From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 03:44:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o212fkWr029088; Mon, 1 Mar 2010 03:42:28 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o212fjOp028669 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 03:41:45 +0100 Received: (qmail 28782 invoked from network); 1 Mar 2010 02:41:46 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 1 Mar 2010 02:41:46 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 01 Mar 2010 02:41:46 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #159: libssh2 takes long time to disconnect from CoreFTP's SFTP service X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/159#comment:3 Message-ID: <048.2f2e8a271c46f5157625623a7b114d9d@libssh2.stuge.se> References: <039.c9b33f1492e7958089fb7727450037b1@libssh2.stuge.se> X-Trac-Ticket-ID: 159 In-Reply-To: <039.c9b33f1492e7958089fb7727450037b1@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #159: libssh2 takes long time to disconnect from CoreFTP's SFTP service -----------------------------------------+---------------------------------- Reporter: qxu | Owner: stuge Type: defect | Status: accepted Priority: normal | Milestone: 1.2.5 Component: protocol | Version: 1.2.4 Keywords: coreftp server idle timeout | Blocking: Blockedby: | -----------------------------------------+---------------------------------- Comment(by qxu): Great, Peter! Yes, the problem is fixed. There is no more visible delay. Thanks a lot for your quick fix! PS: Your fix was sent out 3 days ago. I am sorry to verify it and reply you late, coz I took 2 days' leave last week. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 07:06:18 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o216663S015761; Mon, 1 Mar 2010 07:06:14 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21664nw015756 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 07:06:04 +0100 Received: (qmail 17196 invoked from network); 1 Mar 2010 06:06:03 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 1 Mar 2010 06:06:03 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 01 Mar 2010 06:06:03 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #159: libssh2 takes long time to disconnect from CoreFTP's SFTP service X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/159#comment:4 Message-ID: <048.9fac1e73064f10cee555a23ba8ec4700@libssh2.stuge.se> References: <039.c9b33f1492e7958089fb7727450037b1@libssh2.stuge.se> X-Trac-Ticket-ID: 159 In-Reply-To: <039.c9b33f1492e7958089fb7727450037b1@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #159: libssh2 takes long time to disconnect from CoreFTP's SFTP service -----------------------+---------------------------------------------------- Reporter: qxu | Owner: stuge Type: defect | Status: closed Priority: normal | Milestone: 1.2.5 Component: protocol | Version: 1.2.4 Resolution: fixed | Keywords: coreftp server idle timeout Blocking: | Blockedby: -----------------------+---------------------------------------------------- Changes (by stuge): * status: accepted => closed * resolution: => fixed Comment: Replying to [comment:3 qxu]: > Yes, the problem is fixed. Great! Thank you for testing! This is commit 16af813b812aca09c73045288040ddf1a03a2b86 -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 10:05:46 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2195IJc022486; Mon, 1 Mar 2010 10:05:26 +0100 Received: from av9-1-sn3.vrr.skanova.net (av9-1-sn3.vrr.skanova.net [81.228.9.185]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2195Ggw022355 for ; Mon, 1 Mar 2010 10:05:16 +0100 Received: by av9-1-sn3.vrr.skanova.net (Postfix, from userid 502) id 0FE96380CA; Mon, 1 Mar 2010 10:05:12 +0100 (CET) Received: from smtp3-1-sn3.vrr.skanova.net (smtp3-1-sn3.vrr.skanova.net [81.228.9.101]) by av9-1-sn3.vrr.skanova.net (Postfix) with ESMTP id B89BC380C9 for ; Mon, 1 Mar 2010 10:05:11 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp3-1-sn3.vrr.skanova.net (Postfix) with ESMTP id A3E9F37E4B for ; Mon, 1 Mar 2010 10:04:41 +0100 (CET) Message-ID: <4B8B8323.80606@SDlabs.se> Date: Mon, 01 Mar 2010 10:04:35 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code Content-Type: multipart/mixed; boundary="------------030205060807080103030209" X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --------------030205060807080103030209 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit This is a patch to use OpenSSL's version of aes_*_ctr functions and to not use DSA-code if not supported by OpenSSL From version 0.9.8h, full support for aes_*_ctr was added. I have tested with 0.9.8h, 0.9.8i (using the openssl version of ctr-functions and 0.9.7f (with builtin ctr function). 0.9.8g is not working, but it's the same with a non-patched libssh2. /Lars --------------030205060807080103030209 Content-Type: text/plain; name="libssh2_openssl_aes_ctr_patch.diff" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="libssh2_openssl_aes_ctr_patch.diff" diff -ru libssh2-1.2.5-20100226_orig/src/openssl.c libssh2-1.2.5-20100226/src/openssl.c --- libssh2-1.2.5-20100226_orig/src/openssl.c 2010-02-11 06:06:03.000000000 +0100 +++ libssh2-1.2.5-20100226/src/openssl.c 2010-02-26 10:44:26.000000000 +0100 @@ -111,6 +111,7 @@ return (ret == 1) ? 0 : -1; } +#if LIBSSH2_DSA int _libssh2_dsa_new(libssh2_dsa_ctx ** dsactx, const unsigned char *p, @@ -166,6 +167,7 @@ return (ret == 1) ? 0 : -1; } +#endif /* LIBSSH_DSA */ int _libssh2_cipher_init(_libssh2_cipher_ctx * h, @@ -367,6 +369,7 @@ filename, passphrase); } +#if LIBSSH2_DSA int _libssh2_dsa_new_private(libssh2_dsa_ctx ** dsa, LIBSSH2_SESSION * session, @@ -387,6 +390,7 @@ return read_private_key_from_file((void **) dsa, read_dsa, filename, passphrase); } +#endif /* LIBSSH_DSA */ int _libssh2_rsa_sha1_sign(LIBSSH2_SESSION * session, @@ -419,6 +423,7 @@ return 0; } +#if LIBSSH2_DSA int _libssh2_dsa_sha1_sign(libssh2_dsa_ctx * dsactx, const unsigned char *hash, @@ -453,5 +458,6 @@ return 0; } +#endif /* LIBSSH_DSA */ #endif /* !LIBSSH2_LIBGCRYPT */ diff -ru libssh2-1.2.5-20100226_orig/src/openssl.h libssh2-1.2.5-20100226/src/openssl.h --- libssh2-1.2.5-20100226_orig/src/openssl.h 2009-11-17 06:06:05.000000000 +0100 +++ libssh2-1.2.5-20100226/src/openssl.h 2010-02-26 10:06:38.000000000 +0100 @@ -73,7 +73,11 @@ #endif #if OPENSSL_VERSION_NUMBER >= 0x00907000L && !defined(OPENSSL_NO_AES) +#if OPENSSL_VERSION_NUMBER >= 0x0090808fL +# define LIBSSH2_AES_CTR 0 +#else # define LIBSSH2_AES_CTR 1 +#endif # define LIBSSH2_AES 1 #else # define LIBSSH2_AES 0 @@ -197,9 +201,15 @@ #define _libssh2_cipher_aes256 EVP_aes_256_cbc #define _libssh2_cipher_aes192 EVP_aes_192_cbc #define _libssh2_cipher_aes128 EVP_aes_128_cbc +#if LIBSSH2_AES_CTR #define _libssh2_cipher_aes128ctr _libssh2_EVP_aes_128_ctr #define _libssh2_cipher_aes192ctr _libssh2_EVP_aes_192_ctr #define _libssh2_cipher_aes256ctr _libssh2_EVP_aes_256_ctr +#else +#define _libssh2_cipher_aes128ctr EVP_aes_128_ctr +#define _libssh2_cipher_aes192ctr EVP_aes_192_ctr +#define _libssh2_cipher_aes256ctr EVP_aes_256_ctr +#endif #define _libssh2_cipher_blowfish EVP_bf_cbc #define _libssh2_cipher_arcfour EVP_rc4 #define _libssh2_cipher_cast5 EVP_cast5_cbc --------------030205060807080103030209 