diff --git a/README.md b/README.md index d63a6a1..41ab67e 100644 --- a/README.md +++ b/README.md @@ -1,11 +1,44 @@ **University of Pennsylvania, CIS 565: GPU Programming and Architecture, Project 1 - Flocking** -* (TODO) YOUR NAME HERE - * (TODO) [LinkedIn](), [personal website](), [twitter](), etc. -* Tested on: (TODO) Windows 22, i7-2222 @ 2.22GHz 22GB, GTX 222 222MB (Moore 2222 Lab) +* Joshua Nadel + * https://www.linkedin.com/in/joshua-nadel-379382136/, http://www.joshnadel.com/ +* Tested on: Windows 10, i7-6700HQ @ 2.60GHz 16GB, GTX 970M (Personal laptop) -### (TODO: Your README) +### -Include screenshots, analysis, etc. (Remember, this is public, so don't put -anything here that you don't want to share with the world.) +![](images/boids.png) + +![](images/boidsGIF.gif) + +![](images/boidsGIF2.gif) + +## Part 3: Performance Analysis + +I will be using the non-visualization framerate as a benchmark to measure my implementation's performance. + +* For each implementation, how does changing the number of boids affect performance? Why do you think this is? + +For the naive implementation, increasing the number of boids tremendously slows down performance. This is surely because every boid checks every other boid to determine its neighbors, and so adding more boids exponentially slows down execution. The scattered grid implementation can withstand larger increases in boid count, operating between 50 and 60 fps even at 100,000 boids. The grid-based neighbor check allows the program to skip many boid to boid compares by checking boids against grid cells, of which there are (typically) much fewer. Finally, the coherent grid implementation is able to withstand even more boids. It operates at a consistent 121 fps at 100,000 boids, and can take up to 150,000 before dipping below 60 fps. From my understanding of lower-level memory access optimization, retrieving data from consecutive spaces in memory is much more conducive to the CPU's caching system, through which data can be pre-fetched from memory. + +* For each implementation, how does changing the block count and block size affect performance? Why do you think this is? + +For the naive implementation, increasing the block size improved the performance; at 5000 boids, the simulation went from 130 to 180 to 200 fps at block sizes of 64, 128, and 256 respectively. For the scattered grid implementation, increasing the block size only slightly improved the performance; at 5000 boids, the simulation went from around 450-460 to 460-480 to 460-480 fps at block sizes of 64, 128, and 256 respectively. For the coherent grid implementation, increasing the block size only slightly improved the performance; at 5000 boids, the simulation went from around 520 to 520-530 to 520-550 fps at block sizes of 64, 128, and 256 respectively. These improvements in performance are likely due to the increase in number of available threads that comes with larger block sizes. + +I found no reference to a block count parameter in this project. + +* For the coherent uniform grid: did you experience any performance improvements with the more coherent uniform grid? Was this the outcome you expected? Why or why not? + +I experienced improvements in performance with the coherent grid. However, I did not expect to. My understanding of program execution time was that the number of accesses or memory reads determines efficiency, and that the specific locations in memory were irrelevant. My understanding is still rudimentary, but I'm beginning to see that by cohering memory together in access order, the computer is able to make certain optimizations through low-level caching. + +* Did changing cell width and checking 27 vs 8 neighboring cells affect performance? Why or why not? Be careful: it is insufficient (and possibly incorrect) to say that 27-cell is slower simply because there are more cells to check! + +Changing the cell width did not affect performance. While it is the case that the 27-cell check requires more cell checks, it is also the case that the higher-resolution grid allows more cells to be eliminated by the distance check and can eliminate some potential false boid neighbor checks that the 8-cell check wouldn't catch. + +## Part 4: Write-up + +Boid Count effect on FPS +![](images/boidCountGraph.png) + +Block Size effect on FPS +![](images/blockSizeGraph.png) \ No newline at end of file diff --git a/images/blockSizeGraph.png b/images/blockSizeGraph.png new file mode 100644 index 0000000..c717805 Binary files /dev/null and b/images/blockSizeGraph.png differ diff --git a/images/boidCountGraph.png b/images/boidCountGraph.png new file mode 100644 index 0000000..e5b65c8 Binary files /dev/null and b/images/boidCountGraph.png differ diff --git a/images/boids.png b/images/boids.png new file mode 100644 index 0000000..87f9c5c Binary files /dev/null and b/images/boids.png differ diff --git a/images/boidsGIF.gif b/images/boidsGIF.gif new file mode 100644 index 0000000..422dfb3 Binary files /dev/null and b/images/boidsGIF.gif differ diff --git a/images/boidsGIF2.gif b/images/boidsGIF2.gif new file mode 100644 index 0000000..57c4671 Binary files /dev/null and b/images/boidsGIF2.gif differ diff --git a/src/kernel.cu b/src/kernel.cu index 74dffcb..c923605 100644 --- a/src/kernel.cu +++ b/src/kernel.cu @@ -21,14 +21,14 @@ * Check for CUDA errors; print and exit if there was a problem. */ void checkCUDAError(const char *msg, int line = -1) { - cudaError_t err = cudaGetLastError(); - if (cudaSuccess != err) { - if (line >= 0) { - fprintf(stderr, "Line %d: ", line); - } - fprintf(stderr, "Cuda error: %s: %s.