From owner-freebsd-toolchain@freebsd.org Wed Jun 8 19:11:50 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id F1D39B6FD04 for ; Wed, 8 Jun 2016 19:11:50 +0000 (UTC) (envelope-from gerald@pfeifer.com) Received: from ainaz.pair.com (ainaz.pair.com [209.68.2.66]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 4C1BF1E67; Wed, 8 Jun 2016 19:11:50 +0000 (UTC) (envelope-from gerald@pfeifer.com) Received: from anthias.dhcp.nue.suse.com (nat.nue.novell.com [195.135.221.2]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ainaz.pair.com (Postfix) with ESMTPSA id 52F5A3F479; Wed, 8 Jun 2016 15:11:43 -0400 (EDT) Date: Wed, 8 Jun 2016 21:11:41 +0200 (CEST) From: Gerald Pfeifer To: Andreas Tobler , freebsd-toolchain@freebsd.org Subject: Duplicate OPT_ entries in gcc/options.h Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 19:11:51 -0000 I got a user report, and could reproduce this, that building GCC (lang/gcc, but also current HEAD, so probably pretty much any version) with FreeBSD 11 and LANG = en_US.UTF-8 we get conflicting entires in $BUILDDIR/gcc/options.h such as OPT_d = 135, /* -d */ OPT_D = 136, /* -D */ OPT_d = 137, /* -d */ OPT_D = 138, /* -D */ OPT_d = 141, /* -d */ OPT_D = 142, /* -D */ OPT_d = 143, /* -d */ Using LANG = en_US (without UTF-8), everything works fine. Any ideas what might be going on here? (This is done via AWK scripts from what I can tell, does this trigger any ideas?) Gerald From owner-freebsd-toolchain@freebsd.org Wed Jun 8 21:15:48 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 2AB97B700AC for ; Wed, 8 Jun 2016 21:15:48 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from tensor.andric.com (tensor.andric.com [IPv6:2001:7b8:3a7:1:2d0:b7ff:fea0:8c26]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "tensor.andric.com", Issuer "COMODO RSA Domain Validation Secure Server CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id B3D411F88; Wed, 8 Jun 2016 21:15:47 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from [IPv6:2001:7b8:3a7::35e6:4d9:66a5:188a] (unknown [IPv6:2001:7b8:3a7:0:35e6:4d9:66a5:188a]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by tensor.andric.com (Postfix) with ESMTPSA id 741D61771C; Wed, 8 Jun 2016 23:15:44 +0200 (CEST) Subject: Re: Duplicate OPT_ entries in gcc/options.h Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) Content-Type: multipart/signed; boundary="Apple-Mail=_EE128934-3111-4BA1-80A9-472C0E2D9E01"; protocol="application/pgp-signature"; micalg=pgp-sha1 X-Pgp-Agent: GPGMail 2.6b2 (ebbf3ef) From: Dimitry Andric In-Reply-To: Date: Wed, 8 Jun 2016 23:15:39 +0200 Cc: Andreas Tobler , freebsd-toolchain@freebsd.org Message-Id: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> References: To: Gerald Pfeifer X-Mailer: Apple Mail (2.3124) X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 21:15:48 -0000 --Apple-Mail=_EE128934-3111-4BA1-80A9-472C0E2D9E01 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote: >=20 > I got a user report, and could reproduce this, that building > GCC (lang/gcc, but also current HEAD, so probably pretty much > any version) with FreeBSD 11 and LANG =3D en_US.UTF-8 we get > conflicting entires in $BUILDDIR/gcc/options.h such as >=20 > OPT_d =3D 135, /* -d */ > OPT_D =3D 136, /* -D */ > OPT_d =3D 137, /* -d */ > OPT_D =3D 138, /* -D */ > OPT_d =3D 141, /* -d */ > OPT_D =3D 142, /* -D */ > OPT_d =3D 143, /* -d */ >=20 > Using LANG =3D en_US (without UTF-8), everything works fine. >=20 > Any ideas what might be going on here? (This is done via > AWK scripts from what I can tell, does this trigger any > ideas?) It is definitely something caused by our awk in base, in any case. First opt-gather.awk is run to generate a flat list of all options: /usr/bin/awk -f /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opt-gather.awk = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/ada/gcc-interface/lang.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/fortran/lang.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/go/lang.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/java/lang.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/lto/lang.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/c-family/c.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/common.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/config/fused-madd.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/config/i386/i386.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/config/rpath.opt = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/config/freebsd.opt > = tmp-optionlist Then opt-functions.awk is run to process optionlist into options.h: /usr/bin/awk -f = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opt-functions.awk -f = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opt-read.awk -f = /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opth-gen.awk < optionlist > = options.h If I run the first step using LANG=3DC, or without any LANG setting, = both optionlist and options.h are as expected. If I run the first step using LANG=3Den_US.UTF-8, the optionlist is sorted differently, for example = the "good" optionlist has the uppercase d options first, and much later the lowercase d options: D^\C ObjC C++ ObjC++ Joined Separate MissingArgError(macro name = missing after %qs)^\-D[=3D] Define a with as = its value. If just is given, is taken to be 1 D^\Driver Joined Separate D^\Fortran Joined Separate ... much later in the file, after all options starting with an = uppercase letter ... d^\C ObjC C++ ObjC++ Joined d^\Common Joined^\-d Enable dumps from specific passes of = the compiler d^\Fortran Joined d^\Java Separate SeparateAlias Alias(foutput-class-dir=3D) The "bad" optionlist has the upper and lower case d options sorted together: d^\C ObjC C++ ObjC++ Joined D^\C ObjC C++ ObjC++ Joined Separate MissingArgError(macro name = missing after %qs)^\-D[=3D] Define a with as = its value. If just is given, is taken to be 1 d^\Common Joined^\-d Enable dumps from specific passes of = the compiler D^\Driver Joined Separate defsym=3D^\Driver JoinedOrMissing defsym^\Driver Separate d^\Fortran Joined D^\Fortran Joined Separate d^\Java Separate SeparateAlias Alias(foutput-class-dir=3D) Note that GNU awk does *not* produce a different optionlist file when used with either LANG=3DC or LANG=3Den_US.UTF-8. opt-gather.awk's sorting function looks like this: function sort(ARRAY, ELEMENTS) { for (i =3D 2; i <=3D ELEMENTS; ++i) { for (j =3D i; ARRAY[j-1] > ARRAY[j]; --j) { temp =3D ARRAY[j] ARRAY[j] =3D ARRAY[j-1] ARRAY[j-1] =3D temp } } return } So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works differently in our awk, depending on the LANG settings. No idea when that changed, though, if it changed at all... -Dimitry --Apple-Mail=_EE128934-3111-4BA1-80A9-472C0E2D9E01 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.30 iEYEARECAAYFAldYiwAACgkQsF6jCi4glqNEfwCgjyaa7pD1dwOBJSpksK0JlnHN NGEAnj0pycWr+f2DQBHEa3X5Ro5letdy =GDFJ -----END PGP SIGNATURE----- --Apple-Mail=_EE128934-3111-4BA1-80A9-472C0E2D9E01-- From owner-freebsd-toolchain@freebsd.org Wed Jun 8 21:51:32 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 3E8F2B70948 for ; Wed, 8 Jun 2016 21:51:32 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from tensor.andric.com (tensor.andric.com [87.251.56.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "tensor.andric.com", Issuer "COMODO RSA Domain Validation Secure Server CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id AAC6B16C4; Wed, 8 Jun 2016 21:51:31 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from [IPv6:2001:7b8:3a7::35e6:4d9:66a5:188a] (unknown [IPv6:2001:7b8:3a7:0:35e6:4d9:66a5:188a]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by tensor.andric.com (Postfix) with ESMTPSA id 0181C1779A; Wed, 8 Jun 2016 23:51:22 +0200 (CEST) Subject: Re: Duplicate OPT_ entries in gcc/options.h Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) Content-Type: multipart/signed; boundary="Apple-Mail=_433C8A11-E488-4AF6-AD2A-5E94191B9E70"; protocol="application/pgp-signature"; micalg=pgp-sha1 X-Pgp-Agent: GPGMail 2.6b2 (ebbf3ef) From: Dimitry Andric In-Reply-To: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> Date: Wed, 8 Jun 2016 23:51:15 +0200 Cc: freebsd-toolchain@freebsd.org, Andreas Tobler Message-Id: <9AC91278-90E5-4C74-9FF1-334D16083102@FreeBSD.org> References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> To: Gerald Pfeifer X-Mailer: Apple Mail (2.3124) X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 21:51:32 -0000 --Apple-Mail=_433C8A11-E488-4AF6-AD2A-5E94191B9E70 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii On 08 Jun 2016, at 23:15, Dimitry Andric wrote: > > On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote: ... > Note that GNU awk does *not* produce a different optionlist file when > used with either LANG=C or LANG=en_US.UTF-8. And that phenomenon is explained here: http://www.gnu.org/software/gawk/manual/gawk.html#POSIX-String-Comparison "6.3.2.3 String Comparison with POSIX Rules The POSIX standard says that string comparison is performed based on the locale's collating order. This is the order in which characters sort, as defined by the locale (for more discussion, see Locales). This order is usually very different from the results obtained when doing straight character-by-character comparison.34 Because this behavior differs considerably from existing practice, gawk only implements it when in POSIX mode (see Options)." -Dimitry --Apple-Mail=_433C8A11-E488-4AF6-AD2A-5E94191B9E70 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.30 iEYEARECAAYFAldYk1oACgkQsF6jCi4glqPBXwCeNgPg5b/Rrn8PmpA5Z4XtLMye 0pwAoOc05pY366jfeteql4DO+Zf7v/Ah =EnZ1 -----END PGP SIGNATURE----- --Apple-Mail=_433C8A11-E488-4AF6-AD2A-5E94191B9E70-- From owner-freebsd-toolchain@freebsd.org Wed Jun 8 21:54:28 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 0F72EB70982 for ; Wed, 8 Jun 2016 21:54:28 +0000 (UTC) (envelope-from jkim@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:1900:2254:206c::16:87]) by mx1.freebsd.org (Postfix) with ESMTP id 748FE183F; Wed, 8 Jun 2016 21:54:27 +0000 (UTC) (envelope-from jkim@FreeBSD.org) Subject: Re: Re: Duplicate OPT_ entries in gcc/options.h To: Dimitry Andric , Gerald Pfeifer References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> Cc: Andreas Tobler , freebsd-toolchain@freebsd.org From: Jung-uk Kim Message-ID: <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> Date: Wed, 8 Jun 2016 17:54:20 -0400 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:45.0) Gecko/20100101 Thunderbird/45.1.1 MIME-Version: 1.0 In-Reply-To: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="qg1A47UvPVkdvm1uIQ6tBqxxCAWAmrR2b" X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 21:54:28 -0000 This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --qg1A47UvPVkdvm1uIQ6tBqxxCAWAmrR2b Content-Type: multipart/mixed; boundary="m4Ud1lAVu3MP132uI1fkfpRv0isPT4uke" From: Jung-uk Kim To: Dimitry Andric , Gerald Pfeifer Cc: Andreas Tobler , freebsd-toolchain@freebsd.org Message-ID: <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> Subject: Re: Re: Duplicate OPT_ entries in gcc/options.h References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> In-Reply-To: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> --m4Ud1lAVu3MP132uI1fkfpRv0isPT4uke Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 06/ 8/16 05:15 PM, Dimitry Andric wrote: > On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote: >> >> I got a user report, and could reproduce this, that building >> GCC (lang/gcc, but also current HEAD, so probably pretty much >> any version) with FreeBSD 11 and LANG =3D en_US.UTF-8 we get >> conflicting entires in $BUILDDIR/gcc/options.h such as >> >> OPT_d =3D 135, /* -d */ >> OPT_D =3D 136, /* -D */ >> OPT_d =3D 137, /* -d */ >> OPT_D =3D 138, /* -D */ >> OPT_d =3D 141, /* -d */ >> OPT_D =3D 142, /* -D */ >> OPT_d =3D 143, /* -d */ >> >> Using LANG =3D en_US (without UTF-8), everything works fine. >> >> Any ideas what might be going on here? (This is done via >> AWK scripts from what I can tell, does this trigger any >> ideas?) >=20 > It is definitely something caused by our awk in base, in any case. > First opt-gather.awk is run to generate a flat list of all options: >=20 > /usr/bin/awk -f /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opt-gather.awk= /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/ada/gcc-interface/lang.opt /usr/p= orts/lang/gcc/work/gcc-4.8.5/gcc/fortran/lang.opt /usr/ports/lang/gcc/wor= k/gcc-4.8.5/gcc/go/lang.opt /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/java/l= ang.opt /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/lto/lang.opt /usr/ports/la= ng/gcc/work/gcc-4.8.5/gcc/c-family/c.opt /usr/ports/lang/gcc/work/gcc-4.8= =2E5/gcc/common.opt /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/config/fused-m= add.opt /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/config/i386/i386.opt /usr/= ports/lang/gcc/work/gcc-4.8.5/gcc/config/rpath.opt /usr/ports/lang/gcc/wo= rk/gcc-4.8.5/gcc/config/freebsd.opt > tmp-optionlist >=20 > Then opt-functions.awk is run to process optionlist into options.h: >=20 > /usr/bin/awk -f /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opt-functions.= awk -f /usr/ports/lang/gcc/work/gcc-4.8.5/gcc/opt-read.awk -f /usr/ports/= lang/gcc/work/gcc-4.8.5/gcc/opth-gen.awk < optionlist > options.h >=20 > If I run the first step using LANG=3DC, or without any LANG setting, bo= th > optionlist and options.h are as expected. If I run the first step usin= g > LANG=3Den_US.UTF-8, the optionlist is sorted differently, for example t= he > "good" optionlist has the uppercase d options first, and much later the= > lowercase d options: >=20 > D^\C ObjC C++ ObjC++ Joined Separate MissingArgError(macro name missi= ng after %qs)^\-D[=3D] Define a with as its v= alue. If just is given, is taken to be 1 > D^\Driver Joined Separate > D^\Fortran Joined Separate > ... much later in the file, after all options starting with an upperc= ase letter ... > d^\C ObjC C++ ObjC++ Joined > d^\Common Joined^\-d Enable dumps from specific passes of = the compiler > d^\Fortran Joined > d^\Java Separate SeparateAlias Alias(foutput-class-dir=3D) >=20 > The "bad" optionlist has the upper and lower case d options sorted > together: >=20 > d^\C ObjC C++ ObjC++ Joined > D^\C ObjC C++ ObjC++ Joined Separate MissingArgError(macro name missi= ng after %qs)^\-D[=3D] Define a with as its v= alue. If just is given, is taken to be 1 > d^\Common Joined^\-d Enable dumps from specific passes of = the compiler > D^\Driver Joined Separate > defsym=3D^\Driver JoinedOrMissing > defsym^\Driver Separate > d^\Fortran Joined > D^\Fortran Joined Separate > d^\Java Separate SeparateAlias Alias(foutput-class-dir=3D) >=20 > Note that GNU awk does *not* produce a different optionlist file when > used with either LANG=3DC or LANG=3Den_US.UTF-8. >=20 > opt-gather.awk's sorting function looks like this: >=20 > function sort(ARRAY, ELEMENTS) > { > for (i =3D 2; i <=3D ELEMENTS; ++i) { > for (j =3D i; ARRAY[j-1] > ARRAY[j]; --j) { > temp =3D ARRAY[j] > ARRAY[j] =3D ARRAY[j-1] > ARRAY[j-1] =3D temp > } > } > return > } >=20 > So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works > differently in our awk, depending on the LANG settings. No idea when > that changed, though, if it changed at all... This behaviour is known for very long time: https://svnweb.freebsd.org/changeset/base/173731 and it is not our fault: https://www.gnu.org/software/gawk/manual/html_node/POSIX-String-Compariso= n.html GNU awk produces the same output with "--posix" option. FYI... Jung-uk Kim --m4Ud1lAVu3MP132uI1fkfpRv0isPT4uke-- --qg1A47UvPVkdvm1uIQ6tBqxxCAWAmrR2b Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJXWJQNAAoJEHyflib82/FGnqwIAIeXVDRcuKi3D1N4JRuPr4fx QzExtI/vGEwB3RN05eDI9R+1ME4bVCXJix5nATL3YlUohXC0wUcbCE92R2MJ/xMo LJ5sPsNT73nG2NMGBgyW4ffwnyTrbiyWDKj8lGJFW8extBsgw3E+OmdyKKn+afuF LAsj2qvcRqor5ChhNpTblwCx5PM4BkrYKI3zj3bveW8gfDM1i580xUx5jsWl7gTK mjxi4zc+zFEz2vU2Yhx82gm6leHy+1nU3DCR5ZkMWRpa88EHLeYrVEk5JxRkjYWz bpS2s14Oh8csT8enU4ex1W9+cUFVyeK5jTKlWUEGIr4Q68QH0nhDuQP4vwYcuKQ= =Kh5h -----END PGP SIGNATURE----- --qg1A47UvPVkdvm1uIQ6tBqxxCAWAmrR2b-- From owner-freebsd-toolchain@freebsd.org Wed Jun 8 22:16:41 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 53EF5B70D45 for ; Wed, 8 Jun 2016 22:16:41 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from tensor.andric.com (tensor.andric.com [IPv6:2001:7b8:3a7:1:2d0:b7ff:fea0:8c26]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "tensor.andric.com", Issuer "COMODO RSA Domain Validation Secure Server CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 1A8BE113B; Wed, 8 Jun 2016 22:16:41 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from [IPv6:2001:7b8:3a7::35e6:4d9:66a5:188a] (unknown [IPv6:2001:7b8:3a7:0:35e6:4d9:66a5:188a]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by tensor.andric.com (Postfix) with ESMTPSA id 05118177F9; Thu, 9 Jun 2016 00:16:38 +0200 (CEST) Subject: Re: Duplicate OPT_ entries in gcc/options.h Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) Content-Type: multipart/signed; boundary="Apple-Mail=_D0BD42B3-4C9C-49CC-8EA4-BC24F0E0EEA7"; protocol="application/pgp-signature"; micalg=pgp-sha1 X-Pgp-Agent: GPGMail 2.6b2 (ebbf3ef) From: Dimitry Andric In-Reply-To: <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> Date: Thu, 9 Jun 2016 00:16:31 +0200 Cc: Gerald Pfeifer , freebsd-toolchain@freebsd.org, Andreas Tobler Message-Id: <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> To: Jung-uk Kim X-Mailer: Apple Mail (2.3124) X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 22:16:41 -0000 --Apple-Mail=_D0BD42B3-4C9C-49CC-8EA4-BC24F0E0EEA7 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii On 08 Jun 2016, at 23:54, Jung-uk Kim wrote: >=20 > On 06/ 8/16 05:15 PM, Dimitry Andric wrote: >> On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote: >>>=20 >>> I got a user report, and could reproduce this, that building >>> GCC (lang/gcc, but also current HEAD, so probably pretty much >>> any version) with FreeBSD 11 and LANG =3D en_US.UTF-8 we get >>> conflicting entires in $BUILDDIR/gcc/options.h such as ... >> Note that GNU awk does *not* produce a different optionlist file when >> used with either LANG=3DC or LANG=3Den_US.UTF-8. ... >> So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works >> differently in our awk, depending on the LANG settings. No idea when >> that changed, though, if it changed at all... >=20 > This behaviour is known for very long time: >=20 > https://svnweb.freebsd.org/changeset/base/173731 >=20 > and it is not our fault: >=20 > = https://www.gnu.org/software/gawk/manual/html_node/POSIX-String-Comparison= .html Indeed, so the real question is: why does this only started coming up now, if it is known since 2007? I have been building gcc ports for ages, and never ran into this problem, but I also have never actively used a persistent LANG environment variable, let alone with UTF-8 in it. Is this because more people started using UTF-8 recently? -Dimitry --Apple-Mail=_D0BD42B3-4C9C-49CC-8EA4-BC24F0E0EEA7 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.30 iEYEARECAAYFAldYmUUACgkQsF6jCi4glqO8LgCffANHlFKTtOZX7oZ1EGWY5ZKr QeIAoKe6SX/dcZYepuqi/h6dfXkD3utO =61O0 -----END PGP SIGNATURE----- --Apple-Mail=_D0BD42B3-4C9C-49CC-8EA4-BC24F0E0EEA7-- From owner-freebsd-toolchain@freebsd.org Wed Jun 8 22:30:45 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id B8AFDB6F15B for ; Wed, 8 Jun 2016 22:30:45 +0000 (UTC) (envelope-from jkim@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:1900:2254:206c::16:87]) by mx1.freebsd.org (Postfix) with ESMTP id 46CCD1C7E; Wed, 8 Jun 2016 22:30:45 +0000 (UTC) (envelope-from jkim@FreeBSD.org) Subject: Re: Duplicate OPT_ entries in gcc/options.h To: Dimitry Andric References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> Cc: Gerald Pfeifer , freebsd-toolchain@freebsd.org, Andreas Tobler From: Jung-uk Kim Message-ID: Date: Wed, 8 Jun 2016 18:30:39 -0400 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:45.0) Gecko/20100101 Thunderbird/45.1.1 MIME-Version: 1.0 In-Reply-To: <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="uefrSTQEJuqUI6e1Q84xQuVtXWsUSkGOE" X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 22:30:45 -0000 This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --uefrSTQEJuqUI6e1Q84xQuVtXWsUSkGOE Content-Type: multipart/mixed; boundary="qUs9KkSxj5cJHiP0dcluVohHdXcerXOJE" From: Jung-uk Kim To: Dimitry Andric Cc: Gerald Pfeifer , freebsd-toolchain@freebsd.org, Andreas Tobler Message-ID: Subject: Re: Duplicate OPT_ entries in gcc/options.h References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> In-Reply-To: <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> --qUs9KkSxj5cJHiP0dcluVohHdXcerXOJE Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable On 06/ 8/16 06:16 PM, Dimitry Andric wrote: > On 08 Jun 2016, at 23:54, Jung-uk Kim wrote: >> >> On 06/ 8/16 05:15 PM, Dimitry Andric wrote: >>> On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote: >>>> >>>> I got a user report, and could reproduce this, that building >>>> GCC (lang/gcc, but also current HEAD, so probably pretty much >>>> any version) with FreeBSD 11 and LANG =3D en_US.UTF-8 we get >>>> conflicting entires in $BUILDDIR/gcc/options.h such as > ... >>> Note that GNU awk does *not* produce a different optionlist file when= >>> used with either LANG=3DC or LANG=3Den_US.UTF-8. > ... >>> So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works >>> differently in our awk, depending on the LANG settings. No idea when= >>> that changed, though, if it changed at all... >> >> This behaviour is known for very long time: >> >> https://svnweb.freebsd.org/changeset/base/173731 >> >> and it is not our fault: >> >> https://www.gnu.org/software/gawk/manual/html_node/POSIX-String-Compar= ison.html >=20 >=20 > Indeed, so the real question is: why does this only started coming up > now, if it is known since 2007? I have been building gcc ports for > ages, and never ran into this problem, but I also have never actively > used a persistent LANG environment variable, let alone with UTF-8 in it= =2E >=20 > Is this because more people started using UTF-8 recently? We are doing more correct collation now: https://svnweb.freebsd.org/changeset/base/290494 Jung-uk Kim --qUs9KkSxj5cJHiP0dcluVohHdXcerXOJE-- --uefrSTQEJuqUI6e1Q84xQuVtXWsUSkGOE Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJXWJyUAAoJEHyflib82/FGptYH/jTU93g5GYVxxtHcxTat17Eu 6IkTkuibVG41BiGFG/Nqb3hu/NaV41yqkF5XccPclutmSykMn89VgSquTbaWr4dq kb9wFVCb55mzeSxpHZgXXtjVF90gGXbyH8NQH10ZET9xTD6w8oMcel41VkOSjbgt 3YPzic4hR4e8pE+aKsqFMwswyImIQcXNbXPM3fUYckT0Ozn0pdc/CdZ44XCLzp9m rYKgHqW3UEkRyo8nx4uXpK/BX2UDr2ak94wQsHXNaH5RxZ4ejKAABwzxncy3qMjP mcdz5Ly5dTC9jjHCWU+Pb+lo+y5E0yDUHI4MpPAlAZPSSchA9hnZLT+lnnHRC2E= =2JUu -----END PGP SIGNATURE----- --uefrSTQEJuqUI6e1Q84xQuVtXWsUSkGOE-- From owner-freebsd-toolchain@freebsd.org Wed Jun 8 23:09:42 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 00AF8B6FEA0 for ; Wed, 8 Jun 2016 23:09:42 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from tensor.andric.com (tensor.andric.com [IPv6:2001:7b8:3a7:1:2d0:b7ff:fea0:8c26]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "tensor.andric.com", Issuer "COMODO RSA Domain Validation Secure Server CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id A2C261488; Wed, 8 Jun 2016 23:09:41 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from [IPv6:2001:7b8:3a7::35e6:4d9:66a5:188a] (unknown [IPv6:2001:7b8:3a7:0:35e6:4d9:66a5:188a]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by tensor.andric.com (Postfix) with ESMTPSA id 1161C178B6; Thu, 9 Jun 2016 01:09:40 +0200 (CEST) Subject: Re: Duplicate OPT_ entries in gcc/options.h Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) Content-Type: multipart/signed; boundary="Apple-Mail=_C4FD4980-68D6-46FD-B1D9-BCC64693E3CF"; protocol="application/pgp-signature"; micalg=pgp-sha1 X-Pgp-Agent: GPGMail 2.6b2 (ebbf3ef) From: Dimitry Andric In-Reply-To: Date: Thu, 9 Jun 2016 01:09:32 +0200 Cc: FreeBSD Toolchain , Gerald Pfeifer , Andreas Tobler , Baptiste Daroussin , Tijl Coosemans Message-Id: <823B7E6C-4B47-401F-A254-C88825AE5562@FreeBSD.org> References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> To: Jung-uk Kim X-Mailer: Apple Mail (2.3124) X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 23:09:42 -0000 --Apple-Mail=_C4FD4980-68D6-46FD-B1D9-BCC64693E3CF Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 On 09 Jun 2016, at 00:30, Jung-uk Kim wrote: >=20 > On 06/ 8/16 06:16 PM, Dimitry Andric wrote: >> On 08 Jun 2016, at 23:54, Jung-uk Kim wrote: >>>=20 >>> On 06/ 8/16 05:15 PM, Dimitry Andric wrote: >>>> On 08 Jun 2016, at 21:11, Gerald Pfeifer = wrote: >>>>>=20 >>>>> I got a user report, and could reproduce this, that building >>>>> GCC (lang/gcc, but also current HEAD, so probably pretty much >>>>> any version) with FreeBSD 11 and LANG =3D en_US.UTF-8 we get >>>>> conflicting entires in $BUILDDIR/gcc/options.h such as >> ... >>>> Note that GNU awk does *not* produce a different optionlist file = when >>>> used with either LANG=3DC or LANG=3Den_US.UTF-8. >> ... >>>> So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works >>>> differently in our awk, depending on the LANG settings. No idea = when >>>> that changed, though, if it changed at all... >>>=20 >>> This behaviour is known for very long time: >>>=20 >>> https://svnweb.freebsd.org/changeset/base/173731 >>>=20 >>> and it is not our fault: >>>=20 >>> = https://www.gnu.org/software/gawk/manual/html_node/POSIX-String-Comparison= .html >>=20 >>=20 >> Indeed, so the real question is: why does this only started coming up >> now, if it is known since 2007? I have been building gcc ports for >> ages, and never ran into this problem, but I also have never actively >> used a persistent LANG environment variable, let alone with UTF-8 in = it. >>=20 >> Is this because more people started using UTF-8 recently? >=20 > We are doing more correct collation now: >=20 > https://svnweb.freebsd.org/changeset/base/290494 Indeed. This problem has come up before on the ports mailing list, almost immediately after that commit: = https://lists.freebsd.org/pipermail/freebsd-ports/2015-November/101034.htm= l Apparently some proposals were made to set LANG and LC_ALL to C globally for port builds, but it was never implemented? I guess more people are now noticing it, because they are trying out the 11.0-ALPHA installers. -Dimitry --Apple-Mail=_C4FD4980-68D6-46FD-B1D9-BCC64693E3CF Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.30 iEYEARECAAYFAldYpbMACgkQsF6jCi4glqOvRwCg7FP7GcUrJuQisQeu8n++tTlp JhkAnjadknX5jP5TucYDY3BSCbvOoMIc =N6bR -----END PGP