From: Pedro Alves <palves@redhat.com>
To: Jan Kratochvil <jan.kratochvil@redhat.com>
Cc: Doug Evans <dje@google.com>, gdb-patches <gdb-patches@sourceware.org>
Subject: Re: [patch+doc 1/2] filename-display: 1->4 options {inferior,libs}{,-sepdebug}
Date: Thu, 07 Mar 2013 13:36:00 -0000 [thread overview]
Message-ID: <513897C7.1060709@redhat.com> (raw)
In-Reply-To: <20130307120407.GA21059@host2.jankratochvil.net>
On 03/07/2013 12:04 PM, Jan Kratochvil wrote:
> On Thu, 07 Mar 2013 11:29:48 +0100, Pedro Alves wrote:
>> On 03/06/2013 07:54 PM, Doug Evans wrote:
>>> We have various parameters that together specify where to find separate debug info.
>>> Is it possible to piggyback on that?
>>> E.g., Something minimal for now like specifying which ones are
>>> displayed as relative with the default being none?
>>
>> I've not really been following the discussion closely, but
>> I was also wondering the same. The ultimate goal seems to be to
>> detect system vs non-system binaries ("system" is the word used
>> in the new proposed knobs even). Isn't the definitive answer
>> making GDB consider files under "/usr/"
>> system binaries (the --prefix by default, but could be somewhere
>> else, thus should be tunable), and everything else, non-system?
>
> Besides /usr you need also /bin, /lib, /lib64, /libx32, /opt etc. etc.
True, though I don't think there's much more than that.
> I was thinking rather about the opposite way, consider non-system binaries
> those that are under /usr/local, /root and /home .
The set of system paths is defined by the OS
integrator, so seemed closer to being bounded to me.
The user can well put files wherever it wants.
Other components "know" paths in the system as well, like
gcc with the system includes and the native lib dir paths,
or even the linker and the loader. Note these tools usually
think of /usr/local as "system", and I'd think for this case
that apply too.
Irrespective of glass half empty vs half full, I guess the first
question is whether by-path is a better approach than by
have-sep-debug. It _feels_ more to the point, therefore better to
me, but I'm not really trying to impose it or block the other
approach. I wonder what others think.
> But that is all non-trivial number of directories which will never catch all
> the distros/setups out there.
Using sepdebug for system/non-system also feels like will always
have corner cases, though.
--
Pedro Alves
next prev parent reply other threads:[~2013-03-07 13:36 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-15 20:25 Jan Kratochvil
2013-02-16 8:33 ` Eli Zaretskii
2013-02-18 14:33 ` Jan Kratochvil
2013-02-18 16:27 ` Eli Zaretskii
2013-02-27 19:38 ` Jan Kratochvil
2013-02-27 19:50 ` Doug Evans
2013-02-27 20:06 ` Jan Kratochvil
2013-03-06 19:54 ` Doug Evans
2013-03-07 9:52 ` Jan Kratochvil
2013-03-07 10:26 ` Pedro Alves
2013-03-07 12:01 ` Jan Kratochvil
2013-03-07 10:30 ` Pedro Alves
2013-03-07 12:04 ` Jan Kratochvil
2013-03-07 13:36 ` Pedro Alves [this message]
2013-03-07 15:07 ` Jan Kratochvil
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=513897C7.1060709@redhat.com \
--to=palves@redhat.com \
--cc=dje@google.com \
--cc=gdb-patches@sourceware.org \
--cc=jan.kratochvil@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox