# Target definition: architecture, optimisations, etc...

menu "Target options"

comment "General target options"

choice

    bool

    prompt "Target architecture:"

    default ARCH_x86

config ARCH_ARM

    bool

    prompt "arm"

    select ARCH_SUPPORTS_BE

    select ARCH_SUPPORTS_LE

config ARCH_MIPS

    bool

    prompt "mips"

    select ARCH_SUPPORTS_BE

    select ARCH_SUPPORTS_LE

config ARCH_x86

    bool

    prompt "x86"

    select ARCH_SUPPORTS_LE

config ARCH_x86_64

    bool

    prompt "x86_64"

    select ARCH_SUPPORTS_LE

endchoice

config ARCH_SUPPORTS_BE

    bool

    default n

config ARCH_SUPPORTS_LE

    bool

    default n

choice

    bool

    prompt "Endianness:"

config ARCH_BE

    bool

    prompt "Big endian"

    depends on ARCH_SUPPORTS_BE

config ARCH_LE

    bool

    prompt "Little endian"

    depends on ARCH_SUPPORTS_LE

endchoice

comment "Target optimisations"

config ARCH_CPU

    string

    prompt "Emit assembly for CPU"

    default ""

    help

      This specifies the name of the target ARM processor. GCC uses this name

      to determine what kind of instructions it can emit when generating

      assembly code.

      Pick a value from the gcc manual for your choosen gcc version and your

      target CPU.

      Leave blank if you don't know, or if your target architecture does not

      offer this option.

config ARCH_TUNE

    string

    prompt "Tune for CPU"

    default ""

    help

      This option is very similar to the ARCH_CPU option (above), except

      that instead of specifying the actual target processor type, and hence

      restricting which instructions can be used, it specifies that GCC should

      tune the performance of the code as if the target were of the type

      specified in this option, but still choosing the instructions that it

      will generate based on the cpu specified by the ARCH_CPU option

      (above), or a (command-line) -mcpu= option.

      Pick a value from the gcc manual for your choosen gcc version and your

      target CPU.

      Leave blank if you don't know, or if your target architecture does not

      offer this option.

config ARCH_ARCH

    string

    prompt "Achitecture level"

    default ""

    help

      GCC uses this name to determine what kind of instructions it can emit

      when generating assembly code. This option can be used in conjunction

      with or instead of the ARCH_CPU option (above), or a (command-line)

      -mcpu= option.

      Pick a value from the gcc manual for your choosen gcc version and your

      target CPU.

      Leave blank if you don't know, or if your target architecture does not

      offer this option.

config ARCH_FPU

    string

    prompt "Use FPU"

    default ""

    help

      On some targets (eg. ARM), you can specify the kind of FPU to emit

      code for.

      See below wether to actually emit FP opcodes, or to emulate them.

      Pick a value from the gcc manual for your choosen gcc version and your

      target CPU.

      Leave blank if you don't know, or if your target architecture does not

      offer this option.

choice

    bool

    prompt "Floating point:"

config ARCH_FLOAT_HW

    bool

    prompt "hardware (FPU)"

    help

      Emit hardware floating point opcodes.

      If you've got a processor with a FPU, then you want that.

      If your hardware has no FPU, you still can use HW floating point, but

      need to compile support for FPU emulation in your kernel. Needless to

      say that emulating the FPU is /slooowwwww/...

      One situation you'd want HW floating point without a FPU is if you get

      binary blobs from different vendors that are compiling this way and

      can't (don't wan't to) change.

config ARCH_FLOAT_SW

    bool

    prompt "software"

    help

      Do not emit any hardware floating point opcode.

      If your processor has no FPU, then you most probably want this, as it

      is faster than emulating the FPU in the kernel.

endchoice

config ARCH_FLOAT_SW_LIBFLOAT

    bool

    prompt "Use libfloat"

    default n

    depends on ARCH_FLOAT_SW

    help

      For those targets upporting it, you can use libfloat for the software

      floating point emulation.

      Note that some versions of gcc have support code that supersedes libfloat,

      while others don't. Known version of gcc that don't have support code are

      versions prior to 3.0, and version above 4.0.

      You should check gcc before deciding to use libfloat.

config TARGET_CFLAGS

    string

    prompt "Default target CFLAGS"

    default ""

    help

      Used to add specific options when compiling libraries of the toolchain,

      that will run on the target (eg. libc.so).

      Note that the options above for CPU, tune, arch and FPU will be

      automaticaly used. You don't need to specify them here.

      Leave blank if you don't know better.

comment "Toolchain options"

config USE_SYSROOT

    bool

    prompt "Use sysroot'ed toolchain"

    default y

    help

      Use the 'shinny new' sysroot feature of gcc: libraries split between

      prefix/target/sys-root/lib and prefix/target/sys-root/usr/lib

      You definitely want to say 'Y' here. Yes you do. I know you do. Say 'Y'.

config SHARED_LIBS

    bool

    prompt "Build shared libraries"

    default y

    help

      Say 'y' here, unless you don't want shared libraries.

      You might not want shared librries if you're building for a target that

      don't support it (maybe some nommu targets, for example, or bare metal).

config TARGET_MULTILIB

    bool

#    prompt "Enable 'multilib' support (EXPERIMENTAL)"

    default n

    help

      Enable the so-called 'multilib' support.

      With the same toolchain, and on some architectures, you will be able to

      build big and little endian binaries, soft- and hard-float, etc...

      See the gcc configure manual at http://gcc.gnu.org/install/configure.html

      to see what multilib your target supports.

      It's preferable for now to build two (or more) toolchains, one for each

      configuration you need to support (eg. one for thumb and one for ARM,

      etc...). You can use the vendor string to diferentiate those toolchains.

config TARGET_VENDOR

    string

    prompt "Vendor string"

    default "unknown"

    help

      Vendor part of the machine triplet.

      A triplet is of the form arch-vendor-kernel-system.

      You can set the second part, vendor, to whatever you see fit.

      Use a single word, or use underscores "_" to separate words.

      Keep the default (unkown) if you don't know better.

config TARGET_ALIAS

    string

    prompt "Target alias"

    default ""

    help

      Normaly, you'd call your toolchain component (especially gcc) by

      prefixing the target triplet followed by a dash and the component name

      (eg. armeb-unknown-linux-uclibc-gcc).

      You can enter a shortcut here. This string will be used to create

      symbolic links to the toolchain tools (eg. if you enter "foo-bar" here,

      then gcc for your toolchain will also be available as "foo-bar-gcc" along

      with the original name).

      You shouldn't need to enter anything here, unless you plan to manually

      call the tools (autotools-based ./configure will use the standard name).

config ARCH

    string

    default "arm"     if ARCH_ARM

    default "mips"    if ARCH_MIPS

    default "x86"     if ARCH_x86

    default "x86_64"  if ARCH_x86_64

config BUILD 

    string

    prompt "Build system triplet"

    default ""

    help

      Canonical name of the machine building the toolchain.

      You should leave empty, unless you really now what you're doing.

config CC_NATIVE

    string

    prompt "Native gcc"

    default "gcc"

    help

      The native C compiler.

      You can set this to an alternative compiler if you have more than one

      installed (eg. gcc is gcc-4.1.1 and you want to use gcc-3.4.6).

      You can leave this empty as well, in which case gcc will be used.

config CANADIAN

    bool

    prompt "Canadian build (EXPERIMENTAL)"

    default n

    help

      A canadian build allows to build a compiler on a first machine

      (build system), that will run on second machine (host system),

      targetting a third machine (target system).

      An example where you'd want a candian cross-compiler is to create

      a native compiler for your target. In this case host and target

      are the same.

config HOST

    string

    prompt "Host system triplet"

    default ""

    depends on CANADIAN

    help

      Canonical name of the machine serving as host.

config HOST_CC

    string

    prompt "Host system compiler"

    default "${CT_HOST}-"

    depends on CANADIAN

    help

      C compiler targeting the host system.

      If HOST_CC ends with a dash (-), then it is considered to be the

      prefix to gcc (eg. x86-pc-linuc-gnu-).

      If it is empty, it is formed by appending '-gcc' to HOST.

      Else it is considered to be the complete name of the compiler, with

      full path, or without path (provided that it can be found in PATH).

endmenu