It makes no sense to talk about an "unlimited" address size in this
context.
(gdb) show remoteaddresssize
The maximum size of the address (in bits) in a memory packet is 0.
(gdb) set remoteaddresssize 0
(gdb) show remoteaddresssize
The maximum size of the address (in bits) in a memory packet is unlimited.
"set remoteaddresssize 0" mapping to UINT_MAX means you can't
force gdb through this path twice in the same GDB run:
static CORE_ADDR
remote_address_masked (CORE_ADDR addr)
{
unsigned int address_size = remote_address_size;
/* If "remoteaddresssize" was not set, default to target address size. */
if (!address_size)
address_size = gdbarch_addr_bit (target_gdbarch ());
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* remote.c (_initialize_remote): Make "set remoteaddresssize"
a zuinteger command instead of uinteger.
The "set record full insn-number-max" command is an uinteger command.
If the variable that holds the maximum count of logged instructions is
unsigned, it's better if the variable that holds the current number of
logged instructions is also unsigned. Looking over the code, there's
no case the variable could end up negative.
Then, tests like "if (record_full_insn_max_num)" are always true,
because being a uinteger command means that "set record full
insn-number-max 0" is actually mapped to UINT_MAX internally. IOW,
the command's variable is never 0. The checks might make some sense
if 0 wasn't mapped to UINT_MAX, and 0 meant unlimited, but, that's not
how things work.
Tested on x86_64 Fedora 17.
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* record-full.c (record_full_insn_num): Make it unsigned.
(record_full_check_insn_num, record_full_message)
(record_full_registers_change, record_full_xfer_partial): Remove
record_full_insn_max_num check (it's always != 0).
(record_full_info, record_full_restore): Use %u as format string.
(): Use %u as format string.
(set_record_full_insn_max_num): Remove record_full_insn_max_num
check (it's always != 0).
It doesn't make sense to request an "unlimited" dcache. You want to
configure the cache with specific lines and length of lines.
It doesn't actually work anyway:
(gdb) set dcache line-size 0
Invalid dcache line size: 4294967295 (must be power of 2).
(gdb) set dcache size 0
(gdb) show dcache size
Number of dcache lines is unlimited.
(gdb) info dcache
Dcache 4294967295 lines of 64 bytes each.
No data cache available.
The code already has guards in place to forbid 0s:
static void
set_dcache_size (char *args, int from_tty,
struct cmd_list_element *c)
{
if (dcache_size == 0)
{
dcache_size = DCACHE_DEFAULT_SIZE;
error (_("Dcache size must be greater than 0."));
}
if (last_cache)
dcache_invalidate (last_cache);
}
static void
set_dcache_line_size (char *args, int from_tty,
struct cmd_list_element *c)
{
if (dcache_line_size < 2
|| (dcache_line_size & (dcache_line_size - 1)) != 0)
{
unsigned d = dcache_line_size;
dcache_line_size = DCACHE_DEFAULT_LINE_SIZE;
error (_("Invalid dcache line size: %u (must be power of 2)."), d);
}
if (last_cache)
dcache_invalidate (last_cache);
}
So we now get:
(gdb) set dcache line-size 0
Invalid dcache line size: 0 (must be power of 2).
(gdb) set dcache size 0
Dcache size must be greater than 0.
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* dcache.c (_initialize_dcache): Make the "set dcache line-size"
and "set dcache size" commands zuinteger instead of uinteger.
Being a uinteger means you revert back to having GDB decide the
version. It makes no sense to have an "unlimited" version.
(gdb) show cris-version
The current CRIS version is 0.
(gdb) set cris-version 0
(gdb) show cris-version
The current CRIS version is unlimited.
(gdb)
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* cris-tdep.c (_initialize_cris_tdep): Make the "set cris-version"
command zuinteger instead of uinteger.
Being a uinteger means you can't disable debug output after enabling it...
(gdb) show debug coff_pe_read
Coff PE read debugging is 0.
(gdb) set debug coff_pe_read 0
(gdb) show debug coff_pe_read
Coff PE read debugging is unlimited.
(gdb)
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* coff-pe-read.c (_initialize_coff_pe_read): Make the command
zuinteger instead of uinteger.
While the commands are uinteger, the target interfaces are limited to
INT_MAX. Don't let the user request more than we can handle.
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* record.c (record_insn_history_size_setshow_var)
(record_call_history_size_setshow_var): New globals.
(command_size_to_target_size): New function.
(cmd_record_insn_history, cmd_record_call_history): Use
command_size_to_target_size instead of cast.
(validate_history_size, set_record_insn_history_size)
(set_record_call_history_size): New functions.
(_initialize_record): Install set_record_insn_history_size and
set_record_call_history_size as "set" hooks of "set record
instruction-history-size" and "set record
function-call-history-size".
Ref: http://www.sourceware.org/ml/gdb-patches/2002-08/msg00486.html
We've long since imported a newer readline, no need to use the old
compatibility variable anymore.
Tested on x86_64 Fedora 17.
gdb/
2013-03-26 Pedro Alves <palves@redhat.com>
* top.c (gdb_rl_operate_and_get_next): Replace max_input_history
use with history_max_entries use. Remove FIXME note.
Reading symbols from /bin/true...(no debugging symbols found)...done.
(gdb) b _start
Function "_start" not defined.
Make breakpoint pending on future shared library load? (y or [n]) y
Breakpoint 1 (_start) pending.
(gdb) r
Starting program: /bin/true
Breakpoint 1, 0x00000039a0400af0 in _start () from /lib64/ld-linux-x86-64.so.2
(gdb) rec b
(gdb) r
The program being debugged has been started already.
Start it from the beginning? (y or n) y
Starting program: /bin/true
Breakpoint 1, 0x00000039a0400af0 in _start () from /lib64/ld-linux-x86-64.so.2
(gdb) rec b
gdb/record-btrace.c:154: internal-error: record_btrace_open:
Assertion `record_btrace_thread_observer == NULL' failed.
A problem internal to GDB has been detected,
further debugging may prove unreliable.
Quit this debugging session? (y or n)
gdb/
* record-btrace.c (record_btrace_close): Call
record_btrace_auto_disable.
testsuite/
* gdb.btrace/enable.exp: Add regression test.
* c-exp.y (exp): Add new productions for destructors after '.' and
'->'.
(write_destructor_name): New function.
gdb/testsuite
* gdb.cp/m-static.exp: Add destructor-printing tests.
windows-nat.c (windows_get_absolute_argv0): New function.
windows-nat.h: Add its prototype.
main.c (get_init_files): Use filename_ncmp instead of strncmp.
Use IS_DIR_SEPARATOR instead of looking for a character inside
SLASH_STRING. Include filenames.h.
(captured_main) [__MINGW32__]: Make argv[0] absolute, so that
relocate_gdb_directory works when passed gdb_program_name.
Include windows-nat.h.
* exceptions.h (enum errors): New entry TARGET_CLOSE_ERROR.
* remote.c (trace_error): Remove the special handling of '2'.
(readchar) <SERIAL_EOF>
(readchar) <SERIAL_ERROR>
(getpkt_or_notif_sane_1): Use TARGET_CLOSE_ERROR for them.
(remote_get_trace_status): Call throw_exception if EX is
TARGET_CLOSE_ERROR.
* utils.c (perror_with_name): Rename to ...
(throw_perror_with_name): ... here. New parameter errcode, describe it
in the function comment.
(perror_with_name): New function wrapper.
* utils.h (enum errors): New stub declaration.
(throw_perror_with_name): New declaration.
gdb/testsuite/
* gdb.server/server-kill.c: New file.
* gdb.server/server-kill.exp: New file.
The range validation added by
http://sourceware.org/ml/gdb-patches/2013-03/msg00767.html
Changes things to allow setting the command to INT_MAX or UINT_MAX
directly, with signed and unsigned commands respectively. However,
that went a little bit too far, as in the cases of var_integer and
var_uinteger, those values are actually implementation detail. It's
better to not expose them in the interface, and have users assume
those values mean "unlimited" too, so to be safer to expand the range
of the commands in the future if we want to. Yes, it's pedantic, and
it's not likely users actually will do this, but MI frontends and
Python scripts might.
gdb/
2013-03-22 Pedro Alves <palves@redhat.com>
Yao Qi <yao@codesourcery.com>
Mark Kettenis <kettenis@gnu.org>
* cli/cli-setshow.c (do_set_command) <var_uinteger>:
Don't let the user set the value to UINT_MAX directly.
<var_integer>: Don't let the user set the value to INT_MAX
directly.
The range validation added by
http://sourceware.org/ml/gdb-patches/2013-03/msg00767.html
Changes things to allow setting the command to INT_MAX or UINT_MAX
directly, with signed and unsigned commands respectively. However,
that went a little bit too far, as in the cases of var_integer and
var_uinteger, those values are actually implementation detail. It's
better to not expose them in the interface, and have users assume
those values mean "unlimited" too, so to be safer to expand the range
of the commands in the future if we want to. Yes, it's pedantic, and
it's not likely users actually will do this, but MI frontends and
Python scripts might.
gdb/
2013-03-22 Pedro Alves <palves@redhat.com>
Yao Qi <yao@codesourcery.com>
Mark Kettenis <kettenis@gnu.org>
* cli/cli-setshow.c (do_set_command) <var_uinteger>:
Don't let the user set the value to UINT_MAX directly.
<var_integer>: Don't let the user set the value to INT_MAX
directly.
Oleg Nesterov told me that the Linux kernel copies the parent's ptrace
options to fork/clone children, so there's no need for GDB to do that
manually.
I was actually a bit surprised, since I thought the ptracer had to
always set the ptrace options itself, and GDB is indeed calling
PTRACE_SETOPTIONS for each new fork child, if it'll stay attached.
Looking at the history of that code, I found that is was actually I
who added that set-ptrace-options-in-children bit, back in
http://sourceware.org/ml/gdb-patches/2009-05/msg00656.html. But,
honestly, I don't recall why I needed that. I think I may have just
blindly believed it was necessary.
I then looked back at the history of all the PTRACE_SETOPTIONS code we
have, and found that gdb never did copy the ptrace options before my
patch. But, when gdbserver learnt to use PTRACE_EVENT_CLONE, at
http://sourceware.org/ml/gdb-patches/2007-10/msg00547.html, it was
made to do 'ptrace (PTRACE_SETOPTIONS, new_pid, 0,
PTRACE_O_TRACECLONE)' for all new clones. Hmmm. But, GDB itself
never did that, so it can't really ever have been necessary, I
believe, otherwise GDB should have been doing it too.
(GDBserver doesn't support following forks, and so naturally doesn't
do any PTRACE_SETOPTIONS on fork children.)
So this patch removes the -I believe- unnecessary ptrace syscalls.
Tested on x86_64 Fedora 17, native/gdbserver, and on x86_64 RHEL5
native/gdbserver (Linux 2.6.18, I think a ptrace-on-utrace kernel).
No regressions.
gdb/
2013-03-22 Pedro Alves <palves@redhat.com>
* linux-nat.c (linux_child_follow_fork): Don't call
linux_enable_event_reporting.
