name
string
code
string
asm
string
file
string
deqp::gls::ShaderExecUtil::generateFragShaderOutputDecl(std::ostream&, deqp::gls::ShaderExecUtil::ShaderSpec const&, bool, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>, int, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>>, std::allocat...
static void generateFragShaderOutputDecl (std::ostream& src, const ShaderSpec& shaderSpec, bool useIntOutputs, const std::map<std::string, int>& outLocationMap, const std::string& outputPrefix) { DE_ASSERT(glu::glslVersionUsesInOutQualifiers(shaderSpec.version)); for (int outNdx = 0; outNdx < (int)shaderSpec.outputs...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x2b8, %rsp # imm = 0x2B8 movq %r8, 0xb0(%rsp) movq %rcx, 0xa8(%rsp) movl %edx, 0x1c(%rsp) movq 0x20(%rsi), %r13 movq %rsi, 0xa0(%rsp) movq 0x28(%rsi), %rax subq %r13, %rax shrq $0x3, %rax imull $0xb6db6db7, %eax, %eax # imm = 0xB6DB6DB7...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsShaderExecUtil.cpp
deqp::gls::ShaderExecUtil::generateFragShaderOutAssign(std::ostream&, deqp::gls::ShaderExecUtil::ShaderSpec const&, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>> const&)
static void generateFragShaderOutAssign (std::ostream& src, const ShaderSpec& shaderSpec, bool useIntOutputs, const std::string& valuePrefix, const std::string& outputPrefix) { for (vector<Symbol>::const_iterator output = shaderSpec.outputs.begin(); output != shaderSpec.outputs.end(); ++output) { if (useIntOutputs ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x18, %rsp movl %edx, 0x4(%rsp) movq 0x20(%rsi), %rbx movq %rsi, 0x8(%rsp) cmpq 0x28(%rsi), %rbx je 0x878ea6 movq %r8, %rbp movq %rcx, %r14 movq %rdi, %r12 movq %r8, 0x10(%rsp) cmpb $0x0, 0x4(%rsp) je 0x878b62 movl 0x28(%rbx), %eax decl %eax cmpl $...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsShaderExecUtil.cpp
std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>, int, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>>, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>> const, int>>>::mapped_type cons...
const typename M::mapped_type& lookup (const M& map, const typename M::key_type& key) { const typename M::mapped_type* ptr = tryLookup(map, key); if (ptr == DE_NULL) throw std::out_of_range("key not found in map"); return *ptr; }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx callq 0x879130 addq $0x8, %rbx cmpq %rbx, %rax je 0x878f93 addq $0x40, %rax addq $0x8, %rsp popq %rbx popq %r14 retq movl $0x10, %edi callq 0x325650 movq %rax, %rbx leaq 0x1eeaec(%rip), %rsi # 0xa67a93 movq %rax, %rdi callq 0x3253d0 movq 0x3e393a(%rip), %rsi # 0xc5...
/kaydenl[P]VK-GL-CTS/framework/delibs/decpp/deSTLUtil.hpp
bool de::insert<std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>, int, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>>, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>> const, int>>>>(...
bool insert (M& map, const typename M::key_type& key, const typename M::mapped_type& value) { typename M::value_type entry(key, value); std::pair<typename M::iterator,bool> ret = map.insert(entry); return ret.second; }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x28, %rsp movq %rdx, %rbx movq %rdi, %r14 leaq 0x10(%rsp), %r12 movq %r12, -0x10(%r12) movq (%rsi), %rax movq 0x8(%rsi), %rdx addq %rax, %rdx movq %rsp, %r15 movq %r15, %rdi movq %rax, %rsi callq 0x334442 movl (%rbx), %eax movl %eax, 0x20(%r15) movq %r14, %rdi movq %r1...
/kaydenl[P]VK-GL-CTS/framework/delibs/decpp/deSTLUtil.hpp
deqp::gls::TextureSamplerTest::setTextureState(glw::Functions const&, unsigned int, deqp::gls::TextureSamplerTest::SamplingState)
void TextureSamplerTest::setTextureState (const glw::Functions& gl, GLenum target, SamplingState state) { gl.texParameteri(target, GL_TEXTURE_MIN_FILTER, state.minFilter); GLU_EXPECT_NO_ERROR(gl.getError(), "glTexParameteri(target, GL_TEXTURE_MIN_FILTER, state.minFilter)"); gl.texParameteri(target, GL_TEXTURE_MAG_FI...
pushq %rbp pushq %r14 pushq %rbx movl %esi, %ebp movq %rdi, %rbx movl 0x20(%rsp), %edx movl %esi, %edi movl $0x2801, %esi # imm = 0x2801 callq *0x1360(%rbx) callq *0x800(%rbx) leaq 0x20e007(%rip), %rsi # 0xa878d0 leaq 0x20e040(%rip), %r14 # 0xa87910 movl %eax, %edi movq %r14, %rdx movl $0x56, %ecx callq...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::TextureSamplerTest::selectVertexShader(unsigned int)
const char* TextureSamplerTest::selectVertexShader (GLenum target) { switch (target) { case GL_TEXTURE_2D: return "${VTX_HDR}" "${VTX_IN} ${HIGHP} vec2 a_position;\n" "uniform ${HIGHP} float u_posScale;\n" "${VTX_OUT} ${MEDIUMP} vec2 v_texCoord;\n" "void main (void)\n" "{\n" "\tv_texCoord = ...
cmpl $0xde1, %edi # imm = 0xDE1 je 0x879b72 cmpl $0x8513, %edi # imm = 0x8513 je 0x879b6a cmpl $0x806f, %edi # imm = 0x806F jne 0x879b7a leaq 0x20e216(%rip), %rax # 0xa87d7f retq leaq 0x20e2f0(%rip), %rax # 0xa87e61 retq leaq 0x20e127(%rip), %rax # 0xa87ca0 retq xorl %eax, %eax r...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::TextureSamplerTest::createTexture(glw::Functions const&, unsigned int)
GLuint TextureSamplerTest::createTexture (const glw::Functions& gl, GLenum target) { switch (target) { case GL_TEXTURE_2D: return createTexture2D(gl); case GL_TEXTURE_3D: return createTexture3D(gl); case GL_TEXTURE_CUBE_MAP: return createTextureCube(gl); default: DE_ASSERT(false); return (GL...
cmpl $0x8513, %esi # imm = 0x8513 je 0x87b50f cmpl $0x806f, %esi # imm = 0x806F je 0x87b088 cmpl $0xde1, %esi # imm = 0xDE1 jne 0x87b519 jmp 0x87aed6 pushq %rax callq 0x87b240 addq $0x8, %rsp movl $0xffffffff, %eax # imm = 0xFFFFFFFF retq nop
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::TextureSamplerTest::iterate()
tcu::TestCase::IterateResult TextureSamplerTest::iterate (void) { tcu::TestLog& log = m_testCtx.getLog(); tcu::Surface textureRef(VIEWPORT_WIDTH, VIEWPORT_HEIGHT); tcu::Surface samplerRef(VIEWPORT_WIDTH, VIEWPORT_HEIGHT); tcu::Surface textureResult(VIEWPORT_WIDTH, VIEWPORT_HEIGHT); tcu::Surface samplerResult(VIE...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x70, %rsp movq %rdi, %rbx movq 0x8(%rdi), %rax movq 0x10(%rax), %r14 leaq 0x58(%rsp), %rdi movl $0x80, %esi movl $0x80, %edx callq 0x9a957c leaq 0x40(%rsp), %rdi movl $0x80, %esi movl $0x80, %edx callq 0x9a957c leaq 0x28(%rsp), %rdi movl $0x80, %esi movl $0x...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::MultiTextureSamplerTest::setSamplerState(glw::Functions const&, deqp::gls::MultiTextureSamplerTest::SamplingState, unsigned int)
void MultiTextureSamplerTest::setSamplerState (const glw::Functions& gl, SamplingState state, GLuint sampler) { gl.samplerParameteri(sampler, GL_TEXTURE_MIN_FILTER, state.minFilter); GLU_EXPECT_NO_ERROR(gl.getError(), "glSamplerParameteri(sampler, GL_TEXTURE_MIN_FILTER, state.minFilter)"); gl.samplerParameteri(sampl...
pushq %rbp pushq %r14 pushq %rbx movl %esi, %ebp movq %rdi, %rbx movl 0x20(%rsp), %edx movl %esi, %edi movl $0x2801, %esi # imm = 0x2801 callq *0x1280(%rbx) callq *0x800(%rbx) leaq 0x20bc70(%rip), %rsi # 0xa87ae1 leaq 0x20ba98(%rip), %r14 # 0xa87910 movl %eax, %edi movq %r14, %rdx movl $0x268, %ecx ...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::MultiTextureSamplerTest::render()
void MultiTextureSamplerTest::render (void) { const glw::Functions& gl = m_renderCtx.getFunctions(); GLuint samplerLoc1 = (GLuint)-1; GLuint samplerLoc2 = (GLuint)-1; GLuint scaleLoc = (GLuint)-1; gl.useProgram(m_program->getProgram()); GLU_EXPECT_NO_ERROR(gl.getError(), "glUseProgram(m_program->getProgram())")...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rdi, %rbx movq 0x70(%rdi), %rdi movq (%rdi), %rax callq *0x18(%rax) movq %rax, %r14 movq 0x78(%rbx), %rax movl 0x98(%rax), %edi callq *0x1680(%r14) callq *0x800(%r14) leaq 0x19d02e(%rip), %rsi # 0xa19c78 leaq 0x20acbf(%rip), %rd...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::MultiTextureSamplerTest::createTextureCube(glw::Functions const&, int)
GLuint MultiTextureSamplerTest::createTextureCube (const glw::Functions& gl, int id) { GLuint texture = (GLuint)-1; tcu::TextureCube refTexture (tcu::TextureFormat(tcu::TextureFormat::RGBA, tcu::TextureFormat::UNORM_INT8), CUBEMAP_SIZE); gl.genTextures(1, &texture); GLU_EXPECT_NO_ERROR(gl.getError(), "glGenTex...
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x198, %rsp # imm = 0x198 movl %esi, %ebp movq %rdi, %rbx leaq 0x2c(%rsp), %r14 movl $0xffffffff, (%r14) # imm = 0xFFFFFFFF movabsq $0x300000008, %rax # imm = 0x300000008 movq %rsp, %rsi movq %rax, (%rsi) leaq 0x30(%rsp), %rdi movl $0x20, %edx callq ...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::MultiTextureSamplerTest::renderResults(tcu::Surface&, tcu::Surface&, int, int)
void MultiTextureSamplerTest::renderResults (tcu::Surface& textureResult, tcu::Surface& samplerResult, int x, int y) { const glw::Functions& gl = m_renderCtx.getFunctions(); GLuint texture1 = createTexture(gl, m_target, 0); GLuint texture2 = createTexture(gl, m_target, 1); GLuint sampler = -1; gl.v...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xb8, %rsp movl %r8d, %r12d movl %ecx, %r13d movq %rdx, %rbp movq %rsi, 0x48(%rsp) movq %rdi, %r14 movq 0x70(%rdi), %rdi movq (%rdi), %rax callq *0x18(%rax) movq %rax, %rbx movl 0x80(%r14), %esi movq %rax, %rdi xorl %edx, %edx callq 0x87db76 movl %...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::MultiTextureSamplerTest::iterate()
tcu::TestCase::IterateResult MultiTextureSamplerTest::iterate (void) { tcu::TestLog& log = m_testCtx.getLog(); tcu::Surface textureRef(VIEWPORT_WIDTH, VIEWPORT_HEIGHT); tcu::Surface samplerRef(VIEWPORT_WIDTH, VIEWPORT_HEIGHT); tcu::Surface textureResult(VIEWPORT_WIDTH, VIEWPORT_HEIGHT); tcu::Surface samplerResul...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x70, %rsp movq %rdi, %rbx movq 0x8(%rdi), %rax movq 0x10(%rax), %r14 leaq 0x58(%rsp), %rdi movl $0x80, %esi movl $0x80, %edx callq 0x9a957c leaq 0x40(%rsp), %rdi movl $0x80, %esi movl $0x80, %edx callq 0x9a957c leaq 0x28(%rsp), %rdi movl $0x80, %esi movl $0x...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsSamplerObjectTest.cpp
deqp::gls::ShaderPerformanceMeasurer::render(int)
void ShaderPerformanceMeasurer::render (int numDrawCalls) { const glw::Functions& gl = m_renderCtx.getFunctions(); GLsizei numIndices = (GLsizei)getNumIndices(m_gridSizeX, m_gridSizeY); gl.viewport(0, 0, m_viewportWidth, m_viewportHeight); for (int callNdx = 0; callNdx < numDrawCalls; callNdx++) gl.drawEl...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx movl %esi, %ebx movq %rdi, %r15 movq (%rdi), %rdi movq (%rdi), %rax callq *0x18(%rax) movq %rax, %r14 movl 0x8(%r15), %r12d movl 0x10(%r15), %edx imull 0xc(%r15), %r12d movl 0x14(%r15), %ecx xorl %edi, %edi xorl %esi, %esi callq *0x1a00(%rax) testl %ebx, %ebx jle 0...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsShaderPerformanceMeasurer.cpp
deqp::gls::InteractionTestUtil::getRandomColor(de::Random&)
static Vec4 getRandomColor (de::Random& rnd) { static const float components[] = { 0.0f, 0.2f, 0.4f, 0.5f, 0.6f, 0.8f, 1.0f }; float r = rnd.choose<float>(DE_ARRAY_BEGIN(components), DE_ARRAY_END(components)); float g = rnd.choose<float>(DE_ARRAY_BEGIN(components), DE_ARRAY_END(components)); float b = rnd.choose<fl...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x18, %rsp movq %rsi, %r14 movq %rdi, %rbx xorl %ebp, %ebp leaq 0x8(%rsp), %r13 movl %ebp, (%r13) leaq 0x2086bc(%rip), %r15 # 0xa89070 leaq 0x2086d1(%rip), %r12 # 0xa8908c movq %rsi, %rdi movq %r15, %rsi movq %r12, %rdx movq %r13, %rcx movl $...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsInteractionTestUtil.cpp
deqp::gls::FragOpInteractionCase::deinit()
void FragOpInteractionCase::deinit (void) { delete m_referenceCtx; m_referenceCtx = DE_NULL; delete m_glCtx; m_glCtx = DE_NULL; delete m_program; m_program = DE_NULL; }
pushq %r15 pushq %r14 pushq %rbx movq %rdi, %rbx movq 0x440(%rdi), %r14 testq %r14, %r14 je 0x8812d4 leaq 0xd0(%r14), %rdi callq 0x8a13b4 leaq 0x58(%r14), %r15 leaq 0xa8(%r14), %rdi callq 0x9b433a leaq 0x80(%r14), %rdi callq 0x9b433a movq %r15, %rdi callq 0x9b433a movq 0x38(%r14), %rdi leaq 0x48(%r14), %rax cmpq %rax, ...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsFragOpInteractionCase.cpp
deqp::gls::FragOpInteractionCase::init()
void FragOpInteractionCase::init (void) { de::Random rnd (m_params.seed ^ 0x232faac); const int viewportW = de::min<int>(m_renderCtx.getRenderTarget().getWidth(), VIEWPORT_WIDTH); const int viewportH = de::min<int>(m_renderCtx.getRenderTarget().getHeight(), VIEWPORT_HEIGHT); const int viewportX = ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp movq %rdi, %rbx leaq 0x80(%rdi), %rax movq %rax, 0x10(%rsp) movl $0x232faac, %esi # imm = 0x232FAAC xorl 0x80(%rdi), %esi leaq 0x28(%rsp), %r12 movq %r12, %rdi callq 0x9fa2bc movq 0x70(%rbx), %rdi movq (%rdi), %rax callq *0x20(%ra...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsFragOpInteractionCase.cpp
deqp::gls::StateChangePerformanceCase::deinit()
void StateChangePerformanceCase::deinit (void) { m_indices.clear(); m_interleavedResults.clear(); m_batchedResults.clear(); { const glw::Functions& gl = m_renderCtx.getFunctions(); if (!m_indexBuffers.empty()) { gl.deleteBuffers((GLsizei)m_indexBuffers.size(), &(m_indexBuffers[0])); m_indexBuffers.cle...
pushq %r15 pushq %r14 pushq %rbx movq %rdi, %rbx movq 0x148(%rdi), %rax cmpq %rax, 0x150(%rdi) je 0x883877 movq %rax, 0x150(%rbx) movq 0x160(%rbx), %rax cmpq %rax, 0x168(%rbx) je 0x88388e movq %rax, 0x168(%rbx) movq 0x178(%rbx), %rax cmpq %rax, 0x180(%rbx) je 0x8838a5 movq %rax, 0x180(%rbx) movq 0x70(%rbx), %rdi movq (...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsStateChangePerfTestCases.cpp
deqp::gls::StateChangePerformanceCase::callDraw(glw::Functions const&)
void StateChangePerformanceCase::callDraw (const glw::Functions& gl) { switch (m_drawType) { case DRAWTYPE_NOT_INDEXED: gl.drawArrays(GL_TRIANGLES, 0, m_triangleCount * 3); break; case DRAWTYPE_INDEXED_USER_PTR: gl.drawElements(GL_TRIANGLES, m_triangleCount * 3, GL_UNSIGNED_SHORT, &m_indices[0]); break; ...
movl 0x78(%rdi), %eax cmpl $0x2, %eax je 0x885d2d cmpl $0x1, %eax je 0x885d0a testl %eax, %eax jne 0x885d4b movq 0x538(%rsi), %rax movl 0x84(%rdi), %ecx leal (%rcx,%rcx,2), %edx movl $0x4, %edi xorl %esi, %esi jmpq *%rax movq 0x568(%rsi), %rax movl 0x84(%rdi), %ecx leal (%rcx,%rcx,2), %esi movq 0x148(%rdi), %rcx movl $...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsStateChangePerfTestCases.cpp
deqp::gls::StateChangeCallPerformanceCase::iterate()
tcu::TestCase::IterateResult StateChangeCallPerformanceCase::iterate (void) { if (m_results.empty()) logTestCase(); if ((int)m_results.size() < m_iterationCount) { executeTest(); GLU_EXPECT_NO_ERROR(m_renderCtx.getFunctions().getError(), "Unexpected error"); return CONTINUE; } else { logAndSetTestResul...
pushq %rbx movq %rdi, %rbx movq 0x80(%rdi), %rax cmpq 0x88(%rdi), %rax jne 0x8865f4 movq %rbx, %rdi callq 0x885e58 movq 0x88(%rbx), %rax subq 0x80(%rbx), %rax shrq $0x3, %rax cmpl %eax, 0x78(%rbx) jle 0x886644 movq %rbx, %rdi callq 0x885dce movq 0x70(%rbx), %rdi movq (%rdi), %rax callq *0x18(%rax) callq *0x800(%rax) le...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsStateChangePerfTestCases.cpp
deqp::gls::BufferTestUtil::getUsageHintName(unsigned int)
const char* getUsageHintName (deUint32 hint) { switch (hint) { case GL_STREAM_DRAW: return "stream_draw"; case GL_STREAM_READ: return "stream_read"; case GL_STREAM_COPY: return "stream_copy"; case GL_STATIC_DRAW: return "static_draw"; case GL_STATIC_READ: return "static_read"; case GL_STATIC_COPY: return ...
addl $0xffff7720, %edi # imm = 0xFFFF7720 cmpl $0xa, %edi ja 0x887229 movl %edi, %eax leaq 0x3cf864(%rip), %rcx # 0xc56a88 movq (%rcx,%rax,8), %rax retq xorl %eax, %eax retq
/kaydenl[P]VK-GL-CTS/modules/glshared/glsBufferTestUtil.cpp
deqp::gls::BufferTestUtil::BufferWriter::BufferWriter(glu::RenderContext&, tcu::TestLog&, deqp::gls::BufferTestUtil::WriteType)
BufferWriter::BufferWriter (glu::RenderContext& renderCtx, tcu::TestLog& log, WriteType writeType) : m_writer(DE_NULL) { switch (writeType) { case WRITE_BUFFER_SUB_DATA: m_writer = new BufferSubDataWriter (renderCtx, log); break; case WRITE_BUFFER_WRITE_MAP: m_writer = new BufferWriteMapWriter (renderCtx, log);...
