[PATCH v2 1/2] Fix PR c++/21323: GDB thinks char16_t and char32_t are signed in C++

[PATCH v2 1/2] Fix PR c++/21323: GDB thinks char16_t and char32_t are signed in C++

[Pedro Alves]

New in v2:

 - Make the tests a bit more lax when testing C/C++98, WRT to the
   underlying type of char16_t/char32_t - accept any of unsigned
   long/short/int:

  -	gdb_test "ptype $char16_exp" "type = unsigned short" \
  +	gdb_test "ptype $char16_exp" "type = unsigned (long|int|short)" \
	      "char16_t is typedef"
  -	gdb_test "ptype $char16_exp" "type = unsigned int" \
  +	gdb_test "ptype $char16_exp" "type = unsigned (long|int|short)" \
	      "char32_t is typedef"

  Nothing else changed.

While the C++ standard says that char16_t and char32_t are unsigned types:

 Types char16_t and char32_t denote distinct types with the same size,
 signedness, and alignment as uint_least16_t and uint_least32_t,
 respectively, in <cstdint>, called the underlying types.

... gdb treats them as signed currently:

 (gdb) p (char16_t)-1
 $1 = -1 u'\xffff'

There are actually two places in gdb that hardcode these types:

- gdbtypes.c:gdbtypes_post_init, when creating the built-in types,
  seemingly used by the "x /s" command (judging from commit 9a22f0d0).

- dwarf2read.c, when reading base types with DW_ATE_UTF encoding
  (which is what is used for these types, when compiling for C++11 and
  up).  Despite the comment, the type created does end up used.

Both places need fixing.  But since I couldn't tell why dwarf2read.c
needs to create a new type, I've made it use the per-arch built-in
types instead, so that the types are only created once per arch
instead of once per objfile.  That seems to work fine.

While writting the test, I noticed that the C++ language parser isn't
actually aware of these built-in types, so if you try to use them
without a program that uses them, you get:

 (gdb) set language c++
 (gdb) ptype char16_t
 No symbol table is loaded.  Use the "file" command.
 (gdb) ptype u"hello"
 No type named char16_t.
 (gdb) p u"hello"
 No type named char16_t.

That's fixed by simply adding a couple entries to C++'s built-in types
array in c-lang.c.  With that, we get the expected:

 (gdb) ptype char16_t
 type = char16_t
 (gdb) ptype u"hello"
 type = char16_t [6]
 (gdb) p u"hello"
 $1 = u"hello"

gdb/ChangeLog:
yyyy-mm-dd  Pedro Alves  <palves@redhat.com>

	PR c++/21323
	* c-lang.c (cplus_primitive_types) <cplus_primitive_type_char16_t,
	cplus_primitive_type_char32_t>: New enum values.
	(cplus_language_arch_info): Register cplus_primitive_type_char16_t
	and cplus_primitive_type_char32_t.
	* dwarf2read.c (read_base_type) <DW_ATE_UTF>: If bit size is 16 or
	32, use the archtecture's built-in type for char16_t and char32_t,
	respectively.  Otherwise, fallback to init_integer_type as before,
	but make the type unsigned, and issue a complaint.
	* gdbtypes.c (gdbtypes_post_init): Make char16_t and char32_t unsigned.

gdb/testsuite/ChangeLog:
yyyy-mm-dd  Pedro Alves  <palves@redhat.com>

	PR c++/21323
	* gdb.cp/wide_char_types.c: New file.
	* gdb.cp/wide_char_types.exp: New file.
---
 gdb/c-lang.c                             |   6 ++
 gdb/dwarf2read.c                         |  19 +++-
 gdb/gdbtypes.c                           |   5 +-
 gdb/testsuite/gdb.cp/wide_char_types.c   |  28 ++++++
 gdb/testsuite/gdb.cp/wide_char_types.exp | 143 +++++++++++++++++++++++++++++++
 5 files changed, 195 insertions(+), 6 deletions(-)
 create mode 100644 gdb/testsuite/gdb.cp/wide_char_types.c
 create mode 100644 gdb/testsuite/gdb.cp/wide_char_types.exp

diff --git a/gdb/c-lang.c b/gdb/c-lang.c
index a100199..616aa26 100644
--- a/gdb/c-lang.c
+++ b/gdb/c-lang.c
@@ -895,6 +895,8 @@ enum cplus_primitive_types {
   cplus_primitive_type_decfloat,
   cplus_primitive_type_decdouble,
   cplus_primitive_type_declong,
+  cplus_primitive_type_char16_t,
+  cplus_primitive_type_char32_t,
   nr_cplus_primitive_types
 };
 
@@ -950,6 +952,10 @@ cplus_language_arch_info (struct gdbarch *gdbarch,
     = builtin->builtin_decdouble;
   lai->primitive_type_vector [cplus_primitive_type_declong]
     = builtin->builtin_declong;
+  lai->primitive_type_vector [cplus_primitive_type_char16_t]
+    = builtin->builtin_char16;
+  lai->primitive_type_vector [cplus_primitive_type_char32_t]
+    = builtin->builtin_char32;
 
   lai->bool_type_symbol = "bool";
   lai->bool_type_default = builtin->builtin_bool;
diff --git a/gdb/dwarf2read.c b/gdb/dwarf2read.c
index f1a10c4..1a06f7b 100644
--- a/gdb/dwarf2read.c
+++ b/gdb/dwarf2read.c
@@ -15089,9 +15089,22 @@ read_base_type (struct die_info *die, struct dwarf2_cu *cu)
 	  type = init_integer_type (objfile, bits, 1, name);
 	break;
       case DW_ATE_UTF:
-	/* We just treat this as an integer and then recognize the
-	   type by name elsewhere.  */
-	type = init_integer_type (objfile, bits, 0, name);
+	{
+	  gdbarch *arch = get_objfile_arch (objfile);
+
+	  if (bits == 16)
+	    type = builtin_type (arch)->builtin_char16;
+	  else if (bits == 32)
+	    type = builtin_type (arch)->builtin_char32;
+	  else
+	    {
+	      complaint (&symfile_complaints,
+			 _("unsupported DW_ATE_UTF bit size: '%d'"),
+			 bits);
+	      type = init_integer_type (objfile, bits, 1, name);
+	    }
+	  return set_die_type (die, type, cu);
+	}
 	break;
 
       default:
diff --git a/gdb/gdbtypes.c b/gdb/gdbtypes.c
index 6f3aeab..c1f76fb 100644
--- a/gdb/gdbtypes.c
+++ b/gdb/gdbtypes.c
@@ -5204,10 +5204,9 @@ gdbtypes_post_init (struct gdbarch *gdbarch)
 
   /* Wide character types.  */
   builtin_type->builtin_char16
-    = arch_integer_type (gdbarch, 16, 0, "char16_t");
+    = arch_integer_type (gdbarch, 16, 1, "char16_t");
   builtin_type->builtin_char32
-    = arch_integer_type (gdbarch, 32, 0, "char32_t");
-	
+    = arch_integer_type (gdbarch, 32, 1, "char32_t");
 
   /* Default data/code pointer types.  */
   builtin_type->builtin_data_ptr
diff --git a/gdb/testsuite/gdb.cp/wide_char_types.c b/gdb/testsuite/gdb.cp/wide_char_types.c
new file mode 100644
index 0000000..8337cd4
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/wide_char_types.c
@@ -0,0 +1,28 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2017 Free Software Foundation, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.
+*/
+
+#include <uchar.h>
+
+char16_t u16 = -1;
+char32_t u32 = -1;
+
+int
+main ()
+{
+  return 0;
+}
diff --git a/gdb/testsuite/gdb.cp/wide_char_types.exp b/gdb/testsuite/gdb.cp/wide_char_types.exp
new file mode 100644
index 0000000..df5c8a8f
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/wide_char_types.exp
@@ -0,0 +1,143 @@
+# This testcase is part of GDB, the GNU debugger.
+
+# Copyright 2017 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+# Test GDB's awareness of the char16_t, char32_t (C++11+) built-in
+# types.  We also run most tests here in C mode, and check whether the
+# built-ins are disabled (gdb uses the typedefs in the debug info
+# instead.)
+
+standard_testfile
+
+# Test char16_t/char32_t in language LANG, against symbols in
+# a program.  Lang can be "c", "c++03" or "c++11".  In C++11,
+# char16_t/char32_t are built-in types, and the debug information
+# reflects that (see
+# http://wiki.dwarfstd.org/index.php?title=C%2B%2B0x:_New_string_literals).
+
+proc wide_char_types_program {lang} {
+    global srcfile testfile
+
+    set options {debug}
+    if {$lang == "c++03"} {
+	lappend options c++ additional_flags=-std=c++03
+	set out $testfile-cxx03
+    } elseif {$lang == "c++11"} {
+	lappend options c++ additional_flags=-std=c++11
+	set out $testfile-cxx11
+    } else {
+	set out $testfile-c
+    }
+
+    if { [prepare_for_testing "failed to prepare" \
+	      ${out} [list $srcfile] $options] } {
+	return -1
+    }
+
+    if ![runto_main] then {
+	fail "can't run to main"
+	return 0
+    }
+    do_test_wide_char $lang "u16" "u32"
+}
+
+# Test char16_t/char32_t in language LANG.  Use CHAR16_EXP and
+# CHAR32_EXP as expression for each of the corresponding types.
+# (E.g., CHAR16_EXP will be u16 when testing against the program, and
+# "(char16_t)-1" when testing the built-in types without a program
+# loaded.)
+
+proc do_test_wide_char {lang char16_exp char32_exp} {
+    global gdb_prompt
+
+    # Check that the fixed-width wide types are distinct built-in
+    # types in C++11+.  In other modes, they're instead typedefs,
+    # found in the debug info.
+    if {$lang == "c++11"} {
+	gdb_test "ptype $char16_exp" "type = char16_t" \
+	    "char16_t is distinct"
+	gdb_test "ptype $char32_exp" "type = char32_t" \
+	    "char32_t is distinct"
+    } else {
+	gdb_test "ptype $char16_exp" "type = unsigned (long|int|short)" \
+	    "char16_t is typedef"
+	gdb_test "ptype $char32_exp" "type = unsigned (long|int|short)" \
+	    "char32_t is typedef"
+    }
+
+    # Check that the fixed-width wide char types are unsigned.
+    gdb_test "p $char16_exp" " = 65535 u'\\\\xffff'" \
+	"char16_t is unsigned"
+    gdb_test "p $char32_exp" " = 4294967295 U'\\\\xffffffff'" \
+	"char32_t is unsigned"
+
+    # Check sizeof.  These are fixed-width.
+    gdb_test "p sizeof($char16_exp)" "= 2" \
+	"sizeof($char16_exp) == 2"
+    gdb_test "p sizeof($char32_exp)" "= 4" \
+	"sizeof(char16_t) == 4"
+
+    # Test printing wide literal strings.  Note that when testing with
+    # no program started, this relies on GDB's awareness of the
+    # built-in wide char types.
+    gdb_test {p U"hello"} {= U"hello"}
+    gdb_test {p u"hello"} {= u"hello"}
+}
+
+# Make sure that the char16_t/char32_t types are recognized as
+# distinct built-in types in C++ mode, even with no program loaded.
+# Check that in C mode, the types are not recognized.
+
+proc wide_char_types_no_program {} {
+    global srcfile testfile
+
+    gdb_exit
+    gdb_start
+
+    # These types are not built-in in C.
+    with_test_prefix "c" {
+	gdb_test "set language c"
+
+	gdb_test "p (char16_t) -1" "No symbol table is loaded.*" \
+	    "char16_t is not built-in"
+	gdb_test "p (char32_t) -1" "No symbol table is loaded.*" \
+	    "char32_t is not built-in"
+
+	gdb_test {p U"hello"} "No type named char32_t\\\."
+	gdb_test {p u"hello"} "No type named char16_t\\\."
+    }
+
+    # Note GDB does not distinguish C++ dialects, so the fixed-width
+    # types are always available in C++ mode, even if they were not
+    # built-in types before C++11.
+    with_test_prefix "c++" {
+	gdb_test "set language c++"
+
+	do_test_wide_char "c++11" "(char16_t) -1" "(char32_t) -1"
+    }
+}
+
+# Check wide char types with no program loaded.
+with_test_prefix "no program" {
+    wide_char_types_no_program
+}
+
+# Check types when a program is loaded.
+with_test_prefix "with program" {
+    foreach_with_prefix lang {"c" "c++03" "c++11"} {
+	wide_char_types_program $lang
+    }
+}
-- 
2.5.5

[PATCH v2 0/2] PR c++/21323: char16_t/char32_t/wchar_t built-in C++ types

[Pedro Alves]

This series fixes PR c++/21323, a bug report that shows that GDB
thinks C++11's char16_t and char32_t built-in types are signed, while
the standard clearly says they're unsigned.