(linux_handle_extended_wait): Don't call
linux_enable_event_reporting.
gdb/gdbserver/
2013-03-22 Pedro Alves <palves@redhat.com>
* linux-low.c (handle_extended_wait): Don't call
linux_enable_event_reporting.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" hppa-hpux-tdep.o -k 2>&1 1>/dev/null
../../src/gdb/hppa-hpux-tdep.c: In function ‘hppa_hpux_push_dummy_code’:
../../src/gdb/hppa-hpux-tdep.c:1225:7: error: pointer targets in passing argument 2 of ‘write_memory’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/hppa-hpux-tdep.c:22:0:
../../src/gdb/gdbcore.h:85:13: note: expected ‘const gdb_byte *’ but argument is of type ‘char *’
../../src/gdb/hppa-hpux-tdep.c:1251:7: error: pointer targets in passing argument 2 of ‘write_memory’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/hppa-hpux-tdep.c:22:0:
../../src/gdb/gdbcore.h:85:13: note: expected ‘const gdb_byte *’ but argument is of type ‘char *’
../../src/gdb/hppa-hpux-tdep.c: In function ‘hppa_hpux_supply_save_state’:
../../src/gdb/hppa-hpux-tdep.c:1354:9: error: pointer targets in passing argument 1 of ‘extract_unsigned_integer’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/hppa-hpux-tdep.c:20:0:
../../src/gdb/defs.h:675:22: note: expected ‘const gdb_byte *’ but argument is of type ‘const char *’
Casting to gdb_byte would fix it, however, writing an
unsigned int array like this
static unsigned int hppa64_tramp[] = {
0xeac0f000, /* bve,l (r22),%r2 */
0x0fdf12d1, /* std r31,-8(,sp) */
0x0fd110c2, /* ldd -8(,sp),rp */
0xe840d002, /* bve,n (rp) */
0x08000240 /* nop */
...
directly to target memory assumes the host endianness is the same as
the target's. hppa is big endian, so I believe this patch should be
correct -- it defines the array as a gdb_byte array. It uses a macro
to make the insn bytes a little more readable. I thought of using
write_memory_unsigned_integer once for each element of the unsigned
int array, but this way keeps issuing a single target memory write /
roundtrip for the whole trampoline.
gdb/
2013-03-22 Pedro Alves <palves@redhat.com>
* hppa-hpux-tdep.c (hppa_hpux_push_dummy_code): Define INSN macro,
use it to rewrite the trampoline buffers with type gdb_byte[], and
undefine the macro. Remove char* cast.
* NEWS: Add entry.
* event-top.c: #include "maint.h".
* main.c: #include "maint.h".
* maint.c: #include <sys/time.h>, <time.h>, block.h, top.h,
timeval-utils.h, maint.h, cli/cli-setshow.h.
(per_command_time, per_command_space): New static globals.
(per_command_symtab): New static global.
(per_command_setlist, per_command_showlist): New static globals.
(struct cmd_stats): Move here from utils.c.
(set_per_command_time): Renamed from set_display_time in utils.c
and moved here. All callers updated.
(set_per_command_space): Renamed from set_display_space in utils.c
and moved here. All callers updated.
(count_symtabs_and_blocks): New function.
(report_command_stats): Moved here from utils.c. Add support for
printing symtab stats. Only print data if enabled before command
executed.
(make_command_stats_cleanup): Ditto.
(sert_per_command_cmd, show_per_command_cmd): New functions.
(_initialize_maint_cmds): Add new commands
mt set per-command {space,time,symtab} {on,off}.
* maint.h: New file.
* top.c: #include "maint.h".
* utils.c (reset_prompt_for_continue_wait_time): New function.
(get_prompt_for_continue_wait_time): New function.
* utils.h (reset_prompt_for_continue_wait_time): Declare
(get_prompt_for_continue_wait_time): Declare.
(make_command_stats_cleanup): Moved to maint.h.
(set_display_time, set_display_space): Moved to maint.h and renamed
to set_per_command_time, set_per_command_space.
* cli/cli-setshow.c (parse_cli_boolean_value): Renamed from
parse_binary_operation and made non-static. Don't call error,
just return an error marker. All callers updated.
* cli/cli-setshow.h (parse_cli_boolean_value): Declare.
doc/
* gdb.texinfo (Maintenance Commands): Add docs for
"mt set per-command {space,time,symtab} {on,off}".
testsuite/
* gdb.base/maint.exp: Update tests for per-command stats.
* c-exp.y (yylex): Rewrite to push all tokens onto the FIFO.
Handle FILENAME token.
gdb/testsuite
* gdb.cp/cpexprs.exp: Add test for FILENAME:: case.
* gdb.cp/misc.exp: Add test for FILENAME:: case.
This is a regression from 7.5, introduced/exposed by:
http://sourceware.org/ml/gdb-patches/2012-07/msg00259.html
There are a series of issues with this code.
It does:
unsigned int val = parse_and_eval_long (arg);
^^^^^^^^^^^^
(unsigned, usually 32-bit) while parse_and_eval_long returns a LONGEST
(usually 64-bit), so we lose precision without noticing:
(gdb) set remote hardware-watchpoint-limit 0x100000000
(gdb) show remote hardware-watchpoint-limit 0x100000000
The maximum number of target hardware watchpoints is 0.
While at it, print the invalid number with plongest, so the user sees
what GDB thought the number was:
(gdb) set remote hardware-watchpoint-limit 0x100000000
integer 4294967296 out of range
So with "set remote hardware-watchpoint-limit -1", val ends converted
to 0xffffffff, which then fails the
else if (val >= INT_MAX)
error (_("integer %u out of range"), val);
test.
Looking at that INT_MAX check, we forbid INT_MAX itself, but we
shouldn't, as that does fit in 'int' -- we want to forbid values
_greater_ than INT_MAX (and less than INT_MIN, while at it):
(gdb) set remote hardware-watchpoint-limit 2147483647
integer 2147483647 out of range
The same problem is in the new var_zuinteger_unlimited code, which
also uses "int" for variable.
Also, when printing a 'signed int', we should use %d, not %u.
This adds a couple regression tests. Not completely thorough in checking
all kinds of invalid input; I'm saving more exaustive testing around
zXXinteger commands for something like new test-assisting commands
like "maint test cmd-zinteger -1", where testing would focus on the
command types, and thus be independent of particular user commands of
particular GDB features.
Tested on x86_64 Fedora 17.
gdb/
2013-03-20 Pedro Alves <palves@redhat.com>
PR gdb/15289
* cli/cli-setshow.c (do_set_command)
<var_uinteger, var_zuinteger>: Use LONGEST for variable holding
the result of parsing the command argument. Throw error if the
value is greater than UINT_MAX. Print the invalid value with
plongest.
<var_integer, var_zinteger>: Use LONGEST for variable holding the
result of parsing the command argument. Throw error if the value
is greater than INT_MAX, not greater or equal. Also throw error
if the value is less than INT_MIN. Print the invalid value with
plongest.
<var_zuinteger_unlimited>: Throw error if the value is greater
than INT_MAX, not greater or equal.
(do_show_command) <var_integer, var_zinteger,
var_zuinteger_unlimited>: Use %d for printing int, not %u.
gdb/testsuite/
2013-03-20 Pedro Alves <palves@redhat.com>
PR gdb/15289
* gdb.base/remote.exp: Test
"set remote hardware-watchpoint-limit -1",
"set remote hardware-breakpoint-limit -1",
"set remote hardware-watchpoint-limit 2147483647" and
"set remote hardware-breakpoint-limit 2147483647".
gdb/python/python-internal.h (HAVE_SNPRINTF)
[_WIN32 && HAVE_DECL_SNPRINTF]: Define, to avoid compiler warnings
about redefinition of snprintf by pyerrors.h.
* dwarf2read.c (dw2_map_symtabs_matching_filename): Put continue after
any successful compare_filenames_for_search or FILENAME_CMP.
* psymtab.c (partial_map_symtabs_matching_filename): Likewise.
* symtab.c (iterate_over_some_symtabs): Likewise.
Subject: [PATCH] Fix for PR c++/15203 and PR c++/15210
Date: Sat, 09 Mar 2013 02:50:49 -0300 (5 days, 4 hours, 57 minutes ago)
Message-ID: <m3a9qdnmti.fsf@redhat.com>
Hi,
This bug was reported internally at our Bugzilla, along with a proposed
fix. After talking to Keith about it, he investigated and came up with
another patch needed to really fix the issue on CVS HEAD.
The first part of the fix is the patch to cp-namespace.c. It handles
the case when we are accessing a static variable inside a function
(inside a class) by the full linespec (is it right, Keith?). E.g.:
class foo
{
public:
int bar()
{
static int var = 0;
}
};
And then, printing the value of `var':
(gdb) print 'foo::bar()::var'
GDB would fall in an internal_error:
gdb/cp-namespace.c:816: internal-error: cp_lookup_nested_symbol called on a non-aggregate type.
This is because `cp_lookup_nested_symbol' is not handling the case when
TYPE_CODE is either _FUNC or _METHOD. This patch fixes it by returning
NULL in this case.
The second part of the fix is the patch to elfread.c. It is needed
because the BSF_GNU_UNIQUE flag was added to some symbols in
<http://sourceware.org/ml/binutils/2009-06/msg00016.html>. Because of
that, (still) the command:
(gdb) print 'foo::bar()::var'
where `var' is a static variable returns:
"No symbol "foo::bar()::var" in current context."
So with the second patch applied the command finally DTRT:
(gdb) print 'foo::bar()::var'
$1 = 0
This may not be the ideal solution, according to Keith it would be good
to implement productions on c-exp.y in order to recognize
CLASS::FUNCTION::VARIABLE, but it is a solution which works with what we
have today.
I regtested it in Fedora 17 x86_64 with -m64 and -m32, including
gdbserver, without regressions.
gdb/:
2013-03-14 Keith Seitz <keiths@redhat.com>
Alan Matsuoka <alanm@redhat.com>
PR c++/15203
PR c++/15210
* cp-namespace.c (cp_lookup_nested_symbol): Handle TYPE_CODE_FUNC and
TYPE_CODE_METHOD.
* elfread.c (elf_symtab_read): Handle BSF_GNU_UNIQUE for certain
symbols.
gdb/testsuite/:
2013-03-14 Sergio Durigan Junior <sergiodj@redhat.com>
PR c++/15203
PR c++/15210
* gdb.cp/m-static.cc (keepalive_int): New function.
(gnu_obj_1::method): New variable `sintvar', call `keepalive_int'.
* gdb.cp/m-static.exp: New test for `sintvar'.
As mentioned in the previous patch, I grepped for "\*\*) &" and found
one hit in completer.c.
I was about to post a patch that simply made
current_demangling_style_string const, and cast away constness at the
xfree site. However, looking deeper, it seem to be there's a lot of
dead code in the file.
First, all external callers of set_demangling_style are found in the
stabs reader, commented out for over 12 years:
http://sourceware.org/ml/gdb-patches/2000-12/msg00214.html
I don't think it's likely we'll ever make the older mangling schemes
work for stabs. If we do, we can rediscuss the approach then.
Then, set_demangling_command has special handling for unknown
demangling styles, but "set demangle-style" is an enum command, and
with those, the user can only specify a known enumeration value, by
design:
(gdb) set demangle-style gangnam-style
Undefined item: "gangnam-style".
