This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PING] RE: [RFC][PATCH] - Fix pretty-printers which call the inferiors, mi-interface crashs
- From: Pedro Alves <palves at redhat dot com>
- To: Vörtler, Thilo <thilo dot voertler at coseda-tech dot com>, "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>
- Date: Tue, 4 Oct 2016 18:41:15 +0100
- Subject: Re: [PING] RE: [RFC][PATCH] - Fix pretty-printers which call the inferiors, mi-interface crashs
- Authentication-results: sourceware.org; auth=none
- References: <98af10fc7cb34b1baf4e28408ad6d782@EX132MBOX2B.de2.local>
Hi there,
Sorry for the delay. Thanks much for digging in and writing
the patch. It'd be really great if we could have a testcase in
the testsuite to exercise this. That'd help move the patch along,
and, it'd make sure we don't regress again in the future.
Thanks,
Pedro Alves
On 06/27/2016 03:35 PM, Vörtler, Thilo wrote:
>> -----Ursprüngliche Nachricht-----
>> Von: Vörtler, Thilo
>> Gesendet: Donnerstag, 2. Juni 2016 11:09
>> An: 'gdb-patches@sourceware.org' <gdb-patches@sourceware.org>
>> Betreff: [RFC][PATCH] - Fix pretty-printers which call the inferiors, mi-
>> interface crashs
>>
>> Hi,
>> as described in Bug 17437 the mi interface corrupts cached frames in the
>> list_arg_or_local function of mi-cmd-stack.c when a pretty-printer calls a
>> function of the inferior. To prevent this the frame_id is stored and used to
>> restore the frame after calling list_arg_or_local
>> - MI regressions were run and no effect through the patch was seen
>>
>> Is this patch okay?
>>
>> BR, Thilo
>>
>>
>> gdb/ChangeLog:
>>
>> 2016-06-02 Thilo Vörtler <thilo.voertler@coseda-tech.com>
>>
>> * mi-cmd-stack.c (list_args_or_locals) the mi interface corrupts
>> cached frames in the list_arg_or_local function of mi-cmd-stack.c
>> when
>> a pretty-printer calls a function of the inferior. To prevent this the
>> frame_id is stored and used to restore the frame after calling
>> list_arg_or_local