RTEMS CPU Architecture Supplement (6.a3cfaea).

Copyrights and License

© 2016, 2018 embedded brains GmbH & Co. KG

© 2016, 2018 Sebastian Huber

© 2014, 2015 Hesham Almatary

© 2010 Gedare Bloom

© 1988, 2018 On-Line Applications Research Corporation (OAR)

This document is available under the Creative Commons Attribution-ShareAlike 4.0 International Public License.

The authors have used their best efforts in preparing this material. These efforts include the development, research, and testing of the theories and programs to determine their effectiveness. No warranty of any kind, expressed or implied, with regard to the software or the material contained in this document is provided. No liability arising out of the application or use of any product described in this document is assumed. The authors reserve the right to revise this material and to make changes from time to time in the content hereof without obligation to notify anyone of such revision or changes.

The RTEMS Project is hosted at https://www.rtems.org. Any inquiries concerning RTEMS, its related support components, or its documentation should be directed to the RTEMS Project community.

RTEMS Online Resources

Home	https://www.rtems.org
Documentation	https://docs.rtems.org
Mailing Lists	https://lists.rtems.org
Bug Reporting	https://gitlab.rtems.org
Git Repositories	https://gitlab.rtems.org
Developers	https://gitlab.rtems.org

• 1. Preface
• 2. Port Specific Information
  • 2.1. CPU Model Dependent Features
    • 2.1.1. CPU Model Name
    • 2.1.2. Floating Point Unit
  • 2.2. Multilibs
  • 2.3. Calling Conventions
    • 2.3.1. Calling Mechanism
    • 2.3.2. Register Usage
    • 2.3.3. Parameter Passing
    • 2.3.4. User-Provided Routines
  • 2.4. Memory Model
    • 2.4.1. Flat Memory Model
  • 2.5. Interrupt Processing
    • 2.5.1. Vectoring of an Interrupt Handler
    • 2.5.2. Interrupt Levels
    • 2.5.3. Disabling of Interrupts by RTEMS
  • 2.6. Symmetric Multiprocessing
  • 2.7. Thread-Local Storage
  • 2.8. CPU counter
  • 2.9. Interrupt Profiling
  • 2.10. Board Support Packages
    • 2.10.1. System Reset
• 3. AArch64 Specific Information
  • 3.1. CPU Model Dependent Features
    • 3.1.1. CPU Model Name
    • 3.1.2. Floating Point Unit and SIMD
  • 3.2. Multilibs
  • 3.3. Calling Conventions
  • 3.4. Memory Model
  • 3.5. Interrupt Processing
    • 3.5.1. Interrupt Levels
    • 3.5.2. Interrupt Stack
  • 3.6. Symmetric Multiprocessing
  • 3.7. Thread-Local Storage
• 4. ARM Specific Information
  • 4.1. CPU Model Dependent Features
    • 4.1.1. CPU Model Name
    • 4.1.2. Count Leading Zeroes Instruction
    • 4.1.3. Floating Point Unit
  • 4.2. Multilibs
  • 4.3. Calling Conventions
  • 4.4. Memory Model
  • 4.5. Interrupt Processing
    • 4.5.1. Interrupt Levels
    • 4.5.2. Interrupt Stack
  • 4.6. Symmetric Multiprocessing
  • 4.7. Thread-Local Storage
• 5. Atmel AVR Specific Information
  • 5.1. CPU Model Dependent Features
    • 5.1.1. Count Leading Zeroes Instruction
  • 5.2. Calling Conventions
    • 5.2.1. Processor Background
    • 5.2.2. Register Usage
    • 5.2.3. Parameter Passing
  • 5.3. Memory Model
  • 5.4. Interrupt Processing
    • 5.4.1. Vectoring of an Interrupt Handler
    • 5.4.2. Disabling of Interrupts by RTEMS
    • 5.4.3. Interrupt Stack
  • 5.5. Symmetric Multiprocessing
  • 5.6. Thread-Local Storage
  • 5.7. Board Support Packages
    • 5.7.1. System Reset
• 6. Blackfin Specific Information
  • 6.1. CPU Model Dependent Features
    • 6.1.1. Count Leading Zeroes Instruction
  • 6.2. Calling Conventions
    • 6.2.1. Processor Background
    • 6.2.2. Register Usage
    • 6.2.3. Parameter Passing
  • 6.3. Memory Model
