Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Daniel Jacobowitz <drow@false.org>
To: Eli Zaretskii <eliz@gnu.org>
Cc: gdb-patches@sourceware.org, guitton@adacore.com
Subject: Re: Add support for --with-system-gdbinit
Date: Wed, 14 Jan 2009 23:07:00 -0000	[thread overview]
Message-ID: <20090114230629.GA8400@caradoc.them.org> (raw)
In-Reply-To: <ufxjlwo4s.fsf@gnu.org>

On Wed, Jan 14, 2009 at 11:53:23PM +0200, Eli Zaretskii wrote:
> By specifying the --with-system-gdbinit=$prefix/etc/gdbinit switch to
> configure, we tell GDB to find gdbinit in that specific absolute
> location.  But you also tell that moving GDB into another location
> does not prevent GDB from finding gdbinit in that other location.
> That begs the question: if GDB can find gdbinit even after it is
> moved, why do we need to specify $prefix in the --with-system-gdbinit
> option in the first place?  Evidently, all GDB needs is to know that
> gdbinit will be in the etc/ subdirectory of the top of its tree, so it
> can look there without me telling it where that top will be by
> default.
> 
> IOW, it sounds like --with-system-gdbinit=etc/gdbinit should be
> enough; the $prefix part is redundant.

I understand now.  There's an alternative to keeping it in the same
tree:

  --prefix=/my/working/directory --with-system-gdbinit=/etc/gdbinit

In this version /etc is not relative to $prefix, so it wasn't
redundant.  This example may seem a little weird, but it makes more
sense for --with-debug-file-directory; your native system's libraries
are always in /usr/lib/debug, no matter where GDB goes.

This relocation scheme is a bit complicated.  I bear some blame for
that, though I didn't invent it; I just propogated the logic from its
original home in GCC.

-- 
Daniel Jacobowitz
CodeSourcery


  reply	other threads:[~2009-01-14 23:07 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-01-14 13:56 Daniel Jacobowitz
2009-01-14 19:39 ` Eli Zaretskii
2009-01-14 19:50   ` Daniel Jacobowitz
2009-01-14 20:22     ` Eli Zaretskii
2009-01-14 20:38       ` Daniel Jacobowitz
2009-01-14 21:55         ` Eli Zaretskii
2009-01-14 23:07           ` Daniel Jacobowitz [this message]
2009-01-15  4:10             ` Eli Zaretskii
2009-01-19 14:11     ` Jerome Guitton
2009-01-19 14:20       ` Daniel Jacobowitz
2009-01-19 18:40       ` Eli Zaretskii
2009-01-21 10:54         ` Jerome Guitton
2009-01-21 19:03           ` Eli Zaretskii
2009-01-23 17:24             ` Jerome Guitton
2009-01-23 22:55               ` Eli Zaretskii
2009-01-28 18:00                 ` Jerome Guitton
2009-01-23 17:42           ` Daniel Jacobowitz
2009-01-26  9:12             ` Jerome Guitton
2009-01-26 14:59               ` Daniel Jacobowitz
2009-01-28 15:06                 ` Jerome Guitton
2009-01-16 11:54 ` Jerome Guitton

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=20090114230629.GA8400@caradoc.them.org \
    --to=drow@false.org \
    --cc=eliz@gnu.org \
    --cc=gdb-patches@sourceware.org \
    --cc=guitton@adacore.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