulkan Conformance Tests |
| * ------------------------ |
| * |
| * Copyright (c) 2017 The Khronos Group Inc. |
| * Copyright (c) 2017 Codeplay Software Ltd. |
| * |
| * Licensed under the Apache License, Version 2.0 (the "License"); |
| * you may not use this file except in compliance with the License. |
| * You may obtain a copy of the License at |
| * |
| * http://www.apache.org/licenses/LICENSE-2.0 |
| * |
| * Unless required by applicable law or agreed to in writing, software |
| * distributed under the License is distributed on an "AS IS" BASIS, |
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| * See the License for the specific language governing permissions and |
| * limitations under the License. |
| * |
| */ /*! |
| * \file |
| * \brief Subgroups Tests |
| */ /*--------------------------------------------------------------------*/ |
| |
| #include "vktSubgroupsArithmeticTests.hpp" |
| #include "vktSubgroupsTestsUtils.hpp" |
| |
| #include <string> |
| #include <vector> |
| |
| using namespace tcu; |
| using namespace std; |
| using namespace vk; |
| using namespace vkt; |
| |
| namespace |
| { |
| enum OpType |
| { |
| OPTYPE_ADD = 0, |
| OPTYPE_MUL, |
| OPTYPE_MIN, |
| OPTYPE_MAX, |
| OPTYPE_AND, |
| OPTYPE_OR, |
| OPTYPE_XOR, |
| OPTYPE_INCLUSIVE_ADD, |
| OPTYPE_INCLUSIVE_MUL, |
| OPTYPE_INCLUSIVE_MIN, |
| OPTYPE_INCLUSIVE_MAX, |
| OPTYPE_INCLUSIVE_AND, |
| OPTYPE_INCLUSIVE_OR, |
| OPTYPE_INCLUSIVE_XOR, |
| OPTYPE_EXCLUSIVE_ADD, |
| OPTYPE_EXCLUSIVE_MUL, |
| OPTYPE_EXCLUSIVE_MIN, |
| OPTYPE_EXCLUSIVE_MAX, |
| OPTYPE_EXCLUSIVE_AND, |
| OPTYPE_EXCLUSIVE_OR, |
| OPTYPE_EXCLUSIVE_XOR, |
| OPTYPE_LAST |
| }; |
| |
| static bool checkVertexPipelineStages(std::vector<const void*> datas, |
| deUint32 width, deUint32) |
| { |
| const deUint32* data = |
| reinterpret_cast<const deUint32*>(datas[0]); |
| for (deUint32 x = 0; x < width; ++x) |
| { |
| deUint32 val = data[x]; |
| |
| if (0x3 != val) |
| { |
| return false; |
| } |
| } |
| |
| return true; |
| } |
| |
| static bool checkFragment(std::vector<const void*> datas, |
| deUint32 width, deUint32 height, deUint32) |
| { |
| const deUint32* data = |
| reinterpret_cast<const deUint32*>(datas[0]); |
| for (deUint32 x = 0; x < width; ++x) |
| { |
| for (deUint32 y = 0; y < height; ++y) |
| { |
| deUint32 val = data[x * height + y]; |
| |
| if (0x3 != val) |
| { |
| return false; |
| } |
| } |
| } |
| |
| return true; |
| } |
| |
| static bool checkCompute(std::vector<const void*> datas, |
| const deUint32 numWorkgroups[3], const deUint32 localSize[3], |
| deUint32) |
| { |
| const deUint32* data = |
| reinterpret_cast<const deUint32*>(datas[0]); |
| |
| for (deUint32 nX = 0; nX < numWorkgroups[0]; ++nX) |
| { |
| for (deUint32 nY = 0; nY < numWorkgroups[1]; ++nY) |
| { |
| for (deUint32 nZ = 0; nZ < numWorkgroups[2]; ++nZ) |
| { |
| for (deUint32 lX = 0; lX < localSize[0]; ++lX) |
| { |
| for (deUint32 lY = 0; lY < localSize[1]; ++lY) |
| { |
| for (deUint32 lZ = 0; lZ < localSize[2]; |
| ++lZ) |
| { |
| const deUint32 globalInvocationX = |
| nX * localSize[0] + lX; |
| const deUint32 globalInvocationY = |
| nY * localSize[1] + lY; |
| const deUint32 globalInvocationZ = |
| nZ * localSize[2] + lZ; |
| |
| const deUint32 globalSizeX = |
| numWorkgroups[0] * localSize[0]; |
| const deUint32 globalSizeY = |
| numWorkgroups[1] * localSize[1]; |
| |
| const deUint32 offset = |
| globalSizeX * |
| ((globalSizeY * |
| globalInvocationZ) + |
| globalInvocationY) + |
| globalInvocationX; |
| |
| if (0x3 != data[offset]) |
| { |
| return false; |
| } |
| } |
| } |
| } |
| } |
| } |
| } |
| |
| return true; |
| } |
| |
| std::string getOpTypeName(int opType) |
| { |
| switch (opType) |
| { |
| default: |
| DE_FATAL("Unsupported op type"); |
| case OPTYPE_ADD: |
| return "subgroupAdd"; |
| case OPTYPE_MUL: |
| return "subgroupMul"; |
| case OPTYPE_MIN: |
| return "subgroupMin"; |
| case OPTYPE_MAX: |
| return "subgroupMax"; |
| case OPTYPE_AND: |
| return "subgroupAnd"; |
| case OPTYPE_OR: |
| return "subgroupOr"; |
| case OPTYPE_XOR: |
| return "subgroupXor"; |
| case OPTYPE_INCLUSIVE_ADD: |
| return "subgroupInclusiveAdd"; |
| case OPTYPE_INCLUSIVE_MUL: |
| return "subgroupInclusiveMul"; |
| case OPTYPE_INCLUSIVE_MIN: |
| return "subgroupInclusiveMin"; |
| case OPTYPE_INCLUSIVE_MAX: |
| return "subgroupInclusiveMax"; |
| case OPTYPE_INCLUSIVE_AND: |
| return "subgroupInclusiveAnd"; |
| case OPTYPE_INCLUSIVE_OR: |
| return "subgroupInclusiveOr"; |
| case OPTYPE_INCLUSIVE_XOR: |
| return "subgroupInclusiveXor"; |
| case OPTYPE_EXCLUSIVE_ADD: |
| return "subgroupExclusiveAdd"; |
| case OPTYPE_EXCLUSIVE_MUL: |
| return "subgroupExclusiveMul"; |
| case OPTYPE_EXCLUSIVE_MIN: |
| return "subgroupExclusiveMin"; |
| case OPTYPE_EXCLUSIVE_MAX: |
| return "subgroupExclusiveMax"; |
| case OPTYPE_EXCLUSIVE_AND: |
| return "subgroupExclusiveAnd"; |
| case OPTYPE_EXCLUSIVE_OR: |
| return "subgroupExclusiveOr"; |
| case OPTYPE_EXCLUSIVE_XOR: |
| return "subgroupExclusiveXor"; |
| } |
| } |
| |
| std::string getOpTypeOperation(int opType, vk::VkFormat format, std::string lhs, std::string rhs) |
| { |
| switch (opType) |
| { |
| default: |
| DE_FATAL("Unsupported op type"); |
| case OPTYPE_ADD: |
| case OPTYPE_INCLUSIVE_ADD: |
| case OPTYPE_EXCLUSIVE_ADD: |
| return lhs + " + " + rhs; |
| case OPTYPE_MUL: |
| case OPTYPE_INCLUSIVE_MUL: |
| case OPTYPE_EXCLUSIVE_MUL: |
| return lhs + " * " + rhs; |
| case OPTYPE_MIN: |
| case OPTYPE_INCLUSIVE_MIN: |
| case OPTYPE_EXCLUSIVE_MIN: |
| switch (format) |
| { |
| default: |
| return "min(" + lhs + ", " + rhs + ")"; |
| case VK_FORMAT_R32_SFLOAT: |
| case VK_FORMAT_R64_SFLOAT: |
| return "(isnan(" + lhs + ") ? " + rhs + " : (isnan(" + rhs + ") ? " + lhs + " : min(" + lhs + ", " + rhs + ")))"; |
| case VK_FORMAT_R32G32_SFLOAT: |
| case VK_FORMAT_R32G32B32_SFLOAT: |
| case VK_FORMAT_R32G32B32A32_SFLOAT: |
| case VK_FORMAT_R64G64_SFLOAT: |
| case VK_FORMAT_R64G64B64_SFLOAT: |
| case VK_FORMAT_R64G64B64A64_SFLOAT: |
| return "mix(mix(min(" + lhs + ", " + rhs + "), " + lhs + ", isnan(" + rhs + ")), " + rhs + ", isnan(" + lhs + "))"; |
| } |
| case OPTYPE_MAX: |
| case OPTYPE_INCLUSIVE_MAX: |
| case OPTYPE_EXCLUSIVE_MAX: |
| switch (format) |
| { |
| default: |
| return "max(" + lhs + ", " + rhs + ")"; |
| case VK_FORMAT_R32_SFLOAT: |
| case VK_FORMAT_R64_SFLOAT: |
| return "(isnan(" + lhs + ") ? " + rhs + " : (isnan(" + rhs + ") ? " + lhs + " : max(" + lhs + ", " + rhs + ")))"; |
| case VK_FORMAT_R32G32_SFLOAT: |
| case VK_FORMAT_R32G32B32_SFLOAT: |
| case VK_FORMAT_R32G32B32A32_SFLOAT: |
| case VK_FORMAT_R64G64_SFLOAT: |
| case VK_FORMAT_R64G64B64_SFLOAT: |
| case VK_FORMAT_R64G64B64A64_SFLOAT: |
| return "mix(mix(max(" + lhs + ", " + rhs + "), " + lhs + ", isnan(" + rhs + ")), " + rhs + ", isnan(" + lhs + "))"; |
| } |
| case OPTYPE_AND: |
| case OPTYPE_INCLUSIVE_AND: |
| case OPTYPE_EXCLUSIVE_AND: |
| switch (format) |
| { |
| default: |
| return lhs + " & " + rhs; |
| case VK_FORMAT_R8_USCALED: |
| return lhs + " && " + rhs; |
| case VK_FORMAT_R8G8_USCALED: |
| return "bvec2(" + lhs + ".x && " + rhs + ".x, " + lhs + ".y && " + rhs + ".y)"; |
| case VK_FORMAT_R8G8B8_USCALED: |
| return "bvec3(" + lhs + ".x && " + rhs + ".x, " + lhs + ".y && " + rhs + ".y, " + lhs + ".z && " + rhs + ".z)"; |
| case VK_FORMAT_R8G8B8A8_USCALED: |
| return "bvec4(" + lhs + ".x && " + rhs + ".x, " + lhs + ".y && " + rhs + ".y, " + lhs + ".z && " + rhs + ".z, " + lhs + ".w && " + rhs + ".w)"; |
| } |
| case OPTYPE_OR: |
| case OPTYPE_INCLUSIVE_OR: |
| case OPTYPE_EXCLUSIVE_OR: |
| switch (format) |
| { |
| default: |
| return lhs + " | " + rhs; |
| case VK_FORMAT_R8_USCALED: |
| return lhs + " || " + rhs; |
| case VK_FORMAT_R8G8_USCALED: |
