[ECOS] ANNOUNCEMENT: www.ecosforge.net - a home for eCos packages and modules

[ECOS] ANNOUNCEMENT: www.ecosforge.net - a home for eCos packages and modules

[Øyvind Harboe]

We're a group of eCos developers who have seen the need for an additional
eCos repository for modules that do not belong in the official eCos
repository.

This is by no means an attempt to split the eCos community. We
encourage modules and patches that belong in the official eCos
repository to be posted to ecos-patches as before.

www.ecosforge.net is a place for work in progress that will eventually
make it into the official eCos repository or modules/packages that
simply do not meet the criteria for  the official eCos repository
somehow (work in progress, code up for adoption, pragmatic workarounds,
etc.).

www.ecosforge.net offers a subversion repository, a wiki for the
gathering information, a bug tracker and a mailing list for eCosForge
projects.

Take it for a spin!

To sign in, fill out this form: http://www.ecosforge.net/account.php

Sincerely,
Alexander Neundorf
Øyvind Harboe






-- 
Øyvind Harboe
http://www.zylin.com - eCos ARM & FPGA  developer kit

--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

[ECOS] Placing data at an absolute address

[Laurie Gellatly]

Hi All,
For our eCos application, I'd like to place the MAC address at a particular
address in flash
and have it loaded as part of storing Redboot into the flash.
At production time, a script would be run that finds that address section in
redboot.elf and
updates it before it is stored into the next unit.
Finding the section in the elf is not the hard bit.
I've tried adding a new section on the gcc command line and adding a
variable into the
C source which almost works except the section size is of length 0.
How can I make the section 6 bytes long or is there a better way to do this
whole thing?

Thanks			...Laurie:{)


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

[ECOS] Re: Placing data at an absolute address

[Grant Edwards]

On 2007-04-01, Laurie Gellatly <laurie.gellatly@netic.com> wrote:

> For our eCos application, I'd like to place the MAC address at
> a particular address in flash and have it loaded as part of
> storing Redboot into the flash. At production time, a script
> would be run that finds that address section in redboot.elf
> and updates it before it is stored into the next unit. Finding
> the section in the elf is not the hard bit. I've tried adding
> a new section on the gcc command line and adding a variable
> into the C source which almost works except the section size
> is of length 0. How can I make the section 6 bytes long or is
> there a better way to do this whole thing?

You can use objcopy to add a new section from the command line.

-- 
Grant Edwards                   grante             Yow!  Was my SOY LOAF left
                                  at               out in th'RAIN? It tastes
                               visi.com            REAL GOOD!!


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] Placing data at an absolute address

[Gary Thomas]

Laurie Gellatly wrote:
> Hi All,
> For our eCos application, I'd like to place the MAC address at a particular
> address in flash
> and have it loaded as part of storing Redboot into the flash.
> At production time, a script would be run that finds that address section in
> redboot.elf and
> updates it before it is stored into the next unit.
> Finding the section in the elf is not the hard bit.
> I've tried adding a new section on the gcc command line and adding a
> variable into the
> C source which almost works except the section size is of length 0.
> How can I make the section 6 bytes long or is there a better way to do this
> whole thing?

The best way would be to assign this data to a particular
section of its own, using "attribute".  Then in the linker
file, put that section at the desired physical address.

There was some discussion of this many years ago.  Some
trawling through the archives might find more clues.

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

[ECOS] Re: Placing data at an absolute address

[Grant Edwards]

On 2007-04-01, Gary Thomas <gary@mlbassoc.com> wrote:
> Laurie Gellatly wrote:

>> For our eCos application, I'd like to place the MAC address at
>> a particular address in flash and have it loaded as part of
>> storing Redboot into the flash. At production time, a script
>> would be run that finds that address section in redboot.elf
>> and updates it before it is stored into the next unit. Finding
>> the section in the elf is not the hard bit. I've tried adding
>> a new section on the gcc command line and adding a variable
>> into the C source which almost works except the section size
>> is of length 0. How can I make the section 6 bytes long or is
>> there a better way to do this whole thing?
>
> The best way would be to assign this data to a particular
> section of its own, using "attribute".  Then in the linker
> file, put that section at the desired physical address.

The question is then how to alter the contents of that section
during production in order to uniquely set the MAC address of
each unit.  AFAICT, you'd have to use objcopy to alter the .elf
file each time before you convert to hex/binary and download.

Personally, I think putting the MAC addres into the program
itself before download is way the wrong way to go.  It makes
updating firmware way too difficult.

