This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [WIP] TI msp430 CIO support
- From: Joel Brobecker <brobecker at adacore dot com>
- To: Kevin Buettner <kevinb at redhat dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Fri, 17 May 2013 10:36:03 +0400
- Subject: Re: [WIP] TI msp430 CIO support
- References: <20130516212358 dot 23f3bcdb at mesquite dot lan>
> It may be that this is considered to be (barely) acceptable behavior
> for dprintf. Even if this is acceptable for dprintf, it is completely
> unacceptable for the CIO implementation. It makes no sense whatsoever
> to turn a "next" into a "continue" when attempting to step over a
> statement which performs I/O.
Agreed. Having dprintf with limitations is probably better than
no dprintf at all, but that is a very very serious limitation, IMO.
> In order for this work to go into the GDB sources, the use of the
> shared library machinery needs to be removed and replaced with some
> other kind of breakpoint machinery which still provides correct
> behavior. This same mechanism could be used to improve dprintf
> behavior as well.
Agreed again. I think this option has been delayed and delayed because
it might be hairy to implement right (messing with handle_inferior_event).
--
Joel