This patch removes all that dead code, then makes
current_demangling_style_string point to an element of
demangling_style_names, as the FIXME suggests, and then makes
current_demangling_style_string, removing the need for the 'const char
**' cast.
gdb/
2013-03-13 Pedro Alves <palves@redhat.com>
* dbxread.c (read_ofile_symtab, process_one_symbol): Remove
commented out code.
* demangle.c (current_demangling_style_string): Make it const.
(set_demangling_command): Assert the demangling style is known.
Remove all handling of unknown styles. Set
'current_demangling_style_string' to an element of the
demangling_style_names array.
(set_demangling_style): Delete.
(_initialize_demangler): Set current_demangling_style_string to the
element of the demangling_style_names array that corresponds to
the default demangling style. Remove FIXME note. Don't call
set_demangling_style.
* gdb-demangle.h (set_demangling_style): Remove declaration.
This is sort of a continuation of Keith's parse_exp_1 constification
patch. It started out by undoing these bits:
@@ -754,9 +754,12 @@ validate_actionline (char **line, struct
tmp_p = p;
for (loc = t->base.loc; loc; loc = loc->next)
{
- p = tmp_p;
- exp = parse_exp_1 (&p, loc->address,
+ const char *q;
+
+ q = tmp_p;
+ exp = parse_exp_1 (&q, loc->address,
block_for_pc (loc->address), 1);
+ p = (char *) q;
and progressively making more things const upwards, fixing fallout,
rinse repeat, until GDB built again (--enable-targets=all).
That ended up constifying lookup_cmd/add_cmd and (lots of) friends,
and the completers.
I didn't try to constify the command hooks themselves, because I know
upfront there are commands that write to the command string argument,
and I think I managed to stop at a nice non-hacky split point already.
I think the only non-really-super-obvious changes are
tracepoint.c:validate_actionline, and tracepoint.c:trace_dump_actions.
The rest is just mostly about 'char *' => 'const char *', 'char **'=>
'const char **', and the occasional (e.g., deprecated_cmd_warning)
case of 'char **'=> 'const char *', where/when I noticed that nothing
actually cares about the pointer to pointer output.
Tested on x86_64 Fedora 17, native and gdbserver.
gdb/
2013-03-13 Pedro Alves <palves@redhat.com>
* ada-lang.c (struct add_partial_datum) <text, text0, word>: Make
fields const.
(ada_make_symbol_completion_list): Make "text0" parameter const.
* ax-gdb.c (agent_eval_command_one): Make "exp" parameter const.
* breakpoint.c (condition_completer): Make "text" and "word"
parameters const. Adjust.
(check_tracepoint_command): Adjust to validate_actionline
prototype change.
(catch_syscall_completer): Make "text" and "word" parameters
const.
* cli/cli-cmds.c (show_user): Make "comname" local const.
(valid_command_p): Make "command" parameter const.
(alias_command): Make "alias_prefix" and "command_prefix" locals
const.
* cli/cli-decode.c (add_cmd): Make "name" parameter const.
(add_alias_cmd): Make "name" and "oldname" parameters const.
Adjust. No longer make copy of OLDNAME.
(add_prefix_cmd, add_abbrev_prefix_cmd, add_set_or_show_cmd)
(add_setshow_cmd_full, add_setshow_enum_cmd)
(add_setshow_auto_boolean_cmd, add_setshow_boolean_cmd)
(add_setshow_filename_cmd, add_setshow_string_cmd)
(add_setshow_string_noescape_cmd)
(add_setshow_optional_filename_cmd, add_setshow_integer_cmd)
(add_setshow_uinteger_cmd, add_setshow_zinteger_cmd)
(add_setshow_zuinteger_unlimited_cmd, add_setshow_zuinteger_cmd)
(delete_cmd, add_info, add_info_alias, add_com, add_com_alias):
Make "name" parameter const.
(help_cmd): Rename "command" parameter to "arg". New const local
"command".
(find_cmd): Make "command" parameter const.
(lookup_cmd_1): Make "text" parameter pointer to const. Adjust to
deprecated_cmd_warning prototype change.
(undef_cmd_error): Make "cmdtype" parameter const.
(lookup_cmd): Make "line" parameter const.
(deprecated_cmd_warning): Change type of "text" parameter to
pointer to const char, from pointer to pointer to char. Adjust.
(lookup_cmd_composition): Make "text" parameter const.
(complete_on_cmdlist, complete_on_enum): Make "text" and "word"
parameters const.
* cli/cli-decode.h (struct cmd_list_element) <name>: Make field
const.
* cli/cli-script.c (validate_comname): Make "tem" local const.
(define_command): New const local "tem_c". Use it in calls to
lookup_cmd.
(document_command): Make "tem" and "comfull" locals const.
(show_user_1): Make "prefix" and "name" parameters const.
* cli-script.h (show_user_1): Make "prefix" and "name" parameters
const.
* command.h (add_cmd, add_alias_cmd, add_prefix_cmd)
(add_abbrev_prefix_cmd, completer_ftype, lookup_cmd, lookup_cmd_1)
(deprecated_cmd_warning, lookup_cmd_composition, add_com)
(add_com_alias, add_info, add_info_alias, complete_on_cmdlist)
(complete_on_enum, add_setshow_enum_cmd)
(add_setshow_auto_boolean_cmd, add_setshow_boolean_cmd)
(add_setshow_filename_cmd, add_setshow_string_cmd)
(add_setshow_string_noescape_cmd)
(add_setshow_optional_filename_cmd, add_setshow_integer_cmd)
(add_setshow_uinteger_cmd, add_setshow_zinteger_cmd)
(add_setshow_zuinteger_cmd, add_setshow_zuinteger_unlimited_cmd):
Change prototypes, constifying strings.
* completer.c (noop_completer, filename_completer): Make "text"
and "prefix" parameters const.
(location_completer, expression_completer)
(complete_line_internal): Make "text" and "prefix" parameters
const and adjust.
(command_completer, signal_completer): Make "text" and "prefix"
parameters const.
* completer.h (noop_completer, filename_completer)
(expression_completer, location_completer, command_completer)
(signal_completer): Change prototypes.
* corefile.c (complete_set_gnutarget): Make "text" and "word"
parameters const.
* cp-abi.c (cp_abi_completer): Likewise.
* expression.h (parse_expression_for_completion): Change
prototype.
* f-lang.c (f_make_symbol_completion_list): Make "text" and "word"
parameters const.
* infcmd.c (_initialize_infcmd): Make "cmd_name" local const.
* infrun.c (handle_completer): Make "text" and "word" parameters
const.
* interps.c (interpreter_completer): Make "text" and "word"
parameters const.
* language.h (struct language_defn)
<la_make_symbol_completion_list>: Make "text" and "word"
parameters const.
* parse.c (parse_exp_1): Move const hack to parse_exp_in_context.
(parse_exp_in_context): Rename to ...
(parse_exp_in_context_1): ... this.
(parse_exp_in_context): Reimplement, with const hack from
parse_exp_1.
(parse_expression_for_completion): Make "string" parameter const.
* printcmd.c (decode_format): Make "string_ptr" parameter pointer
to pointer to const char. Adjust.
(print_command_1): Make "exp" parameter const.
(output_command): Rename to ...
(output_command_const): ... this. Make "exp" parameter const.
(output_command): Reimplement.
(x_command): Adjust.
(display_command): Rename "exp" parameter to "arg". New "exp"
local, const version of "arg".
* python/py-auto-load.c (gdbpy_initialize_auto_load): Make
"cmd_name" local const.
* python/py-cmd.c (cmdpy_destroyer): Cast const away in xfree
call.
(cmdpy_completer): Make "text" and "word" parameters const.
(gdbpy_parse_command_name): Make "prefix_text2" local const.
* python/py-param.c (add_setshow_generic): Make "tmp_name" local
const.
* remote.c (_initialize_remote): Make "cmd_name" local const.
* symtab.c (language_search_unquoted_string): Make "text" and "p"
parameters const. Adjust.
(completion_list_add_fields): Make "sym_text", "text" and "word"
parameters const.
(struct add_name_data) <sym_text, text, word>: Make fields const.
(default_make_symbol_completion_list_break_on): Make "text" and
"word" parameters const. Adjust locals.
(default_make_symbol_completion_list)
(make_symbol_completion_list, make_symbol_completion_type)
(make_symbol_completion_list_fn): Make "text" and "word"
parameters const.
(make_file_symbol_completion_list): Make "text", "word" and
"srcfile" parameters const. Adjust locals.
(add_filename_to_list): Make "text" and "word" parameters const.
(struct add_partial_filename_data) <text, word>: Make fields
const.
(make_source_files_completion_list): Make "text" and "word"
parameters const.
* symtab.h (default_make_symbol_completion_list_break_on)
(default_make_symbol_completion_list, make_symbol_completion_list)
(make_symbol_completion_type enum type_code)
(make_symbol_completion_list_fn make_file_symbol_completion_list)
(make_source_files_completion_list): Change prototype.
* top.c (execute_command): Adjust to pass pointer to pointer to
const char to lookup_cmd, and to deprecated_cmd_warning prototype
change.
(set_verbose): Make "cmdname" local const.
* tracepoint.c (decode_agent_options): Make "exp" parameter const,
and adjust.
(validate_actionline): Make "line" parameter a pointer to const
char, and adjust.
(encode_actions_1): Make "action_exp" local const, and adjust.
(encode_actions): Adjust.
(replace_comma): Delete.
(trace_dump_actions): Make "action_exp" and "next_comma" locals
const, and adjust. Don't frob the action string while splitting
it at commas. Instead, make a copy of each split substring in
turn.
(trace_dump_command): Adjust to validate_actionline prototype
change.
* tracepoint.h (decode_agent_options, decode_agent_options)
(encode_actions, validate_actionline): Change prototypes.
* valprint.h (output_command): Delete declaration.
(output_command_const): Declare.
* value.c (function_destroyer): Cast const away in xfree call.
As a follow up to:
http://sourceware.org/ml/gdb-patches/2013-03/msg00449.html
In a nutshell, casts between 'char **' <-> 'unsigned char **' and
'char **' <-> 'const char **' are invalid.
I grepped for "\*\*) &" and found these. There's another one in
demangle.c, but I've split fixing that one to a separate patch.
I think the ada_decode_symbol change is perhaps the one that could be
surprising. The function's description has this comment, which makes
things much clearer:
The GSYMBOL parameter is "mutable" in the C++ sense: logically
const, but nevertheless modified to a semantically equivalent form
when a decoded name is cached in it. */
const char *
ada_decode_symbol (const struct general_symbol_info *gsymbol)
With that out of the way, I think the patch ends up being pretty
obvious.
Tested on x86_64 Fedora 17.
gdb/
2013-03-13 Pedro Alves <palves@redhat.com>
* ada-lang.c (ada_decode_symbol): Cast away constness of GSYMBOL
rather than casting 'const char * const *' to 'const char **'.
* ada-lex.l (processInt): Make "trailer" local const. Remove
'const char **' cast.
* arm-linux-tdep.c (arm_stap_parse_special_token): Add 'char *'
locals, and use those as strtol output pointer, instead than doing
invalid casts to from 'const char **' to 'char **'.
(_initialize_demangle): Remove cast.
* i386-tdep.c (i386_stap_parse_special_token): : Add 'char *'
locals, and use those as strtol output pointer, instead than doing
invalid casts to from 'const char **' to 'char **'.
* solib-dsbt.c (dsbt_get_initial_loadmaps): Remove 'gdb_byte**'
casts.
* stap-probe.c (stap_parse_register_operand)
(stap_parse_single_operand): Likewise.
the last matched 'V' blcok in trace frame.
gdb/gdbserver:
* tracepoint.c (traceframe_read_tsv): Look for the last matched
'V' block in trace frame.
gdb/testsuite:
* gdb.trace/tsv.exp (check_tsv): New.
(top level): Save a tfile on current trace session. Call
check_tsv on live target. Load the tfile with target tfile
and call check_tsv again.
* mi/mi-interp.c (mi_interpreter_exec): Make "command" const.
Remove temporary copy of input string.
(mi_execute_command_wrapper): Make "cmd" const.
* mi/mi-main.c (mi_execute_command): Make "string_ptr" const.
* mi/mi-parse.c (mi_parse_argv): Make "args" const.
Use const strings.
(mi_parse): Make "cmd" const.
Use const strings.
* mi/mi-parse.h (mi_parse): Make "cmd" const.
pointer to expression string to parse_exp_1.
(create_excep_cond_exprs): Likewise.
* ax-gdb.c (agent_eval_command_one): Likewise.
(maint_agent_printf_command): Likewise.
Constify much of the string handling/parsing.
* breakpoint.c (set_breakpoint_condition): Pass const
pointer to expression string to parse_exp_1.
(update_watchpoint): Likewise.
(parse_cmd_to_aexpr): Constify string handling.
Pass const pointer to parse_exp_1.
(init_breakpoint_sal): Pass const pointer to parse_exp_1.
(find_condition_and_thread): Likewise.
Make TOK const.
(watch_command_1): Make "arg" const.
Constify string handling.
Copy the expression string instead of changing the input
string.
(update_breakpoint_location): Pass const pointer to
parse_exp_1.
* eval.c (parse_and_eval_address): Make "exp" const.
(parse_to_comma_and_eval): Make "expp" const.
(parse_and_eval): Make "exp" const.
* expression.h (parse_expression): Make argument const.
(parse_exp_1): Make first argument const.
* findcmd.c (parse_find_args): Treat "args" as const.
* linespec.c (parse_linespec): Pass const pointer to
linespec_expression_to_pc.
(linespec_expression_to_pc): Make "exp_ptr" const.
* parse.c (parse_exp_1): Make "stringptr" const.
Make a copy of the expression to pass to parse_exp_in_context until
this whole interface can be constified.
(parse_expression): Make "string" const.
* printcmd.c (ui_printf): Treat "arg" as const.
Handle const strings.
* tracepoint.c (validate_actionline): Pass const pointer to
all calls to parse_exp_1.
(encode_actions_1): Likewise.
* value.h (parse_to_comma_and_eval): Make argument const.
(parse_and_eval_address): Likewise.
(parse_and_eval): Likewise.
* varobj.c (varobj_create): Pass const pointer to parse_exp_1.
(varobj_set_value): Likewise.
* cli/cli-cmds.c (disassemble_command): Treat "arg" as const and
constify string handling.
Pass const pointers to parse_and_eval_address and
parse_to_comman_and_eval.
* cli/cli-utils.c (skip_to_space): Rename to ...
(skip_to_space_const): ... this. Handle const strings.
* cli/cli-utils.h (skip_to_space): Turn into macro which invokes
skip_to_space_const.
(skip_to_space_const): Declare.
* common/format.c (parse_format_string): Make "arg" const.
Handle const strings.
* common/format.h (parse_format_string): Make "arg" const.
* gdbserver/ax.c (ax_printf): Make "format" const.
* python/python.c (gdbpy_parse_and_eval): Do not make a copy
of the expression string.
Hui Zhu <hui@codesourcery.com>
* dwarf2loc.c (access_memory): Change nbits to nbytes in gdb_assert.
(dwarf2_compile_expr_to_ax): Call access_memory in DW_OP_deref and
DW_OP_deref_size.
GDB treats the identifiers 'if', 'thread', and 'task' unconditionally
as expression delimiters in Ada mode, which is correct for 'if' and 'task',
but wrong for 'thread' in cases such as
print thread
Borrowing from c-exp.y, we observe that 'thread' must be followed by
numerals, whereas identifiers never are and treat them as delimiters
only in that case.
In the process, the current also refactors and incidentally fixes the
code for rewinding the input to before the delimiting tokens. For
example, the code
watch expr if i > 2
fails because the input is only rewound to just before the 'i',
leaving the 'if' as part of the expression (and thus making the
rest look like trailing junk rather than a conditional clause).
gdb/ChangeLog:
* ada-lex.l (rules): Only recognize 'thread' as a
delimiter when followed by numerals, as for c-exp.y.
Use new rewind_to_char function to rewind the input for
expression-delimiting tokens.
(rewind_to_char): New function.
gdb/testsuite/ChangeLog:
* gdb.ada/expr_delims.exp: New file.
* gdb.ada/expr_delims/foo.adb: New file.
* gdb.ada/expr_delims/pck.ads: New file.
* gdb.ada/expr_delims/pck.adb: New file.
Casts between 'char **' <-> 'unsigned char **' and 'char **' <-> const
char **' are actually invalid:
http://gcc.gnu.org/ml/gcc-help/2013-03/msg00118.html
In a nutshell, char (and variants) can alias anything, but pointers to
chars get no special treatment (cf. C99/N1256, 6.5/7).
Turns out older gcc's actually warn/complain on these constructs,
though newer one's don't:
http://sourceware.org/ml/gdb-patches/2013-03/msg00429.htmlhttp://sourceware.org/ml/gdb-patches/2013-03/msg00430.html
This patch fixes the cases I added last week. It also fixes one other
preexisting case in charset.c, though it seems even older gccs don't
complain of char * <-> const char * aliasing.
Tested on x86_64 Fedora 17.
gdb/
2013-03-11 Pedro Alves <palves@redhat.com>
* charset.c (convert_between_encodings): Don't cast between
different pointer to pointer types. Instead, make the 'inp' local
be of the type iconv expects.
(wchar_iterate): Don't cast between different pointer to pointer
types. Instead, use new pointer local of the type iconv expects.
* target.c (target_read_stralloc, target_fileio_read_stralloc):
Add new local of type char pointer, and use it to get a
char/string view of the byte buffer, instead of casting between
pointer to pointer types.
record-full.
Document two new record sub-commands "record instruction-history" and
"record function-call-history" and two associated set/show commands
"set record instruction-history-size" and "set record
function-call-history-size".
Add this to NEWS.
gdb/
* NEWS: Add record changes.
doc/
* gdb.texinfo (Process Record and Replay): Document record
changes.
instruction-history" command of record-btrace.
The pc prefix would appear multiple times in the branch trace disassembly,
which is more confusing than helpful.
gdb/
* record-btrace.c (btrace_insn_history): Omit the pc prefix in
the instruction history disassembly.
* disasm.c (dump_insns): Omit the pc prefix, if requested.
* disasm.h (DISASSEMBLY_OMIT_PC): New.
The target implements the new record sub-commands
"record instruction-history" and
"record function-call-history".
The target does not support reverse execution or navigation in the
recorded execution log.
gdb/
* Makefile.in (SFILES): Add record-btrace.c
(COMMON_OBS): Add record-btrace.o
* record-btrace.c: New.
* objfiles.c: Include btrace.h.
(free_objfile): call btrace_free_objfile.
This command provides a quick high-level overview over the recorded execution
log at function granularity without having to reverse-step.
gdb/
* target.c (target_call_history, target_call_history_from,
target_call_history_range): New.
* target.h (target_ops) <to_call_history, to_call_history_from,
to_call_history_range>: New fields.
(target_call_history, target_call_history_from,
target_call_history_range): New declaration.
* record.c (get_call_history_modifiers, cmd_record_call_history,
record_call_history_size): New.
(_initialize_record): Add the "record function-call-history" command.
Add "set/show record function-call-history-size" commands.
* record.h (record_print_flag): New.
between different record targets.
gdb/
* record.h (record_disconnect): New.
(record_detach): New.
(record_mourn_inferior): New.
(record_kill): New.
* record-full.c (record_disconnect, record_detach,
record_mourn_inferior, record_kill): Move to...
* record.c: ...here.
(DEBUG): New.
(record_stop): New.
(record_unpush): New.
(cmd_record_stop): Call record_stop. Replace unpush_target
call with record_unpush call.
(record_disconnect, record_detach): Assert that the target
is of record stratum. Call record_unpush, record_stop, and
DEBUG.
(record_mourn_inferior, record_kill): Assert that the target
is of record stratum. Call record_unpush and DEBUG.
gdb/
* record-full.h, record-full.c (record_memory_query): Rename
to ...
(record_full_memory_query): ...this. Update all users.
(record_arch_list_add_reg): Rename to ...
(record_full_arch_list_add_reg): ...this. Update all users.
(record_arch_list_add_mem): Rename to ...
(record_full_arch_list_add_mem): ...this. Update all users.
(record_arch_list_add_end): Rename to ...
(record_full_arch_list_add_end): ...this. Update all users.
(record_gdb_operation_disable_set): Rename to ...
(record_full_gdb_operation_disable_set): ...this.
Update all users.
EIST transition, T-states, C1E, or Adaptive Thermal Throttling (AAJ122).
This results in sporadic test fails. Disable btrace on those processors.
gdb/
* common/linux-btrace.c: Include sys/ptrace, sys/types, sys/wait.h,
and signal.h.
(linux_supports_btrace): Add kernel and
cpuid check.
(kernel_supports_btrace): New function.
(cpu_supports_btrace): New function.
(intel_supports_btrace): New function.
We define the following packets:
Qbtrace:bts enable branch tracing for the current thread
returns "OK" or "Enn"
Qbtrace:off disable branch tracing for the current thread
returns "OK" or "Enn"
qXfer:btrace:read read the full branch trace data for the current thread
gdb/
* target.h (enum target_object): Add TARGET_OBJECT_BTRACE.
* remote.c: Include btrace.h.
(struct btrace_target_info): New struct.
(remote_supports_btrace): New function.
(send_Qbtrace): New function.
(remote_enable_btrace): New function.
(remote_disable_btrace): New function.
(remote_teardown_btrace): New function.
(remote_read_btrace): New function.
(init_remote_ops): Add btrace ops.
(enum <unnamed>): Add btrace packets.
(struct protocol_feature remote_protocol_features[]): Add btrace packets.
(_initialize_remote): Add packet configuration for branch tracing.
gdbserver/
* target.h (struct target_ops): Add btrace ops.
(target_supports_btrace): New macro.
(target_enable_btrace): New macro.
(target_disable_btrace): New macro.
(target_read_btrace): New macro.
* gdbthread.h (struct thread_info): Add btrace field.
* server.c: Include btrace-common.h.
