This is the mail archive of the
ecos-devel@sourceware.org
mailing list for the eCos project.
RE: RedBoot patches regarding redboot_getc_terminate
- From: "Doyle, Patrick" <Patrick_Doyle at dtccom dot com>
- To: "'David Vrabel'" <dvrabel at arcom dot com>, Andrew Lunn <andrew at lunn dot ch>
- Cc: "Doyle, Patrick" <Patrick_Doyle at dtccom dot com>, "'ecos-devel at sources dot redhat dot com'" <ecos-devel at sources dot redhat dot com>, "'Andrew Dyer'" <adyer at righthandtech dot com>
- Date: Thu, 18 May 2006 12:23:58 -0400
- Subject: RE: RedBoot patches regarding redboot_getc_terminate
> -----Original Message-----
> From: David Vrabel [mailto:dvrabel@arcom.com]
> Sent: Thursday, May 18, 2006 11:55 AM
> To: Andrew Lunn
> Cc: Doyle, Patrick; 'ecos-devel@sources.redhat.com'; 'Andrew Dyer'
> Subject: Re: RedBoot patches regarding redboot_getc_terminate
>
>
> Andrew Lunn wrote:
> >> I'm not entirely happy with the fix. It terminates the
> download when all
> >> the relevant bits of the ELF have been transferred. The causes the
> >> sender to think that the file transfer has failed which causes some
> >> customer confusion.
> >
> > It is possible to include a text string in the error message.
> >
> > http://www.faqs.org/rfcs/rfc1350.html, figure 5-4.
> >
> > So maybe a message like "Remaining file contents not required" would
> > prevent confusion. This assumes the lame tftp servers actually display
> > the message to the user.
>
> This would take care of the TFTP transfers. But what about
> the Y-modem transfers?
Well, 'retboot_getc_terminate(false)' has always worked for me in the past,
and continues to work for me now that I've put it back in.
Perhaps the 'tftp_stream_terminate()' function should be modified to issue
the "Remaining file contents not required" message in the case of a normal
termination.
>
> >> I think a better solution would be for the downloader to continue to
> >> transfer the remaining portions of the ELF image and just throw them
away."
> >
> > This can be a lot of data, eg the complete debug symbol table if the
> > image has been compiled -g. I would not really be in favor of that.
>
> Agreed. Although I think this is what RedBoot used to do for Y-modem
> transfers.
Actually, as confusing as this is to me, this is what it _still_ does! I'm
pretty sure I remember the behavior you describe, but I'm darn'd if I can
see it in the code (nor in practice).
So if I submit a modified patch that restores the call to
'redboot_getc_terminate()' to its original (pre Andrew Dyer) state and
modify 'tftp_stream_terminate()' to issue different strings based on normal
vs. abnormal termination conditions, is there anybody who can test the TFTP
changes?
--wpd