| return "bvec2(" + lhs + ".x || " + rhs + ".x, " + lhs + ".y || " + rhs + ".y)"; |
| case VK_FORMAT_R8G8B8_USCALED: |
| return "bvec3(" + lhs + ".x || " + rhs + ".x, " + lhs + ".y || " + rhs + ".y, " + lhs + ".z || " + rhs + ".z)"; |
| case VK_FORMAT_R8G8B8A8_USCALED: |
| return "bvec4(" + lhs + ".x || " + rhs + ".x, " + lhs + ".y || " + rhs + ".y, " + lhs + ".z || " + rhs + ".z, " + lhs + ".w || " + rhs + ".w)"; |
| } |
| case OPTYPE_XOR: |
| case OPTYPE_INCLUSIVE_XOR: |
| case OPTYPE_EXCLUSIVE_XOR: |
| switch (format) |
| { |
| default: |
| return lhs + " ^ " + rhs; |
| case VK_FORMAT_R8_USCALED: |
| return lhs + " ^^ " + rhs; |
| case VK_FORMAT_R8G8_USCALED: |
| return "bvec2(" + lhs + ".x ^^ " + rhs + ".x, " + lhs + ".y ^^ " + rhs + ".y)"; |
| case VK_FORMAT_R8G8B8_USCALED: |
| return "bvec3(" + lhs + ".x ^^ " + rhs + ".x, " + lhs + ".y ^^ " + rhs + ".y, " + lhs + ".z ^^ " + rhs + ".z)"; |
| case VK_FORMAT_R8G8B8A8_USCALED: |
| return "bvec4(" + lhs + ".x ^^ " + rhs + ".x, " + lhs + ".y ^^ " + rhs + ".y, " + lhs + ".z ^^ " + rhs + ".z, " + lhs + ".w ^^ " + rhs + ".w)"; |
| } |
| } |
| } |
| |
| std::string getIdentity(int opType, vk::VkFormat format) |
| { |
| bool isFloat = false; |
| bool isInt = false; |
| bool isUnsigned = false; |
| |
| switch (format) |
| { |
| default: |
| DE_FATAL("Unhandled format!"); |
| case VK_FORMAT_R32_SINT: |
| case VK_FORMAT_R32G32_SINT: |
| case VK_FORMAT_R32G32B32_SINT: |
| case VK_FORMAT_R32G32B32A32_SINT: |
| isInt = true; |
| break; |
| case VK_FORMAT_R32_UINT: |
| case VK_FORMAT_R32G32_UINT: |
| case VK_FORMAT_R32G32B32_UINT: |
| case VK_FORMAT_R32G32B32A32_UINT: |
| isUnsigned = true; |
| break; |
| case VK_FORMAT_R32_SFLOAT: |
| case VK_FORMAT_R32G32_SFLOAT: |
| case VK_FORMAT_R32G32B32_SFLOAT: |
| case VK_FORMAT_R32G32B32A32_SFLOAT: |
| case VK_FORMAT_R64_SFLOAT: |
| case VK_FORMAT_R64G64_SFLOAT: |
| case VK_FORMAT_R64G64B64_SFLOAT: |
| case VK_FORMAT_R64G64B64A64_SFLOAT: |
| isFloat = true; |
| break; |
| case VK_FORMAT_R8_USCALED: |
| case VK_FORMAT_R8G8_USCALED: |
| case VK_FORMAT_R8G8B8_USCALED: |
| case VK_FORMAT_R8G8B8A8_USCALED: |
| break; // bool types are not anything |
| } |
| |
| switch (opType) |
| { |
| default: |
| DE_FATAL("Unsupported op type"); |
| case OPTYPE_ADD: |
| case OPTYPE_INCLUSIVE_ADD: |
| case OPTYPE_EXCLUSIVE_ADD: |
| return subgroups::getFormatNameForGLSL(format) + "(0)"; |
| case OPTYPE_MUL: |
| case OPTYPE_INCLUSIVE_MUL: |
| case OPTYPE_EXCLUSIVE_MUL: |
| return subgroups::getFormatNameForGLSL(format) + "(1)"; |
| case OPTYPE_MIN: |
| case OPTYPE_INCLUSIVE_MIN: |
| case OPTYPE_EXCLUSIVE_MIN: |
| if (isFloat) |
| { |
| return subgroups::getFormatNameForGLSL(format) + "(intBitsToFloat(0x7f800000))"; |
| } |
| else if (isInt) |
| { |
| return subgroups::getFormatNameForGLSL(format) + "(0x7fffffff)"; |
| } |
| else if (isUnsigned) |
| { |
| return subgroups::getFormatNameForGLSL(format) + "(0xffffffffu)"; |
| } |
| else |
| { |
| DE_FATAL("Unhandled case"); |
| } |
| case OPTYPE_MAX: |
| case OPTYPE_INCLUSIVE_MAX: |
| case OPTYPE_EXCLUSIVE_MAX: |
| if (isFloat) |
| { |
| return subgroups::getFormatNameForGLSL(format) + "(intBitsToFloat(0xff800000))"; |
| } |
| else if (isInt) |
| { |
| return subgroups::getFormatNameForGLSL(format) + "(0x80000000)"; |
| } |
| else if (isUnsigned) |
| { |
| return subgroups::getFormatNameForGLSL(format) + "(0)"; |
| } |
| else |
| { |
| DE_FATAL("Unhandled case"); |
| } |
| case OPTYPE_AND: |
| case OPTYPE_INCLUSIVE_AND: |
| case OPTYPE_EXCLUSIVE_AND: |
| return subgroups::getFormatNameForGLSL(format) + "(~0)"; |
| case OPTYPE_OR: |
| case OPTYPE_INCLUSIVE_OR: |
| case OPTYPE_EXCLUSIVE_OR: |
| return subgroups::getFormatNameForGLSL(format) + "(0)"; |
| case OPTYPE_XOR: |
| case OPTYPE_INCLUSIVE_XOR: |
| case OPTYPE_EXCLUSIVE_XOR: |
| return subgroups::getFormatNameForGLSL(format) + "(0)"; |
| } |
| } |
| |
| std::string getCompare(int opType, vk::VkFormat format, std::string lhs, std::string rhs) |
| { |
| std::string formatName = subgroups::getFormatNameForGLSL(format); |
| switch (format) |
| { |
| default: |
| return "all(equal(" + lhs + ", " + rhs + "))"; |
| case VK_FORMAT_R8_USCALED: |
| case VK_FORMAT_R32_UINT: |
| case VK_FORMAT_R32_SINT: |
| return "(" + lhs + " == " + rhs + ")"; |
| case VK_FORMAT_R32_SFLOAT: |
| case VK_FORMAT_R64_SFLOAT: |
| switch (opType) |
| { |
| default: |
| return "(abs(" + lhs + " - " + rhs + ") < 0.00001)"; |
| case OPTYPE_MIN: |
| case OPTYPE_INCLUSIVE_MIN: |
| case OPTYPE_EXCLUSIVE_MIN: |
| case OPTYPE_MAX: |
| case OPTYPE_INCLUSIVE_MAX: |
| case OPTYPE_EXCLUSIVE_MAX: |
| return "(" + lhs + " == " + rhs + ")"; |
| } |
| case VK_FORMAT_R32G32_SFLOAT: |
| case VK_FORMAT_R32G32B32_SFLOAT: |
| case VK_FORMAT_R32G32B32A32_SFLOAT: |
| case VK_FORMAT_R64G64_SFLOAT: |
| case VK_FORMAT_R64G64B64_SFLOAT: |
| case VK_FORMAT_R64G64B64A64_SFLOAT: |
| switch (opType) |
| { |
| default: |
| return "all(lessThan(abs(" + lhs + " - " + rhs + "), " + formatName + "(0.00001)))"; |
| case OPTYPE_MIN: |
| case OPTYPE_INCLUSIVE_MIN: |
| case OPTYPE_EXCLUSIVE_MIN: |
| case OPTYPE_MAX: |
| case OPTYPE_INCLUSIVE_MAX: |
| case OPTYPE_EXCLUSIVE_MAX: |
| return "all(equal(" + lhs + ", " + rhs + "))"; |
| } |
| } |
| } |
| |
| struct CaseDefinition |
| { |
| int opType; |
| VkShaderStageFlags shaderStage; |
| VkFormat format; |
| bool noSSBO; |
| }; |
| |
| void initFrameBufferPrograms (SourceCollections& programCollection, CaseDefinition caseDef) |
| { |
| std::string indexVars; |
| switch (caseDef.opType) |
| { |
| default: |
| indexVars = " uint start = 0, end = gl_SubgroupSize;\n"; |
| break; |
| case OPTYPE_INCLUSIVE_ADD: |
| case OPTYPE_INCLUSIVE_MUL: |
| case OPTYPE_INCLUSIVE_MIN: |
| case OPTYPE_INCLUSIVE_MAX: |
| case OPTYPE_INCLUSIVE_AND: |
| case OPTYPE_INCLUSIVE_OR: |
| case OPTYPE_INCLUSIVE_XOR: |
| indexVars = " uint start = 0, end = gl_SubgroupInvocationID + 1;\n"; |
| break; |
| case OPTYPE_EXCLUSIVE_ADD: |
| case OPTYPE_EXCLUSIVE_MUL: |
| case OPTYPE_EXCLUSIVE_MIN: |
| case OPTYPE_EXCLUSIVE_MAX: |
| case OPTYPE_EXCLUSIVE_AND: |
| case OPTYPE_EXCLUSIVE_OR: |
| case OPTYPE_EXCLUSIVE_XOR: |
| indexVars = " uint start = 0, end = gl_SubgroupInvocationID;\n"; |
| break; |
| } |
| |
| std::ostringstream bdy; |
| |
| bdy << indexVars |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " ref = " |
| << getIdentity(caseDef.opType, caseDef.format) << ";\n" |
| << " uint tempResult = 0;\n" |
| << " for (uint index = start; index < end; index++)\n" |
| << " {\n" |
| << " if (subgroupBallotBitExtract(mask, index))\n" |
| << " {\n" |
| << " ref = " << getOpTypeOperation(caseDef.opType, caseDef.format, "ref", "data[index]") << ";\n" |
| << " }\n" |
| << " }\n" |
| << " tempResult = " << getCompare(caseDef.opType, caseDef.format, "ref", |
| getOpTypeName(caseDef.opType) + "(data[gl_SubgroupInvocationID])") << " ? 0x1 : 0;\n" |
| << " if (1 == (gl_SubgroupInvocationID % 2))\n" |
| << " {\n" |
| << " mask = subgroupBallot(true);\n" |
| << " ref = " << getIdentity(caseDef.opType, caseDef.format) << ";\n" |
| << " for (uint index = start; index < end; index++)\n" |
| << " {\n" |
| << " if (subgroupBallotBitExtract(mask, index))\n" |
| << " {\n" |
| << " ref = " << getOpTypeOperation(caseDef.opType, caseDef.format, "ref", "data[index]") << ";\n" |
| << " }\n" |
| << " }\n" |
| << " tempResult |= " << getCompare(caseDef.opType, caseDef.format, "ref", |
| getOpTypeName(caseDef.opType) + "(data[gl_SubgroupInvocationID])") << " ? 0x2 : 0;\n" |
| << " }\n" |
| << " else\n" |
| << " {\n" |
| << " tempResult |= 0x2;\n" |
| << " }\n"; |
| if (VK_SHADER_STAGE_VERTEX_BIT == caseDef.shaderStage) |
| { |
| std::ostringstream src; |
| std::ostringstream fragmentSrc; |
| |
