NVIDIA Tegra Linux Driver Package

Development Guide
28.3 Release

This topic provides an alphabetized list of specialized terms used in the Development Guide, with their definitons.
Boot Configuration Table (BCT)
The NVIDIA platform-specific static binary file used for flashing and updating the device. It is stored on a secondary boot device, such as eMMC, to store information required to:
Initialize and configure the secondary boot device
Initialize and configure the SDRAM
Locate and verify the boot loader
Boot and Power Management Processor (BPMP)
Firmware that runs on Cortex R5 used during boot to execute the boot ROM code and control the system-on-a-chip (SoC) sequence. After booting, the BPMP runs the power management functions.
Compute Unified Code Architecture (CUDA)
An NVIDIA created parallel computing platform and programming model that increases computing performance by harnessing the power of the Graphics Processing Unit (GPU).
Embedded Graphics Library (EGL)
An interface between Khronos rendering Application Programming Interfaces (APIs) and the underlying native platform windowing system. For example EGLStream is an extension that can be used to efficiently transfer a sequence of image frames from one API to another.
Microboot 1 (MB1)
The Microboot 1 bootloader runs after the boot ROM and performs these tasks:
Initializes the SDRAM based on MB1 boot configuration table (MB1-BCT)
Loads the firmware that initializes the CPU complex
Sets the security settings
Configures pinmux/GPIO for the SoC
Programs the PMIC for the VDD_CPU and VDD_DDR rails
Loads the next stage boot loader (MB2)
MB1 runs on the BPMP. It is considered an extended boot ROM because the security level of MB1 is the same as boot ROM.
Microboot 2 (MB2)
Microboot 2 runs after the MB1 bootloader. Its performs these tasks:
• Loads the CPU_BL
• Loads the BPMP-FW and its device tree binary (DTB)
• Loads the Trusted Operating System (TOS), SecureOS
MB2 runs on the BPMP.