This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
Re: gdb/mi and signals vs gdb/cli
- To: Fernando Nasser <fnasser at redhat dot com>
- Subject: Re: gdb/mi and signals vs gdb/cli
- From: Andrew Cagney <ac131313 at cygnus dot com>
- Date: Mon, 01 Oct 2001 11:26:38 -0400
- Cc: Mickael Gicquaire <Mickael dot Gicquaire at st dot com>,"gdb at sources dot redhat dot com" <gdb at sources dot redhat dot com>
- References: <1010928114759.ZM22692@bristol.st.com> <3BB50C41.10807@cygnus.com> <3BB87F6A.F4C2DE5A@redhat.com>
>
> Andrew,
>
> I am afraid we may be missing a ui_out_flush() at the end of all
> of the switch clauses in infrun.c
I think that sounds wrong. I don't think infrun.c needs to know if/when
the output needs to be flushed. Remember, we're trying to separate the
stop output from its analysis. (I'll ignore those print_stop_reason()
calls).
BTW, there were some fixed problems with missing top level flushes and a
missing stop reason case. Check the changelogs. 28/6 is getting pretty
old for a snap, can I suggest trying something more recent.
enjoy,
Andrew
PS: Since the snaps are comming from the 5.1 branch you'll need to use
CVS to get current sources.