STM32: can_api.c files factorization #3737

bcostm · 2017-02-10T10:03:48Z

Description

This PR factorize the can_api.c files which were present in all TARGET_STM32Fx folders.

Changes:

A single can_api.c file is present in TARGET_STM folder. Common to all STM32 targets.
The specific part (mainly defines) are placed in new can_device.h files in TARGET_STM32Fx folders.

This will also update the can_filter function for all STM32 devices. It has been done previously only for F0 devices (PR #2988)

Status

READY

Migrations

If this PR changes any APIs or behaviors, give a short description of what API users should do when this PR is merged.

NO

Related PRs

#2988

Todos

[ ] Verify the can_filter function on other STM32 devices that F0 ones.

CC @martinjaeger @mgiaco

LMESTM

LGTM

LMESTM · 2017-02-10T14:18:29Z

targets/TARGET_STM/can_api.c

could we use
#if defined(TARGET_STM32F0)
..
#else

rather than the family list ?
Or maybe a specific property name that can be defined in can_device.h files ?
In case a new family is added, this would not require a change here.

Good idea. I'll put a "#else" instead of this big list.

sg- · 2017-02-13T16:36:49Z

If Travis doesn't merge with master you'll need to rebase this to get the fix that had to do with USB and STM folder renames.

bridadan · 2017-02-14T20:08:21Z

retest uvisor

bcostm · 2017-02-15T09:01:15Z

Rebased.

0xc0170 · 2017-02-15T11:16:28Z

/morph test-nightly

mbed-bot · 2017-02-15T11:32:03Z

Result: FAILURE

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 1597

Build failed!

0xc0170 · 2017-02-15T11:54:26Z

NUCLEO_F207ZG fails, all 3 toolchains:

05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_filter multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_free multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_frequency multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_init multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_irq_free multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_irq_init multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_irq_set multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_mode multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_monitor multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_rderror multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_read multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_reset multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_tderror multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol can_write multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol CAN1_RX0_IRQHandler multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol CAN1_SCE_IRQHandler multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol CAN1_TX_IRQHandler multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol CAN2_RX0_IRQHandler multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol CAN2_SCE_IRQHandler multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol CAN2_TX_IRQHandler multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol __asm___9_can_api_c_can_init____REV16 multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol __asm___9_can_api_c_can_init____REVSH multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Error: L6200E: Symbol __asm___9_can_api_c_can_init____RRX multiply defined (by can_api.o and can_api.o).
05:25:49         [DEBUG] Errors: Not enough information to list the image map.
05:25:49         [DEBUG] Errors: Finished: 1 information, 0 warning and 23 error messages.

bridadan · 2017-02-15T16:01:33Z

Hi @bcostm, you can also view detailed results by clicking on the mbed Build Number: 1597 link in the mbed-bot's last comment.

bcostm · 2017-02-16T09:12:37Z

I have removed the can_api.c file which was still present in the TARGET_STM32F2 folder. I think this is the root cause of this error. But I don't understand why I don't have this error myself when I build this target ? This is why I didn't see it before. I rebased again also.

bridadan · 2017-02-16T14:56:33Z

/morph test-nightly

mbed-bot · 2017-02-16T17:25:56Z

Result: SUCCESS

Your command has finished executing! Here's what you wrote!

/morph test-nightly

Output

mbed Build Number: 1612

All builds and test passed!

svogl · 2017-02-20T09:26:35Z

Hi all,
I'd really be in favor of merging this; we've a small bugfix here for can_write: the current implementation does not handle the case when no tx mailbox is available.

I can supply a patch for this PR or issue a separate one, whatever's less work.
Best
S

0xc0170 · 2017-02-20T09:46:12Z

I'd really be in favor of merging this; we've a small bugfix here for can_write: the current implementation does not handle the case when no tx mailbox is available.

@svogl Please send separate patch

0xc0170 · 2017-02-20T09:49:17Z

targets/TARGET_STM/can_api.c

Can you please fix formatting { as it was?

I am confused now. I thought the correct coding style was to have the { at the end of the function name ?

No, it is not. There are still some leftovers in the code base, we are trying to fix those.

Create can_device.h files to define specific code for the STM32 family

@mgiaco

Now MBED_A27/A28 tests are OK again. Problem seen also by @mgiaco and discussed in PR 2988.

Change also the coding style for the "{".

…d now)

This reverts commit 32b801b.

bridadan · 2017-02-20T18:48:13Z

retest uvisor

bridadan · 2017-02-20T19:16:56Z

/morph test

mbed-bot · 2017-02-21T00:38:04Z

Result: SUCCESS

Your command has finished executing! Here's what you wrote!

