Skip to content

Latest commit

 

History

History
3198 lines (2363 loc) · 141 KB

Rockchip_Developer_Guide_RGA_EN.md

File metadata and controls

3198 lines (2363 loc) · 141 KB

RGA IM2D API Instruction

ID: RK-KF-YF-403

Release Version: V2.2.3

Release Date: 2024-03-06

Security Level: □Top-Secret □Secret □Internal ■Public

DISCLAIMER

THIS DOCUMENT IS PROVIDED “AS IS”. ROCKCHIP ELECTRONICS CO., LTD.(“ROCKCHIP”)DOES NOT PROVIDE ANY WARRANTY OF ANY KIND, EXPRESSED, IMPLIED OR OTHERWISE, WITH RESPECT TO THE ACCURACY, RELIABILITY, COMPLETENESS,MERCHANTABILITY, FITNESS FOR ANY PARTICULAR PURPOSE OR NON-INFRINGEMENT OF ANY REPRESENTATION, INFORMATION AND CONTENT IN THIS DOCUMENT. THIS DOCUMENT IS FOR REFERENCE ONLY. THIS DOCUMENT MAY BE UPDATED OR CHANGED WITHOUT ANY NOTICE AT ANY TIME DUE TO THE UPGRADES OF THE PRODUCT OR ANY OTHER REASONS.

Trademark Statement

"Rockchip", "瑞芯微", "瑞芯" shall be Rockchip’s registered trademarks and owned by Rockchip. All the other trademarks or registered trademarks mentioned in this document shall be owned by their respective owners.

All rights reserved. ©2022. Rockchip Electronics Co., Ltd.

Beyond the scope of fair use, neither any entity nor individual shall extract, copy, or distribute this document in any form in whole or in part without the written approval of Rockchip.

Rockchip Electronics Co., Ltd.

No.18 Building, A District, No.89, software Boulevard Fuzhou, Fujian,PRC

Website: www.rock-chips.com

Customer service Tel: +86-4007-700-590

Customer service Fax: +86-591-83951833

Customer service e-Mail: fae@rock-chips.com


Intended Audience

This document (this guide) is mainly intended for:

  • Technical support engineers
  • Software development engineers

Revision History

Date Version Author Revision Description
2020/06/24 1.0.0 Chen Cheng, Li Huang Initial version.
2020/10/16 1.0.1 Chen Cheng, Li Huang, Yu Qiaowei Update part of the APIs.
2021/12/07 2.0.0 Chen Cheng, Li Huang, Yu Qiaowei Add RGA3 related support.
2022/01/20 2.1.0 Chen Cheng, Li Huang, Yu Qiaowei - Update im2d api description.
- Updated hardware index description and alignment restrictions.
- Add data structure description.
2022/01/20 2.1.1 Chen Cheng, Li Huang, Yu Qiaowei Supplemental formatting support/alignment instructions.
2022/09/15 2.2.0 Chen Cheng, Li Huang, Yu Qiaowei - Supplementary default value description
- New array api
- New task api
- New rectangle border drawing api
2023/02/09 2.2.1 Yu Qiaowei Format document.
2023/06/28 2.2.2 Yu Qiaowei - Add chip RK3562 introduction
- Improve the precautions for grayscale images
2024/03/06 2.2.3 Yu Qiaowei Add chip RK3576 introduction

Contents

[TOC]


Introductions

RGA (Raster Graphic Acceleration Unit) is an independent 2D hardware accelerator that can be used to speed up point/line drawing, perform image resizing, rotation, bitBlt, alpha blending and other common 2D graphics operations.

Design Index

Version Codename Chip Source Destination Function Pixels/Cycle
min max min max
RGA1 Pagani RK3066 2x2 8192x8192 2x2 2048x2048 90/180/270 Rotate
X/Y Mirror
Crop
1/2~8 scale
Alpha blend
Color key
Color fill
ROP
IOMMU(32bit)
1
Jaguar Plus RK3188
Beetles RK2926/2928
Beetles Plus RK3026/3028
RGA1_plus Audi RK3128 2x2 8192x8192 2x2 2048x2048 90/180/270 Rotate
X/Y Mirror
Crop
1/2~8 scale
Alpha blend
Color key
Color fill
Color palette
IOMMU(32bit)
1
Granite Sofia 3gr
RGA2 Lincoln RK3288/3288w 2x2 8192x8192 2x2 4096x4096 90/180/270 Rotate
X/Y Mirror
Crop
1/16~16 scale
scale-up(bi-linear/bi-cubic)
scale-down(average)
Alpha blend
Color key
Color fill
Color palette
ROP
IOMMU(32bit)
2
Capricorn RK3190
RGA2-Lite0 Maybach RK3368 2x2 8192x8192 2x2 4096x4096 90/180/270 Rotate
X/Y Mirror
Crop
1/8~8 scale
scale-up(bi-linear/bi-cubic)
scale-down(average)
Alpha blend
Color key
Color fill
Color palette
ROP
IOMMU(32bit)
2
BMW RK3366
RGA2-Lite1 Benz RK3228 2x2 8192x8192 2x2 4096x4096 90/180/270 Rotate
X/Y Mirror
Crop
1/8~8 scale
scale-up(bi-linear/bi-cubic)
scale-down(average)
Alpha blend
Color key
Color fill
Color palette
IOMMU(32bit)
2
Infiniti RK3228H
Gemini RK3326
Lion RK1808
RGA2-Enhance Mclaren RK3399 2x2 8192x8192 2x2 4096x4096 90/180/270 Rotate
X/Y Mirror
Crop
1/16~16 scale
scale-up(bi-linear/bi-cubic)
scale-down(average)
Alpha blend
Color key
Color fill
Color palette
ROP(NA for RV1108/RV1109/RK3566)
NN quantize(NA for RK3399/RV1108)
osd (only RV1106/RV1103/RK3562/RK3528)
mosaic(only RV1106/RV1103/RK3562/RK3528)
IOMMU(32bit, RK3528/RK3562为40bit,NA for RV1106/1103)
2
Mercury RK1108
Puma RV1126/RV1109
skylarkV2 RK3566/RK3568
Orion RK3588
Otter RV1106/1103
Bull RK3528
Snipe RK3562
RGA2-Pro Heron RK3576 2x2 8192x8192 2x2 8192x8192 90/180/270 Rotate
X/Y Mirror
Crop
1/16~16 scale
scale-up(bi-linear/bi-cubic)
scale-down(bi-linear/average)
Alpha blend
Color key
Color fill
Color palette
ROP
osd
mosaic
ARGB5551 alpha bit map
rkfbc64x4(only input)
afbc32x8 splice mode(only input)
tile4x4
IOMMU(40bit)
2
RGA3 Orion RK3588 68x2 8176x8176 68x2 8128x8128 90/180/270 Rotate
X/Y Mirror
Crop
1/8~8 scale
scale-up(bi-cubic)
scale-down(average)
Alpha blend
Color key
afbc16x16
IOMMU(40bit)
3 (by pass)
2 (scale)

Note:

1). The capabilities of Pixels/cycle are theoretical data, and the actual operating performance is related to bandwidth, hardware frequency, etc. The list data is for reference only.

2). In addition to the minimum input resolution limit, the x, y, width, and height parameters of the actual operation rectangle that can be set for each channel must be greater than or equal to 2.

3). The addressing capability of RGA is related to the number of bits of IOMMU. For example, the actual physical addressing capability of RGA equipped with 32bit IOMMU only supports 0~4G memory space.

Image Format Supported

  • Pixel Format conversion, BT.601/BT.709/BT.2020(only RGA3)
  • Dither operation
Version Codename Chip Input Data Format Output Data Format
RGA1 Pagani RK3066 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888

RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_BPP1
RK_FORMAT_BPP2
RK_FORMAT_BPP4
RK_FORMAT_BPP8
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP (only for Blur/sharpness)
RK_FORMAT_YCrCb_420_SP (only for Blur/sharpness)
RK_FORMAT_YCbCr_422_SP (only for Blur/sharpness)
RK_FORMAT_YCrCb_422_SP (only for Blur/sharpness)
RK_FORMAT_YCbCr_420_P (only for Blur/sharpness)
RK_FORMAT_YCrCb_420_P (only for Blur/sharpness)
RK_FORMAT_YCbCr_422_P (only for Blur/sharpness)
RK_FORMAT_YCrCb_422_P (only for Blur/sharpness)
Jaguar Plus RK3188
Beetles RK2926/2928
Beetles Plus RK3026/3028
RGA1_plus Audi RK3128 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_BPP1
RK_FORMAT_BPP2
RK_FORMAT_BPP4
RK_FORMAT_BPP8
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP (only for normal Bitblt without alpha)
RK_FORMAT_YCrCb_420_SP (only for normal Bitblt without alpha)
RK_FORMAT_YCbCr_422_SP (only for normal Bitblt without alpha)
RK_FORMAT_YCrCb_422_SP (only for normal Bitblt without alpha)
RK_FORMAT_YCbCr_420_P (only for normal Bitblt without alpha)
RK_FORMAT_YCrCb_420_P (only for normal Bitblt without alpha)
RK_FORMAT_YCbCr_422_P (only for normal Bitblt without alpha)
RK_FORMAT_YCrCb_422_P (only for normal Bitblt without alpha)
Granite Sofia 3gr
RGA2 Lincoln RK3288/3288w RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_BPP1 (only for color palette)
RK_FORMAT_BPP2 (only for color palette)
RK_FORMAT_BPP4 (only for color palette)
RK_FORMAT_BPP8 (only for color palette)
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
Capricorn RK3190
RGA2-Lite0 Maybach RK3368 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_BPP1 (only for color palette)
RK_FORMAT_BPP2 (only for color palette)
RK_FORMAT_BPP4 (only for color palette)
RK_FORMAT_BPP8 (only for color palette)
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
BMW RK3366
RGA2-Lite1 Benz RK3228 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_RGBA_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
RK_FORMAT_BPP1 (only for color palette)
RK_FORMAT_BPP2 (only for color palette)
RK_FORMAT_BPP4 (only for color palette)
RK_FORMAT_BPP8 (only for color palette)
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
Infiniti RK3228H
Gemini RK3326
Lion RK1808
RGA2-Enhance Mclaren RK3399 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
RK_FORMAT_BPP1 (only for color palette)
RK_FORMAT_BPP2 (only for color palette)
RK_FORMAT_BPP4 (only for color palette)
RK_FORMAT_BPP8 (only for color palette)
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
Mercury RK1108
Puma RV1126/ RV1109 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YCbCr_400
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
RK_FORMAT_BPP1 (only for color palette)
RK_FORMAT_BPP2 (only for color palette)
RK_FORMAT_BPP4 (only for color palette)
RK_FORMAT_BPP8 (only for color palette)
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YCbCr_400
RK_FORMAT_Y4
skylarkV2 RK3566/RK3568
Orion RK3588
Otter RV1106/1103
Bull RK3528
Snipe RK3562
RGA2-Pro Heron RK3576 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_444_SP
RK_FORMAT_YCrCb_444_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YCbCr_400
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
RK_FORMA_A8 (only src for alpha blend)
RK_FORMAT_BPP1 (only for color palette)
RK_FORMAT_BPP2 (only for color palette)
RK_FORMAT_BPP4 (only for color palette)
RK_FORMAT_BPP8 (only for color palette)
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_444_SP
RK_FORMAT_YCrCb_444_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YCbCr_400
RK_FORMAT_Y4
RK_FORMAT_Y8
RGA3 Orion RK3588 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B

Note:

1). The "RK_FORMAT_YCbCr_400" format means that the YUV format only takes the Y channel, and is often used in 256 (2 to the 8th power) grayscale images. Here, it should be noted that since the YUV format exists in the RGB/YUV color space conversion, you need to pay attention to the color space configuration , for example, a full 256-level grayscale image needs to be configured as full range during conversion.

