Functions
VXLIB_channelCombine_yuyv_i8u_o8u

Functions

VXLIB_STATUS VXLIB_channelCombine_yuyv_i8u_o8u (const uint8_t src0[restrict], const VXLIB_bufParams2D_t *src0_addr, const uint8_t src1[restrict], const VXLIB_bufParams2D_t *src1_addr, const uint8_t src2[restrict], const VXLIB_bufParams2D_t *src2_addr, uint8_t dst[restrict], const VXLIB_bufParams2D_t *dst_addr, int8_t yidx)
 
VXLIB_STATUS VXLIB_channelCombine_yuyv_i8u_o8u_checkParams (const uint8_t src0[], const VXLIB_bufParams2D_t *src0_addr, const uint8_t src1[], const VXLIB_bufParams2D_t *src1_addr, const uint8_t src2[], const VXLIB_bufParams2D_t *src2_addr, const uint8_t dst[], const VXLIB_bufParams2D_t *dst_addr, int8_t yidx)
 

Detailed Description

Function Documentation

◆ VXLIB_channelCombine_yuyv_i8u_o8u()

VXLIB_STATUS VXLIB_channelCombine_yuyv_i8u_o8u ( const uint8_t  src0[restrict],
const VXLIB_bufParams2D_t *  src0_addr,
const uint8_t  src1[restrict],
const VXLIB_bufParams2D_t *  src1_addr,
const uint8_t  src2[restrict],
const VXLIB_bufParams2D_t *  src2_addr,
uint8_t  dst[restrict],
const VXLIB_bufParams2D_t *  dst_addr,
int8_t  yidx 
)
Description:
Implements the Channel Combine Kernel.
Method:
The YUYV Channel Combine is implemented using the following equation:
   dst[x + yidx]            = src0[x]
   dst[x + (1 - yidx)]      = src1[x]
   dst[x + 2 + yidx]        = src0[x + 1]
   dst[x + 2 + (1 - yidx)]  = src2[x]
Parameters
[in]src0[]Pointer to array containing first input image (UQ8.0)
[in]src0_addr[]Pointer to structure containing dimensional information of src0
[in]src1[]Pointer to array containing second input image (UQ8.0)
[in]src1_addr[]Pointer to structure containing dimensional information of src1
[in]src2[]Pointer to array containing third input image (UQ8.0)
[in]src2_addr[]Pointer to structure containing dimensional information of src2
[out]dst[]Pointer to array containing output image (UQ8.0)
[in]dst_addr[]Pointer to structure containing dimensional information of dst
[in]yidxParameter indicating how images should be interleaved (0: luma is in even byte offsets; 1: luma is in odd byte offsets)
Assumptions:
  • I/O buffer pointers are assumed to be not aliased.
Performance Considerations:
  • For best performance, the following parameter settings are recommended:
    • Set dst_addr->stride_y equal to src0_addr->stride_y * 2
    • Set src0_addr->stride_y and src1_addr->stride_y equal to src0_addr->stride_y/2
    • Align all pointers to 8 byte boundaries
    • Set all stride values to a multiple of 16
    • Set all width values to a multiple of 16

◆ VXLIB_channelCombine_yuyv_i8u_o8u_checkParams()

VXLIB_STATUS VXLIB_channelCombine_yuyv_i8u_o8u_checkParams ( const uint8_t  src0[],
const VXLIB_bufParams2D_t *  src0_addr,
const uint8_t  src1[],
const VXLIB_bufParams2D_t *  src1_addr,
const uint8_t  src2[],
const VXLIB_bufParams2D_t *  src2_addr,
const uint8_t  dst[],
const VXLIB_bufParams2D_t *  dst_addr,
int8_t  yidx 
)
Description:
Checks the parameters for programming errors for the VXLIB_channelCombine_yuyv_i8u_o8u function.
Method:
The following checks are made:
  • There shall be no NULL pointers
  • Both images shall have the same x and y dimensions
  • The stride_y of the src images shall be equal to or greater than dim_x
  • The stride_y of the dst image shall be equal to or greater than dim_x * 2
  • The yidx shall be either 0 or 1.
  • The dst image data type shall be set to VXLIB_UINT16.
Parameters
[in]src0[]Pointer to array containing first input image (UQ8.0)
[in]src0_addr[]Pointer to structure containing dimensional information of src0
[in]src1[]Pointer to array containing second input image (UQ8.0)
[in]src1_addr[]Pointer to structure containing dimensional information of src1
[in]src2[]Pointer to array containing third input image (UQ8.0)
[in]src2_addr[]Pointer to structure containing dimensional information of src2
[out]dst[]Pointer to array containing output image (UQ8.0)
[in]dst_addr[]Pointer to structure containing dimensional information of dst
[in]yidxParameter indicating how images should be interleaved (0: luma is in even byte offsets; 1: luma is in odd byte offsets)

Copyright 2023, Texas Instruments Incorporated