kinetis: adaption for common makefiles by jfischer-no · Pull Request #3099 · RIOT-OS/RIOT

jfischer-no · 2015-05-27T21:12:32Z

like "refactor board makefiles #2852"

@gebart Do you agree with 7704626 and you can test it please? I did not dare to change the Mulle's Makefile :-).

~~3150af0 is also necessary because not every Cortex-M4 has a FPU.~~

haukepetersen · 2015-05-27T21:27:00Z

nice. The FPU issue is fixed in #3023. So I would suggest to merge #3023 first and adapt this...

jfischer-no · 2015-05-27T21:36:08Z

@haukepetersen ok, will rebase/refactor it after #3023 is merged.

jnohlgard · 2015-05-27T22:12:18Z

@jfischer-phytec-iot I will test this asap. You may delete all of the devio stuff as well I guess.

jfischer-no · 2015-05-28T09:44:20Z

rebased on #3023

jfischer-no · 2015-05-29T08:46:00Z

rebased on master, wait until #3100 is done.

jnohlgard · 2015-05-30T07:22:24Z

#3100 is merged now. Please rebase.

jfischer-no · 2015-05-30T11:01:00Z

@gebart rebased, please review and test it. I had a strange behavior of I2C if heap is placed at the end of ram. I initially left it as it was.

jnohlgard · 2015-05-30T11:02:06Z

I have some fixes for mulle, I have opened a PR in your repo at https://github.com/jfischer-phytec-iot/RIOT/pull/15

jnohlgard · 2015-05-30T11:02:56Z

Did you fix the i2c issue?
I don't have any i2c devices with me right now to test with, but I can probably do it on Monday.

jnohlgard · 2015-05-30T11:04:17Z

cpu/kinetis_common/ldscripts/kinetis-base.ld

This seem like a left over after your latest rebase.

I want to keep it for the time being.

jfischer-no · 2015-05-30T11:12:39Z

@gebart I do not have found a issue. If the heap moved backwards, then hdc1000 sernsor no longer works (also on the FRDM k64f-board), everything else works. I have no way to measure on i2c until Monday :-)

jfischer-no · 2015-05-30T11:19:02Z

I mark it WIP (It only applies to c20ef10) until I checked i2c on Monday.

jnohlgard · 2015-05-30T11:32:08Z

A hint: examine the .map file and see if any .data or .bss symbols overlap
with the sheap - eheap area.
On May 30, 2015 1:19 PM, "Johann Fischer" [email protected] wrote:

I mark it WIP (It only applies to c20ef10
c20ef10)
until I checked i2c on Monday.

—
Reply to this email directly or view it on GitHub
#3099 (comment).

jfischer-no · 2015-05-30T12:23:11Z

@gebart Mulle build was broken, look at afe1788, it was necessary to move linkerscripts because they are hardcoded to $(RIOTCPU)/$(CPU)/$(CPU_MODEL)_linkersscript.ld in boards/Makefile.include.cortexm_common.

jnohlgard · 2015-05-30T12:27:38Z

@jfischer-phytec-iot rebase on latest master since #3120, like I wrote in the original PR in your repo, "Depends on #3120"

jfischer-no · 2015-05-30T12:31:04Z

@gebart Sorry, got it too late, will revert and rebase

jfischer-no · 2015-05-30T13:37:28Z

rebased

jfischer-no · 2015-05-30T18:32:10Z

@gebart I have found the reason for the anomaly, #3124. Ready for review.

jnohlgard · 2015-05-30T19:23:46Z

Nice, I will review and test this on mulle tomorrow.

jnohlgard · 2015-05-31T08:18:56Z

Works fine except for make term, I forgot to add the proper include. I opened https://github.com/jfischer-phytec-iot/RIOT/pull/16

jnohlgard · 2015-05-31T11:37:27Z

ACK, please squash all commits marked "squash"

jfischer-no · 2015-05-31T11:46:43Z

@jremmert-phytec-iot We had that before, what was it ? -> failed: driver_kw2xrf, driver_xbee

jfischer-no · 2015-05-31T11:47:49Z

ups

jnohlgard · 2015-05-31T11:53:25Z

the error message is:

riot/sys/auto_init/netif/auto_init_ng_at86rf2xx.c:27:33: fatal error: ng_at86rf2xx_params.h: No such file or directory
 #include "ng_at86rf2xx_params.h"

so it seems like a problem with the ng_at86rf2xx module

jnohlgard · 2015-05-31T11:55:36Z

I think we can just blacklist the xbee and kw2xrf tests for mulle for the time being until I have had time to properly try out the ng_ stack. The kw2xrf driver (not sure about xbee) is not applicable to the mulle platform anyway.

jfischer-no · 2015-05-31T11:58:12Z

The kw2xrf driver (not sure about xbee) is not applicable to the mulle platform anyway.

@gebart maybe yes : http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MCR20A
This is the same modem.

jonas-rem · 2015-05-31T12:49:52Z

@jremmert-phytec-iot We had that before, what was it ? -> failed: driver_kw2xrf, driver_xbee

Without having a deeper look into it, the issues we had before were:

The Makefile of the test application for the radio-drivers defines default GPIO-pins if the Board did not define pins for that radio module (I think the pins are counted for GPIO0 to 4 or so).
And the second problem. The driver module was missing in some Makefile (I guess it was Makefile.include) that the driver was not .

jonas-rem · 2015-05-31T13:45:04Z

@gebart you need to define the pins (SPI, IRQ-pin...) for the radio-driver you want to use. E.g. for the samr21-xpro board these pin assignments are defined in ./boards/samr21-xpro/include/ng_at86rf2xx_params.h

The makefile in ./boards/mulle/Makfile.dep tells to use the ng_at... driver. Without that module definition it should build either.

jnohlgard · 2015-05-31T15:13:50Z

@jremmert-phytec-iot Thank you for the explanation. I think I will make a separate PR with the ng_netif change..

