Uwe Kindler | 8 Apr 2004 07:55
Picon

Re: eCos for Renesas EDOSK-2674

Hello Jonathan,

I have sent you the license agreement for EDOSK-2674 port. Please remove the EDOSK-2674 port from the "Work
in progress" issue - it is finished now and completely documented. You can integrate it into eCos
repository. A port for H8S/2357 is also available and a port for Renesas EDK/2638 including CAN drivers is
currently under development.

> If you think you're close to contributing it back to the eCos source base 
> we'd just need to get a copyright assignment sorted out as per 
> http://ecos.sourceware.org/assign.html so it might be nice to get that 
> formality out of the way as it can take time to arrange.
> 
> Thanks!
>
> Jifl
____________________________________________________________________
Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Sober.A-F!
Kostenfrei fuer FreeMail Nutzer. http://f.web.de/?mc=021158

Andrew Lunn | 8 Apr 2004 09:35
Picon

Re: eCos for Renesas EDOSK-2674

On Thu, Apr 08, 2004 at 07:55:54AM +0200, Uwe Kindler wrote:
> Hello Jonathan,

> I have sent you the license agreement for EDOSK-2674 port. Please
> remove the EDOSK-2674 port from the "Work in progress" issue - it is
> finished now and completely documented. You can integrate it into
> eCos repository.

Just to make sure i've got this correct, you are referring to the
patches from Yoshinori Sato on the 04th March? These patches were
committed on Monday. 

> A port for H8S/2357 is also available and a port
> for Renesas EDK/2638 including CAN drivers is currently under
> development.

Where can i find these? They don't seem to be in the ecos-patches
archive.

      Andrew

Uwe Kindler | 8 Apr 2004 11:01
Picon

Re: eCos for Renesas EDOSK-2674

Hello

> Just to make sure i've got this correct, you are referring to the
> patches from Yoshinori Sato on the 04th March? These patches were
> committed on Monday. 

No, I referring to the EDOSK-2674 port available here:

http://www.htwm.de/ukindler/ecos

I developed this port because when I started using eCos the EDOSK-2674 port of Yoshinori Sato
was very buggy and not well documented.  I also did not agree with making the EDOSK port part
of H8300 architecture.

The new EDOSK-2674 port is well documented, tested (suceeded the ecos tests) and based on
new H8S architecture.

Regards, Uwe 

_____________________________________________________________________
Der WEB.DE Virenschutz schuetzt Ihr Postfach vor dem Wurm Netsky.A-P!
Kostenfrei fuer alle FreeMail Nutzer. http://f.web.de/?mc=021157

Andrew Lunn | 8 Apr 2004 12:11
Picon

Re: eCos for Renesas EDOSK-2674

On Thu, Apr 08, 2004 at 11:01:55AM +0200, Uwe Kindler wrote:
> Hello
> 
> > Just to make sure i've got this correct, you are referring to the
> > patches from Yoshinori Sato on the 04th March? These patches were
> > committed on Monday. 
> 
> No, I referring to the EDOSK-2674 port available here:
> 
> http://www.htwm.de/ukindler/ecos

Ok. It better to send patches to the list. Anything else tends to get
lost, which is what happened here.

> I developed this port because when I started using eCos the
> EDOSK-2674 port of Yoshinori Sato was very buggy and not well
> documented.  I also did not agree with making the EDOSK port part of
> H8300 architecture.

Now we have a problem. We don't want two ports for the same device in
the tree. My suggestion would be for you to work with Yoshinori and
merge the two ports into one. Then post the patches to the
ecos-patches list. 

Please not you have different copyright assignments. Your is to
eCosCentric while Yoshinori is to RedHat. Make sure the copyright
notice on each file states the correct copyright holds.

       Andrew

(Continue reading)

kimai | 10 Apr 2004 05:50
Picon

This address was already removed

This mail account "kimai <at> laser5.co.jp" was already removed.
Please check your mail.

--
LASER5 Co.,Ltd.

Konstantin Azarov | 13 Apr 2004 23:00
Picon
Favicon

What to use instead of pipe?

Hello!