2). The "RK_FORMAT_Y4" format means that the YUV format only takes the Y channel and dithers to 4 bits. It is often used in 16 (2 to the 4th power) grayscale images. The precautions involved in the configuration of the color space conversion are the same as “RK_FORMAT_YCbCr_400”. "RK_FORMAT_Y8" is similar to "RK_FORMAT_Y4", also only 4 bits are valid, the difference is that only the high 4 bits are valid data, the low 4 bits are invalid data.

Image Format Alignment Instructions

Version Byte_stride Format Alignment
RGA1
RGA1-Plus
4 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
width stride does not require alignment
RK_FORMAT_RGBA_4444
RK_FORMAT_BGRA_4444
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_RGBA_5551
RK_FORMAT_BGRA_5551
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
width stride must be 2-aligned
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
width stride must be 4-aligned
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
width stride must be 4-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
RGA2
RGA2-Lite0
RGA2-Lite1
RGA2-Enhance
RGA2-Pro
4 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
width stride does not require alignment
RK_FORMAT_RGBA_4444
RK_FORMAT_BGRA_4444
RK_FORMAT_ARGB_4444
RK_FORMAT_ABGR_4444
RK_FORMAT_RGBA_5551
RK_FORMAT_BGRA_5551
RK_FORMAT_ARGB_5551
RK_FORMAT_ABGR_5551
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
width stride must be 2-aligned
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
RK_FORMAT_YUYV_420
RK_FORMAT_YVYU_420
RK_FORMAT_UYVY_420
RK_FORMAT_VYUY_420
width stride must be 2-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
RK_FORMAT_A8
width stride must be 4-aligned
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
RK_FORMAT_YCbCr_444_SP
RK_FORMAT_YCrCb_444_SP
RK_FORMAT_YCbCr_420_P
RK_FORMAT_YCrCb_420_P
RK_FORMAT_YCbCr_422_P
RK_FORMAT_YCrCb_422_P
RK_FORMAT_YCbCr_400
RK_FORMAT_Y4
RK_FORMAT_Y8
width stride must be 4-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
width stride must be 16-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
RGA3 16 RK_FORMAT_RGBA_8888
RK_FORMAT_BGRA_8888
RK_FORMAT_ARGB_8888
RK_FORMAT_ABGR_8888
RK_FORMAT_RGBX_8888
RK_FORMAT_BGRX_8888
RK_FORMAT_XRGB_8888
RK_FORMAT_XBGR_8888
width stride must be 4-aligned
RK_FORMAT_RGB_565
RK_FORMAT_BGR_565
width stride must be 8-aligned
RK_FORMAT_YUYV_422
RK_FORMAT_YVYU_422
RK_FORMAT_UYVY_422
RK_FORMAT_VYUY_422
width stride must be 8-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
RK_FORMAT_RGB_888
RK_FORMAT_BGR_888
width stride must be 16-aligned
RK_FORMAT_YCbCr_420_SP
RK_FORMAT_YCrCb_420_SP
RK_FORMAT_YCbCr_422_SP
RK_FORMAT_YCrCb_422_SP
width stride must be 16-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
RK_FORMAT_YCbCr_420_SP_10B
RK_FORMAT_YCrCb_420_SP_10B
RK_FORMAT_YCbCr_422_SP_10B
RK_FORMAT_YCrCb_422_SP_10B
width stride must be 64-aligned,x_offset、y_offset、width、height、height stride must be 2-aligned
FBC mode In addition to the format alignment requirements above,width stride、height stride must be 16-aligned
TILE8*8 mode In addition to the format alignment requirements above,width、height must be 8-aligned,input channel width stride、height stride must be 16-aligned。

Note:

1). Alignment requirement formula: lcm(bpp,byte_stride * 8) / pixel_stride.

2). When loaded with multiple versions of hardware, chip platform constraints according to the most strict alignment requirements.


Version Description

RGA's support library librga.so updates the version number according to certain rules, indicating the submission of new features, compatibility, and bug fixes, and provides several ways to query the version number, so that developers can clearly determine whether the current library version is suitable for the current development environment when using librga.so. Detailed version update logs can be found in CHANGLOG.md in the root directory of source code.

librga API Version Description

Version Number Format and Update Rule

Version Number Format
major.minor.revision_[build]

example:

1.0.0_[0]

Update Rule
Name Rule
major Major version number, when submitting a version that is not backward compatible.
minor Minor version number, when the functional API with backward compatibility is added.
revision Revision version number, when submitting backward compatible feature additions or fatal bug fixes.
build Compile version number, when backward compatibility issue is fixed.

Version Number Query

Query by Strings

Take Android R 64bit as example:

:/# strings vendor/lib64/librga.so |grep rga_api |grep version
rga_api version 1.0.0_[0]
Log Print

Version number is printed when each process first calls RGA API.

rockchiprga: rga_api version 1.0.0_[0]
Query by API

Call the following API to query the code version number, compilation version number, and RGA hardware version information. For details, see API Description.

querystring(RGA_VERSION);

String format is as follows:

RGA_api version : v1.0.0_[0] RGA version : RGA_2_Enhance

Query by Property

This method is supported only by Android system, and the property takes effect only after an existing process calls RGA.

:/# getprop |grep rga
[vendor.rga_api.version]: [1.0.0_[0]]

Driver Version Description

librga calls the RGA hardware based on the driver, it must be ensured that the driver version is within the supported range of the used librga library.

Version Number Format and Update Rule

Version Number Format
<driver_name>: v major.minor.revision

example:

RGA2 Device Driver: v2.1.0

RGA multicore Device Driver: v1.2.23

Update Rule
Name Rule
major Major version number, when submitting a version that is not backward compatible.
minor Minor version number, when the functional API with backward compatibility is added.
revision Revision version number, when submitting backward compatible feature additions or fatal bug fixes.

Version Number Query

Boot log query:

Use the following command to query the RGA driver initialization log after booting. Some early drivers do not print the version number, and this method is only applicable to some drivers.

 dmesg |grep rga

example:

[ 2.382393] rga3_core0 fdb60000.rga: Adding to iommu group 2 [ 2.382651] rga: rga3_core0, irq = 33, match scheduler [ 2.383058] rga: rga3_core0 hardware loaded successfully, hw_version:3.0.76831. [ 2.383121] rga: rga3_core0 probe successfully [ 2.383687] rga3_core1 fdb70000.rga: Adding to iommu group 3 [ 2.383917] rga: rga3_core1, irq = 34, match scheduler [ 2.384313] rga: rga3_core1 hardware loaded successfully, hw_version:3.0.76831. [ 2.384412] rga: rga3_core1 probe successfully [ 2.384893] rga: rga2, irq = 35, match scheduler [ 2.385238] rga: rga2 hardware loaded successfully, hw_version:3.2.63318. [ 2.385257] rga: rga2 probe successfully [ 2.385455] rga_iommu: IOMMU binding successfully, default mapping core[0x1] [ 2.385586] rga: Module initialized. v1.2.23

Among them, “v1.2.23” is the driver version number.

debug node query

The version number can be queried through the driver debugging node. If there is no following node, it means that the driver version that does not support query is currently running.

  • Use the kernel with the CONFIG_ROCKCHIP_RGA_DEBUG_FS compile config enabled by default.
cat /sys/kernel/debug/rkrga/driver_version
  • Use the kernel with the CONFIG_ROCKCHIP_RGA_PROC_FS compile config enabled.
cat /proc/rkrga/driver_version

example:

cat /sys/kernel/debug/rkrga/driver_version RGA multicore Device Driver: v1.2.23

Here "RGA multicore Device Driver" means that the driver name is RGA multicore Device Driver, and "v1.2.23" means that the version is 1.2.23, which means that the current driver is the RGA multicore Device Driver(usually referred to as multi_rga driver) of version 1.2.23 .

cat /sys/kernel/debug/rkrga/driver_version RGA2 Device Driver: v2.1.0

Here "RGA2 Device Driver" means that the driver name is RGA2 Device Driver, and "v2.1.0" means that the version number is 2.1.0, which means that the current driver is the RGA2 Device Driver (usually referred to as rga2 driver) of version 2.1.0.

Correspondence between versions

When using RGA, you need to confirm that the current operating environment can work normally. The following table shows the correspondence between commonly used librga and driver versions.

librga version driver version hardware
no version number Drivers in the SDK RGA1、RGA2
1.0.0 ~ 1.3.2 RGA Device Driver(kernel - 4.4 and above)
RGA2 Device Driver(no version number or v2.1.0)
RGA1、RGA2
> 1.4.0 RGA multicore Device Driver(v1.2.0and above) RGA2、RGA3
> 1.9.0 RGA Device Driver(kernel-4.4 and above)
RGA2 Device Driver(no version number or v2.1.0)
RGA multicore Device Driver(v1.2.0and above)
RGA1、RGA2、RGA3

API Description

RGA library librga.so realizes 2D graphics operations through the image buffer structure rga_info configuration. In order to obtain a better development experience, the common 2D image operation API is further encapsulated. The new API mainly contains the following features:

  • API definitions refer to common 2D graphics API definitions in opencv/matlab, reducing the learning cost of secondary development.
  • To eliminate compatibility problems caused by RGA hardware version differences, RGA query is added. Query mainly includes version information, large resolution and image format support.
  • Add improcess API for 2D image compound operations. Compound operations are performed by passing in a set of predefined usage.
  • Before performing image operation, the input and output image buffers need to be processed. The wrapbuffer_T API is called to pass the input and output image information to rga_buffer_t, which contains information such as resolution and image format.
  • It supports to bind the image composite operation that cannot be completed in a single time as an RGA image task, and submit it to the driver and execute it one by one.

Overview

The software support library provides the following API, asynchronous mode only supports C++ implementation.

  • querystring: Query information about the RGA hardware version and supported functions of chip platform, return a string.
  • imcheckHeader: Verify the difference between the currently used header file version and the librga version.
  • importbuffer_T: Import external buffer into RGA driver to achieve hardware fast access to discontinuous physical addresses (dma_fd, virtual address).
  • releasebuffer_handle: Remove reference and map of the external buffer from inside the RGA driver.
  • wrapbuffer_handle Quickly encapsulate the image buffer structure (rga_buffer_t).
  • imbeginJob:Create an RGA image processing job.
  • imendJob: Submit and execute RGA image processing job.
  • imcancelJob: Cancel and delete the RGA image processing job.
  • imcopy: Call RGA for fast image copy.
  • imcopyTask: Added fast image copy operation to RGA image job.
  • imresize: Call RGA for fast image resize.
  • imresizeTask: Added fast image resize operation to RGA image job.
  • impyramind: Call RGA for fast image pyramid.
  • imcrop: Call RGA for fast image cropping.
  • imcropTask: Added fast image cropping operation to RGA image job.
  • imtranslate: Call RGA for fast image translation.
  • imtranslateTask: Added fast image translation operation to RGA image job.
  • imcvtcolor: Call RGA for fast image format conversion.
  • imcvtcolorTask: Added fast image format conversion operation to RGA image job.
  • imrotate: Call RGA for fast image rotation.
  • imrotateTask: Added fast image rotation operation to RGA image job.
  • imflip: Call RGA for fast image flipping.
  • imflipTask: Added fast image flipping operation to RGA image job.
  • imblend: Call RGA for double channel fast image blending.
  • imblendTask: Added double channel fast image blending operation to RGA image job.
  • imcomposite: Call RGA for three-channel fast image composite.
  • imcompositeTask: Added three-channel fast image composite operation to RGA image job.
  • imcolorkey: Call RGA for fast image color key.
  • imcolorkeyTask: Added fast image color key operation to RGA image job.
  • imosd: Call RGA for fast image OSD.
  • imosdTask:Added fast image OSD operation to RGA image job.
  • imquantize: Call RGA for fast image operation point preprocessing (quantization).
  • imquantizeTask: Added fast image operation point preprocessing (quantization) operation to RGA image job.
  • imrop: Call RGA for fast image ROP.
  • imropTask: Added fast image ROP operation to RGA image job.
  • imfill: Call RGA for fast image filling.
  • imfillArray: Call RGA to implement multiple groups of fast image filling.
  • imfillTask: Added fast image filling operation to RGA image job.
  • imfillTaskArray: Added multiple groups of fast image filling operation to RGA image job.
  • imrectangle: Call RGA for fast drawing operation of equidistant rectangle border.
  • imrectangleArray: Call RGA for multiple groups of fast drawing operation of equidistant rectangle border.
  • imrectangleTask: Added fast drawing operation of equidistant rectangle border operation to RGA image job.
  • imrectangleTaskArray: Added multiple groups of fast drawing operation of equidistant rectangle border operation to RGA image job.
  • immakeBorder: Call RGA for fast drawing operation of the border.
  • immosaic: call RGA for fast mosaic masking.
  • immosaicArray: call RGA for fast multiple groups of mosaic masking.
  • immosaicTask:Added fast mosaic masking operation to RGA image job.
  • immosaicTaskArray:Added fast multiple groups of mosaic masking operation to RGA image job.
  • improcess: Call RGA for fast image compound processing.
  • improcessTask: Added fast image compound processing operation to RGA image job.
  • imcheck: Verify whether the parameters are valid and whether the current hardware supports the operation.
  • imsync: Synchronize task completion status in asynchronous mode.
  • imconfig: Add default configuration to current thread context.