pushq %r15 pushq %r14 pushq %r12 pushq %rbx pushq %rax movq %rdx, %r12 movq %rsi, %r15 movq %rdi, %rbx movq $0x0, (%rdi) cmpl $0x1, %ecx je 0x887661 testl %ecx, %ecx jne 0x8876ab movl $0x28, %edi callq 0x325210 movq %rax, %r14 movq (%r15), %rax movq %r15, %rdi callq *0x18(%rax) movq %r14, %rdi addq $0x8, %rdi movq %rax...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsBufferTestUtil.cpp
deqp::gls::BufferTestUtil::VertexArrayVerifier::~VertexArrayVerifier()
VertexArrayVerifier::~VertexArrayVerifier (void) { const glw::Functions& gl = m_renderCtx.getFunctions(); if (m_vao) gl.deleteVertexArrays(1, &m_vao); if (m_positionBuf) gl.deleteBuffers(1, &m_positionBuf); if (m_indexBuf) gl.deleteBuffers(1, &m_indexBuf); delete m_program; }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx leaq 0x3ce870(%rip), %rax # 0xc56858 movq %rax, (%rdi) movq 0x20(%rdi), %rdi movq (%rdi), %rax callq *0x18(%rax) movq %rax, %r14 cmpl $0x0, 0x40(%rbx) je 0x88800e leaq 0x40(%rbx), %rsi movl $0x1, %edi callq *0x490(%r14) cmpl $0x0, 0x44(%rbx) je 0x888024 leaq 0x44(%rbx...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsBufferTestUtil.cpp
deqp::gls::BufferTestUtil::IndexArrayVerifier::~IndexArrayVerifier()
IndexArrayVerifier::~IndexArrayVerifier (void) { const glw::Functions& gl = m_renderCtx.getFunctions(); if (m_vao) gl.deleteVertexArrays(1, &m_vao); if (m_positionBuf) gl.deleteBuffers(1, &m_positionBuf); if (m_colorBuf) gl.deleteBuffers(1, &m_colorBuf); delete m_program; }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx leaq 0x3cd344(%rip), %rax # 0xc56898 movq %rax, (%rdi) movq 0x20(%rdi), %rdi movq (%rdi), %rax callq *0x18(%rax) movq %rax, %r14 cmpl $0x0, 0x40(%rbx) je 0x88957a leaq 0x40(%rbx), %rsi movl $0x1, %edi callq *0x490(%r14) cmpl $0x0, 0x44(%rbx) je 0x889590 leaq 0x44(%rbx...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsBufferTestUtil.cpp
deqp::gls::(anonymous namespace)::logProgram(tcu::TestLog&, glw::Functions const&, unsigned int)
void logProgram (TestLog& log, const glw::Functions& gl, deUint32 program) { const bool programLinkOk = getProgramLinkStatus(gl, program); const string programInfoLog = getProgramInfoLog(gl, program); tcu::ScopedLogSection linkInfo (log, "Program Link Info", "Program Link Info"); { tcu::ScopedLogSection in...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1d8, %rsp # imm = 0x1D8 movl %edx, %ebp movq %rsi, %r14 movq %rdi, %rbx leaq 0x58(%rsp), %r15 movl %edx, %edi movl $0x8b82, %esi # imm = 0x8B82 movq %r15, %rdx callq *0x9d8(%r14) callq *0x800(%r14) leaq 0x1fc740(%rip), %rsi ...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
deqp::gls::(anonymous namespace)::generateTestName[abi:cxx11](deqp::gls::AttributeLocationTestUtil::AttribType const&, int)
string generateTestName (const AttribType& type, int arraySize) { return type.getName() + (arraySize != Attribute::NOT_ARRAY ? "_array_" + de::toString(arraySize) : ""); }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x1c0, %rsp # imm = 0x1C0 movl %edx, %ebp movq %rsi, %r14 movq %rdi, %rbx cmpl $-0x1, %edx setne %r12b je 0x88f581 leaq 0x48(%rsp), %r15 movq %r15, %rdi callq 0x325e00 movq %r15, %rdi movl %ebp, %esi callq 0x325530 leaq 0x50(%rsp), %rsi leaq 0x28(%...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
deqp::gls::BindAliasingAttributeTest::iterate()
tcu::TestCase::IterateResult BindAliasingAttributeTest::iterate (void) { const vector<Bind> noBindings; vector<Attribute> attributes; vector<Bind> bindings; attributes.push_back(Attribute(m_type, "a_0", Attribute::LOC_UNDEF, Cond("A", true), m_arraySize)); attributes.push_back(Attribute(AttribType("vec4", 1, GL...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 movq %rdi, %rbx xorps %xmm0, %xmm0 movaps %xmm0, 0x170(%rsp) xorl %eax, %eax movq %rax, 0x180(%rsp) movq %rax, 0xa0(%rsp) movaps %xmm0, 0x90(%rsp) movq %rax, 0x80(%rsp) movaps %xmm0, 0x70(%rsp) leaq 0x58(%rsp), ...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
deqp::gls::BindMaxAliasingAttributeTest::iterate()
tcu::TestCase::IterateResult BindMaxAliasingAttributeTest::iterate (void) { const vector<Bind> noBindings; const deInt32 maxAttributes = getMaxAttributeLocations(m_renderCtx); const int arrayElementCount = (m_arraySize != Attribute::NOT_ARRAY ? m_arraySize : 1); vector<Attribute> attributes; vector<Bind> bin...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x2e8, %rsp # imm = 0x2E8 movq %rdi, %rbx xorps %xmm0, %xmm0 movaps %xmm0, 0x140(%rsp) movq $0x0, 0x150(%rsp) movq 0x70(%rdi), %rdi callq 0x890121 movq %rax, 0x70(%rsp) movl 0xa0(%rbx), %eax cmpl $-0x1, %eax movl $0x1, %ecx cmovnel %eax,...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
deqp::gls::BindHoleAttributeTest::iterate()
deInt32 getMaxAttributeLocations (glu::RenderContext& renderCtx) { const glw::Functions& gl = renderCtx.getFunctions(); deInt32 maxAttribs; gl.getIntegerv(GL_MAX_VERTEX_ATTRIBS, &maxAttribs); GLU_EXPECT_NO_ERROR(gl.getError(), "glGetIntegerv()"); return maxAttribs; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x2c8, %rsp # imm = 0x2C8 xorps %xmm0, %xmm0 movaps %xmm0, 0x130(%rsp) movq $0x0, 0x140(%rsp) movq %rdi, 0x28(%rsp) movq 0x70(%rdi), %rdi callq 0x890121 movl %eax, 0x24(%rsp) leaq 0x160(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0x180ae0(%r...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
deqp::gls::BindInactiveAliasingAttributeTest::iterate()
tcu::TestCase::IterateResult BindInactiveAliasingAttributeTest::iterate (void) { const vector<Bind> noBindings; const deInt32 maxAttributes = getMaxAttributeLocations(m_renderCtx); const int arrayElementCount = (m_arraySize != Attribute::NOT_ARRAY ? m_arraySize : 1); vector<Attribute> attributes; vector<Bind>...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x2e8, %rsp # imm = 0x2E8 movq %rdi, %rbx xorps %xmm0, %xmm0 movaps %xmm0, 0x140(%rsp) movq $0x0, 0x150(%rsp) movq 0x70(%rdi), %rdi callq 0x890121 movq %rax, 0x50(%rsp) movl 0xa0(%rbx), %eax cmpl $-0x1, %eax movl $0x1, %ecx cmovnel %eax,...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
deqp::gls::MixedAttributeTest::MixedAttributeTest(tcu::TestContext&, glu::RenderContext&, deqp::gls::AttributeLocationTestUtil::AttribType const&, int)
MixedAttributeTest::MixedAttributeTest (tcu::TestContext& testCtx, glu::RenderContext& renderCtx, const AttribType& type, int arraySize) : TestCase (testCtx, generateTestName(type, arraySize).c_str(), generateTestName(type, arraySize).c_str()) , m_renderCtx (renderCtx) , m_type (t...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movl %r8d, %ebp movq %rcx, %r14 movq %rdx, %r15 movq %rsi, %r12 movq %rdi, %rbx leaq 0x28(%rsp), %r13 movq %r13, %rdi movq %rcx, %rsi movl %r8d, %edx callq 0x88f4d7 movq (%r13), %r13 leaq 0x8(%rsp), %rdi movq %r14, %rsi movl %ebp, %edx c...
/kaydenl[P]VK-GL-CTS/modules/glshared/glsAttributeLocationTests.cpp
rsg::AssignOp::~AssignOp()
AssignOp::~AssignOp (void) { delete m_lvalueExpr; delete m_rvalueExpr; }
pushq %rbx movq %rdi, %rbx leaq 0x2d7de7(%rip), %rax # 0xc597d0 movq %rax, (%rdi) movq 0xa0(%rdi), %rdi testq %rdi, %rdi je 0x9819fe movq (%rdi), %rax callq *0x8(%rax) movq 0xa8(%rbx), %rdi testq %rdi, %rdi je 0x981a10 movq (%rdi), %rax callq *0x8(%rax) movq 0x88(%rbx), %rdi testq %rdi, %rdi je 0x981a2b movq 0x98(%r...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::AssignOp::createNextChild(rsg::GeneratorState&)
Expression* AssignOp::createNextChild (GeneratorState& state) { if (m_lvalueExpr == DE_NULL) { // Construct lvalue // \todo [2011-03-14 pyry] Proper l-value generation: // - pure L-value part is generated first // - variable valuerange is made unbound // - R-value is generated // - R-values in L-value...
pushq %r14 pushq %rbx subq $0x38, %rsp movq %rsi, %r14 movq %rdi, %rbx cmpq $0x0, 0xa0(%rdi) je 0x981bac cmpq $0x0, 0xa8(%rbx) je 0x981c1d xorl %eax, %eax jmp 0x981c8c movq 0x58(%rbx), %rax movq 0x70(%rbx), %rcx xorl %edx, %edx cmpq 0x60(%rbx), %rax cmoveq %rdx, %rax cmpq 0x78(%rbx), %rcx cmoveq %rdx, %rcx leaq 0x8(%rb...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::AssignOp::evaluate(rsg::ExecutionContext&)
void AssignOp::evaluate (ExecutionContext& evalCtx) { // Evaluate l-value m_lvalueExpr->evaluate(evalCtx); // Evaluate value m_rvalueExpr->evaluate(evalCtx); m_value.setStorage(m_valueRange.getType()); m_value.getValue(m_valueRange.getType()) = m_rvalueExpr->getValue().value(); // Assign assignMasked(m_lvalue...
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx movq %rsi, %rbx movq %rdi, %r14 movq 0xa0(%rdi), %rdi movq (%rdi), %rax callq *0x20(%rax) movq 0xa8(%r14), %rdi movq (%rdi), %rax movq %rbx, %rsi callq *0x20(%rax) leaq 0x88(%r14), %r12 leaq 0x8(%r14), %r15 movq %r15, %rdi callq 0x97dc5e shll $0x6, %eax movslq %eax...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::(anonymous namespace)::allocateNewVariable(rsg::GeneratorState&, rsg::ConstValueRangeAccess)
Variable* allocateNewVariable (GeneratorState& state, ConstValueRangeAccess valueRange) { Variable* variable = state.getVariableManager().allocate(valueRange.getType()); // Update value range state.getVariableManager().setValue(variable, valueRange); // Random storage \todo [pyry] Check that scalar count in unifo...
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x38, %rsp movq %rdi, %r14 movq 0x28(%rdi), %rdi movq 0x60(%rsp), %r12 movq %r12, %rsi callq 0x97bfce movq %rax, %rbx movq 0x28(%r14), %rdi movq 0x70(%rsp), %rax movq %rax, 0x10(%rsp) movaps 0x60(%rsp), %xmm0 movups %xmm0, (%rsp) movq %rbx, %rsi callq 0x97c2fa movq 0x70...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::VariableRead::getWeight(rsg::GeneratorState const&, rsg::ConstValueRangeAccess)
float VariableRead::getWeight (const GeneratorState& state, ConstValueRangeAccess valueRange) { if (valueRange.getType().isVoid()) { if (state.getVariableManager().hasEntry(IsReadableEntry(state.getExpressionFlags())) || state.getVariableManager().getNumAllocatedScalars() < state.getShaderParameters().maxCombine...
