This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] Add mi-threads-interrupt.exp test (PR 20039)
- From: Pedro Alves <palves at redhat dot com>
- To: Simon Marchi <simon dot marchi at ericsson dot com>, gdb-patches at sourceware dot org
- Date: Tue, 3 May 2016 23:04:24 +0100
- Subject: Re: [PATCH] Add mi-threads-interrupt.exp test (PR 20039)
- Authentication-results: sourceware.org; auth=none
- References: <1462305612-16493-1-git-send-email-simon dot marchi at ericsson dot com> <5983d4d2-016a-8020-c109-cb7ea2cfd179 at redhat dot com>
On 05/03/2016 10:57 PM, Pedro Alves wrote:
> I debugged this a little, and I see that the bug is that -exec-continue ends
> up with thread 3 selected, which generates a =thread-selected event.
Oh, BTW, I think this explains why reversing the thread list order exposed
this bug. Before that patch, we'd issue an -exec-continue with
thread 1 selected, and gdb would resume thread 3, thread 2, and thread 1,
and thus end up with thread 1 selected again, and thus no =thread-selected
event would be output. So I guess that if you select thread 2 before
the -exec-continue, the test exposes the bug even before the thread list
reordering patch. Might be good to do that in the test in case that
detail ever changes again.
Thanks,
Pedro Alves