Simon Marchi [Sun, 27 Nov 2016 03:05:42 +0000 (22:05 -0500)]
Constify wrap_here/wrap_hint code path
Constify the data path between ui_out_wrap_hint and the wrap_indent
global, because we can. It's clearer that the argument passed to
wrap_hint is not intended to be modified by the ui_out implementation.
Simon Marchi [Sun, 27 Nov 2016 03:05:06 +0000 (22:05 -0500)]
Remove ui_out_destroy
It's not actually used, and removing it simplifies the upcoming patches
a bit. After the whole series, destroying an ui_out object will be
simply "delete uiout", which will call the default destructor.
Simon Marchi [Sun, 27 Nov 2016 03:04:49 +0000 (22:04 -0500)]
Rename ui_out_data to mi_ui_out_data
Just a little cleanup, so the name is more consistent with the naming of
the equivalent structures of cli and tui. It goes away in subsequent
patches anyway, but it might help follow the changes in those patches...
gdb/ChangeLog:
* mi/mi-out.c (ui_out_data): Rename to ...
(mi_ui_out_data): ... this.
Jose E. Marchesi [Fri, 25 Nov 2016 11:40:15 +0000 (03:40 -0800)]
gas: fix CBCOND diagnostics for invalid immediate operands.
This patch fixes two problems in the SPARC assembler:
- The diagnostic message
Error: Illegal operands: Immediate value in cbcond is out of range.
is incorrectly issued for non-CBCOND instructions that feature a
simm5 immediate field, such as MPMUL, MONTMUL, etc.
- When an invalid immediate operand is used in a CBCOND
instruction, two redundant error messages are issued to the
user, the second due to a stale fixup (this happens since
commit 85024cd8bcb93f4112470ecdbd6c10fc2aea724f).
Some diagnostic tests for the CBCOND instructions are also
included in the patch.
Tested in both sparc64-linux-gnu and sparcv9-linux-gnu targets.
gas/ChangeLog:
2016-11-25 Jose E. Marchesi <jose.marchesi@oracle.com>
* config/tc-sparc.c (sparc_ip): Avoid emitting a cbcond error
messages for non-cbcond instructions.
* testsuite/gas/sparc/cbcond-diag.s: New file.
* testsuite/gas/sparc/cbcond-diag.l: Likewise.
* testsuite/gas/sparc/sparc.exp (gas_64_check): Run cbcond-diag tests.
John Baldwin [Thu, 24 Nov 2016 20:01:24 +0000 (12:01 -0800)]
Add noexcept to custom non-throwing new operators.
Both libc++ and libstdc++ declare non-throwing new operators as
noexcept and overloads must also be noexcept. This fixes a
-Wmissing-exception-spec warning with clang.
gdb/ChangeLog:
* common/new-op.c (operator new): Mark 'noexcept'.
(operator new[]): Likewise.
Andreas Arnez [Thu, 24 Nov 2016 16:48:04 +0000 (17:48 +0100)]
Optimize byte-aligned copies in copy_bitwise()
The function copy_bitwise used for copying DWARF pieces can potentially
be invoked for large chunks of data. For instance, consider a large
struct one of whose members is currently located in a register. In this
case copy_bitwise would still copy the data bitwise in a loop, which is
much slower than necessary.
This change uses memcpy for the large part instead, if possible.
gdb/ChangeLog:
* dwarf2loc.c (copy_bitwise): Use memcpy for the middle part, if
it is byte-aligned.
Andreas Arnez [Thu, 24 Nov 2016 16:48:04 +0000 (17:48 +0100)]
Add unit test for copy_bitwise
This adds a unit test for the copy_bitwise function in dwarf2loc.c.
With the old (broken) version of copy_bitwise this test would generate
the following failure message:
gdb/ChangeLog:
2016-11-24 Andreas Arnez <arnez@linux.vnet.ibm.com>
Pedro Alves <palves@redhat.com>
* dwarf2loc.c (bits_to_str, check_copy_bitwise)
(copy_bitwise_tests): New functions.
(_initialize_dwarf2loc): Register the new function
copy_bitwise_tests as a unit test.
* selftest.c (run_self_tests): Improve the failure message's
wording and formatting.
Andreas Arnez [Thu, 24 Nov 2016 16:48:03 +0000 (17:48 +0100)]
Fix copy_bitwise()
When the user writes or reads a variable whose location is described
with DWARF pieces (DW_OP_piece or DW_OP_bit_piece), GDB's helper
function copy_bitwise is invoked for each piece. The implementation of
this function has a bug that may result in a corrupted copy, depending
on alignment and bit size. (Full-byte copies are not affected.)
This rewrites copy_bitwise, replacing its algorithm by a fixed version,
and adding an appropriate test case. Without the fix the new test case
fails, e.g.:
Which means that two zero bits have sneaked into the copy of the
original all-one bit pattern. The test uses this simple all-one value
in order to avoid another GDB bug that causes the DWARF piece of a
DW_OP_stack_value to be taken from the wrong end on big-endian
architectures.
gdb/ChangeLog:
* dwarf2loc.c (extract_bits_primitive): Remove.
(extract_bits): Remove.
(copy_bitwise): Rewrite. Fixes a possible corruption that may
occur for non-byte-aligned copies.
gdb/testsuite/ChangeLog:
* gdb.dwarf2/nonvar-access.exp: Add a test for accessing
non-byte-aligned bit fields.
Andreas Arnez [Thu, 24 Nov 2016 16:48:03 +0000 (17:48 +0100)]
Fix PR12616 - gdb does not implement DW_AT_data_bit_offset
The DW_AT_data_bit_offset attribute was introduced by DWARF V4 and
allows specifying the offset of a data member within its containing
entity. But although the new attribute was intended to replace
DW_AT_bit_offset for this purpose, GDB ignores it, and thus GCC still
emits DW_AT_bit_offset instead. See also
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71669.