Just reserve an "external" spot in flash memory that's not
written to when the program is downloaded.  Then download the
MAC address in a separate operation.

-- 
Grant Edwards                   grante             Yow!  PEGGY FLEMMING is
                                  at               stealing BASKET BALLS to
                               visi.com            feed the babies in VERMONT.


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

RE: [ECOS] Re: Placing data at an absolute address

[Laurie Gellatly]

Hi Grant,
objcopy - looked at that and yep, that will do just fine.
The MAC address is not in the code but does live in the same Flash page as
the code so it will be
convenient to be programmed at the same time.
objcopy will allow me to place the MAC into the included file after its been
update.

Excellent thanks.			...Laurie:{)

-----Original Message-----
From: ecos-discuss-owner@ecos.sourceware.org
[mailto:ecos-discuss-owner@ecos.sourceware.org]On Behalf Of Grant
Edwards
Sent: Monday, 2 April 2007 4:52 AM
To: ecos-discuss@sources.redhat.com
Subject: [ECOS] Re: Placing data at an absolute address


On 2007-04-01, Gary Thomas <gary@mlbassoc.com> wrote:
> Laurie Gellatly wrote:

>> For our eCos application, I'd like to place the MAC address at
>> a particular address in flash and have it loaded as part of
>> storing Redboot into the flash. At production time, a script
>> would be run that finds that address section in redboot.elf
>> and updates it before it is stored into the next unit. Finding
>> the section in the elf is not the hard bit. I've tried adding
>> a new section on the gcc command line and adding a variable
>> into the C source which almost works except the section size
>> is of length 0. How can I make the section 6 bytes long or is
>> there a better way to do this whole thing?
>
> The best way would be to assign this data to a particular
> section of its own, using "attribute".  Then in the linker
> file, put that section at the desired physical address.

The question is then how to alter the contents of that section
during production in order to uniquely set the MAC address of
each unit.  AFAICT, you'd have to use objcopy to alter the .elf
file each time before you convert to hex/binary and download.

Personally, I think putting the MAC addres into the program
itself before download is way the wrong way to go.  It makes
updating firmware way too difficult.

Just reserve an "external" spot in flash memory that's not
written to when the program is downloaded.  Then download the
MAC address in a separate operation.

--
Grant Edwards                   grante             Yow!  PEGGY FLEMMING is
                                  at               stealing BASKET BALLS to
                               visi.com            feed the babies in
VERMONT.


--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

RE: [ECOS] Placing data at an absolute address

[Laurie Gellatly]

Hi Gary,
Thanks for the response.
I did try using the "attribute" method but the allocated size for the
segment was always wrong (0).

I looked in the archives but without any success.

			...Laurie:{)

-----Original Message-----
From: Gary Thomas [mailto:gary@mlbassoc.com]
Sent: Monday, 2 April 2007 3:55 AM
To: Laurie Gellatly
Cc: Øyvind Harboe; eCos Discussion
Subject: Re: [ECOS] Placing data at an absolute address


Laurie Gellatly wrote:
> Hi All,
> For our eCos application, I'd like to place the MAC address at a
particular
> address in flash
> and have it loaded as part of storing Redboot into the flash.
> At production time, a script would be run that finds that address section
in
> redboot.elf and
> updates it before it is stored into the next unit.
> Finding the section in the elf is not the hard bit.
> I've tried adding a new section on the gcc command line and adding a
> variable into the
> C source which almost works except the section size is of length 0.
> How can I make the section 6 bytes long or is there a better way to do
this
> whole thing?

The best way would be to assign this data to a particular
section of its own, using "attribute".  Then in the linker
file, put that section at the desired physical address.

There was some discussion of this many years ago.  Some
trawling through the archives might find more clues.

--
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

[ECOS] Dropped characters

[Laurie Gellatly]

Hi all,
I'm having troubles with the UART on our 2212 board.
Its running at 115200 (talking to a 2103) but although the characters are on
the wire OK
some are being dropped.
My code checks for characters being available in eCos and when 'sufficient'
have arrived
they are emptied from eCos's buffer (which is MUCH larger).
I have not been able to isolate any pattern as to why they are being
dropped.
Does this sound familiar to anyone?

Thanks			...Laurie:{)


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] Dropped characters

[Sergei Gavrikov]

В Птн, 01/06/2007 в 20:32 +1000, Laurie Gellatly пишет:
> Hi all,
> I'm having troubles with the UART on our 2212 board.
> Its running at 115200 (talking to a 2103) but although the characters are on
> the wire OK
> some are being dropped.
> My code checks for characters being available in eCos and when 'sufficient'
> have arrived
> they are emptied from eCos's buffer (which is MUCH larger).
> I have not been able to isolate any pattern as to why they are being
> dropped.
> Does this sound familiar to anyone?

