Move code out of 'between TRY and CATCH'

I tried building GDB with TRY/CATCH mapped to raw C++ try/catch (by
defining GDB_XCPT to GDB_XCPT_RAW_TRY in
gdb/common/common-exceptions.h), and that caught a case of code
written between try and catch.  This commit fixes it.

gdb/ChangeLog:
2017-10-04  Pedro Alves  <palves@redhat.com>

* guile/scm-breakpoint.c (gdbscm_breakpoint_commands): Move code
out of 'between TRY and CATCH'.

Add missing-END_CATCH detection/protection (to gdb's TRY/CATCH/END_CATCH)

(Adding missing ChangeLog entry)

While we still have cleanups (i.e., make_cleanup & co), we must be
sure to add END_CATCH at the end of a TRY/CATCH/END_CATCH construct.
However, it's currently too easy to miss adding the END_CATCH, because
the code compiles anyway without it.  I realized this when I noticed
that another patch I was working on missed several adding END_CATCH in
several new TRY/CATCH uses.

This commit fixes that by making TRY open a new scope that is only
closed by END_CATCH.  This way, if you forget to add the END_CATCH,
then compilation fails due to the unbalanced curly braces.

This caught a couple places where we were missing END_CATCH in current
master, also fixed by the patch.

gdb/ChangeLog:
2017-10-04  Pedro Alves  <palves@redhat.com>

* cli/cli-cmds.c (complete_command): Add missing END_CATCH.
* common/common-exceptions.h (TRY): Open an outermost scope.
Expand intro comment.
(CATCH): Reindent.
(END_CATCH): Close the outermost scope.
* completer.c (complete_line_internal): Add missing END_CATCH.

Add missing-END_CATCH detection/protection (to gdb's TRY/CATCH/END_CATCH)

While we still have cleanups (i.e., make_cleanup & co), we must be
sure to add END_CATCH at the end of a TRY/CATCH/END_CATCH construct.
However, it's currently too easy to miss adding the END_CATCH, because
the code compiles anyway without it.  I realized this when I noticed
that another patch I was working on missed several adding END_CATCH in
several new TRY/CATCH uses.

This commit fixes that by making TRY open a new scope that is only
closed by END_CATCH.  This way, if you forget to add the END_CATCH,
then compilation fails due to the unbalanced curly braces.

This caught a couple places where we were missing END_CATCH in current
master, also fixed by the patch.

gdb/ChangeLog:
2017-10-04  Pedro Alves  <palves@redhat.com>

* cli/cli-cmds.c (complete_command): Add missing END_CATCH.
* common/common-exceptions.h (TRY): Open an outermost scope.
Expand intro comment.
(CATCH): Reindent.
(END_CATCH): Close the outermost scope.
* completer.c (complete_line_internal): Add missing END_CATCH.

Extend "set cwd" to work on gdbserver

This is the "natural" extension necessary for the "set cwd" command
(and the whole "set the inferior's cwd" logic) to work on gdbserver.

The idea here is to have a new remote packet, QSetWorkingDir (name
adopted from LLDB's extension to the RSP, as can be seen at
<https://raw.githubusercontent.com/llvm-mirror/lldb/master/docs/lldb-gdb-remote.txt>),
which sends an hex-encoded string representing the working directory
that the remote inferior will use.  There is a slight difference from
the packet proposed by LLDB: GDB's version will accept empty
arguments, meaning that the user wants to clear the previously set
working directory for the inferior (i.e., "set cwd" without arguments
on GDB).

For UNIX-like targets this feature is already implemented on
nat/fork-inferior.c, and all gdbserver has to do is to basically
implement "set_inferior_cwd" and call it whenever such packet arrives.
For other targets, like Windows, it is possible to use the existing
"get_inferior_cwd" function and do the necessary steps to make sure
that the inferior will use the specified working directory.

Aside from that, the patch consists basically of updates to the
testcase (making it available on remote targets) and the
documentation.

No regressions found.

gdb/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* NEWS (Changes since GDB 8.0): Add entry about new
'set-cwd-on-gdbserver' feature.
(New remote packets): Add entry for QSetWorkingDir.
* common/common-inferior.h (set_inferior_cwd): New prototype.
* infcmd.c (set_inferior_cwd): Remove "static".
(show_cwd_command): Expand text to include remote debugging.
* remote.c: Add PACKET_QSetWorkingDir.
(remote_protocol_features) <QSetWorkingDir>: New entry for
PACKET_QSetWorkingDir.
(extended_remote_set_inferior_cwd): New function.
(extended_remote_create_inferior): Call
"extended_remote_set_inferior_cwd".
(_initialize_remote): Call "add_packet_config_cmd" for
QSetWorkingDir.

gdb/gdbserver/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* inferiors.c (set_inferior_cwd): New function.
* server.c (handle_general_set): Handle QSetWorkingDir packet.
(handle_query): Inform that QSetWorkingDir is supported.
* win32-low.c (create_process): Pass the inferior's cwd to
CreateProcess.

gdb/testsuite/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* gdb.base/set-cwd.exp: Make it available on
native-extended-gdbserver.

gdb/doc/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* gdb.texinfo (Starting your Program) <The working directory.>:
Mention remote debugging.
(Working Directory) <Your Program's Working Directory>:
Likewise.
(Connecting) <Remote Packet>: Add "set-working-dir"
and "QSetWorkingDir" to the table.
(Remote Protocol) <QSetWorkingDir>: New item, explaining the
packet.

Implement "set cwd" command on GDB

This commit adds new "set/show cwd" commands, which are used to
set/show the current working directory of the inferior that will be
started.

The idea here is that "set cwd" will become the de facto way of
setting the inferior's cwd.  Currently, the user can use "cd" for
that, but there are side effects: with "cd", GDB also switches to
another directory, and that can impact the loading of scripts and
other files.  With "set cwd", we separate the logic into a new
command.

To maintain backward compatibility, if the user issues a "cd" command
but doesn't use "set cwd", then the inferior's cwd will still be
changed according to what the user specified.  However, "set cwd" has
precedence over "cd", so it can always be used to override it.

"set cwd" works in the following way:

- If the user sets the inferior's cwd by using "set cwd", then this
  directory is saved into current_inferior ()->cwd and is used when
  the inferior is started (see below).

- If the user doesn't set the inferior's cwd by using "set cwd", but
  rather use the "cd" command as before, then this directory is
  inherited by the inferior because GDB will have chdir'd into it.

On Unix-like hosts, the way the directory is changed before the
inferior execution is by expanding the user set directory before the
fork, and then "chdir" after the call to fork/vfork on
"fork_inferior", but before the actual execution.  On Windows, the
inferior cwd set by the user is passed directly to the CreateProcess
call, which takes care of the actual chdir for us.

This way, we'll make sure that GDB's cwd is not affected by the user
set cwd.

gdb/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* NEWS (New commands): Mention "set/show cwd".
* cli/cli-cmds.c (_initialize_cli_cmds): Mention "set cwd" on
"cd" command's help text.
* common/common-inferior.h (get_inferior_cwd): New prototype.
* infcmd.c (inferior_cwd_scratch): New global variable.
(set_inferior_cwd): New function.
(get_inferior_cwd): Likewise.
(set_cwd_command): Likewise.
(show_cwd_command): Likewise.
(_initialize_infcmd): Add "set/show cwd" commands.
* inferior.h (class inferior) <cwd>: New field.
* nat/fork-inferior.c: Include "gdb_tilde_expand.h".
(fork_inferior): Change inferior's cwd before its execution.
* windows-nat.c (windows_create_inferior): Pass inferior's cwd
to CreateProcess.

gdb/gdbserver/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* inferiors.c (current_inferior_cwd): New global variable.
(get_inferior_cwd): New function.
* inferiors.h (struct process_info) <cwd>: New field.

gdb/doc/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* gdb.texinfo (Starting your Program) <The working directory.>:
Mention new "set cwd" command.
(Working Directory) <Your Program's Working Directory>:
Rephrase to explain that "set cwd" exists and is the default
way to change the inferior's cwd.

gdb/testsuite/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* gdb.base/set-cwd.c: New file.
* gdb.base/set-cwd.exp: Likewise.

Introduce gdb_tilde_expand

Currently, whenever we want to handle paths provided by the user and
perform tilde expansion on GDB, we rely on "tilde_expand", which comes
from readline.  This was enough for our use cases so far, but the
situation will change when we start dealing with paths on gdbserver as
well, which is what the next patches implement.

Unfortunately it is not possible to use "tilde_expand" in this case
because gdbserver doesn't use readline.  For that reason I decided to
implement a new "gdb_tilde_expand" function, which is basically a
wrapper for "glob" and its GNU extension, GLOB_TILDE_CHECK.  With the
import of the "glob" module from gnulib, we're sure that "glob" always
supports this extension.

gdb/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* Makefile.in (SFILES): Add gdb_tilde_expand.c.
(HFILES_NO_SRCDIR): Add gdb_tilde_expand.h.
(COMMON_OBS): Add gdb_tilde_expand.o.
* common/gdb_tilde_expand.c: New file.
* common/gdb_tilde_expand.h: Likewise.

gdb/gdbserver/ChangeLog:
2017-10-04  Sergio Durigan Junior  <sergiodj@redhat.com>

* Makefile.in (SFILES): Add $(srcdir)/common/gdb_tilde_expand.c.
(OBS): Add gdb_tilde_expand.o.

PowerPC64 ELFv2 symbols not needed in get_synthetic_symtab

* elf64-ppc.c (ppc64_elf_get_synthetic_symtab): Don't sort or
classify symbols for ELFv2.

PR22245, Fix potential UB in bfd_set_error

Passing enum as a first argument to variadic argument function
may lead to undefined behavior. The explanation on CERT site:
https://www.securecoding.cert.org/confluence/display/cplusplus/
EXP58-CPP.+Pass+an+object+of+the+correct+type+to+va_start

The bug was found by Kirill Nedostoev (nedostoev.ka@phystech.edu)
when he tried to build GNU binutils with Clang 7.

PR 22245
* bfd.c (bfd_set_error): Avoid UB on passing arg to va_start that
undergoes default promotion.
* bfd-in2.h: Regenerate.

Automatic date update in version.in

gdbarch: Remove duplicate `struct objfile' declaration

Remove a duplicate `struct objfile' declaration mistakenly added with
commit 3e29f34a4eef ("MIPS: Keep the ISA bit in compressed code
addresses").

gdb/
* gdbarch.sh (objfile): Remove duplicate declaration.
* gdbarch.h: Regenerate.

Update my email address.

sim/
* MAINTAINERS (aarch64): Update my email address.

Fix incorrect string_printf use in utils.c

I made a mistake earlier and used string_printf where I should have used
string_vprintf.

I'm checking this in as obvious.

2017-10-03  Tom Tromey  <tom@tromey.com>

* utils.c (internal_vproblem): Use string_vprintf.

Use std::string in info_symbol_command

This removes a cleanup by using std::string in info_symbol_command.

2017-10-03  Tom Tromey  <tom@tromey.com>

* printcmd.c (info_symbol_command): Use std::string.

Use std::string in gdb_safe_append_history

This removes a cleanup by using std::string in
gdb_safe_append_history.

2017-10-03  Tom Tromey  <tom@tromey.com>

* top.c (gdb_safe_append_history): Use std::string.

Remove make_delete_ui_cleanup

This removes new_ui and delete_ui in favor of ordinary 'new' and
'delete', and then removes make_delete_ui_cleanup in favor of
std::unique_ptr.

2017-10-03  Tom Tromey  <tom@tromey.com>

* event-top.c (stdin_event_handler): Update.
* main.c (captured_main_1): Update.
* top.h (make_delete_ui_cleanup): Remove.
(struct ui): Add constructor and destructor.
(new_ui, delete_ui): Remove.
* top.c (make_delete_ui_cleanup): Remove.
(new_ui_command): Use std::unique_ptr.
(delete_ui_cleanup): Remove.
(ui::ui): Rename from new_ui.  Update.
(free_ui): Remove.
(ui::~ui): Rename from delete_ui.  Update.

Use gdb::byte_vector in load_progress

This changes load_progress to use gdb::byte_vector, removing a
cleanup.

2017-10-03  Tom Tromey  <tom@tromey.com>

* symfile.c (load_progress): Use gdb::byte_vector.

Remove unused declarations

This removes some unused cleanup declarations.

2017-10-03  Tom Tromey  <tom@tromey.com>

* mi/mi-main.c (mi_cmd_trace_frame_collected): Remove unused
declaration.
* printcmd.c (x_command): Remove unused declaration.
* symfile.c (symbol_file_command): Remove unused declaration.

Use std::string in utils.c

This converts internal_vproblem and defaulted_query to use
std::string.

2017-10-03  Tom Tromey  <tom@tromey.com>

* utils.c (internal_vproblem): Use std::string.
(defaulted_query): Likewise.

Remove set_batch_flag_and_make_cleanup_restore_page_info

This removes set_batch_flag_and_make_cleanup_restore_page_info and
make_cleanup_restore_page_info in favor of a new RAII class.  This
then allows for the removal of make_cleanup_restore_uinteger and
make_cleanup_restore_integer

ChangeLog
2017-10-03  Tom Tromey  <tom@tromey.com>

* guile/scm-ports.c (ioscm_with_output_to_port_worker): Update.
* top.c (execute_command_to_string): Update.
* utils.c (make_cleanup_restore_page_info): Remove.
(do_restore_page_info_cleanup): Remove.
(set_batch_flag_and_restore_page_info):
New.
(make_cleanup_restore_page_info): Remove.
(set_batch_flag_and_make_cleanup_restore_page_info): Remove.
(~set_batch_flag_and_restore_page_info): New
(make_cleanup_restore_uinteger): Remove.
(make_cleanup_restore_integer): Remove.
(struct restore_integer_closure): Remove.
(restore_integer): Remove.
* utils.h (struct set_batch_flag_and_restore_page_info): New
class.
(set_batch_flag_and_make_cleanup_restore_page_info): Remove.
(make_cleanup_restore_page_info): Remove.
(make_cleanup_restore_uinteger) Remove.
(make_cleanup_restore_integer) Remove.

Change record_full_gdb_operation_disable_set not to return a cleanup

This changes record_full_gdb_operation_disable_set to return a
scoped_restore rather than a cleanup, and fixes all the users.

ChangeLog
2017-10-03  Tom Tromey  <tom@tromey.com>

* record-full.h (record_full_gdb_operation_disable_set): Return
scoped_restore_tmpl<int>.
* infrun.c (adjust_pc_after_break): Update.
(handle_signal_stop): Update.
* record-full.c (record_full_gdb_operation_disable_set): Return
scoped_restore_tmpl<int>.
(record_full_wait_1, record_full_insert_breakpoint)
(record_full_remove_breakpoint, record_full_save)
(record_full_goto_insn): Update.

PR21294, Binary size regression on PPC embedded

PR 21294
* NEWS: Note that defaulting to -z relro results in increased
memory and disk size.

Automatic date update in version.in

ld -z relro documentation

* ld.texinfo (-z relro): Expand description.
(DATA_SEGMENT_ALIGN): Note that -z relro is not effective when
running with system page size larger than commonpagesize.
(DATA_SEGMENT_RELRO_END): Be explicit about the alignment.

Fix &str printing in Rust

Printing a string slice ("&str") in Rust would print until the
terminating \0; but that is incorrect because a slice has a length.
This fixes &str printing, and arranges to preserve the type name when
slicing a slice, so that printing a slice of an "&str" works as well.

This is PR rust/22236.

2017-10-02  Tom Tromey  <tom@tromey.com>

PR rust/22236:
* rust-lang.c (rust_val_print_str): New function.
(val_print_struct): Call it.
(rust_subscript): Preserve name of slice type.

2017-10-02  Tom Tromey  <tom@tromey.com>

PR rust/22236:
* gdb.rust/simple.rs (main): New variable "fslice".
* gdb.rust/simple.exp: Add slice tests.  Update string tests.

Fix ptype of Rust slices

Something like "ptype &x[..]" (where "x" was a slice) would crash gdb.
rust_subscript wasn't handling slicing in the EVAL_AVOID_SIDE_EFFECTS
case.

2017-10-02  Tom Tromey  <tom@tromey.com>

* rust-lang.c (rust_subscript): Handle slices in
EVAL_AVOID_SIDE_EFFECTS case.

2017-10-02  Tom Tromey  <tom@tromey.com>

* gdb.rust/simple.exp: Test ptype of a slice.

Allow indexing of &str in Rust

rust_slice_type_p was not recognizing &str as a slice type, so indexing
into (or making a slice of) a slice was not working.

2017-10-02  Tom Tromey  <tom@tromey.com>

* rust-lang.c (rust_slice_type_p): Recognize &str as a slice type.

2017-10-02  Tom Tromey  <tom@tromey.com>

* gdb.rust/simple.exp: Test index of slice.

Add missing "extern" in rust-lang.h

I noticed that one function in rust-lang.h was not declared using
"extern".  In the interested of uniformity, this patch adds it.

Tested by rebuilding.

2017-10-02  Tom Tromey  <tom@tromey.com>

* rust-lang.h (rust_slice_type): Add "extern".

Fix GDB build with G++ 4.8

G++ 4.8 trips on:

  In file included from /opt/gcc-4.8/include/c++/4.8.5/algorithm:62:0,
   from ../../src/gdb/ada-lang.c:65:
  /opt/gcc-4.8/include/c++/4.8.5/bits/stl_algo.h: In instantiation of ‘_RandomAccessIterator std::__unguarded_partition(_RandomAccessIterator, _RandomAccessIterator, const _Tp&) [with _RandomAccessIterator = __gnu_cxx::__normal_iterator<ada_exc_info*, std::vector<ada_exc_info> >; _Tp = ada_exc_info]’:
  /opt/gcc-4.8/include/c++/4.8.5/bits/stl_algo.h:2283:70:   required from ‘_RandomAccessIterator std::__unguarded_partition_pivot(_RandomAccessIterator, _RandomAccessIterator) [with _RandomAccessIterator = __gnu_cxx::__normal_iterator<ada_exc_info*, std::vector<ada_exc_info> >]’
  /opt/gcc-4.8/include/c++/4.8.5/bits/stl_algo.h:2315:54:   required from ‘void std::__introsort_loop(_RandomAccessIterator, _RandomAccessIterator, _Size) [with _RandomAccessIterator = __gnu_cxx::__normal_iterator<ada_exc_info*, std::vector<ada_exc_info> >; _Size = long int]’
  /opt/gcc-4.8/include/c++/4.8.5/bits/stl_algo.h:5461:36:   required from ‘void std::sort(_RAIter, _RAIter) [with _RAIter = __gnu_cxx::__normal_iterator<ada_exc_info*, std::vector<ada_exc_info> >]’
  ../../src/gdb/ada-lang.c:13153:61:   required from here
  /opt/gcc-4.8/include/c++/4.8.5/bits/stl_algo.h:2245:19: error: passing ‘const ada_exc_info’ as ‘this’ argument of ‘bool ada_exc_info::operator<(const ada_exc_info&)’ discards qualifiers [-fpermissive]
      while (__pivot < *__last)
     ^

Seems to be a libstdc++ bug meanwhile fixed by:
  https://gcc.gnu.org/ml/libstdc++/2012-04/msg00074.

In any case, there's no reason these methods can't be const.

gdb/ChangeLog:
2017-10-02  Tom Tromey  <tom@tromey.com>
    Pedro Alves  <palves@redhat.com>

* ada-lang.h (ada_exc_info::operator<): Make const.
(ada_exc_info::operator==): Make const.
* ada-lang.c (ada_exc_info::operator<, ada_exc_info::operator==):
Make const.

nto & lynx x86: call init_target_desc

In gdbserver, target descriptions need to be initialized by calling
init_target_desc.  Because i386_create_target_description is shared with
GDB, it doesn't do that, the callers must take care of it.  These two
platforms currently don't.

I am not able to build them, so I couldn't test.

gdb/gdbserver/ChangeLog:

* lynx-i386-low.c (lynx_i386_arch_setup): Call init_target_desc.
* nto-x86-low.c (nto_x86_arch_setup): Likewise.

Fix powerpc comment typo

* elf32-ppc.c (ppc_elf_relocate_section): Fix comment typo.
* elf64-ppc.c (ppc64_elf_relocate_section): Likewise.

Automatic date update in version.in

PR22232, NULL pointer dereference in load_specific_debug_section

PR 22232
PR 22230
* objdump.c (load_specific_debug_section): Introduce a temp to
stop bfd_get_full_section_contents NULLing out section->start.

PR21970, assertion failed when more than one overlay has subalign specified

PR 21970
* ldlang.c (lang_leave_overlay): Reset overlay_subalign.

PR21957, addr2line incorrectly handles non-increasing sequences in line table

PR 21957
* dwarf2.c (new_line_sorts_after): Remove end_sequence comparison.
(add_line_info): Always put end_sequence last.

Add new mnemonics for VLE multiple load instructions

opcodes/
* ppc-opc.c (vle_opcodes): Add e_lmvsprw, e_lmvgprw,
e_lmvsrrw, e_lmvcsrrw and e_lmvcsrrw as official mnemonics for
VLE multimple load/store instructions. Old e_ldm* variants are
kept as aliases.
Add missing e_lmvmcsrrw and e_stmvmcsrrw.
gas/
* testsuite/gas/ppc/vle-mult-ld-st-insns.s: New file: Tests the
support for the VLE multiple load/store instructions.
* testsuite/gas/ppc/vle-mult-ld-st-insns.d: New file: Test
driver.
* testsuite/gas/ppc/ppc.exp: Run it.

PR22047, Heap out of bounds read in parse_comp_unit

Like the PR22230 fix, we can allocate a buffer with an extra byte
rather than letting bfd_simple_get_relocated_section_contents malloc
and return a buffer.  Much better than allocating another buffer
afterwards.

PR 22047
* dwarf2.c (read_section): Allocate buffer with extra byte for
bfd_simple_get_relocated_section_contents rather than copying
afterwards.

PR22230, buffer overflow in display_debug_macro

PR 22230
* objdump.c (load_specific_debug_section): Allocate an extra byte
for a terminating NUL.

Automatic date update in version.in

Add a test for PR binutils/21978

PR binutils/21978
* testsuite/ld-elf/pr21978.od: New file.
* testsuite/ld-elf/pr21978a.c: Likewise.
* testsuite/ld-elf/pr21978b.c: Likewise.
* testsuite/ld-elf/shared.exp: Run PR binutils/21978 test.

Remove free_memory_read_result_vector

This changes read_memory_robust to return a std::vector, allowing the
removal of free_memory_read_result_vector and associated cleanups.
This patch also changes the functions it touches to be a bit more
robust with regards to deallocation; it's perhaps possible that
read_memory_robust could have leaked in some situations.

This patch is based on my earlier series to remove some MI cleanups.
Regression tested by the buildbot.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* target.c (read_whatever_is_readable): Change type of "result".
Update.
(free_memory_read_result_vector): Remove.
(read_memory_robust): Change return type.  Update.
* mi/mi-main.c (mi_cmd_data_read_memory_bytes): Update.  Use
bin2hex, std::string.
* target.h (memory_read_result_s): Remove typedef.
(free_memory_read_result_vector): Remove.
(read_memory_robust): Return std::vector.

Change captured_mi_execute_command to use scoped_restore

Change captured_mi_execute_command to use a scoped_restore, removing a
cleanup.  The old code copied the current token, but I don't believe
that is necessary.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-main.c (captured_mi_execute_command): Use scope_restore.

Use a std::vector for ada_exceptions_list

Change ada_exceptions_list to return a std::vector and fix up the
users.  This allows removing a cleanup in MI.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-cmd-info.c (mi_cmd_info_ada_exceptions): Update.
* ada-lang.h (struct ada_exc_info): Remove typedef.  Declare
operator< and operator==.
(ada_exceptions_list): Return a std::vector.
* ada-lang.c (ada_exc_info::operator<): Rename from
compare_ada_exception_info.
(ada_exc_info::operator==): New.
(sort_remove_dups_ada_exceptions_list): Change type of
"exceptions".
(ada_add_standard_exceptions, ada_add_exceptions_from_frame)
(ada_add_global_exceptions): Likewise.
(ada_exceptions_list_1): Return a std::vector.
(ada_exceptions_list): Likewise.

Use std::set in mi-main.c

Change a couple of spots in mi-main.c to use std::set.  This
simplifies the code and removes some cleanups.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-main.c (struct print_one_inferior_data) <inferiors>: Now a
'std::set *'.
(print_one_inferior): Update.
(free_vector_of_ints): Remove.
(list_available_thread_groups): Change "ids" to std::set.
(mi_cmd_list_thread_groups): Update.
(struct collect_cores_data) <core>: Now a std::set.
(collect_cores): Update.
(unique): Remove.
(print_one_inferior): Update.

Use std::string in mi-main.c

Change a couple of spots in mi-main.c to use std::string, and change
one place to use field_fmt.  This removes some cleanups.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-main.c (mi_execute_cli_command): Use std::string.
(mi_execute_async_cli_command): Likewise.
(mi_cmd_trace_frame_collected): Use field_fmt.

Use gdb::byte_vector in mi_cmd_data_write_memory_bytes

This changes mi_cmd_data_write_memory_bytes to use gdb::byte_vector,
removing some cleanups.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-main.c (mi_cmd_data_write_memory_bytes): Use
gdb::byte_vector.

Remove unused declaration

There was a leftover cleanup declaration in mi_parse.  Remove it.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-parse.c (mi_parse): Remove unused declaration.

Don't copy a string in mi_cmd_disassemble

This string copy in mi_cmd_disassemble seems not to be needed, so
don't do it.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-cmd-disas.c (mi_cmd_disassemble): Don't copy "oarg".

Remove cleanups from mi-cmd-var.c

This removes some cleanups from mi-cmd-var.c.  varobj_gen_name now
returns a string, simplifying mi_cmd_var_create.  In
mi_cmd_var_delete, a string copy is apparently unnecessary, so it's
simply removed.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* varobj.h (varobj_gen_name): Return std::string.
* varobj.c (varobj_gen_name): Return std::string.
* mi/mi-cmd-var.c (mi_cmd_var_create): Use std::string.
(mi_cmd_var_delete): Don't copy "name".

Remove cleanups from mi_cmd_break_insert_1

This changes mi_argv_to_format to return a string, allowing the
removal of some cleanups.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mi/mi-cmd-break.c (mi_argv_to_format): Return std::string.
(mi_cmd_break_insert_1): Update.

Remove make_cleanup_defer_target_commit_resume

This removes make_cleanup_defer_target_commit_resume in favor of using
scoped_restore.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* target.h (make_scoped_defer_target_commit_resume): Update.
* target.c (make_scoped_defer_target_commit_resume): Rename from
make_cleanup_defer_target_commit_resume.  Return a
scoped_restore.
* infrun.c (proceed): Use make_scoped_defer_target_commit_resume.

Remove some unused declarations

This removes a couple of unused cleanup-related declarations.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* main.c (captured_main_1): Remove unused declaration.
* spu-multiarch.c (parse_spufs_run): Remove unused declaration.

Remove a cleanup from symtab.c

This removes an unused outer cleanup from symtab.c, and an unused
cleanup declaration as well.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* symtab.c (search_symbols): Remove unused outer cleanup.
(make_source_files_completion_list): Remove unused declaration.

Remove cleanup from mt-tdep.c

Remove a cleanup from mt-tdep.c, using gdb::byte_vector.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* mt-tdep.c (mt_push_dummy_call): Use gdb::byte_vector.

Remove cleanup from xstormy16-tdep.c

This removes a cleanup from xstormy16-tdep.c, using gdb::byte_vector.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* xstormy16-tdep.c (xstormy16_push_dummy_call): Use
gdb::byte_vector.

Remove cleanup from complaints.c

This removes a cleanup from complaints.c by using std::string.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* complaints.c (vcomplaint): Use std::string.

Remove some cleanups from tracepoint.c

This removes some cleanups from tracepoint.c by using std::string.  It
also removes some unused cleanup declarations.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* tracepoint.c (trace_variable_command): Use std::string.
(encode_actions_1): Remove unused declarations.
(create_tsv_from_upload): Use std::string.

Remove cleanups from cp-support.c

This removes some cleanups from cp-support.c, using std::string.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* cp-support.c (gdb_demangle): Use std::string.

Remove some cleanups from stack.c

This removes some cleanups from stack.c by using std::string or
gdb::unique_xmalloc_ptr.  One cleanup remains in this file; I did not
remove it here because it is handled in another patch series that has
yet to be resolved.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* stack.c (parse_frame_specification): Use std::string
(info_frame_command): Use gdb::unique_xmalloc_ptr.

Remove cleanup from tilegx-tdep.c

This removes a cleanup from tilegx-tdep.c, by using gdb::byte_vector.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* tilegx-tdep.c (tilegx_push_dummy_call): Use gdb::byte_vector.

Remove cleanups from utils.c

This removes a couple of cleanups from utils.c through the use of
std::string.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* utils.c (vfprintf_maybe_filtered): Use std::string.
(vfprintf_unfiltered): Likewise.

Remove cleanup from display_gdb_prompt

This removes a cleanup from display_gdb_prompt by using std::string.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* event-top.c (top_level_prompt): Return std::string.
(display_gdb_prompt): Update.

Introduce string_vprintf

This adds string_vprintf, a va_list variant of string_printf.  This
will be used in later patches.

gdb/ChangeLog
2017-09-29  Tom Tromey  <tom@tromey.com>

* unittests/common-utils-selftests.c (format): New function.
(string_vprintf_tests): New function.
(_initialize_common_utils_selftests): Register new tests.
* common/common-utils.c (string_vprintf): New function.
* common/common-utils.h (string_vprintf): Declare.

