ID3D10Texture1D::Map method

Get a pointer to the data contained in a subresource, and deny the GPU access to that subresource.


  [in]  UINT      Subresource,
  [in]  D3D10_MAP MapType,
  [in]  UINT      MapFlags,
  [out] void      **ppData


Subresource [in]

Type: UINT

Index number of the subresource. See D3D10CalcSubresource for more details.

MapType [in]

Type: D3D10_MAP

Specifies the CPU's read and write permissions for a resource. For possible values, see D3D10_MAP.

MapFlags [in]

Type: UINT

Flag that specifies what the CPU should do when the GPU is busy. This flag is optional.

ppData [out]

Type: void**

Pointer to the texture resource data.

Return value


If this function succeeds, it returns S_OK. For other restrictions, and a listing of error values that can be returned by any of the Map methods, see Remarks.


Mapping a texture enables the CPU to directly access the underlying data in the subresource of a texture. For the method to succeed, the texture being mapped must be created with the appropriate flags (see D3D10_CPU_ACCESS_FLAG), and its specified usage (see D3D10_USAGE) must be either D3D10_USAGE_DYNAMIC or D3D10_USAGE_STAGING.

Common failures of Map methods are indicated by the following return values:



If MapFlags specifies D3D10_MAP_FLAG_DO_NOT_WAIT and the GPU is not yet finished with the resource, Map returns DXGI_ERROR_WAS_STILL_DRAWING.


Map returns DXGI_ERROR_DEVICE_REMOVED if MapType allows any CPU read access and the video card has been removed.


For more information about the preceding return values, see DXGI_ERROR.

Map has the following restrictions:

  • A single subresource cannot be mapped multiple times; in other words, do not call Map on a subresource that is already mapped.
  • Any subresource that is bound to the pipeline must be unmapped before any render operation (that is, before ID3D10Device::Draw) can be executed.

Applications must cast the void pData pointer to the appropriate type to meaningfully access the underlying subresource data. For example, the following code demonstrates how to read each texel of a 1D subresource. It is assumed that the texture was created using DXGI_FORMAT_R32G32B32A32_FLOAT and that pData is the pointer to the texture resource data output from a successful call to this method.

FLOAT* pTexels = (FLOAT*)pData;
for( UINT col = 0; col < width; col++ )
  pTexels[col*4 + 0]; // Alpha
  pTexels[col*4 + 1]; // Blue
  pTexels[col*4 + 2]; // Green
  pTexels[col*4 + 3]; // Red

Differences between Direct3D 9 and Direct3D 10:

Map in Direct3D 10 is analogous to resource Lock in Direct3D 9.







See also

ID3D10Texture1D Interface