Primary Git Repository for the Zephyr Project. Zephyr is a new generation, scalable, optimized, secure RTOS for multiple hardware architectures.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

68 lines
3.2 KiB

# Common architecture configuration options
# Copyright (c) 2022, CSIRO.
# SPDX-License-Identifier: Apache-2.0
config SEMIHOST
bool "Semihosting support for ARM, RISC-V and Xtensa targets"
depends on ARM || ARM64 || RISCV || (XTENSA && !SIMULATOR_XTENSA)
help
Semihosting is a mechanism that enables code running on an ARM, RISC-V
or Xtensa target to communicate and use the Input/Output facilities on
a host computer that is running a debugger.
Additional information can be found in:
https://developer.arm.com/documentation/dui0471/m/what-is-semihosting-
https://github.com/riscv-non-isa/riscv-semihosting/blob/main/riscv-semihosting.adoc
This option is compatible with hardware and with QEMU, through the
(automatic) use of the -semihosting-config switch when invoking it.
config ISR_TABLE_SHELL
bool "Shell command to dump the ISR tables"
depends on GEN_SW_ISR_TABLE
depends on SHELL
help
This option enables a shell command to dump the ISR tables.
config ARM_MPU
bool "ARM MPU Support"
select MPU
select SRAM_REGION_PERMISSIONS
select THREAD_STACK_INFO
select ARCH_HAS_EXECUTABLE_PAGE_BIT if (CPU_AARCH32_CORTEX_R || CPU_CORTEX_M)
select MPU_REQUIRES_POWER_OF_TWO_ALIGNMENT if !(CPU_HAS_NXP_SYSMPU || ARMV8_M_BASELINE || ARMV8_M_MAINLINE || AARCH32_ARMV8_R)
select MPU_REQUIRES_NON_OVERLAPPING_REGIONS if CPU_HAS_ARM_MPU && (ARMV8_M_BASELINE || ARMV8_M_MAINLINE || AARCH32_ARMV8_R)
select MPU_GAP_FILLING if AARCH32_ARMV8_R
select ARCH_MEM_DOMAIN_SUPPORTS_ISOLATED_STACKS if (CPU_AARCH32_CORTEX_R || CPU_CORTEX_M)
select ARCH_MEM_DOMAIN_SYNCHRONOUS_API if USERSPACE && CPU_AARCH64_CORTEX_R
default y if CPU_AARCH64_CORTEX_R
depends on CPU_HAS_MPU
help
MCU implements Memory Protection Unit.
Notes:
The ARMv6-M, ARMv7-M, and ARMv8-R MPU MPU architecture requires a power-of-two
alignment of MPU region base address and size.
The NXP MPU as well as the ARMv8-M MPU do not require MPU regions
to have power-of-two alignment for base address and region size.
The ARMv8-M and ARMv8-R MPU requires the active MPU regions be non-overlapping.
As a result of this, both respective MPUs needs to fully partition the
memory map when programming dynamic memory regions (e.g. PRIV stack
guard, user thread stack, and application memory domains), if the
system requires PRIV access policy different from the access policy
of the ARMv8-M or ARMv8-R background memory map. The application developer may
enforce full PRIV (kernel) memory partition by enabling the
MPU_GAP_FILLING option.
By not enforcing full partition, MPU may leave part of kernel
SRAM area covered only by the default ARMv8-M or ARMv8-R memory map. This
is fine for User Mode, since the background ARM map does not
allow nPRIV access at all. However, since the background map
policy allows instruction fetches by privileged code, forcing
this Kconfig option off prevents the system from directly
triggering MemManage exceptions upon accidental attempts to
execute code from SRAM in XIP builds.
Since this does not compromise User Mode, we make the skipping
of full partitioning the default behavior for the ARMv8-M and ARMv8-R MPU
driver.