  • 6.4. Interrupt Processing
    • 6.4.1. Vectoring of an Interrupt Handler
    • 6.4.2. Disabling of Interrupts by RTEMS
    • 6.4.3. Interrupt Stack
  • 6.5. Symmetric Multiprocessing
  • 6.6. Thread-Local Storage
  • 6.7. Board Support Packages
    • 6.7.1. System Reset
• 7. Epiphany Specific Information
• 8. Intel/AMD x86 Specific Information
  • 8.1. CPU Model Dependent Features
    • 8.1.1. bswap Instruction
  • 8.2. Calling Conventions
    • 8.2.1. Processor Background
    • 8.2.2. Calling Mechanism
    • 8.2.3. Register Usage
    • 8.2.4. Parameter Passing
  • 8.3. Memory Model
    • 8.3.1. Flat Memory Model
  • 8.4. Interrupt Processing
    • 8.4.1. Vectoring of Interrupt Handler
    • 8.4.2. Interrupt Stack Frame
    • 8.4.3. Interrupt Levels
    • 8.4.4. Interrupt Stack
  • 8.5. Symmetric Multiprocessing
  • 8.6. Thread-Local Storage
  • 8.7. Board Support Packages
    • 8.7.1. System Reset
    • 8.7.2. Processor Initialization
• 9. Lattice Mico32 Specific Information
  • 9.1. CPU Model Dependent Features
  • 9.2. Register Architecture
  • 9.3. Calling Conventions
    • 9.3.1. Calling Mechanism
    • 9.3.2. Register Usage
    • 9.3.3. Parameter Passing
  • 9.4. Memory Model
  • 9.5. Interrupt Processing
  • 9.6. Symmetric Multiprocessing
  • 9.7. Thread-Local Storage
  • 9.8. Board Support Packages
    • 9.8.1. System Reset
• 10. Renesas M32C Specific Information
• 11. M68xxx and Coldfire Specific Information
  • 11.1. CPU Model Dependent Features
    • 11.1.1. BFFFO Instruction
    • 11.1.2. Vector Base Register
    • 11.1.3. Separate Stacks
    • 11.1.4. Pre-Indexing Address Mode
    • 11.1.5. Extend Byte to Long Instruction
  • 11.2. Calling Conventions
    • 11.2.1. Calling Mechanism
    • 11.2.2. Register Usage
    • 11.2.3. Parameter Passing
  • 11.3. Memory Model
  • 11.4. Interrupt Processing
    • 11.4.1. Vectoring of an Interrupt Handler
      • 11.4.1.1. Models Without Separate Interrupt Stacks
      • 11.4.1.2. Models With Separate Interrupt Stacks
    • 11.4.2. CPU Models Without VBR and RAM at 0
    • 11.4.3. Interrupt Levels
  • 11.5. Symmetric Multiprocessing
  • 11.6. Thread-Local Storage
  • 11.7. Board Support Packages
    • 11.7.1. System Reset
    • 11.7.2. Processor Initialization
• 12. Xilinx MicroBlaze Specific Information
  • 12.1. CPU Model Dependent Features
  • 12.2. Calling Conventions
  • 12.3. Interrupt Processing
    • 12.3.1. Interrupt Levels
    • 12.3.2. Interrupt Stack
  • 12.4. Symmetric Multiprocessing
  • 12.5. Thread-Local Storage
• 13. MIPS Specific Information
  • 13.1. CPU Model Dependent Features
    • 13.1.1. Another Optional Feature
  • 13.2. Calling Conventions
    • 13.2.1. Processor Background
    • 13.2.2. Calling Mechanism
    • 13.2.3. Register Usage
    • 13.2.4. Parameter Passing
  • 13.3. Memory Model
    • 13.3.1. Flat Memory Model
  • 13.4. Interrupt Processing
    • 13.4.1. Vectoring of an Interrupt Handler
    • 13.4.2. Interrupt Levels
  • 13.5. Symmetric Multiprocessing
  • 13.6. Thread-Local Storage
  • 13.7. Board Support Packages
    • 13.7.1. System Reset
    • 13.7.2. Processor Initialization
• 14. Altera Nios II Specific Information
  • 14.1. Symmetric Multiprocessing
  • 14.2. Thread-Local Storage
• 15. OpenRISC 1000 Specific Information
  • 15.1. Calling Conventions
    • 15.1.1. Floating Point Unit
  • 15.2. Memory Model
  • 15.3. Interrupt Processing
    • 15.3.1. Interrupt Levels
    • 15.3.2. Interrupt Stack
  • 15.4. Symmetric Multiprocessing
• 16. PowerPC Specific Information
  • 16.1. Multilibs
  • 16.2. Application Binary Interface
  • 16.3. Special Registers
  • 16.4. Memory Model
  • 16.5. Interrupt Processing
    • 16.5.1. Interrupt Levels
    • 16.5.2. Interrupt Stack
  • 16.6. Symmetric Multiprocessing
  • 16.7. Thread-Local Storage
  • 16.8. 64-bit Caveats
• 17. RISC-V Specific Information
  • 17.1. Calling Conventions
  • 17.2. Multilibs
  • 17.3. Interrupt Processing
    • 17.3.1. Interrupt Levels
    • 17.3.2. Interrupt Stack
  • 17.4. Symmetric Multiprocessing
  • 17.5. Thread-Local Storage
• 18. SuperH Specific Information
  • 18.1. CPU Model Dependent Features
    • 18.1.1. Another Optional Feature
  • 18.2. Calling Conventions
    • 18.2.1. Calling Mechanism
    • 18.2.2. Register Usage
    • 18.2.3. Parameter Passing
  • 18.3. Memory Model
    • 18.3.1. Flat Memory Model
  • 18.4. Interrupt Processing
    • 18.4.1. Vectoring of an Interrupt Handler
    • 18.4.2. Interrupt Levels
  • 18.5. Symmetric Multiprocessing
  • 18.6. Thread-Local Storage
  • 18.7. Board Support Packages
    • 18.7.1. System Reset
    • 18.7.2. Processor Initialization
• 19. SPARC Specific Information
  • 19.1. CPU Model Dependent Features
    • 19.1.1. CPU Model Feature Flags
      • 19.1.1.1. CPU Model Name
      • 19.1.1.2. Floating Point Unit
      • 19.1.1.3. Bitscan Instruction
      • 19.1.1.4. Number of Register Windows
      • 19.1.1.5. Low Power Mode
    • 19.1.2. CPU Model Implementation Notes
  • 19.2. Calling Conventions
    • 19.2.1. Programming Model
      • 19.2.1.1. Non-Floating Point Registers
      • 19.2.1.2. Floating Point Registers
      • 19.2.1.3. Special Registers
    • 19.2.2. Register Windows
    • 19.2.3. Call and Return Mechanism
    • 19.2.4. Calling Mechanism
    • 19.2.5. Register Usage
    • 19.2.6. Parameter Passing
    • 19.2.7. User-Provided Routines
  • 19.3. Memory Model
    • 19.3.1. Flat Memory Model
  • 19.4. Interrupt Processing
    • 19.4.1. Synchronous Versus Asynchronous Traps
    • 19.4.2. Trap Table
    • 19.4.3. Vectoring of Interrupt Handler
    • 19.4.4. Traps and Register Windows
    • 19.4.5. Interrupt Levels
    • 19.4.6. Disabling of Interrupts by RTEMS
    • 19.4.7. Interrupt Stack
  • 19.5. Symmetric Multiprocessing
  • 19.6. Thread-Local Storage
  • 19.7. Board Support Packages
    • 19.7.1. System Reset
    • 19.7.2. Processor Initialization
  • 19.8. Stacks and Register Windows
    • 19.8.1. General Structure
    • 19.8.2. Register Semantics
    • 19.8.3. Register Windows and the Stack
    • 19.8.4. Procedure epilogue and prologue
    • 19.8.5. Procedures, stacks, and debuggers
    • 19.8.6. The window overflow and underflow traps
• 20. SPARC-64 Specific Information
  • 20.1. CPU Model Dependent Features
    • 20.1.1. CPU Model Feature Flags
      • 20.1.1.1. CPU Model Name
      • 20.1.1.2. Floating Point Unit
      • 20.1.1.3. Number of Register Windows
    • 20.1.2. CPU Model Implementation Notes
      • 20.1.2.1. sun4u Notes
    • 20.1.3. sun4v Notes
  • 20.2. Calling Conventions
    • 20.2.1. Programming Model
      • 20.2.1.1. Non-Floating Point Registers
      • 20.2.1.2. Floating Point Registers
      • 20.2.1.3. Special Registers
    • 20.2.2. Register Windows
    • 20.2.3. Call and Return Mechanism
    • 20.2.4. Calling Mechanism
    • 20.2.5. Register Usage
    • 20.2.6. Parameter Passing
    • 20.2.7. User-Provided Routines
  • 20.3. Memory Model
    • 20.3.1. Flat Memory Model
  • 20.4. Interrupt Processing
    • 20.4.1. Synchronous Versus Asynchronous Traps
    • 20.4.2. Vectoring of Interrupt Handler
    • 20.4.3. Traps and Register Windows
    • 20.4.4. Interrupt Levels
    • 20.4.5. Disabling of Interrupts by RTEMS
    • 20.4.6. Interrupt Stack
  • 20.5. Symmetric Multiprocessing
  • 20.6. Thread-Local Storage
  • 20.7. Board Support Packages
    • 20.7.1. HelenOS and Open Firmware
• 21. References