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --------------030205060807080103030209-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 10:38:26 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o219cH31010228; Mon, 1 Mar 2010 10:38:21 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o219cEvj010223 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 10:38:15 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o219cAht018851 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 10:38:14 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::29j/LGWtB5WpC82k:B7j6 Date: Mon, 01 Mar 2010 10:38:10 +0100 In-Reply-To: <4B8B8323.80606@SDlabs.se> (Lars Nordin's message of "Mon, 01 Mar 2010 10:04:35 +0100") Message-ID: <87ljec1iwt.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Lars Nordin writes: > This is a patch to use OpenSSL's version of aes_*_ctr functions and to > not use DSA-code if not supported by OpenSSL Thanks. I applied your DSA-code patch as-is, but modified the AES-CTR patch somewhat... > From version 0.9.8h, full support for aes_*_ctr was added. I have > tested with 0.9.8h, 0.9.8i (using the openssl version of ctr-functions > and 0.9.7f (with builtin ctr function). 0.9.8g is not working, but > it's the same with a non-patched libssh2. ...but did you test that AES-CTR worked and not just built? As far as I can tell, your patch would disable AES-CTR completely with newer OpenSSLs since it sets LIBSSH2_AES_CTR to 0 which causes crypt.c to not register it. > +#if OPENSSL_VERSION_NUMBER >= 0x0090808fL > +# define LIBSSH2_AES_CTR 0 > +#else > # define LIBSSH2_AES_CTR 1 > +#endif This is the part that seems wrong to me. Please test the code in git now. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 11:07:01 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21A6puV003929; Mon, 1 Mar 2010 11:06:59 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21A6nab003921 for ; Mon, 1 Mar 2010 11:06:49 +0100 Date: Mon, 1 Mar 2010 11:06:49 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code In-Reply-To: <87ljec1iwt.fsf@mocca.josefsson.org> Message-ID: References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-1470944514-1351047232-1267437920=:12887" Content-ID: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---1470944514-1351047232-1267437920=:12887 Content-Type: TEXT/PLAIN; CHARSET=ISO-8859-7; FORMAT=flowed Content-Transfer-Encoding: 8BIT Content-ID: On Mon, 1 Mar 2010, Simon Josefsson wrote: > Please test the code in git now. Doesn't build for me! crypt.c:128: error: ¡EVP_aes_128_ctr¢ undeclared here (not in a function) crypt.c:129: warning: missing initializer crypt.c:129: warning: (near initialization for ¡libssh2_crypt_method_aes128_ctr.algo¢) crypt.c:140: error: ¡EVP_aes_192_ctr¢ undeclared here (not in a function) crypt.c:141: warning: missing initializer crypt.c:141: warning: (near initialization for ¡libssh2_crypt_method_aes192_ctr.algo¢) crypt.c:152: error: ¡EVP_aes_256_ctr¢ undeclared here (not in a function) crypt.c:153: warning: missing initializer crypt.c:153: warning: (near initialization for ¡libssh2_crypt_method_aes256_ctr.algo¢) make[2]: *** [crypt.lo] Error 1 make[2]: Leaving directory `/home/daniel/src/libssh2/src' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/daniel/src/libssh2/src' make: *** [all-recursive] Error 1 -- / daniel.haxx.se ---1470944514-1351047232-1267437920=:12887 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel ---1470944514-1351047232-1267437920=:12887-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 11:08:28 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21A8P3u004431; Mon, 1 Mar 2010 11:08:27 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21A8OdB004423 for ; Mon, 1 Mar 2010 11:08:24 +0100 Date: Mon, 1 Mar 2010 11:08:24 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings In-Reply-To: <20100301100559.2197.qmail@earth.stuge.se> Message-ID: References: <20100301100559.2197.qmail@earth.stuge.