This change fixes GDB's lack of support for DW_AT_data_bit_offset and
adds an appropriate test case.
Jiong Wang [Thu, 24 Nov 2016 14:01:53 +0000 (14:01 +0000)]
[ARM] Bind defined symbol locally in PIE
bfd/
PR target/20737
* elf32-arm.c (elf32_arm_final_link_relocate): Bind defined symbol
locally in PIE.
ld/
* testsuite/ld-arm/pie-bind-locally-a.s: New test source.
* testsuite/ld-arm/pie-bind-locally-b.s: Likewise.
* testsuite/ld-arm/pie-bind-locally.d: New testcase.
* testsuite/ld-arm/arm-elf.exp: Run new testcase.
Pedro Alves [Wed, 23 Nov 2016 15:36:26 +0000 (15:36 +0000)]
gdb: Use C++11 std::chrono
This patch fixes a few problems with GDB's time handling.
#1 - It avoids problems with gnulib's C++ namespace support
On MinGW, the struct timeval that should be passed to gnulib's
gettimeofday replacement is incompatible with libiberty's
timeval_sub/timeval_add. That's because gnulib also replaces "struct
timeval" with its own definition, while libiberty expects the
system's.
That's currently handled in gdb by not using gnulib's gettimeofday at
all (see common/gdb_sys_time.h), but that #undef hack won't work with
if/when we enable gnulib's C++ namespace support, because that mode
adds compile time warnings for uses of ::gettimeofday, which are hard
errors with -Werror.
#2 - But there's an elephant in the room: gettimeofday is not monotonic...
We're using it to:
a) check how long functions take, for performance analysis
b) compute when in the future to fire events in the event-loop
c) print debug timestamps
But that's exactly what gettimeofday is NOT meant for. Straight from
the man page:
~~~
The time returned by gettimeofday() is affected by
discontinuous jumps in the system time (e.g., if the system
administrator manually changes the system time). If you need a
monotonically increasing clock, see clock_gettime(2).
~~~
std::chrono (part of the C++11 standard library) has a monotonic clock
exactly for such purposes (std::chrono::steady_clock). This commit
switches to use that instead of gettimeofday, fixing all the issues
mentioned above.
gdb/ChangeLog:
2016-11-23 Pedro Alves <palves@redhat.com>
* Makefile.in (SFILES): Add common/run-time-clock.c.
(HFILES_NO_SRCDIR): Add common/run-time-clock.h.
(COMMON_OBS): Add run-time-clock.o.
* common/run-time-clock.c, common/run-time-clock.h: New files.
* defs.h (struct timeval, print_transfer_performance): Delete
declarations.
* event-loop.c (struct gdb_timer) <when>: Now a
std::chrono::steady_clock::time_point.
(create_timer): use std::chrono::steady_clock instead of
gettimeofday. Use new instead of malloc.
(delete_timer): Use delete instead of xfree.
(duration_cast_timeval): New.
(update_wait_timeout): Use std::chrono::steady_clock instead of
gettimeofday.
* maint.c: Include <chrono> instead of "gdb_sys_time.h", <time.h>
and "timeval-utils.h".
(scoped_command_stats::~scoped_command_stats)
(scoped_command_stats::scoped_command_stats): Use
std::chrono::steady_clock instead of gettimeofday. Use
user_cpu_time_clock instead of get_run_time.
* maint.h: Include "run-time-clock.h" and <chrono>.
(scoped_command_stats): <m_start_cpu_time>: Now a
user_cpu_time_clock::time_point.
<m_start_wall_time>: Now a std::chrono::steady_clock::time_point.
* mi/mi-main.c: Include "run-time-clock.h" and <chrono> instead of
"gdb_sys_time.h" and <sys/resource.h>.
(rusage): Delete.
(mi_execute_command): Use new instead of XNEW.
(mi_load_progress): Use std::chrono::steady_clock instead of
gettimeofday.
(timestamp): Rewrite in terms of std::chrono::steady_clock,
user_cpu_time_clock and system_cpu_time_clock.
(timeval_diff): Delete.
(print_diff): Adjust to use std::chrono::steady_clock,
user_cpu_time_clock and system_cpu_time_clock.
* mi/mi-parse.h: Include "run-time-clock.h" and <chrono> instead
of "gdb_sys_time.h".
(struct mi_timestamp): Change fields types to
std::chrono::steady_clock::time_point, user_cpu_time_clock::time
and system_cpu_time_clock::time_point, instead of struct timeval.
* symfile.c: Include <chrono> instead of <time.h> and
"gdb_sys_time.h".
(struct time_range): New.
(generic_load): Use std::chrono::steady_clock instead of
gettimeofday.
(print_transfer_performance): Replace timeval parameters with a
std::chrono::steady_clock::duration parameter. Adjust.
* utils.c: Include <chrono> instead of "timeval-utils.h",
"gdb_sys_time.h", and <time.h>.
(prompt_for_continue_wait_time): Now a
std::chrono::steady_clock::duration.
(defaulted_query, prompt_for_continue): Use
std::chrono::steady_clock instead of
gettimeofday/timeval_sub/timeval_add.
(reset_prompt_for_continue_wait_time): Use
std::chrono::steady_clock::duration instead of struct timeval.
(get_prompt_for_continue_wait_time): Return a
std::chrono::steady_clock::duration instead of struct timeval.
(vfprintf_unfiltered): Use std::chrono::steady_clock instead of
gettimeofday. Use std::string. Use '.' instead of ':'.
* utils.h: Include <chrono>.
(get_prompt_for_continue_wait_time): Return a
std::chrono::steady_clock::duration instead of struct timeval.
gdb/gdbserver/ChangeLog:
2016-11-23 Pedro Alves <palves@redhat.com>
* debug.c: Include <chrono> instead of "gdb_sys_time.h".
(debug_vprintf): Use std::chrono::steady_clock instead of
gettimeofday. Use '.' instead of ':'.
* tracepoint.c: Include <chrono> instead of "gdb_sys_time.h".
(get_timestamp): Use std::chrono::steady_clock instead of
gettimeofday.
Simon Marchi [Tue, 22 Nov 2016 21:14:24 +0000 (16:14 -0500)]
Normalize names of some source files
Most tdep/nat files are named:
<cpu>-<os>-tdep.c
<cpu>-<os>-nat.c
A few files do not respect this scheme. This patch renames them so that
they are consistent with the rest of the files. It builds fine with
--enable-targets=all, but that doesn't test the nat files. I can only
hope that my grep skill is good enough.
Simon Marchi [Tue, 22 Nov 2016 21:14:22 +0000 (16:14 -0500)]
Makefiles: Flatten and sort file lists
I find the big file lists in the Makefiles a bit ugly and not very
practical. Since there are multiple filenames on each line (as much as
fits in 80 columns), it's not easy to add, remove or change a name in
the middle. As a result, we have a mix of long and short lines in no
particular order (ALL_TARGET_OBS is a good example).
I therefore suggest flattening the lists (one name per line) and keeping
them in alphabetical order. The diffs will be much clearer and merge
conflicts will be easier to resolve.
A nice (IMO) side-effect I observed is that the files are compiled
alphabetically by make, so it gives a rough idea of the progress of the
build.
I added a comment in gdb/Makefile.in to mention to keep the file lists
ordered, and gave the general guidelines on what order to respect. I
added a comment in other Makefiles which refers to gdb/Makefile.in, to
avoid duplication.
Running the patch through the buildbot found that gdb.base/default.exp
started to fail. The languages in the error message shown when typing
"set language" have changed order. We could probably improve gdb so
that it prints them in a stable order, regardless of the order of the
object list passed to the linked, but just fixing the test is easier for
now.
New in v2:
- Change ordering style, directories go at the end.
- Cleanup gdbserver's and data-directory's Makefile as well.
- Add comments at top of Makefiles about the ordering.
- Remove wrong trailing backslahes.
- Fix test gdb.base/default.exp.
gdb/ChangeLog:
* Makefile.in: Add comment about file lists ordering.
(SUBDIR_CLI_OBS, SUBDIR_CLI_SRCS, SUBDIR_MI_OBS, SUBDIR_MI_SRCS,
SUBDIR_TUI_OBS, SUBDIR_TUI_SRCS, SUBDIR_GCC_COMPILE_OBS,
SUBDIR_GCC_COMPILE_SRCS, SUBDIR_GUILE_OBS, SUBDIR_GUILE_SRCS,
SUBDIR_PYTHON_OBS, SUBDIR_PYTHON_SRCS, SUBDIR_GDBTK_OBS,
SUBDIR_GDBTK_SRCS, XMLFILES, REMOTE_OBS, ALL_64_TARGET_OBS,
ALL_TARGET_OBS, SFILES, HFILES_NO_SRCDIR, HFILES_WITH_SRCDIR,
COMMON_OBS, YYFILES, YYOBJ, generated_files, ALLDEPFILES):
Flatten list and order alphabetically.
* data-directory/Makefile.in: Add comment about file lists
ordering.
(GEN_SYSCALLS_FILES, PYTHON_FILE_LIST): Flatten list and order
alphabetically.
gdb/gdbserver/ChangeLog:
* Makefile.in (SFILES, OBS): Flatten list and order
alphabetically.
gdb/testsuite/ChangeLog:
* gdb.base/default.exp: Fix output of "set language".
Nick Clifton [Wed, 23 Nov 2016 11:10:39 +0000 (11:10 +0000)]
Fix the linker so that it will not silently generate ELF binaries with invalid program headers. Fix readelf to report such invalid binaries.
PR ld/20815
bfd * elf.c (elf_modify_segment_map): Allow empty LOAD segments if
they contain the program headers.
(_bfd_elf_map_sections_to_segments): If the linker created the
PHDR segment then always attempt to include it in a LOAD segment.
(assign_file_positions_for_non_load_sections): Allow LOAD segments
to overlap PHDR segments.
(phdr_sorter): New function. Sorts program headers.
(assign_file_positions_except_relocs): Sort the program headers
before writing them out. Issue an error if the PHDR segment is
not covered by a LOAD segment, unless the backend allows it.
* elf-bfd.h (struct elf_backend_data): Add
elf_backend_allow_non_load_phdr.
* elfxx-target.h (elf_backend_allow_non_load_phdr): Provide
default definition that returns FALSE.
(elfNN_bed): Initialise the elf_backend_allow_non_load_phdr
field.
* elf64-hppa.c (elf64_hppa_allow_non_load_phdr): New function.
Returns TRUE.
(elf_backend_allow_non_load_phdr): Define.
* elf-m10300.c (_bfd_mn10300_elf_size_dynamic_sections): Do not
place the interpreter string into the .interp section if the
nointerp flag is set in the link info structure.
* elf32-arc.c (elf_arc_size_dynamic_sections): Likewise.
* elf32-score7.c (score_elf_final_link_relocate): Allow for the
_gp symbol not being part of the output.
binutils* readelf.c (process_program_headers): Check PT_LOAD and PT_PHDR
segments for validity.
ld * ld.texinfo: Note that PT_TLS can be used as a segment type.
* testsuite/ld-discard/discard.ld: Add space for program headers.
* testsuite/ld-elf/flags1.ld: Likewise.
* testsuite/ld-elf/maxpage3.t: Likewise.
* testsuite/ld-elf/noload-1.t: Likewise.
* testsuite/ld-elf/orphan.ld: Likewise.
* testsuite/ld-elf/overlay.t: Likewise.
* testsuite/ld-elf/pr14052.t: Likewise.
* testsuite/ld-elf/pr19539.t: Likewise.
* testsuite/ld-elf/provide-hidden-1.ld: Likewise.
