From: Jan Kratochvil <jan.kratochvil@redhat.com>
To: Pedro Alves <palves@redhat.com>
Cc: "H.J. Lu" <hjl.tools@gmail.com>, gdb-patches@sourceware.org
Subject: Re: [patch+7.5.1] Work around PR libc/13097 "linux-vdso.so.1" #3
Date: Sun, 25 Nov 2012 18:15:00 -0000 [thread overview]
Message-ID: <20121125181505.GA26194@host2.jankratochvil.net> (raw)
In-Reply-To: <20121123184322.GA30659@host2.jankratochvil.net>
On Fri, 23 Nov 2012 19:43:22 +0100, Jan Kratochvil wrote:
> On Fri, 23 Nov 2012 19:18:57 +0100, Pedro Alves wrote:
> > It's found in AT_SYSINFO_EHDR too.
> >
> > Alternatively to hard coding the names, maybe we could match the vdso address
> > found through that with the addresses found iterating the dynamic linker list, to
> > know which dynamic linker entry is the vdso.
>
> THis is how elfutils does it.
Implemented below.
But now it discards any shared libraries which match a symbol file loaded via
add-symbol-file-from-memory. Which may be OK but it is more widespread change
than before.
Otherwise there are some configuration/interface difficulties, symfile-mem.c
is not present in every configuration so duplicating its check of
target_auxv_search (AT_SYSINFO_EHDR) in solib-svr4.c may not be correct.
But otherwise the symbol file added via add_vsyscall_page cannot be
differentiated later from those added by add-symbol-file-from-memory.
To reduce the change impact some new field / method / interface would have to
be added to communicate with symfile-mem.c with solib-svr4.c
As is I do not recommend it for 7.5.x.
> But there can be multiple vDSOs I think while
> AT_SYSINFO_EHDR is only one so I have to check that.
There is [vdso] which is listed in _r_debug.r_map and handled above.
There is also [vsyscall] but this page does not have ELF format
(and therefore obviously it is also not listed in _r_debug.r_map).
0: 48 c7 c0 60 00 00 00 mov $0x60,%rax
7: 0f 05 syscall
9: c3 retq
a: cc int3
[...]
400: 48 c7 c0 c9 00 00 00 mov $0xc9,%rax
407: 0f 05 syscall
409: c3 retq
40a: cc int3
[...]
800: 48 c7 c0 35 01 00 00 mov $0x135,%rax
807: 0f 05 syscall
809: c3 retq
80a: cc int3
[...]
fff: cc int3
Jan
gdb/
2012-11-25 Jan Kratochvil <jan.kratochvil@redhat.com>
PR 14466
* solib-svr4.c (svr4_read_so_list): Rename to ...
(svr4_current_sos_1): ... here and change the function comment.
(svr4_current_sos): New function.
gdb/testsuite/
2012-11-25 Jan Kratochvil <jan.kratochvil@redhat.com>
PR 14466
* gdb.base/vdso-warning.c: New file.
* gdb.base/vdso-warning.exp: New file.
diff --git a/gdb/solib-svr4.c b/gdb/solib-svr4.c
index 37cc654..77f4adc 100644
--- a/gdb/solib-svr4.c
+++ b/gdb/solib-svr4.c
@@ -1248,10 +1248,10 @@ svr4_read_so_list (CORE_ADDR lm, struct so_list ***link_ptr_ptr,
}
}
-/* Implement the "current_sos" target_so_ops method. */
+/* Implement the main part of the "current_sos" target_so_ops method. */
static struct so_list *
-svr4_current_sos (void)
+svr4_current_sos_1 (void)
{
CORE_ADDR lm;
struct so_list *head = NULL;
@@ -1322,6 +1322,52 @@ svr4_current_sos (void)
return head;
}
+/* Filter out vDSO module, if present. Its symbol file would not be
+ found on disk. Such OBJFILE would come from add_vsyscall_page. */
+
+static struct so_list *
+svr4_current_sos (void)
+{
+ struct so_list *so_head = svr4_current_sos_1 ();
+ struct objfile *objfile;
+ struct obj_section *osect;
+
+ ALL_OBJSECTIONS (objfile, osect)
+ {
+ const bfd *abfd = objfile->obfd;
+ const struct bfd_section *sect;
+ const char *name;
+ CORE_ADDR sect_vma;
+ struct so_list **sop;
+
+ if ((bfd_get_file_flags (abfd) & BFD_IN_MEMORY) == 0)
+ continue;
+
+ sect = osect->the_bfd_section;
+ name = bfd_get_section_name (abfd, sect);
+ if (name == NULL || strcmp (name, ".dynamic") != 0)
+ continue;
+
+ sect_vma = (bfd_get_section_vma (abfd, sect)
+ + ANOFFSET (objfile->section_offsets, sect->index));
+ sop = &so_head;
+ while (*sop)
+ {
+ struct so_list *so = *sop;
+
+ if (so->lm_info->l_ld == sect_vma)
+ {
+ *sop = so->next;
+ free_so (so);
+ }
+ else
+ sop = &so->next;
+ }
+ }
+
+ return so_head;
+}
+
/* Get the address of the link_map for a given OBJFILE. */
CORE_ADDR
diff --git a/gdb/testsuite/gdb.base/vdso-warning.c b/gdb/testsuite/gdb.base/vdso-warning.c
new file mode 100644
index 0000000..ef89fef
--- /dev/null
+++ b/gdb/testsuite/gdb.base/vdso-warning.c
@@ -0,0 +1,22 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+ Copyright 2012 Free Software Foundation, Inc.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+int
+main (void)
+{
+ return 0;
+}
diff --git a/gdb/testsuite/gdb.base/vdso-warning.exp b/gdb/testsuite/gdb.base/vdso-warning.exp
new file mode 100644
index 0000000..506b376
--- /dev/null
+++ b/gdb/testsuite/gdb.base/vdso-warning.exp
@@ -0,0 +1,43 @@
+# Copyright 2012 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+standard_testfile
+
+# There's no support for passing environment variables in the remote protocol.
+if { [is_remote target] } {
+ return 0
+}
+
+if { [prepare_for_testing ${testfile}.exp ${testfile} $srcfile] } {
+ return -1
+}
+
+gdb_breakpoint "main"
+
+# Fedora/RHEL glibc does not have this problem reproducible without this line.
+# See PR libc/13097 Comment 2.
+gdb_test_no_output "set environment LD_DEBUG=unused"
+
+gdb_run_cmd
+
+set test "stopped"
+gdb_test_multiple "" $test {
+ -re "Could not load shared library symbols .*\r\n$gdb_prompt $" {
+ fail $test
+ }
+ -re "\r\nBreakpoint \[0-9\]+, main .*\r\n$gdb_prompt $" {
+ pass $test
+ }
+}
prev parent reply other threads:[~2012-11-25 18:15 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-22 20:17 Jan Kratochvil
2012-11-23 11:30 ` Luis Gustavo
2012-11-23 11:41 ` Jan Kratochvil
2012-11-23 11:59 ` Luis Machado
2012-11-23 12:03 ` Pedro Alves
2012-11-23 12:40 ` Jan Kratochvil
2012-11-23 14:05 ` Luis Machado
2012-11-23 14:07 ` Jan Kratochvil
2012-11-23 16:01 ` Pedro Alves
2012-11-23 16:12 ` Mark Kettenis
2012-11-23 16:17 ` Jan Kratochvil
2012-11-23 16:23 ` Pedro Alves
2012-11-23 17:28 ` Joel Brobecker
2012-11-23 18:17 ` Jan Kratochvil
2012-11-23 18:22 ` Pedro Alves
2012-11-27 21:58 ` Mike Frysinger
2012-11-27 23:00 ` Matt Rice
2012-11-28 1:12 ` Pedro Alves
2012-11-28 20:39 ` Mike Frysinger
2012-11-28 22:44 ` Jan Kratochvil
2012-11-29 1:07 ` Mike Frysinger
2012-11-23 12:43 ` Mark Kettenis
2012-11-23 16:30 ` H.J. Lu
2012-11-23 18:19 ` Pedro Alves
2012-11-23 18:43 ` Jan Kratochvil
2012-11-25 18:15 ` Jan Kratochvil [this message]
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=20121125181505.GA26194@host2.jankratochvil.net \
--to=jan.kratochvil@redhat.com \
--cc=gdb-patches@sourceware.org \
--cc=hjl.tools@gmail.com \
--cc=palves@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