AVX512 version for POVRAY

avx512_build_setup/ADDITIONAL_README_FOR_AVX512_Version.md

@@ -0,0 +1,65 @@
+Notes for AVX512 Windows build
+==============================
+
+The visual studio version was updated from vs2015 to vs2022 to enable support of AVX512 Version
+1. Set the configuration in solution file to Release-AVX512 | x64
+2. Select the 'Generic POV-Ray > povbase' project and expand 'Backend Headers', then open the
+   file `build.h`(source/base/build.h) listed within it.In it replace with name
+   and email of person who builds the code in `BUILT_BY` flag and comment the #error directive (line 129)
+3. In syspovconfig.h(windows/povconfig/syspovconfig.h) uncomment the #define _CONSOLE. (line 56)
+   The AVX512 version was developed with the console version.
+   The GUI build has been skipped in the solution file.
+   **Note:** (Presently with the updated code the GUI project is skipped for building,
+   as the cmedit64.dll and povcmax64.dll from official windows distribution are
+   incompatible with VS2022. The console version alone is available to build and test).
+4. Build the solution file and in the vs2022/bin64 folder we can run the POVRAY examples with povconsole-avx512.exe.
+```
+         General command example - povconsole-avx512.exe +Ibenchmark.pov
+         Single worker thread - povconsole-avx512.exe +WT1 benchmark.pov
+         Output image - benchmark.png
+```
+5. Results with the AVX512 version has been attached in the same folder.
+
+Notes for UNIX build
+====================
+
+Dependencies for unix build
+```
+   libboost-dev
+   libboost-date-time-dev
+   libboost-thread-dev
+   libz-dev
+   libpng-dev
+   libjpeg-dev
+   libtiff-dev
+   libopenexr-dev
+   pkg-config (if its already not there)
+```
+
+Steps :
+Generating configure and building the code :
+```
+   % cd unix/
+   % ./prebuild.sh
+   % cd ../
+   % ./configure COMPILED_BY="your name <email@address>"
+   % make
+```
+
+To build with icpc :
+```
+   % source /opt/intel/oneapi/setvars.sh
+   % cd unix/
+   % ./prebuild.sh
+   % cd ../
+   % ./configure COMPILED_BY="your name <email@address>" CXX=icpc
+   % make
+```
+
+Sample commands (inside the unix folder) :
+```
+         General command example - ./povray +Ibenchmark.pov
+         Single worker thread - ./povray +WT1 benchmark.pov
+         Output image - benchmark.png
+```
+

platform/x86/avx512/avx512noise.h

@@ -0,0 +1,73 @@
+//******************************************************************************
+///
+/// @file platform/x86/avx512/avx5123noise.h
+///
+/// This file contains declarations related to implementations of the noise
+/// generator optimized for the AVX512 instruction set.
+///
+/// @copyright
+/// @parblock
+///
+/// Persistence of Vision Ray Tracer ('POV-Ray') version 3.8.
+/// Copyright 1991-2017 Persistence of Vision Raytracer Pty. Ltd.
+///
+/// POV-Ray is free software: you can redistribute it and/or modify
+/// it under the terms of the GNU Affero General Public License as
+/// published by the Free Software Foundation, either version 3 of the
+/// License, or (at your option) any later version.
+///
+/// POV-Ray is distributed in the hope that it will be useful,
+/// but WITHOUT ANY WARRANTY; without even the implied warranty of
+/// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+/// GNU Affero General Public License for more details.
+///
+/// You should have received a copy of the GNU Affero General Public License
+/// along with this program.  If not, see <http://www.gnu.org/licenses/>.
+///
+/// ----------------------------------------------------------------------------
+///
+/// POV-Ray is based on the popular DKB raytracer version 2.12.
+/// DKBTrace was originally written by David K. Buck.
+/// DKBTrace Ver 2.0-2.12 were written by David K. Buck & Aaron A. Collins.
+///
+/// @endparblock
+///
+//******************************************************************************
+
+#ifndef POVRAY_AVX512NOISE_H
+#define POVRAY_AVX512NOISE_H
+
+#include "core/configcore.h"
+#include "core/math/vector.h"
+
+#ifdef TRY_OPTIMIZED_NOISE_AVX512
+
+namespace pov
+{
+
+extern const bool kAVX512NoiseEnabled;
+void AVX512NoiseInit();
+
+/// Optimized Noise function for single input for AVX512 architecture
+DBL AVX512Noise(const Vector3d& EPoint, int noise_generator);
+
+/// Optimized DNoise function for single input for AVX512 architecture
+void AVX512DNoise(Vector3d& result, const Vector3d& EPoint);
+
+/// Optimized Noise function for two inputs using AVX512 instructions
+/// @author Optimized by MCW
+void AVX512Noise2D(const Vector3d& EPoint, int noise_generator, double& value);
+
+/// Optimized DNoise function for two inputs using AVX512 instructions
+/// @author Optimized by MCW
+void AVX512DNoise2D(Vector3d& result, const Vector3d& EPoint);
+
+/// Optimized Noise function for 8 multiples of single input using AVX512 instructions.
+/// @author Optimized by MCW
+DBL AVX512Noise8D(const Vector3d& EPoint, int noise_generator);
+
+}
+
+#endif // TRY_OPTIMIZED_NOISE_AVX512
+
+#endif // POVRAY_AVX512NOISE_H

platform/x86/cpuid.cpp

@@ -114,6 +114,7 @@ static unsigned long long getXCR0()
 #define CPUID_00000001_ECX_AVX_MASK     (0x1 << 28)
 #define CPUID_00000001_EDX_SSE2_MASK    (0x1 << 26)
 #define CPUID_00000007_EBX_AVX2_MASK    (0x1 <<  5)
+#define CPUID_00000007_EBX_AVX512_MASK  (0x1 << 16)
 #define CPUID_80000001_ECX_FMA4_MASK    (0x1 << 16)
 
 // Masks for relevant XCR0 register bits.
@@ -170,6 +171,7 @@ struct CPUIDInfo
     bool        sse2   : 1;
     bool        avx    : 1;
     bool        avx2   : 1;
+    bool        avx512 : 1;
     bool        fma3   : 1;
     bool        fma4   : 1;
 #if POV_CPUINFO_DEBUG
@@ -184,6 +186,7 @@ CPUIDInfo::CPUIDInfo() :
     sse2(false),
     avx(false),
     avx2(false),
+    avx512(false),
     fma3(false),
     fma4(false),
     vendorId(kCPUVendor_Unrecognized)
@@ -220,6 +223,11 @@ CPUIDInfo::CPUIDInfo() :
         CPUID(info, 0x7);
         avx2    = ((info[CPUID_EBX] & CPUID_00000007_EBX_AVX2_MASK)    != 0);
     }
+    if (maxLeaf >= 0x7)
+    {
+        CPUID(info, 0x7);
+        avx512 = ((info[CPUID_EBX] & CPUID_00000007_EBX_AVX512_MASK) != 0);
+    }
     CPUID(info, 0x80000000);
     int maxLeafExt = info[CPUID_EAX];
     if (maxLeafExt >= (int)0x80000001)
@@ -233,6 +241,7 @@ struct OSInfo
 {
     bool xcr0_sse : 1;
     bool xcr0_avx : 1;
+    bool xcr0_avx512 : 1;
     OSInfo(const CPUIDInfo& cpuinfo);
 };
 
@@ -278,6 +287,16 @@ bool CPUInfo::SupportsAVX()
         && gpData->osInfo.xcr0_avx;
 }
 
+bool CPUInfo::SupportsAVX512()
+{
+    return gpData->cpuidInfo.osxsave
+        && gpData->cpuidInfo.avx
+        && gpData->cpuidInfo.avx2
+        && gpData->cpuidInfo.avx512
+        && gpData->osInfo.xcr0_sse
+        && gpData->osInfo.xcr0_avx;
+}
+
 bool CPUInfo::SupportsAVX2()
 {
     return gpData->cpuidInfo.osxsave
@@ -329,6 +348,8 @@ std::string CPUInfo::GetFeatures()
         features.push_back("AVX");
     if (SupportsAVX2())
         features.push_back("AVX2");
+    if (SupportsAVX512()) 
+        features.push_back("AVX512");
     if (SupportsFMA3())
         features.push_back("FMA3");
     if (SupportsFMA4())
@@ -356,6 +377,8 @@ std::string CPUInfo::GetDetails()
         cpuidFeatures.push_back("AVX");
     if (gpData->cpuidInfo.avx2)
         cpuidFeatures.push_back("AVX2");
+    if (gpData->cpuidInfo.avx512)
+        cpuidFeatures.push_back("AVX512");
     if (gpData->cpuidInfo.fma3)
         cpuidFeatures.push_back("FMA");
     if (gpData->cpuidInfo.fma4)
@@ -371,6 +394,8 @@ std::string CPUInfo::GetDetails()
 
     if (gpData->osInfo.xcr0_avx)
         xcr0Features.push_back("AVX");
+    if (gpData->osInfo.xcr0_avx)
+        xcr0Features.push_back("AVX512");
     if (gpData->osInfo.xcr0_sse)
         xcr0Features.push_back("SSE");
 

platform/x86/cpuid.h

@@ -45,6 +45,7 @@ class CPUInfo
     static bool SupportsSSE2();             ///< Test whether CPU and OS support SSE2.
     static bool SupportsAVX();              ///< Test whether CPU and OS support AVX.
     static bool SupportsAVX2();             ///< Test whether CPU and OS support AVX2.
+    static bool SupportsAVX512();           ///< Test whether CPU and OS support AVX512.
     static bool SupportsFMA3();             ///< Test whether CPU and OS support FMA3.
     static bool SupportsFMA4();             ///< Test whether CPU and OS support FMA4.
     static bool IsIntel();                  ///< Test whether CPU is genuine Intel product.

platform/x86/optimizednoise.cpp

@@ -39,6 +39,10 @@
 
 #include "core/material/noise.h"
 
+#ifdef TRY_OPTIMIZED_NOISE_AVX512
+#include "avx512/avx512noise.h"
+#endif
+
 #ifdef TRY_OPTIMIZED_NOISE_AVX2FMA3
 #include "avx2fma3/avx2fma3noise.h"
 #endif
@@ -65,19 +69,48 @@ namespace pov
 static bool AVXSupported()      { return CPUInfo::SupportsAVX(); }
 static bool AVXFMA4Supported()  { return CPUInfo::SupportsAVX() && CPUInfo::SupportsFMA4(); }
 static bool AVX2FMA3Supported() { return CPUInfo::SupportsAVX2() && CPUInfo::SupportsFMA3(); }
+static bool AVX512Supported()   { return CPUInfo::SupportsAVX512(); }
 
 /// List of optimized noise implementations.
 ///
 /// @note
 ///     Entries must be listed in descending order of preference.
 ///
 OptimizedNoiseInfo gaOptimizedNoiseInfo[] = {
+#ifdef TRY_OPTIMIZED_NOISE_AVX512
+    {
+        "avx512-mcw",             // name,
+        "hand-optimized by MCW",  // info,
+        AVX512Noise,              // noise,
+        AVX512DNoise,             // dNoise,
+        AVX512Noise2D,            // noise2D,
+        AVX512DNoise2D,           // dNoise2D,
+        AVX512Noise8D,            // noise8D,
+        DTurbulenceAVX512,        // DTurbulence
+        Initialize_WavesAVX512,   // Initalize Waves
+        TurbulenceAVX512,         // Turbulence
+        wrinklesAVX512,           // wrinkles
+        true,                     // value to set versions of WrinklesPattern and GranitePattern
+        &kAVX512NoiseEnabled,     // enabled,
+        AVX512Supported,          // supported,
+        nullptr,                  // recommended,
+        AVX512NoiseInit           // init
+    },
+#endif
 #ifdef TRY_OPTIMIZED_NOISE_AVX2FMA3
     {
         "avx2fma3-intel",           // name,
         "hand-optimized by Intel",  // info,
         AVX2FMA3Noise,              // noise,
         AVX2FMA3DNoise,             // dNoise,
+        nullptr,                    // noise2D
+        nullptr,                    // dnoise2D,
+        nullptr,                    // noise8D,
+        DTurbulenceAVX,             // DTurbulence
+        Initialize_WavesAVX,        // Initalize Waves
+        TurbulenceAVX,              // Turbulence
+        wrinklesAVX,                // wrinkles
+        false,                      // value to set versions of WrinklesPattern and GranitePattern
         &kAVX2FMA3NoiseEnabled,     // enabled,
         AVX2FMA3Supported,          // supported,
         CPUInfo::IsIntel,           // recommended,
@@ -90,6 +123,14 @@ OptimizedNoiseInfo gaOptimizedNoiseInfo[] = {
         "hand-optimized by AMD, 2017-04 update", // info,
         AVXFMA4Noise,               // noise,
         AVXFMA4DNoise,              // dNoise,
+        nullptr,                    // noise2D
+        nullptr,                    // dnoise2D,
+        nullptr,                    // noise8D,
+        DTurbulenceAVX,             // DTurbulence
+        Initialize_WavesAVX,        // Initalize Waves
+        TurbulenceAVX,              // Turbulence
+        wrinklesAVX,                // wrinkles
+        false,                      // value to set versions of WrinklesPattern and GranitePattern
         &kAVXFMA4NoiseEnabled,      // enabled,
         AVXFMA4Supported,           // supported,
         nullptr,                    // recommended,
@@ -102,6 +143,14 @@ OptimizedNoiseInfo gaOptimizedNoiseInfo[] = {
         "hand-optimized by Intel",  // info,
         AVXNoise,                   // noise,
         AVXDNoise,                  // dNoise,
+        nullptr,                    // noise2D
+        nullptr,                    // dnoise2D,
+        nullptr,                    // noise8D,
+        DTurbulenceAVX,             // DTurbulence
+        Initialize_WavesAVX,        // Initalize Waves
+        TurbulenceAVX,              // Turbulence
+        wrinklesAVX,                // wrinkles
+        false,                      // value to set versions of WrinklesPattern and GranitePattern
         &kAVXNoiseEnabled,          // enabled,
         AVXSupported,               // supported,
         CPUInfo::IsIntel,           // recommended,
@@ -114,6 +163,14 @@ OptimizedNoiseInfo gaOptimizedNoiseInfo[] = {
         "compiler-optimized",       // info,
         AVXPortableNoise,           // noise,
         AVXPortableDNoise,          // dNoise,
+        nullptr,                    // noise2D
+        nullptr,                    // dnoise2D,
+        nullptr,                    // noise8D,
+        DTurbulenceAVX,             // DTurbulence
+        Initialize_WavesAVX,        // Initalize Waves
+        TurbulenceAVX,              // Turbulence
+        wrinklesAVX,                // wrinkles
+        false,                      // value to set versions of WrinklesPattern and GranitePattern
         &kAVXPortableNoiseEnabled,  // enabled,
         AVXSupported,               // supported,
         nullptr,                    // recommended,