The documentation for make-breakpoint from the Guile API and the `spec'
variant of the gdb.Breakpoint constructor from the Python API state that
the format acceptable for location strings is the same as that accepted
by the break command. However, using the -probe qualifier at the
beginning of the location string causes a GDB internal error as it
attempts to decode a probe location in the wrong code path. Without this
functionality, there doesn't appear to be another way to set breakpoints
on probe points from Python or Guile scripts.
This patch introduces a new helper function that returns a
breakpoint_ops instance appropriate for a parsed location and updates
the Guile and Python bindings to use said function, rather than the
current hard-coded use of bkpt_breakpoint_ops. Since this logic is
duplicated in the handling of the `break' and `trace' commands, those
are also updated to call into the new helper function.
gdb/ChangeLog:
2019-12-10 George Barrett <bob@bob131.so>
Fix scripted probe breakpoints.
* breakpoint.c (tracepoint_probe_breakpoint_ops): Move
declaration forward.
(breakpoint_ops_for_event_location_type)
(breakpoint_ops_for_event_location): Add function definitions.
(break_command_1, trace_command): Use
breakpoint_ops_for_event_location.
* breakpoint.h (breakpoint_ops_for_event_location): Add function
declarations.
* guile/scm-breakpoint.c (gdbscm_register_breakpoint_x): Use
breakpoint_ops_for_event_location.
* python/py-breakpoint.c (bppy_init): Use
breakpoint_ops_for_event_location.
gdb/testsuite/ChangeLog:
2019-12-10 George Barrett <bob@bob131.so>
Test scripted probe breakpoints.
* gdb.guile/scm-breakpoint.c (main): Add probe point.
* gdb.python/py-breakpoint.c (main): Likewise.
* gdb.guile/scm-breakpoint.exp (test_bkpt_probe): Add probe
specifier test.
* gdb.python/py-breakpoint.exp (test_bkpt_probe): Likewise.
This patch adds . as an allowed character for user defined commands.
Combined with 'define-prefix', this allows to e.g. define a set of Valgrind
specific user command corresponding to the Valgrind monitor commands
(such as check_memory, v.info, v.set, ...).
gdb/ChangeLog
2019-11-30 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* command.h (valid_cmd_char_p): Declare.
* cli/cli-decode.c (valid_cmd_char_p): New function factorizing
the check of valid command char.
(find_command_name_length, valid_user_defined_cmd_name_p): Use
valid_cmd_char_p.
* cli/cli-script.c (validate_comname): Likewise.
* completer.c (gdb_completer_command_word_break_characters):
Do not remove . from the word break char, update comments.
(complete_line_internal_1): Use valid_cmd_char_p.
* guile/scm-cmd.c (gdbscm_parse_command_name): Likewise.
* python/py-cmd.c (gdbpy_parse_command_name): Likewise.
gdb/testsuite/ChangeLog
2019-11-30 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* gdb.base/define.exp: Test . in command names.
* gdb.base/setshow.exp: Update test, as . is now part of
command name.
Introduce a new class to wrap up the parameters needed for the
function search_symbols, which has now become a member function of
this new class.
The motivation is that search_symbols already takes a lot of
parameters, and a future commit is going to add even more. This
commit hopefully makes collecting the state required for a search
easier.
As part of this conversion the list of filenames in which to search
has been converted to a std::vector.
There should be no user visible changes after this commit.
gdb/ChangeLog:
* python/python.c (gdbpy_rbreak): Convert to using
global_symbol_searcher.
* symtab.c (file_matches): Convert return type to bool, change
file list to std::vector, update header comment.
(search_symbols): Rename to...
(global_symbol_searcher::search): ...this and update now its
a member function of global_symbol_searcher. Take account of the
changes to file_matches.
(symtab_symbol_info): Convert to using global_symbol_searcher.
(rbreak_command): Likewise.
(search_module_symbols): Likewise.
* symtab.h (enum symbol_search): Update comment.
(search_symbols): Remove declaration.
(class global_symbol_searcher): New class.
Change-Id: I488ab292a892d9e9e84775c632c5f198b6ad3710
This adds a "name_allocated" field to cmd_list_element, so that
commands can own their "name" when necessary. Then, this changes a
few spots in gdb that currently free the name by hand to instead use
this facility.
gdb/ChangeLog
2019-11-26 Tom Tromey <tom@tromey.com>
* python/py-function.c (fnpy_init): Update.
* value.h (add_internal_function): Adjust declaration.
* value.c (function_destroyer): Remove.
(do_add_internal_function): Don't set destroyer or copy name.
(add_internal_function): Take unique_xmalloc_ptr<char> for name.
Set name_allocated.
* python/py-cmd.c (cmdpy_destroyer): Don't free "name".
(cmdpy_init): Set name_allocated.
* cli/cli-decode.h (struct cmd_list_element) <name_allocated>: New
member.
(~cmd_list_element): Free "name" if needed.
Change-Id: Ie1435cea5bbf4bd92056125f112917c607cbb761
add_internal_function sets a command destroyer that frees the doc
string. However, many callers do not pass in an allocated doc string.
This adds a new overload to clearly differentiate the two cases,
fixing the latent bug.
gdb/ChangeLog
2019-11-26 Tom Tromey <tom@tromey.com>
* value.h (add_internal_function): Add new overload. Move
documentation from value.h.
* value.c (do_add_internal_function): New function.
(add_internal_function): Use it. Add new overload.
(function_destroyer): Don't free doc.
* python/py-function.c (fnpy_init): Update.
Change-Id: I3f6df925bc6b3e1bccbad9eeebc487b908bb5a2a
Python commands manage their "doc" string manually, but
cmd_list_element already has doc_allocated to handle this case. This
changes the Python code to use the existing facility.
gdb/ChangeLog
2019-11-26 Tom Tromey <tom@tromey.com>
* python/py-cmd.c (cmdpy_destroyer): Don't free "doc".
(cmdpy_init): Set "doc_allocated".
Change-Id: I0014edc117b051bba1f4db267687d231e7fe9b56
This changes gdb.post_event to use the new run_on_main_thread
function. This is somewhat tricky because the Python GIL must be held
while manipulating reference counts.
2019-11-26 Tom Tromey <tom@tromey.com>
* python/python.c (class gdbpy_gil): New.
(struct gdbpy_event): Add constructor, destructor, operator().
(gdbpy_post_event): Use run_on_main_thread.
(gdbpy_initialize_events): Remove.
(do_start_initialization): Update.
Change-Id: Ie4431e60f328dae48bd96b6c6a8e875e70bda1de
When running GDB tests under Valgrind, various tests are failing due
to invalid memory access.
