This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [MI non-stop 09/11] Don't care about user-defined thread if --thread is present.
- From: Vladimir Prus <vladimir at codesourcery dot com>
- To: Pedro Alves <pedro at codesourcery dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Sun, 13 Jul 2008 08:34:22 +0400
- Subject: Re: [MI non-stop 09/11] Don't care about user-defined thread if --thread is present.
- References: <200806282058.04805.vladimir@codesourcery.com> <200807111451.59696.pedro@codesourcery.com>
On Friday 11 July 2008 17:51:59 Pedro Alves wrote:
> A Saturday 28 June 2008 17:58:04, Vladimir Prus wrote:
> > If an MI command has --thread parameter, we don't care in any way
> > that user selected thread is. Will commit after core non-stop is in.
> >
>
> Note that the version that was committed doesn't have user_selected_ptid
> anymore,
Oh, that was a nice concept :-)
> and neither an inferior_ptid == null_ptid case, as long as the
> target still has execution. I think that you're already checking
> if the thread is THREAD_EXITED somewhere else before letting the
> command run, so this will disappear.
I don't think I check for THREAD_EXITED. OTOH, given that now the check
for non-exiteness applies not to user-selected thread, but to the thread
we ultimately operate on (after --thread is processed), I don't think
this patch is necessary any more.
- Volodya