\n", msg, cudaGetErrorString(err)); - exit(EXIT_FAILURE); - } + cudaError_t err = cudaGetLastError(); + if (cudaSuccess != err) { + if (line >= 0) { + fprintf(stderr, "Line %d: ", line); + } + fprintf(stderr, "Cuda error: %s: %s.\n", msg, cudaGetErrorString(err)); + exit(EXIT_FAILURE); + } } @@ -85,6 +85,8 @@ int *dev_gridCellEndIndices; // to this cell? // TODO-2.3 - consider what additional buffers you might need to reshuffle // the position and velocity data to be coherent within cells. +glm::vec3 *dev_shuffled_pos; +glm::vec3 *dev_shuffled_vel; // LOOK-2.1 - Grid parameters based on simulation parameters. // These are automatically computed for you in Boids::initSimulation @@ -99,13 +101,13 @@ glm::vec3 gridMinimum; ******************/ __host__ __device__ unsigned int hash(unsigned int a) { - a = (a + 0x7ed55d16) + (a << 12); - a = (a ^ 0xc761c23c) ^ (a >> 19); - a = (a + 0x165667b1) + (a << 5); - a = (a + 0xd3a2646c) ^ (a << 9); - a = (a + 0xfd7046c5) + (a << 3); - a = (a ^ 0xb55a4f09) ^ (a >> 16); - return a; + a = (a + 0x7ed55d16) + (a << 12); + a = (a ^ 0xc761c23c) ^ (a >> 19); + a = (a + 0x165667b1) + (a << 5); + a = (a + 0xd3a2646c) ^ (a << 9); + a = (a + 0xfd7046c5) + (a << 3); + a = (a ^ 0xb55a4f09) ^ (a >> 16); + return a; } /** @@ -113,10 +115,10 @@ __host__ __device__ unsigned int hash(unsigned int a) { * Function for generating a random vec3. */ __host__ __device__ glm::vec3 generateRandomVec3(float time, int index) { - thrust::default_random_engine rng(hash((int)(index * time))); - thrust::uniform_real_distribution unitDistrib(-1, 1); + thrust::default_random_engine rng(hash((int)(index * time))); + thrust::uniform_real_distribution unitDistrib(-1, 1); - return glm::vec3((float)unitDistrib(rng), (float)unitDistrib(rng), (float)unitDistrib(rng)); + return glm::vec3((float)unitDistrib(rng), (float)unitDistrib(rng), (float)unitDistrib(rng)); } /** @@ -124,52 +126,70 @@ __host__ __device__ glm::vec3 generateRandomVec3(float time, int index) { * CUDA kernel for generating boids with a specified mass randomly around the star. */ __global__ void kernGenerateRandomPosArray(int time, int N, glm::vec3 * arr, float scale) { - int index = (blockIdx.x * blockDim.x) + threadIdx.x; - if (index < N) { - glm::vec3 rand = generateRandomVec3(time, index); - arr[index].x = scale * rand.x; - arr[index].y = scale * rand.y; - arr[index].z = scale * rand.z; - } + int index = (blockIdx.x * blockDim.x) + threadIdx.x; + if (index < N) { + glm::vec3 rand = generateRandomVec3(time, index); + arr[index].x = scale * rand.x; + arr[index].y = scale * rand.y; + arr[index].z = scale * rand.z; + } } /** * Initialize memory, update some globals */ void Boids::initSimulation(int N) { - numObjects = N; - dim3 fullBlocksPerGrid((N + blockSize - 1) / blockSize); - - // LOOK-1.2 - This is basic CUDA memory management and error checking. - // Don't forget to cudaFree in Boids::endSimulation. - cudaMalloc((void**)&dev_pos, N * sizeof(glm::vec3)); - checkCUDAErrorWithLine("cudaMalloc dev_pos failed!"); - - cudaMalloc((void**)&dev_vel1, N * sizeof(glm::vec3)); - checkCUDAErrorWithLine("cudaMalloc dev_vel1 failed!"); - - cudaMalloc((void**)&dev_vel2, N * sizeof(glm::vec3)); - checkCUDAErrorWithLine("cudaMalloc dev_vel2 failed!"); - - // LOOK-1.2 - This is a typical CUDA kernel invocation. - kernGenerateRandomPosArray<<>>(1, numObjects, - dev_pos, scene_scale); - checkCUDAErrorWithLine("kernGenerateRandomPosArray failed!"); - - // LOOK-2.1 computing grid params - gridCellWidth = 2.0f * std::max(std::max(rule1Distance, rule2Distance), rule3Distance); - int halfSideCount = (int)(scene_scale / gridCellWidth) + 1; - gridSideCount = 2 * halfSideCount; - - gridCellCount = gridSideCount * gridSideCount * gridSideCount; - gridInverseCellWidth = 1.0f / gridCellWidth; - float halfGridWidth = gridCellWidth * halfSideCount; - gridMinimum.x -= halfGridWidth; - gridMinimum.y -= halfGridWidth; - gridMinimum.z -= halfGridWidth; - - // TODO-2.1 TODO-2.3 - Allocate additional buffers here. - cudaDeviceSynchronize(); + numObjects = N; + dim3 fullBlocksPerGrid((N + blockSize - 1) / blockSize); + + // LOOK-1.2 - This is basic CUDA memory management and error checking. + // Don't forget to cudaFree in Boids::endSimulation. + cudaMalloc((void**)&dev_pos, N * sizeof(glm::vec3)); + checkCUDAErrorWithLine("cudaMalloc dev_pos failed!"); + + cudaMalloc((void**)&dev_vel1, N * sizeof(glm::vec3)); + checkCUDAErrorWithLine("cudaMalloc dev_vel1 failed!"); + + cudaMalloc((void**)&dev_vel2, N * sizeof(glm::vec3)); + checkCUDAErrorWithLine("cudaMalloc dev_vel2 failed!"); + + // LOOK-1.2 - This is a typical CUDA kernel invocation. + kernGenerateRandomPosArray << > > (1, numObjects, + dev_pos, scene_scale); + checkCUDAErrorWithLine("kernGenerateRandomPosArray failed!"); + + // LOOK-2.1 computing grid params + gridCellWidth = 2.0f * std::max(std::max(rule1Distance, rule2Distance), rule3Distance); + int halfSideCount = (int)(scene_scale / gridCellWidth) + 1; + gridSideCount = 2 * halfSideCount; + + gridCellCount = gridSideCount * gridSideCount * gridSideCount; + gridInverseCellWidth = 1.0f / gridCellWidth; + float halfGridWidth = gridCellWidth * halfSideCount; + gridMinimum.x -= halfGridWidth; + gridMinimum.y -= halfGridWidth; + gridMinimum.z -= halfGridWidth; + + // TODO-2.1 TODO-2.3 - Allocate additional buffers here. + cudaMalloc((void**)&dev_particleArrayIndices, N * sizeof(int)); + checkCUDAErrorWithLine("cudaMalloc dev_particleArrayIndices failed!"); + + cudaMalloc((void**)&dev_particleGridIndices, N * sizeof(int)); + checkCUDAErrorWithLine("cudaMalloc dev_particleGridIndices failed!"); + + cudaMalloc((void**)&dev_gridCellStartIndices, gridCellCount * sizeof(int)); + checkCUDAErrorWithLine("cudaMalloc dev_gridCellStartIndices failed!"); + + cudaMalloc((void**)&dev_gridCellEndIndices, gridCellCount * sizeof(int)); + checkCUDAErrorWithLine("cudaMalloc dev_gridCellEndIndices failed!"); + + cudaMalloc((void**)&dev_shuffled_pos, N * sizeof(glm::vec3)); + checkCUDAErrorWithLine("cudaMalloc dev_shuffled_pos failed!"); + + cudaMalloc((void**)&dev_shuffled_vel, N * sizeof(glm::vec3)); + checkCUDAErrorWithLine("cudaMalloc dev_shuffled_vel failed!"); + + cudaDeviceSynchronize(); } @@ -181,41 +201,41 @@ void Boids::initSimulation(int N) { * Copy the boid positions into the VBO so that they can be drawn by OpenGL. */ __global__ void kernCopyPositionsToVBO(int N, glm::vec3 *pos, float *vbo, float s_scale) { - int index = threadIdx.x + (blockIdx.x * blockDim.x); + int index = threadIdx.x + (blockIdx.x * blockDim.x); - float c_scale = -1.0f / s_scale; + float c_scale = -1.0f / s_scale; - if (index < N) { - vbo[4 * index + 0] = pos[index].x * c_scale; - vbo[4 * index + 1] = pos[index].y * c_scale; - vbo[4 * index + 2] = pos[index].z * c_scale; - vbo[4 * index + 3] = 1.0f; - } + if (index < N) { + vbo[4 * index + 0] = pos[index].x * c_scale; + vbo[4 * index + 1] = pos[index].y * c_scale; + vbo[4 * index + 2] = pos[index].z * c_scale; + vbo[4 * index + 3] = 1.0f; + } } __global__ void kernCopyVelocitiesToVBO(int N, glm::vec3 *vel, float *vbo, float s_scale) { - int index = threadIdx.x + (blockIdx.x * blockDim.x); - - if (index < N) { - vbo[4 * index + 0] = vel[index].x + 0.3f; - vbo[4 * index + 1] = vel[index].y + 0.3f; - vbo[4 * index + 2] = vel[index].z + 0.3f; - vbo[4 * index + 3] = 1.0f; - } + int index = threadIdx.x + (blockIdx.x * blockDim.x); + + if (index < N) { + vbo[4 * index + 0] = vel[index].x + 0.3f; + vbo[4 * index + 1] = vel[index].y + 0.3f; + vbo[4 * index + 2] = vel[index].z + 0.3f; + vbo[4 * index + 3] = 1.0f; + } } /** * Wrapper for call to the kernCopyboidsToVBO CUDA kernel. */ void Boids::copyBoidsToVBO(float *vbodptr_positions, float *vbodptr_velocities) { - dim3 fullBlocksPerGrid((numObjects + blockSize - 1) / blockSize); + dim3 fullBlocksPerGrid((numObjects + blockSize - 1) / blockSize); - kernCopyPositionsToVBO << > >(numObjects, dev_pos, vbodptr_positions, scene_scale); - kernCopyVelocitiesToVBO << > >(numObjects, dev_vel1, vbodptr_velocities, scene_scale); + kernCopyPositionsToVBO << > > (numObjects, dev_pos, vbodptr_positions, scene_scale); + kernCopyVelocitiesToVBO << > > (numObjects, dev_vel1, vbodptr_velocities, scene_scale); - checkCUDAErrorWithLine("copyBoidsToVBO failed!"); + checkCUDAErrorWithLine("copyBoidsToVBO failed!"); - cudaDeviceSynchronize(); + cudaDeviceSynchronize(); } @@ -230,10 +250,50 @@ void Boids::copyBoidsToVBO(float *vbodptr_positions, float *vbodptr_velocities) * in the `pos` and `vel` arrays. */ __device__ glm::vec3 computeVelocityChange(int N, int iSelf, const glm::vec3 *pos, const glm::vec3 *vel) { - // Rule 1: boids fly towards their local perceived center of mass, which excludes themselves - // Rule 2: boids try to stay a distance d away from each other - // Rule 3: boids try to match the speed of surrounding boids - return glm::vec3(0.0f, 0.0f, 0.0f); + // Rule 1: boids fly towards their local perceived center of mass, which excludes themselves + // Rule 2: boids try to stay a distance d away from each other + // Rule 3: boids try to match the speed of surrounding boids + + glm::vec3 toReturn = glm::vec3(0, 0, 0); + + glm::vec3 mePos = pos[iSelf]; + + glm::vec3 perceived_center = glm::vec3(0, 0, 0); + glm::vec3 separation = glm::vec3(0, 0, 0); + glm::vec3 alignment = glm::vec3(0, 0, 0); + + int neighborCountRule1 = 0; + int neighborCountRule3 = 0; + for (int i = 0; i < N; i++) { + if (i == iSelf) continue; + glm::vec3 boidPos = pos[i]; + if (glm::distance(mePos, boidPos) < rule1Distance) { + perceived_center += boidPos; + neighborCountRule1++; + } + + if (glm::distance(mePos, boidPos) < rule2Distance) { + separation -= boidPos - mePos; + } + if (glm::distance(boidPos, mePos) < rule3Distance) { + glm::vec3 boidVel = vel[i]; + alignment += boidVel; + neighborCountRule3++; + } + } + + if (neighborCountRule1 != 0) { + perceived_center /= (float)neighborCountRule1; + toReturn += (perceived_center - mePos) * rule1Scale; + } + if (neighborCountRule3 != 0) { + alignment /= (float)neighborCountRule3; + toReturn += alignment * rule3Scale; + } + + toReturn += separation * rule2Scale; + + return toReturn; } /** @@ -241,10 +301,23 @@ __device__ glm::vec3 computeVelocityChange(int N, int iSelf, const glm::vec3 *po * For each of the `N` bodies, update its position based on its current velocity. */ __global__ void kernUpdateVelocityBruteForce(int N, glm::vec3 *pos, - glm::vec3 *vel1, glm::vec3 *vel2) { - // Compute a new velocity based on pos and vel1 - // Clamp the speed - // Record the new velocity into vel2. Question: why NOT vel1? + glm::vec3 *vel1, glm::vec3 *vel2) { + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + glm::vec3 mePos = pos[index]; + + // Compute a new velocity based on pos and vel1 + glm::vec3 newVel = vel1[index] + computeVelocityChange(N, index, pos, vel1); + + // Clamp the speed + if (glm::length(newVel) > maxSpeed) { + newVel = glm::normalize(newVel) * maxSpeed; + } + + // Record the new velocity into vel2. Question: why NOT vel1? + vel2[index] = newVel; } /** @@ -252,24 +325,24 @@ __global__ void kernUpdateVelocityBruteForce(int N, glm::vec3 *pos, * For each of the `N` bodies, update its position based on its current velocity. */ __global__ void kernUpdatePos(int N, float dt, glm::vec3 *pos, glm::vec3 *vel) { - // Update position by velocity - int index = threadIdx.x + (blockIdx.x * blockDim.x); - if (index >= N) { - return; - } - glm::vec3 thisPos = pos[index]; - thisPos += vel[index] * dt; - - // Wrap the boids around so we don't lose them - thisPos.x = thisPos.x < -scene_scale ? scene_scale : thisPos.x; - thisPos.y = thisPos.y < -scene_scale ? scene_scale : thisPos.y; - thisPos.z = thisPos.z < -scene_scale ? scene_scale : thisPos.z; - - thisPos.x = thisPos.x > scene_scale ? -scene_scale : thisPos.x; - thisPos.y = thisPos.y > scene_scale ? -scene_scale : thisPos.y; - thisPos.z = thisPos.z > scene_scale ? -scene_scale : thisPos.z; - - pos[index] = thisPos; + // Update position by velocity + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + glm::vec3 thisPos = pos[index]; + thisPos += vel[index] * dt; + + // Wrap the boids around so we don't lose them + thisPos.x = thisPos.x < -scene_scale ? scene_scale : thisPos.x; + thisPos.y = thisPos.y < -scene_scale ? scene_scale : thisPos.y; + thisPos.z = thisPos.z < -scene_scale ? scene_scale : thisPos.z; + + thisPos.x = thisPos.x > scene_scale ? -scene_scale : thisPos.x; + thisPos.y = thisPos.y > scene_scale ? -scene_scale : thisPos.y; + thisPos.z = thisPos.z > scene_scale ? -scene_scale : thisPos.z; + + pos[index] = thisPos; } // LOOK-2.1 Consider this method of computing a 1D index from a 3D grid index. @@ -279,179 +352,497 @@ __global__ void kernUpdatePos(int N, float dt, glm::vec3 *pos, glm::vec3 *vel) { // for(y) // for(z)? Or some other order? __device__ int gridIndex3Dto1D(int x, int y, int z, int gridResolution) { - return x + y * gridResolution + z * gridResolution * gridResolution; + return x + y * gridResolution + z * gridResolution * gridResolution; } __global__ void kernComputeIndices(int N, int gridResolution, - glm::vec3 gridMin, float inverseCellWidth, - glm::vec3 *pos, int *indices, int *gridIndices) { - // TODO-2.1 - // - Label each boid with the index of its grid cell. - // - Set up a parallel array of integer indices as pointers to the actual - // boid data in pos and vel1/vel2 + glm::vec3 gridMin, float inverseCellWidth, + glm::vec3 *pos, int *indices, int *gridIndices) { + // TODO-2.1 + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + glm::vec3 thisPos = pos[index]; + + // - Label each boid with the index of its grid cell. + glm::vec3 gridIndex3D = glm::floor((thisPos - gridMin) * inverseCellWidth); + int gridIndex = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z, gridResolution); + gridIndices[index] = gridIndex; + + // - Set up a parallel array of integer indices as pointers to the actual + // boid data in pos and vel1/vel2 + indices[index] = index; } // LOOK-2.1 Consider how this could be useful for indicating that a cell // does not enclose any boids __global__ void kernResetIntBuffer(int N, int *intBuffer, int value) { - int index = (blockIdx.x * blockDim.x) + threadIdx.x; - if (index < N) { - intBuffer[index] = value; - } + int index = (blockIdx.x * blockDim.x) + threadIdx.x; + if (index < N) { + intBuffer[index] = value; + } } __global__ void kernIdentifyCellStartEnd(int N, int *particleGridIndices, - int *gridCellStartIndices, int *gridCellEndIndices) { - // TODO-2.1 - // Identify the start point of each cell in the gridIndices array. - // This is basically a parallel unrolling of a loop that goes - // "this index doesn't match the one before it, must be a new cell!" + int *gridCellStartIndices, int *gridCellEndIndices) { + // TODO-2.1 + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + // Identify the start point of each cell in the gridIndices array. + // This is basically a parallel unrolling of a loop that goes + // "this index doesn't match the one before it, must be a new cell!" + int currCell = particleGridIndices[index]; + if (index == 0) { + gridCellStartIndices[currCell] = index; + } + else { + int prevCell = particleGridIndices[index - 1]; + if (currCell != prevCell) { + gridCellEndIndices[prevCell] = index; + gridCellStartIndices[currCell] = index; + } + } } __global__ void kernUpdateVelNeighborSearchScattered( - int N, int gridResolution, glm::vec3 gridMin, - float inverseCellWidth, float cellWidth, - int *gridCellStartIndices, int *gridCellEndIndices, - int *particleArrayIndices, - glm::vec3 *pos, glm::vec3 *vel1, glm::vec3 *vel2) { - // TODO-2.1 - Update a boid's velocity using the uniform grid to reduce - // the number of boids that need to be checked. - // - Identify the grid cell that this particle is in - // - Identify which cells may contain neighbors. This isn't always 8. - // - For each cell, read the start/end indices in the boid pointer array. - // - Access each boid in the cell and compute velocity change from - // the boids rules, if this boid is within the neighborhood distance. - // - Clamp the speed change before putting the new speed in vel2 + int N, int gridResolution, glm::vec3 gridMin, + float inverseCellWidth, float cellWidth, + int *gridCellStartIndices, int *gridCellEndIndices, + int *particleArrayIndices, + glm::vec3 *pos, glm::vec3 *vel1, glm::vec3 *vel2) { + // TODO-2.1 - Update a boid's velocity using the uniform grid to reduce + // the number of boids that need to be checked. + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + // - Identify the grid cell that this particle is in + glm::vec3 thisPos = pos[index]; + glm::vec3 gridIndex3D = glm::floor((thisPos - gridMin) * inverseCellWidth); + glm::vec3 relativePos = ((thisPos - gridMin) * inverseCellWidth) - gridIndex3D; + + // - Identify which cells may contain neighbors. This isn't always 8. + int neighborCells[] = {-1, -1, -1, -1, -1, -1, -1 ,-1}; + neighborCells[0] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z, gridResolution); + if (relativePos.x < 0.5) { + neighborCells[1] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y, gridIndex3D.z, gridResolution); + if (relativePos.y < 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y - 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y - 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z + 1, gridResolution); + } + else if (relativePos.y > 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y + 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y + 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z + 1, gridResolution); + } + } + else if (relativePos.x > 0.5) { + neighborCells[1] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y, gridIndex3D.z, gridResolution); + if (relativePos.y < 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z + 1, gridResolution); + } + else if (relativePos.y > 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z + 1, gridResolution); + } + } + + if (relativePos.y < 0.5) { + neighborCells[2] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y - 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y - 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y - 1, gridIndex3D.z + 1, gridResolution); + } + else if (relativePos.y > 0.5) { + neighborCells[2] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y + 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y + 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y + 1, gridIndex3D.z + 1, gridResolution); + } + + if (relativePos.z < 0.5) { + neighborCells[3] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z - 1, gridResolution); + if (relativePos.x < 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y, gridIndex3D.z - 1, gridResolution); + else if (relativePos.x > 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y, gridIndex3D.z - 1, gridResolution); + } + else if (relativePos.z > 0.5) { + neighborCells[3] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z + 1, gridResolution); + if (relativePos.x < 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y, gridIndex3D.z + 1, gridResolution); + else if (relativePos.x > 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y, gridIndex3D.z + 1, gridResolution); + } + + // - For each cell, read the start/end indices in the boid pointer array. + glm::vec3 velChange = glm::vec3(0, 0, 0); + glm::vec3 perceived_center = glm::vec3(0, 0, 0); + glm::vec3 separation = glm::vec3(0, 0, 0); + glm::vec3 alignment = glm::vec3(0, 0, 0); + int neighborCountRule1 = 0; + int neighborCountRule3 = 0; + + for (int i = 0; i < 8; i++) { + int neighbor = neighborCells[i]; + if (neighbor == -1) continue; + int start = gridCellStartIndices[neighbor]; + int end = gridCellEndIndices[neighbor]; + if (start == -1 || end == -1) continue; + + // - Access each boid in the cell and compute velocity change from + // the boids rules, if this boid is within the neighborhood distance. + for (int j = start; j <= end; j++) { + int boidIndex = particleArrayIndices[j]; + if (boidIndex == index) continue; + glm::vec3 boidPos = pos[boidIndex]; + if (glm::distance(thisPos, boidPos) < rule1Distance) { + perceived_center += boidPos; + neighborCountRule1++; + } + + if (glm::distance(thisPos, boidPos) < rule2Distance) { + separation -= boidPos - thisPos; + } + if (glm::distance(boidPos, thisPos) < rule3Distance) { + glm::vec3 boidVel = vel1[boidIndex]; + alignment += boidVel; + neighborCountRule3++; + } + } + } + + if (neighborCountRule1 != 0) { + perceived_center /= (float)neighborCountRule1; + velChange += (perceived_center - thisPos) * rule1Scale; + } + if (neighborCountRule3 != 0) { + alignment /= (float)neighborCountRule3; + velChange += alignment * rule3Scale; + } + + velChange += separation * rule2Scale; + + // - Clamp the speed change before putting the new speed in vel2 + glm::vec3 newVel = vel1[index] + velChange; + if (glm::length(newVel) > maxSpeed) { + newVel = glm::normalize(newVel) * maxSpeed; + } + vel2[index] = newVel; // Setting vel2[index] specifically to newVel causes FPS slowdown. Hannah says this might be due to a cache line issue and is out of my control. } __global__ void kernUpdateVelNeighborSearchCoherent( - int N, int gridResolution, glm::vec3 gridMin, - float inverseCellWidth, float cellWidth, - int *gridCellStartIndices, int *gridCellEndIndices, - glm::vec3 *pos, glm::vec3 *vel1, glm::vec3 *vel2) { - // TODO-2.3 - This should be very similar to kernUpdateVelNeighborSearchScattered, - // except with one less level of indirection. - // This should expect gridCellStartIndices and gridCellEndIndices to refer - // directly to pos and vel1. - // - Identify the grid cell that this particle is in - // - Identify which cells may contain neighbors. This isn't always 8. - // - For each cell, read the start/end indices in the boid pointer array. - // DIFFERENCE: For best results, consider what order the cells should be - // checked in to maximize the memory benefits of reordering the boids data. - // - Access each boid in the cell and compute velocity change from - // the boids rules, if this boid is within the neighborhood distance. - // - Clamp the speed change before putting the new speed in vel2 + int N, int gridResolution, glm::vec3 gridMin, + float inverseCellWidth, float cellWidth, + int *gridCellStartIndices, int *gridCellEndIndices, + glm::vec3 *pos, glm::vec3 *vel1, glm::vec3 *vel2) { + // TODO-2.3 - This should be very similar to kernUpdateVelNeighborSearchScattered, + // except with one less level of indirection. + // This should expect gridCellStartIndices and gridCellEndIndices to refer + // directly to pos and vel1. + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + + // - Identify the grid cell that this particle is in + glm::vec3 thisPos = pos[index]; + glm::vec3 gridIndex3D = glm::floor((thisPos - gridMin) * inverseCellWidth); + glm::vec3 relativePos = ((thisPos - gridMin) * inverseCellWidth) - gridIndex3D; + + // - Identify which cells may contain neighbors. This isn't always 8. + int neighborCells[] = { -1, -1, -1, -1, -1, -1, -1 ,-1 }; + neighborCells[0] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z, gridResolution); + if (relativePos.x < 0.5) { + neighborCells[1] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y, gridIndex3D.z, gridResolution); + if (relativePos.y < 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y - 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y - 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z + 1, gridResolution); + } + else if (relativePos.y > 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y + 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y + 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z + 1, gridResolution); + } + } + else if (relativePos.x > 0.5) { + neighborCells[1] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y, gridIndex3D.z, gridResolution); + if (relativePos.y < 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y - 1, gridIndex3D.z + 1, gridResolution); + } + else if (relativePos.y > 0.5) { + neighborCells[4] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[7] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y + 1, gridIndex3D.z + 1, gridResolution); + } + } + + if (relativePos.y < 0.5) { + neighborCells[2] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y - 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y - 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y - 1, gridIndex3D.z + 1, gridResolution); + } + else if (relativePos.y > 0.5) { + neighborCells[2] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y + 1, gridIndex3D.z, gridResolution); + if (relativePos.z < 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y + 1, gridIndex3D.z - 1, gridResolution); + else if (relativePos.z > 0.5) neighborCells[5] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y + 1, gridIndex3D.z + 1, gridResolution); + } + + if (relativePos.z < 0.5) { + neighborCells[3] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z - 1, gridResolution); + if (relativePos.x < 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y, gridIndex3D.z - 1, gridResolution); + else if (relativePos.x > 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y, gridIndex3D.z - 1, gridResolution); + } + else if (relativePos.z > 0.5) { + neighborCells[3] = gridIndex3Dto1D(gridIndex3D.x, gridIndex3D.y, gridIndex3D.z + 1, gridResolution); + if (relativePos.x < 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x - 1, gridIndex3D.y, gridIndex3D.z + 1, gridResolution); + else if (relativePos.x > 0.5) neighborCells[6] = gridIndex3Dto1D(gridIndex3D.x + 1, gridIndex3D.y, gridIndex3D.z + 1, gridResolution); + } + + // - For each cell, read the start/end indices in the boid pointer array. + glm::vec3 velChange = glm::vec3(0, 0, 0); + glm::vec3 perceived_center = glm::vec3(0, 0, 0); + glm::vec3 separation = glm::vec3(0, 0, 0); + glm::vec3 alignment = glm::vec3(0, 0, 0); + int neighborCountRule1 = 0; + int neighborCountRule3 = 0; + + for (int i = 0; i < 8; i++) { + int neighbor = neighborCells[i]; + if (neighbor < 0) { + continue; + } + int start = gridCellStartIndices[neighbor]; + int end = gridCellEndIndices[neighbor]; + if (start == -1 || end == -1) continue; + + // DIFFERENCE: For best results, consider what order the cells should be + // checked in to maximize the memory benefits of reordering the boids data. + + // - Access each boid in the cell and compute velocity change from + // the boids rules, if this boid is within the neighborhood distance. + for (int j = start; j <= end; j++) { + int boidIndex = j; // Boid indices are directly the range [start, end] + if (boidIndex == index) continue; + glm::vec3 boidPos = pos[boidIndex]; + if (glm::distance(thisPos, boidPos) < rule1Distance) { + perceived_center += boidPos; + neighborCountRule1++; + } + + if (glm::distance(thisPos, boidPos) < rule2Distance) { + separation -= boidPos - thisPos; + } + if (glm::distance(boidPos, thisPos) < rule3Distance) { + glm::vec3 boidVel = vel1[boidIndex]; + alignment += boidVel; + neighborCountRule3++; + } + } + } + + if (neighborCountRule1 != 0) { + perceived_center /= (float)neighborCountRule1; + velChange += (perceived_center - thisPos) * rule1Scale; + } + if (neighborCountRule3 != 0) { + alignment /= (float)neighborCountRule3; + velChange += alignment * rule3Scale; + } + + velChange += separation * rule2Scale; + + // - Clamp the speed change before putting the new speed in vel2 + glm::vec3 newVel = vel1[index] + velChange; + if (glm::length(newVel) > maxSpeed) { + newVel = glm::normalize(newVel) * maxSpeed; + } + vel2[index] = newVel; } /** * Step the entire N-body simulation by `dt` seconds. */ void Boids::stepSimulationNaive(float dt) { - // TODO-1.2 - use the kernels you wrote to step the simulation forward in time. - // TODO-1.2 ping-pong the velocity buffers + // TODO-1.2 - use the kernels you wrote to step the simulation forward in time. + dim3 fullBlocksPerGrid((numObjects + blockSize - 1) / blockSize); + + kernUpdateVelocityBruteForce << > > (numObjects, dev_pos, dev_vel1, dev_vel2); + kernUpdatePos << > > (numObjects, dt, dev_pos, dev_vel2); + // TODO-1.2 ping-pong the velocity buffers + glm::vec3 *temp = dev_vel1; + dev_vel1 = dev_vel2; + dev_vel2 = temp; } void Boids::stepSimulationScatteredGrid(float dt) { - // TODO-2.1 - // Uniform Grid Neighbor search using Thrust sort. - // In Parallel: - // - label each particle with its array index as well as its grid index. - // Use 2x width grids. - // - Unstable key sort using Thrust. A stable sort isn't necessary, but you - // are welcome to do a performance comparison. - // - Naively unroll the loop for finding the start and end indices of each - // cell's data pointers in the array of boid indices - // - Perform velocity updates using neighbor search - // - Update positions - // - Ping-pong buffers as needed + // TODO-2.1 + dim3 fullBlocksPerGrid((numObjects + blockSize - 1) / blockSize); + + // Uniform Grid Neighbor search using Thrust sort. + // In Parallel: + // - label each particle with its array index as well as its grid index. + // Use 2x width grids. + kernComputeIndices << > > (numObjects, gridSideCount, gridMinimum, gridInverseCellWidth, dev_pos, dev_particleArrayIndices, dev_particleGridIndices); + // - Unstable key sort using Thrust. A stable sort isn't necessary, but you + // are welcome to do a performance comparison. + thrust::device_ptr dev_thrust_particleGridIndices(dev_particleGridIndices); + thrust::device_ptr dev_thrust_particleArrayIndices(dev_particleArrayIndices); + thrust::sort_by_key(dev_thrust_particleGridIndices, dev_thrust_particleGridIndices + numObjects, dev_thrust_particleArrayIndices); + // - Naively unroll the loop for finding the start and end indices of each + // cell's data pointers in the array of boid indices + kernResetIntBuffer << > > (gridCellCount, dev_gridCellStartIndices, -1); + kernResetIntBuffer << > > (gridCellCount, dev_gridCellEndIndices, -1); + kernIdentifyCellStartEnd << > > (numObjects, dev_particleGridIndices, dev_gridCellStartIndices, dev_gridCellEndIndices); + + // - Perform velocity updates using neighbor search + kernUpdateVelNeighborSearchScattered << > > (numObjects, gridSideCount, gridMinimum, gridInverseCellWidth, gridCellWidth, dev_gridCellStartIndices, dev_gridCellEndIndices, dev_particleArrayIndices, dev_pos, dev_vel1, dev_vel2); + + // - Update positions + kernUpdatePos << > > (numObjects, dt, dev_pos, dev_vel2); + // - Ping-pong buffers as needed + glm::vec3 *temp = dev_vel1; + dev_vel1 = dev_vel2; + dev_vel2 = temp; +} + +__global__ void kernShuffleVec3Buffer(int N, int *particleArrayIndices, glm::vec3 *toShuffle, glm::vec3 *tempCopy) { + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + + int arrayIndex = particleArrayIndices[index]; + toShuffle[index] = tempCopy[arrayIndex]; +} + +__global__ void kernUnShuffleVec3Buffer(int N, int *particleArrayIndices, glm::vec3 *toShuffle, glm::vec3 *tempCopy) { + int index = threadIdx.x + (blockIdx.x * blockDim.x); + if (index >= N) { + return; + } + + int arrayIndex = particleArrayIndices[index]; + toShuffle[arrayIndex] = tempCopy[index]; } void Boids::stepSimulationCoherentGrid(float dt) { - // TODO-2.3 - start by copying Boids::stepSimulationNaiveGrid - // Uniform Grid Neighbor search using Thrust sort on cell-coherent data. - // In Parallel: - // - Label each particle with its array index as well as its grid index. - // Use 2x width grids - // - Unstable key sort using Thrust. A stable sort isn't necessary, but you - // are welcome to do a performance comparison. - // - Naively unroll the loop for finding the start and end indices of each - // cell's data pointers in the array of boid indices - // - BIG DIFFERENCE: use the rearranged array index buffer to reshuffle all - // the particle data in the simulation array. - // CONSIDER WHAT ADDITIONAL BUFFERS YOU NEED - // - Perform velocity updates using neighbor search - // - Update positions - // - Ping-pong buffers as needed. THIS MAY BE DIFFERENT FROM BEFORE. + // TODO-2.3 - start by copying Boids::stepSimulationNaiveGrid + // Uniform Grid Neighbor search using Thrust sort on cell-coherent data. + dim3 fullBlocksPerGrid((numObjects + blockSize - 1) / blockSize); + + // In Parallel: + // - Label each particle with its array index as well as its grid index. + // Use 2x width grids + kernComputeIndices << > > (numObjects, gridSideCount, gridMinimum, gridInverseCellWidth, dev_pos, dev_particleArrayIndices, dev_particleGridIndices); + + // - Unstable key sort using Thrust. A stable sort isn't necessary, but you + // are welcome to do a performance comparison. + thrust::device_ptr dev_thrust_particleGridIndices(dev_particleGridIndices); + thrust::device_ptr dev_thrust_particleArrayIndices(dev_particleArrayIndices); + thrust::sort_by_key(dev_thrust_particleGridIndices, dev_thrust_particleGridIndices + numObjects, dev_thrust_particleArrayIndices); + + // - Naively unroll the loop for finding the start and end indices of each + // cell's data pointers in the array of boid indices + kernResetIntBuffer << > > (gridCellCount, dev_gridCellStartIndices, -1); + kernResetIntBuffer << > > (gridCellCount, dev_gridCellEndIndices, -1); + kernIdentifyCellStartEnd << > > (numObjects, dev_particleGridIndices, dev_gridCellStartIndices, dev_gridCellEndIndices); + + // - BIG DIFFERENCE: use the rearranged array index buffer to reshuffle all + // the particle data in the simulation array. + // CONSIDER WHAT ADDITIONAL BUFFERS YOU NEED + kernShuffleVec3Buffer << > > (numObjects, dev_particleArrayIndices, dev_shuffled_pos, dev_pos); + kernShuffleVec3Buffer << > > (numObjects, dev_particleArrayIndices, dev_shuffled_vel, dev_vel1); + + // - Perform velocity updates using neighbor search + kernUpdateVelNeighborSearchCoherent << > > (numObjects, gridSideCount, gridMinimum, gridInverseCellWidth, gridCellWidth, dev_gridCellStartIndices, dev_gridCellEndIndices, dev_shuffled_pos, dev_shuffled_vel, dev_vel2); + + // - Update