Here is the stack trace reported by Valgrind, for gdb.base/freebpcmd.exp :
==18658== Invalid read of size 8
==18658== at 0x7F9107: is_main (signalmodule.c:195)
==18658== by 0x7F9107: PyOS_InterruptOccurred (signalmodule.c:1730)
==18658== by 0x3696E2: check_quit_flag() (extension.c:829)
==18658== by 0x36980B: restore_active_ext_lang(active_ext_lang_state*) (extension.c:782)
==18658== by 0x48F617: gdbpy_enter::~gdbpy_enter() (python.c:235)
==18658== by 0x47BB71: add_thread_object(thread_info*) (object.h:470)
==18658== by 0x53A84D: operator() (std_function.h:687)
==18658== by 0x53A84D: notify (observable.h:106)
==18658== by 0x53A84D: add_thread_silent(ptid_t) (thread.c:311)
==18658== by 0x3CD954: inf_ptrace_target::create_inferior(char const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&
, char**, int) (inf-ptrace.c:139)
==18658== by 0x3FE644: linux_nat_target::create_inferior(char const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&,
char**, int) (linux-nat.c:1094)
==18658== by 0x3D5727: run_command_1(char const*, int, run_how) (infcmd.c:633)
==18658== by 0x2C05D1: cmd_func(cmd_list_element*, char const*, int) (cli-decode.c:1948)
==18658== by 0x53F29F: execute_command(char const*, int) (top.c:639)
==18658== by 0x3638EB: command_handler(char const*) (event-top.c:586)
==18658== by 0x36468C: command_line_handler(std::unique_ptr<char, gdb::xfree_deleter<char> >&&) (event-top.c:771)
==18658== by 0x36407C: gdb_rl_callback_handler(char*) (event-top.c:217)
==18658== by 0x5B2A1F: rl_callback_read_char (callback.c:281)
==18658== by 0x36346D: gdb_rl_callback_read_char_wrapper_noexcept() (event-top.c:175)
==18658== by 0x363F70: gdb_rl_callback_read_char_wrapper(void*) (event-top.c:192)
==18658== by 0x3633AF: stdin_event_handler(int, void*) (event-top.c:514)
==18658== by 0x362504: gdb_wait_for_event (event-loop.c:857)
==18658== by 0x362504: gdb_wait_for_event(int) (event-loop.c:744)
==18658== by 0x362676: gdb_do_one_event() [clone .part.11] (event-loop.c:321)
==18658== by 0x3627AD: gdb_do_one_event (event-loop.c:303)
==18658== by 0x3627AD: start_event_loop() (event-loop.c:370)
==18658== by 0x41D35A: captured_command_loop() (main.c:381)
==18658== by 0x41F2A4: captured_main (main.c:1224)
==18658== by 0x41F2A4: gdb_main(captured_main_args*) (main.c:1239)
==18658== by 0x227D0A: main (gdb.c:32)
==18658== Address 0x10 is not stack'd, malloc'd or (recently) free'd
The problem seems to be created by gdbpy_enter::~gdbpy_enter () releasing the GIL lock
too early:
~gdbpy_enter () does:
...
PyGILState_Release (m_state);
python_gdbarch = m_gdbarch;
python_language = m_language;
restore_active_ext_lang (m_previous_active);
}
So, it releases the GIL lock, does 2 assignments and then leads to the following
call sequence:
restore_active_ext_lang => check_quit_flag => python.c gdbpy_check_quit_flag
=> PyOS_InterruptOccurred => is_main.
is_main code is:
static int
is_main(_PyRuntimeState *runtime)
{
unsigned long thread = PyThread_get_thread_ident();
PyInterpreterState *interp = _PyRuntimeState_GetThreadState(runtime)->interp;
return (thread == runtime->main_thread
&& interp == runtime->interpreters.main);
}
The macros and functions to access the thread state are documented as:
/* Variable and macro for in-line access to current thread
and interpreter state */
#define _PyRuntimeState_GetThreadState(runtime) \
((PyThreadState*)_Py_atomic_load_relaxed(&(runtime)->gilstate.tstate_current))
/* Get the current Python thread state.
Efficient macro reading directly the 'gilstate.tstate_current' atomic
variable. The macro is unsafe: it does not check for error and it can
return NULL.
The caller must hold the GIL.
See also PyThreadState_Get() and PyThreadState_GET(). */
#define _PyThreadState_GET() _PyRuntimeState_GetThreadState(&_PyRuntime)
So, we see that GDB releases the GIL and then potentially calls
_PyRuntimeState_GetThreadState that needs the GIL.
It is not very clear why the problem is only observed when running under
Valgrind. Probably caused by the slowdown due to Valgrind and/or to the 'single
thread' scheduling by Valgrind.
This patch fixes the crashes by releasing the GIT lock later.
2019-11-26 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* python/python.c (gdbpy_enter::~gdbpy_enter): Release GIL after
restore_active_ext_lang, as GIL is needed for (indirectly)
called PyOS_InterruptOccurred.
valgrind reports leaks in many python tests, such as:
==17162== VALGRIND_GDB_ERROR_BEGIN
==17162== 8,208 (5,472 direct, 2,736 indirect) bytes in 57 blocks are definitely lost in loss record 7,551 of 7,679
==17162== at 0x4835753: malloc (vg_replace_malloc.c:307)
==17162== by 0x6EAFD1: _PyObject_New (object.c:279)
==17162== by 0x4720E6: blpy_iter(_object*) (py-block.c:92)
==17162== by 0x698772: PyObject_GetIter (abstract.c:2577)
==17162== by 0x2343BE: _PyEval_EvalFrameDefault (ceval.c:3159)
==17162== by 0x22E9E2: function_code_fastcall (call.c:283)
==17162== by 0x2340A8: _PyObject_Vectorcall (abstract.h:127)
==17162== by 0x2340A8: call_function (ceval.c:4987)
==17162== by 0x2340A8: _PyEval_EvalFrameDefault (ceval.c:3486)
==17162== by 0x22E9E2: function_code_fastcall (call.c:283)
==17162== by 0x82172B: _PyObject_Vectorcall (abstract.h:127)
==17162== by 0x82172B: method_vectorcall (classobject.c:67)
==17162== by 0x6AF474: _PyObject_Vectorcall (abstract.h:127)
==17162== by 0x6AF474: _PyObject_CallNoArg (abstract.h:153)
==17162== by 0x6AF474: _PyObject_CallFunctionVa (call.c:914)
==17162== by 0x6B0673: callmethod (call.c:1010)
==17162== by 0x6B0673: _PyObject_CallMethod_SizeT (call.c:1103)
==17162== by 0x477DFE: gdb_PyObject_CallMethod<> (python-internal.h:182)
==17162== by 0x477DFE: get_py_iter_from_func(_object*, char const*) (py-framefilter.c:272)
==17162== by 0x4791B4: py_print_args (py-framefilter.c:706)
==17162== by 0x4791B4: py_print_frame(_object*, enum_flags<frame_filter_flag>, ext_lang_frame_args, ui_out*, int, htab*) (py-framefilter.c:960)
==17162== by 0x47A130: gdbpy_apply_frame_filter(extension_language_defn const*, frame_info*, enum_flags<frame_filter_flag>, ext_lang_frame_args, ui_out*, int, int) (py-framefilter.c:1236)
==17162== by 0x369C39: apply_ext_lang_frame_filter(frame_info*, enum_flags<frame_filter_flag>, ext_lang_frame_args, ui_out*, int, int) (extension.c:563)
==17162== by 0x4EC9C9: backtrace_command_1 (stack.c:2031)
==17162== by 0x4EC9C9: backtrace_command(char const*, int) (stack.c:2183)
...
Most of the leaks in python tests are due to the fact that many
PyObject xxxxx_dealloc functions are missing the line to free self
or obj such as:
Py_TYPE (self)->tp_free (self);
or
Py_TYPE (obj)->tp_free (obj);
With this patch, the number of python tests leaking decreases from 52 to 12.
gdb/ChangeLog
2019-11-18 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* python/py-block.c (blpy_dealloc): Call tp_free.
(blpy_block_syms_dealloc): Likewise.
* python/py-finishbreakpoint.c (bpfinishpy_dealloc): Likewise.
* python/py-inferior.c (infpy_dealloc): Likewise.
* python/py-lazy-string.c (stpy_dealloc): Likewise.
* python/py-linetable.c (ltpy_iterator_dealloc): Likewise.
* python/py-symbol.c (sympy_dealloc): Likewise.
* python/py-symtab.c (stpy_dealloc): Likewise.
* python/py-type.c (typy_iterator_dealloc): Likewise.
valgrind reports a leak when a breakpoint is created then deleted:
==1313== 40 bytes in 1 blocks are definitely lost in loss record 1,115 of 8,596
==1313== at 0x4835753: malloc (vg_replace_malloc.c:307)
==1313== by 0x6E05BC: _PyObject_New (object.c:255)
==1313== by 0x470E4B: gdbpy_breakpoint_created(breakpoint*) (py-breakpoint.c:1023)
==1313== by 0x2946D9: operator() (std_function.h:687)
==1313== by 0x2946D9: notify (observable.h:106)
==1313== by 0x2946D9: install_breakpoint(int, std::unique_ptr<breakpoint, std::default_delete<breakpoint> >&&, int) (breakpoint.c:8136)
==1313== by 0x295BCA: create_breakpoint_sal (breakpoint.c:8878)
==1313== by 0x295BCA: create_breakpoints_sal (breakpoint.c:8919)
==1313== by 0x295BCA: create_breakpoints_sal_default (breakpoint.c:13671)
...
