GPL debugger for eZ430-RF2500 on Linux/BSD

[Daniel Beer-3]

I've just finished writing a tool for Linux/BSD which can be used to
program and debug the MSP430 in the eZ430-RF2500 from Texas
Instruments.

Source code (GPLv2) and some documentation available here:

    http://homepages.xnet.co.nz/~dlbeer/

Bug reports or any other comments would be appreciated.

Apologies for the off-topic message, but I figured that this would be
of interest to a reasonable number of mspgcc users anyway...

Regards,
Daniel

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[John Porubek]

On Wed, Oct 14, 2009 at 11:36 PM, Daniel Beer <dlbeer@...> wrote:

> I've just finished writing a tool for Linux/BSD which can be used to
> program and debug the MSP430 in the eZ430-RF2500 from Texas
> Instruments.
>
> Source code (GPLv2) and some documentation available here:
>
>    http://homepages.xnet.co.nz/~dlbeer/
>
> Bug reports or any other comments would be appreciated.
>
> Apologies for the off-topic message, but I figured that this would be
> of interest to a reasonable number of mspgcc users anyway...
>
> Regards,
> Daniel

Hi Daniel,

Thanks for making your tool for the eZ430-RF2500 available. I haven't
compiled and run the program yet, but I have a question about
something you wrote in the documentation. Under "USB interface" it
says:

"This device has a product:vendor code of 0451:f432 and provides two
USB interfaces. One is a CDC-ACM class device (this is the interface
to the application UART). This interface is supported by Linux
2.6.31.1, and appears as /dev/ttyACMxx."

My question is - does this apply after compiling and running your
program or does it apply on a stock Linux system? If the latter, what
version of Linux are you running? I'm running Ubuntu 9.04 and it
doesn't seem to be the case. I haven't installed the "libusb-dev"
package yet, either.

I apologize for my ignorance, but I would appreciate whatever you can
do to enlighten me!

Thanks,

--John

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[Daniel Beer-3]

On Thu, Oct 29, 2009 at 05:47:21PM -0400, John Porubek wrote:

> On Wed, Oct 14, 2009 at 11:36 PM, Daniel Beer <dlbeer@...> wrote:
> > I've just finished writing a tool for Linux/BSD which can be used to
> > program and debug the MSP430 in the eZ430-RF2500 from Texas
> > Instruments.
> >
> > Source code (GPLv2) and some documentation available here:
> >
> >    http://homepages.xnet.co.nz/~dlbeer/
> >
> > Bug reports or any other comments would be appreciated.
> >
> > Apologies for the off-topic message, but I figured that this would be
> > of interest to a reasonable number of mspgcc users anyway...
> >
> > Regards,
> > Daniel
>
> Hi Daniel,
>
> Thanks for making your tool for the eZ430-RF2500 available. I haven't
> compiled and run the program yet, but I have a question about
> something you wrote in the documentation. Under "USB interface" it
> says:
>
> "This device has a product:vendor code of 0451:f432 and provides two
> USB interfaces. One is a CDC-ACM class device (this is the interface
> to the application UART). This interface is supported by Linux
> 2.6.31.1, and appears as /dev/ttyACMxx."
>
> My question is - does this apply after compiling and running your
> program or does it apply on a stock Linux system? If the latter, what
> version of Linux are you running? I'm running Ubuntu 9.04 and it
> doesn't seem to be the case. I haven't installed the "libusb-dev"
> package yet, either.
>
> I apologize for my ignorance, but I would appreciate whatever you can
> do to enlighten me!

It applies to a stock Linux system. I've tried this with kernel version
2.6.31.1 (but it may work on earlier kernel versions -- I haven't
tested this).

The program I've written is independent of this. You can use the debug
interface even if you don't have the application UART working, and vice
versa. You only need libusb-dev to compile MSPDebug -- it's not
required for the application UART.

- Daniel

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[John Porubek]

On Thu, Oct 29, 2009 at 7:32 PM, Daniel Beer <dlbeer@...> wrote:

> It applies to a stock Linux system. I've tried this with kernel version
> 2.6.31.1 (but it may work on earlier kernel versions -- I haven't
> tested this).
>
> The program I've written is independent of this. You can use the debug
> interface even if you don't have the application UART working, and vice
> versa. You only need libusb-dev to compile MSPDebug -- it's not
> required for the application UART.
>
> - Daniel

Thanks for the info here and in the "MSP-FET430UIF on linux" thread.
My fully updated Ubuntu 9.04 is running with kernel version 2.6.28-16.
There is no /dev/ttyACMn with the eZ430-RF2500 plugged in.

The just officially released Ubuntu 9.10 includes kernel version
2.6.31 (it took some doing to ferret out this information). I guess
this is ample justification to upgrade soon!

BTW, I just discovered that the eZ430-RF2500 Windows application UART
software apparently doesn't support XON/XOFF flow control. However, if
I connect the RF2500T module's UART lines through an FTDI
serial-to-USB cable, XON/XOFF works under both Windows and Linux. I'm
testing this using Chris Liechti's "miniterm" example app based on his
"pyserial" Python serial port extension. It will be interesting to see
if XON/XOFF works directly with the eZ430-RF2500 USB emulator under
Linux once I upgrade.

--John

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[John Porubek]

On Fri, Oct 30, 2009 at 11:29 AM, John Porubek <jporubek@...> wrote:
> BTW, I just discovered that the eZ430-RF2500 Windows application UART
> software apparently doesn't support XON/XOFF flow control. However, if
> I connect the RF2500T module's UART lines through an FTDI
> serial-to-USB cable, XON/XOFF works under both Windows and Linux. I'm
> testing this using Chris Liechti's "miniterm" example app based on his
> "pyserial" Python serial port extension. It will be interesting to see
> if XON/XOFF works directly with the eZ430-RF2500 USB emulator under
> Linux once I upgrade.

A quick update. I upgraded to Ubuntu 9.10 which includes kernel
version 2.6.31 and now the eZ430-RF2500 emulator does indeed show up
as /dev/ttyACMxx, as you described in your documentation. Further,
XON/XOFF works properly also.

Thus, I find myself doing more of my development work under Linux.
>From time to time what I'm working on hangs the processor. Removing
the emulator from the USB port is inconvenient, and, worse, it doesn't
always reliably get me back to the "fresh from reset" state. I had
thought about installing the IAR Kickstart under Wine, but that seemed
like overkill just to reset the processor and was certainly less than
elegant. Then you announced your MSPDebug tool, and the timing and
suitability to to my needs couldn't have been better! This is a
long-winded way of saying, once again, thank you for making it
available.

I compiled your source code with no real problems. BTW, your coding
style and commenting allowed me to to easily follow your code and
figure out what it was doing. It is proving to be an invaluable tool.
And now that it supports the eZ430-F2013 and FET430UIF devices also,
it's even better.

I have only one real problem. I have to run the program by typing
"sudo ./mspdebug". I tried your suggestion for remounting usbfs,
putting sudo in front of the umount and mount lines, and substituting
my group id for "usb-gid". Here's the error message I get when I run
mspdebug without sudo:

Trying to open interface 1 on 002
usbtr_open_interface: warning: can't detach kernel driver: Operation
not permitted
usbtr_open_interface: can't claim interface: Operation not permitted
usbtr_open: no devices could be found

In your docs you state: "Replace usb-gid with the GID of a group that
should have permission to use the bus." There's the crux - how do I
determine which group has permission to use the bus? I fear I'm
exposing my ignorance here.

Here's what I get when I run "id username" (username substituted) :

uid=1000(username) gid=1000(username)
groups=1000(username),4(adm),20(dialout),24(cdrom),46(plugdev),104(fuse),106(lpadmin),121(admin),122(sambashare)

Anyways, running mspdebug with sudo certainly isn't a huge
inconvenience. It just doesn't feel quite right and I'd like to learn
something along the way. Any light you could shed would be greatly
appreciated.

Thanks again,

John

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[Daniel Beer-3]

On Tue, Nov 17, 2009 at 06:56:30PM -0500, John Porubek wrote:

> A quick update. I upgraded to Ubuntu 9.10 which includes kernel
> version 2.6.31 and now the eZ430-RF2500 emulator does indeed show up
> as /dev/ttyACMxx, as you described in your documentation. Further,
> XON/XOFF works properly also.
>
> Thus, I find myself doing more of my development work under Linux.
> >From time to time what I'm working on hangs the processor. Removing
> the emulator from the USB port is inconvenient, and, worse, it doesn't
> always reliably get me back to the "fresh from reset" state. I had
> thought about installing the IAR Kickstart under Wine, but that seemed
> like overkill just to reset the processor and was certainly less than
> elegant. Then you announced your MSPDebug tool, and the timing and
> suitability to to my needs couldn't have been better! This is a
> long-winded way of saying, once again, thank you for making it
> available.
>
> I compiled your source code with no real problems. BTW, your coding
> style and commenting allowed me to to easily follow your code and
> figure out what it was doing. It is proving to be an invaluable tool.
> And now that it supports the eZ430-F2013 and FET430UIF devices also,
> it's even better.

Glad to have helped :).

> I have only one real problem. I have to run the program by typing
> "sudo ./mspdebug". I tried your suggestion for remounting usbfs,
> putting sudo in front of the umount and mount lines, and substituting
> my group id for "usb-gid". Here's the error message I get when I run
> mspdebug without sudo:
>
> Trying to open interface 1 on 002
> usbtr_open_interface: warning: can't detach kernel driver: Operation
> not permitted
> usbtr_open_interface: can't claim interface: Operation not permitted
> usbtr_open: no devices could be found
>
> In your docs you state: "Replace usb-gid with the GID of a group that
> should have permission to use the bus." There's the crux - how do I
> determine which group has permission to use the bus? I fear I'm
> exposing my ignorance here.
>
> Here's what I get when I run "id username" (username substituted) :
>
> uid=1000(username) gid=1000(username)
> groups=1000(username),4(adm),20(dialout),24(cdrom),46(plugdev),104(fuse),106(lpadmin),121(admin),122(sambashare)
>
> Anyways, running mspdebug with sudo certainly isn't a huge
> inconvenience. It just doesn't feel quite right and I'd like to learn
> something along the way. Any light you could shed would be greatly
> appreciated.

Probably easiest just to use GID 1000, which you're already in. Try
this (as root):

    umount /proc/bus/usb
    mount -t usbfs none /proc/bus/usb -o devgid=1000,devmode=0664

If you want to tidy things up later, you might like to make a "usb"
group, add yourself to it and have usbfs mounted with the GID of
the "usb" group.

- Daniel

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[John Porubek]

On Tue, Nov 17, 2009 at 7:26 PM, Daniel Beer <dlbeer@...> wrote:

> Probably easiest just to use GID 1000, which you're already in. Try
> this (as root):
>
>    umount /proc/bus/usb
>    mount -t usbfs none /proc/bus/usb -o devgid=1000,devmode=0664
>
> If you want to tidy things up later, you might like to make a "usb"
> group, add yourself to it and have usbfs mounted with the GID of
> the "usb" group.
>

As it turns out, I was using the right GID but the wrong mount point.
Under Ubuntu 9.10, instead of "/proc/bus/usb", I had to use
"/dev/bus/usb". Also, I had to add my UID to the mount options
(devuid=1000). I haven't made a "usb" group yet, but I think I will.

Now if I only could make this stick across reboots. As I understand
it, one would normally use fstab to apply mounts at bootup. However,
I'm not sure if that applies here, or, if it does, how to do it. Any
ideas?

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[Daniel Beer-3]

On Mon, Nov 23, 2009 at 04:18:19PM -0500, John Porubek wrote:

> On Tue, Nov 17, 2009 at 7:26 PM, Daniel Beer <dlbeer@...> wrote:
> > Probably easiest just to use GID 1000, which you're already in. Try
> > this (as root):
> >
> >    umount /proc/bus/usb
> >    mount -t usbfs none /proc/bus/usb -o devgid=1000,devmode=0664
> >
> > If you want to tidy things up later, you might like to make a "usb"
> > group, add yourself to it and have usbfs mounted with the GID of
> > the "usb" group.
> >
>
> As it turns out, I was using the right GID but the wrong mount point.
> Under Ubuntu 9.10, instead of "/proc/bus/usb", I had to use
> "/dev/bus/usb". Also, I had to add my UID to the mount options
> (devuid=1000). I haven't made a "usb" group yet, but I think I will.
>
> Now if I only could make this stick across reboots. As I understand
> it, one would normally use fstab to apply mounts at bootup. However,
> I'm not sure if that applies here, or, if it does, how to do it. Any
> ideas?

If the init scripts consult /etc/fstab when mounting procfs, you
could add a line like this:

none /dev/bus/usb usbfs devuid=1000 0 0

I haven't tried this, however.

- Daniel

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users

[John Porubek]

On Mon, Nov 23, 2009 at 7:45 PM, Daniel Beer <dlbeer@...> wrote:

> On Mon, Nov 23, 2009 at 04:18:19PM -0500, John Porubek wrote:
>> On Tue, Nov 17, 2009 at 7:26 PM, Daniel Beer <dlbeer@...> wrote:
>> > Probably easiest just to use GID 1000, which you're already in. Try
>> > this (as root):
>> >
>> >    umount /proc/bus/usb
>> >    mount -t usbfs none /proc/bus/usb -o devgid=1000,devmode=0664
>> >
>> > If you want to tidy things up later, you might like to make a "usb"
>> > group, add yourself to it and have usbfs mounted with the GID of
>> > the "usb" group.
>> >
>>
>> As it turns out, I was using the right GID but the wrong mount point.
>> Under Ubuntu 9.10, instead of "/proc/bus/usb", I had to use
>> "/dev/bus/usb". Also, I had to add my UID to the mount options
>> (devuid=1000). I haven't made a "usb" group yet, but I think I will.
>>
>> Now if I only could make this stick across reboots. As I understand
>> it, one would normally use fstab to apply mounts at bootup. However,
>> I'm not sure if that applies here, or, if it does, how to do it. Any
>> ideas?
>
> If the init scripts consult /etc/fstab when mounting procfs, you
> could add a line like this:
>
> none    /dev/bus/usb    usbfs   devuid=1000     0       0
>
> I haven't tried this, however.
>
> - Daniel

It works! Thank you so much for your help and for creating a great
tool and making it available. Now to compile the latest version and
try it out with an eZ430-F2013 and FET430UIF. I may be back. ;^)

--John

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@...
https://lists.sourceforge.net/lists/listinfo/mspgcc-users