positions + kernUpdatePos << > > (numObjects, dt, dev_shuffled_pos, dev_vel2); + + // - Ping-pong buffers as needed. THIS MAY BE DIFFERENT FROM BEFORE. + kernUnShuffleVec3Buffer << > > (numObjects, dev_particleArrayIndices, dev_pos, dev_shuffled_pos); + kernUnShuffleVec3Buffer << > > (numObjects, dev_particleArrayIndices, dev_vel1, dev_vel2); } void Boids::endSimulation() { - cudaFree(dev_vel1); - cudaFree(dev_vel2); - cudaFree(dev_pos); - - // TODO-2.1 TODO-2.3 - Free any additional buffers here. + cudaFree(dev_vel1); + cudaFree(dev_vel2); + cudaFree(dev_pos); + + // TODO-2.1 TODO-2.3 - Free any additional buffers here. + cudaFree(dev_particleArrayIndices); + cudaFree(dev_particleGridIndices); + cudaFree(dev_gridCellStartIndices); + cudaFree(dev_gridCellEndIndices); + cudaFree(dev_shuffled_pos); + cudaFree(dev_shuffled_vel); } void Boids::unitTest() { - // LOOK-1.2 Feel free to write additional tests here. - - // test unstable sort - int *dev_intKeys; - int *dev_intValues; - int N = 10; - - std::unique_ptrintKeys{ new int[N] }; - std::unique_ptrintValues{ new int[N] }; - - intKeys[0] = 0; intValues[0] = 0; - intKeys[1] = 1; intValues[1] = 1; - intKeys[2] = 0; intValues[2] = 2; - intKeys[3] = 3; intValues[3] = 3; - intKeys[4] = 0; intValues[4] = 4; - intKeys[5] = 2; intValues[5] = 5; - intKeys[6] = 2; intValues[6] = 6; - intKeys[7] = 0; intValues[7] = 7; - intKeys[8] = 5; intValues[8] = 8; - intKeys[9] = 6; intValues[9] = 9; - - cudaMalloc((void**)&dev_intKeys, N * sizeof(int)); - checkCUDAErrorWithLine("cudaMalloc dev_intKeys failed!"); - - cudaMalloc((void**)&dev_intValues, N * sizeof(int)); - checkCUDAErrorWithLine("cudaMalloc dev_intValues failed!"); - - dim3 fullBlocksPerGrid((N + blockSize - 1) / blockSize); - - std::cout << "before unstable sort: " << std::endl; - for (int i = 0; i < N; i++) { - std::cout << " key: " << intKeys[i]; - std::cout << " value: " << intValues[i] << std::endl; - } - - // How to copy data to the GPU - cudaMemcpy(dev_intKeys, intKeys.get(), sizeof(int) * N, cudaMemcpyHostToDevice); - cudaMemcpy(dev_intValues, intValues.get(), sizeof(int) * N, cudaMemcpyHostToDevice); - - // Wrap device vectors in thrust iterators for use with thrust. - thrust::device_ptr dev_thrust_keys(dev_intKeys); - thrust::device_ptr dev_thrust_values(dev_intValues); - // LOOK-2.1 Example for using thrust::sort_by_key - thrust::sort_by_key(dev_thrust_keys, dev_thrust_keys + N, dev_thrust_values); - - // How to copy data back to the CPU side from the GPU - cudaMemcpy(intKeys.get(), dev_intKeys, sizeof(int) * N, cudaMemcpyDeviceToHost); - cudaMemcpy(intValues.get(), dev_intValues, sizeof(int) * N, cudaMemcpyDeviceToHost); - checkCUDAErrorWithLine("memcpy back failed!"); - - std::cout << "after unstable sort: " << std::endl; - for (int i = 0; i < N; i++) { - std::cout << " key: " << intKeys[i]; - std::cout << " value: " << intValues[i] << std::endl; - } - - // cleanup - cudaFree(dev_intKeys); - cudaFree(dev_intValues); - checkCUDAErrorWithLine("cudaFree failed!"); - return; + // LOOK-1.2 Feel free to write additional tests here. + + // test unstable sort + int *dev_intKeys; + int *dev_intValues; + int N = 10; + + std::unique_ptrintKeys{ new int[N] }; + std::unique_ptrintValues{ new int[N] }; + + intKeys[0] = 0; intValues[0] = 0; + intKeys[1] = 1; intValues[1] = 1; + intKeys[2] = 0; intValues[2] = 2; + intKeys[3] = 3; intValues[3] = 3; + intKeys[4] = 0; intValues[4] = 4; + intKeys[5] = 2; intValues[5] = 5; + intKeys[6] = 2; intValues[6] = 6; + intKeys[7] = 0; intValues[7] = 7; + intKeys[8] = 5; intValues[8] = 8; + intKeys[9] = 6; intValues[9] = 9; + + cudaMalloc((void**)&dev_intKeys, N * sizeof(int)); + checkCUDAErrorWithLine("cudaMalloc dev_intKeys failed!"); + + cudaMalloc((void**)&dev_intValues, N * sizeof(int)); + checkCUDAErrorWithLine("cudaMalloc dev_intValues failed!"); + + dim3 fullBlocksPerGrid((N + blockSize - 1) / blockSize); + + std::cout << "before unstable sort: " << std::endl; + for (int i = 0; i < N; i++) { + std::cout << " key: " << intKeys[i]; + std::cout << " value: " << intValues[i] << std::endl; + } + + // How to copy data to the GPU + cudaMemcpy(dev_intKeys, intKeys.get(), sizeof(int) * N, cudaMemcpyHostToDevice); + cudaMemcpy(dev_intValues, intValues.get(), sizeof(int) * N, cudaMemcpyHostToDevice); + + // Wrap device vectors in thrust iterators for use with thrust. + thrust::device_ptr dev_thrust_keys(dev_intKeys); + thrust::device_ptr dev_thrust_values(dev_intValues); + // LOOK-2.1 Example for using thrust::sort_by_key + thrust::sort_by_key(dev_thrust_keys, dev_thrust_keys + N, dev_thrust_values); + + // How to copy data back to the CPU side from the GPU + cudaMemcpy(intKeys.get(), dev_intKeys, sizeof(int) * N, cudaMemcpyDeviceToHost); + cudaMemcpy(intValues.get(), dev_intValues, sizeof(int) * N, cudaMemcpyDeviceToHost); + checkCUDAErrorWithLine("memcpy back failed!"); + + std::cout << "after unstable sort: " << std::endl; + for (int i = 0; i < N; i++) { + std::cout << " key: " << intKeys[i]; + std::cout << " value: " << intValues[i] << std::endl; + } + + // cleanup + cudaFree(dev_intKeys); + cudaFree(dev_intValues); + checkCUDAErrorWithLine("cudaFree failed!"); + return; } diff --git a/src/main.cpp b/src/main.cpp index b82c8c6..56f4ba8 100644 --- a/src/main.cpp +++ b/src/main.cpp @@ -14,11 +14,11 @@ // LOOK-2.1 LOOK-2.3 - toggles for UNIFORM_GRID and COHERENT_GRID #define VISUALIZE 1 -#define UNIFORM_GRID 0 -#define COHERENT_GRID 0 +#define UNIFORM_GRID 1 +#define COHERENT_GRID 1 // LOOK-1.2 - change this to adjust particle count in the simulation -const int N_FOR_VIS = 5000; +const int N_FOR_VIS = 15000; const float DT = 0.2f; /**