glCompressedTexImage2D Function

GL_APICALL void GL_APIENTRY glCompressedTexImage2D(
     GLenum target,
     GLint level,
     GLenum internalformat,
     GLsizei width,
     GLsizei height,
     GLint border,
     GLsizei imageSize,
     const void * data
);

	Name	Description
in	target	Target texture.
in	level	Mipmap level of the data to load.
in	internalformat	Texture format. (GL_ETC1_RGB8_NATIVE_DMP or GL_ETC1_RGB8_A4_NATIVE_DMP.)
in	width	Texture width.
in	height	Texture height
in	border	Border width (0 is the only valid value).
in	imageSize	Texture data size
in	data	Pointer to the texture data.

No values are returned.

Loads texture data in a compressed format.

Set target to one the following values.

GL_TEXTURE_2D	2D texture.
GL_TEXTURE_CUBE_MAP_NEGATIVE_X	Face of a cube-map texture in the negative x direction.
GL_TEXTURE_CUBE_MAP_POSITIVE_X	Face of a cube-map texture in the positive x direction.
GL_TEXTURE_CUBE_MAP_NEGATIVE_Y	Face of a cube-map texture in the negative y direction.
GL_TEXTURE_CUBE_MAP_POSITIVE_Y	Face of a cube-map texture in the positive y direction.
GL_TEXTURE_CUBE_MAP_NEGATIVE_Z	Face of a cube-map texture in the negative z direction.
GL_TEXTURE_CUBE_MAP_POSITIVE_Z	Face of a cube-map texture in the positive z direction.

To specify the region that is accessed by the GPU, set target to a bitwise OR of the aforementioned and following values.

NN_GX_MEM_FCRAM	The GPU accesses FCRAM.
NN_GX_MEM_VRAMA	The GPU accesses VRAMA.
NN_GX_MEM_VRAMB	The GPU accesses VRAMB. You can also specify whether copies of data are generated in FCRAM by including the following values in the bitwise OR.
GL_COPY_FCRAM_DMP	Copies of data are generated in FCRAM.
GL_NO_COPY_FCRAM_DMP	Copies of data are not generated in FCRAM.

These combinations cause data to be transferred as follows.

(NN_GX_MEM_FCRAM | GL_NO_COPY_FCRAM_DMP)	The GPU directly accesses the region specified by data.
(NN_GX_MEM_VRAMA | GL_NO_COPY_FCRAM_DMP)	A region is allocated in VRAMA; data is transferred to VRAMA via DMA; and then the GPU accesses VRAMA.
(NN_GX_MEM_VRAMB | GL_NO_COPY_FCRAM_DMP)	A region is allocated in VRAMB; data is transferred to VRAMB via DMA; and then the GPU accesses VRAMB.
(NN_GX_MEM_FCRAM | GL_COPY_FCRAM_DMP)	A region is allocated in FCRAM; data is copied to FCRAM via the CPU; and then the GPU accesses FCRAM.
(NN_GX_MEM_VRAMA | GL_COPY_FCRAM_DMP)	A region is allocated in VRAMA and FCRAM; data is copied to FCRAM via the CPU, then transferred from FCRAM to VRAMA via DMA; and then the GPU accesses VRAMA.
(NN_GX_MEM_VRAMB | GL_COPY_FCRAM_DMP)	A region is allocated in VRAMB and FCRAM; data is copied to FCRAM via the CPU, then transferred from FCRAM to VRAMB via DMA; and then the GPU accesses VRAMB.

If you do not specify which region is accessed by the GPU or whether data is copied to FCRAM, (NN_GX_MEM_FCRAM | GL_COPY_FCRAM_DMP) is used. (These settings cannot be specified for POD.）

Set level to a negative value that is equal to the number of mipmaps to load for the texture data stored in data. (For example, specify a value of -2 to load two mipmap levels and a value of -3 to load three mipmap levels.)

If you set level to 0 or -1, it is recognized as indicating the minimum single-level mipmap.

For internalformat, you can specify either GL_ETC1_RGB8_NATIVE_DMP or GL_ETC1_ALPHA_RGB8_A4_NATIVE_DMP.

Set width and height to the texture size of the data to load. Specify the size of level 0 when you load multiple mipmap levels of data. With GL_ETC1_RGB8_NATIVE_DMP, the texture size must be a power of 2 between 16 and 1024, inclusive.

Specify 0 for border.

Set imageSize to the number of bytes of data specified by data.

Set data to a pointer to the texture data to load. When data is loaded for multiple mipmap levels, it is stored in ascending level order starting with level 0. You cannot separately load the individual mipmap levels of the data. When data is set to NULL, this function simply allocates a region as large as the data size.

This function generates the following errors.

GL_INVALID_OPERATION	This function was called without any command-list objects bound when DMA transfers were in use.
GL_INVALID_VALUE	The width, height, internalformat, border, or level argument was set to an invalid value.
GL_INVALID_OPERATION	The imageSize argument was set to a value of 0 or less.
GL_INVALID_ENUM	The target argument was set to an invalid value.
GL_INVALID_OPERATION	The target argument was set to an invalid transfer method.

2010/11/29

Added GL_ETC1_ALPHA_RGB8_A4_NATIVE_DMP, which had been omitted by mistake, as one of the specifiable values for internalformat.

2010/01/07

Initial version.