This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
Re: REMOTE_TRANSLATE_XFER_ADDRESS()
- To: jtc at redback dot com
- Subject: Re: REMOTE_TRANSLATE_XFER_ADDRESS()
- From: Andrew Cagney <ac131313 at cygnus dot com>
- Date: Tue, 01 May 2001 12:09:20 -0400
- Cc: gdb at sourceware dot cygnus dot com
- References: <5mpuh1sxwr.fsf@jtc.redback.com>
"J.T. Conklin" wrote:
>
> Assuming we adopt something like the step over range packet for the
> remote protocol, must the addresses that bound the range be translated
> from the GDB's internal representation to the target's representation?
> remote_xfer_memory() does this for memory reads and writes.
>
> If so, how would one go about this. REMOTE_TRANSLATE_XFER_ADDRESS()
> takes an address and a length. I don't know what I would use for a
> length --- actually I don't understand the length part at all, even
> for memory transfers.
The virtual->physical translation may not apply for the entire length of
the transfer. When this occures the target should prune the transfer to
just the correct number of bytes.
Andrew