I'm currently writing a eCOS application and came upon the following
problem. In my application I have the menu, which must be accessible through
UART and Telnet. I'm going to handle the menu via standart files. The first
problem is that I've to write special filesystem to handle telnet sessions.
But this seems quite straitforward. The second problem is more interesting.
UART traffic must be filtered, that is, program must perform certain actions
as the reaction to certain input sequences (for example, if it sees "%%%" in
the UART input, it has, say, to turn on LED. At the same time, Telnet
traffic must not be filtered in this way. The solution, I wanted to
implement, was like this. I was going to create a "filter" thread, wich
would actually read data from UART and would open a pipe. Filtered traffic
would go through the pipe to the menu. The issue is that pipes are not
implemented in eCOS. Could you tell me, what can I use instead of pipe()? Or
may be it will be easier to implement something like pipe()?

By the way, it seems that implementing telnet with pipes will be much
better.

Thanks in advice,
Konstantin Azarov

Jani Monoses | 22 Apr 2004 14:40
Picon

nit

There are 5 subdirs in the tree which do not have a ChangeLog file

jani <at> cow ~/work/ecos $ find . -name current -a -type d |xargs -n1 -i ls -l {}/ChangeLog |grep piggy

ls: ./packages/hal/m68k/arch/current/ChangeLog: No such file or directory
ls: ./packages/hal/m68k/mcf52xx/var/current/ChangeLog: No such file or directory
ls: ./packages/hal/m68k/mcf52xx/mcf5272/proc/current/ChangeLog: No such file or directory
ls: ./packages/hal/m68k/mcf52xx/mcf5272/mcf5272c3/plf/current/ChangeLog: No such file or directory
ls: ./packages/devs/flash/powerpc/moab/current/ChangeLog: No such file or directory

No big deal but saw this while trying to run a script which automatically adds a changelog
entry to dirs affected by a search/replace which touches many CDLs.

Jani

Gary Thomas | 22 Apr 2004 14:57
Favicon

Re: nit

On Thu, 2004-04-22 at 06:40, Jani Monoses wrote:
> There are 5 subdirs in the tree which do not have a ChangeLog file
> 
> jani <at> cow ~/work/ecos $ find . -name current -a -type d |xargs -n1 -i ls -l {}/ChangeLog |grep piggy
> 
> ls: ./packages/hal/m68k/arch/current/ChangeLog: No such file or directory
> ls: ./packages/hal/m68k/mcf52xx/var/current/ChangeLog: No such file or directory
> ls: ./packages/hal/m68k/mcf52xx/mcf5272/proc/current/ChangeLog: No such file or directory
> ls: ./packages/hal/m68k/mcf52xx/mcf5272/mcf5272c3/plf/current/ChangeLog: No such file or directory
> ls: ./packages/devs/flash/powerpc/moab/current/ChangeLog: No such file or directory
> 
> No big deal but saw this while trying to run a script which automatically adds a changelog
> entry to dirs affected by a search/replace which touches many CDLs.

Added (template versions), thanks.

Note: the hal/m68k stuff is probably obsolete anyway if/when eCosCentric
contributes their ColdFire support.  The MOAB one is just one I missed.

--

-- 
Gary Thomas <gary <at> mlbassoc.com>
MLB Associates

Cheng Ping | 26 Apr 2004 11:13
Picon
Favicon

some questions about run-time image

Hello all,

I try to run ecos in a board with MPC8250, SMC1 for console, and 
FCC3/LXT971 for ethernet.
I take the rattler variant for the base, and change some codes for our 
configuration. Now the 
RedBoot seems OK, at least the console and ethernet can work.

But when I want to run the ecos run-time image with the simplest hello 
program, some
questions happened.

1. When CYGSEM_HAL_USE_ROM_MONITOR enabled, load the run-image from tftp 
then run.
There are console outputs. But the network can be accessed. Both DHCP and 
static configuration
are tried.

2. When CYGSEM_HAL_USE_ROM_MONITOR disabled, tftp load and run, the console 
just like
RedBoot> go
$O496E69742064657669636520272F6465762F73657231270A#71
then stalled. The network can't be accessed either.

How to deal with it?

Thanks and best regards,

Rex Cheng

(Continue reading)


Gmane