* testsuite/ld-elf/provide-hidden-s.ld: Likewise.
* testsuite/ld-elf/weak-dyn-1.ld: Likewise.
* testsuite/ld-i386/pr19539.t: Likewise.
* testsuite/ld-scripts/defined.t: Likewise.
* testsuite/ld-scripts/defined6.t: Likewise.
* testsuite/ld-scripts/dynamic-sections.t: Likewise.
* testsuite/ld-scripts/empty-aligned.t: Likewise.
* testsuite/ld-scripts/provide-2.t: Likewise.
* testsuite/ld-scripts/provide-4.t: Likewise.
* testsuite/ld-vax-elf/plt-local.ld: Likewise.
* testsuite/ld-x86-64/pr19539.t: Likewise.
* testsuite/ld-elf/ehdr_start-missing.d: Do not initialise the
dynamic linker.
* testsuite/ld-elf/ehdr_start-weak.d: Likewise.
* testsuite/ld-elf/elf.exp (pr14170, pr17068): Likewise.
* testsuite/ld-elf/loadaddr1.d: Update expected readelf output.
* testsuite/ld-elf/noload-2.d: Likewise.
* testsuite/ld-powerpc/vxworks2.sd: Likewise.
* testsuite/ld-scripts/phdrs3a.d: Likewise.
* testsuite/ld-scripts/size-2.d: Likewise.
* testsuite/ld-elf/group.ld: Add program headers.
* testsuite/ld-elf/overlay.d: Skip for SPU.
* testsuite/ld-elf/flags1.d: Skip for RX.
* testsuite/ld-elf/pr19162.d: Skip for HPPA64.
* testsuite/ld-elf/pr19539.d: Skip for ALPHA.
* testsuite/ld-scripts/empty-orphan.t: Update program headers.
* testsuite/ld-scripts/size-2.t: Likewise.
Alan Modra [Wed, 23 Nov 2016 04:37:17 +0000 (15:07 +1030)]
elf_backend_dtrel_excludes_plt
Now that all targets creating .rel.plt/.rela.plt use the ELF hash
table shortcut srelplt, the generic ELF code can set up DT_RELSZ/
DT_RELASZ and DT_REL/DT_RELA for targets that don't want PLT relocs
included in those tags.
Alan Modra [Wed, 23 Nov 2016 04:36:34 +0000 (15:06 +1030)]
Delete duplicate target short-cuts to dynamic sections
We'd like to have the elf_link_hash_table srelplt field rather than
some private target field used to save short-cuts to a PLT relocation
section. This save a little space but mainly is so that the generic
ELF code can access the field. Ditto for other dynamic sections.
Cary Coutant [Tue, 22 Nov 2016 23:44:56 +0000 (15:44 -0800)]
Print the default for all binary options; clean up --help output.
gold/
PR gold/20346
* options.cc (One_option::print): Print "(default)" when appropriate.
* options.h: Clean up and re-sort options.
(One_option::is_default): New data member.
(One_option::One_option): Add is_default parameter; adjust all calls.
(DEFINE_var): Add is_default__ parameter; adjust all calls.
(DEFINE_bool): Set is_default based on default_value__.
(DEFINE_bool_ignore): New macro.
(--no-eh-frame-hdr): New option.
(--enable-new-dtags): Remove mention of DT_FLAGS.
Yao Qi [Tue, 22 Nov 2016 14:05:06 +0000 (14:05 +0000)]
Change gdbarch software_single_step frame_info to regcache
This patch changes gdbarch method software_single_step's parameter from
"struct frame_info *" to "struct regcache *, IOW, software_single_step
starts to use current regcache rather than current frame for software
single.
Jose E. Marchesi [Tue, 22 Nov 2016 12:40:37 +0000 (04:40 -0800)]
gas,opcodes: fix hardware capabilities bumping in the sparc assembler.
When the assembler finds an instruction which is part of a higher
opcode architecture it bumps the current opcode architecture. For
example:
$ echo "mwait" | as -bump
{standard input}: Assembler messages:
{standard input}:1: Warning: architecture bumped from "v6" to "v9m" on "mwait"
However, when two instructions pertaining to the same opcode
architecture but associated to different SPARC hardware capabilities
are found in the input stream, and no GAS architecture is specified in
the command line, the assembler bangs:
$ echo "mwait; wr %g0,%g1,%mcdper" | as -bump
{standard input}: Assembler messages:
{standard input}:1: Warning: architecture bumped from "v6" to "v9m" on "mwait"
{standard input}:1: Error: Hardware capability "sparc5" not enabled for "wr".
... and it should'nt, as WRMCDPER pertains to the same architecture
level than MWAIT.
This patch fixes this by extending the definition of sparc opcode
architectures to contain a set of hardware capabilities and making the
assembler to take these capabilities into account when updating the
set of allowed hwcaps when an architecture bump is triggered by some
instruction.
This way, hwcaps associated to architecture levels are maintained in
opcodes, while the assembler keeps the flexibiity of defining GAS
architectures including additional hwcaps (like -Asparcfmaf or the
v8plus* variants).
A test covering this failure case is included.
gas/ChangeLog:
2016-11-22 Jose E. Marchesi <jose.marchesi@oracle.com>
* config/tc-sparc.c: Move HWS_* and HWS2_* definitions to
opcodes/sparc-opc.c.
(sparc_arch): Clarify the new role of the hwcap_allowed and
hwcap2_allowed fields.
(sparc_arch_table): Remove HWS_* and HWS2_* instances from
hwcap_allowed and hwcap2_allowed respectively.
(md_parse_option): Include the opcode arch hwcaps when processing
-A.
(sparc_ip): Use the current opcode arch hwcaps to update
hwcap_allowed, as well of the hwcaps of the instruction triggering
the bump.
* testsuite/gas/sparc/hwcaps-bump.s: New file.
* testsuite/gas/sparc/hwcaps-bump.l: Likewise.
* testsuite/gas/sparc/sparc.exp (gas_64_check): Run tests in
hwcaps-bump.
include/ChangeLog:
2016-11-22 Jose E. Marchesi <jose.marchesi@oracle.com>
* opcode/sparc.h (sparc_opcode_arch): New fields hwcaps and
hwcaps2.
opcodes/ChangeLog:
2016-11-22 Jose E. Marchesi <jose.marchesi@oracle.com>
* sparc-opc.c (HWS_V8): Definition moved from
gas/config/tc-sparc.c.
(HWS_V9): Likewise.
(HWS_VA): Likewise.
(HWS_VB): Likewise.
(HWS_VC): Likewise.
(HWS_VD): Likewise.
(HWS_VE): Likewise.
(HWS_VV): Likewise.
(HWS_VM): Likewise.
(HWS2_VM): Likewise.
(sparc_opcode_archs): Initialize hwcaps and hwcaps2 fields of
existing entries.
Alan Modra [Tue, 22 Nov 2016 05:57:12 +0000 (16:27 +1030)]
Use input_bfd in relocate_section
It makes just a little more sense to use input_bfd when retrieving
insns for relocation, since the relocations match the endianness of
the input bfd.
* elf32-ppc.c (ppc64_elf_relocate_section): Calculate d_offset for
input_bfd. Replace occurrences of output_bfd as bfd_get_32 and
bfd_put_32 param with input_bfd.
* elf32-ppc.c (ppc_elf_relocate_section): Likewise. Also
ppc_elf_vle_split16 param.
(ppc_elf_vle_split16): Rename output_bfd param to input_bfd.
Yao Qi [Tue, 22 Nov 2016 08:53:34 +0000 (08:53 +0000)]
Use VALUE_NEXT_FRAME_ID in value_from_component
We renamed VALUE_FRAME_ID to VALUE_NEXT_FRAME_ID recently,
https://sourceware.org/ml/gdb-patches/2016-11/msg00018.html
and we should use VALUE_NEXT_FRAME_ID in value_from_component
too.
gdb:
2016-11-22 Yao Qi <yao.qi@linaro.org>
* value.c (value_from_component): Use VALUE_NEXT_FRAME_ID
instead of VALUE_FROM_ID.
BFD/DWARF2: Correct an `index' global shadowing error
Fix a commit 089e3718bd8d ("Greatly improve the speed if looking up
DWARF line number information.") build regression:
cc1: warnings being treated as errors
.../bfd/dwarf2.c: In function 'build_line_info_table':
.../bfd/dwarf2.c:1614: warning: declaration of 'index' shadows a global declaration
/usr/include/string.h:304: warning: shadowed declaration is here
.../bfd/dwarf2.c: In function 'build_lookup_funcinfo_table':
.../bfd/dwarf2.c:2262: warning: declaration of 'index' shadows a global declaration
/usr/include/string.h:304: warning: shadowed declaration is here
make[4]: *** [dwarf2.lo] Error 1
in a way following commit 91d6fa6a035c ("Add -Wshadow to the gcc command
line options used when compiling the binutils.").
bfd/
* dwarf2.c (build_line_info_table): Rename `index' local
variable to `line_index'.
(build_lookup_funcinfo_table): Rename `index' local variable to
`func_index'.
Yao Qi [Mon, 21 Nov 2016 14:15:06 +0000 (14:15 +0000)]
Create subobject value in pretty printer
Nowadays, we create a value of subobject in pretty printer with 'address'
being used,
value = value_from_contents_and_address (type, valaddr + embedded_offset,
address + embedded_offset);
set_value_component_location (value, val);
/* set_value_component_location resets the address, so we may
need to set it again. */
if (VALUE_LVAL (value) != lval_internalvar
&& VALUE_LVAL (value) != lval_internalvar_component
&& VALUE_LVAL (value) != lval_computed)
set_value_address (value, address + embedded_offset);
value_from_contents_and_address creates a value from memory, but the
value we are pretty-printing may not from memory at all.
Instead of using value_from_contents_and_address, we create a value
of subobject with the same location as object's but different offset.
We avoid using address in this way. As a result, parameter 'address'
in apply_val_pretty_printer is no longer needed, we can remove it in
next step.
We've already had the location of the 'whole' value, so it is safe
to assume we can create a value of 'component' or 'suboject' value
at the same location but with different offset.
Joel Brobecker [Sat, 19 Nov 2016 18:40:17 +0000 (10:40 -0800)]
ARI: Add detection of printf_vma and sprintf_vma
We shouldn't be using these, since their output goes straight to
stdout, which doesn't allow redirection. So this patch updates
the ARI to detect any such use.
gdb/ChangeLog:
* contrib/ari/gdb_ari.sh: Add detection of printf_vma and
sprintf_vma.
As it turns out, GCC (and the assembler) needs additional work in
order to support negative GOT offsets in 64-bit sparc. This is
breaking TLS Local Dynamic in position-independent code.
* testsuite/gas/arc/cl-warn.s: New file.
* testsuite/gas/arc/cpu-pseudop-1.d: Likewise.
* testsuite/gas/arc/cpu-pseudop-1.s: Likewise.
* testsuite/gas/arc/cpu-pseudop-2.d: Likewise.
* testsuite/gas/arc/cpu-pseudop-2.s: Likewise.
* testsuite/gas/arc/cpu-warn2.s: Likewise.
* config/tc-arc.c (selected_cpu): Initialize.
(feature_type): New struct.
(feature_list): New variable.
(arc_check_feature): New function.
(arc_select_cpu): Check for .cpu duplicates. Don't overwrite the
current cpu features. Check if a feature is available for a given
cpu.
(md_parse_option): Test if features are available for a given cpu.
James Clarke [Fri, 18 Nov 2016 11:51:40 +0000 (12:51 +0100)]
bfd: fix negative GOT offsets for non-local references on sparc64
bfd/ChangeLog:
2016-11-18 James Clarke <jrtc27@jrtc27.com>
* elfxx-sparc.c (_bfd_sparc_elf_relocate_section): Don't convert
R_SPARC_GOTDATA_OP_HIX22 and R_SPARC_GOTDATA_OP_LOX10 to
R_SPARC_GOT* for non-local references. Instead, treat them like
R_SPARC_GOTDATA_HIX22/R_SPARC_GOTDATA_LOX10 when filling in the
immediate with the calculated relocation.
David Tolnay [Wed, 16 Nov 2016 23:09:27 +0000 (23:09 +0000)]
libiberty: Add Rust symbol demangling.
Adds Rust symbol demangler. Rust mangles symbols using GNU_V3 style,
adding a hash and various special character subtitutions. This adds
a new rust style to cplus_demangle and adds 3 helper functions
rust_demangle, rust_demangle_sym and rust_is_mangled.
rust-demangle.c was written by David. Mark did the code formatting to
GNU style and integration into the gcc/libiberty build system and
testsuite.
include/ChangeLog:
2016-11-03 David Tolnay <dtolnay@gmail.com>
Mark Wielaard <mark@klomp.org>
* demangle.h (DMGL_RUST): New macro.
(DMGL_STYLE_MASK): Add DMGL_RUST.
(demangling_styles): Add dlang_rust.
(RUST_DEMANGLING_STYLE_STRING): New macro.
(RUST_DEMANGLING): New macro.
(rust_demangle): New prototype.
(rust_is_mangled): Likewise.
(rust_demangle_sym): Likewise.
libiberty/ChangeLog:
2016-11-03 David Tolnay <dtolnay@gmail.com>
Mark Wielaard <mark@klomp.org>
* Makefile.in (CFILES): Add rust-demangle.c.
(REQUIRED_OFILES): Add rust-demangle.o.
* cplus-dem.c (libiberty_demanglers): Add rust_demangling case.
(cplus_demangle): Handle RUST_DEMANGLING.
(rust_demangle): New function.
* rust-demangle.c: New file.
* testsuite/Makefile.in (really-check): Add check-rust-demangle.
(check-rust-demangle): New rule.
* testsuite/rust-demangle-expected: New file.
Mark Wielaard [Tue, 15 Nov 2016 19:31:59 +0000 (19:31 +0000)]
libiberty: demangler crash with missing :? or fold expression component.
When constructing an :? or fold expression that requires a third
expression only the first and second were explicitly checked to
not be NULL. Since the third expression is also required in these
constructs it needs to be explicitly checked and rejected when missing.
Otherwise the demangler will crash once it tries to d_print the
NULL component. Added two examples to demangle-expected of strings
that would crash before this fix.
Mark Wielaard [Tue, 15 Nov 2016 19:31:50 +0000 (19:31 +0000)]
libiberty: Fix some demangler crashes caused by reading past end of input.
In various situations the cplus_demangle () function could read past the
end of input causing crashes. Add checks in various places to not advance
the demangle string location and fail early when end of string is reached.
Add various examples of input strings to the testsuite that would crash
test-demangle before the fixes.
Found by using the American Fuzzy Lop (afl) fuzzer.
libiberty/ChangeLog:
* cplus-dem.c (demangle_signature): After 'H', template function,
no success and don't advance position if end of string reached.
(demangle_template): After 'z', template name, return zero on
premature end of string.
(gnu_special): Guard strchr against searching for zero characters.
(do_type): If member, only advance mangled string when 'F' found.
* testsuite/demangle-expected: Add examples of strings that could
crash the demangler by reading past end of input.
Mark Wielaard [Fri, 4 Nov 2016 23:55:01 +0000 (23:55 +0000)]
libiberty: Fix -Wimplicit-fallthrough warnings.
Adjust some comments, add some explicit fall through comments or explicit
returns where necessary to not get implicit-fallthrough warnings.
All fall throughs were deliberate. In one case I added an explicit return
false for clarity instead of falling through a default case (that also
would return false).
libiberty/ChangeLog:
* cplus-dem.c (demangle_signature): Move fall through comment.
(demangle_fund_type): Add fall through comment between 'G' and 'I'.
* hashtab.c (iterative_hash): Add fall through comments.
* regex.c (regex_compile): Add Fall through comment after '+'/'?'.
(byte_re_match_2_internal): Add Fall through comment after jump_n.
Change "Note fall through" to "Fall through".
(common_op_match_null_string_p): Return false after set_number_at
instead of fall through.
Mark Wielaard [Tue, 1 Nov 2016 23:13:10 +0000 (23:13 +0000)]
libiberty: Fix memory leak in ada_demangle when symbol cannot be demangled.
When a symbol cannot be demangled in ada_demangle a new demangled VEC
will be allocated without deleting the demangled VEC already in use.
Running testsuite/test-demangle under valgrind will show the leak for
this entry in testsuite/demangle-expected:
# Elaborated flag (not demangled)
--format=gnat
x_E
<x_E>
11 bytes in 1 blocks are definitely lost in loss record 1 of 1
at 0x4C27BE3: malloc (vg_replace_malloc.c:299)
by 0x413FE7: xmalloc (xmalloc.c:148)
by 0x4025EC: ada_demangle (cplus-dem.c:930)
by 0x402C59: cplus_demangle (cplus-dem.c:892)
by 0x400FEC: main (test-demangle.c:317)
libiberty/ChangeLog:
* cplus-dem.c (ada_demangle): Initialize demangled to NULL and
XDELETEVEC demangled when unknown.
PR c++/71696
* cplus-dem.c: Prevent infinite recursion when there is a cycle
in the referencing of remembered mangled types.
(work_stuff): New stack to keep track of the remembered mangled
types that are currently being processed.
(push_processed_type): New method to push currently processed
remembered type onto the stack.
(pop_processed_type): New method to pop currently processed
remembered type from the stack.
(work_stuff_copy_to_from): Copy values of new variables.
(delete_non_B_K_work_stuff): Free stack memory.
(demangle_args): Push/Pop currently processed remembered type.
(do_type): Do not demangle a cyclic reference and push/pop
referenced remembered type.
Szabolcs Nagy [Fri, 18 Nov 2016 10:02:16 +0000 (10:02 +0000)]
[AArch64] Add ARMv8.3 FCMLA and FCADD instructions
Add support for FCMLA and FCADD complex arithmetic SIMD instructions.
FCMLA has an indexed element variant where the index range has to be
treated specially because a complex number takes two elements and the
indexed vector size depends on the other operands.
These complex number SIMD instructions are part of ARMv8.3
https://community.arm.com/groups/processors/blog/2016/10/27/armv8-a-architecture-2016-additions
include/
2016-11-18 Szabolcs Nagy <szabolcs.nagy@arm.com>
Add support for ARMv8.3 LDAPRB, LDAPRH and LDAPR weak release
consistency load instructions. (They are equivalent to LDARB,
LDARH and LDAR instructions other than the weaker memory ordering
requirement.)
For more details about weak release consistency see
https://community.arm.com/groups/processors/blog/2016/10/27/armv8-a-architecture-2016-additions
opcodes/
2016-11-18 Szabolcs Nagy <szabolcs.nagy@arm.com>
Add support for ARMv8.3 FJCVTZS floating-point conversion
instruction.
For details about javascript floating-point conversion see
https://community.arm.com/groups/processors/blog/2016/10/27/armv8-a-architecture-2016-additions
opcodes/
2016-11-18 Szabolcs Nagy <szabolcs.nagy@arm.com>
Add support for ARMv8.3 LDRAA and LDRAB combined pointer authentication and
load instructions.
These instructions authenticate the base register and load 8 byte from it plus
a scaled 10-bit offset with optional writeback to update the base register.
A new instruction class (ldst_imm10) and operand type (AARCH64_OPND_ADDR_SIMM10)
were introduced to handle the special addressing form.
include/
2016-11-18 Szabolcs Nagy <szabolcs.nagy@arm.com>
I find it a bit verbose and cumbersome. Since we now require GNU make,
we can change those rules with pattern rules, one for each subdirectory.
For example, the following rule works for all files under mi:
Those pattern rules assume that the source and target files have the
same stem (foo.c and foo.o). In one case, common-agent.o is generated
from common/agent.c, to avoid a conflict with the agent.o in gdb/. In
this case, I kept the explicit rule, which takes precedence over the
pattern rule. We could also rename common/agent.c to
common/common-agent.c to get rid of the special case and still avoid the
clash, as it is done with common/common-regcache.c, for example.
This strategy was the least intrusive I found, as it only requires
changing the rules, not the target names.
I also considered two other solutions, which I did not like because I
would have had to change target names a bit everywhere.
- Replicate the source directory structure in the build directory,
which would generate common/agent.o from common/agent.c. However,
something was not right with the dependency tracking (the .deps
directory). It's probably not hard to fix, but I did not
investigate further.
- Name the object files after the directory they are in, so that
common/agent.c would generate common_agent.c.
GDBserver can benefit from the same treatment, but I'll do it in another
patch.
Built-tested with --enable-targets=all.
New in v2:
- Regroup pattern rules for .c -> .o compilation in a single place.
- Add comment about common-agent.o.
Simon Marchi [Thu, 17 Nov 2016 17:02:13 +0000 (12:02 -0500)]
Makefile: Replace old suffix rules with pattern rules
As mentioned here [1], suffix rules are obsolete and have been
superseeded with pattern rules. People (myself included, before writing
this patch) are more likely to know what pattern rules are than suffix
rules.
AFAIK, .SUFFIXES targets are only used for those rules, and can be
removed as well.
Simon Marchi [Thu, 17 Nov 2016 17:00:10 +0000 (12:00 -0500)]
Remove code that checks for GNU/non-GNU make
Since GNU make is now required to build GDB, we can remove everything
that checks whether the current make implemention is the GNU one or
not. I simply removed the @GMAKE_TRUE@ prefixes and removed the whole
lines that were prefixed with @GMAKE_FALSE@.
I removed the code in the configure scripts that set those variables.
I also removed the following bits from the configure scripts:
AC_CHECK_PROGS(MAKE, make): GNU make already defines a MAKE variable
internally to be used when invoking Makefiles recursively. I don't see
this variable being used anywhere else (in scripts for example), so I
think it's safe for removal.
AC_PROG_MAKE_SET: This macro defines a SET_MAKE output variable, which
is meant to be used in Makefiles to define the MAKE variable when
using an implementation of make that doesn't already define it.
Since we are now requiring GNU make, we don't need it anymore.
Plus, I don't see SET_MAKE being used anywhere, so I don't think it
was actually doing anything...
gdb/ChangeLog:
* Makefile.in: Remove @GMAKE_TRUE@ prefixes and removes lines
prefixed with @GMAKE_FALSE@. Update comment related to non-GNU
make.
* configure.ac: Remove checks for the make program.
* configure: Re-generate.
gdb/gdbserver/ChangeLog:
* Makefile.in: Remove @GMAKE_TRUE@ prefixes and removes lines
prefixed with @GMAKE_FALSE@. Update comment related to non-GNU
make.
* configure.ac: Remove checks for the make program.
* configure: Re-generate.
gdb/testsuite/ChangeLog:
* Makefile.in: Remove @GMAKE_TRUE@ prefixes and removes lines
prefixed with @GMAKE_FALSE@. Update comment related to non-GNU
make.
* configure.ac: Remove checks for the make program.
* configure: Re-generate.
Simon Marchi [Thu, 17 Nov 2016 16:59:18 +0000 (11:59 -0500)]
Document new hard requirement on GNU make
As discussed in [1], it would be benificial for the GDB project to start
requiring GNU make to build its software. It would allow using useful
GNU-specific constructs, such as pattern rules. It would also allow
removing the alternative code paths in the Makefiles (guarded by
GMAKE_TRUE/GMAKE_FALSE), simplifying the Makefile code.
Pedro Alves [Thu, 17 Nov 2016 14:53:02 +0000 (14:53 +0000)]
gdb/c-exp.y: fprintf -> parser_fprintf
Switching GDB to make use of gnulib's C++ namespace support mode
revealed these direct uses of fprintf in the C parser, where
parser_fprintf should be used to handle rewiring stderr to gdb_stderr:
..../src/gdb/c-exp.y: In function ‘void c_print_token(FILE*, int, YYSTYPE)’:
..../src/gdb/c-exp.y:3220:45: error: call to ‘fprintf’ declared with attribute warning: The symbol ::fprintf refers to the system function. Use gnulib::fprintf instead. [-Werror]
pulongest (value.typed_val_int.val));
^
..../src/gdb/c-exp.y:3231:62: error: call to ‘fprintf’ declared with attribute warning: The symbol ::fprintf refers to the system function. Use gnulib::fprintf instead. [-Werror]
fprintf (file, "tsval<type=%d, %s>", value.tsval.type, copy);
^
..../src/gdb/c-exp.y:3237:57: error: call to ‘fprintf’ declared with attribute warning: The symbol ::fprintf refers to the system function. Use gnulib::fprintf instead. [-Werror]
fprintf (file, "sval<%s>", copy_name (value.sval));
^
..../src/gdb/c-exp.y:3243:39: error: call to ‘fprintf’ declared with attribute warning: The symbol ::fprintf refers to the system function. Use gnulib::fprintf instead. [-Werror]
copy_name (value.tsym.stoken));
^
..../src/gdb/c-exp.y:3254:39: error: call to ‘fprintf’ declared with attribute warning: The symbol ::fprintf refers to the system function. Use gnulib::fprintf instead. [-Werror]
value.ssym.is_a_field_of_this);
^
..../src/gdb/c-exp.y:3258:70: error: call to ‘fprintf’ declared with attribute warning: The symbol ::fprintf refers to the system function. Use gnulib::fprintf instead. [-Werror]
fprintf (file, "bval<%s>", host_address_to_string (value.bval));
^
gdb/ChangeLog:
2016-11-17 Pedro Alves <palves@redhat.com>
* c-exp.y (c_print_token): Use parser_fprintf instead of fprintf.
Pedro Alves [Thu, 17 Nov 2016 14:43:02 +0000 (14:43 +0000)]
gdb/ctf.c: Get rid of mkdir redefinition
Making GDB use gnulib's C++ namespace support shows this build error
on mingw:
../../src/gdb/ctf.c: In function 'void ctf_start(trace_file_writer*, const char*)':
../../src/gdb/ctf.c:309:46: error: no match for call to '(const gnulib::_gl_mkdir_wrapper) (const char*&)'
#define mkdir(pathname, mode) mkdir (pathname)
^
../../src/gdb/ctf.c:327:15: note: in expansion of macro 'mkdir'
if (gnulib::mkdir (dirname, hmode) && errno != EEXIST)
^
../../src/gdb/ctf.c:309:46: note: candidate: gnulib::_gl_mkdir_wrapper::type {aka int (*)(const char*, short unsigned int)} <conversion>
#define mkdir(pathname, mode) mkdir (pathname)
^
../../src/gdb/ctf.c:327:15: note: in expansion of macro 'mkdir'
if (gnulib::mkdir (dirname, hmode) && errno != EEXIST)
^
../../src/gdb/ctf.c:309:46: note: candidate expects 3 arguments, 2 provided
#define mkdir(pathname, mode) mkdir (pathname)
^
../../src/gdb/ctf.c:327:15: note: in expansion of macro 'mkdir'
if (gnulib::mkdir (dirname, hmode) && errno != EEXIST)
^
The problem is the '#define mkdir ...'
Fortunately, we can just remove it, since gnulib's sys/stat.h
replacement already takes care of the Windows mkdir prototype quirk:
~~~
/* mingw's _mkdir() function has 1 argument, but we pass 2 arguments.
Additionally, it declares _mkdir (and depending on compile flags, an
alias mkdir), only in the nonstandard includes <direct.h> and <io.h>,
which are included above. */
# if (defined _WIN32 || defined __WIN32__) && ! defined __CYGWIN__
# if !GNULIB_defined_rpl_mkdir
static int
rpl_mkdir (char const *name, mode_t mode)
{
return _mkdir (name);
}
~~~
That's sys_stat.in.h, part of the sys_stat module, which we explictly
pull in nowadays. It wasn't being pulled when this macro was added:
Pedro Alves [Wed, 16 Nov 2016 11:38:49 +0000 (11:38 +0000)]
gdb/ada-lang.c: one malloc -> unique_ptr<[]>
Switching gdb to use gnulib's C++ namespace mode reveals we're calling
malloc instead of xmalloc here:
..../src/gdb/ada-lang.c: In function ‘value* ada_value_primitive_packed_val(value*, const gdb_byte*, long int, int, int, type*)’:
..../src/gdb/ada-lang.c:2592:50: error: call to ‘malloc’ declared with attribute warning: The symbol ::malloc refers to the system function. Use gnulib::malloc instead. [-Werror]
staging = (gdb_byte *) malloc (staging_len);
^
We're unconditionaly using the result afterwards -- so it's not a case
of gracefully handling huge allocations.
Since we want to get rid of all cleanups, fix this by switching to
new[] and unique_ptr<[]> instead, while at it.
Regtested on Fedora 23.
gdb/ChangeLog:
2016-11-16 Pedro Alves <palves@redhat.com>
* ada-lang.c (ada_value_primitive_packed_val): Use unique_ptr and
new gdb_byte[] instead of malloc and cleanups.