RFC: Remote "qSupported" features probe

[Daniel Jacobowitz]

On Tue, Jun 13, 2006 at 09:42:37AM +0100, Nathan Sidwell wrote:
> Daniel Jacobowitz wrote:
> 
> >Comments?  Questions?  Otherwise, I will plan to commit this in a week.
> >I haven't seen any objections to the latest version of the proposal,
> >although I'm somewhat suspicious that I've just talked everyone else
> >into the ground.
> 
> Can you clarify something for me?  If a stub recieves a 'qSupported:foo' 
> packet, should it still respond with any default information it might have 
> -- such as the PacketSize response?  Or should such default responses only 
> be sent for an unadorned qSupported packet?

qSupported:foo means "I'm GDB, I support the "foo" feature, what do you
support?".  The stub's response should be independent of the arguments
in the packet.

-- 
Daniel Jacobowitz
CodeSourcery