This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: nios-elf-objcopy S-record length
- From: Daniel Jacobowitz <drow at false dot org>
- To: "Swiger, Dan" <Dan dot Swiger at drs-ss dot com>
- Cc: "'E. Weddington'" <ericw at evcohs dot com>, binutils at sources dot redhat dot com
- Date: Thu, 24 Feb 2005 13:59:18 -0500
- Subject: Re: nios-elf-objcopy S-record length
- References: <2C52593826C4D611A69F00508B60303F0281CF3D@aego_exch1.sigp.net>
On Thu, Feb 24, 2005 at 01:47:38PM -0500, Swiger, Dan wrote:
>
> Eric wrote:
> > Not to get into a long OT discussion about it, but obviously
> > they should
> > provide the sources if asked....
> > Though I can understand your reluctance when you're just
> > trying to solve
> > your problem...
> >
> > Out of curiousity, why is the SRecord line length a problem?
> > Why do you
> > need 16 data bytes per line?
> >
> The reason is because I will be stuffing the S-records into a 16-bit FLASH.
> We have an established S-Record download mechanism that will be greatly
> put-off by the "odd byte" on the end/beggining of an S-Record. It stands to
> reason, to me at least, that if you are building an S-record for a 16-bit
> device (like NIOS), you'd put an even number of 16-bits in each individual
> S-Record.
>
> The other thing that confuses me is that I wouldn't expect this version of
> the GNU tool chain to be 5 years old, which is when the patch was added to
> the "GNU mainline" (if that is an appropriate term). The "advertised
> version" of the NIOS GNU tools is "2.9", but I can't find quickly when the
> mainline was at 2.9.
Current version is 2.16. 2.9 would be, yes, about that old.
--
Daniel Jacobowitz
CodeSourcery, LLC