Nucleo f072 #4949

jia200x · 2016-03-02T13:19:24Z

Hi!

Added support for Nucleo F072 board. I'm having problems with OpenOCD for Mac OS X (I can flash only once, then stops working). Seems to work in Linux.

haukepetersen · 2016-03-02T13:53:07Z

looks good and works as expected. Could you please rebase and squash your commits? thanks

jia200x · 2016-03-02T15:04:29Z

Rebased and squashed.
Cheers!

haukepetersen · 2016-03-02T15:08:21Z

hm, seems like something went wrong. I am still seeing a number of commits that don't belong here..

jia200x · 2016-03-02T15:18:32Z

Yes, very weird... I have tried several times to rebase, but is not working
as I'm expecting. I think it's faster to create a new branch, add files and
change the variable in vendor header file and PR again. I think I'll do
that, is ok?

2016-03-02 16:08 GMT+01:00 Hauke Petersen notifications@github.com:

hm, seems like something went wrong. I am still seeing a number of commits
that don't belong here..

—
Reply to this email directly or view it on GitHub
#4949 (comment).

jia200x · 2016-03-02T15:47:34Z

Ok, fixed.
Could you test it again just for checking everything went right?

Thank you!

jia200x · 2016-03-03T13:15:58Z

It's seems now is OK

haukepetersen · 2016-03-03T13:35:17Z

looks good and works -> ACK once Travis is green.

jia200x · 2016-03-03T13:44:38Z

Cool!
Thank you.

2016-03-03 14:35 GMT+01:00 Hauke Petersen notifications@github.com:

looks good and works -> ACK once Travis is green.

—
Reply to this email directly or view it on GitHub
#4949 (comment).

OlegHahm · 2016-03-06T19:00:15Z

There are some Travis complaints:

boards/nucleo-f072/board.c:45: new blank line at EOF.

boards/nucleo-f072/include/board.h:68: new blank line at EOF.

boards/nucleo-f072/include/periph_conf.h:109: new blank line at EOF.

cpu/stm32f0/include/cpu_conf.h:33: trailing whitespace.

+ 

cpu/stm32f0/include/stm32f072xb.h:1: trailing whitespace.

+/**^M

cpu/stm32f0/include/stm32f072xb.h:2: trailing whitespace.

+  *****************************************************************************

cpu/stm32f0/include/stm32f072xb.h:3: trailing whitespace.

+  * @file    stm32f072xb.h^M

cpu/stm32f0/include/stm32f072xb.h:4: trailing whitespace.

+  * @author  MCD Application Team^M

cpu/stm32f0/include/stm32f072xb.h:5: trailing whitespace.

+  * @version V2.2.3^M

cpu/stm32f0/include/stm32f072xb.h:6: trailing whitespace.

+  * @date    29-January-2016^M

cpu/stm32f0/include/stm32f072xb.h:7: trailing whitespace.

+  * @brief   CMSIS Cortex-M0 Device Peripheral Access Layer Header File. ^M

cpu/stm32f0/include/stm32f072xb.h:8: trailing whitespace.

+  *          This file contains all the peripheral register's definitions, bits

cpu/stm32f0/include/stm32f072xb.h:9: trailing whitespace.

+  *          definitions and memory mapping for STM32F0xx devices.            

cpu/stm32f0/include/stm32f072xb.h:10: trailing whitespace.

+  *            ^M

cpu/stm32f0/include/stm32f072xb.h:11: trailing whitespace.

+  *          This file contains:^M

cpu/stm32f0/include/stm32f072xb.h:12: trailing whitespace.

+  *           - Data structures and the address mapping for all peripherals^M

cpu/stm32f0/include/stm32f072xb.h:13: trailing whitespace.

+  *           - Peripheral's registers declarations and bits definition^M

cpu/stm32f0/include/stm32f072xb.h:14: trailing whitespace.

+  *           - Macros to access peripheral<92>s registers hardware^M

cpu/stm32f0/include/stm32f072xb.h:15: trailing whitespace.

+  *  ^M

cpu/stm32f0/include/stm32f072xb.h:16: trailing whitespace.

+  *****************************************************************************

cpu/stm32f0/include/stm32f072xb.h:17: trailing whitespace.

+  * @attention^M

:

OlegHahm · 2016-03-06T19:00:50Z

It also needs some blacklisting for several applications.

haukepetersen · 2016-03-16T09:17:08Z

The Travis complains come from dos line endings in the vendor header file, so just convert them to unix ones...

jia200x · 2016-03-16T10:48:51Z

Hi!