While working on it, I noticed that GDB isn't aware that wchar_t is a
built-in type in C++ either (and it was already the case in C++98).
The second patch fixes that.

Tested on x86_64 Fedora 23.

New in v2:

 - Make the tests a bit more lax when testing C/C++98, WRT to the
   underlying type of char16_t/char32_t - accept any of unsigned
   long/short/int.

 - Go through all ports and override wchar_t size/sign appropriately.

Pedro Alves (2):
  Fix PR c++/21323: GDB thinks char16_t and char32_t are signed in C++
  Teach GDB that wchar_t is a built-in type in C++ mode

 gdb/aarch64-tdep.c                       |   2 +
 gdb/alpha-tdep.c                         |   2 +
 gdb/arm-tdep.c                           |   7 ++
 gdb/avr-tdep.c                           |   3 +
 gdb/c-lang.c                             |   9 ++
 gdb/dwarf2read.c                         |  19 +++-
 gdb/gdbarch.c                            |  48 ++++++++
 gdb/gdbarch.h                            |  11 ++
 gdb/gdbarch.sh                           |   6 +
 gdb/gdbtypes.c                           |   8 +-
 gdb/gdbtypes.h                           |   1 +
 gdb/h8300-tdep.c                         |   4 +
 gdb/i386-nto-tdep.c                      |   3 +
 gdb/m32r-tdep.c                          |   3 +
 gdb/moxie-tdep.c                         |   3 +
 gdb/nds32-tdep.c                         |   3 +
 gdb/rs6000-aix-tdep.c                    |   5 +
 gdb/sh-tdep.c                            |   4 +
 gdb/sparc-tdep.c                         |   3 +
 gdb/sparc64-tdep.c                       |   3 +
 gdb/testsuite/gdb.cp/wide_char_types.c   |  30 +++++
 gdb/testsuite/gdb.cp/wide_char_types.exp | 181 +++++++++++++++++++++++++++++++
 gdb/windows-tdep.c                       |   3 +
 gdb/xstormy16-tdep.c                     |   3 +
 gdb/xtensa-tdep.c                        |   3 +
 25 files changed, 361 insertions(+), 6 deletions(-)
 create mode 100644 gdb/testsuite/gdb.cp/wide_char_types.c
 create mode 100644 gdb/testsuite/gdb.cp/wide_char_types.exp

-- 
2.5.5

[PATCH v2 2/2] Teach GDB that wchar_t is a built-in type in C++ mode

[Pedro Alves]

New in v2:

  - Go through all ports and override wchar_t size/sign appropriately.

GDB is currently not aware that wchar_t is a built-in type in C++
mode.  This is usually not a problem because the debug info describes
the type, so when you have a program loaded, you don't notice this.
However, if you try expressions involving wchar_t before a program is
loaded, gdb errors out:

 (gdb) p (wchar_t)-1
 No symbol table is loaded.  Use the "file" command.
 (gdb) p L"hello"
 No type named wchar_t.
 (gdb) ptype L"hello"
 No type named wchar_t.

This commit teaches gdb about the type.  After:

 (gdb) p (wchar_t)-1
 $1 = -1 L'\xffffffff'
 (gdb) p L"hello"
 $2 = L"hello"
 (gdb) ptype L"hello"
 type = wchar_t [6]

Unlike char16_t/char32_t, unfortunately, the underlying type of
wchar_t is implementation dependent, both size and signness.  So this
requires adding a couple new gdbarch hooks.

I grepped the GCC code base for WCHAR_TYPE and WCHAR_TYPE_SIZE, and it
seems to me that the majority of the ABIs have a 4-byte signed
wchar_t, so that's what I made the default for GDB too.  And then I
looked for which ports set the type size to either 16 or to an
unsigned type, and made GDB follow suit.

gdb/ChangeLog:
yyyy-mm-dd  Pedro Alves  <palves@redhat.com>

	PR gdb/21323
	* c-lang.c (cplus_primitive_types) <cplus_primitive_type_wchar_t>:
	New enum value.
	(cplus_language_arch_info): Register cplus_primitive_type_wchar_t.
	* gdbtypes.h (struct builtin_type) <builtin_wchar>: New field.
	* gdbtypes.c (gdbtypes_post_init): Create the "wchar_t" type.
	* gdbarch.sh (wchar_bit, wchar_signed): New per-arch values.
	* gdbarch.h, gdbarch.c: Regenerate.
	* aarch64-tdep.c (aarch64_gdbarch_init): Override
	gdbarch_wchar_bit and gdbarch_wchar_signed.
	* alpha-tdep.c (alpha_gdbarch_init): Likewise.
	* arm-tdep.c (arm_gdbarch_init): Likewise.
	* avr-tdep.c (avr_gdbarch_init): Likewise.
	* h8300-tdep.c (h8300_gdbarch_init): Likewise.
	* i386-nto-tdep.c (i386nto_init_abi): Likewise.
	* m32r-tdep.c (m32r_gdbarch_init): Likewise.
	* moxie-tdep.c (moxie_gdbarch_init): Likewise.
	* nds32-tdep.c (nds32_gdbarch_init): Likewise.
	* rs6000-aix-tdep.c (rs6000_aix_init_osabi): Likewise.
	* sh-tdep.c (sh_gdbarch_init): Likewise.
	* sparc-tdep.c (sparc32_gdbarch_init): Likewise.
	* sparc64-tdep.c (sparc64_init_abi): Likewise.
	* windows-tdep.c (windows_init_abi): Likewise.
	* xstormy16-tdep.c (xstormy16_gdbarch_init): Likewise.

gdb/testsuite/ChangeLog:
yyyy-mm-dd  Pedro Alves  <palves@redhat.com>

	PR gdb/21323
	* gdb.cp/wide_char_types.c: Include <wchar.h>.
	(wchar): New global.
	* gdb.cp/wide_char_types.exp (wide_char_types_program)
	(do_test_wide_char, wide_char_types_no_program, top level): Add
	wchar_t testing.
---
 gdb/aarch64-tdep.c                       |  2 +
 gdb/alpha-tdep.c                         |  2 +
 gdb/arm-tdep.c                           |  7 ++++
 gdb/avr-tdep.c                           |  3 ++
 gdb/c-lang.c                             |  3 ++
 gdb/gdbarch.c                            | 48 +++++++++++++++++++++++
 gdb/gdbarch.h                            | 11 ++++++
 gdb/gdbarch.sh                           |  6 +++
 gdb/gdbtypes.c                           |  3 ++
 gdb/gdbtypes.h                           |  1 +
 gdb/h8300-tdep.c                         |  4 ++
 gdb/i386-nto-tdep.c                      |  3 ++
 gdb/m32r-tdep.c                          |  3 ++
 gdb/moxie-tdep.c                         |  3 ++
 gdb/nds32-tdep.c                         |  3 ++
 gdb/rs6000-aix-tdep.c                    |  5 +++
 gdb/sh-tdep.c                            |  4 ++
 gdb/sparc-tdep.c                         |  3 ++
 gdb/sparc64-tdep.c                       |  3 ++
 gdb/testsuite/gdb.cp/wide_char_types.c   |  2 +
 gdb/testsuite/gdb.cp/wide_char_types.exp | 66 +++++++++++++++++++++++++-------
 gdb/windows-tdep.c                       |  3 ++
 gdb/xstormy16-tdep.c                     |  3 ++
 gdb/xtensa-tdep.c                        |  3 ++
 24 files changed, 180 insertions(+), 14 deletions(-)

diff --git a/gdb/aarch64-tdep.c b/gdb/aarch64-tdep.c
index f340d57..28c2573 100644
--- a/gdb/aarch64-tdep.c
+++ b/gdb/aarch64-tdep.c
@@ -2977,6 +2977,8 @@ aarch64_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_long_long_bit (gdbarch, 64);
   set_gdbarch_ptr_bit (gdbarch, 64);
   set_gdbarch_char_signed (gdbarch, 0);
+  set_gdbarch_wchar_bit (gdbarch, 64);
+  set_gdbarch_wchar_signed (gdbarch, 0);
   set_gdbarch_float_format (gdbarch, floatformats_ieee_single);
   set_gdbarch_double_format (gdbarch, floatformats_ieee_double);
   set_gdbarch_long_double_format (gdbarch, floatformats_ia64_quad);
diff --git a/gdb/alpha-tdep.c b/gdb/alpha-tdep.c
index 4dd65c5..7c521db 100644
--- a/gdb/alpha-tdep.c
+++ b/gdb/alpha-tdep.c
@@ -1771,6 +1771,8 @@ alpha_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_int_bit (gdbarch, 32);
   set_gdbarch_long_bit (gdbarch, 64);
   set_gdbarch_long_long_bit (gdbarch, 64);
+  set_gdbarch_wchar_bit (gdbarch, 64);
+  set_gdbarch_wchar_signed (gdbarch, 0);
   set_gdbarch_float_bit (gdbarch, 32);
   set_gdbarch_double_bit (gdbarch, 64);
   set_gdbarch_long_double_bit (gdbarch, 64);
diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
index b3c3705..008e886 100644
--- a/gdb/arm-tdep.c
+++ b/gdb/arm-tdep.c
@@ -9404,6 +9404,13 @@ arm_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   /* On ARM targets char defaults to unsigned.  */
   set_gdbarch_char_signed (gdbarch, 0);
 
+  /* wchar_t is unsigned under the AAPCS.  */
+  if (tdep->arm_abi == ARM_ABI_AAPCS)
+    set_gdbarch_wchar_signed (gdbarch, 0);
+  else
+    set_gdbarch_wchar_signed (gdbarch, 1);
+  set_gdbarch_wchar_bit (gdbarch, 32);
+
   /* Note: for displaced stepping, this includes the breakpoint, and one word
      of additional scratch space.  This setting isn't used for anything beside
      displaced stepping at present.  */
diff --git a/gdb/avr-tdep.c b/gdb/avr-tdep.c
index 0ae8a08..ec9db1c 100644
--- a/gdb/avr-tdep.c
+++ b/gdb/avr-tdep.c
@@ -1472,6 +1472,9 @@ avr_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_ptr_bit (gdbarch, 2 * TARGET_CHAR_BIT);
   set_gdbarch_addr_bit (gdbarch, 32);
 
+  set_gdbarch_wchar_bit (gdbarch, 2 * TARGET_CHAR_BIT);
+  set_gdbarch_wchar_signed (gdbarch, 1);
+
   set_gdbarch_float_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_long_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
diff --git a/gdb/c-lang.c b/gdb/c-lang.c
index 616aa26..19a8608 100644
--- a/gdb/c-lang.c
+++ b/gdb/c-lang.c
@@ -897,6 +897,7 @@ enum cplus_primitive_types {
   cplus_primitive_type_declong,
   cplus_primitive_type_char16_t,
   cplus_primitive_type_char32_t,
+  cplus_primitive_type_wchar_t,
   nr_cplus_primitive_types
 };
 
@@ -956,6 +957,8 @@ cplus_language_arch_info (struct gdbarch *gdbarch,
     = builtin->builtin_char16;
   lai->primitive_type_vector [cplus_primitive_type_char32_t]
     = builtin->builtin_char32;
+  lai->primitive_type_vector [cplus_primitive_type_wchar_t]
+    = builtin->builtin_wchar;
 
   lai->bool_type_symbol = "bool";
   lai->bool_type_default = builtin->builtin_bool;
diff --git a/gdb/gdbarch.c b/gdb/gdbarch.c
index 87eafb2..8fd9c00 100644
--- a/gdb/gdbarch.c
+++ b/gdb/gdbarch.c
@@ -184,6 +184,8 @@ struct gdbarch
   const struct floatformat ** double_format;
   int long_double_bit;
   const struct floatformat ** long_double_format;
+  int wchar_bit;
+  int wchar_signed;
   gdbarch_floatformat_for_type_ftype *floatformat_for_type;
   int ptr_bit;
   int addr_bit;
@@ -389,6 +391,8 @@ gdbarch_alloc (const struct gdbarch_info *info,
   gdbarch->float_bit = 4*TARGET_CHAR_BIT;
   gdbarch->double_bit = 8*TARGET_CHAR_BIT;
   gdbarch->long_double_bit = 8*TARGET_CHAR_BIT;
+  gdbarch->wchar_bit = 4*TARGET_CHAR_BIT;
+  gdbarch->wchar_signed = -1;
   gdbarch->floatformat_for_type = default_floatformat_for_type;
   gdbarch->ptr_bit = gdbarch->int_bit;
   gdbarch->char_signed = -1;
@@ -533,6 +537,9 @@ verify_gdbarch (struct gdbarch *gdbarch)
   /* Skip verify of long_double_bit, invalid_p == 0 */
   if (gdbarch->long_double_format == 0)
     gdbarch->long_double_format = floatformats_ieee_double;
+  /* Skip verify of wchar_bit, invalid_p == 0 */
+  if (gdbarch->wchar_signed == -1)
+    gdbarch->wchar_signed = 1;
   /* Skip verify of floatformat_for_type, invalid_p == 0 */
   /* Skip verify of ptr_bit, invalid_p == 0 */
   if (gdbarch->addr_bit == 0)
@@ -1457,6 +1464,12 @@ gdbarch_dump (struct gdbarch *gdbarch, struct ui_file *file)
                       "gdbarch_dump: vtable_function_descriptors = %s\n",
                       plongest (gdbarch->vtable_function_descriptors));
   fprintf_unfiltered (file,
+                      "gdbarch_dump: wchar_bit = %s\n",
+                      plongest (gdbarch->wchar_bit));
+  fprintf_unfiltered (file,
+                      "gdbarch_dump: wchar_signed = %s\n",
+                      plongest (gdbarch->wchar_signed));
+  fprintf_unfiltered (file,
                       "gdbarch_dump: gdbarch_write_pc_p() = %d\n",
                       gdbarch_write_pc_p (gdbarch));
   fprintf_unfiltered (file,
@@ -1757,6 +1770,41 @@ set_gdbarch_long_double_format (struct gdbarch *gdbarch,
   gdbarch->long_double_format = long_double_format;
 }
 
+int
+gdbarch_wchar_bit (struct gdbarch *gdbarch)
+{
+  gdb_assert (gdbarch != NULL);
+  /* Skip verify of wchar_bit, invalid_p == 0 */
+  if (gdbarch_debug >= 2)
+    fprintf_unfiltered (gdb_stdlog, "gdbarch_wchar_bit called\n");
+  return gdbarch->wchar_bit;
+}
+
+void
+set_gdbarch_wchar_bit (struct gdbarch *gdbarch,
+                       int wchar_bit)
+{
+  gdbarch->wchar_bit = wchar_bit;
+}
+
+int
+gdbarch_wchar_signed (struct gdbarch *gdbarch)
+{
+  gdb_assert (gdbarch != NULL);
+  /* Check variable changed from pre-default.  */
+  gdb_assert (gdbarch->wchar_signed != -1);
+  if (gdbarch_debug >= 2)
+    fprintf_unfiltered (gdb_stdlog, "gdbarch_wchar_signed called\n");
+  return gdbarch->wchar_signed;
+}
+
+void
+set_gdbarch_wchar_signed (struct gdbarch *gdbarch,
+                          int wchar_signed)
+{
+  gdbarch->wchar_signed = wchar_signed;
+}
+
 const struct floatformat **
 gdbarch_floatformat_for_type (struct gdbarch *gdbarch, const char *name, int length)
 {
diff --git a/gdb/gdbarch.h b/gdb/gdbarch.h
index 34f82a7..5301da1 100644
--- a/gdb/gdbarch.h
+++ b/gdb/gdbarch.h
@@ -188,6 +188,17 @@ extern void set_gdbarch_long_double_bit (struct gdbarch *gdbarch, int long_doubl
 extern const struct floatformat ** gdbarch_long_double_format (struct gdbarch *gdbarch);
 extern void set_gdbarch_long_double_format (struct gdbarch *gdbarch, const struct floatformat ** long_double_format);
 
+/* The ABI default bit-size for "wchar_t".  wchar_t is a built-in type
+   starting with C++11. */
+
+extern int gdbarch_wchar_bit (struct gdbarch *gdbarch);
+extern void set_gdbarch_wchar_bit (struct gdbarch *gdbarch, int wchar_bit);
+
+/* One if `wchar_t' is signed, zero if unsigned. */
+
+extern int gdbarch_wchar_signed (struct gdbarch *gdbarch);
+extern void set_gdbarch_wchar_signed (struct gdbarch *gdbarch, int wchar_signed);
+
 /* Returns the floating-point format to be used for values of length LENGTH.
    NAME, if non-NULL, is the type name, which may be used to distinguish
    different target formats of the same length. */
diff --git a/gdb/gdbarch.sh b/gdb/gdbarch.sh
index 39b1f94..d90755c 100755
--- a/gdb/gdbarch.sh
+++ b/gdb/gdbarch.sh
@@ -383,6 +383,12 @@ v:const struct floatformat **:double_format:::::floatformats_ieee_double::pforma
 v:int:long_double_bit:::8 * sizeof (long double):8*TARGET_CHAR_BIT::0
 v:const struct floatformat **:long_double_format:::::floatformats_ieee_double::pformat (gdbarch->long_double_format)
 
+# The ABI default bit-size for "wchar_t".  wchar_t is a built-in type
+# starting with C++11.
+v:int:wchar_bit:::8 * sizeof (wchar_t):4*TARGET_CHAR_BIT::0
+# One if \`wchar_t' is signed, zero if unsigned.
+v:int:wchar_signed:::1:-1:1
+
 # Returns the floating-point format to be used for values of length LENGTH.
 # NAME, if non-NULL, is the type name, which may be used to distinguish
 # different target formats of the same length.
diff --git a/gdb/gdbtypes.c b/gdb/gdbtypes.c
index c1f76fb..dd3992c 100644
--- a/gdb/gdbtypes.c
+++ b/gdb/gdbtypes.c
@@ -5207,6 +5207,9 @@ gdbtypes_post_init (struct gdbarch *gdbarch)
     = arch_integer_type (gdbarch, 16, 1, "char16_t");
   builtin_type->builtin_char32
     = arch_integer_type (gdbarch, 32, 1, "char32_t");
+  builtin_type->builtin_wchar
+    = arch_integer_type (gdbarch, gdbarch_wchar_bit (gdbarch),
+			 !gdbarch_wchar_signed (gdbarch), "wchar_t");
 
   /* Default data/code pointer types.  */
   builtin_type->builtin_data_ptr
diff --git a/gdb/gdbtypes.h b/gdb/gdbtypes.h
index 4abeaf3..cfb9c28 100644
--- a/gdb/gdbtypes.h
+++ b/gdb/gdbtypes.h
@@ -1509,6 +1509,7 @@ struct builtin_type
   /* Wide character types.  */
   struct type *builtin_char16;
   struct type *builtin_char32;
+  struct type *builtin_wchar;
 
   /* Pointer types.  */
 
diff --git a/gdb/h8300-tdep.c b/gdb/h8300-tdep.c
index 10ef796..c055a9b 100644
--- a/gdb/h8300-tdep.c
+++ b/gdb/h8300-tdep.c
@@ -1381,6 +1381,10 @@ h8300_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_int_bit (gdbarch, 2 * TARGET_CHAR_BIT);
   set_gdbarch_long_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_long_long_bit (gdbarch, 8 * TARGET_CHAR_BIT);
+
+  set_gdbarch_wchar_bit (gdbarch, 2 * TARGET_CHAR_BIT);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   set_gdbarch_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_double_format (gdbarch, floatformats_ieee_single);
   set_gdbarch_long_double_bit (gdbarch, 4 * TARGET_CHAR_BIT);
diff --git a/gdb/i386-nto-tdep.c b/gdb/i386-nto-tdep.c
index a12a56d..3fe715f 100644
--- a/gdb/i386-nto-tdep.c
+++ b/gdb/i386-nto-tdep.c
@@ -362,6 +362,9 @@ i386nto_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
         = nto_in_dynsym_resolve_code;
     }
   set_solib_ops (gdbarch, &nto_svr4_so_ops);
+
+  set_gdbarch_wchar_bit (gdbarch, 32);
+  set_gdbarch_wchar_signed (gdbarch, 0);
 }
 
 /* Provide a prototype to silence -Wmissing-prototypes.  */
diff --git a/gdb/m32r-tdep.c b/gdb/m32r-tdep.c
index 40f29d3..e03b0bd 100644
--- a/gdb/m32r-tdep.c
+++ b/gdb/m32r-tdep.c
@@ -908,6 +908,9 @@ m32r_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   tdep = XNEW (struct gdbarch_tdep);
   gdbarch = gdbarch_alloc (&info, tdep);
 
+  set_gdbarch_wchar_bit (gdbarch, 16);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   set_gdbarch_read_pc (gdbarch, m32r_read_pc);
   set_gdbarch_unwind_sp (gdbarch, m32r_unwind_sp);
 
diff --git a/gdb/moxie-tdep.c b/gdb/moxie-tdep.c
index 7383f27..c5403bb 100644
--- a/gdb/moxie-tdep.c
+++ b/gdb/moxie-tdep.c
@@ -1112,6 +1112,9 @@ moxie_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   tdep = XNEW (struct gdbarch_tdep);
   gdbarch = gdbarch_alloc (&info, tdep);
 
+  set_gdbarch_wchar_bit (gdbarch, 32);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   set_gdbarch_read_pc (gdbarch, moxie_read_pc);
   set_gdbarch_write_pc (gdbarch, moxie_write_pc);
   set_gdbarch_unwind_sp (gdbarch, moxie_unwind_sp);
diff --git a/gdb/nds32-tdep.c b/gdb/nds32-tdep.c
index 05c48aa..e0eea37 100644
--- a/gdb/nds32-tdep.c
+++ b/gdb/nds32-tdep.c
@@ -2069,6 +2069,9 @@ nds32_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
 
   gdbarch = gdbarch_alloc (&info, tdep);
 
+  set_gdbarch_wchar_bit (gdbarch, 16);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   if (fpu_freg == -1)
     num_regs = NDS32_NUM_REGS;
   else if (use_pseudo_fsrs == 1)
diff --git a/gdb/rs6000-aix-tdep.c b/gdb/rs6000-aix-tdep.c
index 9841a60..a14e93c 100644
--- a/gdb/rs6000-aix-tdep.c
+++ b/gdb/rs6000-aix-tdep.c
@@ -1080,6 +1080,11 @@ rs6000_aix_init_osabi (struct gdbarch_info info, struct gdbarch *gdbarch)
   else
     set_gdbarch_frame_red_zone_size (gdbarch, 0);
 
+  if (tdep->wordsize == 8)
+    set_gdbarch_wchar_bit (gdbarch, 32);
+  else
+    set_gdbarch_wchar_bit (gdbarch, 16);
+  set_gdbarch_wchar_signed (gdbarch, 0);
   set_gdbarch_auto_wide_charset (gdbarch, rs6000_aix_auto_wide_charset);
 
   set_solib_ops (gdbarch, &solib_aix_so_ops);
diff --git a/gdb/sh-tdep.c b/gdb/sh-tdep.c
index 9b3692d..e85d94d 100644
--- a/gdb/sh-tdep.c
+++ b/gdb/sh-tdep.c
@@ -2268,6 +2268,10 @@ sh_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_int_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_long_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_long_long_bit (gdbarch, 8 * TARGET_CHAR_BIT);
+
+  set_gdbarch_wchar_bit (gdbarch, 2 * TARGET_CHAR_BIT);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   set_gdbarch_float_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_double_bit (gdbarch, 8 * TARGET_CHAR_BIT);
   set_gdbarch_long_double_bit (gdbarch, 8 * TARGET_CHAR_BIT);
diff --git a/gdb/sparc-tdep.c b/gdb/sparc-tdep.c
index d346aec..078907a 100644
--- a/gdb/sparc-tdep.c
+++ b/gdb/sparc-tdep.c
@@ -1749,6 +1749,9 @@ sparc32_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_long_double_bit (gdbarch, 128);
   set_gdbarch_long_double_format (gdbarch, floatformats_sparc_quad);
 
+  set_gdbarch_wchar_bit (gdbarch, 16);
+  set_gdbarch_wchar_signed (gdbarch, 1);
+
   set_gdbarch_num_regs (gdbarch, SPARC32_NUM_REGS);
   set_gdbarch_register_name (gdbarch, sparc32_register_name);
   set_gdbarch_register_type (gdbarch, sparc32_register_type);
diff --git a/gdb/sparc64-tdep.c b/gdb/sparc64-tdep.c
index bf0da18..58f5bf0 100644
--- a/gdb/sparc64-tdep.c
+++ b/gdb/sparc64-tdep.c
@@ -1268,6 +1268,9 @@ sparc64_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
   set_gdbarch_long_long_bit (gdbarch, 64);
   set_gdbarch_ptr_bit (gdbarch, 64);
 
