mirror-github-bspeice/fractorium
1
0
Fork 0
mirror of https://bitbucket.org/mfeemster/fractorium.git synced 2025-01-21 21:20:07 -05:00
Code Issues Actions Packages Projects Releases Wiki Activity
34d6325be8
fractorium/Source/EmberRender/EmberRender.cpp
mfeemster 018ba26b5f --User changes 
-Add support for multiple GPU devices.
  --These options are present in the command line and in Fractorium.
 -Change scheme of specifying devices from platform,device to just total device index.
  --Single number on the command line.
  --Change from combo boxes for device selection to a table of all devices in Fractorium.
 -Temporal samples defaults to 100 instead of 1000 which was needless overkill.

--Bug fixes
 -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority.
 -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation.
 -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull().

--Code changes
 -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion().
 -EmberMotion.h: Call base constructor.
 -EmberPch.h: #pragma once only on Windows.
 -EmberToXml.h:
  --Handle different types of exceptions.
  --Add default cases to ToString().
 -Isaac.h: Remove unused variable in constructor.
 -Point.h: Call base constructor in Color().
 -Renderer.h/cpp:
  --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU.
  --Make CoordMap() return a const ref, not a pointer.
 -SheepTools.h:
  --Use 64-bit types like the rest of the code already does.
  --Fix some comment misspellings.
 -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions.
 -Utils.h:
  --Add new functions Equal() and Split().
  --Handle more exception types in ReadFile().
  --Get rid of most legacy blending of C and C++ argument parsing.
 -XmlToEmber.h:
  --Get rid of most legacy blending of C and C++ code from flam3.
  --Remove some unused variables.
 -EmberAnimate:
  --Support multi-GPU processing that alternates full frames between devices.
  --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option.
  --Remove bucketT template parameter, and hard code float in its place.
  --If a render fails, exit since there is no point in continuing an animation with a missing frame.
  --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before.
  --Remove some unused variables.
 -EmberGenome, EmberRender:
  --Support multi-GPU processing that alternates full frames between devices.
  --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option.
  --Remove bucketT template parameter, and hard code float in its place.
 -EmberRender:
  --Support multi-GPU processing that alternates full frames between devices.
  --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option.
  --Remove bucketT template parameter, and hard code float in its place.
  --Only print values when not rendering with OpenCL, since they're always 0 in that case.
 -EmberCLPch.h:
  --#pragma once only on Windows.
  --#include <atomic>.
 -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU.
 -OpenCLWrapper.h:
  --Move all OpenCL info related code into its own class OpenCLInfo.
  --Add members to cache the values of global memory size and max allocation size.
 -RendererCL.h/cpp:
  --Redesign to accomodate multi-GPU.
  --Constructor now takes a vector of devices.
  --Remove DumpErrorReport() function, it's handled in the base.
  --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter.
  --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit.
  --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing.
  --m_Calls member removed, as it's now per-device.
  --OpenCLWrapper removed.
  --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device.
  --Added member m_Devices, a vector of unique_ptr of RendererCLDevice.
 -EmberCommon.h
  --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices.
  --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices.
  --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device.
  --Add more comments to some existing functions.
 -EmberCommonPch.h: #pragma once only on Windows.
 -EmberOptions.h:
  --Remove --platform option, it's just sequential device number now with the --device option.
  --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image.
  --Add Devices() function to parse comma separated --device option string and return a vector of device indices.
  --Make int and uint types be 64-bit, so intmax_t and size_t.
  --Make better use of macros.
 -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref.
 -All project files: Turn off buffer security check option in Visual Studio (/Gs-)
 -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems.
 -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link.
 -EmberCL.pro: Add new files for multi-GPU support.
 -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake}
 -shared_settings.pri:
  -Add version string.
  -Remove old DESTDIR definitions.
  -Add the following lines or else nothing would build:
   CONFIG(release, debug|release) {
    CONFIG += warn_off
    DESTDIR = ../../../Bin/release
   }

   CONFIG(debug, debug|release) {
    DESTDIR = ../../../Bin/debug
   }

   QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t))
   LIBS += -L/usr/lib -lpthread
 -AboutDialog.ui: Another futile attempt to make it look correct on Linux.
 -FinalRenderDialog.h/cpp:
  --Add support for multi-GPU.
  --Change from combo boxes for device selection to a table of all devices.
  --Ensure device selection makes sense.
  --Remove "FinalRender" prefix of various function names, it's implied given the context.
 -FinalRenderEmberController.h/cpp:
  --Add support for multi-GPU.
  --Change m_FinishedImageCount to be atomic.
  --Move CancelRender() from the base to FinalRenderEmberController<T>.
  --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread.
  --Consolidate setting various renderer fields into SyncGuiToRenderer().
 -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table.
 -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system.
 -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed.
 -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used.
 -FractoriumSettings.h/cpp:
  --Temporal samples defaults to 100 instead of 1000 which was needless overkill.
  --Add multi-GPU support, remove old device,platform pair.
 -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system.
 -FractoriumOptionsDialog.h/cpp:
  --Add support for multi-GPU.
  --Consolidate more assignments in DataToGui().
  --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event.
 -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 18:33:45 -07:00

370 lines
12 KiB
C++
Raw Blame History

#include "EmberCommonPch.h"
#include "EmberRender.h"
#include "JpegUtils.h"
/// <summary>
/// The core of the EmberRender.exe program.
/// Template argument expected to be float or double.
/// </summary>
/// <param name="opt">A populated EmberOptions object which specifies all program options to be used</param>
/// <returns>True if success, else false.</returns>
template <typename T>
bool EmberRender(EmberOptions& opt)
{
EmberCLns::OpenCLInfo& info(EmberCLns::OpenCLInfo::Instance());
std::cout.imbue(std::locale(""));
if (opt.DumpArgs())
cout << opt.GetValues(OPT_USE_RENDER) << endl;
if (opt.OpenCLInfo())
{
cout << "\nOpenCL Info: " << endl;
cout << info.DumpInfo();
return true;
}
Timing t;
bool writeSuccess = false;
byte* finalImagep;
uint padding;
size_t i, channels;
size_t strips;
size_t iterCount;
string filename;
string inputPath = GetPath(opt.Input());
ostringstream os;
pair<size_t, size_t> p;
vector<Ember<T>> embers;
vector<byte> finalImage;
EmberStats stats;
EmberReport emberReport;
EmberImageComments comments;
XmlToEmber<T> parser;
EmberToXml<T> emberToXml;
vector<QTIsaac<ISAAC_SIZE, ISAAC_INT>> randVec;
const vector<pair<size_t, size_t>> devices = Devices(opt.Devices());
unique_ptr<RenderProgress<T>> progress(new RenderProgress<T>());
unique_ptr<Renderer<T, float>> renderer(CreateRenderer<T>(opt.EmberCL() ? OPENCL_RENDERER : CPU_RENDERER, devices, false, 0, emberReport));
vector<string> errorReport = emberReport.ErrorReport();
if (!errorReport.empty())
emberReport.DumpErrorReport();
if (!renderer.get())
{
cout << "Renderer creation failed, exiting." << endl;
return false;
}
if (opt.EmberCL() && renderer->RendererType() != OPENCL_RENDERER)//OpenCL init failed, so fall back to CPU.
