This is the mail archive of the
mailing list for the GDB project.
Re: [PATCH 5/7] Add support for software single step on ARM aarch32-linux in GDBServer.
- From: Antoine Tremblay <antoine dot tremblay at ericsson dot com>
- To: Yao Qi <qiyaoltc at gmail dot com>
- Cc: <gdb-patches at sourceware dot org>
- Date: Tue, 15 Sep 2015 12:49:19 -0400
- Subject: Re: [PATCH 5/7] Add support for software single step on ARM aarch32-linux in GDBServer.
- Authentication-results: sourceware.org; auth=none
- References: <1441973603-15247-1-git-send-email-antoine dot tremblay at ericsson dot com> <1441973603-15247-6-git-send-email-antoine dot tremblay at ericsson dot com> <8637yh5kpz dot fsf at gmail dot com> <55F6C071 dot 1040104 at ericsson dot com> <86twqx3rty dot fsf at gmail dot com> <55F703AF dot 5010707 at ericsson dot com> <86pp1k406c dot fsf at gmail dot com> <55F8101F dot 1000602 at ericsson dot com>
On 09/15/2015 08:33 AM, Antoine Tremblay wrote:
From reviewer's point of view, I'd like each patch series is
self-contained and independent. Your patch series is about supporting
software breakpoint, software single step and tracepoint in ARM linux
GDBserver. Why do we need to worry about the different endianness of
GDBserver and inferior in your patch series? Your code may be right,
and we may need them sooner or later, but we can't do all of them in one
Alright, here's what I propose in this case, since we may need this
support at one point but we do not now.
I could keep the shared breakpoints definitions , shared operation
signatures and other context that carries the endianness.
(Note that most of this context/signatures just can't go away since it
GDB needs it and the code is shared. )
But I will remove the BE8 detection and ELF endianness detection from
So this way while the architecture of GDBServer will be able to handle
it from the start the feature itself won't be until a separate patch.
Does that sound ok ?
Just FYI after more thought I've decided to remove the shared
breakpoints too so the only thing about endianness that will be there is
what is needed to keep the get_next_pcs operations working with GDB.
I'll post a "v2" with only software breakpoints & software single step soon.
It should address your comments properly.
Thanks for the review :)