This is the mail archive of the ecos-devel@sourceware.org mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: eCos port to Freescale MAC7100


Andrew Lunn wrote:
On Tue, Apr 25, 2006 at 03:12:52PM +0200, Ilija Koco wrote:
Dear colleagues.
We're just concluding eCos port to Freescale's ARM 7 based MAC7100 family.

Port consists of: MAC7100 variant, a platform for our board (MACE 1) and serial device drivers for ESCI (Enhanced Serial Communication Interface found on latest Freescale's auto-motive/industrial controllers).
Following parts are functional at present: boot loader (MAC7100 hardware setup initialization), INTC interrupt controller support, and ESCI serial drivers (both high level and hal_diag).


Needles to say we are willing to contribute it, so I need instructions how to.

Looking forward to hear from you.

Hi


This sounds good.

We will first need a copyright assignment. Please could you take a
look at http://ecos.sourceware.org/assign.html.
OK. We'll do it. Please send me the document/template/instructions.
Once we have received the copyright assignment we will review the
code.
There is a bit of a backlog for big contributions like this at
the moment. So it might take us a while to get around to it.
I tried to follow instructions and fashion in which eCos ports are made so I guess it shall not be hard for you to track it.

Also I have some questions regarding organization of some packages. Namely most of MAC7100 peripherals are (or will be) also implemented on other Freescale chips, but with different architecture: PowerPC (MPC5500 family), ColdFire, so it would be useful to enable sharing of driver sources. Therefore I created ...devs/serial/freescale/esci directory rather than .../mac7100/.... I plan to do so for watchdog too. The question is what to do with interrupt controller (INTC). As far as i have seen in eCos repository interrupt controller software is bonded with respective variant (or platform). The rationale is that interrupt controllers are more closely bonded to processors than other peripherals. I have followed this approach but might try to place INTC sources elsewhere, thus make it sharable with other architectures.
I would like to have opinion from eCos architects (else one is welcome too). Does it pay off? Does it introduce some contradiction with existing eCos organization?


I have some more questions/ suggestions, but this is all for now.

Best regards
Ilija


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]