drivers: add cc2420 support (based on #5314) by kYc0o · Pull Request #5591 · RIOT-OS/RIOT

kYc0o · 2016-07-01T15:08:02Z

This is a reworked version of #5314 with code reorganisation and netdev2 compliant.

The code was tested in a z1 board, please feel free to test it on telosb and wsn430-v1_4.

Communication works for L2 messages with short and long addresses, but IPv6 messages are discarded, so that's why it's WIP.

One major bug is that when dev->netdev.proto = GNRC_NETTYPE_SIXLOWPAN; is enabled random messages are received from random kernel PIDs.

kYc0o · 2016-07-01T16:25:03Z

IPv6 messages are being received now, so no more WIP. However the spurious messages still coming, but apparently they don't bother other threads.

OlegHahm · 2016-07-05T10:09:08Z

I tried with a Z1 and a SAMR21-Xpro. Link layer communication (unicast and broadcast) seems to work fine in both directions. 👍

However, trying to ping the Z1 running gnrc_minimal from SAMR21 with gnrc_networking doesn't work. Also, the L2 address doesn't seem unique, but probably the MSP430 doesn't provide CPUID, right?

kYc0o · 2016-07-05T10:11:20Z

After testing on wsn430-1_4 board I found that the UART0 and SPI share the same buffer, thus it conflicts while initialising the driver. An issue #5599 is open for this

kYc0o · 2016-07-05T10:15:17Z

In order to ping devices using gnrc_minimal you need to disable IPHC with something like ifconfig 7 -iphc.

For the address yes you're right, MSP430 does not provide CPUID.

OlegHahm · 2016-07-05T10:21:47Z

Ah, I should have known this myself. :)

Confirmed working.

kYc0o · 2016-07-05T10:26:47Z

and ACK :P ?

OlegHahm · 2016-07-05T10:29:26Z

Lemmy quickly look at the code.

OlegHahm · 2016-07-05T10:31:47Z

drivers/Makefile.dep

 endif

+ifneq (,$(filter cc2420,$(USEMODULE)))
+  USEMODULE += cc2420


Why is this required?

Oh yeah, it was also used with at86rf2xx so I added it, but you're right, it's not required. Should I remove it?

The reason for the at86rf2xx USEMODULE check is that the driver is checking
for a wildcard and then adding the generic driver if it matches. The
at86rf2xx driver supports the rf231, rf233, rf212b, (possibly more), and is
written in a way that it relies on USEMODULE to tell it which hardware it
should expect.
Den 5 jul 2016 12:33 skrev "kYc0o" notifications@github.com:

In drivers/Makefile.dep
#5591 (comment):

@@ -32,6 +32,21 @@ ifneq (,$(filter cc110x,$(USEMODULE)))
endif
endif

+ifneq (,$(filter cc2420,$(USEMODULE)))

USEMODULE += cc2420

Oh yeah, it was also used with at86rf2xx so I added it, but you're right,
it's not required. Should I remove it?

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/RIOT-OS/RIOT/pull/5591/files/ffd03d77950ecaa2210bc2efb066a12ad9728cfd#r69541037,
or mute the thread
https://github.com/notifications/unsubscribe/AATYQl9hCQ81LrCAtqWHIs1iLu4Rowqzks5qSjL7gaJpZM4JDOnR
.

Yeah once I analysed I saw that, thanks for the explanation. Now USEMODULE += cc2420 it's removed.

OlegHahm · 2016-07-05T10:33:40Z

Shouldn't be ece96d7 and b2def4a be squashed?

OlegHahm · 2016-07-05T10:33:51Z

Apart from the minor comments: ACK.

kYc0o · 2016-07-05T10:36:29Z

Yes I can squash those commits, remove wsn430-v1_4 and add it when #5599 is solved.

kYc0o · 2016-07-05T10:48:54Z

Squashed and waiting for Murdock. I removed the USEMODULE += cc2420 can you check once more if everything works fine?

OlegHahm · 2016-07-05T11:13:23Z

Confirmed, still working. ACK holds.

kYc0o · 2016-07-05T11:56:20Z

Murdock failed because telosb wasn't in some memory insufficient lists for some tests. Fixed and Murdock restarted.

thomaseichinger · 2016-07-06T10:30:18Z

What was it complaining about?

kYc0o · 2016-07-06T10:31:53Z

Sorry, I restarted it too quickly... It was about the cortex_m3 family, there was "no more space left on device" while compiling some unit tests.

kYc0o · 2016-07-06T10:55:25Z

Same error again... @kaspar030 can you take a quick look into that? I'm afraid there's really no more space left on Murdock or something similar...

kaspar030 · 2016-07-06T15:18:31Z

hmm. murdock still has almost 16gig on that drive. I'll watch it a little.

