From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 91200 invoked by alias); 2 Apr 2015 17:34:29 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Received: (qmail 91191 invoked by uid 89); 2 Apr 2015 17:34:28 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-0.0 required=5.0 tests=AWL,BAYES_20,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE,SPF_PASS autolearn=ham version=3.3.2 X-HELO: DUB004-OMC2S4.hotmail.com Received: from dub004-omc2s4.hotmail.com (HELO DUB004-OMC2S4.hotmail.com) (157.55.1.143) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-SHA256 encrypted) ESMTPS; Thu, 02 Apr 2015 17:34:26 +0000 Received: from DUB118-W47 ([157.55.1.138]) by DUB004-OMC2S4.hotmail.com over TLS secured channel with Microsoft SMTPSVC(7.5.7601.22751); Thu, 2 Apr 2015 10:34:23 -0700 X-TMN: [cbFBysjZN4kbWPv4zOFvjQTsdJlczmQe] Message-ID: From: Bernd Edlinger To: Pedro Alves , Eli Zaretskii CC: "gdb-patches@sourceware.org" , "dje@google.com" , "vapier@gentoo.org" Subject: RE: [PATCH] Enable building GDB without installed libtermcap Date: Thu, 02 Apr 2015 17:34:00 -0000 In-Reply-To: <54EF70A1.9000004@redhat.com> References: <54EB4FDF.1060909@redhat.com> <54EB513A.8050706@redhat.com> <54EF5C56.9010101@redhat.com> <83bnkg5z9q.fsf@gnu.org> <54EF69A5.3020902@redhat.com> <83a9005wq5.fsf@gnu.org>,<54EF70A1.9000004@redhat.com> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-SW-Source: 2015-04/txt/msg00111.txt.bz2 Hi, ping... patch looks good for me. On Thu, 26 Feb 2015 19:14:41, Pedro Alves wrote: > > On 02/26/2015 06:55 PM, Eli Zaretskii wrote: >>> From: Pedro Alves >>> Given that this stub file never needed these variables while it was >>> Windows-only, how about we simply not define the variables if >>> compiling for mingw/cygwin, but define them as weak everywhere else? >> >> Works for me, thanks. > > Alright. I recalled that Cygwin doesn't really care about old gcc's, > so I filed it under not-a-problem and limited the #ifdefery to mingw. > > I'd like to hear more opinions on this (whole thing) before > pushing ahead. > > From b99d25c7fc2b10d93b51743f44c08558bf06dfd6 Mon Sep 17 00:00:00 2001 > From: Pedro Alves > Date: Thu, 26 Feb 2015 17:01:06 +0000 > Subject: [PATCH] Fallback to stub-termcap.c on all hosts > > Currently building gdb is impossible without an installed termcap or > curses library. But, GDB already has a very minimal termcap in the > tree to handle this situation for Windows -- gdb/stub-termcap.c. This > patch makes that the fallback for all hosts. > > Testing this on GNU/Linux (by simply hacking away the termcap/curses > detection in gdb/configure.ac), I tripped on: > > ../readline/libreadline.a(terminal.o): In function `_rl_init_terminal_io': > /home/pedro/gdb/mygit/src/readline/terminal.c:527: undefined reference to= `PC' > /home/pedro/gdb/mygit/src/readline/terminal.c:528: undefined reference to= `BC' > /home/pedro/gdb/mygit/src/readline/terminal.c:529: undefined reference to= `UP' > /home/pedro/gdb/mygit/src/readline/terminal.c:538: undefined reference to= `PC' > /home/pedro/gdb/mygit/src/readline/terminal.c:539: undefined reference to= `BC' > /home/pedro/gdb/mygit/src/readline/terminal.c:540: undefined reference to= `UP' > > These are globals that are normally defined by termcap (or ncurses' > termcap emulation). > > Now, we could just define replacements in stub-termcap.c, but > readline/terminal.c (at least the copy in our tree) has this "beauty": > > #if !defined (__linux__) && !defined (NCURSES_VERSION) > # if defined (__EMX__) || defined (NEED_EXTERN_PC) > extern > # endif /* __EMX__ || NEED_EXTERN_PC */ > char PC, *BC, *UP; > #endif /* !__linux__ && !NCURSES_VERSION */ > > which can result in readline defining the globals too. That will > usually work out in C, given that "-fcommon" is usually the default > for C compilers, but that won't work for C++, or C with -fno-common > (link fails with "multiple definition" errors)... > > Mirroring those #ifdef conditions in the stub termcap screams > "brittle" to me -- I can see them changing in latter readline > versions. > > My idea to work around that is to simply use __attribute__((weak)). > Of all supported hosts > (https://sourceware.org/gdb/wiki/Systems#Supported_Hosts), except Windows. > > Windows/PE/COFF's do support weak, but not on gcc 3.4 based toolchains > (4.8.x does work). Given the file never needed the variables while it > was Windows-only, just continue not defining them there. > > gdb/ChangeLog: > 2015-02-26 Bernd Edlinger > Pedro Alves > > * configure.ac: Remove the mingw32-specific stub-termcap.o > fallback, and instead fallback to the stub termcap on all hosts. > * configure: Regenerate. > * stub-termcap.c (PC, BC, UP): Define as weak symbols. > --- > gdb/configure.ac | 7 +------ > gdb/stub-termcap.c | 21 ++++++++++++++++++++- > 2 files changed, 21 insertions(+), 7 deletions(-) > > diff --git a/gdb/configure.ac b/gdb/configure.ac > index 6ac8adb..79fc115 100644 > --- a/gdb/configure.ac > +++ b/gdb/configure.ac > @@ -610,18 +610,13 @@ case $host_os in > go32* | *djgpp*) > ac_cv_search_tgetent=3D"none required" > ;; > - *mingw32*) > - if test x"$curses_found" !=3D xyes; then > - ac_cv_search_tgetent=3D"none required" > - CONFIG_OBS=3D"$CONFIG_OBS stub-termcap.o" > - fi ;; > esac > > # These are the libraries checked by Readline. > AC_SEARCH_LIBS(tgetent, [termcap tinfo curses ncurses]) > > if test "$ac_cv_search_tgetent" =3D no; then > - AC_MSG_ERROR([no termcap library found]) > + CONFIG_OBS=3D"$CONFIG_OBS stub-termcap.o" > fi > > AC_ARG_WITH([system-readline], > diff --git a/gdb/stub-termcap.c b/gdb/stub-termcap.c > index cc8632c..df5f888 100644 > --- a/gdb/stub-termcap.c > +++ b/gdb/stub-termcap.c > @@ -32,9 +32,28 @@ extern char* tgetstr (char *name, char **area); > extern int tputs (char *string, int nlines, int (*outfun) ()); > extern char *tgoto (const char *cap, int col, int row); > > +/* These globals below are global termcap variables that readline > + references. > + > + Actually, depending on preprocessor conditions that we don't want > + to mirror here (as they may change depending on readline versions), > + readline may define these globals as well, relying on the linker > + merging them if needed (-fcommon). That doesn't work with > + -fno-common or C++, so instead we define the symbols as weak. > + Don't do this on Windows though, as MinGW gcc 3.4.2 doesn't support > + weak (later versions, e.g., 4.8, do support it). Given this stub > + file originally was Windows only, and we only needed this when we > + made it work on other hosts, it should be OK. */ > +#ifndef __MINGW32__ > +char PC __attribute__((weak)); > +char *BC __attribute__((weak)); > +char *UP __attribute__((weak)); > +#endif > + > /* Each of the files below is a minimal implementation of the standard > termcap function with the same name, suitable for use in a Windows > - console window. */ > + console window, or when a real termcap/curses library isn't > + available. */ > > int > tgetent (char *buffer, char *termtype) > -- > 1.9.3 > > =20=09=09=20=09=20=20=20=09=09=20=20 >From gdb-patches-return-121524-listarch-gdb-patches=sources.redhat.com@sourceware.org Fri Apr 03 00:39:17 2015 Return-Path: Delivered-To: listarch-gdb-patches@sources.redhat.com Received: (qmail 23070 invoked by alias); 3 Apr 2015 00:39:17 