When I did try up eCos PPP between PC and my target (LPC2294) using a
serial connection, I couldn't achive the desired results for 115,2K
baudrate. But when I turned on the XON/XOFF flow control for eCos serial
driver, that became to work just fine. Do you use any flow control on
your system?

-- Sergei

> 
> Thanks			...Laurie:{)
> 
> 


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

RE: [ECOS] Dropped characters

[Laurie Gellatly]

Hi Sergei,
We've ruled that out.
We tried a several character delay between each character on the transmit side
and still the characters are being dropped.

			...Laurie:{)


-----Original Message-----
From: Sergei Gavrikov [mailto:w3sg@SoftHome.net]
Sent: Friday, 1 June 2007 8:55 PM
To: Laurie Gellatly
Cc: eCos Discussion
Subject: Re: [ECOS] Dropped characters


В Птн, 01/06/2007 в 20:32 +1000, Laurie Gellatly пишет:
> Hi all,
> I'm having troubles with the UART on our 2212 board.
> Its running at 115200 (talking to a 2103) but although the characters are on
> the wire OK
> some are being dropped.
> My code checks for characters being available in eCos and when 'sufficient'
> have arrived
> they are emptied from eCos's buffer (which is MUCH larger).
> I have not been able to isolate any pattern as to why they are being
> dropped.
> Does this sound familiar to anyone?

When I did try up eCos PPP between PC and my target (LPC2294) using a
serial connection, I couldn't achive the desired results for 115,2K
baudrate. But when I turned on the XON/XOFF flow control for eCos serial
driver, that became to work just fine. Do you use any flow control on
your system?

-- Sergei

> 
> Thanks			...Laurie:{)
> 
> 


--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

RE: [ECOS] Dropped characters

[Laurie Gellatly]

An update.
I've found that by turning off FIFO the problem disappears.

Does anyone know of a FIFO problem with the LPC2212 and eCos?
More importantly, anyone know of a fix?

Thanks			...Laurie:{)

-----Original Message-----
From: ecos-discuss-owner@ecos.sourceware.org
[mailto:ecos-discuss-owner@ecos.sourceware.org]On Behalf Of Laurie
Gellatly
Sent: Friday, 1 June 2007 9:57 PM
To: Sergei Gavrikov
Cc: eCos Discussion
Subject: RE: [ECOS] Dropped characters


Hi Sergei,
We've ruled that out.
We tried a several character delay between each character on the transmit side
and still the characters are being dropped.

			...Laurie:{)


-----Original Message-----
From: Sergei Gavrikov [mailto:w3sg@SoftHome.net]
Sent: Friday, 1 June 2007 8:55 PM
To: Laurie Gellatly
Cc: eCos Discussion
Subject: Re: [ECOS] Dropped characters


В Птн, 01/06/2007 в 20:32 +1000, Laurie Gellatly пишет:
> Hi all,
> I'm having troubles with the UART on our 2212 board.
> Its running at 115200 (talking to a 2103) but although the characters are on
> the wire OK
> some are being dropped.
> My code checks for characters being available in eCos and when 'sufficient'
> have arrived
> they are emptied from eCos's buffer (which is MUCH larger).
> I have not been able to isolate any pattern as to why they are being
> dropped.
> Does this sound familiar to anyone?

When I did try up eCos PPP between PC and my target (LPC2294) using a
serial connection, I couldn't achive the desired results for 115,2K
baudrate. But when I turned on the XON/XOFF flow control for eCos serial
driver, that became to work just fine. Do you use any flow control on
your system?

-- Sergei

> 
> Thanks			...Laurie:{)
> 
> 


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss



--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

[ECOS] Eclipse unable to debug threads

[Laurie Gellatly]

Hi again,
I've been using eclipse 3.1.2 to develop my eCos app.
Just recently its stopped allowing me to debug threads.
I can start the app, see it stop at the first line but after that
although I can set a breakpoint in another thread Eclipse won't
stop in that thread and let me debug it.
I've looked at all the configs I can think of and nothing
seems wrongly configured.
I'm seriously considering re-installing eclipse, but before
I take the sledgehammer approach (and probably break something
else in the process) I thought I'd ask here and then try
an eclipse forum somewhere I guess.

Thanks			...Laurie:{)


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss