FP_SHARING duplication.

This commit removes the FP_SHARING and AUTOMATIC_FP_ENABLING
Kconfig symbols.

This Kconfig symbols were defined in arch/x86/Kconfig and
arc/x86/core/Kconfig as well.

Change-Id: I2ed19c32dbb8e16935343b71fb2c2739228e9fb2
Signed-off-by: Juan Manuel Cruz <juan.m.cruz.alcaraz@linux.intel.com>
This commit is contained in:
Juan Manuel Cruz 2015-07-30 16:39:46 -05:00 committed by Anas Nashif
commit f1a3ba3931

View file

@ -111,50 +111,4 @@ config LOCK_INSTRUCTION_UNSUPPORTED
endmenu endmenu
menu "Floating Point Options"
depends on !CPU_FLOAT_UNSUPPORTED && FLOAT && !ENHANCED_SECURITY
config FP_SHARING
bool
prompt "Floating point resource sharing" if !ENHANCED_SECURITY
depends on FLOAT
default n
select AUTOMATIC_FP_ENABLING if ENHANCED_SECURITY
help
This option allows multiple tasks and fibers to safely utilize
floating point hardware resources. Every task or fiber that wishes
to use floating point instructions (i.e. x87 FPU, MMX, or SSEx) must
be created with such support already enabled, or must enable this
support via task_float_enable() or fiber_float_enable() before
executing these instructions.
Enabling this option adds 108 bytes to the stack size requirement
of each task or fiber that utilizes x87 FPU or MMX instructions,
and adds 464 bytes to the stack size requirement of each task or
fiber that utilizes SSEx instructions. (The stack size requirement
of tasks and fibers that do not utilize floating point instructions
remains unchanged.)
Disabling this option means that only a single task or fiber may
utilize x87 FPU, MMX, or SSEx instructions. (The stack size
requirement of all tasks and fibers remains unchanged.)
config AUTOMATIC_FP_ENABLING
bool
prompt "Automatically enable floating point resource sharing" \
if !ENHANCED_SECURITY
depends on FP_SHARING
default n
help
This option allows tasks and fibers to safely utilize floating
point hardware resources without requiring them to first indicate
their intention to do so. The system automatically detects when
a task or fiber that does not currently have floating point support
enabled uses a floating point instruction, and automatically executes
task_float_enable() or fiber_float_enable() on its behalf. The
task or fiber is enabled for using x87 FPU, MMX, or SSEx instructions
if SSE is configured, otherwise it is enabled for using x87 FPU or
MMX instructions only.
endmenu
endmenu endmenu