(handle_btrace_general_set): New function.
(handle_btrace_enable): New function.
(handle_btrace_disable): New function.
(handle_general_set): Call handle_btrace_general_set.
(handle_qxfer_btrace): New function.
(struct qxfer qxfer_packets[]): Add btrace entry.
* inferiors.c (remove_thread): Disable btrace.
* linux-low: Include linux-btrace.h.
(linux_low_enable_btrace): New function.
(linux_low_read_btrace): New function.
(linux_target_ops): Add btrace ops.
* configure.srv (i[34567]86-*-linux*): Add linux-btrace.o.
Add srv_linux_btrace=yes.
(x86_64-*-linux*): Add linux-btrace.o.
Add srv_linux_btrace=yes.
* configure.ac: Define HAVE_LINUX_BTRACE.
* config.in: Regenerated.
* configure: Regenerated.
Add a function to parse a btrace xml document into a vector of branch trace
blocks.
gdb/
* features/btrace.dtd: New file.
* Makefile.in (XMLFILES): Add btrace.dtd.
* btrace.h (parse_xml_btrace): New declaration.
* btrace.c: Include xml-support.h.
(parse_xml_btrace): New function.
(parse_xml_btrace_block): New function.
(block_attributes): New struct.
(btrace_attributes): New struct.
(btrace_children): New struct.
(btrace_elements): New struct.
* common/linux-ptrace.c (linux_ptrace_test_ret_to_nx): Call also kill
for CHILD, ignore PTRACE_KILL errors, move the inner block variable
kill_status to outer block.
Fix entry-values if the callee called a noreturn function.
* dwarf2-frame-tailcall.c (dwarf2_tailcall_sniffer_first): Use
get_frame_address_in_block. Add new comment.
gdb/testsuite/
Fix entry-values if the callee called a noreturn function.
* gdb.arch/amd64-tailcall-noret.S: New file.
* gdb.arch/amd64-tailcall-noret.c: New file.
* gdb.arch/amd64-tailcall-noret.exp: New file.
Fix entry-values in C++ across CUs.
* dwarf2loc.c (call_site_to_target_addr) <FIELD_LOC_KIND_PHYSNAME>: Use
lookup_minimal_symbol. Add a comment.
* dwarf2read.c
(read_call_site_scope) <is_ref_attr> <die_is_declaration>: Prefer
DW_AT_linkage_name.
gdb/testsuite/
Fix entry-values in C++ across CUs.
* gdb.arch/amd64-tailcall-cxx.exp: New file.
* gdb.arch/amd64-tailcall-cxx1.S: New file.
* gdb.arch/amd64-tailcall-cxx1.cc: New file.
* gdb.arch/amd64-tailcall-cxx2.S: New file.
* gdb.arch/amd64-tailcall-cxx2.cc: New file.
From: Pedro Alves <palves@redhat.com>
The find command's patter/buffer that is passed to the target is a
binary blob, not a string.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" findcmd.o -k 2>&1 1>/dev/null
../../src/gdb/findcmd.c: In function ‘find_command’:
../../src/gdb/findcmd.c:278:6: error: pointer targets in passing argument 3 of ‘target_search_memory’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/findcmd.c:26:0:
../../src/gdb/target.h:1582:12: note: expected ‘const gdb_byte *’ but argument is of type ‘char *’
gdb/
2013-03-08 Pedro Alves <palves@redhat.com>
* findcmd.c (put_bits): Change type of parameter to 'gdb_byte *'.
(parse_find_args, find_command): Change type of pattern buffer
locals to 'gdb_byte *'.
Hafiz Abid Qadeer <abidh@codesourcery.com>
gdb/
* NEWS: Mention set and show trace-buffer-size commands.
Mention new packet.
* target.h (struct target_ops): New method
to_set_trace_buffer_size.
(target_set_trace_buffer_size): New macro.
* target.c (update_current_target): Set up new method.
* tracepoint.c (trace_buffer_size): New global.
(start_tracing): Send it to the target.
(set_trace_buffer_size): New function.
(_initialize_tracepoint): Add new setshow for trace-buffer-size.
* remote.c (remote_set_trace_buffer_size): New function.
(_initialize_remote): Use it.
(QTBuffer:size) New remote command.
(PACKET_QTBuffer_size): New enum.
(remote_protocol_features): Add an entry for
PACKET_QTBuffer_size.
gdb/gdbserver/
* tracepoint.c (trace_buffer_size): New global.
(DEFAULT_TRACE_BUFFER_SIZE): New define.
(init_trace_buffer): Change to one-argument function. Allocate
trace buffer memory.
(handle_tracepoint_general_set): Call cmd_bigqtbuffer_size to
handle QTBuffer:size packet.
(cmd_bigqtbuffer_size): New function.
(initialize_tracepoint): Call init_trace_buffer with
DEFAULT_TRACE_BUFFER_SIZE.
* server.c (handle_query): Add QTBuffer:size in the
supported packets.
gdb/doc/
* gdb.texinfo (Starting and Stopping Trace Experiments): Document
trace-buffer-size set and show commands.
(Tracepoint Packets): Document QTBuffer:size.
(General Query Packets): Document QTBuffer:size.
gdb/testsuite/
* gdb.trace/trace-buffer-size.exp: New file.
* gdb.trace/trace-buffer-size.c: New file.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" target.o -k 2>&1 1>/dev/null
../../src/gdb/target.c: In function ‘target_read_stralloc’:
../../src/gdb/target.c:2376:3: error: pointer targets in passing argument 1 of ‘strlen’ differ in signedness [-Werror=pointer-sign]
In file included from build-gnulib/import/string.h:27:0,
from ../../src/gdb/common/gdb_string.h:24,
from ../../src/gdb/target.c:24:
/usr/include/string.h:399:15: note: expected ‘const char *’ but argument is of type ‘gdb_byte *’
...
This is about the same as the previous patch.
Functions that take or return ascii-ish string arguments usually use
char* for parameters/return. That means that at points we call into
target methods that work with binary blobs, we need casts to/from
gdb_byte*/char*. To choose which type for the variables, I usually go
based on which requires the fewer casts, and what the contents of the
variable are supposed to hold, which often gives the same answer.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* target.c (target_read_stralloc, target_fileio_read_alloc):
*Cast pointer to 'gdb_byte *' in target call.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" corefile.o -k 2>&1 1>/dev/null
../../src/gdb/corefile.c: In function ‘read_memory_string’:
../../src/gdb/corefile.c:334:7: error: pointer targets in passing argument 2 of ‘read_memory’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/corefile.c:217:1: note: expected ‘gdb_byte *’ but argument is of type ‘char *’
Functions that take or return ascii-ish string arguments usually use
char* for parameters/return. That means that at points we call into
target methods that work with binary blobs, we need casts to
gdb_byte*.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* corefile.c (read_memory_string): Cast pointer to gdb_byte* in
call.
(trace_pass_command): Likewise.
* cli/cli-cmds.c: Include cli/cli-utils.h.
(source_command): Use skip-spaces.
(disassemble_command): Likewise.
* findcmd.c: Include cli/cli-utils.h.
(parse_find_args): Use skip_spaces.
* go32-nat.c: Include cli/cli-utils.h.
(go32_sldt): Use skip_spaces.
(go32_sgdt): Likewise.
(go32_sidt): Likewise.
(go32_pde): Likewise.
(go32_pte): Likewise.
(go32_pte_for_address): Likewise.
* infcmd.c: Include cli/cli-utils.h.
(registers_info): Use skip_spaces.
* linux-tdep.c (read_mapping): Use skip_spaces_const.
(linux_info_proc): Likewise.
* linux-thread-db.c: Include cli/cli-utils.h.
(info_auto_load_libthread_db): Use skip_spaces_const.
* m32r-rom.c: Include cli/cli-utils.h.
(m32r_upload_command): Use skip_spaces.
* maint.c: Include cli/cli-utils.h.
(maintenance_translate_address): Use skip_spaces.
* mi/mi-parse.c: Include cli/cli-utils.h.
(mi_parse_argv): Use skip_spaces.
(mi_parse): Likewise.
* minsyms.c: Include cli/cli-utils.h.
(msymbol_hash_iw): Use skip_spaces_const.
* objc-lang.c: Include cli/cli-utils.h.
(parse_selector): Use skip_spaces.
(parse_method): Likewise.
* python/python.c: Include cli/cli-utils.h.
(python_interactive_command)[HAVE_PYTHON]: Use skip_spaces.
(python_command)[HAVE_PYTHON]: Likewise.
(python_interactive_command)[!HAVE_PYTHON]: Likewise.
* remote-m32r-sdi.c: Include cli/cli-utils.h.
(m32r_load): Use skip_spaces.
* serial.c: Include cli/cli-utils.h.
(serial_open): Use skip_spaces_const.
* stack.c: Include cli/cli-utils.h.
(parse_frame_specification_1): Use skip_spaces_const.
* symfile.c: Include cli/cli-utils.h.
(set_ext_lang_command): Use skip_spaces.
* symtab.c: Include cli/cli-utils.h.
(rbreak_command): Use skip_spaces.
* thread.c (thread_name_command): Use skip_spaces.
* tracepoint.c (validate_actionline): Use skip_spaces.
(encode_actions_1): Likewise.
(trace_find_range_command): Likewise.
(trace_find_outside_command): Likewise.
(trace_dump_actions): Likewise.
Trimmed for brevity:
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" c-lang.o expprint.o utils.o valprint.o varobj.o -k 2>&1 1>/dev/null
../../src/gdb/c-lang.c: In function ‘parse_one_string’:
../../src/gdb/c-lang.c:540:8: error: pointer targets in passing argument 3 of ‘convert_between_encodings’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/c-lang.c:30:0:
../../src/gdb/charset.h:64:6: note: expected ‘const gdb_byte *’ but argument is of type ‘char *’
../../src/gdb/expprint.c: In function ‘print_subexp_standard’:
../../src/gdb/expprint.c:205:2: error: pointer targets in passing argument 3 of ‘current_language->la_printstr’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/expprint.c:205:2: note: expected ‘const gdb_byte *’ but argument is of type ‘char *’
cc1: all warnings being treated as errors
make: *** [expprint.o] Error 1
../../src/gdb/utils.c: In function ‘host_char_to_target’:
../../src/gdb/utils.c:1474:9: error: pointer targets in passing argument 3 of ‘convert_between_encodings’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/varobj.c: In function ‘value_get_print_value’:
../../src/gdb/varobj.c:2934:8: error: pointer targets in return differ in signedness [-Werror=pointer-sign]
../../src/gdb/varobj.c:2968:12: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
../../src/gdb/varobj.c:2971:3: error: pointer targets in return differ in signedness [-Werror=pointer-sign]
cc1: all warnings being treated as errors
make: *** [varobj.o] Error 1
As with the previous patch, the encoding conversion code works with
gdb_byte arrays as the generic buffers that hold strings of any
encoding/width. This patch adds casts where appropriate.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* c-lang.c (parse_one_string): Cast argument to gdb_byte *.
* expprint.c (print_subexp_standard): Likewise.
* utils.c (host_char_to_target): Likewise.
* valprint.c (generic_emit_char, generic_printstr): Likewise.