kaspar030 · 2016-07-06T15:34:36Z

Yep, the ramdisk is getting full. Every package get's checked out ~40 times, so if more than four tests using packages get's built, there's no space left. I'll think of something.

kYc0o · 2016-07-06T15:36:10Z

Ups... sorry for break Murdock, if I can do something on it just let me know :)

kaspar030 · 2016-07-06T15:46:23Z

@kYc0o please help quicktesting #5610. :)

kYc0o · 2016-07-06T15:47:15Z

Great! doing it now!

kYc0o · 2016-07-06T16:15:50Z

Let's see if #5610 works ;)

kYc0o · 2016-07-06T16:36:48Z

Yeeeeeah!!! Murdock rules! Maybe it's needed to squash the doc fix?

OlegHahm · 2016-07-06T16:48:56Z

Yes, please squash and then merge as soon as Murdock confirms again.

kYc0o · 2016-07-06T18:39:06Z

And GO!

kYc0o added this to the Release 2016.07 milestone Jul 1, 2016

kYc0o assigned OlegHahm Jul 1, 2016

kYc0o force-pushed the cc2420_driver branch from 2f95e91 to ffd03d7 Compare July 1, 2016 15:20

kYc0o removed the State: WIP State: The PR is still work-in-progress and its code is not in its final presentable form yet label Jul 1, 2016

OlegHahm reviewed Jul 5, 2016
View reviewed changes

kYc0o force-pushed the cc2420_driver branch from ffd03d7 to 1475c7d Compare July 5, 2016 10:47

kYc0o added the CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR label Jul 5, 2016

emmanuelsearch removed the Discussion: RFC The issue/PR is used as a discussion starting point about the item of the issue/PR label Jul 5, 2016

kYc0o force-pushed the cc2420_driver branch 4 times, most recently from 3b2c136 to c89279d Compare July 5, 2016 17:36

kYc0o force-pushed the cc2420_driver branch from d9f715e to a3dd0b3 Compare July 6, 2016 12:38

kaspar030 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 Jul 6, 2016

haukepetersen added 5 commits July 6, 2016 18:14

drivers: added support for CC2420 radio

abc6b5c

sys/auto_init: added initialization for cc2420

8d3cb3a

boards/z1: add support for CC2420 radio

2047028

boards/telosb: added support for CC2420

dfe52cf

examples/default: add z1 and telosb for radio

32be958

kYc0o force-pushed the cc2420_driver branch from a3dd0b3 to 1b21626 Compare July 6, 2016 16:15

kYc0o added 4 commits July 6, 2016 20:23

drivers/cc2420: factorise/reorganise overall code

62d8b9a

drivers/include: add cc2420 doxygen doc

374f4a1

auto_init/netif/auto_init_cc2420: change variable names

124bddb

tests: add telosb to memory insufficient boards

0a70094

kYc0o force-pushed the cc2420_driver branch from 1b21626 to 0a70094 Compare July 6, 2016 18:26

kYc0o merged commit 2cbc8fb into RIOT-OS:master Jul 6, 2016

kYc0o mentioned this pull request Jul 7, 2016

drivers: added support for CC2420 [WIP] #5314

Closed

Conversation

kYc0o commented Jul 1, 2016

Uh oh!

kYc0o commented Jul 1, 2016

Uh oh!

OlegHahm commented Jul 5, 2016

Uh oh!

kYc0o commented Jul 5, 2016

Uh oh!

kYc0o commented Jul 5, 2016

Uh oh!

OlegHahm commented Jul 5, 2016

Uh oh!

kYc0o commented Jul 5, 2016

Uh oh!

OlegHahm commented Jul 5, 2016

Uh oh!

OlegHahm Jul 5, 2016

Choose a reason for hiding this comment

Uh oh!

kYc0o Jul 5, 2016

Choose a reason for hiding this comment

Uh oh!

jnohlgard Jul 5, 2016

Choose a reason for hiding this comment

Uh oh!

kYc0o Jul 5, 2016

Choose a reason for hiding this comment

Uh oh!

OlegHahm commented Jul 5, 2016

Uh oh!

OlegHahm commented Jul 5, 2016

Uh oh!

kYc0o commented Jul 5, 2016

Uh oh!

kYc0o commented Jul 5, 2016

Uh oh!

OlegHahm commented Jul 5, 2016

Uh oh!

kYc0o commented Jul 5, 2016

Uh oh!

thomaseichinger commented Jul 6, 2016 via email

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

kaspar030 commented Jul 6, 2016

Uh oh!

kaspar030 commented Jul 6, 2016

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

kaspar030 commented Jul 6, 2016

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

OlegHahm commented Jul 6, 2016

Uh oh!

kYc0o commented Jul 6, 2016

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants