From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <libc-hacker-return-6467-listarch-libc-hacker=sources.redhat.com@sources.redhat.com>
Received: (qmail 14695 invoked by alias); 15 Apr 2003 15:02:39 -0000
Mailing-List: contact libc-hacker-help@sources.redhat.com; run by ezmlm
Precedence: bulk
List-Subscribe: <mailto:libc-hacker-subscribe@sources.redhat.com>
List-Archive: <http://sources.redhat.com/ml/libc-hacker/>
List-Post: <mailto:libc-hacker@sources.redhat.com>
List-Help: <mailto:libc-hacker-help@sources.redhat.com>, <http://sources.redhat.com/ml/#faqs>
Sender: libc-hacker-owner@sources.redhat.com
Received: (qmail 14678 invoked from network); 15 Apr 2003 15:02:39 -0000
Received: from unknown (HELO localhost.localdomain) (195.113.19.66)
  by sources.redhat.com with SMTP; 15 Apr 2003 15:02:39 -0000
Received: from sunsite.ms.mff.cuni.cz (sunsite.mff.cuni.cz [127.0.0.1])
	by localhost.localdomain (8.12.8/8.12.8) with ESMTP id h3FF2aqO007194;
	Tue, 15 Apr 2003 17:02:36 +0200
Received: (from jakub@localhost)
	by sunsite.ms.mff.cuni.cz (8.12.8/8.12.8/Submit) id h3FF2ZjZ007192;
	Tue, 15 Apr 2003 17:02:35 +0200
Date: Tue, 15 Apr 2003 15:02:00 -0000
From: Jakub Jelinek <jakub@redhat.com>
To: Ulrich Drepper <drepper@redhat.com>, Roland McGrath <roland@redhat.com>
Cc: Glibc hackers <libc-hacker@sources.redhat.com>
Subject: [PATCH] Fix prelinking on ppc32
Message-ID: <20030415150235.GR16629@sunsite.ms.mff.cuni.cz>
Reply-To: Jakub Jelinek <jakub@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.4i
X-SW-Source: 2003-04/txt/msg00045.txt.bz2

Hi!

When calling elf_machine_rela from dl-conflict.c, sym is NULL, so we
certainly cannot do checks on what it points to.

BTW: Shouldn't most of the TLS handling go into dl-machine.c on PPC32?
I mean, unless there are -fno-pic shared libraries, linker certainly
shouldn't keep any TLS relocs but R_PPC_DTPMOD32, R_PPC_DTPREL32 and
R_PPC_TPREL32 around and thus the rest doesn't hurt if it is slower.

2003-04-15  Jakub Jelinek  <jakub@redhat.com>

	* sysdeps/powerpc/powerpc32/dl-machine.h (elf_machine_rela): Avoid
	checking R_PPC_RELATIVE, R_PPC_NONE and whether relocation is
	against local symbol in conflict processing.

--- libc/sysdeps/powerpc/powerpc32/dl-machine.h.jj	2003-04-01 04:20:54.000000000 -0500
+++ libc/sysdeps/powerpc/powerpc32/dl-machine.h	2003-04-15 09:08:04.000000000 -0400
@@ -365,6 +365,7 @@ elf_machine_rela (struct link_map *map, 
   struct link_map *sym_map;
 #endif
 
+#ifndef RESOLVE_CONFLICT_FIND_MAP
   if (r_type == R_PPC_RELATIVE)
     {
       *reloc_addr = map->l_addr + reloc->r_addend;
@@ -381,18 +382,21 @@ elf_machine_rela (struct link_map *map, 
     value = map->l_addr;
   else
     {
-#if defined USE_TLS && !defined RTLD_BOOTSTRAP
+# if defined USE_TLS && !defined RTLD_BOOTSTRAP
       sym_map = RESOLVE_MAP (&sym, version, r_type);
       value = sym == NULL ? 0 : sym_map->l_addr + sym->st_value;
-#else
+# else
       value = RESOLVE (&sym, version, r_type);
-# ifndef RTLD_BOOTSTRAP
+#  ifndef RTLD_BOOTSTRAP
       if (sym != NULL)
-# endif
+#  endif
 	value += sym->st_value;
-#endif
+# endif
     }
   value += reloc->r_addend;
+#else
+  value = reloc->r_addend;
+#endif
 
   /* A small amount of code is duplicated here for speed.  In libc,
      more than 90% of the relocs are R_PPC_RELATIVE; in the X11 shared

	Jakub