opt.EmberCL(false);
if (!InitPaletteList<T>(opt.PalettePath()))
return false;
if (!ParseEmberFile(parser, opt.Input(), embers))
return false;
if (!opt.EmberCL())
{
if (opt.ThreadCount() == 0)
{
cout << "Using " << Timing::ProcessorCount() << " automatically detected threads." << endl;
opt.ThreadCount(Timing::ProcessorCount());
}
else
{
cout << "Using " << opt.ThreadCount() << " manually specified threads." << endl;
}
renderer->ThreadCount(opt.ThreadCount(), opt.IsaacSeed() != "" ? opt.IsaacSeed().c_str() : nullptr);
}
else
{
cout << "Using OpenCL to render." << endl;
if (opt.Verbose())
{
for (auto& device : devices)
{
cout << "Platform: " << info.PlatformName(device.first) << endl;
cout << "Device: " << info.DeviceName(device.first, device.second) << endl;
}
}
if (opt.ThreadCount() > 1)
cout << "Cannot specify threads with OpenCL, using 1 thread." << endl;
opt.ThreadCount(1);
renderer->ThreadCount(opt.ThreadCount(), opt.IsaacSeed() != "" ? opt.IsaacSeed().c_str() : nullptr);
if (opt.BitsPerChannel() != 8)
{
cout << "Bits per channel cannot be anything other than 8 with OpenCL, setting to 8." << endl;
opt.BitsPerChannel(8);
}
}
if (opt.Format() != "jpg" &&
opt.Format() != "png" &&
opt.Format() != "ppm" &&
opt.Format() != "bmp")
{
cout << "Format must be jpg, png, ppm, or bmp not " << opt.Format() << ". Setting to jpg." << endl;
}
channels = opt.Format() == "png" ? 4 : 3;
if (opt.BitsPerChannel() == 16 && opt.Format() != "png")
{
cout << "Support for 16 bits per channel images is only present for the png format. Setting to 8." << endl;
opt.BitsPerChannel(8);
}
else if (opt.BitsPerChannel() != 8 && opt.BitsPerChannel() != 16)
{
cout << "Unexpected bits per channel specified " << opt.BitsPerChannel() << ". Setting to 8." << endl;
opt.BitsPerChannel(8);
}
if (opt.InsertPalette() && opt.BitsPerChannel() != 8)
{
cout << "Inserting palette only supported with 8 bits per channel, insertion will not take place." << endl;
opt.InsertPalette(false);
}
if (opt.AspectRatio() < 0)
{
cout << "Invalid pixel aspect ratio " << opt.AspectRatio() << endl << ". Must be positive, setting to 1." << endl;
opt.AspectRatio(1);
}
if (!opt.Out().empty() && (embers.size() > 1))
{
cout << "Single output file " << opt.Out() << " specified for multiple images. Changing to use prefix of badname-changethis instead. Always specify prefixes when reading a file with multiple embers." << endl;
opt.Out("");
opt.Prefix("badname-changethis");
}
//Final setup steps before running.
os.imbue(std::locale(""));
padding = uint(log10(double(embers.size()))) + 1;
renderer->EarlyClip(opt.EarlyClip());
renderer->YAxisUp(opt.YAxisUp());
renderer->LockAccum(opt.LockAccum());
renderer->InsertPalette(opt.InsertPalette());
renderer->PixelAspectRatio(T(opt.AspectRatio()));
renderer->Transparency(opt.Transparency());
renderer->NumChannels(channels);
renderer->BytesPerChannel(opt.BitsPerChannel() / 8);
renderer->Priority(eThreadPriority(Clamp<int>(int(opt.Priority()), int(eThreadPriority::LOWEST), int(eThreadPriority::HIGHEST))));
renderer->Callback(opt.DoProgress() ? progress.get() : nullptr);
for (i = 0; i < embers.size(); i++)
{
if (opt.Verbose() && embers.size() > 1)
cout << "\nFlame = " << i + 1 << "/" << embers.size() << endl;
else if (embers.size() > 1)
VerbosePrint(endl);
if (opt.Supersample() > 0)
embers[i].m_Supersample = opt.Supersample();
if (opt.SubBatchSize() != DEFAULT_SBS)
embers[i].m_SubBatchSize = opt.SubBatchSize();
embers[i].m_TemporalSamples = 1;//Force temporal samples to 1 for render.