@jfischer-phytec-iot could you remove https://github.com/jfischer-phytec-iot/RIOT/commit/1d06ddebaebbe2f53c9db7b2aedb3ba534659408 (mulle: Add dependency check for ng_netif) from this PR?

I'll open a specific PR for the transceiver settings for mulle.

Add a exception for common startup

jfischer-no · 2015-05-31T16:34:21Z

@gebart removed 1d06dde, squashed and rebased on master

jnohlgard · 2015-06-01T04:56:03Z

ACK
@haukepetersen do you want to say anything or can we merge this so that all Cortex M platforms are finally aligned with regards to Makefiles and linker?

jfischer-no · 2015-06-02T11:37:55Z

broadcast: Can we merge it?

kaspar030 · 2015-06-02T12:07:52Z

Yeah, merge!

…on-makefile kinetis: adaption for common makefiles

jfischer-no added Type: enhancement The issue suggests enhanceable parts / The PR enhances parts of the codebase / documentation Type: cleanup The issue proposes a clean-up / The PR cleans-up parts of the codebase / documentation labels May 27, 2015

jfischer-no assigned kaspar030 May 27, 2015

jfischer-no added the State: waiting for other PR State: The PR requires another PR to be merged first label May 27, 2015

jfischer-no assigned haukepetersen and unassigned kaspar030 May 28, 2015

OlegHahm modified the milestone: Release 2015.06 May 28, 2015

jnohlgard removed the State: waiting for other PR State: The PR requires another PR to be merged first label May 30, 2015

jnohlgard reviewed May 30, 2015
View reviewed changes

jfischer-no added the State: WIP State: The PR is still work-in-progress and its code is not in its final presentable form yet label May 30, 2015

jfischer-no removed the State: WIP State: The PR is still work-in-progress and its code is not in its final presentable form yet label May 30, 2015

jnohlgard mentioned this pull request May 31, 2015

pkg/cmsis-dsp: Add CMSIS-DSP as a package #3028

Merged

jnohlgard added the CI: needs squashing Commits in this PR need to be squashed; If set, CI systems will mark this PR as unmergable label May 31, 2015

jfischer-no closed this May 31, 2015

jfischer-no reopened this May 31, 2015

jfischer-no and others added 7 commits May 31, 2015 18:24

Makefile.include.cortex_common: enhancement for kinetis_common

7bc468a

Add a exception for common startup

cpu/kinetis_common/ldscripts: adapt for newlib module

072d0a8

cpu/kw2x: use cortex common makefile and newlib module

002ba9b

cpu/k60: use cortex common makefile and newlib module

1454276

boards/pba-d-01-kw2x: adapt to common board Makefile

cbfc236

mulle: remove device map

6e12503

mulle: Use common Cortex M build enviornment

612044d

jfischer-no removed the CI: needs squashing Commits in this PR need to be squashed; If set, CI systems will mark this PR as unmergable label May 31, 2015

kaspar030 added a commit that referenced this pull request Jun 2, 2015

Merge pull request #3099 from jfischer-phytec-iot/pr@kinetis-use-comm…

369a47a

…on-makefile kinetis: adaption for common makefiles

kaspar030 merged commit 369a47a into RIOT-OS:master Jun 2, 2015

jnohlgard mentioned this pull request Jun 4, 2015

sys: factor stdio uart functions out of newlib's syscalls #3161

Merged

jfischer-no deleted the pr@kinetis-use-common-makefile branch July 6, 2015 15:51

Conversation

jfischer-no commented May 27, 2015

Uh oh!

haukepetersen commented May 27, 2015

Uh oh!

jfischer-no commented May 27, 2015

Uh oh!

jnohlgard commented May 27, 2015

Uh oh!

jfischer-no commented May 28, 2015

Uh oh!

jfischer-no commented May 29, 2015

Uh oh!

jnohlgard commented May 30, 2015

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jnohlgard commented May 30, 2015

Uh oh!

jnohlgard commented May 30, 2015

Uh oh!

jnohlgard May 30, 2015

Choose a reason for hiding this comment

Uh oh!

jfischer-no May 30, 2015

Choose a reason for hiding this comment

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jnohlgard commented May 30, 2015

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jnohlgard commented May 30, 2015

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jfischer-no commented May 30, 2015

Uh oh!

jnohlgard commented May 30, 2015

Uh oh!

jnohlgard commented May 31, 2015

Uh oh!

jnohlgard commented May 31, 2015

Uh oh!

jfischer-no commented May 31, 2015

Uh oh!

jfischer-no commented May 31, 2015

Uh oh!

jnohlgard commented May 31, 2015

Uh oh!

jnohlgard commented May 31, 2015

Uh oh!

jfischer-no commented May 31, 2015

Uh oh!

jonas-rem commented May 31, 2015

Uh oh!

jonas-rem commented May 31, 2015

Uh oh!

jnohlgard commented May 31, 2015

Uh oh!

jfischer-no commented May 31, 2015

Uh oh!

jnohlgard commented Jun 1, 2015

Uh oh!

jfischer-no commented Jun 2, 2015

Uh oh!

kaspar030 commented Jun 2, 2015

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!