* varobj.c (value_get_print_value): Change type of local to char*.
Cast it gdb_byte * in call to language printer.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" charset.o 2>&1 1>/dev/null
../../src/gdb/charset.c: In function ‘wchar_iterate’:
../../src/gdb/charset.c:665:13: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
../../src/gdb/charset.c:691:13: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
../../src/gdb/charset.c:706:12: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
The encoding conversion code works with gdb_byte arrays as the generic
buffers that hold strings of any encoding/width. Changing the type of
this field to gdb_byte* removes the need for one cast, and makes
everything work with the same types. That's good -- WRT to strings,
"char *" is (almost) consistently throughout GDB only used for
ascii-ish strings.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* charset.c (struct wchar_iterator) <input>: Change type to 'const
gdb_byte *'.
(make_wchar_iterator): Remove cast to char*.
(wchar_iterate): Change type of local.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" regcache.o 2>&1 1>/dev/null
../../src/gdb/regcache.c: In function ‘regcache_xmalloc_1’:
../../src/gdb/regcache.c:228:2: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
../../src/gdb/regcache.c:235:2: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
cc1: all warnings being treated as errors
regcache->register_status is "signed char".
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* regcache.c (regcache_xmalloc_1): Call XCALLOC with signed char
for 'regcache->register_status'.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" breakpoint.o 2>&1 1>/dev/null
../../src/gdb/breakpoint.c: In function ‘breakpoint_xfer_memory’:
../../src/gdb/breakpoint.c:1578:2: error: pointer targets in passing argument 3 of ‘gdbarch_breakpoint_from_pc’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/defs.h:644:0,
from ../../src/gdb/breakpoint.c:20:
../../src/gdb/gdbarch.h:495:25: note: expected ‘int *’ but argument is of type ‘unsigned int *’
target_info.placed_size is an 'int', and gdbarch_breakpoint_from_pc
takes an int too.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* breakpoint.c.c (breakpoint_xfer_memory): Change type of local to
int.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" stap-probe.o 2>&1 1>/dev/null
../../src/gdb/stap-probe.c: In function ‘handle_stap_probe’:
../../src/gdb/stap-probe.c:1306:19: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
cc1: all warnings being treated as errors
make: *** [stap-probe.o] Error 1
provider is a string, so it's rightfully a char*. 'data' holds raw
bytes (bfd_byte), so a cast is the right thing to do.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* stap-probe.c (handle_stap_probe): Add cast to char*.
$ make WERROR_CFLAGS="-Wpointer-sign -Werror" linux-record.o 2>&1 1>/dev/null
...
../../src/gdb/linux-record.c: In function ‘record_linux_system_call’:
../../src/gdb/linux-record.c:1152:9: error: pointer targets in passing argument 3 of ‘regcache_raw_read_signed’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/linux-record.c:23:0:
../../src/gdb/regcache.h:76:3: note: expected ‘long int *’ but argument is of type ‘long unsigned int *’
../../src/gdb/linux-record.c:1186:13: error: pointer targets in passing argument 3 of ‘regcache_raw_read_signed’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/linux-record.c:23:0:
../../src/gdb/regcache.h:76:3: note: expected ‘long int *’ but argument is of type ‘long unsigned int *’
I believe the read_signed calls are correct, and the variables are in
the wrong.
Tested on x86_64 Fedora 17.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* linux-record.c (record_linux_system_call) <gdb_sys_msgrcv,
RECORD_MSGRCV>: Pass a signed variable to
regcache_raw_read_signed, instead of an unsigned one.
tcp_retry_limit is installed as variable of an unsigned command:
add_setshow_uinteger_cmd ("connect-timeout", class_obscure,
&tcp_retry_limit, _("\
and I found no uses of the variable treating it as signed (like < 0
checks or some such).
2013-03-07 Pedro Alves <palves@redhat.com>
* ser-tcp.c (tcp_retry_limit): Change type to unsigned int.
Move the declarations to a header, rather than declaring them in
(multiple) .c files.
gdb/
2013-03-07 Pedro Alves <palves@redhat.com>
* remote.c (hex2bin, bin2hex): Move extern declarations to ...
* remote.h (hex2bin, bin2hex): ... here.
* tracepoint.c (hex2bin, bin2hex): Remove extern declarations.
The strlen in this function looked suspicious, for the function
documents that STR is in the target's charset (what if it is utf-32,
etc.).
On closer inspection, nothing is calling the function, and it's been
that way ever since the function was added.
gdb/
2013-03-06 Pedro Alves <palves@redhat.com>
* python/py-utils.c (target_string_to_unicode): Delete function.
* python/python-internal.h (target_string_to_unicode): Delete
declaration.
* cli/cli-decode.c (add_setshow_zuinteger_unlimited_cmd): Change
parameter VAR's type from "unsigned int" to "int".
* command.h (var_zuinteger_unlimited): Update its comments.
(add_setshow_zuinteger_unlimited_cmd): Update the declaration.
* linespec.c (struct linespec_canonical_name): New.
(struct linespec_state): Change canonical_names type to it.
(add_sal_to_sals): Change variable canonical_name to canonical. Change
xrealloc element size. Initialize the different CANONICAL fields.
(canonical_to_fullform): New.
(filter_results): Use it. Add variables canonical, fullform and
cleanup.
(struct decode_line_2_item, decode_line_2_compare_items): New.
(decode_line_2): Remove variables iter and item_names, add variables
items and items_count. Modify the code for these new variables.
gdb/testsuite/
* gdb.linespec/base/one/thefile.cc (twodup): New.
(m): Call it.
* gdb.linespec/base/two/thefile.cc (dupname): New.
(n): Call it.
* gdb.linespec/break-ask.exp: New file.
* gdb.linespec/lspec.cc (body_elsewhere): New comment marker.
gdb_byte should be used for bytes from the program being debugged. We
have many places using char or unsigned char instead all over the
existing ports, and more ends up added over time due to copy/paste as
new code is based on old code.
I've greped the tree for "char buf[", and fixed all I found.
Tested by building with --enable-targets=all.
2013-03-01 Pedro Alves <palves@redhat.com>
Use gdb_byte for bytes from the program being debugged.
* arm-tdep.c (arm_store_return_value, arm_get_longjmp_target):
Change type of local 'buf' to gdb_byte.
* avr-tdep.c (avr_frame_prev_register, avr_push_dummy_call): Likewise.
* bfin-tdep.c (bfin_push_dummy_call): Likewise.
* cris-tdep.c (cris_sigcontext_addr)
(cris_sigtramp_frame_unwind_cache): Likewise.
* frv-linux-tdep.c (frv_linux_pc_in_sigtramp)
(frv_linux_sigcontext_reg_addr, frv_linux_sigtramp_frame_cache):
Likewise.
* frv-tdep.c (frv_pseudo_register_write, frv_analyze_prologue): Likewise.
* hppa-hpux-tdep.c (hppa32_hpux_find_global_pointer)
(hppa32_hpux_search_dummy_call_sequence)
(hppa_hpux_supply_save_state): Likewise.
* hppa-linux-tdep.c (insns_match_pattern)
(hppa_linux_find_global_pointer): Likewise.
* hppa-tdep.c (hppa_in_function_epilogue_p)
(skip_prologue_hard_way, hppa_frame_cache): Likewise.
* i386-nto-tdep.c (i386nto_sigcontext_addr): Likewise.
* i386fbsd-tdep.c (i386fbsd_supply_uthread)
(i386fbsd_collect_uthread): Likewise.
* ia64-hpux-tdep.c (ia64_hpux_push_dummy_code): Likewise.
* ia64-linux-tdep.c (ia64_linux_sigcontext_register_address): Likewise.
* ia64-tdep.c (examine_prologue, ia64_frame_cache)
(ia64_frame_prev_register, ia64_sigtramp_frame_cache)
(ia64_sigtramp_frame_prev_register, ia64_access_reg)
(ia64_access_rse_reg, ia64_libunwind_frame_this_id)
(ia64_libunwind_frame_prev_register)
(ia64_libunwind_sigtramp_frame_this_id)
(ia64_find_global_pointer_from_dynamic_section)
(find_extant_func_descr, find_func_descr, ia64_dummy_id)
(ia64_unwind_pc): Likewise.
* iq2000-tdep.c (iq2000_store_return_value): Likewise.
* m68hc11-tdep.c (m68hc11_push_dummy_call)
(m68hc11_extract_return_value): Likewise.
* m68klinux-nat.c (fetch_register, store_register): Likewise.
* mep-tdep.c (mep_pseudo_cr32_read, mep_pseudo_cr32_write)
(mep_get_insn, mep_push_dummy_call): Likewise.
* mips-linux-tdep.c (mips_linux_get_longjmp_target)
(mips_linux_in_dynsym_stub): Likewise.
* mn10300-tdep.c (mep_pseudo_cr32_write): Likewise.
* ppc-linux-nat.c (fetch_register, store_register): Likewise.
* regcache.c (dump_endian_bytes): Change type of parameter 'buf'
to gdb_byte.
* remote-mips.c (mips_set_register): Likewise.
* remote-sim.c (gdbsim_fetch_register): Likewise.
* score-tdep.c (score7_fetch_inst): Change type of parameter
'memblock' and local 'buf' to gdb_byte.
(score7_malloc_and_get_memblock): Change return type to gdb_byte.
Change type of local 'buf' to gdb_byte. Adjust.
(score7_adjust_memblock_ptr): Change type of parameter 'memblock'
to gdb_byte**.
(score7_analyze_prologue): Change type of 'memblock' and
'memblock_ptr' locals to gdb_byte*.
* sh64-tdep.c (sh64_extract_return_value)
(sh64_store_return_value): Change type of local 'buf' to gdb_byte.
* solib-darwin.c (darwin_current_sos, darwin_read_exec_load_addr):
* solib-pa64.c (pa64_solib_create_inferior_hook)
(pa64_open_symbol_file_object): Remove local 'buf'.
* solib-som.c (som_solib_create_inferior_hook, link_map_start)
(som_open_symbol_file_object): Likewise.
* solib-spu.c (spu_current_sos): Likewise.
* spu-linux-nat.c (spu_fetch_inferior_registers): Likewise.
* spu-multiarch.c (parse_spufs_run, spu_fetch_registers)
(spu_store_registers): Likewise.
* target.c (debug_print_register): Likewise.
* tic6x-tdep.c (tic6x_get_longjmp_target): Likewise.
* xstormy16-tdep.c (xstormy16_store_return_value)
(xstormy16_push_dummy_call, xstormy16_resolve_jmp_table_entry)
(xstormy16_find_jmp_table_entry): Likewise.
printf-like functions to avoid type related warnings on all
platforms.
(handle_output_debug_string): Fetch context information address
from debug string using string_to_core_addr.
* regformats/reg-tilegx.dat (name): Change abi name to "tilegx".
* regformats/reg-tilegx32.dat: New.
gdbserver/
* Makefile.in (clean): Remove reg-tilegx.c, reg-tilegx32.c.
(reg-tilegx32.c): New rule.
* configure.srv (tilegx-*-linux*): Add reg-tilegx32.o to srv_regobj.
* linux-tile-low.c (tile_arch_setup): New function. Invoke
different register info initializer according to elf class.
(init_registers_tilgx32): New function. The tilegx32 register info
initializer.
(tile_fill_gregset): Use "uint_reg_t" to represent register size.
(tile_store_gregset): Likewise.
2013-02-27 Yao Qi <yao@codesourcery.com>
Pedro Alves <palves@redhat.com>
* tracepoint.c (tfile_trace_find): For tfind
pc/tp/range/outside, look for the next trace frame instead of
always starting from frame 0.
The "you have broken" bit of this text reads to me as if I had broken
it myself somehow. This patch eliminates that ambiguity.
2013-02-25 Pedro Alves <palves@redhat.com>
* common/linux-ptrace.c (linux_ptrace_test_ret_to_nx): Change
warning text.
symbol size.
* ppc64-tdep.c (ppc64_elf_make_msymbol_special): New function.
* ppc64-tdep.h (ppc64_elf_make_msymbol_special): Declare.
* ppc-linux-tdep.c (ppc_linux_init_abi): Set up to use the above.
* ppcfbsd-tdep.c (ppcfbsd_init_abi): Likewise.
Code cleanup.
* elfread.c (build_id_bfd_get): Make the return type const.
(build_id_verify): Make the check parameter const.
(build_id_to_debug_filename): Make the build_id parameter and variable
data const.
(find_separate_debug_file_by_buildid): Make the variable build_id const.
* python/py-arch.c (archpy_disassmble): Implementation of the
new method gdb.Architecture.disassemble.
(arch_object_methods): Add entry for the new method.
* doc/gdb.texinfo (Architectures In Python): Add description
about the new method gdb.Architecture.disassemble.
* testsuite/gdb.python/py-arch.c: New test case
* testsuite/gdb.python/py-arch.exp: New tests to test
gdb.Architecture.disassemble
* testsuite/gdb.python/Makefile.in: Add py-arch to the list of
EXECUTABLES.
While touching monitor_load in a previous patch, I noticed this method
is no longer set to anything other than NULL in the tree, so we can
remove it.
Tested by building with --enable-targets=all.
2013-02-19 Pedro Alves <palves@redhat.com>
Garbage collect 'struct monitor_ops'::load_routine.
* monitor.h (struct monitor_ops) <load_routine>: Remove field.
* monitor.c (monitor_load): No longer call
current_monitor->load_routine.
* dbug-rom.c (init_dbug_cmds): Don't set 'load_routine'.
* m32r-rom.c (init_m32r_cmds): Don't set 'load_routine'.
* ppcbug-rom.c (init_ppc_cmds): Don't set 'load_routine'.
Harmonize this old-looking code with generic_load, which fixes several
issues.
2013-02-19 Pedro Alves <palves@redhat.com>
PR gdb/15161
Harmonize with generic_load.
* monitor.c: Include "readline/readline.h".
(monitor_load): Rename parameter 'file' to 'args'. Use build_argv
instead of sscanf. Use CORE_ADDR/strtoulst instead of unsigned
long/strtol for the 'load_offset' local. Error out if no argument
is given or if too many arguments are given. Tilde expand the
passed in file name.
* symfile.c (load_section_data): Change type of load_offset
to CORE_ADDR.
(generic_load): User strtoulst instead of strtoul for conversion
of load_offset.
(tilegx_write_pc): New function.
(tilegx_cannot_reference_register): Return zero if REGNO
is TILEGX_FAULTNUM_REGNUM.
(tilegx_gdbarch_init): Add call to set_gdbarch_write_pc.
(tilegx_register_name): Add handling of "faultnum" register.
* tilegx-tdep.h (enum tilegx_regnum): Add TILEGX_FAULTNUM_REGNUM.
* tilegx-linux-tdep.c (tilegx_linux_supply_regset): Add
handling of TILEGX_FAULTNUM_REGNUM.
* tilegx-linux-nat.c (regmap): Add entry for TILEGX_FAULTNUM_REGNUM.
Hafiz Abid Qadeer <abidh@codesourcery.com>
gdb/
* NEWS: Mention new field "trace-file".
* tracepoint.c (trace_status_mi): Output "trace-file" field.
(tfile_open): Record the trace file's filename in the trace
status.
(tfile_files_info): Mention the name of the trace file.
Check the "filename" field explicitely.
(trace_status_command): Explicitely check "filename" field.
(trace_find_command): Ditto.
(trace_find_pc_command): Ditto.
(trace_find_tracepoint_command): Ditto.
(trace_find_line_command): Ditto.
(trace_find_range_command): Ditto.
(trace_find_outside_command): Ditto.
* tracepoint.h (struct trace_status) <from_file>: Rename it
to "filename" and make it hold the trace file's filename
instead of a boolean.
* remote.c (remote_get_trace_status): Initialize "filename"
field with NULL instead of 0.
gdb/doc/
* gdb.texinfo (GDB/MI Tracepoint Commands) <-trace-status>:
Document the "trace-file" field.
gdb/testsuite/
* gdb.trace/tfile.exp: Add test for -trace-status command.
This makes gdbserver share gdb's savestring, instead of baking its own.
Tested on x86_64 Fedora 17.
gdb/
2013-02-14 Pedro Alves <palves@redhat.com>
* utils.c (savestring): Don't #undef it. Move function to
common/common-utils.c.
* common/common-utils.c: Include gdb_string.h.
(savestring): Move here from utils.c.
* common/common-utils.h (savestring): Declare.
gdb/gdbserver/
2013-02-14 Pedro Alves <palves@redhat.com>
* tracepoint.c (save_string): Delete.
(add_tracepoint_action): Use savestring instead of save_string.
It's better to avoid needless confusion, and call string length,
length, instead of size, which is usually used to refer to sizeof of
the string (len+1):
size_t len = strlen (str);
size_t size = sizeof (str);
Tested on x86_64 Fedora 17.
2013-02-14 Pedro Alves <palves@redhat.com>
* utils.c (savestring): Rename parameter 'size' to 'len'.
Yufeng Zhang <yufeng.zhang@arm.com>
* aarch64-linux-nat.c (aarch64_init_debug_reg_state): Delete.
(aarch64_inferior_data, struct aarch64_inferior_data):
Delete.
(struct aarch64_process_info): New.
(aarch64_process_list): New global.
(aarch64_find_process_pid, aarch64_add_process)
(aarch64_process_info_get): New functions.
(aarch64_inferior_data_get): Delete.
(aarch64_process_info_get): New function.
(aarch64_forget_process): New function.
(aarch64_get_debug_reg_state): New parameter 'pid'. Reimplement.
(aarch64_linux_prepare_to_resume): Pass the lwp's pid to
aarch64_get_debug_reg_state.
(aarch64_notify_debug_reg_change): Use iterate_over_lwps
instead of linux_nat_iterate_watchpoint_lwps.
(aarch64_linux_new_fork): New function.
(aarch64_linux_child_post_startup_inferior): Use
aarch64_forget_process instead of aarch64_init_debug_reg_state.
(aarch64_handle_breakpoint, aarch64_linux_insert_hw_breakpoint)
(aarch64_linux_remove_hw_breakpoint)
(aarch64_handle_aligned_watchpoint)
(aarch64_handle_unaligned_watchpoint)
(aarch64_linux_insert_watchpoint)
(aarch64_linux_remove_watchpoint)
(aarch64_linux_stopped_data_address): Adjust to pass the current
process id to aarch64_debug_reg_state.
(_initialize_aarch64_linux_nat): Install aarch64_linux_new_fork as
linux_nat_new_fork hook, and aarch64_forget_process as
linux_nat_forget_process hook; remove the call to
register_inferior_data_with_cleanup.
I happened to notice a bug with ptype &Ref, and found out userdef.exp
actually exercises the bug. With:
class Container
{
public:
Member m;
Member& operator* ();
};
Member& Container::operator* ()
{
return this->m;
}
And 'c' is of type Container:
(gdb) p c
$1 = {m = {z = -9192}}
(gdb) p *c
$2 = (Member &) @0x7fffffffda20: {z = -9192}
(gdb) ptype *c
type = class Member {
public:
int z;
} &
(gdb) p &*c
$3 = (Member *) 0x7fffffffda20
(gdb) ptype &*c
type = class Member {
public:
int z;
} &*
(gdb)
Notice that last print (&*c) on says the type is a pointer - that's
how you get the address behind a reference. But notice the last ptype
instead says the type of the same expression is a pointer _reference_.
This looks like a bug to me.
This patch fixes it. The issue is that we're entering the VALUE_LVAL
(x) == lval_memory branch by mistake for references. The fix is just
to swap the tests so references are checked first, like value_addr
also handles references first.
Tested on x86_64 Fedora 17.
2013-02-14 Pedro Alves <palves@redhat.com>
* eval.c (evaluate_subexp_for_address) <default_case_after_eval,
EVAL_AVOID_SIDE_EFFECTS>: Swap and handle TYPE_CODE_REF before
lval_memory.
2013-02-14 Pedro Alves <palves@redhat.com>
* gdb.cp/userdef.exp (ptype &*c): Don't expect an &.
Hafiz Abid Qadeer <abidh@codesourcery.com>
gdb/
* tracepoint.h (validate_trace_state_variable_name): Declare.
* tracepoint.c (validate_trace_state_variable_name): New.
(trace_variable_command): Parse the trace state variable's name
without using parse_expression. Do several validations.
* mi/mi-main.c (mi_cmd_trace_define_variable): Don't parse the
trace state variable's name with parse_expression. Validate it.
gdb/testsuite/
* gdb.trace/tsv.exp: Adjust tests, and add a few more.
While reviewing the native AArch64 patch, I noticed a problem:
On 02/06/2013 08:46 PM, Pedro Alves wrote:
>
>> > +static void
>> > +aarch64_linux_prepare_to_resume (struct lwp_info *lwp)
>> > +{
>> > + struct arch_lwp_info *info = lwp->arch_private;
>> > +
>> > + /* NULL means this is the main thread still going through the shell,
>> > + or, no watchpoint has been set yet. In that case, there's
>> > + nothing to do. */
>> > + if (info == NULL)
>> > + return;
>> > +
>> > + if (DR_HAS_CHANGED (info->dr_changed_bp)
>> > + || DR_HAS_CHANGED (info->dr_changed_wp))
>> > + {
>> > + int tid = GET_LWP (lwp->ptid);
>> > + struct aarch64_debug_reg_state *state = aarch64_get_debug_reg_state ();
> Hmm. This is always fetching the debug_reg_state of
> the current inferior, but may not be the inferior of lwp.
> I see the same bug on x86. Sorry about that. I'll fix it.
A natural fix would be to make xxx_get_debug_reg_state take an
inferior argument, but that doesn't work because of the case where we
detach breakpoints/watchpoints from the child fork, at a time there's
no inferior for the child fork at all. We do a nasty hack in
i386_inferior_data_get, but that relies on all callers pointing the
current inferior to the correct inferior, which isn't actually being
done by all callers, and I don't think we want to enforce that -- deep
in the bowls of linux-nat.c, there are many cases we resume lwps
behind the scenes, and it's be better to not have that code rely on
global state (as it doesn't today).
