This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
[PATCH 0/2] Consolidate target_{wait,resume} calls between GDB and gdbserver
- From: Sergio Durigan Junior <sergiodj at redhat dot com>
- To: GDB Patches <gdb-patches at sourceware dot org>
- Cc: Sergio Durigan Junior <sergiodj at redhat dot com>
- Date: Mon, 29 Aug 2016 00:11:49 -0400
- Subject: [PATCH 0/2] Consolidate target_{wait,resume} calls between GDB and gdbserver
- Authentication-results: sourceware.org; auth=none
Hi,
This simple patch series proposes a consolidation of the calls to
target_{wait,resume} functions between GDB and gdbserver. This is a
necessary and helpful step in the direction that we're heading, and is
particularly important for my work on the unification of fork_inferior.
The first patch implements a new function, target_continue, on top of
target_resume (or the_target->resume, depending on where you're
looking), and replaces all calls to target_resume (or
the_target->resume) by either target_continue or
target_continue_no_signal (this last one has been a GDB citizen for
some time). Note that I haven't touched infrun.c's
"do_target_resume", which is a wrapper around target_resume and does
many other things.
The second patch, which is even simpler, is meant to pay a technical
debt on gdbserver: the lack of the "target_wait" function. So there
it is now, properly implemented, even though everyone else will still
use mywait because it accounts for many aspects that the regular
target_wait doesn't. This is actually another place for future
improvement: make the two versions of target_wait identical API-wise,
so that sharing code is easier.
Both patches do not introduce regressions (I've tested them using the
Try server feature on BuildBot, BTW).
OK to apply?
Thanks,