embers[i].m_Quality *= T(opt.QualityScale());
embers[i].m_FinalRasW = uint(T(embers[i].m_FinalRasW) * opt.SizeScale());
embers[i].m_FinalRasH = uint(T(embers[i].m_FinalRasH) * opt.SizeScale());
embers[i].m_PixelsPerUnit *= T(opt.SizeScale());
if (embers[i].m_FinalRasW == 0 || embers[i].m_FinalRasH == 0)
{
cout << "Output image " << i << " has dimension 0: " << embers[i].m_FinalRasW << ", " << embers[i].m_FinalRasH << ". Setting to 1920 x 1080." << endl;
embers[i].m_FinalRasW = 1920;
embers[i].m_FinalRasH = 1080;
}
//Cast to double in case the value exceeds 2^32.
double imageMem = double(renderer->NumChannels()) * double(embers[i].m_FinalRasW)
* double(embers[i].m_FinalRasH) * double(renderer->BytesPerChannel());
double maxMem = pow(2.0, double((sizeof(void*) * 8) - 1));
if (imageMem > maxMem)//Ensure the max amount of memory for a process is not exceeded.
{
cout << "Image " << i << " size > " << maxMem << ". Setting to 1920 x 1080." << endl;
embers[i].m_FinalRasW = 1920;
embers[i].m_FinalRasH = 1080;
}
stats.Clear();
renderer->SetEmber(embers[i]);
renderer->PrepFinalAccumVector(finalImage);//Must manually call this first because it could be erroneously made smaller due to strips if called inside Renderer::Run().
if (opt.Strips() > 1)
{
strips = opt.Strips();
}
else
{
p = renderer->MemoryRequired(1, true, false);//No threaded write for render, only for animate.
strips = CalcStrips(double(p.second), double(renderer->MemoryAvailable()), opt.UseMem());
if (strips > 1)
VerbosePrint("Setting strips to " << strips << " with specified memory usage of " << opt.UseMem());
}
strips = VerifyStrips(embers[i].m_FinalRasH, strips,
[&](const string& s) { cout << s << endl; },//Greater than height.
[&](const string& s) { cout << s << endl; },//Mod height != 0.
[&](const string& s) { cout << s << endl; });//Final strips value to be set.
//For testing incremental renderer.
//int sb = 1;
//bool resume = false, success = false;
//do
//{
// success = renderer->Run(finalImage, 0, sb, false/*resume == false*/) == RENDER_OK;
// sb++;
// resume = true;
//}
//while (success && renderer->ProcessState() != ACCUM_DONE);
StripsRender<T>(renderer.get(), embers[i], finalImage, 0, strips, opt.YAxisUp(),
[&](size_t strip)//Pre strip.
{
if (opt.Verbose() && (strips > 1) && strip > 0)
cout << endl;
if (strips > 1)
VerbosePrint("Strip = " << (strip + 1) << "/" << strips);
},
[&](size_t strip)//Post strip.
{
progress->Clear();
stats += renderer->Stats();
},
[&](size_t strip)//Error.
{
cout << "Error: image rendering failed, skipping to next image." << endl;
renderer->DumpErrorReport();//Something went wrong, print errors.
},
//Final strip.
//Original wrote every strip as a full image which could be very slow with many large images.
//Only write once all strips for this image are finished.
[&](Ember<T>& finalEmber)
{
if (!opt.Out().empty())
{
filename = opt.Out();
}
else if (opt.NameEnable() && !finalEmber.m_Name.empty())
{
filename = inputPath + opt.Prefix() + finalEmber.m_Name + opt.Suffix() + "." + opt.Format();
}
else
{
ostringstream fnstream;
fnstream << inputPath << opt.Prefix() << setfill('0') << setw(padding) << i << opt.Suffix() << "." << opt.Format();
filename = fnstream.str();
}
//TotalIterCount() is actually using ScaledQuality() which does not get reset upon ember assignment,
//so it ends up using the correct value for quality * strips.
