binutils_os216

Fork

Update w/HEAD

Problems:
gdb/compile/compile.c
gdb/cp-support.c
gdb/cp-support.h
gdb/gdbtypes.h
gdb/language.c
gdb/linespec.c

Automatic date update in version.in

Add new NT_PPC_* available since Linux 4.8

Add new note types available since Linux 4.8 to elf/common.h and
make the 'readelf' tool aware of it.

include/
* elf/common.h (NT_PPC_TAR): New macro.
(NT_PPC_PPR): Likewise.
(NT_PPC_DSCR): Likewise.
(NT_PPC_EBB): Likewise.
(NT_PPC_PMU): Likewise.
(NT_PPC_TM_CGPR): Likewise.
(NT_PPC_TM_CFPR): Likewise.
(NT_PPC_TM_CVMX): Likewise.
(NT_PPC_TM_CVSX): Likewise.
(NT_PPC_TM_SPR): Likewise.
(NT_PPC_TM_CTAR): Likewise.
(NT_PPC_TM_CPPR): Likewise.
(NT_PPC_TM_CDSCR): Likewise.
binutils/
* readelf.c (get_note_type): Handle PPC note types available
since Linux 4.8.

Add myself as a write-after-approval GDB maintainer.

gdb/ChangeLog:

* MAINTAINERS (Write After Approval): Add Stafford Horne.

xtensa: Properly strdup string when building reggroup

I noticed this while looking at the reggroup intializations. It seems
for xtensa the "cpN" reggroup->name is getting assigned to the same text
pointer for each iteration of XTENSA_MAX_COPROCESSOR.

Note, internally reggroup_new() does not do any xstrdup().

gdb/ChangeLog:
2017-08-15 Stafford Horne <shorne@gmail.com>

* xtensa-tdep.c (xtensa_init_reggroups): Use xstrdup for cpname.

Fix PR gdb/21954: make 'unset environment' work again

When I made commit 9a6c7d9c021cfeb290d76584db7a01e57e7c3d4e, which
C++-fied gdb/common/environ.[ch], I mistakenly altered the behaviour
of the 'unset environment' command. This command, which should delete
all environment variables, is now resetting the list of variables to
the state they were when GDB was started.

This commit fixes this regression, and also adds a test on
gdb.base/environ.exp which really checks if 'unset environment'
worked.

gdb/ChangeLog:
2017-08-15 Sergio Durigan Junior <sergiodj@redhat.com>

PR gdb/21954
* infcmd.c (unset_environment_command): Use the 'clear' method on
the environment instead of resetting it.

gdb/testsuite/ChangeLog:
2017-08-15 Sergio Durigan Junior <sergiodj@redhat.com>

PR gdb/21954
* gdb.base/environ.exp: Add test to check if 'unset environment'
works.

Fix compile on big-endian platforms in siginfo_t converter.

gdb/ChangeLog:

* fbsd-nat.c (fbsd_convert_siginfo): Fix compile on big-endian
platforms.

[Patch AArch64] Turn lr, fp, ip0 and ip1 into proper aliases

We got a report from the linux-arm-kernel folks about getting spurious
warnings when building the kernel with binutils 2.29. See
https://www.spinics.net/lists/arm-kernel/msg599929.html

which boils down to this testcase.

$> cat /tmp/tst.s
lr .req x30
/tmp/tst.s: Assembler messages:
/tmp/tst.s:1: Warning: ignoring attempt to redefine built-in register 'lr'

Instead let's treat this as a proper alias at startup time thus
avoiding the problem and treating these as proper aliases
rather than new registers. This means that attempts to redefine
the alias with the same "name" will provoke no warning and attempts
to redefine the alias to something else will provoke the above mentioned
warning.

Tested make check-gas and no regressions.

Ok to apply to trunk (and backport to 2.29 branch)?

Regards
Ramana

GDB testsuite: Suppress GCC's colored output

Newer GCC versions yield colored diagnostic messages by default, which may
be useful when executing GDB interactively from a terminal. But when run
from a GDB test case, the compiler output is written into gdb.log, where
such escape sequences are usually more inhibiting than helpful to the
evaluation of test results. So this patch suppresses that.

gdb/testsuite/ChangeLog:

* lib/gdb.exp (universal_compile_options): New caching proc.
(gdb_compile): Suppress GCC's coloring of messages.

Remove BITS_IN_BYTES define

While working on the previous patch, I noticed that BITS_IN_BYTES can be
replaced by HOST_CHAR_BIT, which is used more widely in gdb.

ChangeLog
2017-08-14 Tom Tromey <tom@tromey.com>

* valprint.c (print_octal_chars): Use HOST_CHAR_BIT.
(print_binary_chars): Likewise.
(BITS_IN_BYTES): Remove.

Fix two regressions in scalar printing

PR gdb/21675 points out a few regressions in scalar printing.

One type of regression is due to not carrying over the old handling of
floating point printing -- where a format like "/d" causes a floating
point number to first be cast to a signed integer. This patch restores
this behavior.

The other regression is a longstanding bug in print_octal_chars: one of
the constants was wrong. This patch fixes the constant and adds static
asserts to help catch this sort of error.

ChangeLog
2017-08-14 Tom Tromey <tom@tromey.com>

PR gdb/21675
* valprint.c (LOW_ZERO): Change value to 034.
(print_octal_chars): Add static_asserts for octal constants.
* printcmd.c (print_scalar_formatted): Add 'd' case.

testsuite/ChangeLog
2017-08-14 Tom Tromey <tom@tromey.com>

PR gdb/21675:
* gdb.base/printcmds.exp (test_radices): New function.
* gdb.dwarf2/var-access.exp: Use p/u, not p/d.
* gdb.base/sizeof.exp (check_valueof): Use p/d.
* lib/gdb.exp (get_integer_valueof): Use p/d.

Fix memory leak in add_symbol_file_command

I happened to notice that add_symbol_file_command leaks "sect_opts".
This patch fixes the leak by changing sect_opts to be a std::vector.

I had to change the logic in the loop a little bit. Previously, it
was incrementing section_index after completing an entry; but this
changes it to push a new entry when the name is seen.

I believe the argument parsing here is mildly incorrect, in that
nothing checks whether the -s option actually had any arguments.
Maybe gdb can crash if "-s NAME" is given without an argument. I
didn't try to fix this in this patch, but I do have another patch I
can send later that fixes it up.

Regression tested on the buildbot.

ChangeLog
2017-08-11 Tom Tromey <tom@tromey.com>

* symfile.c (add_symbol_file_command): Use std::vector.

Use std::move in a few places

This patch adds std::move to few spots where it seems to be missing.

Regression tested by the buildbot.

ChangeLog
2017-08-14 Tom Tromey <tom@tromey.com>

* break-catch-throw.c (handle_gnu_v3_exceptions): Use std::move.
* break-catch-syscall.c (create_syscall_event_catchpoint): Use
std::move.
* break-catch-sig.c (create_signal_catchpoint): Use std::move.

Fix null pointer dereference when parsing a corrupt ELF binary.

PR 21957
* elf.c (setup_group): Check for an empty or very small group
section.
* po/bfd.pot: Regenerate.

Automatic date update in version.in

PR21441, Unnecessary padding of .eh_frame section

Until all .eh_frame sections have been edited we don't know their
sizes. So it isn't possible to properly decide whether a non-empty
.eh_frame section follows a given section until editing is complete.

bfd/
PR 21441
* elf-eh-frame.c (_bfd_elf_discard_section_eh_frame): Don't add
alignment padding here.
* elflink.c (bfd_elf_discard_info): Add .eh_frame padding here
in a reverse pass over sections.
ld/
PR 21441
* testsuite/ld-x86-64/pr21038a.d: Adjust.
* testsuite/ld-x86-64/pr21038a-now.d: Adjust.

ld: Restore linker scripts in PR ld/21884 tests

OUTPUT_FORMAT in linker script of PR ld/21884 tests is needed to trigger
PR ld/21884. Restore linker scripts and add nacl versions of the same
tests.

* testsuite/ld-i386/i386.exp: Run pr21884-nacl.
* testsuite/ld-x86-64/x86-64.exp: Likewise.
* testsuite/ld-i386/pr21884.d: Don't run on nacl targets.
* testsuite/ld-x86-64/pr21884.d: Likewise.
* testsuite/ld-i386/pr21884.t: Revert the last change.
* testsuite/ld-x86-64/pr21884.t: Likewise.
* testsuite/ld-i386/pr21884-nacl.d: New file.
* testsuite/ld-i386/pr21884-nacl.t: Likewise.
* testsuite/ld-x86-64/pr21884-nacl.d: Likewise.
* testsuite/ld-x86-64/pr21884-nacl.t: Likewise.

Fix i686-nacl and x86_64-nacl pr21884 failures

OUTPUT_FORMAT in the script results in "./ld-new: target elf32-i386
not found" and similarly for the x86_64 test.

* testsuite/ld-i386/pr21884.t: Remove unneeded format, arch and entry.
* testsuite/ld-x86-64/pr21884.t: Likewise.