PR21978, objdump does not display line numbers in certain cases

Same line but different file ought to display file and line.

PR 21978
* objdump.c: Formatting.
(show_line): Reset prev_line when function name changes.

Automatic date update in version.in

Constify unpack_varlen_hex & fix fallout

I ran into non-const unpack_varlen_hex while working on something
else, and decided to just fix it first.  Ends up constifying a good
deal of remote packet parsing.

gdb/ChangeLog:
2017-09-29  Pedro Alves  <palves@redhat.com>

* common/rsp-low.c (unpack_varlen_hex): Constify.
* common/rsp-low.h (unpack_varlen_hex): Constify.
* linux-nat.c (linux_child_static_tracepoint_markers_by_strid):
Constify.
* remote.c (remote_set_permissions, read_ptid)
(remote_current_thread, remote_get_threads_with_qthreadinfo)
(remote_static_tracepoint_marker_at)
(remote_static_tracepoint_markers_by_strid)
(stop_reply_extract_thread, remote_parse_stop_reply): Constify.
* tracepoint.c (parse_trace_status, parse_tracepoint_status)
(parse_tracepoint_definition, parse_tsv_definition)
(parse_static_tracepoint_marker_definition): Constify.
* tracepoint.h (parse_static_tracepoint_marker_definition)
(parse_trace_status, parse_tracepoint_status)
(parse_tracepoint_definition, parse_tsv_definition): Constify.

gdb/gdbserver/ChangeLog:
2017-09-29  Pedro Alves  <palves@redhat.com>

* ax.c (gdb_parse_agent_expr): Constify.
* ax.h (gdb_parse_agent_expr): Constify.
* mem-break.c (add_breakpoint_condition, add_breakpoint_commands):
Constify.
* mem-break.h (add_breakpoint_condition, add_breakpoint_commands): Constify.
* remote-utils.c (hex_or_minus_one, read_ptid): Constify.
* remote-utils.h (read_ptid): Constify.
* server.c (handle_qxfer_exec_file, handle_query, handle_v_cont)
(process_point_options, process_serial_event): Constify.
* tracepoint.c (add_tracepoint_action, cmd_qtdp, cmd_qtdpsrc)
(cmd_qtdv, cmd_qtenable_disable, cmd_qtro, cmd_qtframe, cmd_qtp)
(cmd_qtbuffer): Constify.

gdb/remote.c: Eliminate target_buf/target_buf_size hack

This finally eliminates an old hack left in place when tracepoint RSP
support was migrated from tracepoint.c to remote.c, back in
35b1e5cca081 ("Make tracepoint operations go through target vector.")
over 7 years ago.

Tested on x86_64 GNU/Linux.

gdb/ChangeLog:
2017-09-29  Pedro Alves  <palves@redhat.com>

* remote.c (target_buf, target_buf_size): Delete.
(remote_get_noisy_reply): Remove buf_p and sizeof_buf parameters.
Use the connection's packet buffer instead.
All callers adjusted.
(_initialize_remote): Remove references to target_buf and
target_buf_size.

gdbserver/libthread_db: Don't ignore memory reading failures

If we had this in place before, then the regression fixed by the
previous commit would have been been visible is all test runs.  E.g.:

  Running src/gdb/testsuite/gdb.threads/multi-create-ns-info-thr.exp ...
  FAIL: gdb.threads/multi-create-ns-info-thr.exp: continue to breakpoint 6

Debugging manually we'd see this:
  gdbserver: Cannot get thread handle for LWP 1467: generic error

Instead of:
  gdbserver: PID mismatch!  Expected 27472, got 27471

which is misleading - gdbserver didn't 27471, that was stale stack
data from previous function invocations.

gdb/gdbserver/ChangeLog:
2017-09-29  Pedro Alves  <palves@redhat.com>

* proc-service.c (ps_pdread): Return PS_ERR if reading memory
fails.

Fix gdbserver regression exposed by gdb.threads/multi-create-ns-info-thr.exp

Commit 8629910955a7 ("Add thread_db_notice_clone to gdbserver")
introduced calls into libthread_db without making sure that the
current thread is pointing to a know-stopped thread.  This resulted in
sometimes thread_db_notice_clone failing->find_one_thread failing like
this, as seen when running gdb.threads/multi-create-ns-info-thr.exp:

~~~
  Thread <6> executing
  Thread <7> executing
  gdbserver: PID mismatch!  Expected 27472, got 27471
  gdbserver: Cannot find thread after clone.

  Thread <1000> executing
  Thread <1001> executing
~~~

Things go south from here and sometimes that ends up resulting in
gdbserver crashing and the test failing.

gdb/gdbserver/ChangeLog:
2017-09-29  Pedro Alves  <palves@redhat.com>

* linux-low.c (handle_extended_wait): Pass parent thread instead
of process to thread_db_notice_clone.
* linux-low.h (thread_db_notice_clone): Replace parent process
parameter with parent thread parameter.
* thread-db.c (find_one_thread): Add comment.
(thread_db_notice_clone): Replace parent process parameter with
parent thread parameter.  Temporarily switch to the parent thread.

Fail when string merge can't alloc memory

I was looking at Debian bug #874674 again today, and think I might
have spotted the problem.  It appears that merge.c tries to cope with
memory allocation failures in some circumstances, but doesn't quite
manage to get everything right.  This patch will make ld report memory
allocation failures instead of silently not merging strings.

* merge.c (merge_strings): Return FALSE on malloc failure.
(_bfd_merge_sections): Return failures from record_section and
merge_strings.

Automatic date update in version.in

Move utils-selftests.c -> gdb/unittests/

This file was only under gdb/ currently because it predates the
gdb/unittests/ directory.

gdb/ChangeLog:
2017-09-28  Pedro Alves  <palves@redhat.com>

* Makefile.in (SUBDIR_UNITTESTS_SRCS): Add
unittests/common-utils-selftests.c.
(SUBDIR_UNITTESTS_OBS): Add common-utils-selftests.o.
(COMMON_OBS): Remove utils-selftests.o.
* utils-selftests.c: Move to ...
* unittests/common-utils-selftests.c: ... here and rename self
test to "string_printf".

(open_and_init_dwp_file): Protect against some segvs

This is a "tiny patch", no assignment required.

2017-09-28  Alexander Shaposhnikov <alexander.v.shaposhnikov@gmail.com>

* dwarf2read.c (open_and_init_dwp_file): Protect against dwp_file
having NULL cus or tus.

Skip the PR 14918 linker test for ARM targets.

 I am applying a patch that has been lying around in the Fedora
  binutils sources for a while.  It skips the PR14918 linker test for
  ARM based targets.  This test checks that libgcc is not included in a
  link of an empty executable.  This works for most targets, but on the
  ARM the crt1.o startup code calls __libc_csu_init which is in
  /usr/lib/libc_nonshared.a(elf-init.oS).  This in turn needs
  __aeabi_unwind_cpp_pr0@@GCC_3.5 which is provided by libgcc_s.so.1,
  and so the test fails.

PR22220, BFD linker wrongly marks symbols as PREVAILING_DEF_IRONLY

non_ir_ref_dynamic wasn't being set in the case where we have a
versioned dynamic symbol definition with a non-versioned matching IR
symbol.

bfd/
PR 22220
* elflink.c (_bfd_elf_merge_symbol): Set non_ir_ref_dynamic in
a case where plugin_notice isn't called.
ld/
* testsuite/ld-plugin/pr22220.h,
* testsuite/ld-plugin/pr22220lib.cc,
* testsuite/ld-plugin/pr22220lib.ver,
* testsuite/ld-plugin/pr22220main.cc: New test.
* testsuite/ld-plugin/lto.exp: Run it.

Automatic date update in version.in

Complete tdep move to convert_typed_floating

Many tdep files need to perform conversions between two floating-point
types, usually when accessing FP registers.  Most targets now use the
convert_typed_floating helper routine to do so.  However, a small number
still use the old method of converting via a DOUBLEST.  Since we want
to get rid of DOUBLEST, these targets need to be moved to the new
method as well.

The main obstacle is that for convert_typed_floating we need an actual
*type*, not just a floatformat.

In arm-tdep.c, this is very straightforward, since there is already a
type using the ARM extended floatformat.

For sh-tdep.c and sh64-tdep.c, no such type already exists, so I've
added one to the gdbarch_tdep struct as done on other targets.

gdb/ChangeLog
2017-09-27  Ulrich Weigand  <uweigand@de.ibm.com>

* arm-tdep.c: (convert_from_extended): Remove.
(convert_to_extended): Likewise.
(arm_extract_return_value): Use convert_typed_floating.
(arm_store_return_value): Likewise.

* sh-tdep.h (struct gdbarch_tdep): Add sh_littlebyte_bigword_type.
* sh-tdep.c: Do not include "floatformat.h".
(sh_littlebyte_bigword_type): New function.
(sh_register_convert_to_virtual): Use convert_typed_floating.
(sh_register_convert_to_raw): Likewise.
* sh64-tdep.c: (struct gdbarch_tdep): Add sh_littlebyte_bigword_type.
(sh64_littlebyte_bigword_type): New function.
(sh64_extract_return_value): Use convert_typed_floating.
(sh64_register_convert_to_virtual): Likewise.
(sh64_register_convert_to_raw): Likewise.

Simplify floatformat_from_type

For historical reasons, the TYPE_FLOATFORMAT element is still set to hold
an array of two floatformat structs, one for big-endian and the other for
little-endian.  When accessing the element via floatformat_from_type,
the code would check the type's byte order and return the appropriate
floatformat.

However, these days this is quite unnecessary, since the type's byte order
is already known at the time the type is allocated and the floatformat is
installed into TYPE_FLOATFORMAT.  Therefore, we can just install the correct
version here.

Also, moves the (now trivially simple) floatformat_from_type accessor to
gdbtypes.{c,h}, since it doesn't really need to be in doublest.c now.

gdb/ChangeLog
2017-09-27  Ulrich Weigand  <uweigand@de.ibm.com>

* doublest.h (floatformat_from_type): Move to gdbtypes.h.
* doublest.c (floatformat_from_type): Move to gdbtypes.c.

* gdbtypes.h (union type_specific): Make field floatformat hold
just a single struct floatformat, not an array.
(floatformat_from_type): Move here.
* gdbtypes.c (floatformat_from_type): Move here.  Update to
changed TYPE_FLOATFORMAT definition.
(verify_floatformat): Update to changed TYPE_FLOATFORMAT.
(recursive_dump_type): Likewise.
(init_float_type): Install correct floatformat for byte order.
(arch_float_type): Likewise.

Make init_type/arch_type take a size in bits

This changes the interfaces to init_type and arch_type to take the
type length in bits as input (instead of as bytes).  The routines
assert that the length is a multiple of TARGET_CHAR_BIT.

For consistency, arch_flags_type is changed likewise, so that now
all type creation interfaces always use length in bits.

All callers are updated in the straightforward manner.

The assert actually found a bug in read_range_type, where the
init_integer_type routine was called with a wrong argument (probably
a bug introduced with the conversion to use init_integer_type).

gdb/ChangeLog
2017-09-27  Ulrich Weigand  <uweigand@de.ibm.com>

* gdbtypes.c (init_type): Change incoming argument from
length-in-bytes to length-in-bits.  Assert length is a
multiple of TARGET_CHAR_BITS.
(arch_type, arch_flags_type): Likewise.
(init_integer_type): Update call to init_type.
(init_character_type): Likewise.
(init_boolean_type): Likewise.
(init_float_type): Likewise.
(init_decfloat_type): Likewise.
(init_complex_type): Likewise.
(init_pointer_type): Likewise.
(objfile_type): Likewise.
(arch_integer_type): Update call to arch_type.
(arch_character_type): Likewise.
(arch_boolean_type): Likewise.
(arch_float_type): Likewise.
(arch_decfloat_type): Likewise.
(arch_complex_type): Likewise.
(arch_pointer_type): Likewise.
(gdbtypes_post_init): Likewise.

* dwarf2read.c (dwarf2_init_float_type): Update call to init_type.
(read_base_type): Likewise.
* mdebugread.c (basic_type): Likewise.
* stabsread.c (dbx_init_float_type): Likewise.
(rs6000_builtin_type): Likewise.
(read_range_type): Likewise.  Also, fix call to init_integer_type
with erroneous length argument.

* ada-lang.c (ada_language_arch_info): Update call to arch_type.
* d-lang.c (build_d_types): Likewise.
* f-lang.c (build_fortran_types): Likewise.
* go-lang.c (build_go_types): Likewise.
* opencl-lang.c (build_opencl_types): Likewise.
* jit.c (finalize_symtab): Likewise.
* gnu-v3-abi.c (build_gdb_vtable_type): Likewise.
(build_std_type_info_type): Likewise.
* target-descriptions.c (tdesc_gdb_type): Likewise.  Also,
update call to arch_flags_type.

* linux-tdep.c (linux_get_siginfo_type_with_fields): Update call to
arch_type.
* fbsd-tdep.c (fbsd_get_siginfo_type): Likewise.
* windows-tdep.c (windows_get_tlb_type): Likewise.

* avr-tdep.c (avr_gdbarch_init): Update call to arch_type.
* ft32-tdep.c (ft32_gdbarch_init): Likewise.
* m32c-tdep.c (make_types): Likewise.
* rl78-tdep.c (rl78_gdbarch_init): Likewise.
(rl78_psw_type): Update call to arch_flags_type.
* m68k-tdep.c (m68k_ps_type): Update call to arch_flags_type.
* rx-tdep.c (rx_psw_type): Likewise.
(rx_fpsw_type): Likewise.
* sparc-tdep.c (sparc_psr_type): Likewise.
(sparc_fsr_type): Likewise.
* sparc64-tdep.c (sparc64_pstate_type): Likewise.
(sparc64_ccr_type): Likewise.
(sparc64_fsr_type): Likewise.
(sparc64_fprs_type): Likewise.

Add support for the new names of the RISC-V fmv.x.s and fmv.s.x instructions, vis: fmv.x.w and fmv.w.x.

PR 22179
opcodes * riscv-opc.c (riscv_opcodes): Add fmv.x.w and fmv.w.x as the new
names for the fmv.x.s and fmv.s.x instructions respectively.

gas * testsuite/gas/riscv/fmv.x.s: New file: Tests the support for the
renamed fmv.x.s and fmv.s.x instructions.
* testsuite/gas/riscv/fmv.x.d: New file: Test driver.

Constify find_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* findcmd.c (find_command): Constify.

Constify some commands in ada-tasks.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* ada-tasks.c (task_command_1, task_command): Constify.

Constify some commands in symtab.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* symtab.c (maintenance_print_symbol_cache)
(maintenance_flush_symbol_cache)
(maintenance_print_symbol_cache_statistics): Constify.

Constify some commands in inferior.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* inferior.c (detach_inferior_command, kill_inferior_command)
(inferior_command): Constify.

Constify some commands in regcache.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* regcache.c (regcache_print, maintenance_print_registers)
(maintenance_print_raw_registers)
(maintenance_print_cooked_registers)
(maintenance_print_register_groups)
(maintenance_print_remote_registers): Constify.

Constify some commands in printcmd.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* printcmd.c (map_display_numbers, undisplay_command)
(enable_disable_display_command, enable_display_command)
(disable_display_command): Constify.

Constify some commands in breakpoint.c

This also makes delete_command static; but now I wonder if it is used
in Insight and should not be touched.

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* breakpoint.h (delete_command): Don't declare.
* breakpoint.c (delete_command, enable_once_command)
(enable_count_command, enable_delete_command, breakpoint_1)
(maintenance_info_breakpoints, stopin_command, stopat_command)
(delete_command, delete_trace_command, save_breakpoints)
(save_breakpoints_command, save_tracepoints_command): Constify.

Constify some commands in macrocmd.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* macrocmd.c (macro_expand_command, macro_expand_once_command)
(skip_ws, extract_identifier, macro_define_command)
(macro_undef_command, macro_list_command): Constify.

Constify some commands in infcmd.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* infcmd.c (environment_info, set_environment_command)
(unset_environment_command, path_info, info_proc_cmd_1)
(info_proc_cmd_mappings, info_proc_cmd_stat)
(info_proc_cmd_status, info_proc_cmd_cwd, info_proc_cmd_cmdline)
(info_proc_cmd_exe, info_proc_cmd_all): Constify.

Constify some commands in i386-tdep.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* i386-tdep.c (i386_mpx_info_bounds, i386_mpx_set_bounds):
Constify.

Constify add_symbol_file_from_memory_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* symfile-mem.c (add_symbol_file_from_memory_command): Constify.

Constify demangle_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* demangle.c (demangle_command): Constify.

Constify maintenance_info_program_spaces_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* progspace.c (maintenance_info_program_spaces_command):
Constify.

Constify some commands in compile.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* compile/compile.c (check_raw_argument, compile_file_command)
(compile_code_command, compile_print_command): Constify.

Constify maintenance_print_reggroups

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* reggroups.c (maintenance_print_reggroups): Constify.

Constify save_gdb_index_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* dwarf2read.c (save_gdb_index_command): Constify.

Constify info_probes_stap_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* stap-probe.c (info_probes_stap_command): Constify.

Constify unset_exec_wrapper_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* fork-child.c (unset_exec_wrapper_command): Constify.

Constify some commands in btrace.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* btrace.c (get_uint, get_context_size, no_chunk)
(maint_btrace_packet_history_cmd)
(maint_btrace_clear_packet_history_cmd, maint_btrace_clear_cmd)
(maint_info_btrace_cmd): Constify.

Constify delete_bookmark_command

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* reverse.c (delete_bookmark_command): Constify.

Constify some commands in remote.c

gdb/ChangeLog
2017-09-27  Tom Tromey  <tom@tromey.com>

* remote.c (set_memory_packet_size)
(set_memory_write_packet_size, show_memory_write_packet_size)
(set_memory_read_packet_size, show_memory_read_packet_size)
(compare_sections_command, packet_command, remote_put_command)
(remote_get_command, remote_delete_command): Constify.