D3d10和D3d11中Map()函数用法一点心得总结

  Directx10和Directx11很多resource类都有个Map()函数，个人感觉它是一个蛮重要的CPU和GPU沟通桥梁。下面以ID3D11DeviceContext::Map()为例讲下。

  ID3D11DeviceContext::Map官方解释是：ID3D11DeviceContext::Map Method   Get a pointer to the data contained in a subresource, and deny the GPU access to that subresource.

  Syntax HRESULT Map(     ID3D11Resource *pResource,     UINT Subresource,     D3D11_MAP MapType,     UINT MapFlags,     D3D11_MAPPED_SUBRESOURCE *pMappedResource );

  个人理解总结是当需要用CPU读写（GPU的）subresouce（最常用如buffer）时，就用Map()得到该subresource的pointer（D3D11_MAPPED_SUBRESOURCE*），然后将D3D11_MAPPED_SUBRESOURCE::pData强制转换成CPU理解的类型（struct，class），CPU就可以对得到的struct*或class*读写了。

附录D3D11_MAPPED_SUBRESOURCE结构体 typedef struct D3D11_MAPPED_SUBRESOURCE     {     void *pData;     UINT RowPitch;     UINT DepthPitch;     }  D3D11_MAPPED_SUBRESOURCE;

 

Map函数使用的几个经典场景1：debug时用CPU读写GPU之前读写的buffer（比如GPU计算输出的结果），然后再通过CPU输出来检验对错。 //将GPU读或读写的buffer *pGBuffer拷贝到可给CPU读的buffer并返回，以便调试 ID3D11Buffer* GPUBaseApp::CreateAndCopyToDebugBuf(ID3D11Buffer* pGBuffer) {  D3D11_BUFFER_DESC bufferDesc;  ZeroMemory(&bufferDesc,sizeof(bufferDesc));  pGBuffer->GetDesc(&bufferDesc);  bufferDesc.CPUAccessFlags=D3D11_CPU_ACCESS_READ;  bufferDesc.Usage=D3D11_USAGE_STAGING;  bufferDesc.BindFlags=0;  bufferDesc.MiscFlags=0;    ID3D11Buffer* pDebugBuffer=NULL;

 if( SUCCEEDED(m_pDevice->CreateBuffer(&bufferDesc,NULL,&pDebugBuffer)) )  {   m_pDeviceContext->CopyResource(pDebugBuffer,pGBuffer);  }  return pDebugBuffer; }

HRESULT Function() {               //==============================================       /

 //===============================================  //看Direct Compute的计算结果  ID3D11Buffer* debugOutBuffer=NULL;  ID3D11Buffer* debugInBuffer=NULL;

 debugOutBuffer=CreateAndCopyToDebugBuf(m_pOutBuffer);  debugInBuffer=CreateAndCopyToDebugBuf(m_pInputBuffer);

 VoronoiOutBufType* pOut=NULL;  VoronoiInBufType* pIn=NULL;

 D3D11_MAPPED_SUBRESOURCE mappedOutResource;  D3D11_MAPPED_SUBRESOURCE mappedInResource;

 m_pDeviceContext->Map(debugOutBuffer,0,D3D11_MAP_READ,0,&mappedOutResource);  pOut=(VoronoiOutBufType*)mappedOutResource.pData;

 m_pDeviceContext->Map(debugInBuffer,0,D3D11_MAP_READ,0,&mappedInResource);  pIn=(VoronoiInBufType*)mappedInResource.pData;

 //printf("OutBuffer的结果如下:/n");  //for(int i=0;i<30;i++)  // printf("%d %f/n",pOut[i].index,pOut[i].dist);

 m_pDeviceContext->Unmap(debugOutBuffer,0);  SAFE_RELEASE(debugOutBuffer);

 m_pDeviceContext->Unmap(debugInBuffer,0);  SAFE_RELEASE(debugInBuffer);

 //===============================================

 

              //==============================================       /

}

Map函数使用的几个经典场景2：更新GPU Shader中Constant buffer，把CPU中新的数据struct CB_MouseInfo mouseInfo写入GPU中Constant Buffer。 struct CB_MouseInfo {  UINT x;  UINT y;  UINT btDown; };

CB_MouseInfo mouseInfo；

HRESULT GPUBaseApp::UpDateMouseInfoCB() {  HRESULT hr=S_OK;

 D3D11_MAPPED_SUBRESOURCE subData;

 ZeroMemory(&subData,sizeof(subData));

 m_pDeviceContext->Map(m_pConstantBuffer,0,D3D11_MAP_WRITE_DISCARD,0,&subData);

 memcpy(subData.pData,&mouseInfo,sizeof(CB_MouseInfo));

 m_pDeviceContext->Unmap(&subData,0);

 return hr; }

注意点，当调用了Map()，最后记得调用ID3D11DeviceContext::Unmap Method使得GPU重新获取对该resouce的读写权。 Invalidate the pointer to a resource and re-enable the GPU's access to that resource.

Syntax void Unmap(     ID3D11Resource *pResource,     UINT Subresource );

Parameters：

pResource ID3D11Resource A pointer to a ID3D11Resource interface.

Subresource UINT A subresource to be unmapped.