| src << glu::getGLSLVersionDeclaration(glu::GLSL_VERSION_450)<<"\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout(location = 0) in highp vec4 in_position;\n" |
| << "layout(location = 0) out float out_color;\n" |
| << "layout(set = 0, binding = 0) uniform Buffer1\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[" << subgroups::maxSupportedSubgroupSize() << "];\n" |
| << "};\n" |
| << "\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " out_color = float(tempResult);\n" |
| << " gl_Position = in_position;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("vert") << glu::VertexSource(src.str()); |
| |
| fragmentSrc << glu::getGLSLVersionDeclaration(glu::GLSL_VERSION_450)<<"\n" |
| << "layout(location = 0) in float in_color;\n" |
| << "layout(location = 0) out uint out_color;\n" |
| << "void main()\n" |
| <<"{\n" |
| << " out_color = uint(in_color);\n" |
| << "}\n"; |
| programCollection.glslSources.add("fragment") << glu::FragmentSource(fragmentSrc.str()); |
| } |
| else |
| { |
| DE_FATAL("Unsupported shader stage"); |
| } |
| } |
| |
| void initPrograms(SourceCollections& programCollection, CaseDefinition caseDef) |
| { |
| std::string indexVars; |
| switch (caseDef.opType) |
| { |
| default: |
| indexVars = " uint start = 0, end = gl_SubgroupSize;\n"; |
| break; |
| case OPTYPE_INCLUSIVE_ADD: |
| case OPTYPE_INCLUSIVE_MUL: |
| case OPTYPE_INCLUSIVE_MIN: |
| case OPTYPE_INCLUSIVE_MAX: |
| case OPTYPE_INCLUSIVE_AND: |
| case OPTYPE_INCLUSIVE_OR: |
| case OPTYPE_INCLUSIVE_XOR: |
| indexVars = " uint start = 0, end = gl_SubgroupInvocationID + 1;\n"; |
| break; |
| case OPTYPE_EXCLUSIVE_ADD: |
| case OPTYPE_EXCLUSIVE_MUL: |
| case OPTYPE_EXCLUSIVE_MIN: |
| case OPTYPE_EXCLUSIVE_MAX: |
| case OPTYPE_EXCLUSIVE_AND: |
| case OPTYPE_EXCLUSIVE_OR: |
| case OPTYPE_EXCLUSIVE_XOR: |
| indexVars = " uint start = 0, end = gl_SubgroupInvocationID;\n"; |
| break; |
| } |
| |
| std::ostringstream bdy; |
| |
| bdy << indexVars |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " ref = " |
| << getIdentity(caseDef.opType, caseDef.format) << ";\n" |
| << " uint tempResult = 0;\n" |
| << " for (uint index = start; index < end; index++)\n" |
| << " {\n" |
| << " if (subgroupBallotBitExtract(mask, index))\n" |
| << " {\n" |
| << " ref = " << getOpTypeOperation(caseDef.opType, caseDef.format, "ref", "data[index]") << ";\n" |
| << " }\n" |
| << " }\n" |
| << " tempResult = " << getCompare(caseDef.opType, caseDef.format, "ref", |
| getOpTypeName(caseDef.opType) + "(data[gl_SubgroupInvocationID])") << " ? 0x1 : 0;\n" |
| << " if (1 == (gl_SubgroupInvocationID % 2))\n" |
| << " {\n" |
| << " mask = subgroupBallot(true);\n" |
| << " ref = " << getIdentity(caseDef.opType, caseDef.format) << ";\n" |
| << " for (uint index = start; index < end; index++)\n" |
| << " {\n" |
| << " if (subgroupBallotBitExtract(mask, index))\n" |
| << " {\n" |
| << " ref = " << getOpTypeOperation(caseDef.opType, caseDef.format, "ref", "data[index]") << ";\n" |
| << " }\n" |
| << " }\n" |
| << " tempResult |= " << getCompare(caseDef.opType, caseDef.format, "ref", |
| getOpTypeName(caseDef.opType) + "(data[gl_SubgroupInvocationID])") << " ? 0x2 : 0;\n" |
| << " }\n" |
| << " else\n" |
| << " {\n" |
| << " tempResult |= 0x2;\n" |
| << " }\n"; |
| |
| if (VK_SHADER_STAGE_COMPUTE_BIT == caseDef.shaderStage) |
| { |
| std::ostringstream src; |
| |
| src << "#version 450\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout (local_size_x_id = 0, local_size_y_id = 1, " |
| "local_size_z_id = 2) in;\n" |
| << "layout(set = 0, binding = 0, std430) buffer Buffer1\n" |
| << "{\n" |
| << " uint result[];\n" |
| << "};\n" |
| << "layout(set = 0, binding = 1, std430) buffer Buffer2\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[];\n" |
| << "};\n" |
| << "\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec3 globalSize = gl_NumWorkGroups * gl_WorkGroupSize;\n" |
| << " highp uint offset = globalSize.x * ((globalSize.y * " |
| "gl_GlobalInvocationID.z) + gl_GlobalInvocationID.y) + " |
| "gl_GlobalInvocationID.x;\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " result[offset] = tempResult;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("comp") |
| << glu::ComputeSource(src.str()) << vk::ShaderBuildOptions(vk::SPIRV_VERSION_1_3, 0u);; |
| } |
| else if (VK_SHADER_STAGE_FRAGMENT_BIT == caseDef.shaderStage) |
| { |
| programCollection.glslSources.add("vert") |
| << glu::VertexSource(subgroups::getVertShaderForStage(caseDef.shaderStage)); |
| |
| std::ostringstream frag; |
| |
| frag << "#version 450\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout(location = 0) out uint result;\n" |
| << "layout(set = 0, binding = 0, std430) readonly buffer Buffer2\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[];\n" |
| << "};\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " result = tempResult;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("frag") |
| << glu::FragmentSource(frag.str())<< vk::ShaderBuildOptions(vk::SPIRV_VERSION_1_3, 0u); |
| } |
| else if (VK_SHADER_STAGE_VERTEX_BIT == caseDef.shaderStage) |
| { |
| std::ostringstream src; |
| |
| src << "#version 450\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout(set = 0, binding = 0, std430) buffer Buffer1\n" |
| << "{\n" |
| << " uint result[];\n" |
| << "};\n" |
| << "layout(set = 0, binding = 1, std430) buffer Buffer2\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[];\n" |
| << "};\n" |
| << "\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " result[gl_VertexIndex] = tempResult;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("vert") |
| << glu::VertexSource(src.str()) << vk::ShaderBuildOptions(vk::SPIRV_VERSION_1_3, 0u); |
| } |
| else if (VK_SHADER_STAGE_GEOMETRY_BIT == caseDef.shaderStage) |
| { |
| programCollection.glslSources.add("vert") |
| << glu::VertexSource(subgroups::getVertShaderForStage(caseDef.shaderStage)); |
| |
| std::ostringstream src; |
| |
| src << "#version 450\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout(points) in;\n" |
| << "layout(points, max_vertices = 1) out;\n" |
| << "layout(set = 0, binding = 0, std430) buffer Buffer1\n" |
| << "{\n" |
| << " uint result[];\n" |
| << "};\n" |
| << "layout(set = 0, binding = 1, std430) buffer Buffer2\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[];\n" |
| << "};\n" |
| << "\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " result[gl_PrimitiveIDIn] = tempResult;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("geom") |
| << glu::GeometrySource(src.str()) << vk::ShaderBuildOptions(vk::SPIRV_VERSION_1_3, 0u); |
| } |
| else if (VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT == caseDef.shaderStage) |
| { |
| programCollection.glslSources.add("vert") |
| << glu::VertexSource(subgroups::getVertShaderForStage(caseDef.shaderStage)); |
| |
| programCollection.glslSources.add("tese") |
| << glu::TessellationEvaluationSource("#version 450\nlayout(isolines) in;\nvoid main (void) {}\n"); |
| |
| std::ostringstream src; |
| |
| src << "#version 450\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout(vertices=1) out;\n" |
| << "layout(set = 0, binding = 0, std430) buffer Buffer1\n" |
| << "{\n" |
| << " uint result[];\n" |
| << "};\n" |
| << "layout(set = 0, binding = 1, std430) buffer Buffer2\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[];\n" |
| << "};\n" |
| << "\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " result[gl_PrimitiveID] = tempResult;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("tesc") |
| << glu::TessellationControlSource(src.str()) << vk::ShaderBuildOptions(vk::SPIRV_VERSION_1_3, 0u); |
| } |
| else if (VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT == caseDef.shaderStage) |
| { |
| programCollection.glslSources.add("vert") |
| << glu::VertexSource(subgroups::getVertShaderForStage(caseDef.shaderStage)); |
| |
| programCollection.glslSources.add("tesc") |
| << glu::TessellationControlSource("#version 450\nlayout(vertices=1) out;\nvoid main (void) { for(uint i = 0; i < 4; i++) { gl_TessLevelOuter[i] = 1.0f; } }\n"); |
| |
| std::ostringstream src; |
| |
| src << "#version 450\n" |
| << "#extension GL_KHR_shader_subgroup_arithmetic: enable\n" |
| << "#extension GL_KHR_shader_subgroup_ballot: enable\n" |
| << "layout(isolines) in;\n" |
| << "layout(set = 0, binding = 0, std430) buffer Buffer1\n" |
| << "{\n" |
| << " uint result[];\n" |
| << "};\n" |
| << "layout(set = 0, binding = 1, std430) buffer Buffer2\n" |
| << "{\n" |
| << " " << subgroups::getFormatNameForGLSL(caseDef.format) << " data[];\n" |
| << "};\n" |
| << "\n" |
| << "void main (void)\n" |
| << "{\n" |
| << " uvec4 mask = subgroupBallot(true);\n" |
| << bdy.str() |
| << " result[gl_PrimitiveID * 2 + uint(gl_TessCoord.x + 0.5)] = tempResult;\n" |
| << "}\n"; |
| |
| programCollection.glslSources.add("tese") |
| << glu::TessellationEvaluationSource(src.str()) << vk::ShaderBuildOptions(vk::SPIRV_VERSION_1_3, 0u); |
| } |
| else |
| { |
| DE_FATAL("Unsupported shader stage"); |
| } |
| } |
| |
| tcu::TestStatus test(Context& context, const CaseDefinition caseDef) |
| { |
| if (!subgroups::isSubgroupSupported(context)) |
| TCU_THROW(NotSupportedError, "Subgroup operations are not supported"); |
| |
| if (!subgroups::areSubgroupOperationsSupportedForStage( |
| context, caseDef.shaderStage)) |
| { |
| if (subgroups::areSubgroupOperationsRequiredForStage( |
| caseDef.shaderStage)) |
| { |
| return tcu::TestStatus::fail( |
| "Shader stage " + |
| subgroups::getShaderStageName(caseDef.shaderStage) + |
| " is required to support subgroup operations!"); |
| } |
| else |
| { |
| TCU_THROW(NotSupportedError, "Device does not support subgroup operations for this stage"); |
| } |
| } |
| |
| if (!subgroups::isSubgroupFeatureSupportedForDevice(context, VK_SUBGROUP_FEATURE_ARITHMETIC_BIT)) |
| { |
| TCU_THROW(NotSupportedError, "Device does not support subgroup arithmetic operations"); |
| } |
| |
| if (subgroups::isDoubleFormat(caseDef.format) && |
| !subgroups::isDoubleSupportedForDevice(context)) |
| { |
| TCU_THROW(NotSupportedError, "Device does not support subgroup double operations"); |
| } |
| |
| //Tests which don't use the SSBO |
| if (caseDef.noSSBO && VK_SHADER_STAGE_VERTEX_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeVertexFrameBufferTest(context, VK_FORMAT_R32_UINT, &inputData, 1, checkVertexPipelineStages); |
| } |
| |
| if ((VK_SHADER_STAGE_FRAGMENT_BIT != caseDef.shaderStage) && |
| (VK_SHADER_STAGE_COMPUTE_BIT != caseDef.shaderStage)) |
| { |
| if (!subgroups::isVertexSSBOSupportedForDevice(context)) |
| { |
| TCU_THROW(NotSupportedError, "Device does not support vertex stage SSBO writes"); |
| } |
| } |
| |
| if (VK_SHADER_STAGE_FRAGMENT_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeFragmentTest(context, VK_FORMAT_R32_UINT, |
| &inputData, 1, checkFragment); |
| } |
| else if (VK_SHADER_STAGE_COMPUTE_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeComputeTest(context, VK_FORMAT_R32_UINT, &inputData, |
| 1, checkCompute); |
| } |
| else if (VK_SHADER_STAGE_VERTEX_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeVertexTest(context, VK_FORMAT_R32_UINT, &inputData, |
| 1, checkVertexPipelineStages); |
| } |
| else if (VK_SHADER_STAGE_GEOMETRY_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeGeometryTest(context, VK_FORMAT_R32_UINT, &inputData, |
| 1, checkVertexPipelineStages); |
| } |
| else if (VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeTessellationControlTest(context, VK_FORMAT_R32_UINT, &inputData, |
| 1, checkVertexPipelineStages); |
| } |
| else if (VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT == caseDef.shaderStage) |
| { |
| subgroups::SSBOData inputData; |
| inputData.format = caseDef.format; |
| inputData.numElements = subgroups::maxSupportedSubgroupSize(); |
| inputData.initializeType = subgroups::SSBOData::InitializeNonZero; |
| |
| return subgroups::makeTessellationEvaluationTest(context, VK_FORMAT_R32_UINT, &inputData, |
| 1, checkVertexPipelineStages); |
| } |
| else |
| { |
| TCU_THROW(InternalError, "Unhandled shader stage"); |
| } |
| } |
| } |
| |
| namespace vkt |
| { |
| namespace subgroups |
| { |
| tcu::TestCaseGroup* createSubgroupsArithmeticTests(tcu::TestContext& testCtx) |
| { |
| de::MovePtr<tcu::TestCaseGroup> group(new tcu::TestCaseGroup( |
| testCtx, "arithmetic", "Subgroup arithmetic category tests")); |
| |
| const VkShaderStageFlags stages[] = |
| { |
| VK_SHADER_STAGE_TESSELLATION_EVALUATION_BIT, |
| VK_SHADER_STAGE_TESSELLATION_CONTROL_BIT, |
| VK_SHADER_STAGE_GEOMETRY_BIT, |
| VK_SHADER_STAGE_VERTEX_BIT, |
| VK_SHADER_STAGE_FRAGMENT_BIT, |
| VK_SHADER_STAGE_COMPUTE_BIT |
| }; |
| |
| const VkFormat formats[] = |
| { |
| VK_FORMAT_R32_SINT, VK_FORMAT_R32G32_SINT, VK_FORMAT_R32G32B32_SINT, |
| VK_FORMAT_R32G32B32A32_SINT, VK_FORMAT_R32_UINT, VK_FORMAT_R32G32_UINT, |
| VK_FORMAT_R32G32B32_UINT, VK_FORMAT_R32G32B32A32_UINT, |
| VK_FORMAT_R32_SFLOAT, VK_FORMAT_R32G32_SFLOAT, |
| VK_FORMAT_R32G32B32_SFLOAT, VK_FORMAT_R32G32B32A32_SFLOAT, |
| VK_FORMAT_R64_SFLOAT, VK_FORMAT_R64G64_SFLOAT, |
| VK_FORMAT_R64G64B64_SFLOAT, VK_FORMAT_R64G64B64A64_SFLOAT, |
| VK_FORMAT_R8_USCALED, VK_FORMAT_R8G8_USCALED, |
| VK_FORMAT_R8G8B8_USCALED, VK_FORMAT_R8G8B8A8_USCALED, |
| }; |
| |
| for (int stageIndex = 0; stageIndex < DE_LENGTH_OF_ARRAY(stages); ++stageIndex) |
| { |
| const VkShaderStageFlags stage = stages[stageIndex]; |
| |
| for (int formatIndex = 0; formatIndex < DE_LENGTH_OF_ARRAY(formats); ++formatIndex) |
| { |
| const VkFormat format = formats[formatIndex]; |
| |
| for (int opTypeIndex = 0; opTypeIndex < OPTYPE_LAST; ++opTypeIndex) |
| { |
| bool isBool = false; |
| bool isFloat = false; |
| |
| switch (format) |
| { |
| default: |
| break; |
| case VK_FORMAT_R32_SFLOAT: |
| case VK_FORMAT_R32G32_SFLOAT: |
| case VK_FORMAT_R32G32B32_SFLOAT: |
| case VK_FORMAT_R32G32B32A32_SFLOAT: |
| case VK_FORMAT_R64_SFLOAT: |
| case VK_FORMAT_R64G64_SFLOAT: |
| case VK_FORMAT_R64G64B64_SFLOAT: |
| case VK_FORMAT_R64G64B64A64_SFLOAT: |
| isFloat = true; |
| break; |
| case VK_FORMAT_R8_USCALED: |
| case VK_FORMAT_R8G8_USCALED: |
| case VK_FORMAT_R8G8B8_USCALED: |
| case VK_FORMAT_R8G8B8A8_USCALED: |
| isBool = true; |
| break; |
| } |
| |
| bool isBitwiseOp = false; |
| |
| switch (opTypeIndex) |
| { |
| default: |
| break; |
| case OPTYPE_AND: |
| case OPTYPE_INCLUSIVE_AND: |
| case OPTYPE_EXCLUSIVE_AND: |
| case OPTYPE_OR: |
| case OPTYPE_INCLUSIVE_OR: |
| case OPTYPE_EXCLUSIVE_OR: |
| case OPTYPE_XOR: |
| case OPTYPE_INCLUSIVE_XOR: |
| case OPTYPE_EXCLUSIVE_XOR: |
| isBitwiseOp = true; |
| break; |
| } |
| |
| if (isFloat && isBitwiseOp) |
| { |
| // Skip float with bitwise category. |
| continue; |
| } |
| |
| if (isBool && !isBitwiseOp) |
| { |
| // Skip bool when its not the bitwise category. |
| continue; |
| } |
| |
| CaseDefinition caseDef = {opTypeIndex, stage, format, false}; |
| |
| std::string op = getOpTypeName(opTypeIndex); |
| |
| addFunctionCaseWithPrograms(group.get(), |
| de::toLower(op) + "_" + |
| subgroups::getFormatNameForGLSL(format) + |
| "_" + getShaderStageName(stage), |
| "", initPrograms, test, caseDef); |
| |
| if (VK_SHADER_STAGE_VERTEX_BIT == stage) |
| { |
| caseDef.noSSBO = true; |
| addFunctionCaseWithPrograms(group.get(), de::toLower(op) + "_" + subgroups::getFormatNameForGLSL(format) + |
| "_" + getShaderStageName(stage) + "_framebuffer", "", |
| initFrameBufferPrograms, test, caseDef); |
| } |
| } |
| } |
| } |
| |
| return group.release(); |
| } |
| |
| } // subgroups |
| } // vkt |