This is the mail archive of the
ecos-patches@sources.redhat.com
mailing list for the eCos project.
Re: at91 HAL patch
- From: Laurent GONZALEZ <laurent dot gonzalez at silicomp dot fr>
- To: "Koeller, T." <Thomas dot Koeller at baslerweb dot com>
- Cc: "'Andrew Lunn'" <andrew dot lunn at ascom dot ch>, ecos-patches at sources dot redhat dot com
- Date: Tue, 19 Aug 2003 16:14:19 +0200
- Subject: Re: at91 HAL patch
- Organization: SILICOMP RESEARCH INSTITUTE
- References: <850597605E79D21182830008C7A4B9CF1EB4264F@COMM1>
On Tue, 19 Aug 2003 15:39:34 +0200
"Koeller, T." <Thomas.Koeller@baslerweb.com> wrote:
> Just another related thought:
> Wouldn't it be more sensible to include the platform header
> (plf_io.h) at the end of var_io.h instead of near the beginning,
> as is the case now? Then its contents could refer to the
> definitions contained in var_io.h, for example, assign
> meaningful (in the platform context) names to I/O pins.
>
> tk
Yes, it is possible. But if you do that, all board (at91/ebxx or at91/custom_board) that uses the same mcu (say at91r40807) will provide the same definitions in their own plf_io.h .
PIO bit definitions should definitively be in var_io.h . As at91/var provide support for several mcus, it should also provide PIO bit definitions for each of them.
IMHO, at91_var shall contain the user API (eg register definitions) and include processor specific var_io.h like file for bit field definition (include/x40/x40_pio_pins.h, include/x55/x55_pio_pins.h, ...).
regards,
--
GONZALEZ Laurent
Silicomp Research Institute
Tel: 04 76 41 66 98