rpms/gdb
3
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Backport "Exclude debuginfo files from 'outside ELF segments' warning".

Browse Source 
(Keith Seitz, RH BZ 1898252)
Backport "Fix crash when expanding partial symtab..."
  (Tom Tromey. gdb/27743)
Backport "[gdb/server] Don't overwrite fs/gs_base with -m32"
(Tom de Vries)
This commit is contained in:
Kevin Buettner 2021-06-14 14:51:25 -07:00
parent 506c836503
commit 8f22c5cbd3
7 changed files with 514 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
_gdb.spec.Patch.include
View File

@ -372,3 +372,15 @@ Patch089: gdb-rhbz1964167-fortran-array-slices-at-prompt.patch
# problem in gdb/f-lang.c.
Patch090: gdb-rhbz1964167-fortran-fix-type-format-mismatch-in-f-lang.c.patch
# Backport of "Exclude debuginfo files from 'outside of ELF segments'
# warning" (Keith Seitz)
Patch091: gdb-rhbz1898252-loadable-section-outside-ELF-segments.patch
# Backport "Fix crash when expanding partial symtabs with DW_TAG_imported_unit"
# (Tom Tromey, gdb/27743)
Patch092: gdb-gdb27743-psymtab-imported-unit.patch
# Backport "[gdb/server] Don't overwrite fs/gs_base with -m32"
# (Tom de Vries)
Patch093: gdb-dont-overwrite-fsgsbase-m32.patch

3
_gdb.spec.patch.include
View File

@ -88,3 +88,6 @@
%patch088 -p1
%patch089 -p1
%patch090 -p1
%patch091 -p1
%patch092 -p1
%patch093 -p1

3
_patch_order
View File

@ -88,3 +88,6 @@ gdb-rhbz1964167-fortran-range_type-to-range_flag.patch
gdb-rhbz1964167-fortran-array-strides-in-expressions.patch
gdb-rhbz1964167-fortran-array-slices-at-prompt.patch
gdb-rhbz1964167-fortran-fix-type-format-mismatch-in-f-lang.c.patch
gdb-rhbz1898252-loadable-section-outside-ELF-segments.patch
gdb-gdb27743-psymtab-imported-unit.patch
gdb-dont-overwrite-fsgsbase-m32.patch