Obtain RGA Version and Support Information

querystring

const char* querystring(int name);

Query RGA basic information and resolution format.

Parameters Description
name RGA_VENDOR - vendor information
RGA_VERSION - RGA version
RGA_MAX_INPUT - max input resolution
RGA_MAX_OUTPUT - max output resolution
RGA_BYTE_STRIDE - stride alignment requirements
RGA_SCALE_LIMIT - scale limit
RGA_INPUT_FORMAT - input formats supported
RGA_OUTPUT_FORMAT - output formats supported
RGA_EXPECTED - expected performance
RGA_ALL - print all information

Returns a string describing properties of RGA.

header version check

imcheckHeader

IM_API IM_STATUS imcheckHeader(im_api_version_t header_version = RGA_CURRENT_API_HEADER_VERSION);

Verify the difference between the currently used header file version and the librga version.

Parameters Description
header_version Header file version, usually use the macro RGA_CURRENT_API_HEADER_VERSION.

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Buffer Preprocessing

importbuffer_T

For external memory that requires RGA processing, you can use importbuffer_T to map physical address of buffer to RGA driver and obtain the corresponding address of buffer, facilitating the subsequent stable and fast RGA call to complete the work.

Parameters(T) Data Type Description
virtual address void * image buffer virtual address
physical address uint64_t contiguous physical address of image buffer
fd int image buffer DMA file descriptor
GraphicBuffer handle buffer_handle_t image buffer handle, containing buffer address, file descriptor, resolution and format
GraphicBuffer GraphicBuffer android graphic buffer
AHardwareBuffer AHardwareBuffer chunks of memory that can be accessed by various hardware components in the system. https://developer.android.com/ndk/reference/group/a-hardware-buffer

Performance varies when different buffer types call RGA, and the performance order is shown below:

physical address > fd > virtual address

The recommended buffer type is fd.

IM_API rga_buffer_handle_t importbuffer_fd(int fd, int size);
IM_API rga_buffer_handle_t importbuffer_virtualaddr(void *va, int size);
IM_API rga_buffer_handle_t importbuffer_physicaladdr(uint64_t pa, int size);
Parameter Description
fd/va/pa [required] external buffer
size [required] memory size
IM_API rga_buffer_handle_t importbuffer_fd(int fd, int width, int height, int format);
IM_API rga_buffer_handle_t importbuffer_virtualaddr(void *va, int width, int height, int format);
IM_API rga_buffer_handle_t importbuffer_physicaladdr(uint64_t pa, int width, int height, int format);
Parameter Description
fd/va/pa [required] external buffer
width [required] pixel width stride of the image buffer
height [required] pixel height stride of the image buffer
format [required] pixel format of the image buffer
IM_API rga_buffer_handle_t importbuffer_fd(int fd, im_handle_param_t *param);
IM_API rga_buffer_handle_t importbuffer_virtualaddr(void *va, im_handle_param_t *param);
IM_API rga_buffer_handle_t importbuffer_physicaladdr(uint64_t pa, im_handle_param_t *param);
Parameter Description
fd/va/pa [required] external buffer
param [required] configure buffer parameters
IM_API rga_buffer_handle_t importbuffer_GraphicBuffer_handle(buffer_handle_t hnd);
IM_API rga_buffer_handle_t importbuffer_GraphicBuffer(sp<GraphicBuffer> buf);
IM_API rga_buffer_handle_t importbuffer_AHardwareBuffer(AHardwareBuffer *buf);
Parameter Description
hnd/buf [required] external buffer

Returns rga_buffer_handle_t to describe the memory handle.

releasebuffer_handle

After finishing calling RGA using external memory, you need to call releasebuffer_handle through memory handle to remove the mapping and binding between buffer and RGA driver, and release the resource inside RGA driver.

IM_API IM_STATUS releasebuffer_handle(rga_buffer_handle_t handle);

Return IM_STATUS_SUCCESS on success or else negative error code.

wrapbuffer_handle

In IM2D library API parameters, input image and output image should support multiple types, which mainly include memory, image format, image width and height. Before performing corresponding image operation, you need to call wrapbuffer_handle to convert the input and output image parameters into rga_buffer_t structure as the input parameter of user API.

rga_buffer_t wrapbuffer_handle(rga_buffer_handle_t handle,
                               int width,
                               int height,
                               int format,
                               int wstride = width,
                               int hstride = height);
Parameter Description
handle [required] RGA buffer handle
width [required] pixel width of the image that needs to be processed
height [required] pixel height of the image that needs to be processed
format [required] pixel format
wtride [optional] pixel width stride of the image
hstride [optional] pixel width stride of the image

Returns a rga_buffer_t to desribe image information.

Image processing job create

imbeginJob

IM_API im_job_handle_t imbeginJob(uint64_t flags = 0);

Create an RGA image processing job, which will return a job handle, job_handle can be used to add/remove RGA image operations, submit/execute the job.

Parameter Description
flags [optional] job flags

Returns im_job_handle_t to describe the job handle.

Image processing job submit

imendJob

IM_API IM_STATUS imendJob(im_job_handle_t job_handle,
                          int sync_mode = IM_SYNC,
                          int acquire_fence_fd = 0,
                          int *release_fence_fd = NULL);

Submit and execute RGA image processing job. When the job is complete, the currently completed RGA image processing job resource is deleted.

Parameter Description
job_handle [required] job handle
sync_mode [optional] wait until operation complete
acquire_fence_fd [optional] Used in async mode, run the job after waiting foracquire_fence signal
release_fence_fd [optional] Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

Image processing job cancel

imcancelJob

IM_API IM_STATUS imcancelJob(im_job_handle_t job_handle);

cancel and delete RGA image processing job.

Parameter Description
job_handle [required] job handle

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Copy

imcopy

IM_STATUS imcopy(const rga_buffer_t src,
                 rga_buffer_t dst,
                 int sync = 1),
                 int *release_fence_fd = NULL);

Copy the image, RGA basic operation. Its function is similar to memcpy.

Parameter Description
src [required] input image
dst [required] output image
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imcopyTask

IM_API IM_STATUS imcopyTask(im_job_handle_t job_handle,
                            const rga_buffer_t src,
                            rga_buffer_t dst);

Add an image copy operation to the specified job through job_handle. The configuration parameters are the same as imcopy.

Parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Resizing and Image Pyramid

imresize

IM_STATUS
imresize(const rga_buffer_t src,
         rga_buffer_t dst,
         double fx = 0,
         double fy = 0,
         int interpolation = INTER_LINEAR,
         int sync = 1,
         int *release_fence_fd = NULL);

According to different scenario, you can choose to configure dst to describe the output image size of resizing, or configure the scale factor fx/fy to resize at a specified ratio. When dst and fx/fy are configured at the same time, the calculated result of fx/fy is used as the output image size.

Only hardware version RGA1/RGA1 plus supports interpolation configuration.

Note: When resizing with fx/fy, format such as YUV that requires width and height alignment will force downward alignment to meet the requirements. Using this feature may affect the expected resizing effect.

Parameters Description
src [required] input image
dst [required] output image; it has the size dsize (when it is non-zero) or the size computed from src.size(), fx, and fy; the type of dst is the same as of src.
fx [optional] scale factor along the horizontal axis; when it equals 0, it is computed as:
fx = (double) dst.width / src.width
fy [optional] scale factor along the vertical axis; when it equals 0, it is computed as:
fy = (double) dst.height / src.height
interpolation [optional] interpolation method:
INTER_NEAREST - a nearest-neighbor interpolation
INTER_LINEAR - a bilinear interpolation (used by default)
INTER_CUBIC - a bicubic interpolation over 4x4 pixel neighborhood
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

impyramid

IM_STATUS impyramid (const rga_buffer_t src,
                     rga_buffer_t dst,
                     IM_SCALE direction)

Pyramid scaling. Scale by 1/2 or twice, depending on the direction width and height.

Parameters Description
src [required] input image
dst [required] output image;
direction [required] scale mode
IM_UP_SCALE —— up scale
IM_DOWN_SCALE —— down scale
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imresizeTask

IM_API IM_STATUS imresizeTask(im_job_handle_t job_handle,
                              const rga_buffer_t src,
                              rga_buffer_t dst,
                              double fx = 0,
                              double fy = 0,
                              int interpolation = 0);

Add an image resize operation to the specified job through job_handle. The configuration parameters are the same as imresize.

Parameters Description
job_handle [required] job handle
src [required] input image
dst [required] output image; it has the size dsize (when it is non-zero) or the size computed from src.size(), fx, and fy; the type of dst is the same as of src.
fx [optional] scale factor along the horizontal axis; when it equals 0, it is computed as:
fx = (double) dst.width / src.width
fy [optional] scale factor along the vertical axis; when it equals 0, it is computed as:
fy = (double) dst.height / src.height
interpolation [optional] interpolation method:
INTER_NEAREST - a nearest-neighbor interpolation
INTER_LINEAR - a bilinear interpolation (used by default)
INTER_CUBIC - a bicubic interpolation over 4x4 pixel neighborhood

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Cropping

imcrop

IM_STATUS imcrop(const rga_buffer_t src,
                 rga_buffer_t dst,
                 im_rect rect,
                 int sync = 1,
                 int *release_fence_fd = NULL);

Perform image clipping by specifying the size of the region.

Parameter Description
src **[required] **input image
dst [required] output image
rect [required] crop region
x - upper-left x coordinate
y - upper-left y coordinate
width - region width
height - region height
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imcropTask

IM_API IM_STATUS imcropTask(im_job_handle_t job_handle,
                            const rga_buffer_t src,
                            rga_buffer_t dst,
                            im_rect rect);

Add an image crop operation to the specified job through job_handle. The configuration parameters are the same as imcrop.

Parameter Description
job_handle [required] job handle
src **[required] **input image
dst [required] output image
rect [required] crop region
x - upper-left x coordinate
y - upper-left y coordinate
width - region width
height - region height

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Translation

imtranslate

IM_STATUS imtranslate(const rga_buffer_t src,
                      rga_buffer_t dst,
                      int x,
                      int y,
                      int sync = 1,
                      int *release_fence_fd = NULL);

Image translation. Move to (x, y) position, the width and height of src and dst must be the same, the excess part will be clipped.

Parameter Description
src **[required]**input image
dst [required] output image
x [optional] horizontal translation
y [optional] vertical translation
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imtranslateTask

IM_API IM_STATUS imtranslateTask(im_job_handle_t job_handle,
                                 const rga_buffer_t src,
                                 rga_buffer_t dst,
                                 int x,
                                 int y);

Add an image translation operation to the specified job through job_handle. The configuration parameters are the same as imtranslate.

Parameter Description
job_handle [required] job handle
src **[required]**input image
dst [required] output image
x [required] horizontal translation
y [required] vertical translation

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Format Conversion

imcvtcolor

IM_STATUS imcvtcolor(rga_buffer_t src,
                     rga_buffer_t dst,
                     int sfmt,
                     int dfmt,
                     int mode = IM_COLOR_SPACE_DEFAULT,
                     int sync = 1,
                     int *release_fence_fd = NULL);

Image format conversion,specific format support varies according to soc, please refer to the Image Format Supported section.

The format can be set by rga_buffer_t, or configure the input image and output image formats respectively by sfmt/dfmt. When it comes to YUV/RGB color gamut conversion, you can configure the converted color gamut through mode, and the conversion is performed according to the BT.601 limit range by default.

parameter Description
src [required] input image
dst [required] output image
sfmt [optional] source image format
dfmt [optional] destination image format
Mode [optional] color space mode:
IM_YUV_TO_RGB_BT601_LIMIT
IM_YUV_TO_RGB_BT601_FULL
IM_YUV_TO_RGB_BT709_LIMIT
IM_RGB_TO_YUV_BT601_LIMIT
IM_RGB_TO_YUV_BT601_FULL
IM_RGB_TO_YUV_BT709_LIMIT
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imcvtcolorTask

IM_API IM_STATUS imcvtcolorTask(im_job_handle_t job_handle,
                                rga_buffer_t src,
                                rga_buffer_t dst,
                                int sfmt,
                                int dfmt,
                                int mode = IM_COLOR_SPACE_DEFAULT);

Add an image format conversion operation to the specified job through job_handle. The configuration parameters are the same as imcvtcolor.

parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image
sfmt [optional] source image format
dfmt [optional] destination image format
Mode [optional] color space mode:
IM_YUV_TO_RGB_BT601_LIMIT
IM_YUV_TO_RGB_BT601_FULL
IM_YUV_TO_RGB_BT709_LIMIT
IM_RGB_TO_YUV_BT601_LIMIT
IM_RGB_TO_YUV_BT601_FULL
IM_RGB_TO_YUV_BT709_LIMIT

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Rotation

imrotate

IM_STATUS imrotate(const rga_buffer_t src,
                   rga_buffer_t dst,
                   int rotation,
                   int sync = 1,
                   int *release_fence_fd = NULL);

Support image rotation 90,180,270 degrees.

Parameter Description
src [required] input image
dst [required] output image
rotation [required] rotation angle:
0
IM_HAL_TRANSFORM_ROT_90
IM_HAL_TRANSFORM_ROT_180
IM_HAL_TRANSFORM_ROT_270
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imrotateTask

IM_API IM_STATUS imrotateTask(im_job_handle_t job_handle,
                              const rga_buffer_t src,
                              rga_buffer_t dst,
                              int rotation);

Add an image rotation operation to the specified job through job_handle. The configuration parameters are the same as imrotate.

Parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image
rotation [required] rotation angle:
IM_HAL_TRANSFORM_ROT_90
IM_HAL_TRANSFORM_ROT_180
IM_HAL_TRANSFORM_ROT_270

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Mirror Flip

imfilp

IM_STATUS imflip (const rga_buffer_t src,
                  rga_buffer_t dst,
                  int mode,
                  int sync = 1,
                  int *release_fence_fd = NULL);

Support image to do horizontal, vertical mirror flip.

Parameter Description
src [required] input image
dst [required] output image
mode [optional] flip mode:
0
IM_HAL_TRANSFORM_FLIP_H
IM_HAL_TRANSFORM_FLIP_V
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imflipTask

IM_API IM_STATUS imflipTask(im_job_handle_t job_handle,
                            const rga_buffer_t src,
                            rga_buffer_t dst,
                            int mode);

Add an image flip operation to the specified job through job_handle. The configuration parameters are the same as imflip.

Parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image
mode [required] flip mode:
IM_HAL_TRANSFORM_FLIP_H_V
IM_HAL_TRANSFORM_FLIP_H
IM_HAL_TRANSFORM_FLIP_V

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Blending

imblend/imcomposite

IM_STATUS imblend(const rga_buffer_t srcA,
                  rga_buffer_t dst,
                  int mode = IM_ALPHA_BLEND_SRC_OVER,
                  int sync = 1,
                  int *release_fence_fd = NULL);

RGA uses A+B -> B image two-channel blending mode to perform Alpha superposition for foreground image (srcA channel) and background image (dst channel) according to the configured blending model, and output the blending result to dst channel. When no mode is configured, it is set to src-over mode by default.

IM_STATUS imcomposite(const rga_buffer_t srcA,
                      const rga_buffer_t srcB,
                      rga_buffer_t dst,
                      int mode = IM_ALPHA_BLEND_SRC_OVER,
                      int sync = 1,
                      int *release_fence_fd = NULL);

RGA uses A+B -> C image three-channel blending mode to perform Alpha superposition for foreground image (srcA channel) and background image (srcB channel) according to the configured blending model, and output the blending result to dst channel.

mode in the two image blending modes can be configured with different Porter-Duff blending model

Before describing the Porter-Duff blending model, give the following definitions:

  • S -Marks the source image in two blended images,namely the foreground image, short for source.
  • D -Marks the destination image in two blended images,namely the background image, short for destination.
  • R -Marks the result of two images blending,short for result.
  • c -Marks the color of the pixel,the RGB part of RGBA, which describes the color of the image itself, short for color.(Note,Color values (RGB) in the Porter-Duff blending model are all left-multiplied results, that is, the product of original color and transparency. If the color values are not left-multiplied, pre-multiplied operations(Xc = Xc * Xa) are required.)
  • a -Marks the Alpha of the pixel,Namely the A part of RGBA, describe the transparency of the image itself, short for Alpha.
  • f -Marks factors acting on C or A,short for factor.

The core formula of Porter-Duff blending model is as follows:

Rc = Sc * Sf + Dc * Df;

that is: result color = source color * source factor + destination color * destination factor.

Ra = Sa * Sf + Da * Df;

that is: result Alpha = source Alpha * source factor + destination Alpha * destination factor.

RGA supports following blending models:

SRC:

​ Sf = 1, Df = 0;

​ [Rc,Ra] = [Sc,Sa];

DST:

​ Sf = 0, Df = 1;

​ [Rc,Ra] = [Dc,Da];

SRC_OVER:

​ Sf = 1, Df = (1 - Sa);

​ [Rc,Ra] = [ Sc + (1 - Sa) * Dc, Sa + (1 - Sa) * Da ];

DST_OVER:

​ Sf = (1 - Da) , Df = 1;

​ [Rc,Ra] = [ Sc * (1 - Da) + Dc, Sa * (1 - Da) + Da ] ;

【Note】Image blending model does not support the YUV format image blending, the YUV format is not support in dst image of imblend , the YUV format is not support in srcB image of imcomposite.

Parameter Description
srcA [required] input image A
srcB [required] input image B
dst [required] output image
mode [optional] blending mode:
IM_ALPHA_BLEND_SRC —— SRC mode
IM_ALPHA_BLEND_DST —— DST mode
IM_ALPHA_BLEND_SRC_OVER —— SRC OVER mode
IM_ALPHA_BLEND_DST_OVER —— DST OVER mode
IM_ALPHA_BLEND_PRE_MUL —— Enable premultipling. When premultipling is required, this identifier must be processed with other mode identifiers, then assign to mode
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imblendTask/imcompositeTask

IM_API IM_STATUS imblendTask(im_job_handle_t job_handle,
                             const rga_buffer_t fg_image,
                             rga_buffer_t bg_image,
                             int mode = IM_ALPHA_BLEND_SRC_OVER);

Add an A+B -> B image two-channel blending operation to the specified job through job_handle. The configuration parameters are the same as imblend. When no mode is configured, it is set to src-over mode by default.

IM_API IM_STATUS imcompositeTask(im_job_handle_t job_handle,
                                 const rga_buffer_t fg_image,
                                 const rga_buffer_t bg_image,
                                 rga_buffer_t output_image,
                                 int mode = IM_ALPHA_BLEND_SRC_OVER);

Add an A+B -> C image three-channel blending operation to the specified job through job_handle. The configuration parameters are the same as imcomposite.

【Note】Image blending model does not support the YUV format image blending, the YUV format is not support in dst image of imblend , the YUV format is not support in srcB image of imcomposite.

Parameter Description
job_handle [required] job handle
fg_image [required] foreground image
bg_image [required] background image, when A+B->B it is also the output destination image.
output_image [required] output destination image.
mode [optional] blending mode:
IM_ALPHA_BLEND_SRC —— SRC mode
IM_ALPHA_BLEND_DST —— DST mode
IM_ALPHA_BLEND_SRC_OVER —— SRC OVER mode
IM_ALPHA_BLEND_DST_OVER —— DST OVER mode
IM_ALPHA_BLEND_PRE_MUL —— Enable premultipling. When premultipling is required, this identifier must be processed with other mode identifiers, then assign to mode

Return IM_STATUS_SUCCESS on success or else negative error code.

Color Key

imcolorkey

IM_STATUS imcolorkey(const rga_buffer_t src,
                     rga_buffer_t dst,
                     im_colorkey_range range,
                     int mode = IM_ALPHA_COLORKEY_NORMAL,
                     int sync = 1,
                     int *release_fence_fd = NULL);

Color Key is to preprocesses the source image, zeros the alpha component of pixels that meet the Color Key filtering conditions, wherein the Color Key filtering conditions are non-transparent color values, and performs the alpha blending mode between the preprocessed source image and the destination image.

This mode only supports the Color Key operation on the source image (src) region of the image for the set Color range, and overlays on the destination image (dst) region.

IM_ALPHA_COLORKEY_NORMAL is the normal mode, that is, the colors in the set color range are used as the filtering condition. The Alpha components of pixels in this color range are set zero; IM_ALPHA_COLORKEY_INVERTED is inverted. When no mode is configured, it is set to IM_ALPHA_COLORKEY_NORMAL mode by default.

Parameters Range Description
max 0x0 ~ 0xFFFFFFFF The max color range to cancel/scratch, arranged as ABGR
min 0x0 ~ 0xFFFFFFFF The min color range to cancel/scratch, arranged as ABGR
parameter Description
src [required] input image
dst [required] output image
range [required] Target color range
typedef struct im_colorkey_range {
int max;
int min;
} im_colorkey_value;
Mode [required] Color Key mode:
IM_ALPHA_COLORKEY_NORMAL
IM_ALPHA_COLORKEY_INVERTED
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imcolorkeyTask

IM_API IM_STATUS imcolorkeyTask(im_job_handle_t job_handle,
                                const rga_buffer_t fg_image,
                                rga_buffer_t bg_image,
                                im_colorkey_range range,
                                int mode = IM_ALPHA_COLORKEY_NORMAL);

Add an image Color Key operation to the specified job through job_handle. The configuration parameters are the same as imcolorkey.

Parameters Range Description
max 0x0 ~ 0xFFFFFFFF The max color range to cancel/scratch, arranged as ABGR
min 0x0 ~ 0xFFFFFFFF The min color range to cancel/scratch, arranged as ABGR
parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image
range [required] Target color range
typedef struct im_colorkey_range {
int max;
int min;
} im_colorkey_value;
Mode [required] Color Key mode:
IM_ALPHA_COLORKEY_NORMAL
IM_ALPHA_COLORKEY_INVERTED

Return IM_STATUS_SUCCESS on success or else negative error code.

OSD

imosd

IM_API IM_STATUS imosd(const rga_buffer_t osd,
                       const rga_buffer_t bg_image,
                       const im_rect osd_rect,
                       im_osd_t *osd_config,
                       int sync = 1,
                       int *release_fence_fd = NULL);

OSD (On-Screen-Display), can superimpose text information on video pictures, and perform brightness statistics and automatic color inversion functions for fonts.

parameter Description
OSD [required] osd block image
bg_image [required] output image
osd_rect [required] image region to OSD
osd_config [required] OSD function config
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imosdTask

IM_API IM_STATUS imosdTask(im_job_handle_t job_handle,
                           const rga_buffer_t osd,
                           const rga_buffer_t bg_image,
                           const im_rect osd_rect,
                           im_osd_t *osd_config);

Add an OSD operation to the specified job through job_handle. The configuration parameters are the same as imosd.

parameter Description
job_handle [required] job handle
OSD [required] osd block image
dst [required] output image
osd_rect [required] image region to OSD
osd_config [required] OSD function config

Return IM_STATUS_SUCCESS on success or else negative error code.

Pre-processing of NN Operation Points (Quantization)

imquantize

IM_STATUS imquantize(const rga_buffer_t src,
                     rga_buffer_t dst,
                     rga_nn_t nn_info,
                     int sync = 1,
                     int *release_fence_fd = NULL);

Currently supported only on RV1126 / RV1109. NN operation point pre-processing, three channels of RGB of image can be separately configured with offset and scale.

Formula:

dst = 【(src + offset) * scale 】

Parameters range:

Parameters Range Description
scale 0 ~ 3.99 10bit,From left to right, the highest 2 bits indicate the integer part and the lowest 8 bits indicate the decimal part
offset -255 ~ 255 9bit,From left to right, the high indicates the sign bit, and the low indicates the offset from 0 to 255
Parameter Description
src [required] input image
dst [required] output image
nn_info [required] rga_nn_t struct configures the offset and scale of the three RGB channels respectively
typedef struct rga_nn {
int nn_flag;
int scale_r;
int scale_g;
int scale_b;
int offset_r;
int offset_g;
int offset_b;
} rga_nn_t;
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imquantizeTask

IM_API IM_STATUS imquantizeTask(im_job_handle_t job_handle,
                                const rga_buffer_t src,
                                rga_buffer_t dst,
                                im_nn_t nn_info);

Add an image quantization conversion operation to the specified job through job_handle. The configuration parameters are the same as imquantize.

parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image
nn_info [required] rga_nn_t结构体对RGB三个通道offset及scale进行单独配置
typedef struct rga_nn {
int nn_flag;
int scale_r;
int scale_g;
int scale_b;
int offset_r;
int offset_g;
int offset_b;
} rga_nn_t;

Return IM_STATUS_SUCCESS on success or else negative error code.

Image ROP

imrop

IM_STATUS imrop(const rga_buffer_t src,
                rga_buffer_t dst,
                int rop_code,
                int sync = 1,
                int *release_fence_fd = NULL);

Perform ROP, and, or, not operations on two images

Parameter Description
src [required] input image
dst [required] output image
rop_code [required] rop code mode

IM_ROP_AND : dst = dst AND src;
IM_ROP_OR : dst = dst OR src
IM_ROP_NOT_DST : dst = NOT dst
IM_ROP_NOT_SRC : dst = NOT src
IM_ROP_XOR : dst = dst XOR src
IM_ROP_NOT_XOR : dst = NOT (dst XOR src)
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code

imropTask

IM_API IM_STATUS imropTask(im_job_handle_t job_handle,
                           const rga_buffer_t src,
                           rga_buffer_t dst,
                           int rop_code);

Add an image ROP conversion operation to the specified job through job_handle. The configuration parameters are the same as imrop.

parameter Description
job_handle [required] job handle
src [required] input image
dst [required] output image
rop_code [required] rop code mode

IM_ROP_AND : dst = dst AND src;
IM_ROP_OR : dst = dst OR src
IM_ROP_NOT_DST : dst = NOT dst
IM_ROP_NOT_SRC : dst = NOT src
IM_ROP_XOR : dst = dst XOR src
IM_ROP_NOT_XOR : dst = NOT (dst XOR src)

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Color Filling

imfill

IM_STATUS imfill(rga_buffer_t buf,
                 im_rect rect,
                 int color = 0x00000000,
                 int sync = 1);

Color fills the specified area rect of the image.

Color parameters from high to low are respectively R,G,B,A. For example, red: color = 0xff000000.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
src [required] input image
dst [required] output image
rect [required] image region to fill specified color
width and height of rect must be greater than or equal to 2
color [required] fill with color, default=0x00000000
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imfillArray

IM_API IM_STATUS imfillArray(rga_buffer_t dst,
                             im_rect *rect_array,
                             int array_size,
                             uint32_t color,
                             int sync = 1,
                             int *release_fence_fd = NULL);

Color fills multiple areas of the image one by one.

Color parameters from high to low are respectively R,G,B,A. For example, red: color = 0xff000000.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
dst [required] target image
rect_array [required] image region array_ptr to fill specified color
width and height of rect must be greater than or equal to 2
array_size [required] size of region arrays.
color [required] fill with color
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imfillTask

IM_API IM_STATUS imfillTask(im_job_handle_t job_handle,
                            rga_buffer_t dst,
                            im_rect rect,
                            uint32_t color);

Add an image color fill operation to the specified job through job_handle. The configuration parameters are the same as imfill.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
job_handle [required] job handle
dst [required] target image
rect [required] image region to fill specified color
width and height of rect must be greater than or equal to 2
color [required] fill with color

Return IM_STATUS_SUCCESS on success or else negative error code.

imfillTaskArray

IM_API IM_STATUS imfillTaskArray(im_job_handle_t job_handle,
                                 rga_buffer_t dst,
                                 im_rect *rect_array,
                                 int array_size,
                                 uint32_t color);

Add an image color fill multiple areas operation to the specified job through job_handle. The configuration parameters are the same as imfillArray.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
job_handle [required] job handle
dst [required] target image
rect_array [required] image region array_ptr to fill specified color
width and height of rect must be greater than or equal to 2
array_size [required] size of region arrays.
color [required] fill with color

Return IM_STATUS_SUCCESS on success or else negative error code.

imrectangle

IM_API IM_STATUS imrectangle(rga_buffer_t dst,
                             im_rect rect,
                             uint32_t color,
                             int thickness,
                             int sync = 1,
                             int *release_fence_fd = NULL);

Draw a border with a thickness of "thickness" to the specified area rect of the image (described as the outer diameter of the border) according to the specified color by "color", and fill a solid rectangle when the thickness is negative.

Color parameters from high to low are respectively R,G,B,A. For example, red: color = 0xff000000.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
dst [required] target image
rect [required] image region to fill specified color
width and height of rect must be greater than or equal to 2
color [required] fill with color
thickness [required] Thickness of lines that make up the rectangle.
Negative values, like -1, mean that the function has to draw a filled rectangle.
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imrectangleArray

IM_API IM_STATUS imrectangleArray(rga_buffer_t dst,
                                  im_rect *rect_array,
                                  int array_size,
                                  uint32_t color,
                                  int thickness,
                                  int sync = 1,
                                  int *release_fence_fd = NULL);

Draw multiple border with a thickness of "thickness" to the specified area rect of the image (described as the outer diameter of the border) according to the specified color by "color", and fill a solid rectangle when the thickness is negative.

Color parameters from high to low are respectively R,G,B,A. For example, red: color = 0xff000000.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
dst [required] target image
rect_array [required] image region array_ptr to fill specified color
width and height of rect must be greater than or equal to 2
array_size [required] size of region arrays.
color [required] fill with color
thickness [required] Thickness of lines that make up the rectangle.
Negative values, like -1, mean that the function has to draw a filled rectangle.
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

imrectangleTask

IM_API IM_STATUS imrectangleTask(im_job_handle_t job_handle,
                                 rga_buffer_t dst,
                                 im_rect rect,
                                 uint32_t color,
                                 int thickness);

Add an Draw border operation to the specified job through job_handle. The configuration parameters are the same as imrectangle.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
job_handle [required] job handle
dst [required] target image
rect [required] image region to fill specified color
width and height of rect must be greater than or equal to 2
color [required] fill with color
thickness [required] Thickness of lines that make up the rectangle.
Negative values, like -1, mean that the function has to draw a filled rectangle.

Return IM_STATUS_SUCCESS on success or else negative error code.

imrectangleTaskArray

IM_API IM_STATUS imrectangleTaskArray(im_job_handle_t job_handle,
                                      rga_buffer_t dst,
                                      im_rect *rect_array,
                                      int array_size,
                                      uint32_t color,
                                      int thickness);

Add an Draw multiple border operation to the specified job through job_handle. The configuration parameters are the same as imrectangleArray.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
job_handle [required] job handle
dst [required] target image
rect_array [required] image region array_ptr to fill specified color
width and height of rect must be greater than or equal to 2
array_size [required] size of region arrays.
color [required] fill with color
thickness [required] Thickness of lines that make up the rectangle.
Negative values, like -1, mean that the function has to draw a filled rectangle.

Return IM_STATUS_SUCCESS on success or else negative error code.

immakeBorder

IM_API IM_STATUS immakeBorder(rga_buffer_t src,
                              rga_buffer_t dst,
                              int top,
                              int bottom,
                              int left,
                              int right,
                              int border_type,
                              int value = 0,
                              int sync = 1,
                              int acquir_fence_fd = -1,
                              int *release_fence_fd = NULL);

According to the configured top/bottom/left/right pixels, draw a border to the input image and output it to the output target image buffer.

【 Note 】 The width and height of the rect must be greater than or equal to 2

Parameter Description
src [required] input source image
dst [required] output target image
top [required] number of top pixels
bottom [required] number of bottom pixels
left [required] number of left pixels
right **[required] **number of right pixels
border_type [required] Border type
IM_BORDER_CONSTANT // iiiiii abcdefgh iiiiiii with some specified value 'i'
IM_BORDER_REFLECT //fedcba abcdefgh hgfedcb
IM_BORDER_WRAP //cdefgh abcdefgh abcdefg
value [optional] the pixel value at which the border is filled
sync [optional] wait until operation complete
acquire_fence_fd [required] used in async mode, run the job after waiting foracquire_fence signal
release_fence_fd [required] used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Mosaic

immosaic

IM_API IM_STATUS immosaic(const rga_buffer_t image,
                          im_rect rect,
                          int mosaic_mode,
                          int sync = 1,
                          int *release_fence_fd = NULL);

Mosaic masking the specified area of the image.

Parameter Description
image [required] souce image
rect [required] image region to mosaic
mosaic_mode [required] set mosaic mode
IM_MOSAIC_8
IM_MOSAIC_16
IM_MOSAIC_32
IM_MOSAIC_64
IM_MOSAIC_128
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

immosaicArray

IM_API IM_STATUS immosaicArray(const rga_buffer_t image,
                               im_rect *rect_array,
                               int array_size,
                               int mosaic_mode,
                               int sync = 1,
                               int *release_fence_fd = NULL);

Mosaic masking the specified multiple area of the image.

Parameter Description
image [required] target image
rect_array [required] image region array_ptr to mosaic
array_size [required] size of region arrays.
mosaic_mode [required] set mosaic mode
IM_MOSAIC_8
IM_MOSAIC_16
IM_MOSAIC_32
IM_MOSAIC_64
IM_MOSAIC_128
sync [optional] wait until operation complete
release_fence_fd **[optional]**Used in async mode, as a parameter of imsync()

Return IM_STATUS_SUCCESS on success or else negative error code.

immosaicTask

IM_API IM_STATUS immosaicTask(im_job_handle_t job_handle,
                              const rga_buffer_t image,
                              im_rect rect,
                              int mosaic_mode);

Add an image mosaic masking operation to the specified job through job_handle. The configuration parameters are the same as immosaic.

