Bring your game to Mac, Part 3: Render with Metal

3:55 - Encode the texture tables.

// Encode the texture tables outside of the rendering loop.


id<MTLBuffer> textureTable  = [device newBufferWithLength:sizeof(MTLResourceID) * texturesCount
                                                  options:MTLResourceStorageModeShared];


MTLResourceID* textureTableCPUPtr = (MTLResourceID*)textureTable.contents;
for (uint32_t i = 0; i < texturesCount; ++i)
{
    // create the textures.
    id<MTLTexture> texture = [device newTextureWithDescriptor:textureDesc[i]];

    // encode texture in argument buffer
    textureTableCPUPtr[i] = texture.gpuResourceID;
}

4:33 - Encode the sampler tables.

// Encode the sampler tables outside of the rendering loop.


id<MTLBuffer> samplerTable  = [device newBufferWithLength:sizeof(MTLResourceID) * samplersCount
                                                  options:MTLResourceStorageModeShared];

MTLResourceID* samplerTableCPUPtr = (MTLResourceID*)samplerTable.contents;
for (uint32_t i = 0; i < samplersCount; ++i)
{
    // create sampler descriptor
    MTLSamplerDescriptor* desc  = [MTLSamplerDescriptor new];
    desc.supportArgumentBuffers = YES;
    . . .

    // create a sampler
    id<MTLSamplerState> sampler = [device newSamplerStateWithDescriptor:desc];

    // encode the sampler in argument buffer
    samplerTableCPUPtr[i] = sampler.gpuResourceID;
}

5:05 - Encode the top level argument buffer.

// Encode the top level argument buffer.


struct TopLevelAB
{
    MTLResourceID* textureTable;
    float*         myBuffer;
    uint32_t       myConstant;
    MTLResourceID* samplerTable;
};

id<MTLBuffer> topAB = [device newBufferWithLength:sizeof(TopLevelAB)
                                          options:MTLResourceStorageModeShared];


TopLevelAB* topABCPUPtr     = (TopLevelAB*)topAB.contents;
topABCPUPtr->textureTable   = (MTLResourceID*)textureTable.gpuAddress;
topABCPUPtr->myBuffer       = (float*)myBuffer.gpuAddress;
topABCPUPtr->myConstant     = 128;
topABCPUPtr->samplerTable   = (MTLResourceID*)samplerTable.gpuAddress;

6:49 - Allocate the read-only resources.

// Allocate the read-only resources from a heap.

MTLHeapDescriptor* heapDesc = [MTLHeapDescriptor new];
heapDesc.size               = requiredSize;
heapDesc.type               = MTLHeapTypeAutomatic;

id<MTLHeap> heap = [device newHeapWithDescriptor:heapDesc];



// Allocate the textures and the buffers from the heap.

id<MTLTexture> texture = [heap newTextureWithDescriptor:desc];
id<MTLBuffer>  buffer = [heap newBufferWithLength:length options:options];
. . .


// Make the heap resident once for each encoder that uses it.

[encoder useHeap:heap];

7:34 - Allocate the writable resources.

// Allocate the writable resources individually.

id<MTLTexture> textureRW = [device newTextureWithDescriptor:desc];
id<MTLBuffer>  bufferRW  = [device newBufferWithLength:length options:options];



// Mark these resources resident when they're needed in the current encoder.
// Specify the resource usage in the encoder using MTLResourceUsage.

[encoder useResource:textureRW usage:MTLResourceUsageWrite stages:stage];
[encoder useResource:bufferRW  usage:MTLResourceUsageRead  stages:stage];

19:31 - Encode the execute indirect

// Encode the execute indirect command as a series of indirect draw calls.

for (uint32_t i = 0; i < maxDrawCount; ++i)
{
    // Encode the current indirect draw call.
    [renderEncoder drawIndexedPrimitives:MTLPrimitiveTypeTriangle
                       				 indexType:MTLIndexTypeUInt16
                             indexBuffer:indexBuffer
                       indexBufferOffset:indexBufferOffset
                          indirectBuffer:drawArgumentsBuffer
                    indirectBufferOffset:drawArgumentsBufferOffset];
    
    // Advance the draw arguments buffer offset to the next indirect arguments.
    drawArgumentsBufferOffset += sizeof(MTLDrawIndexedPrimitivesIndirectArguments);
}

21:48 - Translate the indirect draw arguments to ICB.

// Kernel written in Metal Shading Language to translate the indirect draw arguments to an ICB. 


kernel void translateToICB(device const Command* indirectCommands [[ buffer(0) ]],
                           device const ICBContainerAB* icb [[ buffer(1) ]],
                           . . .)
{
    . . .
   
    device const Command* indirectCommand = &indirectCommands[commandIndex];
    device const MTLDrawIndexedPrimitivesIndirectArguments* args =
    &command->mdiBuffer[mdiIndex];
    
    render_command drawCall(icb->buffer, indirectCommand->mdiCmdStart + mdiIndex);

    if(args->indexCount > 0 && args->instanceCount > 0) {
        encodeCommand(indirectCommand, args, drawCall);
    }
    else {
        cmd.reset();
    }
}

// Encode a render command on the GPU.
void encodeCommand(device const Command* indirectCommand,
                   device const MTLDrawIndexedPrimitivesIndirectArguments* args,
                   thread render_command& drawCall)
{
    drawCall.set_render_pipeline_state(indirectCommand->pso);
    
    for(ushort i = 0; i < indirectCommand->vertexBuffersCount; ++i) {
        drawCall.set_vertex_buffer(indirectCommand->vertexBuffer[i].buffer,
                              indirectCommand->vertexBuffer[i].slot);
    }
    
    for(ushort i = 0; i < indirectCommand->fragmentBuffersCount; ++i) {
        drawCall.set_fragment_buffer(indirectCommand->fragmentBuffer[i].buffer,
                                indirectCommand->fragmentBuffer[i].slot);
    }

    drawCall.draw_indexed_primitives(primitive_type::triangle,
                                args->indexCount,
                                indirectCommand->indexBuffer + args->indexStart,
                                args->instanceCount,
                                args->baseVertex,
                                args->baseInstance);
}