On Wed, May 18, 2011 at 3:06 AM, Derek Baker <
derek.ba...@claritas-solutions.com> wrote:

> Hi Eric,
>
>
>
> Thanks for all that and sorry yes I’m using the em-cc430f5137-900
> development board from TI, I need the driver quite urgent, to a point where
> I’ve started to develop a driver myself but I’m now going to stop if there
> is a working/semi working driver in the trunk you’ve listed below, I got my
> source code from the OSIAN git rep and it doesn’t have the MSP430X in it, is
> it fare to say the google rep is more up to date in terms of development
> work?.
>

It is fair to say it is confusing.   One of the things I'm trying to fix.

The SVN trunk (google rep) has a directory tos/chips/msp430X which reflects
the state of the Zolertia Z1 port to the x2 (msp430f2617).  It is
essentially their code from about a year ago or so.   I'm working with
Antonio (from Zolertia) on getting this cleaned up.    We are eliminating
duplicates and consolidating the code.

I am also bringing over Peter Bigots x5 work and it is forming the initial
cut for the x5 part of the consolidated msp430 core support for the x5.
When that is done I will add support for the msp430f5438 and its flavors.
That is the processor that I am using on the next interation of my next mote
(the mm5).

Now that said....    Peter's osian work didn't include an i2c driver.   He
didn't need it and had nothing to test it on.   The i2c driver from the Z1
works from what I understand.   SVN trunk, tos/chips/msp430X/usci

But it would be more helpful if you looked at/used the current Z1 i2c driver
from the msp430 integration branch.   Its on a git repo.  Do you git?

mkdir msp430
cd msp430
git clone git://hinrg.cs.jhu.edu/git/cire/msp430.git tinyos-2.x
cd tinyos-2.x
git checkout -t origin/mm-z1
cd tos/chips/msp430/x2xxx/usci

That will put you on the mm-z1 integration branch.   That has the merge
between the existing x1 code and my mm4 (2618 based) and the Z1 (2617) based
code bases.  tos/chips/msp430/x2xxx/usci has the current merged drivers for
the usci on the x2 parts.   There are enough differences between how the
usci interfaces on the x2 vs. the x5 parts that for the time being the x2
and x5 usci drivers have their own homes.

I also know that the shimmer/span i2c drivers work and are in production
systems.   So you might want to take a look in the shimmer/span code
directories.   Try down in tos/platforms/shimmer...   or
tos/platforms/span...    and see what you can find that touches i2c.

I do know from Antonio who is actively testing the integration branch on the
Z1 motes he has that the i2c driver there works.

But you'll run into another problem.    We found that using the mspgcc4 tool
chain (4.4.5, 20110312) that the i2c driver broke.   When Antonio reverted
back to the 3.2.3 based Z1 toolchain it works fine.   Now you can't use the
3.2.3 toolchain because it doesn't support your processor.   So there is
another gotcha there.

Antonio is going to see if the i2c driver works with the current work Peter
is doing which is the uniarch toolchain.

You should add yourself to the tinyos-msp430 mailing list.


>
> Regards
>
>
>
> Derek
>
>
>
> *From:* Eric Decker [mailto:cire...@gmail.com]
> *Sent:* 18 May 2011 10:52
> *To:* Derek Baker
> *Cc:* tinyos-help@millennium.berkeley.edu
> *Subject:* Re: [Tinyos-help] Uart v Usci
>
>
>
> Hi Derek,
>
>
>
> I'm currently working on integrating the msp430 core code for the three
> major msp430 families (the x1, x2, and x5).
>
>
>
> You are using an x5 part.   I'm guessing the cc430f5137 because I beleive
> you are basing off of the osian/surf board code.
>
>
>
> When you post question onto tinyos-help (or for that matter anywhere) it
> would be helpful if you included the specifics of exactly what you are
> working with so we don't have to guess.
>
>
>
> That also applies to what flavor of em430 board you are using.  TI makes
> many different eval boards for their processors.   For example, I have an
> eval board for the 5438 that is significantly different from the eval board
> that the surf folks called the em430.   So calling it the em430 doesn't
> convey alot of information.
>
>
>
> On Wed, May 18, 2011 at 2:14 AM, Derek Baker <
> derek.ba...@claritas-solutions.com> wrote:
>
> Hi, can anybody explain to me the difference between the msp430/usci and
> msp430/uart drivers,
>
>
>
> The devil is in the details.    I think you are talking about
> tos/chips/msp430/usart vs. tos/chips/msp430/usci.
>
>
>
> The usart driver (tos/chips/msp430/usart) is the driver for the x1 family
> peripheral (the USART).   On the x2 and x5 family parts there is a different
> peripheral called the USCI.  It is a different beast and has a different
> driver.   There are basically differnt flavors of what these modules do.
> The USART on the x1 parts only has I2C on port 0 and this is reflected in
> the code.
>
>
>
> The USCI situation is also convoluted.   The X2 implementation of USCI vs.
> the x5 implementation has some significant warts.  In particular how
> interrupts are dealt with and where the register are mapped and how they are
> accessed.
>
>
>
>
>
> I noticed the /msp430/uart has i2c support. I’m using the em430 mote which
> is configured to use the usci but I need the i2c part has anybody developed
> the i2c part of the /msp/usci?
>
>
>
> Peter who did the initial support for the tinyos cc430f5137 port didn't add
> the i2c functionality because he didn't have any h/w that used it.  The
> parts will support it.    But the driver hasn't been integrated yet.
>
>
>
> i'm pretty sure that the Z1 folks have a working i2c driver.    Keep in
> mind that all of this code is in flux.   The code that is currently in the
> trunk for the msp430X (tos/chips/msp430X) is being heavily reworked.   That
> said there is a version of the usci drivers in there that worked for the Z1
> folks.   But it is very ugly and unfinished.   A more modern driver is being
> worked on in the msp430 integration branch.   That is also where the merged
> x5 code is coming in.
>
>
>
> How soon do you need an i2c driver?
>
>
>
>
>
> Regards
>
>
>
> Derek
>
>
>
>
> ------------------------------
>
> This email message has been delivered safely and archived online by
> Mimecast.
> A true SaaS solution, Mimecast provides the security, continuity and
> archiving for millions of emails, across thousands of customers every day.
> For more information please visit http://www.mimecast.co.uk
> ------------------------------
>
>
> _______________________________________________
> Tinyos-help mailing list
> Tinyos-help@millennium.berkeley.edu
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>
>
>
>
> --
> Eric B. Decker
> Senior (over 50 :-) Researcher
>
>
>
>  ------------------------------
> This email message has been delivered safely and archived online by
> Mimecast.
> A true SaaS solution, Mimecast provides the security, continuity and
> archiving for millions of emails, across thousands of customers every day.
> For more information please visit http://www.mimecast.co.uk
> ------------------------------
>



-- 
Eric B. Decker
Senior (over 50 :-) Researcher
_______________________________________________
Tinyos-help mailing list
Tinyos-help@millennium.berkeley.edu
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to