139
gdb-dont-overwrite-fsgsbase-m32.patch Normal file
View File

@ -0,0 +1,139 @@
From FEDORA_PATCHES Mon Sep 17 00:00:00 2001
From: Tom de Vries <tdevries@suse.de>
Date: Tue, 1 Jun 2021 10:14:31 -0700
Subject: gdb-dont-overwrite-fsgsbase-m32.patch
;; Backport "[gdb/server] Don't overwrite fs/gs_base with -m32"
;; (Tom de Vries)
Consider a minimal test-case test.c:
...
int main (void) { return 0; }
...
compiled with -m32:
...
$ gcc test.c -m32
...
When running the exec using gdbserver on openSUSE Factory (currently running a
linux kernel version 5.10.5):
...
$ gdbserver localhost:12345 a.out
...
to which we connect in a gdb session, we run into a segfault in the inferior:
...
$ gdb -batch -q -ex "target remote localhost:12345" -ex continue
Program received signal SIGSEGV, Segmentation fault.
0xf7dd8bd2 in init_cacheinfo () at ../sysdeps/x86/cacheinfo.c:761
...
The segfault is caused by gdbserver overwriting $gs_base with 0 using
PTRACE_SETREGS. After it is overwritten, the next use of $gs in the inferior
will trigger the segfault.
Before linux kernel version 5.9, the value used by PTRACE_SETREGS for $gs_base
was ignored, but starting version 5.9, the linux kernel has support for
intel architecture extension FSGSBASE, which allows users to modify $gs_base,
and consequently PTRACE_SETREGS can no longer ignore the $gs_base value.
The overwrite of $gs_base with 0 is done by a memset in x86_fill_gregset,
which was added in commit 9e0aa64f551 "Fix gdbserver qGetTLSAddr for
x86_64 -m32". The memset intends to zero-extend 32-bit registers that are
tracked in the regcache to 64-bit when writing them into the PTRACE_SETREGS
data argument. But in addition, it overwrites other registers that are
not tracked in the regcache, such as $gs_base.
Fix the segfault by redoing the fix from commit 9e0aa64f551 in minimal form.
Tested on x86_64-linux:
- openSUSE Leap 15.2 (using kernel version 5.3.18):
- native
- gdbserver -m32
- -m32
- openSUSE Factory (using kernel version 5.10.5):
- native
- m32
gdbserver/ChangeLog:
2021-01-20 Tom de Vries <tdevries@suse.de>
* linux-x86-low.cc (collect_register_i386): New function.
(x86_fill_gregset): Remove memset. Use collect_register_i386.
diff --git a/gdbserver/linux-x86-low.cc b/gdbserver/linux-x86-low.cc
--- a/gdbserver/linux-x86-low.cc
+++ b/gdbserver/linux-x86-low.cc
@@ -397,6 +397,35 @@ x86_target::low_cannot_fetch_register (int regno)
return regno >= I386_NUM_REGS;
}
+static void
+collect_register_i386 (struct regcache *regcache, int regno, void *buf)
+{
+ collect_register (regcache, regno, buf);
+
+#ifdef __x86_64__
+ /* In case of x86_64 -m32, collect_register only writes 4 bytes, but the
+ space reserved in buf for the register is 8 bytes. Make sure the entire
+ reserved space is initialized. */
+
+ gdb_assert (register_size (regcache->tdesc, regno) == 4);
+
+ if (regno == RAX)
+ {
+ /* Sign extend EAX value to avoid potential syscall restart
+ problems.
+
+ See amd64_linux_collect_native_gregset() in
+ gdb/amd64-linux-nat.c for a detailed explanation. */
+ *(int64_t *) buf = *(int32_t *) buf;
+ }
+ else
+ {
+ /* Zero-extend. */
+ *(uint64_t *) buf = *(uint32_t *) buf;
+ }
+#endif
+}
+
static void
x86_fill_gregset (struct regcache *regcache, void *buf)
{
@@ -411,32 +440,14 @@ x86_fill_gregset (struct regcache *regcache, void *buf)
return;
}
-
- /* 32-bit inferior registers need to be zero-extended.
- Callers would read uninitialized memory otherwise. */
- memset (buf, 0x00, X86_64_USER_REGS * 8);
#endif
for (i = 0; i < I386_NUM_REGS; i++)
- collect_register (regcache, i, ((char *) buf) + i386_regmap[i]);
-
- collect_register_by_name (regcache, "orig_eax",
- ((char *) buf) + ORIG_EAX * REGSIZE);
+ collect_register_i386 (regcache, i, ((char *) buf) + i386_regmap[i]);
-#ifdef __x86_64__
- /* Sign extend EAX value to avoid potential syscall restart
- problems.
-
- See amd64_linux_collect_native_gregset() in gdb/amd64-linux-nat.c
- for a detailed explanation. */
- if (register_size (regcache->tdesc, 0) == 4)
- {
- void *ptr = ((gdb_byte *) buf
- + i386_regmap[find_regno (regcache->tdesc, "eax")]);
-
- *(int64_t *) ptr = *(int32_t *) ptr;
- }
-#endif
+ /* Handle ORIG_EAX, which is not in i386_regmap. */
+ collect_register_i386 (regcache, find_regno (regcache->tdesc, "orig_eax"),
+ ((char *) buf) + ORIG_EAX * REGSIZE);
}
static void