The leak is due to a superfluous Py_INCREF when the python object
is allocated inside gdbpy_breakpoint_created, when the python object
is allocated locally: this object has already a refcount of 1, and
the only reference is the reference from the C breakpoint object.
The Py_INCREF is however needed when the python object was created from
python: the python object was stored in bppy_pending_object, and
gdbpy_breakpoint_created creates a new reference to this object.
Solve the leak by calling 'Py_INCREF (newbp);' only in the bppy_pending_object
case.
Regression tested on debian/amd64 natively and under valgrind on centos/amd64.
Before the patch, 795 tests have a definite leak.
After the patch, 197 have a definite leak.
Thanks to Tom, that helped on irc with the python refcount logic.
gdb/ChangeLog
2019-11-14 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* python/py-finishbreakpoint.c (gdbpy_breakpoint_created):
only call Py_INCREF (newbp) in the bppy_pending_object case.
If gdb.lookup_static_symbol is going to return a single symbol then it
makes sense (I think) for it to return a context sensitive choice of
symbol, that is the global static symbol that would be visible to the
program at that point.
However, if the user of the python API wants to instead get a
consistent set of global static symbols, no matter where they stop,
then they have to instead consider all global static symbols with a
given name - there could be many. That is what this new API function
offers, it returns a list (possibly empty) of all global static
symbols matching a given name (and optionally a given symbol domain).
gdb/ChangeLog:
* python/py-symbol.c (gdbpy_lookup_static_symbols): New
function.
* python/python-internal.h (gdbpy_lookup_static_symbols):
Declare new function.
* python/python.c (python_GdbMethods): Add
gdb.lookup_static_symbols method.
* NEWS: Mention gdb.lookup_static_symbols.
gdb/testsuite/ChangeLog:
* gdb.python/py-symbol.exp: Add test for
gdb.lookup_static_symbols.
gdb/doc/ChangeLog:
* python.texi (Symbols In Python): Add documentation for
gdb.lookup_static_symbols.
Change-Id: I1153b0ae5bcbc43b3dcf139043c7a48bf791e1a3
When using gdb.lookup_static_symbol I think that GDB should find
static symbols (global symbol with static linkage) from the current
object file ahead of static symbols from other object files.
This means that if we have two source files f1.c and f2.c, and both
files contains 'static int foo;', then when we are stopped in f1.c a
call to 'gdb.lookup_static_symbol ("foo")' will find f1.c::foo, and if
we are stopped in f2.c we would find 'f2.c::foo'.
Given that gdb.lookup_static_symbol always returns a single symbol,
but there can be multiple static symbols with the same name GDB is
always making a choice about which symbols to return. I think that it
makes sense for the choice GDB makes in this case to match what a user
would get on the command line if they asked to 'print foo'.
gdb/testsuite/ChangeLog:
* gdb.python/py-symbol.c: Declare and call function from new
py-symbol-2.c file.
* gdb.python/py-symbol.exp: Compile both source files, and add new
tests for gdb.lookup_static_symbol.
* gdb.python/py-symbol-2.c: New file.
gdb/doc/ChangeLog:
* python.texi (Symbols In Python): Extend documentation for
gdb.lookup_static_symbol.
gdb/ChangeLog:
* python/py-symbol.c (gdbpy_lookup_static_symbol): Lookup in
static block of current object file first. Also fix typo in
header comment.
Change-Id: Ie55dbeb8806f35577b46015deecde27a0ca2ab64
This changes command_line_input to return a "const char *", which is
appropriate because the memory is owned by command_line_input. Then
it fixes up the users.
I looked at making command_line_input transfer ownership to its caller
instead, but this is complicated due to the way read_next_line is
called, so I decided against it.
Tested by rebuilding.
gdb/ChangeLog
2019-11-08 Tom Tromey <tromey@adacore.com>
* top.c (read_command_file): Update.
(command_line_input): Make return type const.
* python/py-gdb-readline.c: Update.
* linespec.c (decode_line_2): Update.
* defs.h (command_line_input): Make return type const.
* cli/cli-script.c (read_next_line): Make return type const.
* ada-lang.c (get_selections): Update.
Change-Id: I27e6c9477fd1005ab5b16e0d337e4c015b6e6248
The la_get_string member of struct language_defn was intended to
provide a way to fetch string data from a "string" object in a
language-dependent way. However, it turned out that this was never
needed, and was only ever implemented for C. This patch removes the
language hook entirely.
gdb/ChangeLog
2019-11-05 Tom Tromey <tom@tromey.com>
* rust-lang.c (rust_language_defn): Update.
* python/py-value.c (valpy_string): Call c_get_string.
* p-lang.c (pascal_language_defn): Update.
* opencl-lang.c (opencl_language_defn): Update.
* objc-lang.c (objc_language_defn): Update.
* m2-lang.c (m2_language_defn): Update.
* language.c (unknown_language_defn, auto_language_defn): Update.
(default_get_string): Remove.
* guile/scm-value.c (gdbscm_value_to_string): Use c_get_string.
* go-lang.c (go_language_defn): Update.
* f-lang.c (f_language_defn): Update.
* d-lang.c (d_language_defn): Update.
* c-lang.c (c_language_defn, cplus_language_defn)
(asm_language_defn, minimal_language_defn): Update.
* ada-lang.c (ada_language_defn): Update.
* language.h (struct language_defn) <la_get_string>: Remove.
(LA_GET_STRING): Remove.
(default_get_string): Don't declare.
Change-Id: Ia97763dfe34dc8ecb46587f7a651f8af9be8fdbd
The documentation for Progspace.block_for_pc says:
Return the innermost gdb.Block containing the given pc value. If the
block cannot be found for the pc value specified, the function will
return None.
However, the implementation actually throws an error for invalid
addresses, like this:
(gdb) python print gdb.current_progspace ().block_for_pc (1)
Traceback (most recent call last):
File "<string>", line 1, in <module>
RuntimeError: Cannot locate object file for block.
Error while executing Python code.
(gdb)
This has been the behaviour since the command was first added (when
the documentation was still as above) in this commit:
commit f3e9a8177c
Date: Wed Feb 24 21:18:28 2010 +0000
Since that commit the code in question has moved around, but the
important parts are largely unchanged. The function in question is
now in py-progspace.c:pspy_block_for_pc.
Examining the code shows that the real state is more complex than just
the function throws an error instead of returning None, instead the
real situation is:
1. If we can't find a compilation unit for the $pc value then we
throw an error, but
2. If we can find a compilation unit, but can't find a block within
the compilation unit for the $pc then return None.
I suspect for most users of the Python API this distinction is
irrelevant, and I propose that we standardise on one single failure
mechanism.
Given the function can currently return None in some cases, and is
documented to return None on error, I propose we make that the case
for all error paths, which is what this patch does.
As the Progspace.block_for_pc method is currently untested, I've added
some basic tests including for a call with an invalid $pc.
This is potentially an API breaking change, though an undocumented
part of the API. Also, users should have been checking and handling a
None return value anyway, so my hope is that this shouldn't be too
disruptive.
gdb/ChangeLog:
* python/py-progspace.c (pspy_block_for_pc): Return None for all
error paths.
gdb/testsuite/ChangeLog:
* gdb.python/py-progspace.exp: Add tests for the
Progspace.block_for_pc method.
Change-Id: I9cea8d2132902bcad0013d1fd39080dd5423cc57
I'm seeing this failure:
...
(gdb) print /x $bnd0 = {0x10, 0x20}^M
$23 = {lbound = 0x10, ubound = 0x20}^M
(gdb) FAIL: gdb.arch/i386-mpx.exp: verify size for bnd0
...
The test expects a pretty printer to be actived printing 'size 17':
...
set test_string ".*\\\: size 17.*"
gdb_test "print /x \$bnd0 = {0x10, 0x20}" "$test_string" "verify size for bnd0"
...
but that doesn't happen.
The pretty printer is for the type of the $bnd0 register, which is created
here in i386_bnd_type:
...
t = arch_composite_type (gdbarch,
"__gdb_builtin_type_bound128", TYPE_CODE_STRUCT);
append_composite_type_field (t, "lbound", bt->builtin_data_ptr);
append_composite_type_field (t, "ubound", bt->builtin_data_ptr);
TYPE_NAME (t) = "builtin_type_bound128";
...
And the pretty-printer is registered here in
gdb/python/lib/gdb/printer/bound_registers.py:
...
gdb.printing.add_builtin_pretty_printer ('mpx_bound128',
'^__gdb_builtin_type_bound128',
MpxBound128Printer)
...
Fix the pretty printer by changing the regexp argument of
add_builtin_pretty_printer to match "builtin_type_bound128", the TYPE_NAME.
Tested on x86_64-linux.
gdb/ChangeLog:
2019-10-09 Tom de Vries <tdevries@suse.de>
* python/lib/gdb/printer/bound_registers.py: Use
'^builtin_type_bound128' as regexp argument for
add_builtin_pretty_printer.
This introduces a new "metadata" style and changes many places in gdb
to use it. The idea here is to let the user distinguish gdb output
from output that (conceptually at least) comes directly from the
inferior. The newly-styled category includes text that gdb
traditionally surrounds in "<...>", like "<unavailable>".
I only added a single test for this. In many cases this output is
difficult to test. Also, while developing this errors in the
implementation of the new printf formats showed up as regressions.
gdb/ChangeLog
2019-10-01 Tom Tromey <tom@tromey.com>
* p-lang.c (pascal_printstr): Use metadata style.
* value.c (show_convenience): Use metadata style.
* valprint.c (valprint_check_validity, val_print_optimized_out)
(val_print_not_saved, val_print_unavailable)
(val_print_invalid_address, generic_val_print, val_print)
(value_check_printable, val_print_array_elements): Use metadata
style.
* ui-out.h (class ui_out) <field_fmt>: New overload.
<do_field_fmt>: Add style parameter.
* ui-out.c (ui_out::field_fmt): New overload.
* typeprint.c (type_print_unknown_return_type)
(val_print_not_allocated, val_print_not_associated): Use metadata
style.
* tui/tui-out.h (class tui_ui_out) <do_field_fmt>: Add style
parameter.
* tui/tui-out.c (tui_ui_out::do_field_fmt): Update.
* tracepoint.c (tvariables_info_1): Use metadata style.
* stack.c (print_frame_arg, print_frame_info, print_frame)
(info_frame_command_core): Use metadata style.
* skip.c (info_skip_command): Use metadata style.
* rust-lang.c (rust_print_enum): Use metadata style.
* python/py-prettyprint.c (print_stack_unless_memory_error): Use
metadata style.
* python/py-framefilter.c (py_print_single_arg): Use metadata
style.
* printcmd.c (do_one_display, print_variable_and_value): Use
metadata style.
* p-valprint.c (pascal_val_print)
(pascal_object_print_value_fields): Use metadata style.
* p-typeprint.c (pascal_type_print_base): Use metadata style.
* mi/mi-out.h (class mi_ui_out) <do_field_fmt>: Add style
parameter.
* mi/mi-out.c (mi_ui_out::do_field_fmt): Update.
* m2-valprint.c (m2_print_long_set): Use metadata style.
* m2-typeprint.c (m2_print_type): Use metadata style.
* infcmd.c (print_return_value_1): Use metadata style.
* gnu-v3-abi.c (print_one_vtable): Use metadata style.
* f-valprint.c (info_common_command_for_block): Use metadata
style.
* f-typeprint.c (f_type_print_base): Use metadata style.
* expprint.c (print_subexp_standard): Use metadata style.
* cp-valprint.c (cp_print_value_fields): Use metadata style.
* cli/cli-style.h (class cli_style_option): Add constructor.
(metadata_style): Declare.
* cli/cli-style.c (metadata_style): New global.
(_initialize_cli_style): Register metadata style.
* cli-out.h (class cli_ui_out) <do_field_fmt>: Add style
parameter.
* cli-out.c (cli_ui_out::do_field_fmt): Update.
* c-typeprint.c (c_type_print_base_struct_union)
(c_type_print_base_1): Use metadata style.
* breakpoint.c (watchpoint_value_print)
(print_one_breakpoint_location): Use metadata style.
* break-catch-syscall.c (print_one_catch_syscall): Use metadata
style.
* break-catch-sig.c (signal_catchpoint_print_one): Use metadata
style.
* ada-valprint.c (val_print_packed_array_elements, printstr)
(print_field_values, ada_val_print_ref, ada_val_print): Use
metadata style.
* ada-typeprint.c (print_array_type, ada_print_type): Use metadata
style.
* ada-tasks.c (print_ada_task_info, info_task): Use metadata
style.
* ada-lang.c (user_select_syms): Use metadata style.
gdb/testsuite/ChangeLog
2019-10-01 Tom Tromey <tom@tromey.com>
* lib/gdb-utils.exp (style): Handle "metadata" argument.
* gdb.base/style.exp: Add metadata style test.
dwarf2read.c will create stub types for Ada "Taft Amendment" types.
These stub types can currently be exposed to Python code, where they
show up as TYPE_CODE_VOID types (but that, mysteriously, can sometimes
be used in other ways).
While it's possible to work with such types by using strip_typedefs,
this seemed unpleasant to me. This patch takes another approach
instead, which is to try not to expose stub types to Python users.
gdb/ChangeLog
2019-09-26 Tom Tromey <tromey@adacore.com>
* python/py-type.c (type_to_type_object): Call check_typedef
for stub types.
gdb/testsuite/ChangeLog
2019-09-26 Tom Tromey <tromey@adacore.com>
* gdb.ada/py_taft.exp: New file.
* gdb.ada/py_taft/main.adb: New file.
* gdb.ada/py_taft/pkg.adb: New file.
* gdb.ada/py_taft/pkg.ads: New file.
I noticed that make_hex_string does essentially the same thing as
bin2hex, and furthermore is only called in a single spot. This patch
removes make_hex_string.
Tested by the builtbot.
gdb/ChangeLog
2019-09-25 Tom Tromey <tom@tromey.com>
* python/py-objfile.c (objfpy_get_build_id): Use bin2hex.
* utils.h (make_hex_string): Don't declare.
* utils.c (make_hex_string): Remove.
This simplifies memory management. I've also changed some global variables
to std::string accordingly (which store the result of these functions),
but not all because some are used with add_setshow_optional_filename_cmd
which requires a char*.
gdb/ChangeLog:
2019-09-11 Christian Biesinger <cbiesinger@google.com>
* auto-load.c (auto_load_expand_dir_vars): Update.
* defs.h (gdb_datadir): Change to std::string.
(python_libdir): Likewise.
(relocate_gdb_directory): Change return type to std::string.
* guile/guile.c (gdbscm_data_directory): Update.
(initialize_scheme_side): Update.
* jit.c (jit_reader_dir): Change to std::string.
(jit_reader_load_command): Update.
* main.c (gdb_datadir): Change to std::string.
(python_libdir): Likewise.
(set_gdb_data_directory): Update.
(relocate_path): Change to return std::string.
(relocate_gdb_directory): Change to return std::string.
(relocate_gdbinit_path_maybe_in_datadir): Update.
(captured_main_1): Update.
* python/python.c (do_start_initialization): Update.
* top.c (show_gdb_datadir): Update.
* xml-syscall.c (xml_init_syscalls_info): Update.
(init_syscalls_info): Update.
Valgrind reports the below leak.
Make the variable progname_copy static, so that Valgrind continues
to find a pointer to the memory given to Python.
Note that the comment in do_start_initialization and the Python documentation
indicates that the progname given to Py_SetProgramName cannot be freed.
However, in Python 3.7.4, Py_SetProgramName does:
void
Py_SetProgramName(const wchar_t *program_name)
{
...
PyMem_RawFree(_Py_path_config.program_name);
_Py_path_config.program_name = _PyMem_RawWcsdup(program_name);
So, it looks like 3.7.4 Python duplicates its argument, which explains
the leak found by Valgrind.
It looks better to respect the doc and not have GDB freeing the string
given to Py_SetProgramName, and avoid the leak error by declaring
the progname_copy static.
This will work with Python versions that really use this string without
duplicating it, and avoids a leak report for Python version that duplicates
it.
==4023== 200 bytes in 1 blocks are definitely lost in loss record 4,545 of 7,116^M
==4023== at 0x4C29F33: malloc (vg_replace_malloc.c:307)^M
==4023== by 0x446D27: xmalloc (alloc.c:60)^M
==4023== by 0x657C77: do_start_initialization (python.c:1610)^M
==4023== by 0x657C77: _initialize_python() (python.c:1823)^M
==4023== by 0x75FE24: initialize_all_files() (init.c:231)^M
==4023== by 0x708A94: gdb_init(char*) (top.c:2242)^M
==4023== by 0x5E7460: captured_main_1 (main.c:857)^M
==4023== by 0x5E7460: captured_main (main.c:1161)^M
==4023== by 0x5E7460: gdb_main(captured_main_args*) (main.c:1186)^M
==4023== by 0x4122D4: main (gdb.c:32)^M
gdb/ChangeLog
2019-09-09 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* python/python.c (do_start_initialization): Make progname_copy static,
to avoid a leak report.
The 'info variables', its alias 'whereis', and 'info functions' all
include non-debug symbols in the output by default. The list of
non-debug symbols can sometimes be quite long, resulting in the
debug symbol based results being scrolled off the screen.
This commit adds a '-n' flag to all of the commands listed above that
excludes the non-debug symbols from the results, leaving just the
debug symbol based results.
gdb/ChangeLog:
* cli/cli-utils.c (info_print_options_defs): Delete.
(make_info_print_options_def_group): Delete.
(extract_info_print_options): Delete.
(info_print_command_completer): Delete.
(info_print_args_help): Add extra parameter, and optionally
include text about -n flag.
* cli/cli-utils.h (struct info_print_options): Delete.
(extract_info_print_options): Delete declaration.
(info_print_command_completer): Delete declaration.
(info_print_args_help): Add extra parameter, extend header
comment.
* python/python.c (gdbpy_rbreak): Pass additional parameter to
search_symbols.
* stack.c (struct info_print_options): New type.
(info_print_options_defs): New file scoped variable.
(make_info_print_options_def_group): New static function.
(info_print_command_completer): New static function.
(info_locals_command): Update to use new local functions.
(info_args_command): Likewise.
(_initialize_stack): Add extra parameter to calls to
info_print_args_help.
* symtab.c (search_symbols): Add extra parameter, use this to
possibly excluse non-debug symbols.
(symtab_symbol_info): Add extra parameter, which is passed on to
search_symbols.
(struct info_print_options): New type.
(info_print_options_defs): New file scoped variable.
(make_info_print_options_def_group): New static function.
(info_print_command_completer): New static function.
(info_variables_command): Update to use local functions, and pass
extra parameter through to symtab_symbol_info.
(info_functions_command): Likewise.
(info_types_command): Pass additional argument through to
symtab_symbol_info.
(rbreak_command): Pass extra argument to search_symbols.
(_initialize_symtab): Add extra arguments for calls to
info_print_args_help, and update help text for 'info variables',
'whereis', and 'info functions' commands.
* symtab.h (search_symbols): Add extra argument to declaration.
* NEWS: Mention new flags.
gdb/doc/ChangeLog:
* gdb.texinfo (Symbols): Add information about the -n flag to
"info variables" and "info functions".
gdb/testsuite/ChangeLog:
* gdb.base/info-fun.exp: Extend to test the -n flag for 'info
functions'. Reindent as needed.
* gdb.base/info-var-f1.c: New file.
* gdb.base/info-var-f2.c: New file.
* gdb.base/info-var.exp: New file.
* gdb.base/info-var.h: New file.
PyFile_FromString and PyFile_AsFile have been removed in Python 3.
There is no obvious replacement that works here, and we can't just
pass our FILE* to a DLL in Windows because it may use a different
C runtime.
So we just call a Python function which reads and executes file
contents. Care must be taken to execute it in the context of
__main__.
Tested by inverting the ifdef and running the testsuite on Debian
Linux (even without the patch, I failed at running the testsuite
on Windows). I did test with both Python 2 and 3.
gdb/ChangeLog:
2019-08-22 Christian Biesinger <cbiesinger@google.com>
* python/lib/gdb/__init__.py (_execute_file): New function.
* python/python.c (python_run_simple_file): Call gdb._execute_file
on Windows.
value_has_field had a mis-indented line. This fixes it.
gdb/ChangeLog
2019-08-19 Tom Tromey <tromey@adacore.com>
* python/py-value.c (value_has_field): Fix indentation.
I could not tell if GdbSetPythonDirectory is internal or not because
I could not find any references to it, so I left it as-is.
Tested by running the testsuite on gdb.python/*.exp; everything still
passes.
2019-08-15 Christian Biesinger <cbiesinger@google.com>
* python/lib/gdb/__init__.py (GdbOutputFile): Rename to have a
leading underscore.
(GdbOutputErrorFile): Likewise.
(global scope): Adjust constructor calls to GdbOutput{,Error}File
accordingly.
(execute_unwinders): Rename to have a leading underscore.
(auto_load_packages): Likewise.
(global scope): Adjust call to auto_load_packages accordingly.
(GdbSetPythonDirectory): Likewise.
* python/py-unwind.c (pyuw_sniffer): Call _execute_unwinders
instead of execute_unwinders.
gdb/testsuite/ChangeLog:
2019-08-15 Christian Biesinger <cbiesinger@google.com>
* gdb.python/python.exp: Expect a leading underscore on
GdbOutput{,Error}File.
This changes struct frame_arg to be self-managing and then fixes the
various users.
Tested by the buildbot.
gdb/ChangeLog
2019-08-07 Tom Tromey <tromey@adacore.com>
* stack.c (print_frame_arg, read_frame_local, read_frame_arg)
(print_frame_args): Update.
* python/py-framefilter.c (py_print_single_arg, enumerate_args):
Update.
* mi/mi-cmd-stack.c (list_arg_or_local): Update.
* frame.h (struct frame_arg): Add initializers.
<error>: Now a unique_xmalloc_ptr.
Armv8.3-a Pointer Authentication causes the function return address to be
obfuscated on entry to some functions. GDB must unmask the link register in
order to produce a backtrace.
The following patch adds markers of [PAC] to the bracktrace, to indicate
which addresses needed unmasking. This includes the backtrace when using MI.
For example, consider the following backtrace:
(gdb) bt
0 0x0000000000400490 in puts@plt ()
1 0x00000000004005dc in foo ("hello") at cbreak-lib.c:6
2 0x0000000000400604 [PAC] in bar () at cbreak-lib.c:12
3 0x0000000000400620 [PAC] in main2 () at cbreak.c:17
4 0x00000000004005b4 in main () at cbreak-3.c:10
The functions in cbreak-lib use pointer auth, which masks the return address
to the previous function, causing the addresses of bar (in the library) and main2
(in the main binary) to require unmasking in order to unwind the backtrace.
An extra bool is added alongside the prev_pc in the frame structure. At the
point at which the link register is unmasked, the AArch64 port calls into frame
to sets the bool. This is the most efficient way of doing it.
The marker is also added to the python frame printer, which is always printed if
set. The marker is not explicitly exposed to the python code.
I expect this will potentially cause issues with some tests in the testsuite
when Armv8.3 pointer authentication is used. This should be fixed up in the
the future once real hardware is available for full testsuite testing.
gdb/ChangeLog:
* NEWS: Expand the Pointer Authentication entry.
* aarch64-tdep.c (aarch64_frame_unmask_address): Rename from this.
(aarch64_frame_unmask_lr): ... to this.
(aarch64_prologue_prev_register, aarch64_dwarf2_prev_register):
Call aarch64_frame_unmask_lr.
* frame.c (struct frame_info): Add "masked" variable.
(frame_set_previous_pc_masked) (frame_get_pc_masked): New functions.
(fprint_frame): Check for masked pc.
* frame.h (frame_set_previous_pc_masked) (frame_get_pc_masked): New
declarations.
* python/py-framefilter.c (py_print_frame): Check for masked pc.
* stack.c (print_frame): Check for masked pc.
gdb/doc/ChangeLog:
* gdb.texinfo (AArch64 Pointer Authentication): New subsection.
With this patch, the help docs now respect 2 invariants:
* The first line of a command help is terminated by a '.' character.
* The last character of a command help is not a newline character.
Note that the changes for the last invariant were done by Tom, as part of :
[PATCH] Remove trailing newlines from help text
https://sourceware.org/ml/gdb-patches/2019-06/msg00050.html
but some occurrences have been re-introduced since then.
Some help docs had to be rephrased/restructured to respect the above
invariants.
Before this patch, print_doc_line was printing the first line
of a command help documentation, but stopping at the first '.'
or ',' character.
This was giving inconsistent results :
* The first line of command helps was sometimes '.' terminated,
sometimes not.
* The first line of command helps was not always designed to be
readable/understandable/unambiguous when stopping at the first
'.' or ',' character.
This e.g. created the following inconsistencies/problems:
< catch exception -- Catch Ada exceptions
< catch handlers -- Catch Ada exceptions
< catch syscall -- Catch system calls by their names
< down-silently -- Same as the `down' command
while the new help is:
> catch exception -- Catch Ada exceptions, when raised.
> catch handlers -- Catch Ada exceptions, when handled.
> catch syscall -- Catch system calls by their names, groups and/or numbers.
> down-silently -- Same as the `down' command, but does not print anything.
Also, the command help doc should not be terminated by a newline
character, but this was not respected by all commands.
The cli-option -OPT framework re-introduced some occurences.
So, the -OPT build help framework was changed to not output newlines at the
end of %OPTIONS% replacement.
This patch changes the help documentations to ensure the 2 invariants
given above.
It implied to slightly rephrase or restructure some help docs.
Based on the above invariants, print_doc_line (called by
'apropos' and 'help' commands to print the first line of a command
help) now outputs the full first line of a command help.
This all results in a lot of small changes in the produced help docs.
There are less code changes than changes in the help docs, as a lot
of docs are produced by some code (e.g. the remote packet usage settings).
gdb/ChangeLog
2019-08-07 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* cli/cli-decode.h (print_doc_line): Add for_value_prefix argument.
* cli/cli-decode.c (print_doc_line): Likewise. It now prints
the full first line, except when FOR_VALUE_PREFIX. In this case,
the trailing '.' is not output, and the first character is uppercased.
(print_help_for_command): Update call to print_doc_line.
(print_doc_of_command): Likewise.
* cli/cli-setshow.c (deprecated_show_value_hack): Likewise.
* cli/cli-option.c (append_indented_doc): Do not append newline.
(build_help_option): Append newline after first appended_indented_doc
only if a second call is done.
(build_help): Append 2 new lines before each option, except the first
one.
* compile/compile.c (_initialize_compile): Add new lines after
%OPTIONS%, when not at the end of the help.
Change help doc or code
producing the help doc to respect the invariants.
* maint-test-options.c (_initialize_maint_test_options): Likewise.
Also removed the new line after 'Options:', as all other commands
do not put an empty line between 'Options:' and the first option.
* printcmd.c (_initialize_printcmd): Likewise.
* stack.c (_initialize_stack): Likewise.
* interps.c (interpreter_exec_cmd): Fix "Usage:" line that was
incorrectly telling COMMAND is optional.
* ada-lang.c (_initialize_ada_language): Change help doc or code
producing the help doc to respect the invariants.
* ada-tasks.c (_initialize_ada_tasks): Likewise.
* breakpoint.c (_initialize_breakpoint): Likewise.
* cli/cli-cmds.c (_initialize_cli_cmds): Likewise.
* cli/cli-logging.c (_initialize_cli_logging): Likewise.
* cli/cli-setshow.c (_initialize_cli_setshow): Likewise.
* cli/cli-style.c (cli_style_option::add_setshow_commands,
_initialize_cli_style): Likewise.
* corelow.c (core_target_info): Likewise.
* dwarf-index-cache.c (_initialize_index_cache): Likewise.
* dwarf2read.c (_initialize_dwarf2_read): Likewise.
* filesystem.c (_initialize_filesystem): Likewise.
* frame.c (_initialize_frame): Likewise.
* gnu-nat.c (add_task_commands): Likewise.
* infcall.c (_initialize_infcall): Likewise.
* infcmd.c (_initialize_infcmd): Likewise.
* interps.c (_initialize_interpreter): Likewise.
* language.c (_initialize_language): Likewise.
* linux-fork.c (_initialize_linux_fork): Likewise.
* maint-test-settings.c (_initialize_maint_test_settings): Likewise.
* maint.c (_initialize_maint_cmds): Likewise.
* memattr.c (_initialize_mem): Likewise.
* printcmd.c (_initialize_printcmd): Likewise.
* python/lib/gdb/function/strfns.py (_MemEq, _StrLen, _StrEq,
_RegEx): Likewise.
* ravenscar-thread.c (_initialize_ravenscar): Likewise.
* record-btrace.c (_initialize_record_btrace): Likewise.
* record-full.c (_initialize_record_full): Likewise.
* record.c (_initialize_record): Likewise.
* regcache-dump.c (_initialize_regcache_dump): Likewise.
* regcache.c (_initialize_regcache): Likewise.
* remote.c (add_packet_config_cmd, init_remote_threadtests,
_initialize_remote): Likewise.
* ser-tcp.c (_initialize_ser_tcp): Likewise.
* serial.c (_initialize_serial): Likewise.
* skip.c (_initialize_step_skip): Likewise.
* source.c (_initialize_source): Likewise.
* stack.c (_initialize_stack): Likewise.
* symfile.c (_initialize_symfile): Likewise.
* symtab.c (_initialize_symtab): Likewise.
* target-descriptions.c (_initialize_target_descriptions): Likewise.
* top.c (init_main): Likewise.
* tracefile-tfile.c (tfile_target_info): Likewise.
* tracepoint.c (_initialize_tracepoint): Likewise.
* tui/tui-win.c (_initialize_tui_win): Likewise.
* utils.c (add_internal_problem_command): Likewise.
* valprint.c (value_print_option_defs): Likewise.
gdb/testsuite/ChangeLog
2019-08-07 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* gdb.base/style.exp: Update tests for help doc new invariants.
* gdb.base/help.exp: Likewise.
Currently we support iteration on blocks; this patch extends that to make
subscript access work as well.
gdb/ChangeLog:
2019-08-05 Christian Biesinger <cbiesinger@google.com>
* NEWS: Mention dictionary access on blocks.
* python/py-block.c (blpy_getitem): New function.
(block_object_as_mapping): New struct.
(block_object_type): Use new struct for tp_as_mapping field.
gdb/doc/ChangeLog:
2019-08-05 Christian Biesinger <cbiesinger@google.com>
* python.texi (Blocks In Python): Document dictionary access on blocks.
gdb/testsuite/ChangeLog:
2019-08-05 Christian Biesinger <cbiesinger@google.com>
* gdb.python/py-block.exp: Test dictionary access on blocks.
Similar to lookup_global_symbol, except that it checks the
STATIC_SCOPE.
gdb/ChangeLog:
2019-07-30 Christian Biesinger <cbiesinger@google.com>
PR/24474: Add a function to lookup static variables.
* NEWS: Mention this new function.
* python/py-symbol.c (gdbpy_lookup_static_symbol): New function.
* python/python-internal.h (gdbpy_lookup_static_symbol): New function.
* python/python.c (python_GdbMethods): Add new function.
gdb/doc/ChangeLog:
2019-07-30 Christian Biesinger <cbiesinger@google.com>
* python.texi (Symbols In Python): Document new function
gdb.lookup_static_symbol.
gdb/testsuite/ChangeLog:
2019-07-30 Christian Biesinger <cbiesinger@google.com>
* gdb.python/py-symbol.c: Add a static variable and one in an anonymous
namespace.
* gdb.python/py-symbol.exp: Test gdb.lookup_static_symbol.
