This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH 0/8] [AArch64] Add support for tracepoints
- From: Pedro Alves <palves at redhat dot com>
- To: Pierre Langlois <pierre dot langlois at arm dot com>, gdb-patches at sourceware dot org
- Date: Wed, 08 Jul 2015 11:57:00 +0100
- Subject: Re: [PATCH 0/8] [AArch64] Add support for tracepoints
- Authentication-results: sourceware.org; auth=none
- References: <1436273518-5959-1-git-send-email-pierre dot langlois at arm dot com>
On 07/07/2015 01:51 PM, Pierre Langlois wrote:
> Hi all,
>
> These patches enable tracepoints for AArch64. Although tracepoints are
> enabled in GDBServer with the last patch, most of the changes are in GDB.
> The most important changes teach AArch64's frame unwinders to report when
> the inferior is unavailable.
>
> The first three patches refactor the frame caches. The idea is to keep
> accesses to the inferior's registers in aarch64_make_prologue_cache and
> aarch64_make_stub_cache. This way the following patches can easily catch
> exceptions when the inferior is unavailable.
>
> The following two patches teach AArch64's unwinders to terminate
> gracefully, in a similar way as it was done for x86 here:
>
> https://sourceware.org/ml/gdb-patches/2011-02/msg00611.html
>
> It fixes cases where we do not have debugging information and AArch64's
> unwinders need to be used when examining a trace buffer. In this context
> we cannot assume that the inferior's memory and registers are available.
>
This all looks like what I'd expect to see. LGTM.
Yao may want to take a look too.
Just a few nits:
- In a few patches, you had a spurious empty line in the middle
of the ChangeLog entry. Remove it please.
- Several new functions are missing an intro comment.
- aarch64_prologue_frame_unwind_stop_reason in patch 4 has
odd indentation in the "We've hit a wall" case.
Thanks,
Pedro Alves