/morph test

Output

mbed Build Number: 1661

All builds and test passed!

svogl · 2017-02-22T08:15:10Z

Cool,
@0xc0170 when will 5.4.0 be released? I'll have to wait to apply our fixes until then.

Lynda96 · 2022-03-22T15:24:09Z

Hi
I am working on CAN bus with NUCLEO BOARD ,so I am coding with mbed studio , I am just doing a loop test (send data from CAN1TD and receive it from CAN2RD with the same board ) this is the code that I have tested

it works with frame standard but if i change the format on Extended , it doesn't work , what should you ?

bcostm force-pushed the dev_can_factorization branch from 410b13b to 3bcd827 Compare February 10, 2017 10:11

LMESTM reviewed Feb 10, 2017

View reviewed changes

sg- added the needs: CI label Feb 13, 2017

bcostm force-pushed the dev_can_factorization branch from 2b089e1 to d1911db Compare February 15, 2017 08:59

0xc0170 added needs: work and removed needs: CI labels Feb 15, 2017

bridadan added needs: review and removed needs: work labels Feb 16, 2017

0xc0170 requested changes Feb 20, 2017

View reviewed changes

0xc0170 added needs: work and removed needs: review labels Feb 20, 2017

bcostm added 7 commits February 20, 2017 11:19

Initial commit: copy can_api.c from F0 target

45c7e0d

Add F1 differences.

311648a

Create can_device.h files to define specific code for the STM32 family

Remove can_api.c file for F1 targets

5fa0c4e

Add F2 differences. Mainly add support for CAN2 instance.

1f47619

Add can_device.h for F2 targets

0fd596e

Add F3 differences.

b6b2167

Add support to F4 targets.

0c4688e

bcostm added 7 commits February 20, 2017 11:19

Add support to F7 targets.

caef97c

Add support to L4 targets.

f527ff9

F103 devices: Set CAN_RD pins in Input mode instead of A/F Push-Pull

8568b40

Now MBED_A27/A28 tests are OK again. Problem seen also by @mgiaco and discussed in PR 2988.

Add a #else to cover default STM32 devices for the IRQHandler functions.

eec6e10

Change also the coding style for the "{".

Coding style

32b801b

Remove can_api.c file present in TARGET_STM32F2 folder (no more neede…

02653ad

…d now)

Revert "Coding style"

30565cb

This reverts commit 32b801b.

bcostm force-pushed the dev_can_factorization branch from a8b6d2a to 30565cb Compare February 20, 2017 10:44

bridadan added needs: CI and removed needs: work labels Feb 20, 2017

jeromecoutant mentioned this pull request Feb 21, 2017

STM32 incoming CAN message wrong type and format #3792

Closed

0xc0170 approved these changes Feb 21, 2017

View reviewed changes

0xc0170 added ready for merge release-version: 5.4.0 and removed needs: CI labels Feb 21, 2017

sg- merged commit a8a31da into ARMmbed:master Feb 22, 2017

sg- removed the ready for merge label Feb 22, 2017

bcostm mentioned this pull request Feb 23, 2017

STM32F103 Nucleo CAN init hangs #3826

Closed

bcostm deleted the dev_can_factorization branch February 27, 2017 07:39

STM32: can_api.c files factorization #3737

STM32: can_api.c files factorization #3737

Uh oh!

Conversation

bcostm commented Feb 10, 2017

Description

Status

Migrations

Related PRs

Todos

Uh oh!

LMESTM left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

sg- commented Feb 13, 2017

Uh oh!

bridadan commented Feb 14, 2017

Uh oh!

bcostm commented Feb 15, 2017

Uh oh!

0xc0170 commented Feb 15, 2017

Uh oh!

mbed-bot commented Feb 15, 2017

Output

Uh oh!

0xc0170 commented Feb 15, 2017

Uh oh!

bridadan commented Feb 15, 2017

Uh oh!

bcostm commented Feb 16, 2017

Uh oh!

bridadan commented Feb 16, 2017

Uh oh!

mbed-bot commented Feb 16, 2017

Output

Uh oh!

svogl commented Feb 20, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

0xc0170 commented Feb 20, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bridadan commented Feb 20, 2017

Uh oh!

bridadan commented Feb 20, 2017

Uh oh!

mbed-bot commented Feb 21, 2017

Output

Uh oh!

svogl commented Feb 22, 2017

Uh oh!

Lynda96 commented Mar 22, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

svogl commented Feb 20, 2017 •

edited

Loading

0xc0170 commented Feb 20, 2017 •

edited

Loading