This is essentially the inverse of Symbol.objfile. This allows
handling different symbols with the same name (but from different
objfiles) and can also be faster if the objfile is known.
gdb/ChangeLog:
2019-07-29 Christian Biesinger <cbiesinger@google.com>
* NEWS: Mention new functions Objfile.lookup_{global,static}_symbol.
* python/py-objfile.c (objfpy_lookup_global_symbol): New function.
(objfpy_lookup_static_symbol): New function.
(objfile_object_methods): Add new functions.
gdb/doc/ChangeLog:
2019-07-29 Christian Biesinger <cbiesinger@google.com>
* python.texi (Objfiles In Python): Document new functions
Objfile.lookup_{global,static}_symbol.
gdb/testsuite/ChangeLog:
2019-07-29 Christian Biesinger <cbiesinger@google.com>
* gdb.python/py-objfile.c: Add global and static vars.
* gdb.python/py-objfile.exp: Test new functions Objfile.
lookup_global_symbol and lookup_static_symbol.
New settings allow to better control what frame information is printed.
'set print frame-info' allows to override the default frame information
printed when a GDB command prints a frame.
The backtrace command has a new option -frame-info to override
this global setting.
It is now possible to have very short frame information by using the
new 'set print frame-arguments presence' and
'set print frame-info short-location'.
Combined with 'set print address off', a backtrace will only show
the essential information to see the function call chain, e.g.:
(gdb) set print address off
(gdb) set print frame-arguments presence
(gdb) set print frame-info short-location
(gdb) bt
#0 break_me ()
#1 call_me (...)
#2 main ()
(gdb)
This is handy in particular for big backtraces with functions having
many arguments.
Python frame filter printing logic has been updated to respect the new
setting in non MI mode.
Also, the default frame information printed was inconsistent when
backtrace was printing the frame information itself, or when the python
frame filtering code was printing the frame information.
This patch changes the default of python frame filtering to have a
consistent behaviour regarding printed frame-information, whatever
the presence/activity/matches of python filters.
2019-07-29 Philippe Waroquiers <philippe.waroquiers@skynet.be>
* frame.h (enum print_what): New value 'SHORT_LOCATION', update
comments.
(print_frame_info_auto, print_frame_info_source_line,
print_frame_info_location, print_frame_info_source_and_location,
print_frame_info_location_and_address, print_frame_info_short_location):
New declarations.
(struct frame_print_options): New member print_frame_info.
* extension.h (enum ext_lang_frame_args): New value CLI_PRESENCE.
* stack.h (get_user_print_what_frame_info): New declaration.
(frame_show_address): New declaration.
* stack.c (print_frame_arguments_choices): New value 'presence'.
(print_frame_info_auto, print_frame_info_source_line,
print_frame_info_location, print_frame_info_source_and_location,
print_frame_info_location_and_address, print_frame_info_short_location,
print_frame_info_choices, print_frame_info_print_what): New definitions.
(print_frame_args): Only print dots for args if print frame-arguments
is 'presence'.
(frame_print_option_defs): New element for "frame-info".
(get_user_print_what_frame_info): New function.
(frame_show_address): Make non static. Move comment to stack.h.
(print_frame_info_to_print_what): New function.
(print_frame_info): Update comment. Use fp_opts.print_frame_info
to decide what to print.
(backtrace_command_1): Handle the new print_frame_arguments_presence
value.
(_initialize_stack): Call add_setshow_enum_cmd for frame-info.
* python/py-framefilter.c (py_print_args): Handle CLI_PRESENCE.
(py_print_frame): In non-mi mode, use LOCATION as default for
print_what, similarly to frame information printed directly by
backtrace command. Handle frame-info user option in non MI mode.
Also fixes the date in the changelog of my last commit.
gdb/ChangeLog:
2019-07-25 Christian Biesinger <cbiesinger@google.com>
* python/py-objfile.c (add_separate_debug_file): Fix comment about
this function's Python signature.
commit 05b08ac160 ("Reduce manual reference counting in
py-inferior.c") introduced a use-after-move bug in add_thread_object,
causing a test suite failure. This patch fixes the bug.
Tested on x86-64 Fedora 29.
gdb/ChangeLog
2019-07-11 Tom Tromey <tromey@adacore.com>
* python/py-inferior.c (add_thread_object): Don't use thread_obj
after it has been moved.
I touched symtab.h and was surprised to see how many files were
rebuilt. I looked into it a bit, and found that defs.h includes
gdbarch.h, which in turn includes many things.
gdbarch.h is only needed by a minority ofthe files in gdb, so this
patch removes the include from defs.h and updates the fallout.
I did "wc -l" on the files in build/gdb/.deps; this patch reduces the
line count from 139935 to 137030; so there are definitely future
build-time savings here.
Note that while I configured with --enable-targets=all, it's possible
that some *-nat.c file needs an update. I could not test all of
these. The buildbot caught a few problems along these lines.
gdb/ChangeLog
2019-07-10 Tom Tromey <tom@tromey.com>
* defs.h: Don't include gdbarch.h.
* aarch64-ravenscar-thread.c, aarch64-tdep.c, alpha-bsd-tdep.h,
alpha-linux-tdep.c, alpha-mdebug-tdep.c, arch-utils.h, arm-tdep.h,
ax-general.c, btrace.c, buildsym-legacy.c, buildsym.h, c-lang.c,
cli/cli-decode.h, cli/cli-dump.c, cli/cli-script.h,
cli/cli-style.h, coff-pe-read.h, compile/compile-c-support.c,
compile/compile-cplus.h, compile/compile-loc2c.c, corefile.c,
cp-valprint.c, cris-linux-tdep.c, ctf.c, d-lang.c, d-namespace.c,
dcache.c, dicos-tdep.c, dictionary.c, disasm-selftests.c,
dummy-frame.c, dummy-frame.h, dwarf2-frame-tailcall.c,
dwarf2expr.c, expression.h, f-lang.c, frame-base.c,
frame-unwind.c, frv-linux-tdep.c, gdbarch-selftests.c, gdbtypes.h,
go-lang.c, hppa-nbsd-tdep.c, hppa-obsd-tdep.c, i386-dicos-tdep.c,
i386-tdep.h, ia64-vms-tdep.c, interps.h, language.c,
linux-record.c, location.h, m2-lang.c, m32r-linux-tdep.c,
mem-break.c, memattr.c, mn10300-linux-tdep.c, nios2-linux-tdep.c,
objfiles.h, opencl-lang.c, or1k-linux-tdep.c, p-lang.c,
parser-defs.h, ppc-tdep.h, probe.h, python/py-record-btrace.c,
record-btrace.c, record.h, regcache-dump.c, regcache.h,
riscv-fbsd-tdep.c, riscv-linux-tdep.c, rust-exp.y,
sh-linux-tdep.c, sh-nbsd-tdep.c, source-cache.c,
sparc-nbsd-tdep.c, sparc-obsd-tdep.c, sparc-ravenscar-thread.c,
sparc64-fbsd-tdep.c, std-regs.c, target-descriptions.h,
target-float.c, tic6x-linux-tdep.c, tilegx-linux-tdep.c, top.c,
tracefile.c, trad-frame.c, type-stack.h, ui-style.c, utils.c,
utils.h, valarith.c, valprint.c, varobj.c, x86-tdep.c,
xml-support.h, xtensa-linux-tdep.c, cli/cli-cmds.h: Update.
* s390-linux-nat.c, procfs.c, inf-ptrace.c: Likewise.
This patch changes py-inferior.c to use gdbpy_ref<> when possible,
reducing the amount of manual reference counting.
Tested on x86-64 Fedora 29.
gdb/ChangeLog
2019-07-10 Tom Tromey <tromey@adacore.com>
* python/python-internal.h (create_thread_object): Return
gdbpy_ref.
* python/py-infthread.c (create_thread_object): Return gdbpy_ref.
* python/py-inferior.c (struct threadlist_entry): Add
constructor.
<thread_obj>: Now a gdbpy_ref.
(thread_to_thread_object): Update.
(add_thread_object): Use new.
(delete_thread_object): Use delete.