Parameter Description
job_handle [required] job handle
image [required] target image
rect [required] image region to mosaic
mosaic_mode [required] set mosaic mode
IM_MOSAIC_8
IM_MOSAIC_16
IM_MOSAIC_32
IM_MOSAIC_64
IM_MOSAIC_128

Return IM_STATUS_SUCCESS on success or else negative error code.

immosaicTaskArray

IM_API IM_STATUS immosaicTaskArray(im_job_handle_t job_handle,
                                   const rga_buffer_t image,
                                   im_rect *rect_array,
                                   int array_size,
                                   int mosaic_mode);

Add multiple image mosaic masking operation to the specified job through job_handle. The configuration parameters are the same as immosaicArray.

Parameter Description
job_handle [required] job handle
image [required] target image
rect_array [required] image region array_ptr to mosaic
array_size [required] size of region arrays.
mosaic_mode [required] set mosaic mode
IM_MOSAIC_8
IM_MOSAIC_16
IM_MOSAIC_32
IM_MOSAIC_64
IM_MOSAIC_128

Return IM_STATUS_SUCCESS on success or else negative error code.

Image Process

improcess

IM_STATUS improcess(rga_buffer_t src,
                    rga_buffer_t dst,
                    rga_buffer_t pat,
                    im_rect srect,
                    im_rect drect,
                    im_rect prect,
                    int acquire_fence_fd,
                    int *release_fence_fd,
                    im_opt_t *opt,
                    int usage);

RGA image compound operation. Other APIs are developed based on this API, improcess can achieve more complex compound operations.

Image processes are configured by usage.

usage definitions:

typedef enum {
    /* Rotation */
    IM_HAL_TRANSFORM_ROT_90     = 1 << 0,
    IM_HAL_TRANSFORM_ROT_180    = 1 << 1,
    IM_HAL_TRANSFORM_ROT_270    = 1 << 2,
    IM_HAL_TRANSFORM_FLIP_H     = 1 << 3,
    IM_HAL_TRANSFORM_FLIP_V     = 1 << 4,
    IM_HAL_TRANSFORM_FLIP_H_V   = 1 << 5,
    IM_HAL_TRANSFORM_MASK       = 0x3f,

    /*
     * Blend
     * Additional blend usage, can be used with both source and target configs.
     * If none of the below is set, the default "SRC over DST" is applied.
     */
    IM_ALPHA_BLEND_SRC_OVER     = 1 << 6,     /* Default, Porter-Duff "SRC over DST" */
    IM_ALPHA_BLEND_SRC          = 1 << 7,     /* Porter-Duff "SRC" */
    IM_ALPHA_BLEND_DST          = 1 << 8,     /* Porter-Duff "DST" */
    IM_ALPHA_BLEND_SRC_IN       = 1 << 9,     /* Porter-Duff "SRC in DST" */
    IM_ALPHA_BLEND_DST_IN       = 1 << 10,    /* Porter-Duff "DST in SRC" */
    IM_ALPHA_BLEND_SRC_OUT      = 1 << 11,    /* Porter-Duff "SRC out DST" */
    IM_ALPHA_BLEND_DST_OUT      = 1 << 12,    /* Porter-Duff "DST out SRC" */
    IM_ALPHA_BLEND_DST_OVER     = 1 << 13,    /* Porter-Duff "DST over SRC" */
    IM_ALPHA_BLEND_SRC_ATOP     = 1 << 14,    /* Porter-Duff "SRC ATOP" */
    IM_ALPHA_BLEND_DST_ATOP     = 1 << 15,    /* Porter-Duff "DST ATOP" */
    IM_ALPHA_BLEND_XOR          = 1 << 16,    /* Xor */
    IM_ALPHA_BLEND_MASK         = 0x1ffc0,

    IM_ALPHA_COLORKEY_NORMAL    = 1 << 17,
    IM_ALPHA_COLORKEY_INVERTED  = 1 << 18,
    IM_ALPHA_COLORKEY_MASK      = 0x60000,

    IM_SYNC                     = 1 << 19,
    IM_ASYNC                    = 1 << 26,
    IM_CROP                     = 1 << 20,    /* Unused */
    IM_COLOR_FILL               = 1 << 21,
    IM_COLOR_PALETTE            = 1 << 22,
    IM_NN_QUANTIZE              = 1 << 23,
    IM_ROP                      = 1 << 24,
    IM_ALPHA_BLEND_PRE_MUL      = 1 << 25,
} IM_USAGE;
Parameter Description
src [required] input imageA
dst [required] output image
pat [required] input imageB
srect [required] src crop region
drect [required] dst crop region
prect [required] pat crop region
acquire_fence_fd [required] Used in async mode, run the job after waiting foracquire_fence signal
release_fence_fd [required] Used in async mode, as a parameter of imsync()
opt [required] operation options
typedef struct im_opt {
int color;
im_colorkey_range colorkey_range;
im_nn_t nn;
int rop_code;

int priority;
int core;
} im_opt_t;
usage [required] image operation usage

Return IM_STATUS_SUCCESS on success or else negative error code.

improcessTask

IM_API IM_STATUS improcessTask(im_job_handle_t job_handle,
                               rga_buffer_t src,
                               rga_buffer_t dst,
                               rga_buffer_t pat,
                               im_rect srect,
                               im_rect drect,
                               im_rect prect,
                               im_opt_t *opt_ptr,
                               int usage);

Add an image compound operation to the specified job through job_handle. The configuration parameters are the same as improcess.

Parameter Description
job_handle [required] job handle
src [required] input imageA
dst [required] output image
pat [required] input imageB
srect [required] src crop region
drect [required] dst crop region
prect [required] pat crop region
acquire_fence_fd [required] Used in async mode, run the job after waiting foracquire_fence signal
release_fence_fd [required] Used in async mode, as a parameter of imsync()
opt [required] operation options
typedef struct im_opt {
int color;
im_colorkey_range colorkey_range;
im_nn_t nn;
int rop_code;

int priority;
int core;
} im_opt_t;
usage [required] image operation usage

Return IM_STATUS_SUCCESS on success or else negative error code.

Parameter Check

imcheck

IM_API IM_STATUS imcheck(const rga_buffer_t src, const rga_buffer_t dst,
                         const im_rect src_rect, const im_rect dst_rect,
                         const int mode_usage);
IM_API IM_STATUS imcheck_composite(const rga_buffer_t src, const rga_buffer_t dst, const rga_buffer_t pat,
                                   const im_rect src_rect, const im_rect dst_rect, const im_rect pat_rect,
                                   const int mode_usage);

After RGA parameters are configured, users can use this API to verify whether the current parameters are valid and determine whether the hardware supports them based on the current hardware conditions.

Users are advised to use this API only during development and debugging to avoid performance loss caused by multiple verification.

Parameter Description
src [required] input imageA
dst [required] output image
pat [optional] input imageB
srect [required] src crop region
drect [required] dst crop region
prect [optional] pat crop region
usage [optional] image operation usage

Return IM_STATUS_NOERROR on success or else negative error code.

Synchronous operation

imsync

IM_STATUS imsync(int fence_fd);

RGA asynchronous mode requires this API to be called, passing the returned release_fence_fd as parameter.

Other API enable asynchronous call mode when sync is set to 0, which is equivalent to glFlush in opengl. Further calls to imsync is equivalent to glFinish.

Parameter Description
fence_fd [required] fence_fd to wait

Return IM_STATUS_SUCCESS on success or else negative error code.

Thread Context Configuration

imconfig

IM_STATUS  imconfig(IM_CONFIG_NAME name, uint64_t value);

The context for the current thread is configured through different configuration name, which will be the default configuration for the thread.

The thread context configuration has a lower priority than the parameter configuration of the API. If no related parameters are configured for API, the local call uses the default context configuration. If related parameters are configured for API, the call uses the API parameter configuration.

Parameter Description
name [required] context config name:
IM_CONFIG_SCHEDULER_CORE —— Specify the task processing core
IM_CONFIG_PRIORITY —— Specify the task priority
IM_CHECK_CONFIG —— Check enable
value [required] config value
IM_CONFIG_SCHEDULER_CORE :
IM_SCHEDULER_RGA3_CORE0
IM_SCHEDULER_RGA3_CORE1
IM_SCHEDULER_RGA2_CORE0
IM_SCHEDULER_RGA3_DEFAULT
IM_SCHEDULER_RGA2_DEFAULT
IM_CONFIG_PRIORITY:
0 ~ 6
IM_CHECK_CONFIG:
TRUE
FALSE

Note:Permissions of priority and core are very high. Improper operations may cause system crash or deadlock. Therefore, users are advised to configure them only during development and debugging. Users are not advised to perform this configuration in actual product

Return IM_STATUS_SUCCESS on success or else negative error code


Data Structure

This section describes the data structures involved in API in detail.

Overview

Data structure Description
rga_buffer_t image buffer information
im_rect the actual operating area of the image
im_opt_t image manipulation options
im_job_handle_t RGA job handle
rga_buffer_handle_t RGA driver image buffer handle
im_handle_param_t buffer parameters of image to be imported
im_context_t default context for the current thread
im_nn_t operation point preprocessing parameters
im_colorkey_range Colorkey range

Detailed Descriptions

rga_buffer_t

  • descriptions

Buffer information of image with single channel.

  • path

im2d_api/im2d_type.h

  • definitions
typedef struct {
    void* vir_addr;                     /* virtual address */
    void* phy_addr;                     /* physical address */
    int fd;                             /* shared fd */
    rga_buffer_handle_t handle;         /* buffer handle */

    int width;                          /* width */
    int height;                         /* height */
    int wstride;                        /* wstride */
    int hstride;                        /* hstride */
    int format;                         /* format */

    int color_space_mode;               /* color_space_mode */
    int global_alpha;                   /* global_alpha */
    int rd_mode;
} rga_buffer_t;
Parameter Description
vir_addr Virtual address of image buffer.
phy_addr Contiguous physical address of the image buffer.
fd File descriptor of image buffer DMA.
handle Import handle corresponding to the image buffer of the RGA driver.
width The width of the actual operating area of image,in pixels.
height The height of the actual operating area of image,in pixels.
wstride The stride of the image width, in pixels.
hstride The stride of the image height, in pixels.
format Image format.
color_space_mode Image color space mode.
global_alpha Global Alpha configuration.
rd_mode The mode in which the current channel reads data.
  • Note

Simply selects either one of vir_addr、phy_addr、fd、handle as the description of image buffer, if multiple values are assigned, only one of them is selected as the image buffer description according to the default priority, which is as follows:handle > phy_addr > fd > vir_addr.

im_rect

  • descriptions

Describes the actual operation area of image with single channel.

  • path

im2d_api/im2d_type.h

  • definitions
typedef struct {
    int x;        /* upper-left x */
    int y;        /* upper-left y */
    int width;    /* width */
    int height;   /* height */
} im_rect;
Parameters Description
x The starting abscissa of the actual operation area of the image, in pixels.
y The starting ordinate of the actual operating area of an image, in pixels.
width The width of the actual operating area of the image, in pixels.
height The height of the actual operating area of the image, in pixels.
  • Note

The actual operating area cannot exceed the image size, i.e(x + width)<= wstride, (y + height) <= hstride。

im_opt_t

  • description

Describes operation options of current image.

  • path

im2d_api/im2d_type.h

  • definitions
typedef struct im_opt {
    int color;                          /* color, used by color fill */
    im_colorkey_range colorkey_range;   /* range value of color key */
    im_nn_t nn;
    int rop_code;

    int priority;
    int core;
} im_opt_t;
Parameter Description
color Image color-fill configuration.
colorkey_range Colorkey range configuration.
nn Operation point preprocessing (quantization) configuration.
rop_code ROP operation code configuration.
priority Current task priority configuration.
core Specify the hardware core of current task.
  • Note

Permissions of priority and core are very high. Improper operations may cause system crash or deadlock. Therefore, users are advised to configure them only during development and debugging.Users are not advised to perform this configuration in actual product.

im_job_handle_t

  • 说明

RGA jobhandle, used to identify the currently configured RGA job.

  • 路径

im2d_api/im2d_type.h

  • 定义
typedef uint32_t im_job_handle_t;
  • 注意事项

After the configuration fails, imcancelJob must be used to release the current task handle to avoid memory leaks.

rga_buffer_handle_t

  • description

RGA driver image buffer handle.

  • path

im2d_api/im2d_type.h

  • definitions
typedef int rga_buffer_handle_t;
  • Note

When the buffer is used up, releasebuffer_handle must be used to release the memory to avoid memory leaks.

im_handle_param_t

  • description

Describe parameters of the image buffer to be imported.

  • path

im2d_api/im2d_type.h

  • definitions
typedef struct rga_memory_parm im_handle_param_t;

struct rga_memory_parm {
    uint32_t width_stride;
    uint32_t height_stride;
    uint32_t format;
};
Parameter Description
width_stride Describes the horizontal stride of the image buffer to be imported, in pixels.
height_stride Describes the vertical stride of the image buffer to be imported, in pixels.
format Describes the format of the buffer of the image to be imported.
  • Note

If the actual size of buffer memory is smaller than the configured size, the importbuffer_T API error occurs.

im_nn_t

  • description

Parameter of operation point preprocessing (quantization).

  • path

im2d_api/im2d_type.h

  • definitions
typedef struct im_nn {
    int scale_r;                /* scaling factor on R channal */
    int scale_g;                /* scaling factor on G channal */
    int scale_b;                /* scaling factor on B channal */
    int offset_r;               /* offset on R channal */
    int offset_g;               /* offset on G channal */
    int offset_b;               /* offset on B channal */
} im_nn_t;
Parameter Description
scale_r Scaling factor on red channel.
scale_g Scaling factor on green channel.
scale_b Scaling factor on blue channel.
offset_r Offset on red channel.
offset_g Offset on green channel.
offset_b Offset on blue channel.
  • Note

null

im_colorkey_range

  • description

Colorkey range.

  • path

im2d_api/im2d_type.h

  • definitions
typedef struct {
    int max;                    /* The Maximum value of the color key */
    int min;                    /* The minimum value of the color key */
} im_colorkey_range;
Parameter Description
max The Maximum value of the color key.
min The minimum value of the color key.
  • Note

null


Test Cases and Debugging Methods

In order to make developers get started with the above new API more quickly, here by running demo and parsing the source code to help developers to understand and use the API.

Test File Description

Input and output binary file for testing should be prepared in advance. The default source image file in RGBA8888 format is stored in directory /sample/sample_file.

In Android system, the source image should be stored in /data/ directory of device,in Linux system, the source image should be stored in/usr/data directory of device. The file naming rules are as follows:

in%dw%d-h%d-%s.bin
out%dw%d-h%d-%s.bin

Example:
1280×720 RGBA8888 input image: in0w1280-h720-rgba8888.bin
1280×720 RGBA8888 output image: out0w1280-h720-rgba8888.bin

Parameter descriptions:

The input is in ,the output is out. --->The first%d is the index of files, usually 0, used to distinguish files of the same format, width and height. --->The second%d is width, usually indicates virtual width. --->The third%d is height, usually indicates virtual height. --->The fourth%s is the name of format.

Some common image formats for preset tests are as follows. You can view names of other formats in rgaUtils.cpp:

format(Android) format(Linux) name
HAL_PIXEL_FORMAT_RGB_565 RK_FORMAT_RGB_565 "rgb565"
HAL_PIXEL_FORMAT_RGB_888 RK_FORMAT_RGB_888 "rgb888"
HAL_PIXEL_FORMAT_RGBA_8888 RK_FORMAT_RGBA_8888 "rgba8888"
HAL_PIXEL_FORMAT_RGBX_8888 RK_FORMAT_RGBX_8888 "rgbx8888"
HAL_PIXEL_FORMAT_BGRA_8888 RK_FORMAT_BGRA_8888 "bgra8888"
HAL_PIXEL_FORMAT_YCrCb_420_SP RK_FORMAT_YCrCb_420_SP "crcb420sp"
HAL_PIXEL_FORMAT_YCrCb_NV12 RK_FORMAT_YCbCr_420_SP "nv12"
HAL_PIXEL_FORMAT_YCrCb_NV12_VIDEO / "nv12"
HAL_PIXEL_FORMAT_YCrCb_NV12_10 RK_FORMAT_YCbCr_420_SP_10B "nv12_10"

The default resolution of input image of demo is 1280x720, format is RGBA8888, in0w1280-h720-rgba8888.bin source image should be prepared in advance in the /data or /usr/data directory, in1w1280-h720-rgba8888.bin source image should be additionally prepared in advance in the /data or /usr/data directory in image blending mode.

Debugging Method Description

After running demo, print log as follows (in image copying, for example):

Log is printed in Android system as follows:

# rgaImDemo --copy

librga:RGA_GET_VERSION:3.02,3.020000					//RGA version
ctx=0x7ba35c1520,ctx->rgaFd=3							//RGA context
Start selecting mode
im2d copy ..											//RGA running mode
GraphicBuffer check ok
GraphicBuffer check ok
lock buffer ok
open file ok											//src file status, if there is no corresponding file in /data/ directory, an error will be reported here
unlock buffer ok
lock buffer ok
unlock buffer ok
copying .... successfully								//indicates successful running
open /data/out0w1280-h720-rgba8888.bin and write ok		//output filename and directory

Log is printed in Linux system as follows:

# rgaImDemo --copy

librga:RGA_GET_VERSION:3.02,3.020000					//RGA version
ctx=0x2b070,ctx->rgaFd=3								//RGA context
Rga built version:version:1.00
Start selecting mode
im2d copy ..											//RGA running mode
open file												//src file status, if there is no corresponding file in /usr/data/ directory, an error will be reported here
copying .... Run successfully							//indicates successful running
open /usr/data/out0w1280-h720-rgba8888.bin and write ok	//output filename and directory

To view more detailed logs of RGA running, the Android system can enable RGA configuration log print by setting vendor.rga.log (Android 8 and below is sys.rga.log):

setprop vendor.rga.log 1		enable RGA log print
logcat -s librga				enable and filter log print
setprop vendor.rga.log 0		 cancel RGA log print

In Linux system, you should open core/NormalRgaContext.h, set __DEBUG to 1 and recompile.

#ifdef LINUX

-#define __DEBUG 0
+#define __DEBUG 1

Generally, the printed log is as follows, which can be uploaded to RedMine for analysis by relevant engineers of RK:

Log is printed in Android system as follows:

 D librga  : <<<<-------- print rgaLog -------->>>>
 D librga  : src->hnd = 0x0 , dst->hnd = 0x0
 D librga  : srcFd = 11 , phyAddr = 0x0 , virAddr = 0x0
 D librga  : dstFd = 15 , phyAddr = 0x0 , virAddr = 0x0
 D librga  : srcBuf = 0x0 , dstBuf = 0x0
 D librga  : blend = 0 , perpixelAlpha = 1
 D librga  : scaleMode = 0 , stretch = 0;
 D librga  : rgaVersion = 3.020000  , ditherEn =0
 D librga  : srcMmuFlag = 1 , dstMmuFlag = 1 , rotateMode = 0
 D librga  : <<<<-------- rgaReg -------->>>>
 D librga  : render_mode=0 rotate_mode=0
 D librga  : src:[b,0,e1000],x-y[0,0],w-h[1280,720],vw-vh[1280,720],f=0
 D librga  : dst:[f,0,e1000],x-y[0,0],w-h[1280,720],vw-vh[1280,720],f=0
 D librga  : pat:[0,0,0],x-y[0,0],w-h[0,0],vw-vh[0,0],f=0
 D librga  : ROP:[0,0,0],LUT[0]
 D librga  : color:[0,0,0,0,0]
 D librga  : MMU:[1,0,80000521]
 D librga  : mode[0,0,0,0]

Log is printed in Linux system as follows:

render_mode=0 rotate_mode=0
src:[0,a681a008,a68fb008],x-y[0,0],w-h[1280,720],vw-vh[1280,720],f=0
dst:[0,a6495008,a6576008],x-y[0,0],w-h[1280,720],vw-vh[1280,720],f=0
pat:[0,0,0],x-y[0,0],w-h[0,0],vw-vh[0,0],f=0
ROP:[0,0,0],LUT[0]
color:[0,0,0,0,0]
MMU:[1,0,80000521]
mode[0,0,0,0,0]
gr_color_x [0, 0, 0]
gr_color_x [0, 0, 0]

Test Case Descriptions

  • The test path is sample/im2d_API_demo. Developers can modify the demo configuration as required. It is recommended to use the default configuration when running demo for the first time.

  • The compilation of test cases varies on different platforms. On the Android platform, the 'mm' command can be used to compile the test cases. On the Linux platform, when librga.so is compiled using cmake, the corresponding test cases will be generated in the same directory

  • Import the executable file generated by compiling the corresponding test case into the device through adb, add the execution permission, execute demo, and check the printed log.

  • Check the output file to see if it matches your expectations.

Apply for Image Buffer

The demo provides two types of buffer for RGA synthesis: Graphicbuffer and AHardwareBuffer. The two buffers are distinguished by the macro USE_AHARDWAREBUFFER.

Directory:librga/samples/im2d_api_demo/Android.mk
(line +15ifeq (1,$(strip $(shell expr $(PLATFORM_SDK_VERSION) \> 25)))
/*if USE_AHARDWAREBUFFER is set to 1 then use AHardwareBuffer, if USE_AHARDWAREBUFFER is set to 0 then use Graphicbuffer*/
LOCAL_CFLAGS += -DUSE_AHARDWAREBUFFER=1
endif
Graphicbuffer

Graphicbuffer is initialized, filled/emptied, and filling rga_buffer_t structure through three functions.

	/*Passing in width, height, and image formats of src/dst and initialize Graphicbuffer*/
	src_buf = GraphicBuffer_Init(SRC_WIDTH, SRC_HEIGHT, SRC_FORMAT);
	dst_buf = GraphicBuffer_Init(DST_WIDTH, DST_HEIGHT, DST_FORMAT);

	/*Fill/empty Graphicbuffer by enumerating FILL_BUFF/EMPTY_BUFF*/
	GraphicBuffer_Fill(src_buf, FILL_BUFF, 0);
	if(MODE == MODE_BLEND)
		GraphicBuffer_Fill(dst_buf, FILL_BUFF, 1);
	else
		GraphicBuffer_Fill(dst_buf, EMPTY_BUFF, 1);

	/*Fill rga_buffer_t structure: src、dst*/
	src = wrapbuffer_GraphicBuffer(src_buf);
	dst = wrapbuffer_GraphicBuffer(dst_buf);
AHardwareBuffer

AHardwareBuffer is initialized, filled/emptied, and filling rga_buffer_t structure through three functions.

	/*Passing in width, height, and image formats of src/dst and initialize AHardwareBuffer*/
	AHardwareBuffer_Init(SRC_WIDTH, SRC_HEIGHT, SRC_FORMAT, &src_buf);
	AHardwareBuffer_Init(DST_WIDTH, DST_HEIGHT, DST_FORMAT, &dst_buf);

	/*Fill/empty AHardwareBuffer by enumerating FILL_BUFF/EMPTY_BUFF*/
	AHardwareBuffer_Fill(&src_buf, FILL_BUFF, 0);
	if(MODE == MODE_BLEND)
		AHardwareBuffer_Fill(&dst_buf, FILL_BUFF, 1);
	else
		AHardwareBuffer_Fill(&dst_buf, EMPTY_BUFF, 1);

	/*Fill rga_buffer_t structure: src、dst*/
	src = wrapbuffer_AHardwareBuffer(src_buf);
	dst = wrapbuffer_AHardwareBuffer(dst_buf);

Viewing Help Information

Run the following command to obtain the help information about the test case:

rgaImDemo -h
rgaImDemo --help
rgaImDemo

You can use demo according to the help information. The following information is printed:

rk3399_Android10:/ # rgaImDemo
librga:RGA_GET_VERSION:3.02,3.020000
ctx=0x7864d7c520,ctx->rgaFd=3

=============================================================================================
   usage: rgaImDemo [--help/-h] [--while/-w=(time)] [--querystring/--querystring=<options>]
                    [--copy] [--resize=<up/down>] [--crop] [--rotate=90/180/270]
                    [--flip=H/V] [--translate] [--blend] [--cvtcolor]
                    [--fill=blue/green/red]
         --help/-h     Call help
         --while/w     Set the loop mode. Users can set the number of cycles by themselves.
         --querystring You can print the version or support information corresponding to the current version of RGA according to the options.
                       If there is no input options, all versions and support information of the current version of RGA will be printed.
                       <options>:
                       vendor           Print vendor information.
                       version          Print RGA version, and librga/im2d_api version.
                       maxinput         Print max input resolution.
                       maxoutput        Print max output resolution.
                       scalelimit       Print scale limit.
                       inputformat      Print supported input formats.
                       outputformat     Print supported output formats.
                       expected         Print expected performance.
                       all              Print all information.
         --copy        Copy the image by RGA.The default is 720p to 720p.
         --resize      resize the image by RGA.You can choose to up(720p->1080p) or down(720p->480p).
         --crop        Crop the image by RGA.By default, a picture of 300*300 size is cropped from (100,100).
         --rotate      Rotate the image by RGA.You can choose to rotate 90/180/270 degrees.

         --flip        Flip the image by RGA.You can choice of horizontal flip or vertical flip.
         --translate   Translate the image by RGA.Default translation (300,300).
         --blend       Blend the image by RGA.Default, Porter-Duff 'SRC over DST'.
         --cvtcolor    Modify the image format and color space by RGA.The default is RGBA8888 to NV12.
         --fill        Fill the image by RGA to blue, green, red, when you set the option to the corresponding color.
=============================================================================================

Parameter parsing is in the directory /librga/demo/im2d_api_demo/args.cpp.

Executing Demo in Loop

Run the following command to loop demo. The loop command must precede all parameters. The number of cycles are of the type int and the default interval is 200ms.

rgaImDemo -w6 --copy
rgaImDemo --while=6 --copy

Obtain RGA Version and Support Information

Run the following command to obtain the version and support information:

rgaImDemo --querystring
rgaImDemo --querystring=<options>

If there is no input options, all versions and support information of current version of RGA will be printed.

options:
	=vendor				  Print vendor information.
	=version			  Print RGA version, and librga/im2d_api version.
	=maxinput			  Print max input resolution.
	=maxoutput			Print max output resolution.
	=scalelimit			Print scale limit.
	=inputformat		Print supported input formats.
	=outputformat		Print supported output formats.
	=expected			  Print expected performance.
	=all				    Print all information.
Code Parsing

According to parameters of main() to print different information.

	/*Convert the parameters of main() into QUERYSTRING_INFO enumeration values*/
	IM_INFO = (QUERYSTRING_INFO)parm_data[MODE_QUERYSTRING];
	/*Print the string returned by querystring(), which is the required information*/
	printf("\n%s\n", querystring(IM_INFO));

Image Resizing

Use the following command to test image resizing.

rgaImDemo --resize=up
rgaImDemo --resize=down

This function must be filled withoptions as follows:

options:
	=up					image resolution scale up to 1920x1080
	=down 			image resolution scale down to 720x480
Code Parsing

According to the parameters (up/down) of main() to choose to up(720p->1080p) or down(720p->480p), that is, for different scenarios, the buffer is re-initialized, emptied, or fills rga_buffer_t structure, and the rga_buffer_t structure that stores src and dst image data is passed to imresize().

	switch(parm_data[MODE_RESIZE])
	{
        /*scale up the image*/
		case IM_UP_SCALE :
            /*re-initialize Graphicbuffer to corresponding resolution 1920x1080*/
			dst_buf = GraphicBuffer_Init(1920, 1080, DST_FORMAT);
            /*empty the buffer*/
			GraphicBuffer_Fill(dst_buf, EMPTY_BUFF, 1);
            /*refill rga_buffer_t structure that stores dst data*/
			dst = wrapbuffer_GraphicBuffer(dst_buf);
			break;

		case IM_DOWN_SCALE :
            /*re-initialize Graphicbuffer to corresponding resolution 720x480**/
			dst_buf = GraphicBuffer_Init(720, 480, DST_FORMAT);
            /*empty the buffer*/
			GraphicBuffer_Fill(dst_buf, EMPTY_BUFF, 1);
            /*refill rga_buffer_t structure that stores dst data*/
			dst = wrapbuffer_GraphicBuffer(dst_buf);
			break;
	}
	/*pass src and dst of rga_buffer_t structure to imresize()*/
	STATUS = imresize(src, dst);
	/*print running status according to IM_STATUS enumeration values*/
	printf("resizing .... %s\n", imStrError(STATUS));

Image Cropping

Test image cropping using the following command.

rgaImDemo --crop

Options are not available for this feature. By default, crop the image within the coordinate LT(100,100),RT(400,100),LB(100,400),RB(400,400).

Code Parsing

Assign the size of clipped area in the im_rect structure that stores the src rectangle data, and pass the rga_buffer_t structure that stores the src and dst image data to imcrop().

	/*The coordinates of the clipped vertex are determined by x and y,the size of the clipped area is determined by width and height*/
	src_rect.x      = 100;
	src_rect.y      = 100;
	src_rect.width  = 300;
	src_rect.height = 300;

	/*pass src and dst of src_rect structure and rga_buffer_t structure format to imcrop()*/
	STATUS = imcrop(src, dst, src_rect);
	/*print the running status according to the returned IM_STATUS enumeration values*/
	printf("cropping .... %s\n", imStrError(STATUS));

Image Rotation

Test image rotation using the following command.

rgaImDemo --rotate=90
rgaImDemo --rotate=180
rgaImDemo --rotate=270

This function must be filled with options, which are as follows:

options:
	=90					Image rotation by 90°, exchange the width and height of output image resolution.
	=180 				Image rotation by 180°, output image resolution unchanged.
	=270 				Image rotation by 270°, exchange the width and height of output image resolution.
Code Parsing

According to the arguments (up/down) of main() to choose the rotation degrees(90/180/270). IM_USAGE enumeration transformed from arguments values, together with the rga_buffer_t structure that stores src and dst image data is passed to imrotate().

	/*convert the parameters of main() into IM_USAGE enumeration values*/
	ROTATE = (IM_USAGE)parm_data[MODE_ROTATE];

	/*pass both IM_USAGE enumeration value that identifies the rotation degrees and src and dst of rga_buffer_t structure format to imrotate()*/
	STATUS = imrotate(src, dst, ROTATE);
	/*print the running status according to the returned IM_STATUS enumeration values*/
	printf("rotating .... %s\n", imStrError(STATUS));

Image Mirror Flip

Use the following command to test mirror flipping

rgaImDemo --flip=H
rgaImDemo --flip=V

This function must be filled with options, which are as follows:

options:
	=H					Image horizontal mirror flip.
	=V 					Image vertical mirror flip.
Code Parsing

According to the arguments (H/V) of main() to choose the flipped direction, transform the arguments to IM_USAGE enumeration values, and the rga_buffer_t structure that stores src and dst image data is passed to imflip().

	/*convert the parameters of main() into IM_USAGE enumeration values*/
	FLIP = (IM_USAGE)parm_data[MODE_FLIP];

	/*pass both IM_USAGE enumeration value that identifies the flipped direction and src and dst of rga_buffer_t structure format to imflip()*/
	STATUS = imflip(src, dst, FLIP);
	/*print the running status according to the returned IM_STATUS enumeration value*/
	printf("flipping .... %s\n", imStrError(STATUS));

Image Color Fill

Use the following command to test the color fill.

rgaImDemo --fill=blue
rgaImDemo --fill=green
rgaImDemo --fill=red

This function must be filled with options. By default, fill the color of image within the coordinate LT(100,100),RT(400,100),LB(100,400),RB(400,400), options are as follows:

options:
	=blue				Fill the image color as blue.
	=green			Fill the image color as green.
	=red				Fill the image color as red.
Code Parsing

The filled color is determined according to the (blue/green/red) parameters of main(), and the size to be filled is assigned to the im_rect structure that stores the dst rectangle data, and the passed parameter is converted to the hexadecimal number of the corresponding color, which is passed to imfill() along with rga_buffer_t that stores the dst image data.

	/*Convert parameter of main() to hexadecimal number of the corresponding color*/
	COLOR = parm_data[MODE_FILL];

	/*The coordinates of clipping vertex are determined by x and y, and size of color-filled area is determined by width and height*/
	dst_rect.x      = 100;
	dst_rect.y      = 100;
	dst_rect.width  = 300;
	dst_rect.height = 300;

	/*Pass dst_rect of im_rect format and hexadecimal number of the corresponding color together with src and dst of rga_buffer_t format to imfill().*/
	STATUS = imfill(dst, dst_rect, COLOR);
	/*print the running status according to the returned IM_STATUS enumeration value*/
	printf("filling .... %s\n", imStrError(STATUS));

Image Translation

Use the following command to test image translation.

rgaImDemo --translate

This feature has no options. By default, the vertex (upper-left coordinate) is shifted to (300,300), that is, shifted 300 pixels to the right and 300 pixels down.

Code Parsing

Assign the offset of translation to the im_rect that stores the src rectangle data, and pass the rga_buffer_t structure that stores the src and dst image data to imtranslate().

	/*The coordinates of vertices of translated image are determined by x and y*/
	src_rect.x = 300;
	src_rect.y = 300;

	/*pass the src_rect of im_rect format along with src and dst of rga_buffer_t format into imtranslate()*/
	STATUS = imtranslate(src, dst, src_rect.x, src_rect.y);
	/*print the running status according to the returned IM_STATUS enumeration value*/
	printf("translating .... %s\n", imStrError(STATUS));

Image Copying

Use the following command to test image copying.

rgaImDemo --copy

This feature has no options. The default copy resolution is 1280x720 and the format is RGBA8888.

Code Parsing

Passing rga_buffer_t that stores src and dst image data to imcopy().

	/*passing src and dst of rga_buffer_t format to imcopy()*/
	STATUS = imcopy(src, dst);
	/*print the running status according to the returned IM_STATUS enumeration value*/
	printf("copying .... %s\n", imStrError(STATUS));

Image Blending

Use the following command to test image blending.

rgaImDemo --blend

This feature has no options. By default, the blending mode is IM_ALPHA_BLEND_DST.

Code Parsing

Passing rga_buffer_t that stores src and dst image data to imblend().

	/*passing src and dst of rga_buffer_t format to imblend()*/
	STATUS = imblend(src, dst);
	/*print the running status according to the returned IM_STATUS enumeration value*/
	printf("blending .... %s\n", imStrError(STATUS));

Image Format Conversion

Use the following command to test image format conversion.

rgaImDemo --cvtcolor

This feature has no options. By default, images with resolution of 1280x720 will be converted from RGBA8888 to NV12.

Code Parsing

Assign the format to be converted in the format variable member of rga_buffer_t, and pass the rga_buffer_t structure that stores src and dst image data to imcvtcolor().

	/*Assign the format in the format variable member*/
	src.format = HAL_PIXEL_FORMAT_RGBA_8888;
	dst.format = HAL_PIXEL_FORMAT_YCrCb_NV12;

	/*passing the format to be converted and src and dst of rga_buffer_t format to  imcvtcolor()*/
	STATUS = imcvtcolor(src, dst, src.format, dst.format);
	/*print the running status according to the returned IM_STATUS enumeration value*/
	printf("cvtcolor .... %s\n", imStrError(STATUS));