From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <archer-return-2021-listarch-archer=sourceware.org@sourceware.org>
Received: (qmail 13952 invoked by alias); 5 Jul 2010 09:36:44 -0000
Mailing-List: contact archer-help@sourceware.org; run by ezmlm
Sender: <archer@sourceware.org>
Precedence: bulk
List-Post: <mailto:archer@sourceware.org>
List-Help: <mailto:archer-help@sourceware.org>
List-Subscribe: <mailto:archer-subscribe@sourceware.org>
List-Id: <archer.sourceware.org>
Received: (qmail 13937 invoked by uid 22791); 5 Jul 2010 09:36:42 -0000
X-SWARE-Spam-Status: No, hits=-6.9 required=5.0
	tests=BAYES_00,RCVD_IN_DNSWL_HI,SPF_HELO_PASS,T_RP_MATCHES_RCVD
X-Spam-Check-By: sourceware.org
Date: Mon, 05 Jul 2010 09:36:00 -0000
From: Panu Matilainen <pmatilai@redhat.com>
To: Tom Tromey <tromey@redhat.com>
cc: Roland McGrath <roland@redhat.com>, Project Archer <archer@sourceware.org>
Subject: Re: find-debuginfo.sh change for gdb index
In-Reply-To: <m339w1lkuu.fsf@fleche.redhat.com>
Message-ID: <alpine.LFD.2.00.1007051223290.8757@localhost>
References: <m3vd91tqpl.fsf@fleche.redhat.com> <20100629232147.C019548255@magilla.sf.frob.com> <m3r5joscvs.fsf@fleche.redhat.com> <20100630181436.518364C33C@magilla.sf.frob.com> <m31vbos7lc.fsf@fleche.redhat.com> <m3y6dws151.fsf@fleche.redhat.com>
 <20100630204424.3DCE34C33C@magilla.sf.frob.com> <m3lj9wrz60.fsf@fleche.redhat.com> <20100630221406.254AC4C33E@magilla.sf.frob.com> <m339w1lkuu.fsf@fleche.redhat.com>
User-Agent: Alpine 2.00 (LFD 1167 2008-08-23)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
X-SW-Source: 2010-q3/txt/msg00004.txt.bz2

On Fri, 2 Jul 2010, Tom Tromey wrote:

> Tom> For a symbol file X, it looks for X.gdb-index.
>
> Roland> It's not clear to me what that means in the separate .debug case.
> Roland> If "symbol file" means the file with the DWARF, then that file is
> Roland> foo.debug so you will be looking for foo.debug.gdb-index.
>
> Right, that is what we do.
>
> Roland> I think it is preferable for the packaging not to add the new symlinks.
> Roland> It doesn't seem like we really need them, since you can look at the
> Roland> .build-id/xx/yyy.debug symlink target name instead.
>
> Sounds good.
>
> In this case I think the most recent patch I sent is the one to use.

So it would be this patch, right?

--- find-debuginfo.sh.orig	2010-06-29 16:19:42.000000000 -0600
+++ find-debuginfo.sh	2010-06-30 14:41:19.000000000 -0600
@@ -96,6 +96,15 @@
    chmod 444 "$1" || exit
  }

+# Create a .gdb-index file for $1.
+make_gdb_index()
+{
+  local f="$1"
+  local d="${f%/*}"
+  # We don't care if gdb gives an error.
+  gdb --batch-silent -ex "file $f" -ex "maintenance save-gdb-index $d" > /dev/null 2>&1
+}
+
  # Make a relative symlink to $1 called $3$2
  shopt -s extglob
  link_relative()
@@ -224,6 +233,8 @@
      chmod u-w "$f"
    fi

+  make_gdb_index "$debugfn"
+
    if [ -n "$id" ]; then
      make_id_link "$id" "$dn/$(basename $f)"
      make_id_link "$id" "/usr/lib/debug$dn/$bn" .debug

Is the necessary patch(es) already in rawhide gdb, I dont see anything 
obviously related in gdb changelogs?

One thing this does is that it forces rpm-build to depend on gdb, which 
hardly is the end of the world, just something to note.

 	- Panu -