The fix is to decouple the watchpoints code from inferiors, making it
track target processes instead. This way, we can freely keep track of
the watchpoint mirrors for these processes behind the core's back.
Checkpoints also play dirty tricks with swapping the process behind
the inferior, so they get special treatment too in the patch (which
just amounts to calling a new hook). Instead of the old hack in
i386_inferior_data_get, where we returned a copy of the current
inferior's debug registers mirror, as soon as we detect a fork in the
target, we copy the debug register mirror from the parent to the child
process.
I don't have an old kernel handy to test, but I stepped through gdb doing
the watchpoint removal in the fork child in the watchpoint-fork test
seeing that the debug registers end up cleared in the child.
I didn't find the need for linux_nat_iterate_watchpoint_lwps. If
we use plain iterate_over_lwps instead, what happens is that
when removing watchpoints, that iterate_over_lwps doesn't actually
iterate over anything, since the fork child is not added to the
lwp list until later, at detach time, in linux_child_follow_fork.
And if we don't iterate over that lwp, we don't mark its debug
registers as needing update. But linux_child_follow_fork takes
care of doing that explicitly:
child_lp = add_lwp (inferior_ptid);
child_lp->stopped = 1;
child_lp->last_resume_kind = resume_stop;
make_cleanup (delete_lwp_cleanup, child_lp);
/* CHILD_LP has new PID, therefore linux_nat_new_thread is not called for it.
See i386_inferior_data_get for the Linux kernel specifics.
Ensure linux_nat_prepare_to_resume will reset the hardware debug
registers. It is done by the linux_nat_new_thread call, which is
being skipped in add_lwp above for the first lwp of a pid. */
gdb_assert (num_lwps (GET_PID (child_lp->ptid)) == 1);
if (linux_nat_new_thread != NULL)
linux_nat_new_thread (child_lp);
if (linux_nat_prepare_to_resume != NULL)
linux_nat_prepare_to_resume (child_lp);
ptrace (PTRACE_DETACH, child_pid, 0, 0);
so unless I'm missing something (quite possible) it ends up all
the same. But, the !detach-on-fork, and the "follow-fork child" paths
should also call linux_nat_new_thread, and they don't presently. It
seems to me in those cases we're not clearing debug regs correctly
when that's needed. Instead of copying that bit that works around
add_lwp bypassing the linux_nat_new_thread call, I thought it'd
be better to add an add_initial_lwp call to be used in the case we
really need to bypass linux_nat_new_thread, and make
add_lwp always call linux_nat_new_thread.
i386_cleanup_dregs is rewritten to forget about the current process
debug mirrors, which takes cares of other i386 ports. Only a couple
of extra tweaks here and there were needed, as some targets wheren't
actually calling i386_cleanup_dregs.
Tested on Fedora 17 x86_64 -m64/-m32.
GDBserver already fetches the i386_debug_reg_state from the right
process, and, it doesn't handle forks at all, so no fix is needed over
there.
gdb/
2013-02-13 Pedro Alves <palves@redhat.com>
* amd64-linux-nat.c (update_debug_registers_callback):
Update comment.
(amd64_linux_dr_set_control, amd64_linux_dr_set_addr): Use
iterate_over_lwps.
(amd64_linux_prepare_to_resume): Pass the lwp's pid to
i386_debug_reg_state.
(amd64_linux_new_fork): New function.
(_initialize_amd64_linux_nat): Install amd64_linux_new_fork as
linux_nat_new_fork hook, and i386_forget_process as
linux_nat_forget_process hook.
* i386-linux-nat.c (update_debug_registers_callback):
Update comment.
(amd64_linux_dr_set_control, amd64_linux_dr_set_addr): Use
iterate_over_lwps.
(i386_linux_prepare_to_resume): Pass the lwp's pid to
i386_debug_reg_state.
(i386_linux_new_fork): New function.
(_initialize_i386_linux_nat): Install i386_linux_new_fork as
linux_nat_new_fork hook, and i386_forget_process as
linux_nat_forget_process hook.
* i386-nat.c (i386_init_dregs): Delete.
(i386_inferior_data, struct i386_inferior_data):
Delete.
(struct i386_process_info): New.
(i386_process_list): New global.
(i386_find_process_pid, i386_add_process, i386_process_info_get):
New functions.
(i386_inferior_data_get): Delete.
(i386_process_info_get): New function.
(i386_debug_reg_state): New parameter 'pid'. Reimplement.
(i386_forget_process): New function.
(i386_cleanup_dregs): Rewrite.
(i386_update_inferior_debug_regs, i386_insert_watchpoint)
(i386_remove_watchpoint, i386_region_ok_for_watchpoint)
(i386_stopped_data_address, i386_insert_hw_breakpoint)
(i386_remove_hw_breakpoint): Adjust to pass the current process id
to i386_debug_reg_state.
(i386_use_watchpoints): Don't register inferior data.
* i386-nat.h (i386_debug_reg_state): Add new 'pid' parameter, and
adjust comment.
(i386_forget_process): Declare.
* linux-fork.c (delete_fork): Call linux_nat_forget_process.
* linux-nat.c (linux_nat_new_fork, linux_nat_forget_process_hook):
New static globals.
(linux_child_follow_fork): Don't call linux_nat_new_thread here.
(add_initial_lwp): New, factored out from ...
(add_lwp): ... this. Don't check the number of lwps before
calling linux_nat_new_thread.
(linux_nat_iterate_watchpoint_lwps): Delete.
(linux_nat_attach): Use add_initial_lwp instead of add_lwp.
(linux_handle_extended_wait): Call the linux_nat_new_fork hook on
forks and vforks.
(linux_nat_wait_1): Use add_initial_lwp instead of add_lwp for the
initial lwp.
(linux_nat_kill, linux_nat_mourn_inferior): Call
linux_nat_forget_process.
(linux_nat_set_new_fork, linux_nat_set_forget_process)
(linux_nat_forget_process): New functions.
* linux-nat.h (linux_nat_iterate_watchpoint_lwps_ftype): Delete
type.
(linux_nat_iterate_watchpoint_lwps): Delete declaration.
(linux_nat_new_fork_ftype, linux_nat_forget_process_ftype): New
types.
(linux_nat_set_new_fork, linux_nat_set_forget_process)
(linux_nat_forget_process): New declarations.
* amd64fbsd-nat.c (super_mourn_inferior): New global.
(amd64fbsd_mourn_inferior): New function.
(_initialize_amd64fbsd_nat): Override to_mourn_inferior.
* windows-nat.c (windows_detach): Call i386_cleanup_dregs.
2013-02-13 Marcus Shawcroft <marcus.shawcroft@arm.com>
* aarch64-linux-nat.c (debug_reg_change_callback)
(aarch64_linux_get_debug_reg_capacity): ARI fix: Replace %llx with
%s and phex().
* c-exp.y (lex_one_token): Initialize other fields of yylval on
NAME return.
(classify_inner_name): Remove 'first_name' argument, add
'context'. Remove unused variable.
(yylex): Explicitly maintain the context type. Exit loop earlier
if NAME result is seen.
gdb/testsuite
* gdb.cp/m-static.cc (gnu_obj_1::~gnu_obj_1): New destructor.
* gdb.cp/m-static.exp: Add tests to print quoted destructor.
Some files managed to get in the tree with outdated copyright years.
This fixes it. Applied.
gdb/
2013-02-12 Pedro Alves <palves@redhat.com>
* break-catch-sig.c: Update copyright years.
gdb/testsuite/
2013-02-12 Pedro Alves <palves@redhat.com>
* gdb.base/catch-signal.c: Update copyright years.
* gdb.base/catch-signal.exp: Update copyright years.
* gdb.dwarf2/dw2-dir-file-name.c: Update copyright years.
* gdb.dwarf2/dw2-dir-file-name.exp: Update copyright years.
* gdb.dwarf2/dw2-empty-pc-range.S: Update copyright years.
* gdb.dwarf2/dw2-error.S: Update copyright years.
* gdb.dwarf2/dw2-error.c: Update copyright years.
* gdb.dwarf2/dw2-restrict.S: Update copyright years.
* gdb.dwarf2/dw2-restrict.c: Update copyright years.
* gdb.dwarf2/dw2-restrict.exp: Update copyright years.
provide a ui_out destructor.
* ui-out.h: Declare the new ui_out destructor.
(ui_out_impl): Add a field for data destructor in ui_out_impl.
* ui-out.c (default_data_destroy): Add a default data destructor
which does nothing.
(default_ui_out_impl): Set the new data_destroy field to
default_data_destroy
(uo_data_destroy): Local function which invokes the data
destructor if present.
(clear_table): Local function which clears the table data of a
ui_out object.
(ui_out_destroy): Public function which frees a ui_out object.
(ui_out_table_end): Use the new clear_table function.
* cli-out.c (cli_ui_out_impl): Set the new data_destroy field to
NULL.
* mi/mi-out.c (mi_ui_out_impl): Set the new data_destroy field
to NULL.
(printf_decfloat): New function. Broken out from ui_printf.
Remove unnecessary code to shift the entire format string down.
(printf_pointer): New function.
(ui_printf): Code to print C strings, wide C strings, decfloats,
and pointers moved to separate functions.
2013-02-08 Yufeng Zhang <yufeng.zhang@arm.com>
* aarch64-linux-tdep.c (AARCH64_LINUX_SIZEOF_GREGSET): Change the
number of the registers from 36 to 34.
gdb/
* sparc-tdep.c (sparc32_return_value): Handle writing return value when
using RETURN_VALUE_ABI_PRESERVES_ADDRESS.
* value.c (struct_return_convention): New function.
(using_struct_return): Implement in terms of struct_return_convention.
* value.h (struct_return_convention): Declare.
* stack.c (return_command): Allow successful overriding of the return
value when RETURN_VALUE_ABI_PRESERVES_ADDRESS.
2013-02-06 Yao Qi <yao@codesourcery.com>
* gdb.texinfo (GDB/MI Async Records): Document new MI
notification "=tsv-modified". Update the document of MI
notification "=tsv-created".
* observer.texi (GDB Observers): New observer tsv_modified.
Update observer tsv_created and tsv_deleted.
gdb:
2013-02-06 Yao Qi <yao@codesourcery.com>
* mi/mi-interp.c: Include "tracepoint.h".
(mi_tsv_modified): Declare.
(mi_tsv_created, mi_tsv_deleted): Update declaration.
(mi_interpreter_init): Call observer_attach_tsv_modified.
(mi_tsv_modified): New.
(mi_tsv_created, mi_tsv_deleted): Update.
* tracepoint.c (trace_variable_command): Call
observer_notify_tsv_modified if the initial value of tsv is
changed.
(delete_trace_state_variable): Call
observer_notify_tsv_deleted earlier.
(trace_variable_command): Caller update.
(create_tsv_from_upload): Likewise.
* observer.sh: Declare "struct trace_state_variable".
* NEWS: Mention the new MI notification "=tsv-modified".
gdb/testsuite:
2013-02-06 Yao Qi <yao@codesourcery.com>
* gdb.trace/mi-tsv-changed.exp (test_create_delete_tsv): Rename
to ...
(test_create_delete_modify_tsv): ... here. New test on modifying
the initial value of a tsv.