16. PowerPC Specific Information

16.1. Multilibs

The following multilibs are available:

  1. .: 32-bit PowerPC with FPU

  2. nof: 32-bit PowerPC with software floating point support

  3. m403: Instruction set for PPC403 with FPU

  4. m505: Instruction set for MPC505 with FPU

  5. m603e: Instruction set for MPC603e with FPU

  6. m603e/nof: Instruction set for MPC603e with software floating point support

  7. m604: Instruction set for MPC604 with FPU

  8. m604/nof: Instruction set for MPC604 with software floating point support

  9. m860: Instruction set for MPC860 with FPU

  10. m7400: Instruction set for MPC7500 with FPU

  11. m7400/nof: Instruction set for MPC7500 with software floating point support

  12. m8540: Instruction set for e200, e500 and e500v2 cores with single-precision FPU and SPE

  13. m8540/gprsdouble: Instruction set for e200, e500 and e500v2 cores with double-precision FPU and SPE

  14. m8540/nof/nospe: Instruction set for e200, e500 and e500v2 cores with software floating point support and no SPE

  15. me6500/m32: 32-bit instruction set for e6500 core with FPU and AltiVec

  16. me6500/m32/nof/noaltivec: 32-bit instruction set for e6500 core with software floating point support and no AltiVec

  17. me6500/m64: 64-bit instruction set for e6500 core with FPU and AltiVec

  18. me6500/m64/nof/noaltivec: 64-bit instruction set for e6500 core with software floating point support and no AltiVec

16.2. Application Binary Interface

In 32-bit PowerPC configurations the ABI defined by Power Architecture 32-bit Application Binary Interface Supplement 1.0 - Embedded is used.

In 64-bit PowerPC configurations the ABI defined by Power Architecture 64-Bit ELF V2 ABI Specification, Version 1.1 is used.

16.3. Special Registers

The following special-purpose registers are used by RTEMS:

Special-Purpose Register General 0 (SPRG0)

In SMP configurations, this register contains the address of the per-CPU control of the processor.

Special-Purpose Register General 1 (SPRG1)

This register contains the interrupt stack pointer for the outer-most interrupt service routine.

Special-Purpose Register General 2 (SPRG2)

This register contains the address of interrupt stack area begin.

16.4. Memory Model

The memory model is flat.

16.5. Interrupt Processing

16.5.1. Interrupt Levels

There are exactly two interrupt levels on PowerPC with respect to RTEMS. Level zero corresponds to interrupts enabled. Level one corresponds to interrupts disabled.

16.5.2. Interrupt Stack

The interrupt stack size can be configured via the CONFIGURE_INTERRUPT_STACK_SIZE application configuration option.

16.6. Default Fatal Error Processing

The default fatal error handler is BSP-specific.

16.7. Symmetric Multiprocessing

SMP is supported. Available platforms are the Freescale QorIQ P series (e.g. P1020) and T series (e.g. T2080, T4240).

16.8. Thread-Local Storage

Thread-local storage is supported.

16.9. 64-bit Caveats

  • The thread pointer is r13 in contrast to r2 used in the 32-bit ABI.

  • The TOC pointer is r2. It must be initialized as part of the C run-time setup. A valid stack pointer is not enough to call C functions. They may use the TOC to get addresses and constants.

  • The TOC must be within the first 2GiB of the address space. This simplifies the interrupt prologue, since the r2 can be set to .TOC. via the usual lis followed by ori combination. The lis is subject to sign-extension.

  • The PPC_REG_LOAD, PPC_REG_STORE, PPC_REG_STORE_UPDATE, and PPC_REG_CMP macros are available for assembly code to provide register size operations selected by the GCC -m32 and -m64 options.

  • The MSR[CM] bit must be set all the time, otherwise the MMU translation my yield unexpected results. The EPCR[ICM] or EPCR[GICM] bits may be used to enable the 64-bit compute mode for exceptions.