This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] "tfind" across unavailable-stack frames.
- From: Yao Qi <yao at codesourcery dot com>
- To: Pedro Alves <palves at redhat dot com>
- Cc: <gdb-patches at sourceware dot org>
- Date: Tue, 17 Dec 2013 20:37:14 +0800
- Subject: Re: [PATCH] "tfind" across unavailable-stack frames.
- Authentication-results: sourceware.org; auth=none
- References: <1366214779 dot 30939 dot 1 at abidh-ubunto1104> <516F11B9 dot 8030202 at redhat dot com> <52AB48B6 dot 6040206 at redhat dot com> <52ABF8D7 dot 1050805 at codesourcery dot com> <52AF27F7 dot 2060500 at redhat dot com> <52B01328 dot 8030707 at codesourcery dot com> <52B022C8 dot 3090504 at redhat dot com>
On 12/17/2013 06:09 PM, Pedro Alves wrote:
> It is clear, but it is not as precise or sufficient. A wild
> card means that given these frame ids:
>
> fid1: {code_p,stack_p,special_p}
> fid2: {!code_p,stack_p,!special_p}
>
> fid3: {code_p,stack_p,special_p}
> fid4: {code_p,stack_p,!special_p}
>
> {fid1, fid2} with same stack addresses,
> and {fid3, fid4} with same code and stack addresses,
> both:
>
> frame_id_eq(fid1, fid2)
> frame_id_eq(fid3, fid4)
>
> return true.
>
>> > I don't see
>> > any extra information the last sentence "Otherwise, xxxxx" delivered
>> > except confusion.
> The extra information indicates that e.g., frame_id_eq(fid3, fid4)
> above returns true, not false, as one might at first expect.
Thanks for giving these examples.
>
> Whether this whole wildcarding business is a good idea, is
> another story.
Anyway, I've got use to it now. :)
--
Yao (éå)