Move symfile.c:generic_load()? Was: RFA: patch to remote.c for larger download packet support (part 1)

Andrew Cagney ac131313@cygnus.com
Sat Oct 16 23:17:00 GMT 1999


Michael Snyder wrote:
> 
> Frank Ch. Eigler wrote:
> 
> > >       o       verify download
> > >
> > >               To, at run time, enable/disable
> > >               the verification of a download.
> >
> > In the same vane, consider an "verify crc" option that uses the "qCRC"
> > packet instead of explicit reads.
> 
> Rather than make it a separate option, just let GDB attempt to
> use the qCRC message to do the verify.  If the target doesn't
> support that, it will reply with an empty packet, and then
> GDB can do the verify the hard way.  Cost: one packet exchange.

Taking it further, why not do the verify _before_ the download? :-)
I'll add a comment suggesting both as enhancements.  As J.T.C. notes:

> generic_load() is used by a many more targets than just the remote
> protocol.  Perhaps a target_verify_memory() function in the target
> vector that can use whatever verification method (crc, checksums,
> etc.) that may be available for a target before falling back to 
> memory read & compare..

Which is true. Something for a later round of changes.

Enjoy,

	Andrew


More information about the Gdb-patches mailing list