Automatic date update in version.in

Fix typo on documentation ("show set startup-with-shell")

The documentation was erroneously saying that there is a command named
"show set startup-with-shell", while the correct version is "show
startup-with-shell". This commit fixes obvious mistake.

gdb/doc/ChangeLog:
2017-08-12 Sergio Durigan Junior <sergiodj@redhat.com>

PR gdb/21925
* gdb.texinfo (Starting) <startup-with-shell>: Fix typo ("show
set...").

testsuite: Exclude end-of-line characters from get_valueof result

The get_valueof procedure allows tests to conveniently make gdb evaluate
an expression an return the value as a string. However, it includes an
end-of-line character in its result. I stumbled on this when trying to
use that result as part of a regex further in a test.

You can see this for example by adding a puts in
gdb.dwarf2/implref-struct.exp:get_members:

set members [get_valueof "" ${var} ""]
puts "<$members>"

The output is

<{a = 0, b = 1, c = 2}
>

This is because the regex in get_valueof is too greedy, the captured
portion matches anything up to the gdb_prompt, including the end of line
characters. This patch changes it to capture everything but end of line
characters.

The output of the puts becomes:

<{a = 0, b = 1, c = 2}>

I tested this by running gdb.dwarf2/implref-array.exp and
gdb.dwarf2/implref-struct.exp, the two only current users of that
procedure.

gdb/testsuite/ChangeLog:

* lib/gdb.exp (get_valueof): Don't capture end-of-line
characters.

Collision between NT_GNU_BUILD_ATTRIBUTE_OPEN and NT_PPC_VMX

* readelf.c (process_note): Qualify NT_GNU_BUILD_ATTRIBUTE notes
by name data.

Automatic date update in version.in

x86: Allocate space for symbol names with symbol table

When synthesizing symbols for PLT entries, allocate space for symbol
names with @plt suffixes together with symbol table so that all memory
is returned when symbol table is freed.

PR binutils/21943
* elf32-i386.c (elf_i386_get_synthetic_symtab): Allocate space
for @plt suffixes first.
* elf64-x86-64.c (elf_x86_64_get_synthetic_symtab): Likewise.

Add -z globalaudit linker command line option to set the DF_GLOBALAUDIT flag bit in the dynamic tags.

* emultempl/elf32.em (handle_option): Accept the -z globalaudit
command line option.
* lexsup.c (elf_static_list_options): Add -z globalaudit.
* ld.texinfo: Document the support for the new command line
option.
* NEWS: Mention the new feature.
* testsuite/ld-elf/audit.exp: Add a test of the -z globalaudit
command line option.
* testsuite/ld-elf/globalaudit.rd: New file: Expected output from
readelf.

Add 2 more tests for PR ld/21884

PR ld/21884
* testsuite/ld-i386/i386.exp: Run pr21884.
* testsuite/ld-x86-64/x86-64.exp: Likewise.
* testsuite/ld-i386/pr21884.d: New file.
* testsuite/ld-i386/pr21884.t: Likewise.
* testsuite/ld-x86-64/pr21884.d: Likewise.
* testsuite/ld-x86-64/pr21884.t: Likewise.

Also disallow global alias of common symbol

We can't create alias of common symbol. Local alias of common symbol has
been disallowed. But global alias of common symbol is disallowed when the
common symbol is seen first and silently dropped otherwise. This patch
disallows alias of common symbol in all cases.

gas/

PR gas/21667
* read.c (pseudo_set): Update error message for alias of common
symbol.
* write.c (write_object_file): Disallow both local and global
aliases of common symbol.
* testsuite/gas/elf/common5a.d: New file.
* testsuite/gas/elf/common5a.l: Likewise.
* testsuite/gas/elf/common5a.s: Likewise.
* testsuite/gas/elf/common5b.d: Likewise.
* testsuite/gas/elf/common5b.l: Likewise.
* testsuite/gas/elf/common5b.s: Likewise.
* testsuite/gas/elf/common5c.d: Likewise.
* testsuite/gas/elf/common5c.s: Likewise.
* testsuite/gas/elf/common5d.d: Likewise.
* testsuite/gas/elf/common5d.s: Likewise.
* testsuite/gas/elf/elf.exp: Run common5a, common5b, common5c
and common5d.

More gdb/skip.c C++ification

- Make skiplist_entry a class with private data members.
- Move all construction logic to the ctor.
- Make skip_file_p etc be methods of skiplist_entry.
- Use std::list for the skip entries chain. Make the list own its
elements.
- Get rid of the ALL_SKIPLIST_ENTRIES/ALL_SKIPLIST_ENTRIES_SAFE
macros, use range-for / iterators instead.
- function_name_is_marked_for_skip 'function_sal' argument must be
non-NULL, so make it a reference instead.

All skiplist_entry invariants are now controlled by skiplist_entry
methods/internals. Some gdb_asserts disappear for being redundant.

gdb/ChangeLog:
2017-08-11 Pedro Alves <palves@redhat.com>

* infrun.c (process_event_stop_test): Adjust
function_name_is_marked_for_skip call.
* skip.c: Include <list>.
(skiplist_entry): Make it a class with private fields, and
getters/setters.
(skiplist_entry_chain): Delete.
(skiplist_entries): New.
(skiplist_entry_count): Delete.
(highest_skiplist_entry_num): New.
(ALL_SKIPLIST_ENTRIES, ALL_SKIPLIST_ENTRIES_SAFE): Delete.
(add_skiplist_entry): Delete.
(skiplist_entry::skiplist_entry): New.
(skiplist_entry::add_entry): New.
(skip_file_command, skip_function): Adjust.
(compile_skip_regexp): Delete.
(skip_command): Don't compile regexp here. Adjust to use
skiplist_entry::add_entry.
(skip_info): Adjust to use range-for and getters.
(skip_enable_command, skip_disable_command): Adjust to use
range-for and setters.
(skip_delete_command): Adjust to use std::list.
(add_skiplist_entry): Delete.
(skip_file_p): Delete, refactored as ...
(skiplist_entry::do_skip_file_p): ... this new method.
(skip_gfile_p): Delete, refactored as ...
(skiplist_entry::do_gskip_file_p): ... this new method.
(skip_function_p, skip_rfunction_p): Delete, refactored as ...
(skiplist_entry::skip_function_p): ... this new method.
(function_name_is_marked_for_skip): Now returns bool, and takes
the function sal by const reference. Adjust to use range-for and
skiplist_entry methods.
(_initialize_step_skip): Remove references to
skiplist_entry_chain, skiplist_entry_count.
* skip.h (function_name_is_marked_for_skip): Now returns bool, and
takes the function sal by const reference.

Reset *THIS_CACHE in frame_unwind_try_unwinder in case of exception

It is required that unwinder->sniffer should set *this_cache to NULL if
the unwinder is not applicable or exception is thrown, so
78ac5f831692f70b841044961069e50d4ba6a76f adds clear_pointer_cleanup to set
*this_cache to NULL in case of exception in order to fix PR 14100.
https://sourceware.org/ml/gdb-patches/2012-08/msg00075.html

This patch removes that clear_pointer_cleanup, and catch all exception in
the caller of unwinder->sniffer. In case of exception, reset *this_case.

gdb:

2017-08-11 Yao Qi <yao.qi@linaro.org>

* dwarf2-frame.c (clear_pointer_cleanup): Remove.
(dwarf2_frame_cache): Remove reset_cache_cleanup.
(dwarf2_frame_cache):
* frame-unwind.c (frame_unwind_try_unwinder): Catch
RETURN_MASK_ALL and set *this_case to NULL.
* frame-unwind.h: Update comments.

Class-fy dwarf2_frame_state_reg_info

This patch adds dwarf2_frame_state_reg_info ctor, dtor, copy ctor,
assignment operator, and move assignment. This patch also adds unit test
to execute_cfa_program to cover the changes.

gdb:

2017-08-11 Yao Qi <yao.qi@linaro.org>

* dwarf2-frame.c (dwarf2_frame_state_alloc_regs): Remove.
(dwarf2_frame_state_copy_regs): Remove.
(dwarf2_frame_state_free_regs): Remove.
(dwarf2_frame_state::~dwarf2_frame_state): Remove.
(dwarf2_restore_rule): Call method .alloc_regs instead of
dwarf2_frame_state_alloc_regs.
(execute_cfa_program): Likewise. Call dwarf2_frame_state_reg_info
constructor. Call std::move.
(dwarf2_fetch_cfa_info): Don't call dwarf2_frame_state_copy_regs.
(dwarf2_frame_cache): Likewise.

[GDB_SELF_TEST]: Include selftest.h and
selftest-arch.h.
[GDB_SELF_TEST] (execute_cfa_program_test): New function.
(_initialize_dwarf2_frame) [GDB_SELF_TEST]: Register
execute_cfa_program_test.

* dwarf2-frame.h (dwarf2_frame_state_reg_info): Add ctor, dtor,
copy ctor, assignment operator, move assignment.
<alloc_regs>: New method.
<swap>: New method.
(struct dwarf2_frame_state): Delete dtor.
(dwarf2_frame_state_alloc_regs): Remove declaration.
* sparc-tdep.c (sparc_execute_dwarf_cfa_vendor_op): Don't call
dwarf2_frame_state_alloc_regs, use .alloc_regs instead.