Re: [avr-gcc-list] IAR vs GCC (Larry Barello)

[Jesper Hansen]

This may not be feasible for your specific application(s), but
why not just communicate the data over the SPI link to the
field programmer device ? No need for absolute addresses, as
the CPU always "knows" where to look.

In my AVR programmer I've built a complete debugging terminal
using the SPI. Makes it easy to debug, as you already have the cable
connected.


/Jesper
http://www.yampp.com
http://www.myplace.nu/avr


----- Original Message -----
From: "TODD BATZLER" <address@hidden>
To: <address@hidden>
Cc: <address@hidden>
Sent: Wednesday, June 13, 2001 6:38 PM
Subject: Re: [avr-gcc-list] IAR vs GCC (Larry Barello)


> Thanks for the reply.  The reason I'm looking into a string in an absolute
location is so I can go to the field and query my deeply embedded micro to
determine what software revision it contains.  We're using AVR's in a wide
variety of applications and a "standard ID string" at a fixed location in
the program space seemed to be the simplest solution for a limited field
programmer device to ID the software revision level if it always knew where
to look, regardless of the actual micro application.  I'm certainly open to
any other suggestions it anyone has a better system for field level revision
identification where the service personnel are far from technically
competent and the embedded application has no interface but the spi
programming port.
>
> Todd Batzler
> Senior Staff Engineer
> Miller Electric Mfg. Co.
> P.O. Box 1079
> Appleton, WI  54912-1079
> Phone:  (920) 735-4230
> Fax:  (920) 735-4488
> Email:  address@hidden
>
>
> >>> "Larry Barello" <address@hidden> 06/13/01 10:27AM >>>
> Sorry, I can't help you with either of your questions.  IAR is certainly a
> professional development environment.  But, as I use GCC I find that most
> everything I need to do has been covered.  The extra's in IAR seem to be
> archaic or there to support other architectures.  I am sure, if you knew
> how to run the gnu linker, you could do what you want. But I am the wrong
> person to ask.
>
> I question the need to place strings at an absolute location.  But, what
do
> I know...
>
> ----- Original Message -----
> From: "TODD BATZLER" <address@hidden>
> To: <address@hidden>
> Sent: Wednesday, June 13, 2001 5:26 AM
> Subject: [avr-gcc-list] IAR vs GCC (Larry Barello)
>
>
> My company is currently using the IAR Embedded WorkBench as the "official"
> compiler for product development. I've been using GCC for everything of a
> personal nature.  I've been struggling with GCC over a feature that IAR
> makes very pain free, which is placing a string of bytes at an absolute
> location in the code segment, then having the linker to be smart enough to
> "wrap" the other code segments around this absolute segment so as to not
> waste large chunks of the CODE section.  I've managed to dig up enough
> information on linker scripts to modify one to create the required
absolute
> field in the rom image and place my required data there.  However, if my
> relocatable code gets larger than my absolute address minus the start of
> the .text section, I get a warning that the linker location counter can
not
> be "backed up".  Which makes sense.  Any ideas on how to instruct the
> linker to check the available space between defined section starts and any
> absolute sections and then have it figure out when it can place code in
> that section (if it will fit) or place it after the absolute section (if
it
> will not)?
>
> Also, do you know how to find out what was specifically done to the "avr-"
> variants of the GCC binaries to make them different from the "standard"
GCC
> release files.  I'm assuming that support for some of the architectures
was
> removed, plus a bunch of other things.  There's no specific documentation
> for the avr-ld, avr-objcopy, etc files.
>
> Thanks for the attention.
>
> Todd Batzler
> address@hidden
>
>
>
>
>
> _______________________________________________
> avr-gcc-list mailing list
> address@hidden
> http://avr.jpk.co.nz/mailman/listinfo/avr-gcc-list