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x48, %rsp movq %rdi, %rbx movq 0x70(%rsp), %rsi cmpl $0x0, (%rsi) je 0x982a27 movq %rbx, %rdi callq 0x9819a9 testb %al, %al jne 0x982a1d leaq 0x70(%rsp), %rax movq 0x28(%rbx), %r14 movq 0x48(%rbx), %rcx movl -0x4(%rcx), %ecx movq 0x10(%rax), %rdx movq %rdx, 0x40(%rsp) ...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::VariableWrite::VariableWrite(rsg::GeneratorState&, rsg::ConstValueRangeAccess)
VariableWrite::VariableWrite (GeneratorState& state, ConstValueRangeAccess valueRange) { DE_ASSERT(!valueRange.getType().isVoid()); // Find variable with range that is superset of given range IsWritableSupersetEntry::Iterator first = state.getVariableManager().getBegin(IsWritableSupersetEntry(valueRange)); IsWrita...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x118, %rsp # imm = 0x118 movq %rdi, %r12 leaq 0x150(%rsp), %rbx xorps %xmm0, %xmm0 movups %xmm0, 0x8(%rdi) movq $0x0, 0x18(%rdi) leaq 0x2d7132(%rip), %rax # 0xc59c00 movq %rax, (%rdi) movq %rsi, 0x18(%rsp) movq 0x28(%rsi), %r14 movq ...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::ParenOp::tokenize(rsg::GeneratorState&, rsg::TokenStream&) const
void ParenOp::tokenize (GeneratorState& state, TokenStream& str) const { str << Token::LEFT_PAREN; m_child->tokenize(state, str); str << Token::RIGHT_PAREN; }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x18, %rsp movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r15 movl $0x10, 0x8(%rsp) movq 0x8(%rdx), %rax movq 0x18(%rdx), %rsi subq (%rdx), %rax sarq $0x4, %rax cmpq %rsi, %rax jne 0x983132 addq $0x40, %rsi movq %rbx, %rdi callq 0x974778 movq 0x18(%rbx), %rdi shlq $0x4, %r...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::SwizzleOp::createNextChild(rsg::GeneratorState&)
Expression* SwizzleOp::createNextChild (GeneratorState& state) { if (m_child) return DE_NULL; // Compute input value range. VariableType inVarType = VariableType(m_outValueRange.getType().getBaseType(), m_numInputElements); ValueRange inValueRange = ValueRange(inVarType); // Initialize all inputs to -inf..in...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x178, %rsp # imm = 0x178 cmpq $0x0, 0x90(%rdi) je 0x9834a7 xorl %r15d, %r15d jmp 0x9837d7 movq %rsi, %rbx movq %rdi, %rbp movl 0x8(%rdi), %eax movl 0x88(%rdi), %ecx leaq 0x140(%rsp), %rdx movl %eax, -0x18(%rdx) movl $0x0, -0x14(%rdx) mo...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
rsg::TexLookup::createNextChild(rsg::GeneratorState&)
Expression* TexLookup::createNextChild (GeneratorState& state) { bool hasLodBias = m_type == TYPE_TEXTURE2D_LOD || m_type == TYPE_TEXTURE2D_PROJ_LOD || m_type == TYPE_TEXTURECUBE_LOD; if (hasLodBias && !m_lodBiasExpr) { ValueRange lodRange(VariableType(VariableType::TYPE_FLOAT, 1)); setInfinite...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x128, %rsp # imm = 0x128 movq %rsi, %rbx movq %rdi, %r13 movl 0x8(%rdi), %eax cmpl $0x5, %eax ja 0x98408a movl $0x2a, %ecx btl %eax, %ecx jae 0x98408a cmpq $0x0, 0x20(%r13) je 0x98409d cmpq $0x0, 0x18(%r13) je 0x9841d9 xorl %r14d, %r14d...
/kaydenl[P]VK-GL-CTS/framework/randomshaders/rsgExpression.cpp
tcu::intThresholdCompare(tcu::TestLog&, char const*, char const*, tcu::ConstPixelBufferAccess const&, tcu::ConstPixelBufferAccess const&, tcu::Vector<unsigned int, 4> const&, tcu::CompareLogMode)
bool intThresholdCompare (TestLog& log, const char* imageSetName, const char* imageSetDesc, const ConstPixelBufferAccess& reference, const ConstPixelBufferAccess& result, const UVec4& threshold, CompareLogMode logMode) { int width = reference.getWidth(); int height = reference.getHeight(); int dept...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x4b8, %rsp # imm = 0x4B8 movq %r9, %r13 movq %r8, %r12 movq %rdx, %rbp movq %rsi, 0xe0(%rsp) movq %rdi, %r15 movl 0x8(%rcx), %ebx movl 0xc(%rcx), %r9d movq %rcx, 0x28(%rsp) movl 0x10(%rcx), %r8d movabsq $0x300000007, %rax # imm = 0...
/kaydenl[P]VK-GL-CTS/framework/common/tcuImageCompare.cpp
tcu::intThresholdPositionDeviationCompare(tcu::TestLog&, char const*, char const*, tcu::ConstPixelBufferAccess const&, tcu::ConstPixelBufferAccess const&, tcu::Vector<unsigned int, 4> const&, tcu::Vector<int, 3> const&, bool, tcu::CompareLogMode)
bool intThresholdPositionDeviationCompare (TestLog& log, const char* imageSetName, const char* imageSetDesc, const ConstPixelBufferAccess& reference, const ConstPixelBufferAccess& result, const UVec4& threshold, const tcu::IVec3& maxPositionDeviation, bool acceptOutOfBoundsAsAnyValue, CompareLogMode logMode) { const i...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x478, %rsp # imm = 0x478 movq %r9, %rbx movq %r8, %r15 movq %rcx, %r12 movq %rdx, 0x78(%rsp) movq %rsi, 0x80(%rsp) movq %rdi, %r14 movl 0x8(%rcx), %edx movl 0xc(%rcx), %ecx movl 0x10(%r12), %r8d movabsq $0x300000007, %rax # imm = 0...
/kaydenl[P]VK-GL-CTS/framework/common/tcuImageCompare.cpp
tcu::(anonymous namespace)::findNumPositionDeviationFailingPixels(tcu::PixelBufferAccess const&, tcu::ConstPixelBufferAccess const&, tcu::ConstPixelBufferAccess const&, tcu::Vector<unsigned int, 4> const&, tcu::Vector<int, 3> const&, bool)
static int findNumPositionDeviationFailingPixels (const PixelBufferAccess& errorMask, const ConstPixelBufferAccess& reference, const ConstPixelBufferAccess& result, const UVec4& threshold, const tcu::IVec3& maxPositionDeviation, bool acceptOutOfBoundsAsAnyValue) { const tcu::IVec4 okColor (0, 255, 0, 255); const t...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xd8, %rsp movq %r8, 0x8(%rsp) movq %rcx, %r14 movq %rdi, 0x68(%rsp) movaps 0x9de4c(%rip), %xmm0 # 0xa41fa0 movups %xmm0, 0xc8(%rsp) movaps 0x9de2d(%rip), %xmm0 # 0xa41f90 movups %xmm0, 0xb8(%rsp) movq %rsi, 0x60(%rsp) movl 0x8(%rsi), %ebx mo...
/kaydenl[P]VK-GL-CTS/framework/common/tcuImageCompare.cpp
tcu::getPositionOfIEEEFloatWithoutDenormals(float)
static deInt32 getPositionOfIEEEFloatWithoutDenormals (float x) { DE_ASSERT(!deIsNaN(x)); // not sane if (x == 0.0f) return 0; else if (x < 0.0f) return -getPositionOfIEEEFloatWithoutDenormals(-x); else { DE_ASSERT(x > 0.0f); const tcu::Float32 f(x); if (f.isDenorm()) { // Denorms are flushed to ...
xorl %eax, %eax xorps %xmm1, %xmm1 ucomiss %xmm1, %xmm0 jne 0x9a5f64 jp 0x9a5f64 retq ucomiss %xmm0, %xmm1 jbe 0x9a5f7d pushq %rax xorps 0x5af5f(%rip), %xmm0 # 0xa00ed0 callq 0x9a5f57 negl %eax addq $0x8, %rsp retq movd %xmm0, %ecx leal -0x7fffff(%rcx), %edx xorl %eax, %eax testl $0x7fffff, %ecx # imm = 0x7F...
/kaydenl[P]VK-GL-CTS/framework/common/tcuImageCompare.cpp
tcu::applyMonotone(tcu::Interval (&)(double, double), tcu::Interval const&, tcu::Interval const&)
Interval applyMonotone (DoubleIntervalFunc2& func, const Interval& arg0, const Interval& arg1) { double lo0 = arg0.lo(), hi0 = arg0.hi(), lo1 = arg1.lo(), hi1 = arg1.hi(); return Interval(Interval(func(lo0, lo1), func(lo0, hi1)), Interval(func(hi0, lo1), func(hi0, hi1))); }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0xa0, %rsp movq %rsi, %r14 movq %rdi, %rbx movsd 0x8(%rdx), %xmm2 movsd %xmm2, 0x10(%rsp) movsd 0x10(%rdx), %xmm0 movsd %xmm0, 0x8(%rsp) movsd 0x8(%rcx), %xmm1 movsd %xmm1, 0x28(%rsp) movsd 0x10(%rcx), %xmm0 movsd %xmm0, (%rsp) leaq 0x88(%rsp), %r15 movq %r15...
/kaydenl[P]VK-GL-CTS/framework/common/tcuInterval.cpp
tcu::RGBA::toVec() const
int getAlpha (void) const { return (int)((m_value >> (deUint32)ALPHA_SHIFT) & 0xFFu); }
movl (%rsi), %ecx movl $0xff, %edx movl %ecx, %eax andl %edx, %eax cvtsi2ss %eax, %xmm0 movq %rdi, %rax movl %ecx, %esi shrl $0x8, %esi andl %edx, %esi cvtsi2ss %esi, %xmm1 movss 0x598d1(%rip), %xmm2 # 0xa02500 divss %xmm2, %xmm0 movl %ecx, %esi shrl $0x10, %esi andl %edx, %esi cvtsi2ss %esi, %xmm3 shrl $0x18, %ecx ...
/kaydenl[P]VK-GL-CTS/framework/common/tcuRGBA.hpp
tcu::FileResource::FileResource(char const*)
FileResource::FileResource (const char* filename) : Resource(std::string(filename)) { m_file = fopen(filename, "rb"); if (!m_file) throw ResourceError("Failed to open file", filename, __FILE__, __LINE__); }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x30, %rsp movq %rsi, %r14 movq %rdi, %rbx leaq 0x10(%rsp), %rdi leaq 0xf(%rsp), %rdx callq 0x333da2 leaq 0x2b2a59(%rip), %r13 # 0xc5b9a8 movq %r13, (%rbx) leaq 0x8(%rbx), %rdi leaq 0x18(%rbx), %r12 movq %r12, 0x8(%rbx) movq 0x10(%rsp), %rsi movq 0x18(%rsp...
/kaydenl[P]VK-GL-CTS/framework/common/tcuResource.cpp
tcu::ResourcePrefix::getResource(char const*) const
Resource* ResourcePrefix::getResource (const char* name) const { return m_archive.getResource((m_prefix + name).c_str()); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x28, %rsp movq %rsi, %r14 movq 0x8(%rdi), %rbx leaq 0x18(%rsp), %r12 movq %r12, -0x10(%r12) movq 0x10(%rdi), %rsi movq 0x18(%rdi), %rdx addq %rsi, %rdx leaq 0x8(%rsp), %r15 movq %r15, %rdi callq 0x334442 movq %r15, %rdi movq %r14, %rsi callq 0x3259c0 movq 0x8(%rsp), %r...