iterCount = renderer->TotalIterCount(1);
comments = renderer->ImageComments(stats, opt.PrintEditDepth(), opt.IntPalette(), opt.HexPalette());
os.str("");
os << comments.m_NumIters << " / " << iterCount << " (" << std::fixed << std::setprecision(2) << ((double(stats.m_Iters) / double(iterCount)) * 100) << "%)";
VerbosePrint("\nIters ran/requested: " + os.str());
if (!opt.EmberCL()) VerbosePrint("Bad values: " << stats.m_Badvals);
VerbosePrint("Render time: " + t.Format(stats.m_RenderMs));
VerbosePrint("Pure iter time: " + t.Format(stats.m_IterMs));
VerbosePrint("Iters/sec: " << size_t(stats.m_Iters / (stats.m_IterMs / 1000.0)) << endl);
VerbosePrint("Writing " + filename);
if ((opt.Format() == "jpg" || opt.Format() == "bmp") && renderer->NumChannels() == 4)
RgbaToRgb(finalImage, finalImage, renderer->FinalRasW(), renderer->FinalRasH());
finalImagep = finalImage.data();
writeSuccess = false;
if (opt.Format() == "png")
writeSuccess = WritePng(filename.c_str(), finalImagep, finalEmber.m_FinalRasW, finalEmber.m_FinalRasH, opt.BitsPerChannel() / 8, opt.PngComments(), comments, opt.Id(), opt.Url(), opt.Nick());
else if (opt.Format() == "jpg")
writeSuccess = WriteJpeg(filename.c_str(), finalImagep, finalEmber.m_FinalRasW, finalEmber.m_FinalRasH, int(opt.JpegQuality()), opt.JpegComments(), comments, opt.Id(), opt.Url(), opt.Nick());
else if (opt.Format() == "ppm")
writeSuccess = WritePpm(filename.c_str(), finalImagep, finalEmber.m_FinalRasW, finalEmber.m_FinalRasH);
else if (opt.Format() == "bmp")
writeSuccess = WriteBmp(filename.c_str(), finalImagep, finalEmber.m_FinalRasW, finalEmber.m_FinalRasH);
if (!writeSuccess)
cout << "Error writing " << filename << endl;
});
if (opt.EmberCL() && opt.DumpKernel())
{
if (auto rendererCL = dynamic_cast<RendererCL<T, float>*>(renderer.get()))
{
cout << "Iteration kernel: \n" <<
rendererCL->IterKernel() << "\n\n" <<
rendererCL->DEKernel() << "\n\n" <<
rendererCL->FinalAccumKernel() << endl;
}
}
VerbosePrint("Done.");
}
t.Toc("\nFinished in: ", true);
return true;
}
/// <summary>
/// Main program entry point for EmberRender.exe.
/// </summary>
/// <param name="argc">The number of command line arguments passed</param>
/// <param name="argv">The command line arguments passed</param>
/// <returns>0 if successful, else 1.</returns>
int _tmain(int argc, _TCHAR* argv[])
{
bool b = false;
EmberOptions opt;
//Required for large allocs, else GPU memory usage will be severely limited to small sizes.
//This must be done in the application and not in the EmberCL DLL.
#ifdef WIN32
_putenv_s("GPU_MAX_ALLOC_PERCENT", "100");
//_putenv_s("GPU_FORCE_64BIT_PTR", "1");
#else
putenv(const_cast<char*>("GPU_MAX_ALLOC_PERCENT=100"));
#endif
if (!opt.Populate(argc, argv, OPT_USE_RENDER))
{
#ifdef DO_DOUBLE
if (opt.Bits() == 64)
{
b = EmberRender<double>(opt);
}
else
#endif
if (opt.Bits() == 33)
{
b = EmberRender<float>(opt);
}
else if (opt.Bits() == 32)
{
cout << "Bits 32/int histogram no longer supported. Using bits == 33 (float)." << endl;
b = EmberRender<float>(opt);
}
}
return b ? 0 : 1;
}
Reference in New Issue View Git Blame Copy Permalink