This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH v4 3/6] Refactor arm_software_single_step to use regcache.
- 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: Thu, 3 Dec 2015 09:41:53 -0500
- Subject: Re: [PATCH v4 3/6] Refactor arm_software_single_step to use regcache.
- Authentication-results: sourceware.org; auth=none
- References: <1449062264-18565-1-git-send-email-antoine dot tremblay at ericsson dot com> <1449062264-18565-4-git-send-email-antoine dot tremblay at ericsson dot com> <86egf3u8py dot fsf at gmail dot com> <56603F64 dot 6050906 at ericsson dot com> <861tb3tx7s dot fsf at gmail dot com>
On 12/03/2015 09:37 AM, Yao Qi wrote:
Antoine Tremblay <antoine.tremblay@ericsson.com> writes:
Unfortunately we do for the syscall_next_pc target dependent call.
The reason I did not change syscall_next_pc for regcache is that this
would require the implementation of a syscall_next_pc that is not
using the frame unwinders and directly inspects the stack for multiple
targets.
This would need to be done for arm, mips, nios2, and tix6x.
software_single_step would also have to be adapted for mips, nios2 and
tix6x for this new api.
We don't need to touch targets other than arm. syscall_next_pc is a
function pointer in struct gdbarch_tdep in arm-tdep.h.
Oops indeed sorry my bad, I'll change it then.
/* Target-dependent structure in gdbarch. */
struct gdbarch_tdep
{
...
/* Return the expected next PC if FRAME is stopped at a syscall
instruction. */
CORE_ADDR (*syscall_next_pc) (struct frame_info *frame);
...
};
we can change frame to regcache.
The indentation looks odd in my mailer, if it is right to you, that is
fine. Maybe, my mailer doesn't show the code format correctly.
That happens all the time to me with thunderbird in a reply window...