/kaydenl[P]VK-GL-CTS/framework/common/tcuResource.cpp
tcu::ResourcePrefix::~ResourcePrefix()
virtual ~ResourcePrefix (void) {}
pushq %rbx movq %rdi, %rbx leaq 0x2b26ab(%rip), %rax # 0xc5b918 movq %rax, (%rdi) movq 0x10(%rdi), %rdi leaq 0x20(%rbx), %rax cmpq %rax, %rdi je 0x9a9288 movq (%rax), %rsi incq %rsi callq 0x3251a0 movl $0x30, %esi movq %rbx, %rdi popq %rbx jmp 0x3251a0
/kaydenl[P]VK-GL-CTS/framework/common/tcuResource.hpp
tcu::TextureCube::operator=(tcu::TextureCube const&)
TextureCube& TextureCube::operator= (const TextureCube& other) { if (this == &other) return *this; const int numLevels = computeMipPyramidLevels(other.m_size); const ConstPixelBufferAccess* levels[CUBEFACE_LAST]; for (int face = 0; face < CUBEFACE_LAST; face++) { m_data[face].resize(numLevels); m_acc...
movq %rdi, %rcx cmpq %rsi, %rdi je 0x9b74cd pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rsi, %r13 movl 0x8(%rsi), %ebp movl $0x20, %r15d movl $0x20, %eax testl %ebp, %ebp je 0x9b7328 bsrl %ebp, %eax xorl $0x1f, %eax subl %eax, %r15d movq %rcx, 0x8(%rsp) leaq 0xa0(%rcx), %r12...
/kaydenl[P]VK-GL-CTS/framework/common/tcuTexture.cpp
tcu::Texture2DArray::allocLevel(int)
void Texture2DArray::allocLevel (int levelNdx) { DE_ASSERT(de::inBounds(levelNdx, 0, getNumLevels())); const int width = getMipPyramidLevelSize(m_width, levelNdx); const int height = getMipPyramidLevelSize(m_height, levelNdx); TextureLevelPyramid::allocLevel(levelNdx, width, height, m_numLayers); }
movl %esi, %ecx movl 0x38(%rdi), %edx movl 0x3c(%rdi), %eax sarl %cl, %edx cmpl $0x2, %edx movl $0x1, %esi cmovll %esi, %edx sarl %cl, %eax cmpl $0x2, %eax cmovll %esi, %eax movl 0x40(%rdi), %r8d movl %ecx, %esi movl %eax, %ecx jmp 0x9b5b64
/kaydenl[P]VK-GL-CTS/framework/common/tcuTexture.cpp
tcu::Texture3D::operator=(tcu::Texture3D const&)
Texture3D& Texture3D::operator= (const Texture3D& other) { if (this == &other) return *this; TextureLevelPyramid::operator=(other); m_width = other.m_width; m_height = other.m_height; m_depth = other.m_depth; m_view = Texture3DView(getNumLevels(), getLevels()); return *this; }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx cmpq %rsi, %rdi je 0x9b7e00 movq %rsi, %r14 movq %rbx, %rdi callq 0x9b5958 movl 0x38(%r14), %eax movl %eax, 0x38(%rbx) movl 0x3c(%r14), %eax movl %eax, 0x3c(%rbx) movl 0x40(%r14), %eax movl %eax, 0x40(%rbx) movq 0x20(%rbx), %rax movq 0x28(%rbx), %rcx subq %rax, %rcx shrq...
/kaydenl[P]VK-GL-CTS/framework/common/tcuTexture.cpp
van_kampen::Graph::mergeNodes(int, int, std::unordered_set<int, std::hash<int>, std::equal_to<int>, std::allocator<int>> const&)
void Graph::mergeNodes(nodeId_t alive, nodeId_t dead, const std::unordered_set<nodeId_t> &untouchable) { for (Transition &edgeFromDead : node(dead).transitions()) { if (untouchable.count(edgeFromDead.to)) { continue; } node(alive).addTransition(edgeFromDead.to, edgeFr...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x18, %rsp movq %rcx, %r14 movl %esi, %ebp movq %rdi, %rbx movl %edx, 0xc(%rsp) movl %edx, %esi callq 0x4c996 movq 0x20(%rax), %r15 movq 0x30(%rax), %r12 movq 0x38(%rax), %rcx movq %rcx, 0x10(%rsp) movq 0x40(%rax), %r13 cmpq %r13, %r15 je 0x4cd93 m...
/Glebanister[P]van_kampen_diagramm/src/Graph.cpp
cxxopts::values::abstract_value<unsigned long>::implicit_value(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>> const&)
std::shared_ptr<Value> implicit_value(const std::string& value) override { m_implicit = true; m_implicit_value = value; return shared_from_this(); }
pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq %rdi, %r14 movb $0x1, 0x31(%rsi) leaq 0x58(%rsi), %rdi movq %rdx, %rsi callq 0x3e200 addq $0x8, %rbx movq %r14, %rdi movq %rbx, %rsi callq 0x5970a movq %r14, %rax addq $0x8, %rsp popq %rbx popq %r14 retq
/Glebanister[P]van_kampen_diagramm/extern/cxxopts/include/cxxopts.hpp
cxxopts::values::abstract_value<unsigned long>::abstract_value(cxxopts::values::abstract_value<unsigned long> const&)
abstract_value(const abstract_value& rhs) { if (rhs.m_result) { m_result = std::make_shared<T>(); m_store = m_result.get(); } else { m_store = rhs.m_store; } m_default = rhs.m_default; m_implicit = rhs.m_implicit; ...
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x10, %rsp movq %rsi, %r15 movq %rdi, %rbx xorps %xmm0, %xmm0 movups %xmm0, 0x8(%rdi) leaq 0x1b5e5(%rip), %rax # 0x77ae8 movq %rax, (%rdi) movups %xmm0, 0x18(%rdi) andw $0x0, 0x30(%rdi) leaq 0x48(%rdi), %rax movq %rax, 0x38(%rdi) andq $0x0, 0x40(%rdi) lea...
/Glebanister[P]van_kampen_diagramm/extern/cxxopts/include/cxxopts.hpp
van_kampen::LargeFirstAlgorithm::generate(std::vector<std::vector<van_kampen::GroupElement, std::allocator<van_kampen::GroupElement>>, std::allocator<std::vector<van_kampen::GroupElement, std::allocator<van_kampen::GroupElement>>>> const&)
void LargeFirstAlgorithm::generate(const std::vector<std::vector<GroupElement>> &words) { std::size_t totalIterations = words.size(); if (cellsLimit) { totalIterations = std::min(totalIterations, cellsLimit); } ProcessLogger logger(totalIterations, std::clog, "Relations used", quiet); s...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x108, %rsp # imm = 0x108 movq %rsi, %r14 movq %rdi, %r13 movq 0x8(%rsi), %rax subq (%rsi), %rax pushq $0x18 popq %rbx cqto idivq %rbx movq 0x18(%rdi), %rcx cmpq %rax, %rcx movq %rax, %r15 cmovbq %rcx, %r15 testq %rcx, %rcx cmoveq %rax, ...
/Glebanister[P]van_kampen_diagramm/src/LargeFirstAlgorithm.cpp
van_kampen::MergingAlgorithm::MergingAlgorithm()
MergingAlgorithm::MergingAlgorithm() : result_(graph_) {}
pushq %r15 pushq %r14 pushq %rbx subq $0x10, %rsp movq %rdi, %rbx callq 0x5ce98 leaq 0x1a340(%rip), %rax # 0x77cd8 movq %rax, (%rbx) andq $0x0, 0x18(%rbx) movb $0x0, 0x20(%rbx) leaq 0x28(%rbx), %r14 leaq 0x8(%rbx), %rsi movq %rsp, %r15 movq %r15, %rdi callq 0x4f9ce movq %r14, %rdi movq %r15, %rsi callq 0x4e9ae leaq...
/Glebanister[P]van_kampen_diagramm/src/MergingAlgorithm.cpp
generate_anchors(int, ncnn::Mat const&, ncnn::Mat const&)
static ncnn::Mat generate_anchors(int base_size, const ncnn::Mat& ratios, const ncnn::Mat& scales) { int num_ratio = ratios.w; int num_scale = scales.w; ncnn::Mat anchors; anchors.create(4, num_ratio * num_scale); const float cx = 0; const float cy = 0; for (int i = 0; i < num_ratio; i++)...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp movq %rdi, %rbx movq %rdx, 0x10(%rsp) movl 0x2c(%rdx), %r13d movq %rcx, 0x18(%rsp) movslq 0x2c(%rcx), %r12 andq $0x0, 0x40(%rdi) movl %esi, %r14d movl %r12d, %edx imull %r13d, %edx xorps %xmm0, %xmm0 movups %xmm0, (%rdi) movups %xmm0, 0x...
/csukuangfj[P]ncnn/examples/scrfd_crowdhuman.cpp
ncnn::Mat::create(int, int, int, unsigned long, int, ncnn::Allocator*)
void Mat::create(int _w, int _h, int _c, size_t _elemsize, int _elempack, Allocator* _allocator) { if (dims == 3 && w == _w && h == _h && c == _c && elemsize == _elemsize && elempack == _elempack && allocator == _allocator) return; release(); elemsize = _elemsize; elempack = _elempack; all...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movl %r9d, %r14d movq %r8, %r15 movl %ecx, %ebp movl %edx, %r12d movl %esi, %r13d movq %rdi, %rbx movq 0x40(%rsp), %rdi cmpl $0x3, 0x28(%rbx) jne 0x62940 cmpl %r13d, 0x2c(%rbx) jne 0x62940 cmpl %r12d, 0x30(%rbx) jne 0x62940 cmpl %ebp, 0x38(%rb...
/csukuangfj[P]ncnn/src/mat.cpp
ncnn::Mat::reshape(int, int, ncnn::Allocator*) const
Mat Mat::reshape(int _w, int _h, Allocator* _allocator) const { if (w * h * d * c != _w * _h) return Mat(); if (dims >= 3 && cstep != (size_t)w * h * d) { Mat m; m.create(_w, _h, elemsize, elempack, _allocator); // flatten for (int i = 0; i < c; i++) { ...
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx movq %r8, %r9 movq %rsi, %r14 movq %rdi, %rbx movl 0x2c(%rsi), %eax movl 0x30(%rsi), %esi movl %esi, %r10d imull %eax, %r10d movl 0x34(%r14), %edi movl 0x38(%r14), %r8d movl %edi, %r11d imull %r8d, %r11d imull %r10d, %r11d movl %ecx, %r10d imull %edx, %r10d cmpl %r...
