This commit is contained in:
2025-11-28 23:13:44 +05:30
commit a3a8e79709
7360 changed files with 1156074 additions and 0 deletions

View File

@ -0,0 +1,264 @@
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions
// are met:
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
// * Neither the name of NVIDIA CORPORATION nor the names of its
// contributors may be used to endorse or promote products derived
// from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ''AS IS'' AND ANY
// EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
// PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
// CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
// EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
// PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
// OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Copyright (c) 2008-2021 NVIDIA Corporation. All rights reserved.
// Copyright (c) 2004-2008 AGEIA Technologies, Inc. All rights reserved.
// Copyright (c) 2001-2004 NovodeX AG. All rights reserved.
// ****************************************************************************
// This snippet shows how to coordinate threads performing asynchronous
// work during the scene simulation. After simulate() is called, user threads
// are started that perform ray-casts against the scene. The call to
// fetchResults() is delayed until all ray-casts have completed.
// ****************************************************************************
#include <ctype.h>
#include "PxPhysicsAPI.h"
#include "../snippetutils/SnippetUtils.h"
#include "../snippetcommon/SnippetPrint.h"
#include "../snippetcommon/SnippetPVD.h"
using namespace physx;
PxDefaultAllocator gAllocator;
PxDefaultErrorCallback gErrorCallback;
PxFoundation* gFoundation = NULL;
PxPhysics* gPhysics = NULL;
PxDefaultCpuDispatcher* gDispatcher = NULL;
PxScene* gScene = NULL;
PxMaterial* gMaterial = NULL;
PxPvd* gPvd = NULL;
struct RaycastThread
{
SnippetUtils::Sync* mWorkReadySyncHandle;
SnippetUtils::Thread* mThreadHandle;
};
const PxU32 gNumThreads = 1;
RaycastThread gThreads[gNumThreads];
SnippetUtils::Sync* gWorkDoneSyncHandle;
const PxI32 gRayCount = 1024;
volatile PxI32 gRaysAvailable;
volatile PxI32 gRaysCompleted;
static PxVec3 randVec3()
{
return (PxVec3(float(rand())/float(RAND_MAX),
float(rand())/float(RAND_MAX),
float(rand())/float(RAND_MAX))*2.0f - PxVec3(1.0f)).getNormalized();
}
static void threadExecute(void* data)
{
RaycastThread* raycastThread = static_cast<RaycastThread*>(data);
// Perform random raycasts against the scene until stop.
for(;;)
{
// Wait here for the sync to be set then reset the sync
// to ensure that we only perform raycast work after the
// sync has been set again.
SnippetUtils::syncWait(raycastThread->mWorkReadySyncHandle);
SnippetUtils::syncReset(raycastThread->mWorkReadySyncHandle);
// If the thread has been signaled to quit then exit this function.
if (SnippetUtils::threadQuitIsSignalled(raycastThread->mThreadHandle))
break;
// Perform a fixed number of random raycasts against the scene
// and share the work between multiple threads.
while (SnippetUtils::atomicDecrement(&gRaysAvailable) >= 0)
{
PxVec3 dir = randVec3();
PxRaycastBuffer buf;
gScene->raycast(PxVec3(0.0f), dir.getNormalized(), 1000.0f, buf, PxHitFlag::eDEFAULT);
// If this is the last raycast then signal this to the main thread.
if (SnippetUtils::atomicIncrement(&gRaysCompleted) == gRayCount)
{
SnippetUtils::syncSet(gWorkDoneSyncHandle);
}
}
}
// Quit the current thread.
SnippetUtils::threadQuit(raycastThread->mThreadHandle);
}
void createStack(const PxTransform& t, PxU32 size, PxReal halfExtent)
{
PxShape* shape = gPhysics->createShape(PxBoxGeometry(halfExtent, halfExtent, halfExtent), *gMaterial);
for(PxU32 i=0; i<size;i++)
{
for(PxU32 j=0;j<size-i;j++)
{
PxTransform localTm(PxVec3(PxReal(j*2) - PxReal(size-i), PxReal(i*2+1), 0) * halfExtent);
PxRigidDynamic* body = gPhysics->createRigidDynamic(t.transform(localTm));
body->attachShape(*shape);
PxRigidBodyExt::updateMassAndInertia(*body, 10.0f);
gScene->addActor(*body);
}
}
shape->release();
}
void createPhysicsAndScene()
{
gFoundation = PxCreateFoundation(PX_PHYSICS_VERSION, gAllocator, gErrorCallback);
gPvd = PxCreatePvd(*gFoundation);
PxPvdTransport* transport = PxDefaultPvdSocketTransportCreate(PVD_HOST, 5425, 10);
gPvd->connect(*transport,PxPvdInstrumentationFlag::eALL);
gPhysics = PxCreatePhysics(PX_PHYSICS_VERSION, *gFoundation, PxTolerancesScale(),true,gPvd);
gMaterial = gPhysics->createMaterial(0.5f, 0.5f, 0.6f);
PxSceneDesc sceneDesc(gPhysics->getTolerancesScale());
sceneDesc.gravity = PxVec3(0.0f, -9.81f, 0.0f);
PxU32 numCores = SnippetUtils::getNbPhysicalCores();
gDispatcher = PxDefaultCpuDispatcherCreate(numCores == 0 ? 0 : numCores - 1);
sceneDesc.cpuDispatcher = gDispatcher;
sceneDesc.filterShader = PxDefaultSimulationFilterShader;
gScene = gPhysics->createScene(sceneDesc);
PxRigidStatic* groundPlane = PxCreatePlane(*gPhysics, PxPlane(0,1,0,0), *gMaterial);
gScene->addActor(*groundPlane);
for(PxU32 i=0;i<5;i++)
createStack(PxTransform(PxVec3(0,0,i*10.0f)), 10, 2.0f);
}
void createRaycastThreads()
{
// Create and start threads that will perform raycasts.
// Create a sync for each thread so that a signal may be sent
// from the main thread to the raycast thread that it can start
// performing raycasts.
for (PxU32 i=0; i < gNumThreads; ++i)
{
//Create a sync.
gThreads[i].mWorkReadySyncHandle = SnippetUtils::syncCreate();
//Create and start a thread.
gThreads[i].mThreadHandle = SnippetUtils::threadCreate(threadExecute, &gThreads[i]);
}
// Create another sync so that the raycast threads can signal to the main
// thread that they have finished performing their raycasts.
gWorkDoneSyncHandle = SnippetUtils::syncCreate();
}
void initPhysics()
{
createPhysicsAndScene();
createRaycastThreads();
}
void stepPhysics()
{
// Start simulation
gScene->simulate(1.0f/60.0f);
// Start ray-cast threads
gRaysAvailable = gRayCount;
gRaysCompleted = 0;
// Signal to each raycast thread that they can start performing raycasts.
for (PxU32 i=0; i < gNumThreads; ++i)
{
SnippetUtils::syncSet(gThreads[i].mWorkReadySyncHandle);
}
// Wait for raycast threads to finish.
SnippetUtils::syncWait(gWorkDoneSyncHandle);
SnippetUtils::syncReset(gWorkDoneSyncHandle);
// Fetch simulation results
gScene->fetchResults(true);
}
void cleanupPhysics()
{
// Signal threads to quit.
for (PxU32 i=0; i < gNumThreads; ++i)
{
SnippetUtils::threadSignalQuit(gThreads[i].mThreadHandle);
SnippetUtils::syncSet(gThreads[i].mWorkReadySyncHandle);
}
// Clean up raycast threads and syncs.
for (PxU32 i=0; i < gNumThreads; ++i)
{
SnippetUtils::threadWaitForQuit(gThreads[i].mThreadHandle);
SnippetUtils::threadRelease(gThreads[i].mThreadHandle);
SnippetUtils::syncRelease(gThreads[i].mWorkReadySyncHandle);
}
// Clean up the sync for the main thread.
SnippetUtils::syncRelease(gWorkDoneSyncHandle);
PX_RELEASE(gScene);
PX_RELEASE(gDispatcher);
PX_RELEASE(gPhysics);
if(gPvd)
{
PxPvdTransport* transport = gPvd->getTransport();
gPvd->release(); gPvd = NULL;
PX_RELEASE(transport);
}
PX_RELEASE(gFoundation);
printf("SnippetMultiThreading done.\n");
}
int snippetMain(int, const char*const*)
{
initPhysics();
for(PxU32 i=0; i<100; ++i)
stepPhysics();
cleanupPhysics();
return 0;
}