This is the mail archive of the
mailing list for the GDB project.
Re: [PATCH 0/3] Fix fails in gdb.mi on arm-none-eabi
- From: Yao Qi <yao at codesourcery dot com>
- To: <gdb-patches at sourceware dot org>
- Date: Tue, 30 Sep 2014 22:00:28 +0800
- Subject: Re: [PATCH 0/3] Fix fails in gdb.mi on arm-none-eabi
- Authentication-results: sourceware.org; auth=none
- References: <1409533072-6152-1-git-send-email-yao at codesourcery dot com> <87vbolzqq9 dot fsf at codesourcery dot com>
On 09/18/2014 09:01 AM, Yao Qi wrote:
> Yao Qi <email@example.com> writes:
>> This series fixes some fails in mi-var-child.exp and
>> mi-var-display.exp, which are caused by printed symbol on address
>> zero. One of them is fixed by "set print symbol" off, and that is
>> what patch 1/3 does. Patch 2/3 is a refactor patch, which uses
>> mi_varobj_update in tests to simplify them.
>> The cause of the rest of fails is that function pointers are NULL,
>> and they point to a code symbol _ftext on address zero. We can
>> relax pattern to match both "0x0" and "0x0 <_ftext>", but I didn't
>> fix it in this way. Instead, I explicitly assign function address
>> to these function pointers, so the output becomes "$hex <$func_name>"
>> on all the targets. See details in patch 3/3.
>> The patch series is tested on x86_64-linux with native and gdbserver,
>> and arm-none-eabi.
> Patch 1 is pushed in. Ping patch 2 and patch 3.