http://sourceware.org/ml/gdb-patches/2009-11/msg00388.html
http://sourceware.org/ml/gdb-cvs/2009-11/msg00156.html
Subject: [patch] Fix crash on reading stabs
Hi,
there is a crash on reading stabs fpc binary:
https://bugzilla.redhat.com/show_bug.cgi?id=537837
Program received signal SIGSEGV, Segmentation fault.
0x000000000069db3d in read_dbx_symtab (objfile=0x1daf5f0) at dbxread.c:1369
1369 if ((namestring[0] == '-' && namestring[1] == 'l')
(gdb) p/x nlist.n_strx
$7 = 0xfffffff8
(gdb) p sizeof(nlist.n_strx)
$10 = 8
Below the patch context is:
namestring = (nlist->n_strx + file_string_table_offset
+ DBX_STRINGTAB (objfile));
so IMO the `(unsigned)' cast is excessive as it does not match the expression
below. Such cast is there since the GDB "Initial revision" (1999).
`n_strx' type:
struct internal_nlist
{
unsigned long n_strx; /* Index into string table of name. */
...
};
Regression tested on {x86_64,x86_64-m32,i686}-fedora12-linux-gnu which does not
mean anything with the default DWARF debug info. It was hanging for stabs so
tried just a large part of gdb.base/*.exp on x86_64-m32 - `unix/-gstabs+/-m32'.
If it isn't obviously approved please feel free to drop it as one should not
use STABS in the first place.
Regards,
Jan
gdb/
2009-11-17 Jan Kratochvil <jan.kratochvil@redhat.com>
* dbxread.c (set_namestring): Remove cast to unsigned. Check N_STRX
overflow.
--- a/gdb/dbxread.c
+++ b/gdb/dbxread.c
@@ -965,8 +965,9 @@ set_namestring (struct objfile *objfile, const struct internal_nlist *nlist)
{
char *namestring;
- if (((unsigned) nlist->n_strx + file_string_table_offset)
- >= DBX_STRINGTAB_SIZE (objfile))
+ if (nlist->n_strx + file_string_table_offset
+ >= DBX_STRINGTAB_SIZE (objfile)
+ || nlist->n_strx + file_string_table_offset < nlist->n_strx)
{
complaint (&symfile_complaints, _("bad string table offset in symbol %d"),
symnum);
Re: [RFA] Fix "show convenience" test
http://sourceware.org/ml/gdb-patches/2009-09/msg00565.html
http://sourceware.org/ml/gdb-cvs/2009-09/msg00099.html
### src/gdb/testsuite/ChangeLog 2009/09/15 18:51:25 1.1960
### src/gdb/testsuite/ChangeLog 2009/09/17 17:49:46 1.1961
## -1,3 +1,7 @@
+2009-09-17 Paul Pluzhnikov <ppluzhnikov@google.com>
+
+ * gdb.base/default.exp: Fix "show convenience".
+
2009-09-15 Tom Tromey <tromey@redhat.com>
* lib/mi-support.exp (mi_create_varobj): Update.
--- src/gdb/testsuite/gdb.base/default.exp 2009/08/13 14:58:27 1.31
+++ src/gdb/testsuite/gdb.base/default.exp 2009/09/17 17:49:46 1.32
@@ -598,7 +598,7 @@
#test show confirm
gdb_test "show confirm" "Whether to confirm potentially dangerous operations is o\[a-z\]*." "show confirm"
#test show convenience
-gdb_test "show convenience" "No debugger convenience variables now defined.(\[^\r\n\]*\[\r\n\])+Convenience variables have names starting with \".\";(\[^\r\n\]*\[\r\n\])+use \"set\" as in \"set .foo = 5\" to define them." "show convenience"
+gdb_test "show convenience" "\\\$_siginfo = void" "show convenience"
#test show directories
gdb_test "show directories" "Source directories searched: .cdir\[:;\].cwd" "show directories"
#test show editing
### src/gdb/doc/ChangeLog 2009/09/15 18:51:24 1.953
### src/gdb/doc/ChangeLog 2009/09/17 17:49:46 1.954
## -1,3 +1,8 @@
+2009-09-17 Paul Pluzhnikov <ppluzhnikov@google.com>
+
+ * gdb.texinfo (convenince variables): Mention
+ $_siginfo could be empty.
+
2009-09-15 Tom Tromey <tromey@redhat.com>
* gdb.texinfo (GDB/MI Variable Objects): Document
--- src/gdb/doc/gdb.texinfo 2009/09/15 18:51:25 1.624
+++ src/gdb/doc/gdb.texinfo 2009/09/17 17:49:46 1.625
@@ -7819,8 +7819,10 @@
@item $_siginfo
@vindex $_siginfo@r{, convenience variable}
-The variable @code{$_siginfo} is bound to extra signal information
-inspection (@pxref{extra signal information}).
+The variable @code{$_siginfo} contains extra signal information
+(@pxref{extra signal information}). Note that @code{$_siginfo}
+could be empty, if the application has not yet received any signals.
+For example, it will be empty before you execute the @code{run} command.
@end table
On HP-UX systems, if you refer to a function or variable name that
http://sourceware.org/ml/gdb-cvs/2009-12/msg00128.html
### src/gdb/testsuite/ChangeLog 2009/12/23 23:18:08 1.2054
### src/gdb/testsuite/ChangeLog 2009/12/24 21:57:06 1.2055
## -1,3 +1,10 @@
+2009-12-24 Jan Kratochvil <jan.kratochvil@redhat.com>
+
+ Fix compatibility with G++-4.5.
+ * gdb.cp/expand-sals.cc (main): Remove the "exit-line" comment.
+ * gdb.cp/expand-sals.exp: Remove breakpoint on "exit-line".
+ (uncaught return): Remove.
+
2009-12-23 Jan Kratochvil <jan.kratochvil@redhat.com>
Phil Muldoon <pmuldoon@redhat.com>
--- src/gdb/testsuite/gdb.cp/expand-sals.cc 2009/05/11 15:05:56 1.1
+++ src/gdb/testsuite/gdb.cp/expand-sals.cc 2009/12/24 21:57:06 1.2
@@ -49,5 +49,5 @@
A a;
B b;
- return 0; /* exit-line */
+ return 0;
}
--- src/gdb/testsuite/gdb.cp/expand-sals.exp 2009/05/11 15:05:56 1.1
+++ src/gdb/testsuite/gdb.cp/expand-sals.exp 2009/12/24 21:57:06 1.2
@@ -23,8 +23,6 @@
return -1
}
-gdb_breakpoint [gdb_get_line_number "exit-line"]
-
gdb_breakpoint [gdb_get_line_number "func-line"]
gdb_continue_to_breakpoint "func" ".*func-line.*"
@@ -52,7 +50,3 @@
"bt from A"
gdb_continue_to_breakpoint "next caller func" ".*func-line.*"
-
-# Verify GDB really could not catch any other breakpoint location.
-
-gdb_continue_to_breakpoint "uncaught return" ".*exit-line.*"
http://sourceware.org/ml/gdb-cvs/2009-11/msg00213.html
[ cut ]
--- src/gdb/testsuite/gdb.base/condbreak.exp 2009/01/03 05:58:03 1.13
+++ src/gdb/testsuite/gdb.base/condbreak.exp 2009/11/25 20:43:29 1.14
@@ -207,10 +207,10 @@
setup_xfail hppa2.0w-*-* 11512CLLbs
send_gdb "continue\n"
gdb_expect {
- -re "Continuing\\..*Breakpoint \[0-9\]+, marker2 \\(a=43\\) at .*$srcfile1:($bp_location8|$bp_location9).*($bp_location8|$bp_location9)\[\t \]+.*" {
+ -re "Continuing\\..*Breakpoint \[0-9\]+, marker2 \\(a=43\\) at .*$srcfile1:($bp_location8|$bp_location9).*($bp_location8|$bp_location9)\[\t \]+.*$gdb_prompt $" {
pass "run until breakpoint at marker2"
}
- -re "Continuing\\..*Breakpoint \[0-9\]+, $hex in marker2 \\(a=43\\) at .*$srcfile1:($bp_location8|$bp_location9).*($bp_location8|$bp_location9)\[\t \]+.*" {
+ -re "Continuing\\..*Breakpoint \[0-9\]+, $hex in marker2 \\(a=43\\) at .*$srcfile1:($bp_location8|$bp_location9).*($bp_location8|$bp_location9)\[\t \]+.*$gdb_prompt $" {
xfail "run until breakpoint at marker2"
}
-re "$gdb_prompt $" {
Fix for gfortran-4.1:
-PASS: gdb.mi/mi-var-child-f.exp: mi runto MAIN__
+FAIL: gdb.mi/mi-var-child-f.exp: mi runto MAIN__ (unknown output after running)
RFC: Move language-changed message to verbose
http://sourceware.org/ml/gdb-patches/2009-11/msg00031.html
http://sourceware.org/ml/gdb-cvs/2009-11/msg00034.html
### src/gdb/ChangeLog 2009/11/05 19:53:03 1.11030
### src/gdb/ChangeLog 2009/11/05 20:43:52 1.11031
## -1,3 +1,8 @@
+2009-11-05 Daniel Jacobowitz <dan@codesourcery.com>
+
+ * top.c (execute_command): Select a frame before checking the current
+ language. Only output a message if verbose.
+
2009-11-05 Tom Tromey <tromey@redhat.com>
* symtab.h (SYMBOL_SET_LINKAGE_NAME): Update comment.
--- src/gdb/top.c 2009/10/19 09:51:42 1.172
+++ src/gdb/top.c 2009/11/05 20:43:52 1.173
@@ -457,10 +457,13 @@
}
- /* Tell the user if the language has changed (except first time). */
+ /* Tell the user if the language has changed (except first time).
+ First make sure that a new frame has been selected, in case this
+ command or the hooks changed the program state. */
+ deprecated_safe_get_selected_frame ();
if (current_language != expected_language)
{
- if (language_mode == language_mode_auto)
+ if (language_mode == language_mode_auto && info_verbose)
{
language_info (1); /* Print what changed. */
}