incubator-nuttx/arch
Gregory Nutt 22cd0d47fa This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE.
This makes the user interface a little hostile.  People thing of an MTU of 1500 bytes, but the corresponding packet is really 1514 bytes (including the 14 byte Ethernet header).  A more friendly solution would configure the MTU (as before), but then derive the packet buffer size by adding the MAC header length.  Instead, we define the packet buffer size then derive the MTU.

The MTU is not common currency in networking.  On the wire, the only real issue is the MSS which is derived from MTU by subtracting the IP header and TCP header sizes (for the case of TCP).  Now it is derived for the PKTSIZE by subtracting the IP header, the TCP header, and the MAC header sizes.  So we should be all good and without the recurring 14 byte error in MTU's and MSS's.

Squashed commit of the following:

    Trivial update to fix some spacing issues.
    net/: Rename several macros containing _MTU to _PKTSIZE.
    net/: Rename CONFIG_NET_SLIP_MTU to CONFIG_NET_SLIP_PKTSIZE and similarly for CONFIG_NET_TUN_MTU.  These are not the MTU which does not include the size of the link layer header.  These are the full size of the packet buffer memory (minus any GUARD bytes).
    net/: Rename CONFIG_NET_6LOWPAN_MTU to CONFIG_NET_6LOWPAN_PKTSIZE and similarly for CONFIG_NET_TUN_MTU.  These are not the MTU which does not include the size of the link layer header.  These are the full size of the packet buffer memory (minus any GUARD bytes).
    net/: Rename CONFIG_NET_ETH_MTU to CONFIG_NET_ETH_PKTSIZE.  This is not the MTU which does not include the size of the link layer header.  This is the full size of the packet buffer memory (minus any GUARD bytes).
    net/: Rename the file d_mtu in the network driver structure to d_pktsize.  That value saved there is not the MTU.  The packetsize is the memory large enough to hold the maximum packet PLUS the size of the link layer header.  The MTU does not include the link layer header.
2018-07-04 14:10:40 -06:00
..
arm This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE. 2018-07-04 14:10:40 -06:00
avr syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
hc This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE. 2018-07-04 14:10:40 -06:00
mips This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE. 2018-07-04 14:10:40 -06:00
misoc This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE. 2018-07-04 14:10:40 -06:00
or1k syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
renesas Various locations: Fix typos due to missing right parenthese. Thanks to hyungsubkim for posting Issue #102 2018-06-16 16:19:32 -06:00
risc-v syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
sim This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE. 2018-07-04 14:10:40 -06:00
x86 syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
xtensa syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
z16 syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
z80 This commit attempts remove some long standard confusion in naming and some actual problems that result from the naming confusion. The basic problem is the standard MTU does not include the size of the Ethernet header. For clarity, I changed the naming of most things called MTU to PKTSIZE. For example, CONFIG_NET_ETH_MTU is now CONFIG_NET_ETH_PKTSIZE. 2018-07-04 14:10:40 -06:00
Kconfig arch/mips/src: All MIPS architctures now support CONFIG_ARCH_IDLE_CUSTOM 2018-05-07 11:17:19 -06:00
README.txt This commit adds basic architectural support for the OpenRISK mor1kx architecture. The initial commit was verified on a Qemu simulation but is otherwise untested. 2018-04-26 11:22:28 -06:00

README.txt

Architecture-Specific Code
^^^^^^^^^^^^^^^^^^^^^^^^^^
Table of Contents
^^^^^^^^^^^^^^^^^

  o Architecture-Specific Code
  o Summary of Files
  o Supported Architectures
  o Configuring NuttX

Architecture-Specific Code
^^^^^^^^^^^^^^^^^^^^^^^^^^

The NuttX configuration consists of:

o Processor architecture specific files.  These are the files contained
  in the arch/<arch-name>/ directory discussed in this README.

o Chip/SoC specific files.  Each processor processor architecture
  is embedded in chip or System-on-a-Chip (SoC) architecture.  The
  full chip architecture includes the processor architecture plus
  chip-specific interrupt logic, general purpose I/O (GIO) logic, and
  specialized, internal peripherals (such as UARTs, USB, etc.).

  These chip-specific files are contained within chip-specific
  sub-directories in the arch/<arch-name>/ directory and are selected
  via the CONFIG_ARCH_name selection

o Board specific files.  In order to be usable, the chip must be
  contained in a board environment.  The board configuration defines
  additional properties of the board including such things as
  peripheral LEDs, external peripherals (such as network, USB, etc.).

  These board-specific configuration files can be found in the
  configs/<board-name>/ sub-directories.

This README will address the processor architecture specific files
that are contained in the arch/<arch-name>/ directory. The file
include/nuttx/arch.h identifies all of the APIs that must
be provided by this architecture specific logic.  (It also includes
arch/<arch-name>/arch.h as described below).

Directory Structure
^^^^^^^^^^^^^^^^^^^

The arch directory contains architecture specific logic.  The complete
board port in is defined by the architecture-specific code in this
directory (plus the board-specific configurations in the config/
subdirectory).  Each architecture must provide a subdirectory <arch-name>
under arch/ with the following characteristics:


        <arch-name>/
        |-- include/
        |   |--<chip-name>/
        |   |  `-- (chip-specific header files)
        |   |--<other-chips>/
        |   |-- arch.h
        |   |-- irq.h
        |   `-- types.h
        `-- src/
            |--<chip-name>/
            |  `-- (chip-specific source files)
            |--<other-chips>/
            |-- Makefile
            `-- (architecture-specific source files)

Summary of Files
^^^^^^^^^^^^^^^^

include/<chip-name>/
  This sub-directory contains chip-specific header files.

include/arch.h
  This is a hook for any architecture specific definitions that may
  be needed by the system.  It is included by include/nuttx/arch.h

include/types.h
  This provides architecture/toolchain-specific definitions for
  standard types.  This file should typedef:

    _int8_t, _uint8_t, _int16_t, _uint16_t, _int32_t, _uint32_t

  and if the architecture supports 64-bit integers.

    _int24_t, _uint24_t, int64_t, uint64_t

  NOTE that these type names have a leading underscore character.  This
  file will be included(indirectly) by include/stdint.h and typedef'ed to
  the final name without the underscore character.  This roundabout way of
  doings things allows the stdint.h to be removed from the include/
  directory in the event that the user prefers to use the definitions
  provided by their toolchain header files

    irqstate_t

  Must be defined to the be the size required to hold the interrupt
  enable/disable state.

  This file will be included by include/sys/types.h and be made
  available to all files.

include/irq.h
  This file needs to define some architecture specific functions (usually
  inline if the compiler supports inlining) and structure.  These include:

  - struct xcptcontext.  This structures represents the saved context
    of a thread.

  - irqstate_t up_irq_save(void) -- Used to disable all interrupts.

  - void upirq_restore(irqstate_t flags) -- Used to restore interrupt
    enables to the same state as before up_irq_save was called.

  NOTE: These interfaces are not available to application code but can
  only be used withint the operating system code.  And, in general,
  these functions should *never* be called directly, not unless you
  know absolutely well what you are doing.  Rather you shoudl typically
  use the wrapper functions enter_critical_section() and leave_critical_section()
  as prototyped in include/nuttx/irq.h.

  This file must also define NR_IRQS, the total number of IRQs supported
  by the board.

src/<chip-name>/
  This sub-directory contains chip-specific source files.

src/Makefile
  This makefile will be executed to build the targets src/libup.a and
  src/up_head.o.  The up_head.o file holds the entry point into the system
  (power-on reset entry point, for example).  It will be used in
  the final link with libup.a and other system archives to generate the
  final executable.

Supported Architectures
^^^^^^^^^^^^^^^^^^^^^^^
NOTE: nuttx/Documentation/NuttX.html for current information about the
state of these MCU ports.

arch/sim - Linux/Cygwin simulation
    A user-mode port of NuttX to the x86 Linux platform is available.
    The purpose of this port is primarily to support OS feature development.
    This port does not support interrupts or a real timer (and hence no
    round robin scheduler)  Otherwise, it is complete.

arch/arm - ARM-based micro-controllers
    This directory holds common ARM architectures.  At present, this includes
    the following subdirectories:

    Architecture Support
      arch/arm/include and arch/arm/src/common
      arch/arm/src/arm and arch/arm/include/arm
      arch/arm/src/armv7-a and arch/arm/include/armv6-m
      arch/arm/src/armv7-a and arch/arm/include/armv7-a
      arch/arm/src/armv7-m and arch/arm/include/armv7-m
      arch/arm/src/armv7-r and arch/arm/include/armv7-r

    MCU support
      arch/arm/include/a1x and arch/arm/src/a1x
      arch/arm/include/c5471 and arch/arm/src/c5471
      arch/arm/include/dm320 and arch/arm/src/dm320
      arch/arm/include/efm32 and arch/arm/src/efm32
      arch/arm/include/imx1 and arch/arm/src/imx1
      arch/arm/include/imx6 and arch/arm/src/imx6
      arch/arm/include/kinetis and arch/arm/src/kinetis
      arch/arm/include/kl and arch/arm/src/kl
      arch/arm/include/lpc11xx and arch/arm/src/lpc11xx
      arch/arm/include/lpc17xx and arch/arm/src/lpc17xx
      arch/arm/include/lpc214x and arch/arm/src/lpc214x
      arch/arm/include/lpc2378 and arch/arm/src/lpc2378.
      arch/arm/include/lpc31xx and arch/arm/src/lpc31xx
      arch/arm/include/lpc43xx and arch/arm/src/lpc43xx
      arch/arm/include/moxart and arch/arm/src/moxart
      arch/arm/include/nuc1xx and arch/arm/src/nuc1xx
      arch/arm/include/sam34 and arch/arm/src/sam34
      arch/arm/include/sama45 and arch/arm/src/sama5
      arch/arm/include/samdl and arch/arm/src/samdl
      arch/arm/include/samv7 and arch/arm/src/samv7
      arch/arm/include/stm32 and arch/arm/src/stm32
      arch/arm/include/stm32f7 and arch/arm/src/stm32f7
      arch/arm/include/stm32l4 and arch/arm/src/stm32l4
      arch/arm/include/str71x and arch/arm/src/str71x
      arch/arm/include/tiva and arch/arm/src/tiva
      arch/arm/include/tms570 and arch/arm/src/tms570

arch/avr
    This directory is dedicated to ports to the Atmel AVR (8-bit) and AVR32 (32-bit)
    MCU families.  STATUS:  Under development.

    Architecture Support
      arch/avr/include/avr and arch/avr/src/avr
      arch/avr/include/avr32 and arch/avr/src/avr32

    MCU support
      arch/avr/include/atmega and arch/avr/src/atmega
      arch/avr/include/at90usb and arch/avr/src/at90usb
      arch/avr/include/at32uc3 and arch/avr/src/at32uc3

arch/hc
    This directory is dedicated to ports to the Freescale HC family.

    arch/arm/include/m9s12 and arch/arm/src/m9s12

arch/mips
    This directory is dedicated to ports to the MIPS family.

    Architecture Support
      arch/mips/include/mips32 and arch/mips/src/mips32

    MCU support
      arch/mips/include/pic32mx and arch/mips/src/pic32mx
      arch/mips/include/pic32mz and arch/mips/src/pic32mz

arch/renesas - Support for Renesas and legacy Hitachi microcontrollers.
    This include SuperH and M16C.

    Architecture Support
      arch/renesas/include and arch/renesas/src/common

    MCU support
      arch/renesas/include/m16c and arch/renesas/src/m16c
      arch/renesas/include/sh1 and arch/renesas/src/sh1

arch/or1k
    This directory is dedicated to ports to OpenRISC architectures.

    arch/or1k/include/mor1k and arch/or1k/src/mor1k

arch/risc-v
    This directory is dedicated to ports to the RISC-V family.

    Architecture Support
      arch/risc-v/include/rv32im

    MCU support
      arch/risc-v/include/nr5m100

arch/x86 - Intel x86 architectures
    This directory holds related, 32- and 64-bit architectures from Intel.
    At present, this includes the following subdirectories:

    Architecture Support
      arch/x86/include and arch/x86/src/common

    MCU support
      arch/x86/include/i486 and arch/x86/src/i486
      arch/x86/include/qemu and arch/x86/src/qemu

arch/xtensa

    Implementations based on the Cadence® Tensilica® Xtensa® processors,
    such as the Xtensa LX6 dataplane processing units (DPUs).  At
    present, this includes the following subdirectories:

    Common XTENSA support:
      arch/xtensa/include and arch/xtensa/src/common

    LX6 DPU support:
      arch/xtensa/include/lx6 and arch/xtensa/xtensa/lx6

    Expressif ESP32 implemenation of the LX6 DPU:
      arch/xtensa/include/esp32 and arch/xtensa/xtensa/esp32

arch/z16 - ZiLOG 16-bit processors
    This directory holds related, 16-bit architectures from ZiLOG.  At
    present, this includes the following subdirectories:

    Architecture Support
      arch/z16/include and arch/z16/src/common

    MCU support
      arch/z16/include/z16f and arch/z16/src/z16f

arch/z80 - ZiLOG 8-bit microcontrollers
    This directory holds related, 8-bit architectures from ZiLOG.  At
    present, this includes the following subdirectories:

    Architecture Support
      arch/z80/include and arch/z80/src/common

    MCU support
      arch/z80/include/z80 and arch/z80/src/z80
      arch/z80/include/z8 and arch/z80/src/z8
      arch/z80/include/ez80 and arch/z80/src/ez80