This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [patch] New m32r remote target, m32rsdi
- From: Andrew Cagney <ac131313 at redhat dot com>
- To: Daniel Jacobowitz <drow at mvista dot com>,Kei Sakamoto <sakamoto dot kei at renesas dot com>
- Cc: gdb-patches at sources dot redhat dot com
- Date: Mon, 22 Sep 2003 03:14:09 -0400
- Subject: Re: [patch] New m32r remote target, m32rsdi
- References: <02e201c35a5e$883a8cb0$5169910a@KEI> <20030817210347.GF11300@nevyn.them.org>
* remote-m32r-sdi.c : New file, interface to m32r on-chip
debug interface, SDI (Scalable Debug Interface).
* Makefile.in (remote-m32r-sdi.o): Add build rule.
* config/m32r/m32r.mt (TDEPFILES) : Add remote-m32r-sdi.o.
I don't feel qualified to review a new remote target, so you'll have to
ask the remote maintainer.
I don't feel exactly qualified either. I think, in terms of the
protocol, GDB we'll have to take it at face value.
Kei, Sorry for the delay, I think 6.0 is done for ....
From the external interface interface viewpoint, I've noticed a few things:
- it has a custom serial [socket] interface
Is there a reason to not use serial.[hc]? All new targets should use a
common interface (in fact I think I just deleted the last target that
didn't :-) as by doing this the user gets more consistent behavior.
- it isn't documented
Probably just a line or two after the doco for "target m32r" in
doc/gdb.texinfo.
- it isn't async
Look at remote.c:remote_async_wait (although as far examples go it may
not help much) and notice how, for 'O' packets, it returns:
/* Return immediately to the event loop. The event loop will
still be waiting on the inferior afterwards. */
status->kind = TARGET_WAITKIND_IGNORE;
If this target is going to last, it's going to need to eventually become
async (just not immediatly). Please don't be suprized if, at some stage
in the not to distant future, I start making noises about this.
- it will need a brief NEWS entry!
enjoy,
Andrew