This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [PATCH]: Update thread list
- To: Andrew Cagney <ac131313 at cygnus dot com>
- Subject: Re: [PATCH]: Update thread list
- From: Michael Snyder <msnyder at redhat dot com>
- Date: Wed, 26 Jul 2000 12:33:40 -0700
- CC: Michael Snyder <msnyder at cygnus dot com>, gdb-patches at sourceware dot cygnus dot com
- Organization: Red Hat, Inc.
- References: <3978E783.37F9@cygnus.com> <397E6CE0.AD61DDD8@cygnus.com>
Andrew Cagney wrote:
>
> Michael Snyder wrote:
> >
> > I can think of two times when it is both safe and useful
> > for GDB to go query the target to find out if there are
> > any new threads.
> > 1) "info threads" and
> > 2) "thread apply all xxx"
> >
> > This change adds the second (the first is already in place).
>
> Should this (and the other) call be wrapped in something that stops GDB
> updating the thread list more than once? Doing a
> target_find_new_threads() can get expensive and, assuming the target has
> stopped, shouldn't change between each call.
>
> But you're the thread maintainer.
Good idea, I guess. I'd wanna think about what "level" it
belongs at. Should we have the core GDB maintain a "once" flag,
or should we go ahead and call the target_find_new_threads
function, and let the target decide? Certainly there may
be future targets where "all-stop" is not true, and in that
day there may be some "all-stop" targets and some "some-stop"
ones.