Add the framework to have architecture-specific configuration and functions.
API is not yet defined.
1 # Target definition: architecture, optimisations, etc...
5 comment "General target options"
9 default "arm" if ARCH_ARM
10 default "ia64" if ARCH_IA64
11 default "mips" if ARCH_MIPS
12 default "x86" if ARCH_x86
13 default "x86_64" if ARCH_x86_64
17 prompt "Target architecture:"
23 select ARCH_SUPPORTS_BOTH_ENDIAN
24 select ARCH_DEFAULT_LE
28 prompt "ia64 (EXPERIMENTAL)"
29 depends on EXPERIMENTAL
30 select ARCH_SUPPORTS_BOTH_ENDIAN
35 select ARCH_SUPPORTS_BOTH_ENDIAN
36 select ARCH_DEFAULT_BE
48 config ARCH_SUPPORTS_BOTH_ENDIAN
52 config ARCH_DEFAULT_BE
56 config ARCH_DEFAULT_LE
63 depends on ARCH_SUPPORTS_BOTH_ENDIAN
64 default ARCH_BE if ARCH_DEFAULT_BE
65 default ARCH_LE if ARCH_DEFAULT_LE
73 prompt "Little endian"
77 # Include architecture-specific configuration
79 source config/arch/arm/config.in
82 source config/arch/ia64/config.in
85 source config/arch/mips/config.in
88 source config/arch/x86/config.in
91 source config/arch/x86_64/config.in
94 comment "Target optimisations"
98 prompt "Achitecture level"
101 GCC uses this name to determine what kind of instructions it can emit
102 when generating assembly code. This option can be used in conjunction
103 with or instead of the ARCH_CPU option (above), or a (command-line)
106 This is the configuration flag --with-arch=XXXX, and the runtime flag
109 Pick a value from the gcc manual for your choosen gcc version and your
112 Leave blank if you don't know, or if your target architecture does not
117 prompt "Generate code for the specific ABI"
120 Generate code for the given ABI.
122 This is the configuration flag --with-abi=XXXX, and the runtime flag
125 Pick a value from the gcc manual for your choosen gcc version and your
128 Leave blank if you don't know, or if your target architecutre does not
133 prompt "Emit assembly for CPU"
136 This specifies the name of the target processor. GCC uses this name
137 to determine what kind of instructions it can emit when generating
140 This is the configuration flag --with-cpu=XXXX, and the runtime flag
143 Pick a value from the gcc manual for your choosen gcc version and your
146 Leave blank if you don't know, or if your target architecture does not
151 prompt "Tune for CPU"
154 This option is very similar to the ARCH_CPU option (above), except
155 that instead of specifying the actual target processor type, and hence
156 restricting which instructions can be used, it specifies that GCC should
157 tune the performance of the code as if the target were of the type
158 specified in this option, but still choosing the instructions that it
159 will generate based on the cpu specified by the ARCH_CPU option
160 (above), or a (command-line) -mcpu= option.
162 This is the configuration flag --with-tune=XXXX, and the runtime flag
165 Pick a value from the gcc manual for your choosen gcc version and your
168 Leave blank if you don't know, or if your target architecture does not
173 prompt "Use specific FPU"
176 On some targets (eg. ARM), you can specify the kind of FPU to emit
179 This is the configuration flag --with-fpu=XXX, and the runtime flag
182 See below wether to actually emit FP opcodes, or to emulate them.
184 Pick a value from the gcc manual for your choosen gcc version and your
187 Leave blank if you don't know, or if your target architecture does not
192 prompt "Floating point:"
196 prompt "hardware (FPU)"
198 Emit hardware floating point opcodes.
200 If you've got a processor with a FPU, then you want that.
201 If your hardware has no FPU, you still can use HW floating point, but
202 need to compile support for FPU emulation in your kernel. Needless to
203 say that emulating the FPU is /slooowwwww/...
205 One situation you'd want HW floating point without a FPU is if you get
206 binary blobs from different vendors that are compiling this way and
207 can't (don't wan't to) change.
213 Do not emit any hardware floating point opcode.
215 If your processor has no FPU, then you most probably want this, as it
216 is faster than emulating the FPU in the kernel.
222 prompt "Target CFLAGS"
225 Used to add specific options when compiling libraries of the toolchain,
226 that will run on the target (eg. libc.so).
228 Note that the options above for CPU, tune, arch and FPU will be
229 automaticaly used. You don't need to specify them here.
231 Leave blank if you don't know better.