/csukuangfj[P]ncnn/src/mat.cpp
ncnn::Mat::reshape(int, int, int, ncnn::Allocator*) const
Mat Mat::reshape(int _w, int _h, int _c, Allocator* _allocator) const { if (w * h * d * c != _w * _h * _c) return Mat(); if (dims < 3) { if ((size_t)_w * _h != alignSize((size_t)_w * _h * elemsize, 16) / elemsize) { Mat m; m.create(_w, _h, _c, elemsize, elemp...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x68, %rsp movq %r9, %r13 movl %edx, %r12d movq %rdi, %rbx movl 0x2c(%rsi), %edi movl 0x30(%rsi), %r9d movl %r9d, %eax imull %edi, %eax movl 0x34(%rsi), %r10d movl 0x38(%rsi), %r11d movl %r10d, %edx imull %r11d, %edx imull %eax, %edx movl %ecx, %ea...
/csukuangfj[P]ncnn/src/mat.cpp
ncnn::Blob::~Blob()
class NCNN_EXPORT Blob { public: // empty Blob(); public: #if NCNN_STRING // blob name std::string name; #endif // NCNN_STRING // layer index which produce this blob as output int producer; // layer index which need this blob as input int consumer; // shape hint Mat shape; }
pushq %rbx movq %rdi, %rbx movq 0x30(%rdi), %rax testq %rax, %rax je 0x6ff89 lock decl (%rax) jne 0x6ff89 movq 0x28(%rbx), %rsi movq 0x48(%rbx), %rdi testq %rdi, %rdi je 0x6ff81 movq (%rdi), %rax callq *0x18(%rax) jmp 0x6ff89 movq %rsi, %rdi callq 0x5f3e0 leaq 0x28(%rbx), %rax andq $0x0, 0x68(%rbx) xorps %xmm0, %xmm0 m...
/csukuangfj[P]ncnn/src/blob.h
ncnn::ParamDict::ParamDict(ncnn::ParamDict const&)
ParamDict::ParamDict(const ParamDict& rhs) : d(new ParamDictPrivate) { for (int i = 0; i < NCNN_MAX_PARAM_COUNT; i++) { int type = rhs.d->params[i].type; d->params[i].type = type; if (type == 1 || type == 2 || type == 3) { d->params[i].i = rhs.d->params[i].i; ...
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx movq %rsi, %rbx movq %rdi, %r14 leaq 0x3ecc1e(%rip), %rax # 0x45e288 movq %rax, (%rdi) movl $0xa00, %edi # imm = 0xA00 callq 0x5f3d0 movq %rax, %r15 movq %rax, %rdi callq 0x72028 movq %r15, 0x8(%r14) xorl %r15d, %r15d xorps %xmm1, %xmm1 cmpq $0xa00, %...
/csukuangfj[P]ncnn/src/paramdict.cpp
ncnn::vstr_is_float(char const*)
static bool vstr_is_float(const char vstr[16]) { // look ahead for determine isfloat for (int j = 0; j < 16; j++) { if (vstr[j] == '\0') break; if (vstr[j] == '.' || tolower(vstr[j]) == 'e') return true; } return false; }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx xorl %r14d, %r14d cmpq $0x10, %r14 je 0x71d26 movzbl (%rbx,%r14), %eax testl %eax, %eax je 0x71d26 cmpl $0x2e, %eax je 0x71d22 movsbl %al, %edi callq 0x5f500 incq %r14 cmpl $0x65, %eax jne 0x71cfe movb $0x1, %al jmp 0x71d28 xorl %eax, %eax addq $0x8, %rsp popq %rbx popq ...
/csukuangfj[P]ncnn/src/paramdict.cpp
ncnn::cpu_support_x86_fma()
int cpu_support_x86_fma() { try_initialize_global_cpu_info(); #if defined(__i386__) || defined(__x86_64__) || defined(_M_IX86) || defined(_M_X64) return g_cpu_support_x86_fma; #else return 0; #endif }
pushq %rax callq 0x72ac4 movl 0x41d187(%rip), %eax # 0x490460 popq %rcx retq
/csukuangfj[P]ncnn/src/cpu.cpp
ncnn::cpu_support_x86_xop()
int cpu_support_x86_xop() { try_initialize_global_cpu_info(); #if defined(__i386__) || defined(__x86_64__) || defined(_M_IX86) || defined(_M_X64) return g_cpu_support_x86_xop; #else return 0; #endif }
pushq %rax callq 0x72ac4 movl 0x41d17d(%rip), %eax # 0x490464 popq %rcx retq
/csukuangfj[P]ncnn/src/cpu.cpp
ncnn::cpu_support_riscv_v()
int cpu_support_riscv_v() { try_initialize_global_cpu_info(); #if defined __ANDROID__ || defined __linux__ #if __riscv return g_hwcaps & COMPAT_HWCAP_ISA_V; #else return 0; #endif #else return 0; #endif }
pushq %rax callq 0x72ac4 xorl %eax, %eax popq %rcx retq
/csukuangfj[P]ncnn/src/cpu.cpp
ncnn::test_prefer_winograd63(int, int, int, int)
static bool test_prefer_winograd63(int num_input, int num_output, int w, int h) { // winograd selection strategy (profiled on i7-7700 single thread) int minwh = std::min(w, h); if (num_input >= 64) { return false; } if (num_input >= 32) { if (num_output >= 64) return false;...
cmpl %edx, %ecx cmovll %ecx, %edx cmpl $0x3f, %edi jle 0x9ba1d xorl %eax, %eax retq cmpl $0x20, %edi jl 0x9ba66 cmpl $0x3f, %esi jg 0x9ba1a cmpl $0x20, %esi jl 0x9baed movd %edx, %xmm0 pshufd $0x0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] paddd 0x35614f(%rip), %xmm0 # 0x3f1b90 pxor 0x352647(%rip), %xmm0 # 0x3ee090 ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_x86.cpp
ncnn::test_prefer_winograd23(int, int, int, int)
static bool test_prefer_winograd23(int num_input, int num_output, int w, int h) { int minwh = std::min(w, h); if (num_input >= 512) { if (num_output >= 512) return (minwh >= 3 && minwh <= 14); if (num_output >= 256) return (minwh >= 3 && minwh <= 14); if (num_output >= 128) return (...
cmpl %edx, %ecx cmovll %ecx, %edx cmpl $0x200, %edi # imm = 0x200 jl 0x9bc47 cmpl $0x200, %esi # imm = 0x200 jge 0x9bc83 cmpl $0x100, %esi # imm = 0x100 jge 0x9bc83 cmpl $0x80, %esi jge 0x9bc83 cmpl $0x40, %esi jge 0x9bcbe jmp 0x9bd01 cmpl $0x100, %edi # imm = 0x100 jl 0x9bc7...
/csukuangfj[P]ncnn/src/layer/x86/convolution_x86.cpp
ncnn::Convolution_x86::destroy_pipeline(ncnn::Option const&)
int Convolution_x86::destroy_pipeline(const Option& opt) { if (activation) { activation->destroy_pipeline(opt); delete activation; activation = 0; } if (convolution_dilation1) { convolution_dilation1->destroy_pipeline(opt); delete convolution_dilation1; ...
pushq %r14 pushq %rbx pushq %rax movq %rsi, %r14 movq %rdi, %rbx movq 0x8(%rdi), %rdi testq %rdi, %rdi je 0x9bf50 movq (%rdi), %rax movq %r14, %rsi callq *0x28(%rax) movq 0x8(%rbx), %rdi testq %rdi, %rdi je 0x9bf4b movq (%rdi), %rax callq *0x8(%rax) andq $0x0, 0x8(%rbx) movq 0x180(%rbx), %rdi testq %rdi, %rdi je 0x9bf7...
/csukuangfj[P]ncnn/src/layer/x86/convolution_x86.cpp
virtual thunk to ncnn::Convolution_x86::destroy_pipeline(ncnn::Option const&)
int Convolution_x86::destroy_pipeline(const Option& opt) { if (activation) { activation->destroy_pipeline(opt); delete activation; activation = 0; } if (convolution_dilation1) { convolution_dilation1->destroy_pipeline(opt); delete convolution_dilation1; ...
pushq %rax movq (%rdi), %rax addq -0x38(%rax), %rdi callq 0x9bf20 xorl %eax, %eax popq %rcx retq nop
/csukuangfj[P]ncnn/src/layer/x86/convolution_x86.cpp
ncnn::Convolution_x86::forward_int8_x86(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Convolution_x86::forward_int8_x86(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int elembits = bottom_blob.elembits(); Mat bottom_blob_int8 = bottom_blob; if (elembits != 8) { Option opt_q = opt; opt_q.blob_allocator = opt.workspace_allocator; quantize_to...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x6f8, %rsp # imm = 0x6F8 movq %rcx, %r14 movq %rdx, %rbx movq %rdi, 0x20(%rsp) movl 0x18(%rsi), %ecx movq 0x10(%rsi), %rdi testl %ecx, %ecx je 0xa62eb leal (,%rdi,8), %eax cltd idivl %ecx cmpl $0x8, %eax sete %al jmp 0xa62ed xorl %eax, ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_x86.cpp
ncnn::im2col_sgemm_pack8to1_int8_sse(ncnn::Mat const&, ncnn::Mat&, ncnn::Mat const&, ncnn::Option const&)
static void im2col_sgemm_pack8to1_int8_sse(const Mat& bottom_im2col, Mat& top_blob, const Mat& kernel, const Option& opt) { #if !(__AVX512VNNI__ || __AVXVNNI__ || __AVX2__ || __XOP__) #if NCNN_RUNTIME_CPU && NCNN_AVX512VNNI && __AVX512F__ && !__AVX512VNNI__ if (ncnn::cpu_support_x86_avx512_vnni()) { im2...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x88, %rsp movq %rcx, %r13 movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r12 callq 0x732db testl %eax, %eax je 0xb3945 movq %r12, %rdi movq %r14, %rsi movq %rbx, %rdx movq %r13, %rcx addq $0x88, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to1_int8.h
ncnn::im2col_sgemm_int8_sse(ncnn::Mat const&, ncnn::Mat&, ncnn::Mat const&, ncnn::Option const&)
static void im2col_sgemm_int8_sse(const Mat& bottom_im2col, Mat& top_blob, const Mat& kernel, const Option& opt) { #if !(__AVX512VNNI__ || __AVXVNNI__ || __AVX2__ || __XOP__) #if NCNN_RUNTIME_CPU && NCNN_AVX512VNNI && __AVX512F__ && !__AVX512VNNI__ if (ncnn::cpu_support_x86_avx512_vnni()) { im2col_sgemm...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rcx, %r13 movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r15 callq 0x732db testl %eax, %eax je 0xb413a movq %r15, %rdi movq %r14, %rsi movq %rbx, %rdx movq %r13, %rcx addq $0xa8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_int8.h
ncnn::convolution_im2col_sgemm_transform_kernel_pack8to1_int8_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int, int)
static void convolution_im2col_sgemm_transform_kernel_pack8to1_int8_sse(const Mat& _kernel, Mat& kernel_tm, int inch, int outch, int kernel_w, int kernel_h) { const int maxk = kernel_w * kernel_h; // interleave // src = maxk-inch-outch // dst = 8a-4b-maxk-inch/8a-outch/4b Mat kernel = _kernel.resha...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movl %r8d, %ebx movl %ecx, %ebp movl %edx, %r15d movq %rsi, %r14 movq %rdi, %rsi imull %r9d, %ebx movq %rsp, %rdi movl %ebx, %edx movl %r15d, %ecx movl %ebp, %r8d xorl %r9d, %r9d callq 0x63020 pushq $0x8 popq %rcx movl %r15d, %eax cltd i...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to1_int8.h
ncnn::Convolution_x86_fma::create_pipeline(ncnn::Option const&)
int Convolution_x86_fma::create_pipeline(const Option& opt) { if (dynamic_weight) return 0; activation = create_activation_layer(activation_type, activation_params, opt); nT = opt.num_threads; #if NCNN_INT8 if (opt.use_int8_inference && weight_data.elemsize == (size_t)1u) { return ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x228, %rsp # imm = 0x228 movq (%rdi), %rax movq -0x18(%rax), %rbx cmpl $0x0, 0x158(%rdi,%rbx) je 0xe34e0 xorl %eax, %eax addq $0x228, %rsp # imm = 0x228 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %r...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
ncnn::conv3x3s1_winograd43_transform_kernel(ncnn::Mat const&, ncnn::Mat&, int, int, ncnn::Option const&)
static void conv3x3s1_winograd43_transform_kernel(const Mat& kernel, Mat& AT, int inch, int outch, const Option& opt) { const int M = outch; const int K = inch; const int B = 36; int TILE_M, TILE_N, TILE_K; get_optimal_tile_mnk(M, 0, K, TILE_M, TILE_N, TILE_K, opt.num_threads); const int nn_M ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x148, %rsp # imm = 0x148 movq %r8, %r13 movl %ecx, %r14d movl %edx, %r15d movq %rsi, 0x18(%rsp) movq %rdi, %r12 movl 0x4(%r8), %eax movl %eax, (%rsp) leaq 0xac(%rsp), %rbx leaq 0xa8(%rsp), %r8 leaq 0xa4(%rsp), %rbp movl %ecx, %edi xorl ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_3x3_winograd.h
ncnn::conv3x3s1_winograd23_transform_kernel(ncnn::Mat const&, ncnn::Mat&, int, int, ncnn::Option const&)
static void conv3x3s1_winograd23_transform_kernel(const Mat& kernel, Mat& AT, int inch, int outch, const Option& opt) { const int M = outch; const int K = inch; const int B = 16; int TILE_M, TILE_N, TILE_K; get_optimal_tile_mnk(M, 0, K, TILE_M, TILE_N, TILE_K, opt.num_threads); const int nn_M ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x158, %rsp # imm = 0x158 movq %r8, %r13 movl %ecx, %r14d movl %edx, %r15d movq %rsi, 0x28(%rsp) movq %rdi, %r12 movl 0x4(%r8), %eax movl %eax, (%rsp) leaq 0xbc(%rsp), %rbx leaq 0xb8(%rsp), %r8 leaq 0xb4(%rsp), %rbp movl %ecx, %edi xorl ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_3x3_winograd.h
ncnn::test_prefer_winograd63(int, int, int, int)
static bool test_prefer_winograd63(int num_input, int num_output, int w, int h) { // winograd selection strategy (profiled on i7-7700 single thread) int minwh = std::min(w, h); if (num_input >= 64) { return false; } if (num_input >= 32) { if (num_output >= 64) return false;...
cmpl %edx, %ecx cmovll %ecx, %edx cmpl $0x3f, %edi jle 0xe8733 xorl %eax, %eax retq cmpl $0x20, %edi jl 0xe8778 cmpl $0x3f, %esi jg 0xe8730 cmpl $0x20, %esi jl 0xe87f7 vmovd %edx, %xmm0 vpshufd $0x0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vpaddd 0x309439(%rip), %xmm0, %xmm0 # 0x3f1b90 vpminud 0x30a150(%rip), %xmm0, %...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
ncnn::test_prefer_winograd23(int, int, int, int)
static bool test_prefer_winograd23(int num_input, int num_output, int w, int h) { int minwh = std::min(w, h); if (num_input >= 512) { if (num_output >= 512) return (minwh >= 3 && minwh <= 14); if (num_output >= 256) return (minwh >= 3 && minwh <= 14); if (num_output >= 128) return (...
cmpl %edx, %ecx cmovll %ecx, %edx cmpl $0x200, %edi # imm = 0x200 jl 0xe894d cmpl $0x200, %esi # imm = 0x200 jge 0xe8989 cmpl $0x100, %esi # imm = 0x100 jge 0xe8989 cmpl $0x80, %esi jge 0xe8989 cmpl $0x40, %esi jge 0xe89c4 jmp 0xe8a07 cmpl $0x100, %edi # imm = 0x100 jl 0xe897...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
ncnn::Convolution_x86_fma::forward_int8_x86(ncnn::Mat const&, ncnn::Mat&, ncnn::Option const&) const
int Convolution_x86_fma::forward_int8_x86(const Mat& bottom_blob, Mat& top_blob, const Option& opt) const { int elembits = bottom_blob.elembits(); Mat bottom_blob_int8 = bottom_blob; if (elembits != 8) { Option opt_q = opt; opt_q.blob_allocator = opt.workspace_allocator; quantiz...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x6e8, %rsp # imm = 0x6E8 movq %rcx, %r14 movq %rdx, %rbx movq %rdi, 0x20(%rsp) movl 0x18(%rsi), %ecx movq 0x10(%rsi), %rdi testl %ecx, %ecx je 0xf8f2b leal (,%rdi,8), %eax cltd idivl %ecx cmpl $0x8, %eax sete %al jmp 0xf8f2d xorl %eax, ...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
ncnn::Convolution_x86_fma::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Convolution_x86_fma::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const Mat& _weight_data = bottom_blobs[1]; Mat& top_blob = top_blobs[0]; const int _kernel_w = _weight_data.w; const int _kernel_h...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x168, %rsp # imm = 0x168 movq %rsi, %rbp movq %rdi, %r13 movq (%rsi), %r14 leaq 0x48(%r14), %rdi movq (%rdx), %rax movq %rax, 0xb8(%rsp) movl 0x60(%r14), %ebx movl 0x74(%r14), %eax movl %eax, 0x1c(%rsp) movl 0x78(%r14), %eax movl %eax, ...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
virtual thunk to ncnn::Convolution_x86_fma::forward(std::vector<ncnn::Mat, std::allocator<ncnn::Mat>> const&, std::vector<ncnn::Mat, std::allocator<ncnn::Mat>>&, ncnn::Option const&) const
int Convolution_x86_fma::forward(const std::vector<Mat>& bottom_blobs, std::vector<Mat>& top_blobs, const Option& opt) const { const Mat& bottom_blob = bottom_blobs[0]; const Mat& _weight_data = bottom_blobs[1]; Mat& top_blob = top_blobs[0]; const int _kernel_w = _weight_data.w; const int _kernel_h...
movq (%rdi), %rax addq -0x40(%rax), %rdi jmp 0x100126
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
ncnn::convolution_im2col_sgemm_transform_kernel_pack8to4_int8_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int, int)
static void convolution_im2col_sgemm_transform_kernel_pack8to4_int8_sse(const Mat& _kernel, Mat& kernel_tm, int inch, int outch, int kernel_w, int kernel_h) { const int maxk = kernel_w * kernel_h; // interleave // src = maxk-inch-outch // dst = 8a-4b-maxk-inch/8a-outch/4b Mat kernel = _kernel.resha...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movl %r8d, %ebx movl %ecx, %ebp movl %edx, %r15d movq %rsi, %r14 movq %rdi, %rsi imull %r9d, %ebx xorl %r12d, %r12d movq %rsp, %rdi movl %ebx, %edx movl %r15d, %ecx movl %ebp, %r8d xorl %r9d, %r9d callq 0x63020 movl %ebx, %esi shll $0x5,...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to4_int8.h
ncnn::convolution_transform_kernel_packed_int8_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int, int, int, int)
static void convolution_transform_kernel_packed_int8_sse(const Mat& weight_data, Mat& weight_data_tm, int num_input, int num_output, int kernel_w, int kernel_h, int elempack, int out_elempack) { const int maxk = kernel_w * kernel_h; // src = kw-kh-inch-outch // dst = pa-pb-kw-kh-inch/pa-outch/pb { ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x78, %rsp movl %r8d, %ebx movl %ecx, %r12d movl %edx, %r13d movq %rsi, %r14 movq %rdi, %rsi movl 0xb0(%rsp), %ebp imull %r9d, %ebx xorl %r15d, %r15d leaq 0x30(%rsp), %rdi movl %ebx, %edx movl %r13d, %ecx movl %r12d, %r8d xorl %r9d, %r9d callq 0x63...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_fma.cpp
ncnn::convolution_im2col_sgemm_transform_kernel_pack8to1_int8_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int, int)
static void convolution_im2col_sgemm_transform_kernel_pack8to1_int8_sse(const Mat& _kernel, Mat& kernel_tm, int inch, int outch, int kernel_w, int kernel_h) { const int maxk = kernel_w * kernel_h; // interleave // src = maxk-inch-outch // dst = 8a-4b-maxk-inch/8a-outch/4b Mat kernel = _kernel.resha...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movl %r8d, %ebx movl %ecx, %ebp movl %edx, %r15d movq %rsi, %r14 movq %rdi, %rsi imull %r9d, %ebx movq %rsp, %rdi movl %ebx, %edx movl %r15d, %ecx movl %ebp, %r8d xorl %r9d, %r9d callq 0x63020 pushq $0x8 popq %rcx movl %r15d, %eax cltd i...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to1_int8.h
ncnn::convolution_im2col_sgemm_transform_kernel_int8_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int, int)
static void convolution_im2col_sgemm_transform_kernel_int8_sse(const Mat& _kernel, Mat& kernel_tm, int inch, int outch, int kernel_w, int kernel_h) { const int maxk = kernel_w * kernel_h; #if __SSE2__ // interleave // src = maxk-inch-outch // dst = 4a-4b-maxk-inch/4a-outch/4b Mat kernel = _kernel.r...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movl %r8d, %ebx movl %ecx, %ebp movl %edx, %r15d movq %rsi, %r14 movq %rdi, %rsi imull %r9d, %ebx movq %rsp, %rdi movl %ebx, %edx movl %r15d, %ecx movl %ebp, %r8d xorl %r9d, %r9d callq 0x63020 cmpl $0x4, %ebp jl 0x1010a9 cmpl $0x4, %r15d...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_int8.h
ncnn::get_optimal_tile_mnk(int, int, int, int&, int&, int&, int)
static void get_optimal_tile_mnk(int M, int N, int K, int& TILE_M, int& TILE_N, int& TILE_K, int nT) { // resolve optimal tile size from cache size const size_t l2_cache_size = get_cpu_level2_cache_size(); if (nT == 0) nT = get_physical_big_cpu_count(); // solve M { int tile_size =...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp movq %r9, 0x20(%rsp) movq %r8, 0x18(%rsp) movq %rcx, %r15 movl %edx, %r13d movq %rsi, 0x10(%rsp) movl %edi, %ebx movl 0x60(%rsp), %ebp callq 0x73479 movl %eax, %r12d movslq %eax, %r14 testl %ebp, %ebp jne 0x101710 callq 0x7345d movl %eax...
/csukuangfj[P]ncnn/src/layer/x86/convolution_3x3_winograd.h
ncnn::pack_A_tile(ncnn::Mat const&, ncnn::Mat&, int, int, int)
static void pack_A_tile(const Mat& A, Mat& AT, int batch, int max_ii, int max_kk) { const int N = max_kk * batch; for (int b = 0; b < batch; b++) { float* pp = AT.row(b); int ii = 0; #if __SSE2__ #if __AVX__ #if __AVX512F__ for (; ii + 15 < max_ii; ii += 16) { c...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx movl %r8d, %r10d movq %rdi, -0x68(%rsp) movl %r8d, %eax imull %edx, %eax movq (%rsi), %r8 movq %r8, -0x30(%rsp) movslq 0x2c(%rsi), %r8 imulq 0x10(%rsi), %r8 movq %r8, -0x38(%rsp) movslq %eax, %r11 leal (%r11,%r11), %eax movslq %eax, %r14 leal (%r11,%r11,...
/csukuangfj[P]ncnn/src/layer/x86/convolution_3x3_winograd.h
ncnn::im2col_sgemm_pack8to4_int8_sse(ncnn::Mat const&, ncnn::Mat&, ncnn::Mat const&, ncnn::Option const&)
static void im2col_sgemm_pack8to4_int8_sse(const Mat& bottom_im2col, Mat& top_blob, const Mat& kernel, const Option& opt) { #if !(__AVX512VNNI__ || __AVXVNNI__ || __AVX2__ || __XOP__) #if NCNN_RUNTIME_CPU && NCNN_AVX512VNNI && __AVX512F__ && !__AVX512VNNI__ if (ncnn::cpu_support_x86_avx512_vnni()) { im2...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x78, %rsp movq %rcx, %r13 movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r12 callq 0x732f7 testl %eax, %eax je 0x107af0 movq %r12, %rdi movq %r14, %rsi movq %rbx, %rdx movq %r13, %rcx addq $0x78, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to4_int8.h
ncnn::im2col_sgemm_pack8to1_int8_sse(ncnn::Mat const&, ncnn::Mat&, ncnn::Mat const&, ncnn::Option const&)
static void im2col_sgemm_pack8to1_int8_sse(const Mat& bottom_im2col, Mat& top_blob, const Mat& kernel, const Option& opt) { #if !(__AVX512VNNI__ || __AVXVNNI__ || __AVX2__ || __XOP__) #if NCNN_RUNTIME_CPU && NCNN_AVX512VNNI && __AVX512F__ && !__AVX512VNNI__ if (ncnn::cpu_support_x86_avx512_vnni()) { im2...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x88, %rsp movq %rcx, %r13 movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r12 callq 0x732f7 testl %eax, %eax je 0x1086a3 movq %r12, %rdi movq %r14, %rsi movq %rbx, %rdx movq %r13, %rcx addq $0x88, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to1_int8.h
ncnn::im2col_sgemm_int8_sse(ncnn::Mat const&, ncnn::Mat&, ncnn::Mat const&, ncnn::Option const&)
static void im2col_sgemm_int8_sse(const Mat& bottom_im2col, Mat& top_blob, const Mat& kernel, const Option& opt) { #if !(__AVX512VNNI__ || __AVXVNNI__ || __AVX2__ || __XOP__) #if NCNN_RUNTIME_CPU && NCNN_AVX512VNNI && __AVX512F__ && !__AVX512VNNI__ if (ncnn::cpu_support_x86_avx512_vnni()) { im2col_sgemm...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rcx, %r13 movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r15 callq 0x732f7 testl %eax, %eax je 0x108ddb movq %r15, %rdi movq %r14, %rsi movq %rbx, %rdx movq %r13, %rcx addq $0xa8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_int8.h
ncnn::Convolution_x86_avx::Convolution_x86_avx()
Convolution_x86_avx::Convolution_x86_avx() { #if __SSE2__ support_packing = true; #endif // __SSE2__ activation = 0; nT = 0; convolution_dilation1 = 0; gemm = 0; }
movq (%rsi), %rax movq %rax, (%rdi) movq 0x8(%rsi), %rcx movq -0x18(%rax), %rax movq %rcx, (%rdi,%rax) andq $0x0, 0x178(%rdi) andq $0x0, 0x1d0(%rdi) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x18(%rdi) vmovups %xmm0, 0x24(%rdi) vmovups %xmm0, 0x38(%rdi) vmovups %xmm0, 0x44(%rdi) vxorps %xmm1, %xmm1, %xmm1 andl $0x0, 0x...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_avx.cpp
ncnn::Convolution_x86_avx::Convolution_x86_avx()
Convolution_x86_avx::Convolution_x86_avx() { #if __SSE2__ support_packing = true; #endif // __SSE2__ activation = 0; nT = 0; convolution_dilation1 = 0; gemm = 0; }
pushq %rbx movq %rdi, %rbx addq $0x1d8, %rdi # imm = 0x1D8 callq 0x94b00 leaq 0x377a8f(%rip), %rax # 0x481318 movq %rax, (%rbx) leaq 0x377b0d(%rip), %rax # 0x4813a0 movq %rax, 0x1d8(%rbx) andq $0x0, 0x178(%rbx) andq $0x0, 0x1d0(%rbx) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x18(%rbx) vmovups %xmm0, 0...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_avx.cpp
ncnn::Convolution_x86_avx::create_pipeline(ncnn::Option const&)
int Convolution_x86_avx::create_pipeline(const Option& opt) { if (dynamic_weight) return 0; activation = create_activation_layer(activation_type, activation_params, opt); nT = opt.num_threads; #if NCNN_INT8 if (opt.use_int8_inference && weight_data.elemsize == (size_t)1u) { return ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x228, %rsp # imm = 0x228 movq (%rdi), %rax movq -0x18(%rax), %rbx cmpl $0x0, 0x158(%rdi,%rbx) je 0x1099b0 xorl %eax, %eax addq $0x228, %rsp # imm = 0x228 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_avx.cpp
ncnn::Convolution_x86_avx::create_pipeline_int8_x86(ncnn::Option const&)
int Convolution_x86_avx::create_pipeline_int8_x86(const Option& opt) { const int maxk = kernel_w * kernel_h; const int num_input = weight_data_size / maxk / num_output; int elempack = 1; int out_elempack = 1; #if __SSE2__ if (opt.use_packing_layout) { elempack = num_input % 8 == 0 ? 8 :...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x108, %rsp # imm = 0x108 movq %rdi, %r14 movq (%rdi), %rax movq -0x18(%rax), %r12 movl 0xd4(%rdi,%r12), %r8d movl 0xd8(%rdi,%r12), %r9d movl %r9d, %ecx imull %r8d, %ecx movl 0x104(%rdi,%r12), %eax cltd idivl %ecx movslq 0xd0(%rdi,%r12),...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_avx.cpp
ncnn::conv3x3s1_winograd23_transform_kernel(ncnn::Mat const&, ncnn::Mat&, int, int, ncnn::Option const&)
static void conv3x3s1_winograd23_transform_kernel(const Mat& kernel, Mat& AT, int inch, int outch, const Option& opt) { const int M = outch; const int K = inch; const int B = 16; int TILE_M, TILE_N, TILE_K; get_optimal_tile_mnk(M, 0, K, TILE_M, TILE_N, TILE_K, opt.num_threads); const int nn_M ...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x158, %rsp # imm = 0x158 movq %r8, %r13 movl %ecx, %r14d movl %edx, %r15d movq %rsi, 0x28(%rsp) movq %rdi, %r12 movl 0x4(%r8), %eax movl %eax, (%rsp) leaq 0xbc(%rsp), %rbx leaq 0xb8(%rsp), %r8 leaq 0xb4(%rsp), %rbp movl %ecx, %edi xorl ...
/csukuangfj[P]ncnn/src/layer/x86/convolution_3x3_winograd.h
ncnn::test_prefer_winograd63(int, int, int, int)
static bool test_prefer_winograd63(int num_input, int num_output, int w, int h) { // winograd selection strategy (profiled on i7-7700 single thread) int minwh = std::min(w, h); if (num_input >= 64) { return false; } if (num_input >= 32) { if (num_output >= 64) return false;...
cmpl %edx, %ecx cmovll %ecx, %edx cmpl $0x3f, %edi jle 0x10ec06 xorl %eax, %eax retq cmpl $0x20, %edi jl 0x10ec4b cmpl $0x3f, %esi jg 0x10ec03 cmpl $0x20, %esi jl 0x10ecca vmovd %edx, %xmm0 vpshufd $0x0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vpaddd 0x2e2f66(%rip), %xmm0, %xmm0 # 0x3f1b90 vpminud 0x2e3c7d(%rip), %xmm...
/csukuangfj[P]ncnn/build_O2/src/layer/x86/convolution_x86_avx.cpp
ncnn::convolution_im2col_sgemm_transform_kernel_pack8to1_int8_sse(ncnn::Mat const&, ncnn::Mat&, int, int, int, int)
static void convolution_im2col_sgemm_transform_kernel_pack8to1_int8_sse(const Mat& _kernel, Mat& kernel_tm, int inch, int outch, int kernel_w, int kernel_h) { const int maxk = kernel_w * kernel_h; // interleave // src = maxk-inch-outch // dst = 8a-4b-maxk-inch/8a-outch/4b Mat kernel = _kernel.resha...
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movl %r8d, %ebx movl %ecx, %ebp movl %edx, %r15d movq %rsi, %r14 movq %rdi, %rsi imull %r9d, %ebx movq %rsp, %rdi movl %ebx, %edx movl %r15d, %ecx movl %ebp, %r8d xorl %r9d, %r9d callq 0x63020 pushq $0x8 popq %rcx movl %r15d, %eax cltd i...
/csukuangfj[P]ncnn/src/layer/x86/convolution_sgemm_pack8to1_int8.h
ncnn::conv3x3s1_winograd43_transform_kernel_pack8to4_int8_sse_avx2(ncnn::Mat const&, ncnn::Mat&, int, int, ncnn::Option const&)
void conv3x3s1_winograd43_transform_kernel_pack8to4_int8_sse_avx2(const Mat& kernel, Mat& kernel_tm, int inch, int outch, const Option& opt) { conv3x3s1_winograd43_transform_kernel_pack8to4_int8_sse(kernel, kernel_tm, inch, outch, opt); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xd8, %rsp movl %ecx, %ebx movl %edx, %r14d movq %rsi, %r15 movq %rdi, 0x38(%rsp) leaq 0x60(%rsp), %rdi andq $0x0, 0x40(%rdi) vpxor %xmm0, %xmm0, %xmm0 vmovdqa %xmm0, (%rdi) vmovdqu %xmm0, 0xc(%rdi) vmovdqa %xmm0, 0x20(%rdi) vmovdqu %xmm0, 0x2c(%rd...
/csukuangfj[P]ncnn/src/layer/x86/convolution_x86_avx2.cpp