Oops sorry, I forgot to finish this.
What kind of blacklisting do this model need?

I'm fixing it

2016-03-16 10:17 GMT+01:00 Hauke Petersen notifications@github.com:

The Travis complains come from dos line endings in the vendor header file,
so just convert them to unix ones...

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#4949 (comment)

OlegHahm · 2016-03-16T10:53:59Z

E.g.

gnrc_border_router:nucleo-f072:
Building application "gnrc_border_router" for "nucleo-f072" with MCU "stm32f0".
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: /home/travis/build/RIOT-OS/RIOT/examples/gnrc_border_router/bin/nucleo-f072/gnrc_border_router.elf section `.bss' will not fit in region `ram'
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: region `ram' overflowed by 3424 bytes
collect2: error: ld returned 1 exit status
make: *** [all] Error 1

this means, that you will have to add the board to the BOARD_INSUFFICIENT_MEMORY list in examples/gnrc_border_router/Makefile.

jia200x · 2016-03-16T10:55:21Z

Got it.
Thank you!

2016-03-16 11:54 GMT+01:00 Oleg Hahm notifications@github.com:

E.g.

gnrc_border_router:nucleo-f072:
Building application "gnrc_border_router" for "nucleo-f072" with MCU "stm32f0".
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: /home/travis/build/RIOT-OS/RIOT/examples/gnrc_border_router/bin/nucleo-f072/gnrc_border_router.elf section .bss' will not fit in regionram'
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: region `ram' overflowed by 3424 bytes
collect2: error: ld returned 1 exit status
make: *** [all] Error 1

this means, that you will have to add the board to the
BOARD_INSUFFICIENT_MEMORY list in examples/gnrc_border_router/Makefile.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#4949 (comment)

jia200x · 2016-03-16T12:07:49Z

Done.

I blacklisted the examples, but make command compiles everything and throws
"region 'ram' overflow". Is this a normal behavior?

Cheers!

2016-03-16 11:55 GMT+01:00 Jose Alamos jialamos@uc.cl:

Got it.
Thank you!

2016-03-16 11:54 GMT+01:00 Oleg Hahm notifications@github.com:

E.g.

gnrc_border_router:nucleo-f072:
Building application "gnrc_border_router" for "nucleo-f072" with MCU "stm32f0".
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: /home/travis/build/RIOT-OS/RIOT/examples/gnrc_border_router/bin/nucleo-f072/gnrc_border_router.elf section .bss' will not fit in regionram'
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: region `ram' overflowed by 3424 bytes
collect2: error: ld returned 1 exit status
make: *** [all] Error 1

this means, that you will have to add the board to the
BOARD_INSUFFICIENT_MEMORY list in examples/gnrc_border_router/Makefile.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#4949 (comment)

jia200x · 2016-03-21T18:19:23Z

Hi.

Failed again. The examples are not being build. I blacklisted some of them,
but make still tries to build them. It's normal?

Also don't know what happened with the vendor header file, It seems I
commited the old version. I'll fix it and try again.

2016-03-16 13:07 GMT+01:00 Jose Alamos jialamos@uc.cl:

Done.

I blacklisted the examples, but make command compiles everything and
throws "region 'ram' overflow". Is this a normal behavior?

Cheers!

2016-03-16 11:55 GMT+01:00 Jose Alamos jialamos@uc.cl:

Got it.
Thank you!

2016-03-16 11:54 GMT+01:00 Oleg Hahm notifications@github.com:

E.g.

gnrc_border_router:nucleo-f072:
Building application "gnrc_border_router" for "nucleo-f072" with MCU "stm32f0".
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: /home/travis/build/RIOT-OS/RIOT/examples/gnrc_border_router/bin/nucleo-f072/gnrc_border_router.elf section .bss' will not fit in regionram'
/usr/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld: region `ram' overflowed by 3424 bytes
collect2: error: ld returned 1 exit status
make: *** [all] Error 1

this means, that you will have to add the board to the
BOARD_INSUFFICIENT_MEMORY list in examples/gnrc_border_router/Makefile.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#4949 (comment)

OlegHahm · 2016-03-21T20:07:44Z

Somehow you seemed to have reverted the blacklisting in jia200x@06dbcdf

OlegHahm · 2016-03-30T21:37:11Z

I'm somewhat scared by the number of added lines, but since most of them are in the (bloated) STM CPU header, I guess this is fine. Additionally this PR introduces a new feature - so it cannot break much.

