This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH][PR python/17364] Cleanup registration of __gdb_builtin_type_bound128 pretty-printer
- From: Doug Evans <dje at google dot com>
- To: "Tedeschi, Walfred" <walfred dot tedeschi at intel dot com>
- Cc: "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>
- Date: Thu, 9 Oct 2014 09:23:09 -0700
- Subject: Re: [PATCH][PR python/17364] Cleanup registration of __gdb_builtin_type_bound128 pretty-printer
- Authentication-results: sourceware.org; auth=none
- References: <yjt21tqn7jvk dot fsf at ruffy dot mtv dot corp dot google dot com> <AC542571535E904D8E8ADAE745D60B192F92CA48 at IRSMSX104 dot ger dot corp dot intel dot com>
On Mon, Oct 6, 2014 at 12:33 AM, Tedeschi, Walfred
<walfred.tedeschi@intel.com> wrote:
> Hi Doug,
>
> I have also submitted a patch for this last week. Though I was not creating a new directory for pretty printers.
> https://sourceware.org/ml/gdb-patches/2014-09/msg00867.html
>
> I would like though to understand a bit more your comment on having x86 on the name of the register. What is the purpose of that?
> (I could change it in case)
Imagine a multi-arch gdb with several arches having bounds registers.
How do we distinguish them in the pretty-printer namespace?
> About registration of the pretty printer. We could do the following:
> 1. Register an event listener for new obj_files.
> 2. Query at every event the architecture of the inferior. (In case we had an event new_inferior loaded it would be cheaper.)
> 3. In case of x86 or amd64 we can then register the pretty-printer for the bound registers.
We could do something like that for 7.9 alright.
At the moment it feels like a bit of overkill just for one pretty-printer.
And for 7.8.1 I want to do something simple.