From: Tom Tromey <tromey@redhat.com>
To: kazu@codesourcery.com (Kazu Hirata)
Cc: gdb-patches@sourceware.org
Subject: Re: Your INTERMEDIATE_ENCODING patch for Solaris
Date: Tue, 17 Aug 2010 18:06:00 -0000 [thread overview]
Message-ID: <m3eidxi069.fsf@fleche.redhat.com> (raw)
In-Reply-To: <m37hk4x5r0.fsf@fleche.redhat.com> (Tom Tromey's message of "Thu, 05 Aug 2010 12:40:35 -0600")
>>>>> "Tom" == Tom Tromey <tromey@redhat.com> writes:
Kazu> As Andrew replied to your message above, __STDC_ISO_10646__ is not
Kazu> defined on Solaris. Could we use __sun__ instead?
Tom> I am not a Solaris expert, but some digging through the online
Tom> OpenSolaris libc sources a few months ago convinced me that the Solaris
Tom> wchar_t is in fact not UCS-4.
Tom> If that is true, then this patch would be incorrect.
While looking at this more today, I found:
http://defect.opensolaris.org/bz/show_bug.cgi?id=11076
Some of the comments make it clear that wchar_t != UCS-4 for Solaris.
Tom> This particular problem has come up multiple times. I am not completely
Tom> sure what to do about it, but my current inclination is to change gdb so
Tom> that most hosts fall back to PHONY_ICONV, with exceptions for libiconv
Tom> and for Linux. This will at least have gdb default to working, and is
Tom> also attractive because I have some hope of actually testing it.
Daniel> Sounds good to me.
Here is what I am testing. If someone can try it out on Solaris, I
would appreciate that.
Tom
b/gdb/ChangeLog:
2010-08-17 Tom Tromey <tromey@redhat.com>
* gdb_wchar.h: Check HAVE_LIBICONV, __STDC_ISO_10646__. Change
how INTERMEDIATE_ENCODING is defined.
* configure, config.in: Rebuild.
* acinclude.m4 (AM_ICONV): Define HAVE_LIBICONV.
diff --git a/gdb/acinclude.m4 b/gdb/acinclude.m4
index 1942ef4..082a9bc 100644
--- a/gdb/acinclude.m4
+++ b/gdb/acinclude.m4
@@ -256,6 +256,7 @@ AC_DEFUN([AM_ICONV],
LIBICONV=
if test "$am_cv_lib_iconv" = yes; then
LIBICONV="-liconv"
+ AC_DEFINE(HAVE_LIBICONV, 1, [Define if using libiconv.])
else
LIBICONV_LIBDIR=
LIBICONV_INCLUDE=
diff --git a/gdb/config.in b/gdb/config.in
index 7659181..35272a2 100644
--- a/gdb/config.in
+++ b/gdb/config.in
@@ -214,6 +214,9 @@
/* Define if you have the expat library. */
#undef HAVE_LIBEXPAT
+/* Define if using libiconv. */
+#undef HAVE_LIBICONV
+
/* Define to 1 if you have the `libiconvlist' function. */
#undef HAVE_LIBICONVLIST
diff --git a/gdb/configure b/gdb/configure
index 485c904..852c5c1 100755
--- a/gdb/configure
+++ b/gdb/configure
@@ -9450,6 +9450,9 @@ $as_echo "$am_cv_func_iconv" >&6; }
LIBICONV=
if test "$am_cv_lib_iconv" = yes; then
LIBICONV="-liconv"
+
+$as_echo "#define HAVE_LIBICONV 1" >>confdefs.h
+
else
LIBICONV_LIBDIR=
LIBICONV_INCLUDE=
diff --git a/gdb/gdb_wchar.h b/gdb/gdb_wchar.h
index fca3fe4..1ff7a1c 100644
--- a/gdb/gdb_wchar.h
+++ b/gdb/gdb_wchar.h
@@ -23,7 +23,11 @@
Capable systems have the full suite: wchar_t support and iconv
(perhaps via GNU libiconv). On these machines, full functionality
- is available.
+ is available. Note that full functionality is dependent on us
+ being able to convert from an arbitrary encoding to wchar_t. In
+ practice this means we look for __STDC_ISO_10646__ (where we know
+ the name of the wchar_t encoding) or GNU libiconv, where we can use
+ "wchar_t".
DJGPP is known to have libiconv but not wchar_t support. On
systems like this, we use the narrow character functions. The full
@@ -35,8 +39,6 @@
wrappers for the wchar_t functionality we use. */
-#define INTERMEDIATE_ENCODING "wchar_t"
-
#if defined (HAVE_ICONV)
#include <iconv.h>
#else
@@ -45,9 +47,11 @@
#define PHONY_ICONV
#endif
-/* We use "btowc" as a sentinel to detect functioning wchar_t
- support. */
-#if defined (HAVE_ICONV) && defined (HAVE_WCHAR_H) && defined (HAVE_BTOWC)
+/* We use "btowc" as a sentinel to detect functioning wchar_t support.
+ We check for either __STDC_ISO_10646__ or libiconv in order to
+ ensure we can convert to and from wchar_t. */
+#if defined (HAVE_ICONV) && defined (HAVE_WCHAR_H) && defined (HAVE_BTOWC) \
+ && (defined (__STDC_ISO_10646__) || defined (HAVE_LIBICONV))
#include <wchar.h>
#include <wctype.h>
@@ -63,6 +67,25 @@ typedef wint_t gdb_wint_t;
#define LCST(X) L ## X
+/* If __STDC_ISO_10646__ is defined, then the host wchar_t is UCS-4.
+ We exploit this fact in the hope that there are hosts that define
+ this but which do not support "wchar_t" as an encoding argument to
+ iconv_open. We put the endianness into the encoding name to avoid
+ hosts that emit a BOM when the unadorned name is used. */
+#if defined (__STDC_ISO_10646__)
+#if WORDS_BIGENDIAN
+#define INTERMEDIATE_ENCODING "UCS-4BE"
+#else
+#define INTERMEDIATE_ENCODING "UCS-4LE"
+#endif
+#elif defined (HAVE_LIBICONV)
+#define INTERMEDIATE_ENCODING "wchar_t"
+#else
+/* This shouldn't happen, because the earlier #if should have filtered
+ out this case. */
+#error "Neither __STDC_ISO_10646__ nor HAVE_LIBICONV defined"
+#endif
+
#else
typedef char gdb_wchar_t;
next prev parent reply other threads:[~2010-08-17 18:06 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-31 16:25 Kazu Hirata
2010-08-05 18:40 ` Tom Tromey
2010-08-10 20:28 ` Daniel Jacobowitz
2010-08-17 18:06 ` Tom Tromey [this message]
2010-08-17 18:44 ` Joel Brobecker
2010-08-17 19:03 ` Tom Tromey
2010-08-18 10:14 ` Joel Brobecker
2010-08-18 14:43 ` Pierre Muller
2010-08-18 14:52 ` Joel Brobecker
2010-08-18 15:10 ` Pierre Muller
2010-08-18 15:35 ` Tom Tromey
2010-08-18 15:44 ` Tom Tromey
2010-08-18 16:36 ` Joel Brobecker
2010-08-18 17:35 ` Tom Tromey
2010-08-18 17:41 ` Joel Brobecker
[not found] ` <15264.6257346079$1282142643@news.gmane.org>
2010-08-18 16:12 ` Tom Tromey
2010-08-19 15:03 ` Pierre Muller
2010-08-19 16:09 ` Tom Tromey
2010-08-30 18:01 ` Tom Tromey
2010-08-31 9:25 ` Pierre Muller
2010-08-31 16:47 ` Tom Tromey
2010-09-01 7:30 ` Pierre Muller
[not found] ` <44796.6229789474$1283326243@news.gmane.org>
2010-09-01 22:35 ` Tom Tromey
2010-09-02 14:21 ` Pierre Muller
2010-09-02 15:39 ` Tom Tromey
2010-09-14 21:29 ` Tom Tromey
2010-09-15 17:20 ` Pierre Muller
2010-09-16 7:55 ` Tom Tromey
2010-09-16 9:49 ` Pierre Muller
2010-09-16 20:22 ` Tom Tromey
2010-09-16 22:31 ` Pierre Muller
[not found] ` <20078.2261243605$1284672670@news.gmane.org>
2010-09-17 9:21 ` Tom Tromey
2010-09-17 13:48 ` Pierre Muller
2010-09-23 13:00 ` Tom Tromey
2010-09-23 14:48 ` Pierre Muller
2010-09-27 18:42 ` Tom Tromey
2010-09-16 17:51 ` [patch] Regression on py-prettyprint.exp: print estring [Re: Your INTERMEDIATE_ENCODING patch for Solaris] Jan Kratochvil
2010-09-16 20:33 ` Tom Tromey
2010-09-16 20:44 ` Jan Kratochvil
2010-08-18 11:52 ` Your INTERMEDIATE_ENCODING patch for Solaris Pierre Muller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=m3eidxi069.fsf@fleche.redhat.com \
--to=tromey@redhat.com \
--cc=gdb-patches@sourceware.org \
--cc=kazu@codesourcery.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox