Factorisation of NUCLEO_F302R8 and F334R8 cmsis folders

[bcostm]

• STM32Cube driver moved to TARGET_STM32F3 folder

[0xc0170]

Hi,

seems the previous merge for NUCLEO_L053R8 broke the mbed library. We should look at it. I'll run the build from master and check the latest merges.

[ohagendorf]

Yes, it is the same error message. And these two pull requests (#639 and 640) are the first after adding the L053 to build_travis.py.
The strange thing is, that both pull requests run successfully through a local build_travis.

[bcostm]

I have launched this command:
build.py -m NUCLEO_F334R8 -t uARM -j 4 -c --dsp --fat
as Bogdan suggested and every thing is ok. I did also the same for F302R8 and L053R8 and also ok.
Is this what you mean by Travis test ??? If not what is the command ?

[ohagendorf]

build_travis.py is a Python script in workspace_tools folder. It defines in a build_list variable several targets, toolchains and libraries. These are build automatically after a pull request.

The command line above is just one of these combinations. I our case the NUCLEO_L053R8 with gcc_arm build failes. The target is new in the script. This pull request and #639 are the first with it.

I had the same situation. My local builds, also the complete test with build_travis are OK.
I don't think that your pull request or my one in #639 are the problem. It has something to do with l053.

[0xc0170]

I just run it locally, it builds.. have to investigate

[bcostm]

Can you read the Details of the Travis CI build in GitHub ? When I click on the Details link nothing is displayed for me. So I cannot help too much...

Update: It was due to Internet Explorer. Using Chrome I can see the log.

I know. It is due to the file name. There is a "L" instead of "l"... We need to change the line:
include "stm32l0xx_hal_rcc_ex.h"

I remember having this error in the past... and I think if we use Windows we don't have any issue. If we use Linux yes...

[bogdanm]

This is the same issue as the one described in pull request #639