[stm32f7] Add initial support of the family, GPIO support.
Most changes are noise from doxygen. Readme udpated to explain newer FP_FLAGS for m7 stm32f7 library is skipped if the toolchain doesn't support it yet.
This commit is contained in:
committed by
Karl Palsson
parent
2e25d678ba
commit
3ef2c38120
15
README.md
15
README.md
@@ -67,6 +67,9 @@ Other toolchains _should_ work, but have not been nearly as well tested.
|
||||
Toolchains targetting linux, such as "gcc-arm-linux-gnu" or the like are
|
||||
_not_ appropriate.
|
||||
|
||||
_NOTE_ We recommend, that you use g-a-c version 2.8 2014q3 or newer
|
||||
to build all platforms covered by libopencm3 succesfully.
|
||||
|
||||
Building
|
||||
--------
|
||||
|
||||
@@ -92,9 +95,15 @@ them as environment variables, for example:
|
||||
* `FP_FLAGS` - Control the floating-point ABI
|
||||
|
||||
If the Cortex-M core supports a hard float ABI, it will be compiled with
|
||||
floating-point support by default. In cases where this is not desired, the
|
||||
behavior can be specified by setting `FP_FLAGS` Currently, M4F cores default
|
||||
to `-mfloat-abi=hard -mfpu=fpv4-sp-d16` and others to no FP flags
|
||||
best floating-point support by default. In cases where this is not desired, the
|
||||
behavior can be specified by setting `FP_FLAGS`.
|
||||
|
||||
Currently, M4F cores default to `-mfloat-abi=hard -mfpu=fpv4-sp-d16`,
|
||||
M7 core defaults to double precision `-mfloat-abi=hard -mfpu=fpv5-d16`
|
||||
and other architectures to no FP flags forcing to use software computation.
|
||||
|
||||
You may find which FP_FLAGS you can use in particular architecture in readme.txt
|
||||
shipped with gcc-arm-embedded package.
|
||||
|
||||
Examples:
|
||||
|
||||
|
||||
Reference in New Issue
Block a user