(infpy_threads): Update.
(py_free_inferior): Update. Construct "inf_obj" after acquiring
GIL.
Currently, with:
(gdb) catch catch
Catchpoint 1 (catch)
(gdb) catch throw
Catchpoint 2 (throw)
(gdb) catch rethrow
Catchpoint 3 (rethrow)
You get:
(gdb) info breakpoints
Num Type Disp Enb Address What
1 breakpoint keep y 0x0000000000b122af exception catch
2 breakpoint keep y 0x0000000000b1288d exception throw
3 breakpoint keep y 0x0000000000b12931 exception rethrow
I think it doesn't make much sense usability-wise, to show a
catchpoint as a breakpoint. The fact that GDB sets a breakpoint at
some magic address in the C++ run time is an implementation detail,
IMO. And as seen in the previous patch, such a catchpoint can end up
with more than one location/address even, so showing a single address
isn't entirely accurate.
This commit hides the addresses from view, and makes GDB show
"catchpoint" for type as well:
(gdb) info breakpoints
Num Type Disp Enb Address What
1 catchpoint keep y exception catch
2 catchpoint keep y exception throw
3 catchpoint keep y exception rethrow
This comment in the code seems telling:
/* We need to reset 'type' in order for code in breakpoint.c to do
the right thing. */
cp->type = bp_breakpoint;
It kind of suggests that the reason catchpoints end up shown as
breakpoints was that it was easier to implement them that way, rather
than a desired property.
This commit fixes things up to make it possible to have bp_catch
breakpoints have software/hardware breakpoint locations, thus
eliminating the need for that hack:
- redo breakpoint_address_is_meaningful in terms of the location's
type rather than breakpoint type.
- teach bpstat_what about stepping over the catchpoint locations.
- install a allocate_location method for "catch catch/throw/rethrow",
one that forces the location type.
Note that this also reverts the gdb hunk from:
commit 2a8be20359
Commit: Tom Tromey <tom@tromey.com>
CommitDate: Sat Oct 6 22:17:45 2018 -0600
Fix Python gdb.Breakpoint.location crash
because now "catch throw" catchpoints hit the
if (obj->bp->type != bp_breakpoint)
Py_RETURN_NONE;
check above, and, adjusts the testcase to no longer expect to see the
catchpoint in the gdb.breakpoints() list.
(Note: might make sense to do the same to Ada exception catchpoints.)
gdb/ChangeLog:
2019-07-09 Pedro Alves <palves@redhat.com>
* break-catch-throw.c (print_one_exception_catchpoint): Skip the
"addr" field.
(allocate_location_exception_catchpoint): New.
(handle_gnu_v3_exceptions): Don't reset 'type' to bp_breakpoint.
(initialize_throw_catchpoint_ops): Install
allocate_location_exception_catchpoint as allocate_location
method.
* breakpoint.c (bpstat_what) <bp_catch>: Set action to
BPSTAT_WHAT_SINGLE if not stopping and the location's type is not
bp_loc_other.
(breakpoint_address_is_meaningful): Delete.
(bl_address_is_meaningful): New.
(breakpoint_locations_match): Adjust comment.
(bp_location_from_bp_type): New, factored out of...
(bp_location::bp_location(breakpoint *)): ... this.
(bp_location::bp_location(breakpoint *, bp_loc_type)): New,
factored out of...
(bp_location::bp_location(breakpoint *)): ... this. Reimplement.
(bp_loc_is_permanent): Use bl_address_is_meaningful instead of
breakpoint_address_is_meaningful.
(bp_locations_compare): Adjust comment.
(update_global_location_list): Use bl_address_is_meaningful
instead of breakpoint_address_is_meaningful.
* breakpoint.h (bp_location::bp_location(breakpoint *)): New
explicit.
(bp_location::bp_location(breakpoint *, bp_loc_type)): Declare.
* python/py-breakpoint.c (bppy_get_location): No longer check
whether location is null.
gdb/doc/ChangeLog:
2019-07-09 Pedro Alves <palves@redhat.com>
* gdb.texinfo (C++ Exception GDB/MI Catchpoint Commands): Adjust
examples to show type=catchpoint instead of type=breakpoint and an
address.
gdb/testsuite/ChangeLog:
2019-07-09 Pedro Alves <palves@redhat.com>
* gdb.cp/catch-multi-stdlib.exp: Adjust expected "info
breakpoints" output.
* gdb.cp/exception.exp: Adjust expected "info breakpoints" output.
* gdb.python/py-breakpoint.exp: No longer expect that "catch
throw" creates breakpoint.
* gdb.mi/mi-catch-cpp-exceptions.exp (setup_catchpoint): Expect
'type="catchpoint"'.
This bug was originally reported against Fedora GDB:
https://bugzilla.redhat.com/show_bug.cgi?id=1723564
The problem is that GDB will crash in the following scenario:
- PYTHONMALLOC=debug or PYTHONDEVMODE=1 is set.
- The Python debuginfo is installed.
- GDB is used to debug Python.
The crash looks like this:
$ PYTHONMALLOC=debug gdb -args python3 -c pass
GNU gdb (GDB) Fedora 8.3-3.fc30
Reading symbols from python3...
Reading symbols from /usr/lib/debug/usr/bin/python3.7m-3.7.3-3.fc30.x86_64.debug...
(gdb) run
Starting program: /usr/bin/python3 -c pass
Missing separate debuginfos, use: dnf debuginfo-install glibc-2.29-9.fc30.x86_64
Debug memory block at address p=0x5603977bf330: API ''
8098648152243306496 bytes originally requested
The 7 pad bytes at p-7 are not all FORBIDDENBYTE (0xfb):
at p-7: 0x03 *** OUCH
at p-6: 0x00 *** OUCH
at p-5: 0x00 *** OUCH
at p-4: 0x00 *** OUCH
at p-3: 0x00 *** OUCH
at p-2: 0x00 *** OUCH
at p-1: 0x00 *** OUCH
Because memory is corrupted at the start, the count of bytes requested
may be bogus, and checking the trailing pad bytes may segfault.
The 8 pad bytes at tail=0x706483999ad1f330 are Segmentation fault (core dumped)
It's hard to determine what happens, but after doing some
investigation and talking to Victor Stinner I found that GDB should
not use the Python memory allocation functions before the Python
interpreter is initialized (which makes sense). However, we do just
that on python/python.c:do_start_initialization:
...
progsize = strlen (progname.get ());
progname_copy = (wchar_t *) PyMem_Malloc ((progsize + 1) * sizeof (wchar_t));
...
/* Note that Py_SetProgramName expects the string it is passed to
remain alive for the duration of the program's execution, so
it is not freed after this call. */
Py_SetProgramName (progname_copy);
...
Py_Initialize ();
PyEval_InitThreads ();
Upon reading the Python 3 C API documentation, I
found (https://docs.python.org/3.5/c-api/memory.html):
To avoid memory corruption, extension writers should never try to
operate on Python objects with the functions exported by the C
library: malloc(), calloc(), realloc() and free(). This will result in
mixed calls between the C allocator and the Python memory manager with
fatal consequences, because they implement different algorithms and
operate on different heaps. However, one may safely allocate and
release memory blocks with the C library allocator for individual
purposes[...]
And Py_SetProgramName seems like a very simple call that doesn't need
a Python-allocated memory to work on. So I'm proposing this patch,
which simply replaces PyMem_Malloc by xmalloc.
Testing this is more complicated. First, the crash is completely
non-deterministic; I was able to reproduce it 10 times in a row, and
then I wasn't able to reproduce it anymore. I found that if you
completely remove your build directory and rebuild GDB from scratch,
you can reproduce it again confidently. And with my patch, I
confirmed that the bug doesn't manifest even in this situation.
No regressions found.
OK to apply?
gdb/ChangeLog:
2019-06-28 Sergio Durigan Junior <sergiodj@redhat.com>
PR python/24742
https://bugzilla.redhat.com/show_bug.cgi?id=1723564
* python/python.c (do_start_initialization): Use 'xmalloc'
instead of 'PyMem_Malloc'.
Joel Brobecker