281
gdb-gdb27743-psymtab-imported-unit.patch Normal file
View File

@ -0,0 +1,281 @@
From FEDORA_PATCHES Mon Sep 17 00:00:00 2001
From: Tom Tromey <tromey@adacore.com>
Date: Fri, 23 Apr 2021 11:28:48 -0600
Subject: gdb-gdb27743-psymtab-imported-unit.patch
;; Backport "Fix crash when expanding partial symtabs with DW_TAG_imported_unit"
;; (Tom Tromey, gdb/27743)
From e7d77ce0c408e7019f9885b8be64c9cdb46dd312 Mon Sep 17 00:00:00 2001
Subject: [PATCH] Fix crash when expanding partial symtabs with
DW_TAG_imported_unit
PR gdb/27743 points out a gdb crash when expanding partial symtabs,
where one of the compilation units uses DW_TAG_imported_unit.
The bug is that partial_map_expand_apply expects only to be called for
the outermost psymtab. However, filename searching doesn't (and
probably shouldn't) guarantee this. The fix is to walk upward to find
the outermost CU.
A new test case is included. It is mostly copied from other test
cases, which really sped up the effort.
This bug does not occur on trunk. There,
psym_map_symtabs_matching_filename is gone, replaced by
psymbol_functions::expand_symtabs_matching. When this find a match,
it calls psymtab_to_symtab, which does this same upward walk.
Tested on x86-64 Fedora 32.
I propose checking in this patch on the gdb-10 branch, and just the
new test case on trunk.
gdb/ChangeLog
2021-04-23 Tom Tromey <tromey@adacore.com>
PR gdb/27743:
* psymtab.c (partial_map_expand_apply): Expand outermost psymtab.
gdb/testsuite/ChangeLog
2021-04-23 Tom Tromey <tromey@adacore.com>
PR gdb/27743:
* gdb.dwarf2/imported-unit-bp.exp: New file.
* gdb.dwarf2/imported-unit-bp-main.c: New file.
* gdb.dwarf2/imported-unit-bp-alt.c: New file.
diff --git a/gdb/psymtab.c b/gdb/psymtab.c
--- a/gdb/psymtab.c
+++ b/gdb/psymtab.c
@@ -127,9 +127,10 @@ partial_map_expand_apply (struct objfile *objfile,
{
struct compunit_symtab *last_made = objfile->compunit_symtabs;
- /* Shared psymtabs should never be seen here. Instead they should
- be handled properly by the caller. */
- gdb_assert (pst->user == NULL);
+ /* We may see a shared psymtab here, but we want to expand the
+ outermost symtab. */
+ while (pst->user != nullptr)
+ pst = pst->user;
/* Don't visit already-expanded psymtabs. */
if (pst->readin_p (objfile))
diff --git a/gdb/testsuite/gdb.dwarf2/imported-unit-bp-alt.c b/gdb/testsuite/gdb.dwarf2/imported-unit-bp-alt.c
new file mode 100644
--- /dev/null
+++ b/gdb/testsuite/gdb.dwarf2/imported-unit-bp-alt.c
@@ -0,0 +1,50 @@
+/* Copyright 2020-2021 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/>. */
+
+/* Used to insert labels with which we can build a fake line table. */
+#define LL(N) asm ("line_label_" #N ": .globl line_label_" #N)
+
+volatile int var;
+volatile int bar;
+
+/* Generate some code to take up some space. */
+#define FILLER do { \
+ var = 99; \
+} while (0)
+
+int
+func (void)
+{ /* func prologue */
+ asm ("func_label: .globl func_label");
+ LL (1); // F1, Ln 16
+ FILLER;
+ LL (2); // F1, Ln 17
+ FILLER;
+ LL (3); // F2, Ln 21
+ FILLER;
+ LL (4); // F2, Ln 22 // F1, Ln 18, !S
+ FILLER;
+ LL (5); // F1, Ln 19 !S
+ FILLER;
+ LL (6); // F1, Ln 20
+ FILLER;
+ LL (7);
+ FILLER;
+ return 0; /* func end */
+}
+
+#ifdef WITHMAIN
+int main () { return 0; }
+#endif
diff --git a/gdb/testsuite/gdb.dwarf2/imported-unit-bp-main.c b/gdb/testsuite/gdb.dwarf2/imported-unit-bp-main.c
new file mode 100644
--- /dev/null
+++ b/gdb/testsuite/gdb.dwarf2/imported-unit-bp-main.c
@@ -0,0 +1,24 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+ Copyright 2004-2021 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/>. */
+
+extern int func (void);
+
+int
+main()
+{
+ return func ();
+}
diff --git a/gdb/testsuite/gdb.dwarf2/imported-unit-bp.exp b/gdb/testsuite/gdb.dwarf2/imported-unit-bp.exp
new file mode 100644
--- /dev/null
+++ b/gdb/testsuite/gdb.dwarf2/imported-unit-bp.exp
@@ -0,0 +1,128 @@
+# Copyright 2020-2021 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/>.
+
+# Test that "break /absolute/file:line" works ok with imported CUs.
+
+load_lib dwarf.exp
+
+# This test can only be run on targets which support DWARF-2 and use gas.
+if {![dwarf2_support]} {
+ return 0
+}
+
+# The .c files use __attribute__.
+if [get_compiler_info] {
+ return -1
+}
+if !$gcc_compiled {
+ return 0
+}
+
+standard_testfile imported-unit-bp-alt.c .S imported-unit-bp-main.c
+
+set build_options {nodebug optimize=-O1}
+
+set asm_file [standard_output_file $srcfile2]
+Dwarf::assemble $asm_file {
+ global srcdir subdir srcfile srcfile
+ global build_options
+ declare_labels lines_label callee_subprog_label cu_label
+
+ get_func_info func "$build_options additional_flags=-DWITHMAIN"
+
+ cu {} {
+ compile_unit {
+ {language @DW_LANG_C}
+ {name "<artificial>"}
+ } {
+ imported_unit {
+ {import %$cu_label}
+ }
+ }
+ }
+
+ cu {} {
+ cu_label: compile_unit {
+ {producer "gcc"}
+ {language @DW_LANG_C}
+ {name ${srcfile}}
+ {comp_dir "/tmp"}
+ {low_pc 0 addr}
+ {stmt_list ${lines_label} DW_FORM_sec_offset}
+ } {
+ callee_subprog_label: subprogram {
+ {external 1 flag}
+ {name callee}
+ {inline 3 data1}
+ }
+ subprogram {
+ {external 1 flag}
+ {name func}
+ {low_pc $func_start addr}
+ {high_pc "$func_start + $func_len" addr}
+ } {
+ }
+ }
+ }
+
+ lines {version 2 default_is_stmt 1} lines_label {
+ include_dir "/tmp"
+ file_name "$srcfile" 1
+
+ program {
+ {DW_LNE_set_address line_label_1}
+ {DW_LNS_advance_line 15}
+ {DW_LNS_copy}
+
+ {DW_LNE_set_address line_label_2}
+ {DW_LNS_advance_line 1}
+ {DW_LNS_copy}
+
+ {DW_LNE_set_address line_label_3}
+ {DW_LNS_advance_line 4}
+ {DW_LNS_copy}
+
+ {DW_LNE_set_address line_label_4}
+ {DW_LNS_advance_line 1}
+ {DW_LNS_copy}
+
+ {DW_LNS_advance_line -4}
+ {DW_LNS_negate_stmt}
+ {DW_LNS_copy}
+
+ {DW_LNE_set_address line_label_5}
+ {DW_LNS_advance_line 1}
+ {DW_LNS_copy}
+
+ {DW_LNE_set_address line_label_6}
+ {DW_LNS_advance_line 1}
+ {DW_LNS_negate_stmt}
+ {DW_LNS_copy}
+
+ {DW_LNE_set_address line_label_7}
+ {DW_LNE_end_sequence}
+ }
+ }
+}
+
+if { [prepare_for_testing "failed to prepare" ${testfile} \
+ [list $srcfile $asm_file $srcfile3] $build_options] } {
+ return -1
+}
+
+gdb_reinitialize_dir /tmp
+
+# Using an absolute path is important to see the bug.
+gdb_test "break /tmp/${srcfile}:19" "Breakpoint .* file $srcfile, line .*"

67
gdb-rhbz1898252-loadable-section-outside-ELF-segments.patch Normal file
View File

@ -0,0 +1,67 @@
From FEDORA_PATCHES Mon Sep 17 00:00:00 2001
From: Keith Seitz <keiths@redhat.com>
Date: Mon, 16 Nov 2020 12:42:09 -0500
Subject: gdb-rhbz1898252-loadable-section-outside-ELF-segments.patch
;; Backport of "Exclude debuginfo files from 'outside of ELF segments'
;; warning" (Keith Seitz)
Exclude debuginfo files from "outside of ELF segments" warning
When GDB loads an ELF file, it will warn when a section is not located
in an ELF segment:
$ ./gdb -q -iex "set build-id-verbose 0" --ex "b systemctl_main" -ex "r" -batch --args systemctl kexec
Breakpoint 1 at 0xc24d: file ../src/systemctl/systemctl.c, line 8752.
warning: Loadable section ".note.gnu.property" outside of ELF segments
in .gnu_debugdata for /lib64/libgcc_s.so.1
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
warning: Loadable section ".note.gnu.property" outside of ELF segments
in .gnu_debugdata for /lib64/libcap.so.2
warning: Loadable section ".note.gnu.property" outside of ELF segments
in .gnu_debugdata for /lib64/libacl.so.1
warning: Loadable section ".note.gnu.property" outside of ELF segments
in .gnu_debugdata for /lib64/libcryptsetup.so.12
warning: Loadable section ".note.gnu.property" outside of ELF segments
in .gnu_debugdata for /lib64/libgcrypt.so.20
warning: Loadable section ".note.gnu.property" outside of ELF segments
in .gnu_debugdata for /lib64/libip4tc.so.2
[snip]
This has feature has also been reported by various users, most notably
the Fedora-EOL'd bug 1553086.
Mark Wielaard explains the issue quite nicely in
https://sourceware.org/bugzilla/show_bug.cgi?id=24717#c2
The short of it is, the ELF program headers for debuginfo files are
not suited to this particular use case. Consequently, the warning
generated above really is useless and should be ignored.
This patch follows the same heuristic that BFD itself uses.
gdb/ChangeLog
2020-11-13 Keith Seitz <keiths@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1553086
* elfread.c (elf_symfile_segments): Omit "Loadable section ...
outside of ELF segments" warning for debugin
diff --git a/gdb/elfread.c b/gdb/elfread.c
--- a/gdb/elfread.c
+++ b/gdb/elfread.c
@@ -147,7 +147,12 @@ elf_symfile_segments (bfd *abfd)
RealView) use SHT_NOBITS for uninitialized data. Since it is
uninitialized, it doesn't need a program header. Such
binaries are not relocatable. */
- if (bfd_section_size (sect) > 0 && j == num_segments
+
+ /* Exclude debuginfo files from this warning, too, since those
+ are often not strictly compliant with the standard. See, e.g.,
+ ld/24717 for more discussion. */
+ if (!is_debuginfo_file (abfd)
+ && bfd_section_size (sect) > 0 && j == num_segments
&& (bfd_section_flags (sect) & SEC_LOAD) != 0)
warning (_("Loadable section \"%s\" outside of ELF segments"),
bfd_section_name (sect));

10
gdb.spec
View File

@ -37,7 +37,7 @@ Version: 10.2
# The release always contains a leading reserved number, start it at 1.
# `upstream' is not a part of `name' to stay fully rpm dependencies compatible for the testing.
Release: 2%{?dist}
Release: 3%{?dist}
License: GPLv3+ and GPLv3+ with exceptions and GPLv2+ and GPLv2+ with exceptions and GPL+ and LGPLv2+ and LGPLv3+ and BSD and Public Domain and GFDL
# Do not provide URL for snapshots as the file lasts there only for 2 days.
@ -1154,6 +1154,14 @@ fi
%endif
%changelog
* Fri Jun 11 2021 Keith Seitz <keiths@redhat.com> - 10.2-3
- Backport "Exclude debuginfo files from 'outside ELF segments' warning".
(Keith Seitz, RH BZ 1898252)
- Backport "Fix crash when expanding partial symtab..."
(Tom Tromey. gdb/27743)
- Backport "[gdb/server] Don't overwrite fs/gs_base with -m32"
- (Tom de Vries)
* Sun Jun 06 2021 Python Maint <python-maint@redhat.com>
- Rebuilt for Python 3.10