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Delivered-To: mailing list gdb-patches@sourceware.org Received: (qmail 23045 invoked by uid 89); 3 Apr 2015 00:39:16 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.8 required=5.0 tests=AWL,BAYES_00,SPF_PASS,T_RP_MATCHES_RCVD autolearn=ham version=3.3.2 X-HELO: smtp.gentoo.org Received: from smtp.gentoo.org (HELO smtp.gentoo.org) (140.211.166.183) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-GCM-SHA384 encrypted) ESMTPS; Fri, 03 Apr 2015 00:39:15 +0000 Received: from vapier (localhost [127.0.0.1]) by smtp.gentoo.org (Postfix) with SMTP id 2D09B340823; Fri, 3 Apr 2015 00:39:13 +0000 (UTC) Date: Fri, 03 Apr 2015 00:39:00 -0000 From: Mike Frysinger To: Hans-Peter Nilsson Cc: gdb-patches@sourceware.org Subject: Re: [PATCH] sim: d10v: link in missing testsuite Message-ID: <20150403003913.GF22171@vapier> Mail-Followup-To: Hans-Peter Nilsson , gdb-patches@sourceware.org References: <1427692153-12656-1-git-send-email-vapier@gentoo.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="0rSojgWGcpz+ezC3" Content-Disposition: inline In-Reply-To: X-IsSubscribed: yes X-SW-Source: 2015-04/txt/msg00112.txt.bz2 --0rSojgWGcpz+ezC3 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Content-length: 1506 On 02 Apr 2015 12:53, Hans-Peter Nilsson wrote: > On Mon, 30 Mar 2015, Mike Frysinger wrote: > > Looks like historical restructuring in this dir lost the d10v-elf subdir > > and no one noticed in the meantime. >=20 > Wow... And no port maintainer to frown at. bisecting down, it was a general refactor that broke it ... 10 years ago ;) > > Re-add it to the testsuite. > > > > There are some failures, but better some tests get run than none at all. >=20 > Meh... I agree it should be enabled, but as this particular > harness doesn't summarize the results, the value of a passing > subset is much lessened... it helped me verify my nrun refactor didn't screw things up. i was missing= some=20 parts which would lead to a segfault early on. > (I have to disable d10v sim testing -i.e. just building it- > in my sim autotester, which is all-must-pass-or-nothing anyway.) see the patch i posted & pushed > Is the attached sim d10v-elf testsuite run log consistent with > your results? A "grep -c Error log" yields 25, FWIW. all the *-elf subdirs of sim/testsuite/ are like this and suck horribly. i= =20 think i'll pull all the sim/ subdirs out into the higher level and rewrite = these=20 to use the same dejagnu framework. obnoxiously, this crappy framework supports parallel test running while dej= agnu=20 does not. i'm not sure if there's any way to make the dejagnu do parallel = ...=20 it's really really painful for some targets like frv & bfin which have hund= reds=20 of tests. -mike --0rSojgWGcpz+ezC3 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-length: 819 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJVHeExAAoJEEFjO5/oN/WBCecP/37kO6tE/ffOSJ9Pf1vfrJlp IBmUJdbDWty8C6mOv64PQVOS5FWw8fg5DVf6umHbvNu2dO9Xpn4KlE5UqyCwNbvN X/qGGz7sWVJJ3e2MYW83qJQxfLDUK5lYFn21TEpU/meANNyfZU8XWlNQ/0vGStkc tgfT4JzL33r7T2qCTzWzQWRM+EDjArZGRNAuXQZ4cepvkd1a7wOXg/60yQ6Sej6K skxECQXh1UmMKzVzRlEl5aQxJFM3HI8SduyIakjrffhsR7Wl1vwSMrSKWLgnQ2oE l/oYBLGXaZ9q/ULDeeBWvdAhvEAd9MgwkilWNBG/EjbSJa7zYpnt3+UADWWrfJml onBtl4UI0BRYly8ulsaMZDJGRZZ/SbzG7maSU4FNbi1+a9fRmEYMRQAXVZDCYfVf 6PIUYOb/u8K81SIMzxuvaqJrNxwRgYxOb25PZE9CuCZKlvoheR2lgI0LUGimRZWu eO3HZqYhfUpsm/h3fVejDFE+m70lAcLd1g3ufg4UyhamPyHRz2RpfxIlzU5j6r3r 8oWBEo/DsU+GlMMe78y2SJ+N3JNDloKMR/mKa6ZbFTKN/2m9KbIKxoFr1Wt+mLnA fF5SnugjQMFEDpJUHpoNcZ7ou6FPx5MTrxPigr4r07VsZjsQO0HXRt1LzUHRvnuO ZrdpUqnkVuZrhr3gLwE9 =CDyt -----END PGP SIGNATURE----- --0rSojgWGcpz+ezC3--