OlegHahm · 2016-03-30T21:39:23Z

boards/nucleo-f072/include/board.h

+
+#define LED_GREEN_ON        (LED_GREEN_PORT->BSRR = (1 << LED_GREEN_PIN))
+#define LED_GREEN_OFF       (LED_GREEN_PORT->BSRR = ((1 << LED_GREEN_PIN) << 16))
+#define LED_GREEN_TOGGLE    (LED_GREEN_PORT->ODR ^= (1 << LED_GREEN_PIN))


needs to be adapted to #5045.

I forgot to remove these lines.
Just done.

jia200x · 2016-03-31T10:02:09Z

Beside of the number of lines, the header files changed common variable names and added some legacy variables that breaks compiling. I had to remove the #define DAC1 line due to the new implementation of DAC.

I will be aware if breaks again with a new change.

haukepetersen · 2016-03-31T15:24:29Z

boards/nucleo-f072/include/periph_conf.h

+#define UART_1_AF           0
+/** @} */
+
+#define ADC_CONFIG {            \


missing doxygen for the ADC block

haukepetersen · 2016-03-31T15:26:04Z

I don't quite understand, why you removed the DAC1 from the vendor header. The device clearly has a DAC included, so you should not remove it. This error should be solved in a different way.

jia200x · 2016-03-31T15:32:10Z

Hi.

According to the vendor header file, DAC was defined in the same way as
DAC1 due to legacy reasons. When the DAC1 was present, dac.c file from
stm32_common [1] failed at line 61. (RCC_APB1ENR_DACEN is defined, not
RCC_APB1ENR_DAC1EN).

That was the reason why I removed DAC1.
Should I instead define RCC_APB1ENR_DAC1EN?

2016-03-31 17:26 GMT+02:00 Hauke Petersen notifications@github.com:

I don't quite understand, why you removed the DAC1 from the vendor header.
The device clearly has a DAC included, so you should not remove it. This
error should be solved in a different way.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#4949 (comment)

jia200x · 2016-03-31T15:41:03Z

[1]
https://github.com/RIOT-OS/RIOT/blob/d411387bec067098f2692de8332d07af82973137/cpu/stm32_common/periph/dac.c

2016-03-31 17:32 GMT+02:00 Jose Alamos jialamos@uc.cl:

Hi.

According to the vendor header file, DAC was defined in the same way as
DAC1 due to legacy reasons. When the DAC1 was present, dac.c file from
stm32_common [1] failed at line 61. (RCC_APB1ENR_DACEN is defined, not
RCC_APB1ENR_DAC1EN).

That was the reason why I removed DAC1.
Should I instead define RCC_APB1ENR_DAC1EN?

2016-03-31 17:26 GMT+02:00 Hauke Petersen notifications@github.com:

I don't quite understand, why you removed the DAC1 from the vendor
header. The device clearly has a DAC included, so you should not remove it.
This error should be solved in a different way.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#4949 (comment)

haukepetersen · 2016-03-31T15:45:40Z

Oh, I see, then never mind the comment about the DAC...

haukepetersen · 2016-03-31T15:47:23Z

would just be nice if you could squash your commits a little bit, I would suggest 1 commit for adding the CPU, 1 for editing the vendor header, 1 for adding the board, and 1 for the blacklisting...

jia200x · 2016-03-31T16:10:43Z

I'm on it. Had some conflicts with rebase, but I'm fixing them.

…variables.

haukepetersen · 2016-03-31T17:03:49Z

looks good, just one more thing that Murdock complains about, there are some whitespace errors in the board files:


BUILD OUTPUT of group static-tests:
Current branch HEAD is up to date.
Current branch HEAD is up to date.
boards/nucleo-f072/board.c:34: new blank line at EOF.
boards/nucleo-f072/include/board.h:48: new blank line at EOF.
boards/nucleo-f072/include/periph_conf.h:136: new blank line at EOF.
boards/nucleo-f072/board.c:34: new blank line at EOF.
boards/nucleo-f072/include/board.h:48: new blank line at EOF.
boards/nucleo-f072/include/periph_conf.h:136: new blank line at EOF.
ERROR: This change introduces new whitespace errors

jia200x · 2016-03-31T17:08:34Z

Done. It seems it was introducing when I tried to solve conflicts of rebase.

haukepetersen · 2016-03-31T17:18:09Z

Alright, Murdock is green, code looks good, I'll make an exception here and merge this (as feature freeze is very young and the release branch is not created yet...) -> ACK and go

jia200x · 2016-03-31T17:19:09Z

:)

haukepetersen · 2016-03-31T17:20:00Z

nice work by the way!

jia200x · 2016-03-31T17:21:22Z

Thank you!
Helped me a lot to understand PR<->CI-> ACK loop.