SIGNATURE----- --Apple-Mail=_C4FD4980-68D6-46FD-B1D9-BCC64693E3CF-- From owner-freebsd-toolchain@freebsd.org Wed Jun 8 23:21:27 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id A3CBAB703EE for ; Wed, 8 Jun 2016 23:21:27 +0000 (UTC) (envelope-from jkim@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:1900:2254:206c::16:87]) by mx1.freebsd.org (Postfix) with ESMTP id 0E2D510FB; Wed, 8 Jun 2016 23:21:26 +0000 (UTC) (envelope-from jkim@FreeBSD.org) Subject: Re: Duplicate OPT_ entries in gcc/options.h To: Dimitry Andric References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> <823B7E6C-4B47-401F-A254-C88825AE5562@FreeBSD.org> Cc: FreeBSD Toolchain , Gerald Pfeifer , Andreas Tobler , Baptiste Daroussin , Tijl Coosemans From: Jung-uk Kim Message-ID: <65f5f485-3ca8-4b22-7087-8793934f470d@FreeBSD.org> Date: Wed, 8 Jun 2016 19:21:16 -0400 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:45.0) Gecko/20100101 Thunderbird/45.1.1 MIME-Version: 1.0 In-Reply-To: <823B7E6C-4B47-401F-A254-C88825AE5562@FreeBSD.org> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="jm42IhHsbVuhf0Kk7NNQBMwScHFaN37Uv" X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2016 23:21:27 -0000 This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --jm42IhHsbVuhf0Kk7NNQBMwScHFaN37Uv Content-Type: multipart/mixed; boundary="1hWNeDVpF1L2EfIVrsEXeQI2QTsfm1OFe" From: Jung-uk Kim To: Dimitry Andric Cc: FreeBSD Toolchain , Gerald Pfeifer , Andreas Tobler , Baptiste Daroussin , Tijl Coosemans Message-ID: <65f5f485-3ca8-4b22-7087-8793934f470d@FreeBSD.org> Subject: Re: Duplicate OPT_ entries in gcc/options.h References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> <823B7E6C-4B47-401F-A254-C88825AE5562@FreeBSD.org> In-Reply-To: <823B7E6C-4B47-401F-A254-C88825AE5562@FreeBSD.org> --1hWNeDVpF1L2EfIVrsEXeQI2QTsfm1OFe Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable On 06/ 8/16 07:09 PM, Dimitry Andric wrote: > On 09 Jun 2016, at 00:30, Jung-uk Kim wrote: >> >> On 06/ 8/16 06:16 PM, Dimitry Andric wrote: >>> On 08 Jun 2016, at 23:54, Jung-uk Kim wrote: >>>> >>>> On 06/ 8/16 05:15 PM, Dimitry Andric wrote: >>>>> On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote= : >>>>>> >>>>>> I got a user report, and could reproduce this, that building >>>>>> GCC (lang/gcc, but also current HEAD, so probably pretty much >>>>>> any version) with FreeBSD 11 and LANG =3D en_US.UTF-8 we get >>>>>> conflicting entires in $BUILDDIR/gcc/options.h such as >>> ... >>>>> Note that GNU awk does *not* produce a different optionlist file wh= en >>>>> used with either LANG=3DC or LANG=3Den_US.UTF-8. >>> ... >>>>> So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works >>>>> differently in our awk, depending on the LANG settings. No idea wh= en >>>>> that changed, though, if it changed at all... >>>> >>>> This behaviour is known for very long time: >>>> >>>> https://svnweb.freebsd.org/changeset/base/173731 >>>> >>>> and it is not our fault: >>>> >>>> https://www.gnu.org/software/gawk/manual/html_node/POSIX-String-Comp= arison.html >>> >>> >>> Indeed, so the real question is: why does this only started coming up= >>> now, if it is known since 2007? I have been building gcc ports for >>> ages, and never ran into this problem, but I also have never actively= >>> used a persistent LANG environment variable, let alone with UTF-8 in = it. >>> >>> Is this because more people started using UTF-8 recently? >> >> We are doing more correct collation now: >> >> https://svnweb.freebsd.org/changeset/base/290494 >=20 > Indeed. This problem has come up before on the ports mailing list, > almost immediately after that commit: >=20 > https://lists.freebsd.org/pipermail/freebsd-ports/2015-November/101034.= html >=20 > Apparently some proposals were made to set LANG and LC_ALL to C globall= y > for port builds, but it was never implemented? tijl proposed it but never implemented because of difficulties to exp-run the change, it seems. http://docs.freebsd.org/cgi/mid.cgi?20151127145552.GA19655 > I guess more people are now noticing it, because they are trying out th= e > 11.0-ALPHA installers. Which is good. :-) Jung-uk Kim --1hWNeDVpF1L2EfIVrsEXeQI2QTsfm1OFe-- --jm42IhHsbVuhf0Kk7NNQBMwScHFaN37Uv Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJXWKhsAAoJEHyflib82/FGjBgH/RbjCWABfhgIcwfUubGwgeQF S3Z/V1OZHov+CT7FOMMil0sisELskc2FaRD6EJqKciDalDmuoupMShfb2xJAueIO K2RpSA6zw1nTWSGEJM+Y1ONZebUGISvEqaG3j2ACLIPRbj5D3dfZARCAZglR/K7j /J3qVlx8p0DI/YeynqLPfT2w0YQQyGUb96dRui9fym6UPnOwJO5i66dYw1kU/K4/ N1hIl0HkuDJOnwxN4zCZnDf7TEvVKEC3kcfbhIwxOHTc2jkZay20RICoJjOfFjOF FLpRiDW6e55jMdejSM4ZH93pP6TjsmFIFeaUWSQZFlBffqP407pRX1RDV6jkQ8A= =L15j -----END PGP SIGNATURE----- --jm42IhHsbVuhf0Kk7NNQBMwScHFaN37Uv-- From owner-freebsd-toolchain@freebsd.org Thu Jun 9 07:12:43 2016 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id D4CA8AC8151 for ; Thu, 9 Jun 2016 07:12:43 +0000 (UTC) (envelope-from tijl@freebsd.org) Received: from mailrelay111.isp.belgacom.be (mailrelay111.isp.belgacom.be [195.238.20.138]) (using TLSv1.2 with cipher RC4-SHA (128/128 bits)) (Client CN "relay.skynet.be", Issuer "GlobalSign Organization Validation CA - SHA256 - G2" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 254621B11; Thu, 9 Jun 2016 07:12:41 +0000 (UTC) (envelope-from tijl@freebsd.org) X-Belgacom-Dynamic: yes X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: A2BcDQAnFllX/4VasVtdgz5JDW4PuxSBehcNgjyDMwKBRDkUAQEBAQEBAWUnQQEBAQIBCgKDcwEBBAECNxwYCxALGAklDxkRHgYTiDMKvX8BAQEBAQEBAwEBAQEBASGKdIoaBZhShgOFNIJlcIFRjGhGjx0eNoIHHIFNOjIBAYNmgU+CI4IuAQEB Received: from 133.90-177-91.adsl-dyn.isp.belgacom.be (HELO kalimero.tijl.coosemans.org) ([91.177.90.133]) by relay.skynet.be with ESMTP; 09 Jun 2016 09:11:28 +0200 Received: from kalimero.tijl.coosemans.org (kalimero.tijl.coosemans.org [127.0.0.1]) by kalimero.tijl.coosemans.org (8.15.2/8.15.2) with ESMTP id u597BRW4001062; Thu, 9 Jun 2016 09:11:28 +0200 (CEST) (envelope-from tijl@FreeBSD.org) Date: Thu, 9 Jun 2016 09:11:27 +0200 From: Tijl Coosemans To: Jung-uk Kim Cc: Dimitry Andric , FreeBSD Toolchain , Gerald Pfeifer , Andreas Tobler , Baptiste Daroussin Subject: Re: Duplicate OPT_ entries in gcc/options.h Message-ID: <20160609091127.4795b484@kalimero.tijl.coosemans.org> In-Reply-To: <65f5f485-3ca8-4b22-7087-8793934f470d@FreeBSD.org> References: <75411813-0C9B-4CEF-BEE4-8B26DD8346F7@FreeBSD.org> <0610816e-2675-1abf-a4ee-274807317932@FreeBSD.org> <5A8D0A1F-6EA3-4DF2-82A2-EC1309C33E1E@FreeBSD.org> <823B7E6C-4B47-401F-A254-C88825AE5562@FreeBSD.org> <65f5f485-3ca8-4b22-7087-8793934f470d@FreeBSD.org> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Jun 2016 07:12:43 -0000 On Wed, 8 Jun 2016 19:21:16 -0400 Jung-uk Kim wrote: > On 06/ 8/16 07:09 PM, Dimitry Andric wrote: >> On 09 Jun 2016, at 00:30, Jung-uk Kim wrote: >>> On 06/ 8/16 06:16 PM, Dimitry Andric wrote: >>>> On 08 Jun 2016, at 23:54, Jung-uk Kim wrote: >>>>> On 06/ 8/16 05:15 PM, Dimitry Andric wrote: >>>>>> On 08 Jun 2016, at 21:11, Gerald Pfeifer wrote: >>>>>>> I got a user report, and could reproduce this, that building >>>>>>> GCC (lang/gcc, but also current HEAD, so probably pretty much >>>>>>> any version) with FreeBSD 11 and LANG = en_US.UTF-8 we get >>>>>>> conflicting entires in $BUILDDIR/gcc/options.h such as >>>> ... >>>>>> Note that GNU awk does *not* produce a different optionlist file when >>>>>> used with either LANG=C or LANG=en_US.UTF-8. >>>> ... >>>>>> So I am assuming that the ARRAY[j-1] > ARRAY[j] comparison works >>>>>> differently in our awk, depending on the LANG settings. No idea when >>>>>> that changed, though, if it changed at all... >>>>> >>>>> This behaviour is known for very long time: >>>>> >>>>> https://svnweb.freebsd.org/changeset/base/173731 >>>>> >>>>> and it is not our fault: >>>>> >>>>> https://www.gnu.org/software/gawk/manual/html_node/POSIX-String-Comparison.html >>>> >>>> >>>> Indeed, so the real question is: why does this only started coming up >>>> now, if it is known since 2007? I have been building gcc ports for >>>> ages, and never ran into this problem, but I also have never actively >>>> used a persistent LANG environment variable, let alone with UTF-8 in it. >>>> >>>> Is this because more people started using UTF-8 recently? >>> >>> We are doing more correct collation now: >>> >>> https://svnweb.freebsd.org/changeset/base/290494 >> >> Indeed. This problem has come up before on the ports mailing list, >> almost immediately after that commit: >> >> https://lists.freebsd.org/pipermail/freebsd-ports/2015-November/101034.html >> >> Apparently some proposals were made to set LANG and LC_ALL to C globally >> for port builds, but it was never implemented? > > tijl proposed it but never implemented because of difficulties to > exp-run the change, it seems. > > http://docs.freebsd.org/cgi/mid.cgi?20151127145552.GA19655 There are over 300 files in the ports tree that match "[[:<:]]LANG[[:>:]]". I just don't have the time to go over all of them.