+  set_gdbarch_wchar_bit (gdbarch, 16);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   set_gdbarch_num_regs (gdbarch, SPARC64_NUM_REGS);
   set_gdbarch_register_name (gdbarch, sparc64_register_name);
   set_gdbarch_register_type (gdbarch, sparc64_register_type);
diff --git a/gdb/testsuite/gdb.cp/wide_char_types.c b/gdb/testsuite/gdb.cp/wide_char_types.c
index 8337cd4..c899b71 100644
--- a/gdb/testsuite/gdb.cp/wide_char_types.c
+++ b/gdb/testsuite/gdb.cp/wide_char_types.c
@@ -17,9 +17,11 @@
 */
 
 #include <uchar.h>
+#include <wchar.h>
 
 char16_t u16 = -1;
 char32_t u32 = -1;
+wchar_t wchar = -1;
 
 int
 main ()
diff --git a/gdb/testsuite/gdb.cp/wide_char_types.exp b/gdb/testsuite/gdb.cp/wide_char_types.exp
index df5c8a8f..6a91350 100644
--- a/gdb/testsuite/gdb.cp/wide_char_types.exp
+++ b/gdb/testsuite/gdb.cp/wide_char_types.exp
@@ -15,14 +15,14 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-# Test GDB's awareness of the char16_t, char32_t (C++11+) built-in
-# types.  We also run most tests here in C mode, and check whether the
-# built-ins are disabled (gdb uses the typedefs in the debug info
-# instead.)
+# Test GDB's awareness of the wchar_t (C++98+) and char16_t, char32_t
+# (C++11+) built-in types.  We also run most tests here in C mode, and
+# check whether the built-ins are disabled (gdb uses the typedefs in
+# the debug info instead.)
 
 standard_testfile
 
-# Test char16_t/char32_t in language LANG, against symbols in
+# Test char16_t/char32_t/wchar_t in language LANG, against symbols in
 # a program.  Lang can be "c", "c++03" or "c++11".  In C++11,
 # char16_t/char32_t are built-in types, and the debug information
 # reflects that (see
@@ -51,16 +51,16 @@ proc wide_char_types_program {lang} {
 	fail "can't run to main"
 	return 0
     }
-    do_test_wide_char $lang "u16" "u32"
+    do_test_wide_char $lang "u16" "u32" "wchar"
 }
 
-# Test char16_t/char32_t in language LANG.  Use CHAR16_EXP and
-# CHAR32_EXP as expression for each of the corresponding types.
-# (E.g., CHAR16_EXP will be u16 when testing against the program, and
-# "(char16_t)-1" when testing the built-in types without a program
-# loaded.)
+# Test char16_t/char32_t/wchar_t in language LANG.  Use CHAR16_EXP,
+# CHAR32_EXP, and WCHAR_EXP as expression for each of the
+# corresponding types.  (E.g., CHAR16_EXP will be u16 when testing
+# against the program, and "(char16_t)-1" when testing the built-in
+# types without a program loaded.)
 
-proc do_test_wide_char {lang char16_exp char32_exp} {
+proc do_test_wide_char {lang char16_exp char32_exp wchar_exp} {
     global gdb_prompt
 
     # Check that the fixed-width wide types are distinct built-in
@@ -78,26 +78,60 @@ proc do_test_wide_char {lang char16_exp char32_exp} {
 	    "char32_t is typedef"
     }
 
+    # wchar_t is a disctinct built-in type in C++03+.
+    if {$lang != "c"} {
+	gdb_test "ptype $wchar_exp" "type = wchar_t" \
+	    "wchar_t is distinct"
+    } else {
+	gdb_test "ptype $wchar_exp" "type = (unsigned )?(long|int|short)" \
+	    "wchar_t is typedef"
+    }
+
     # Check that the fixed-width wide char types are unsigned.
     gdb_test "p $char16_exp" " = 65535 u'\\\\xffff'" \
 	"char16_t is unsigned"
     gdb_test "p $char32_exp" " = 4294967295 U'\\\\xffffffff'" \
 	"char32_t is unsigned"
 
+    # Whether wchar_t is signed is implementation-dependent.  While we
+    # ignore whether GDB got the ABI size/sign details right here,
+    # this at least verifies that the value isn't garbage, and that
+    # GDB correctly outputs the character using the "L" prefix.
+    set test "wchar_t sign"
+    gdb_test_multiple "p $wchar_exp" $test {
+	-re " = 4294967295 L'\\\\xffffffff'\r\n$gdb_prompt $" {
+	    pass "$test (unsigned)"
+	}
+	-re " = 65535 L'\\\\xffff'\r\n$gdb_prompt $" {
+	    pass "$test (unsigned)"
+	}
+	-re " = -1 L'\\\\xffffffff'\r\n$gdb_prompt $" {
+	    pass "$test (signed)"
+	}
+	-re " = -1 L'\\\\xffff'\r\n$gdb_prompt $" {
+	    pass "$test (signed)"
+	}
+    }
+
     # Check sizeof.  These are fixed-width.
     gdb_test "p sizeof($char16_exp)" "= 2" \
 	"sizeof($char16_exp) == 2"
     gdb_test "p sizeof($char32_exp)" "= 4" \
 	"sizeof(char16_t) == 4"
 
+    # Size of wchar_t depends on ABI.
+    gdb_test "p sizeof($wchar_exp)" "= (2|4)" \
+	"sizeof(wchar_t)"
+
     # Test printing wide literal strings.  Note that when testing with
     # no program started, this relies on GDB's awareness of the
     # built-in wide char types.
     gdb_test {p U"hello"} {= U"hello"}
     gdb_test {p u"hello"} {= u"hello"}
+    gdb_test {p L"hello"} {= L"hello"}
 }
 
-# Make sure that the char16_t/char32_t types are recognized as
+# Make sure that the char16_t/char32_t/wchar_t types are recognized as
 # distinct built-in types in C++ mode, even with no program loaded.
 # Check that in C mode, the types are not recognized.
 
@@ -116,8 +150,12 @@ proc wide_char_types_no_program {} {
 	gdb_test "p (char32_t) -1" "No symbol table is loaded.*" \
 	    "char32_t is not built-in"
 
+	gdb_test "p (wchar_t) -1" "No symbol table is loaded.*" \
+	    "wchar_t is not built-in"
+
 	gdb_test {p U"hello"} "No type named char32_t\\\."
 	gdb_test {p u"hello"} "No type named char16_t\\\."
+	gdb_test {p L"hello"} "No type named wchar_t\\\."
     }
 
     # Note GDB does not distinguish C++ dialects, so the fixed-width
@@ -126,7 +164,7 @@ proc wide_char_types_no_program {} {
     with_test_prefix "c++" {
 	gdb_test "set language c++"
 
-	do_test_wide_char "c++11" "(char16_t) -1" "(char32_t) -1"
+	do_test_wide_char "c++11" "(char16_t) -1" "(char32_t) -1" "(wchar_t) -1"
     }
 }
 
diff --git a/gdb/windows-tdep.c b/gdb/windows-tdep.c
index 62a303d..3e7e8f8 100644
--- a/gdb/windows-tdep.c
+++ b/gdb/windows-tdep.c
@@ -466,6 +466,9 @@ init_w32_command_list (void)
 void
 windows_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
 {
+  set_gdbarch_wchar_bit (gdbarch, 16);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   /* Canonical paths on this target look like
      `c:\Program Files\Foo App\mydll.dll', for example.  */
   set_gdbarch_has_dos_based_file_system (gdbarch, 1);
diff --git a/gdb/xstormy16-tdep.c b/gdb/xstormy16-tdep.c
index 91983f8..952f952 100644
--- a/gdb/xstormy16-tdep.c
+++ b/gdb/xstormy16-tdep.c
@@ -809,6 +809,9 @@ xstormy16_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   set_gdbarch_long_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_long_long_bit (gdbarch, 8 * TARGET_CHAR_BIT);
 
+  set_gdbarch_wchar_bit (gdbarch, 2 * TARGET_CHAR_BIT);
+  set_gdbarch_wchar_signed (gdbarch, 1);
+
   set_gdbarch_float_bit (gdbarch, 4 * TARGET_CHAR_BIT);
   set_gdbarch_double_bit (gdbarch, 8 * TARGET_CHAR_BIT);
   set_gdbarch_long_double_bit (gdbarch, 8 * TARGET_CHAR_BIT);
diff --git a/gdb/xtensa-tdep.c b/gdb/xtensa-tdep.c
index 13f1514..d990be0 100644
--- a/gdb/xtensa-tdep.c
+++ b/gdb/xtensa-tdep.c
@@ -3200,6 +3200,9 @@ xtensa_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
   xtensa_verify_config (gdbarch);
   xtensa_session_once_reported = 0;
 
+  set_gdbarch_wchar_bit (gdbarch, 2 * TARGET_CHAR_BIT);
+  set_gdbarch_wchar_signed (gdbarch, 0);
+
   /* Pseudo-Register read/write.  */
   set_gdbarch_pseudo_register_read (gdbarch, xtensa_pseudo_register_read);
   set_gdbarch_pseudo_register_write (gdbarch, xtensa_pseudo_register_write);
-- 
2.5.5

Re: [PATCH v2 2/2] Teach GDB that wchar_t is a built-in type in C++ mode

[Eli Zaretskii]

> From: Pedro Alves <palves@redhat.com>
> Date: Fri, 31 Mar 2017 12:03:15 +0100
> 
> New in v2:
> 
>   - Go through all ports and override wchar_t size/sign appropriately.

I think go32 needs the same treatment as Windows, since its system
headers say

  typedef unsigned short wchar_t;

Thanks.

Re: [PATCH v2 2/2] Teach GDB that wchar_t is a built-in type in C++ mode

[Pedro Alves]

On 03/31/2017 01:28 PM, Eli Zaretskii wrote:
>> From: Pedro Alves <palves@redhat.com>
>> Date: Fri, 31 Mar 2017 12:03:15 +0100
>>
>> New in v2:
>>
>>   - Go through all ports and override wchar_t size/sign appropriately.
> 
> I think go32 needs the same treatment as Windows, since its system
> headers say
> 
>   typedef unsigned short wchar_t;
> 

Ah, yes, missed that one.  Consider this added to
i386-tdep.c:i386_go32_init_abi:

  set_gdbarch_wchar_bit (gdbarch, 16);
  set_gdbarch_wchar_signed (gdbarch, 0);

Thanks,
Pedro Alves

Re: [PATCH v2 2/2] Teach GDB that wchar_t is a built-in type in C++ mode

[Eli Zaretskii]

> Cc: gdb-patches@sourceware.org
> From: Pedro Alves <palves@redhat.com>
> Date: Fri, 31 Mar 2017 13:38:44 +0100
> 
> > I think go32 needs the same treatment as Windows, since its system
> > headers say
> > 
> >   typedef unsigned short wchar_t;
> > 
> 
> Ah, yes, missed that one.  Consider this added to
> i386-tdep.c:i386_go32_init_abi:
> 
>   set_gdbarch_wchar_bit (gdbarch, 16);
>   set_gdbarch_wchar_signed (gdbarch, 0);

Thanks!

Re: [PATCH v2 0/2] PR c++/21323: char16_t/char32_t/wchar_t built-in C++ types

[Pedro Alves]

On 03/31/2017 12:03 PM, Pedro Alves wrote:
> This series fixes PR c++/21323, a bug report that shows that GDB
> thinks C++11's char16_t and char32_t built-in types are signed, while
> the standard clearly says they're unsigned.
> 
> While working on it, I noticed that GDB isn't aware that wchar_t is a
> built-in type in C++ either (and it was already the case in C++98).
> The second patch fixes that.
> 
> Tested on x86_64 Fedora 23.
> 
> New in v2:
> 
>  - Make the tests a bit more lax when testing C/C++98, WRT to the
>    underlying type of char16_t/char32_t - accept any of unsigned
>    long/short/int.
> 
>  - Go through all ports and override wchar_t size/sign appropriately.
> 
> Pedro Alves (2):
>   Fix PR c++/21323: GDB thinks char16_t and char32_t are signed in C++
>   Teach GDB that wchar_t is a built-in type in C++ mode

I've pushed this in now.

Thanks,
Pedro Alves