I will check for missing documentation about this.
Cheers!

haukepetersen added the Type: new feature The issue requests / The PR implemements a new feature for RIOT label Mar 2, 2016

haukepetersen added this to the Release 2016.04 milestone Mar 2, 2016

haukepetersen self-assigned this Mar 2, 2016

haukepetersen added CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR and removed CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR labels Mar 2, 2016

jia200x force-pushed the nucleo-f072 branch from 486f028 to 2382d54 Compare March 2, 2016 15:46

OlegHahm added the Platform: ARM Platform: This PR/issue effects ARM-based platforms label Mar 3, 2016

jia200x force-pushed the nucleo-f072 branch from ff2c6c8 to 17c46a8 Compare March 3, 2016 13:14

haukepetersen added the CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR label Mar 3, 2016

jia200x force-pushed the nucleo-f072 branch 4 times, most recently from 546b2e3 to 06dbcdf Compare March 16, 2016 12:04

OlegHahm reviewed Mar 30, 2016
View reviewed changes

jia200x force-pushed the nucleo-f072 branch from b0df2ff to d14e5ca Compare March 31, 2016 09:55

haukepetersen reviewed Mar 31, 2016
View reviewed changes

jia200x force-pushed the nucleo-f072 branch from d14e5ca to da89584 Compare March 31, 2016 15:38

jia200x added 2 commits March 31, 2016 18:20

stm32f072rb: Added support for CPU

20d6ead

stm32f072rb: Adapter vendor header file due to inconsistency in some …

4ea6d26

…variables.

jia200x force-pushed the nucleo-f072 branch from da89584 to a743275 Compare March 31, 2016 16:22

jia200x force-pushed the nucleo-f072 branch from a743275 to cc0e94f Compare March 31, 2016 17:08

jia200x added 2 commits March 31, 2016 19:08

nucleo-f072: Added support for board.

62de9c9

blacklist: nucleo-f072: blacklisted several examples and tests

cc0e94f

haukepetersen merged commit 02df871 into RIOT-OS:master Mar 31, 2016

jia200x deleted the nucleo-f072 branch April 1, 2016 13:54

OlegHahm modified the milestones: Release 2016.04, Release 2016.07 Apr 1, 2016

aabadie mentioned this pull request Sep 16, 2016

boards: initial support of ST Nucleo f446 #5706

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Nucleo f072 #4949

Nucleo f072 #4949

jia200x commented Mar 2, 2016

haukepetersen commented Mar 2, 2016

jia200x commented Mar 2, 2016

haukepetersen commented Mar 2, 2016

jia200x commented Mar 2, 2016

jia200x commented Mar 2, 2016

jia200x commented Mar 3, 2016

haukepetersen commented Mar 3, 2016

jia200x commented Mar 3, 2016

OlegHahm commented Mar 6, 2016

OlegHahm commented Mar 6, 2016

haukepetersen commented Mar 16, 2016

jia200x commented Mar 16, 2016

OlegHahm commented Mar 16, 2016

jia200x commented Mar 16, 2016

jia200x commented Mar 16, 2016

jia200x commented Mar 21, 2016

OlegHahm commented Mar 21, 2016

OlegHahm commented Mar 30, 2016

OlegHahm Mar 30, 2016

jia200x Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

Nucleo f072 #4949

Nucleo f072 #4949

Conversation

jia200x commented Mar 2, 2016

haukepetersen commented Mar 2, 2016

jia200x commented Mar 2, 2016

haukepetersen commented Mar 2, 2016

jia200x commented Mar 2, 2016

jia200x commented Mar 2, 2016

jia200x commented Mar 3, 2016

haukepetersen commented Mar 3, 2016

jia200x commented Mar 3, 2016

OlegHahm commented Mar 6, 2016

OlegHahm commented Mar 6, 2016

haukepetersen commented Mar 16, 2016

jia200x commented Mar 16, 2016

OlegHahm commented Mar 16, 2016

jia200x commented Mar 16, 2016

jia200x commented Mar 16, 2016

jia200x commented Mar 21, 2016

OlegHahm commented Mar 21, 2016

OlegHahm commented Mar 30, 2016

OlegHahm Mar 30, 2016

Choose a reason for hiding this comment

jia200x Mar 31, 2016

Choose a reason for hiding this comment

jia200x commented Mar 31, 2016

haukepetersen Mar 31, 2016

Choose a reason for hiding this comment

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016

haukepetersen commented Mar 31, 2016

jia200x commented Mar 31, 2016