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, libssh2@git.stuge.se wrote: > use const to silence a bazillion warnings My advice to committers: use configure --enable-debug to get a wide range of picky compiler options enabled. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 11:41:11 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Af37C018321; Mon, 1 Mar 2010 11:41:08 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Aexsn018229 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 11:41:00 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21AevIA021100 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 11:40:59 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::GlVNMpvnrrrhaSv6:Cd5B Date: Mon, 01 Mar 2010 11:40:57 +0100 In-Reply-To: (Daniel Stenberg's message of "Mon, 1 Mar 2010 11:06:49 +0100 (CET)") Message-ID: <87sk8kz5mu.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se RGFuaWVsIFN0ZW5iZXJnIDxkYW5pZWxAaGF4eC5zZT4gd3JpdGVzOgoKPiBPbiBNb24sIDEgTWFy IDIwMTAsIFNpbW9uIEpvc2Vmc3NvbiB3cm90ZToKPgo+PiBQbGVhc2UgdGVzdCB0aGUgY29kZSBp biBnaXQgbm93Lgo+Cj4gRG9lc24ndCBidWlsZCBmb3IgbWUhCj4KPiBjcnlwdC5jOjEyODogZXJy b3I6IOKAmEVWUF9hZXNfMTI4X2N0cuKAmSB1bmRlY2xhcmVkIGhlcmUgKG5vdCBpbiBhIGZ1bmN0 aW9uKQo+IGNyeXB0LmM6MTI5OiB3YXJuaW5nOiBtaXNzaW5nIGluaXRpYWxpemVyCj4gY3J5cHQu YzoxMjk6IHdhcm5pbmc6IChuZWFyIGluaXRpYWxpemF0aW9uIGZvcgo+IOKAmGxpYnNzaDJfY3J5 cHRfbWV0aG9kX2FlczEyOF9jdHIuYWxnb+KAmSkKPiBjcnlwdC5jOjE0MDogZXJyb3I6IOKAmEVW UF9hZXNfMTkyX2N0cuKAmSB1bmRlY2xhcmVkIGhlcmUgKG5vdCBpbiBhIGZ1bmN0aW9uKQo+IGNy eXB0LmM6MTQxOiB3YXJuaW5nOiBtaXNzaW5nIGluaXRpYWxpemVyCj4gY3J5cHQuYzoxNDE6IHdh cm5pbmc6IChuZWFyIGluaXRpYWxpemF0aW9uIGZvcgo+IOKAmGxpYnNzaDJfY3J5cHRfbWV0aG9k X2FlczE5Ml9jdHIuYWxnb+KAmSkKPiBjcnlwdC5jOjE1MjogZXJyb3I6IOKAmEVWUF9hZXNfMjU2 X2N0cuKAmSB1bmRlY2xhcmVkIGhlcmUgKG5vdCBpbiBhIGZ1bmN0aW9uKQo+IGNyeXB0LmM6MTUz OiB3YXJuaW5nOiBtaXNzaW5nIGluaXRpYWxpemVyCj4gY3J5cHQuYzoxNTM6IHdhcm5pbmc6IChu ZWFyIGluaXRpYWxpemF0aW9uIGZvcgo+IOKAmGxpYnNzaDJfY3J5cHRfbWV0aG9kX2FlczI1Nl9j dHIuYWxnb+KAmSkKCkl0IHNlZW1zIHNvbWUgT3BlblNTTCBoYXZlIEFFUy1DVFIgYW5kIHNvbWUg ZG8gbm90LCBldmVuIGlmIHRoZSB2ZXJzaW9uCmlzIG5ldyBlbm91Z2guICBJIGNhbiByZXByb2R1 Y2UgdGhlIGFib3ZlLiAgT24gbXkgZGViaWFuIHN5c3RlbQovdXNyL2luY2x1ZGUvb3BlbnNzbC9l dnAuaCBoYXM6CgojaWYgMApjb25zdCBFVlBfQ0lQSEVSICpFVlBfYWVzXzEyOF9jdHIodm9pZCk7 CiNlbmRpZgoKVGhlIHZlcnNpb24gaXMgcmVjZW50IGVub3VnaDoKCiNkZWZpbmUgT1BFTlNTTF9W RVJTSU9OX05VTUJFUiAgMHgwMDkwODBiZkwKCkkgYmVsaWV2ZSB3ZSBuZWVkIHRvIGNoZWNrIHRo YXQgRVZQX2Flc18xMjhfY3RyIGFjdHVhbGx5IGlzIHByZXNlbnQgaW4KdGhlIGxpYnJhcnksIHNv IEkndmUgYXBwbGllZCBhIHBhdGNoIHRvIGRvIHRoYXQuICBQbGVhc2UgdGVzdCBhZ2Fpbi4KCkkg bm93IHJlY2FsbCB0aGF0IHRoaXMgd2FzIHRoZSByZWFzb24gSSBkaWQgbm90IHVzZSBPcGVuU1NM J3MgYnVpbHQtaW4KQUVTLUNUUiBBUElzIGJ1dCBoYWQgdG8gcmUtaW1wbGVtZW50IGl0LiAgQUVT LUNUUiBkb2Vzbid0IGFwcGVhciB3aWRlbHkKYXZhaWxhYmxlIGluIE9wZW5TU0wgaW4gZGlzdHJp YnV0aW9ucywgZXZlbiBpZiBpdCBpcyBpbiBPcGVuU1NMCnVwc3RyZWFtLgoKL1NpbW9uCl9fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fCmxpYnNzaDItZGV2ZWwg aHR0cDovL2Nvb2wuaGF4eC5zZS9jZ2ktYmluL21haWxtYW4vbGlzdGluZm8vbGlic3NoMi1kZXZl bAo= From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 11:43:34 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21AhUan021995; Mon, 1 Mar 2010 11:43:32 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21AhTCF021967 for ; Mon, 1 Mar 2010 11:43:29 +0100 Date: Mon, 1 Mar 2010 11:43:29 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code In-Reply-To: <87sk8kz5mu.fsf@mocca.josefsson.org> Message-ID: References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, Simon Josefsson wrote: > I believe we need to check that EVP_aes_128_ctr actually is present in the > library, so I've applied a patch to do that. Please test again. I updated, did a little #if fixing (pushed) and now it builds fine. Thanks! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 12:27:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21BRIHQ022649; Mon, 1 Mar 2010 12:27:25 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21BRHpv022643 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 12:27:17 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21BRFch022062 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 12:27:17 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::Vjnk5LYNvfdS+atI:HIjD Date: Mon, 01 Mar 2010 12:27:15 +0100 In-Reply-To: (Daniel Stenberg's message of "Mon, 1 Mar 2010 11:43:29 +0100 (CET)") Message-ID: <87ljecz3ho.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Mon, 1 Mar 2010, Simon Josefsson wrote: > >> I believe we need to check that EVP_aes_128_ctr actually is present >> in the library, so I've applied a patch to do that. Please test >> again. > > I updated, did a little #if fixing (pushed) and now it builds fine. Thanks! Great. Btw, any particular reason you needed those fixes? They don't look required to me. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 12:30:51 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21BUm1C023880; Mon, 1 Mar 2010 12:30:49 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21BUlTD023874 for ; Mon, 1 Mar 2010 12:30:47 +0100 Date: Mon, 1 Mar 2010 12:30:47 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code In-Reply-To: <87ljecz3ho.fsf@mocca.josefsson.org> Message-ID: References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, Simon Josefsson wrote: > Great. Btw, any particular reason you needed those fixes? They don't look > required to me. Because the first one caused a compiler failure, and the second a compiler warning... -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 12:37:09 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Bb3TX026143; Mon, 1 Mar 2010 12:37:06 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Bb21U026119 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 12:37:02 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21Bb0jK022322 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 12:37:02 +0100 X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::WSF2zApGmcuzPzIz:5vEd From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: ./configure summary? OpenPGP: id=B565716F; url=http://josefsson.org/key.txt Date: Mon, 01 Mar 2010 12:37:00 +0100 Message-ID: <87hbp0z31f.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX, FB_WORD1_END_DOLLAR, RDNS_DYNAMIC, SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Any objections to pushing the patch below, so that when ./configure has completed it will print something like this: ... config.status: executing depfiles commands config.status: executing libtool commands configure: summary of build options: version: 1.2.5_DEV Host type: i686-pc-linux-gnu Install prefix: /usr/local Compiler: gcc Compiler flags: -g -O2 Library types: Shared=yes, Static=yes Crypto library: openssl: yes (AES-CTR: no) libgcrypt: Debug build: no Path to sshd: /usr/sbin/sshd (only for self-tests) libz compression: yes jas@mocca:~/src/libssh2 master$ /Simon diff --git a/configure.ac b/configure.ac index a9b54b0..4debd24 100644 --- a/configure.ac +++ b/configure.ac @@ -152,6 +152,7 @@ dnl ************************************************************ dnl option to switch on compiler debug options dnl AC_MSG_CHECKING([whether to enable pedantic and debug compiler options]) +enable_debug=no AC_ARG_ENABLE(debug, AC_HELP_STRING([--enable-debug],[Enable pedantic and debug options]) AC_HELP_STRING([--disable-debug],[Disable debug options]), @@ -160,7 +161,7 @@ AC_HELP_STRING([--disable-debug],[Disable debug options]), AC_MSG_RESULT(no) ;; *) AC_MSG_RESULT(yes) - + enable_debug=yes CPPFLAGS="$CPPFLAGS -DLIBSSH2DEBUG" CFLAGS="$CFLAGS -g" @@ -267,3 +268,17 @@ AC_CONFIG_FILES([Makefile docs/Makefile libssh2.pc]) AC_OUTPUT + +AC_MSG_NOTICE([summary of build options: + + version: ${VERSION} + Host type: ${host} + Install prefix: ${prefix} + Compiler: ${CC} + Compiler flags: ${CFLAGS} + Library types: Shared=${enable_shared}, Static=${enable_static} + Crypto library: openssl: $ac_cv_libssl (AES-CTR: $ac_cv_func_EVP_aes128_ctr) libgcrypt: $ac_cv_libgcrypt + Debug build: $enable_debug + Path to sshd: $ac_cv_path_SSHD (only for self-tests) + libz compression: $ac_cv_libz +]) _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 12:59:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21BxYJN008588; Mon, 1 Mar 2010 12:59:39 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21BxWvM008584 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 12:59:33 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21BxVMl022973 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 12:59:32 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::UhzHK7hQ5pWjpFx0:FVFf Date: Mon, 01 Mar 2010 12:59:31 +0100 In-Reply-To: (Daniel Stenberg's message of "Mon, 1 Mar 2010 12:30:47 +0100 (CET)") Message-ID: <87vddgxnfg.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Mon, 1 Mar 2010, Simon Josefsson wrote: > >> Great. Btw, any particular reason you needed those fixes? They >> don't look required to me. > > Because the first one caused a compiler failure, and the second a > compiler warning... I couldn't reproduce a compiler failure, but the warning was likely due to -Wundef (which IMHO is pointless -- undeclared CPP constants must evaluate to 0). Just curious; I don't care strongly. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 13:01:37 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21C1YEb011361; Mon, 1 Mar 2010 13:01:35 +0100 Received: from fg-out-1718.google.com (fg-out-1718.google.com [72.14.220.159]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21C1V5a011332 for ; Mon, 1 Mar 2010 13:01:32 +0100 Received: by fg-out-1718.google.com with SMTP id e21so571575fga.11 for ; Mon, 01 Mar 2010 04:01:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=u64JdC/b3Xrd0IK0QLMBOfS6ihHeOZrF2XjJlsePqks=; b=MTpLIgWXaPmGVkyUKqz2f6wWOzbbWSyrUHOa56WWzm8LVhybm5l2HUtBouZPxUtRbG Ee0q8KkYPcfMi4GbUihxvRNuyjDTXAZz/6cpXJUbDaFGnOjxShN71tGmwdXIjskZHJcH ckAVYSgjBoqs8Jg8W7/t4R+mPdCbKWhtc/uVE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=cfGM1LLGSRwr48Qzd9ac/vpOxEBEQJOqkFBvmXrGhwHj1KZhzQ/arvh1SaI4/dFYtd 6Ovs9QDdUVGXzUQWmN0tz1/Qx6JfffmqkAYsuHI0T5lbnf0N9S/iaM8jT1oeyuTS9N6o Z4Gtc+HeUfiHXbvHAigqOhAi8AgGTLHbAMLCU= MIME-Version: 1.0 Received: by 10.239.166.84 with SMTP id a20mr434039hbe.116.1267444890869; Mon, 01 Mar 2010 04:01:30 -0800 (PST) In-Reply-To: References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> Date: Mon, 1 Mar 2010 13:01:30 +0100 X-Google-Sender-Auth: 8271515cfa38a77e Message-ID: Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code From: Tor Arntsen To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, Mar 1, 2010 at 12:30, Daniel Stenberg wrote: > On Mon, 1 Mar 2010, Simon Josefsson wrote: > >> Great. =A0Btw, any particular reason you needed those fixes? =A0They don= 't >> look required to me. > > Because the first one caused a compiler failure, and the second a compiler > warning... But if it's this change: -#if LIBSSH2_AES_CTR && !HAVE_EVP_AES128_CTR +#if defined(LIBSSH2_AES_CTR) && !defined(HAVE_EVP_AES128_CTR) then there's something else wrong, because the code has: src/openssl.h: #if OPENSSL_VERSION_NUMBER >=3D 0x00907000L && !defined(OPENSSL_NO_AES) # define LIBSSH2_AES_CTR 1 # define LIBSSH2_AES 1 #else # define LIBSSH2_AES_CTR 0 # define LIBSSH2_AES 0 #endif so the test _should_ test for 1, and not just for defined (i.e. the change ef27023.. is wrong). If it fails then that means cpp didn't go through the path above - i.e. src/openssl.c didn't read src/openssl.h... -Tor _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 13:08:23 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21C8JhY018461; Mon, 1 Mar 2010 13:08:21 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21C8Iw7018453 for ; Mon, 1 Mar 2010 13:08:18 +0100 Date: Mon, 1 Mar 2010 13:08:18 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code In-Reply-To: <87vddgxnfg.fsf@mocca.josefsson.org> Message-ID: References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> <87vddgxnfg.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, Simon Josefsson wrote: > I couldn't reproduce a compiler failure, but the warning was likely due to > -Wundef (which IMHO is pointless -- undeclared CPP constants must evaluate > to 0). Just curious; I don't care strongly. Ah, yes it was a warning only, the error must've been some mistake of mine. I like -Wundef because it makes you write stricter code that leaves less mistakes around. Like when you do #if [misspelled define] which always will evaluate false but with -Wundef will give you a clear warning. That specific mistake is also very hard to spot and track down (and yes I've seen it happen more than once). -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 13:11:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CB2T0019622; Mon, 1 Mar 2010 13:11:04 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CB0MQ019580 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 13:11:00 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21CAxi0023311 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 13:11:00 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::PZ0Foa9+ey4YZPSn:QYOM Date: Mon, 01 Mar 2010 13:10:59 +0100 In-Reply-To: (Tor Arntsen's message of "Mon, 1 Mar 2010 13:01:30 +0100") Message-ID: <87r5o4xmwc.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Tor Arntsen writes: > On Mon, Mar 1, 2010 at 12:30, Daniel Stenberg wrote: >> On Mon, 1 Mar 2010, Simon Josefsson wrote: >> >>> Great. =A0Btw, any particular reason you needed those fixes? =A0They do= n't >>> look required to me. >> >> Because the first one caused a compiler failure, and the second a compil= er >> warning... > > But if it's this change: > > -#if LIBSSH2_AES_CTR && !HAVE_EVP_AES128_CTR > +#if defined(LIBSSH2_AES_CTR) && !defined(HAVE_EVP_AES128_CTR) > > then there's something else wrong, because the code has: > > src/openssl.h: > #if OPENSSL_VERSION_NUMBER >=3D 0x00907000L && !defined(OPENSSL_NO_AES) > # define LIBSSH2_AES_CTR 1 > # define LIBSSH2_AES 1 > #else > # define LIBSSH2_AES_CTR 0 > # define LIBSSH2_AES 0 > #endif > > so the test _should_ test for 1, and not just for defined (i.e. the > change ef27023.. is wrong). > If it fails then that means cpp didn't go through the path above - > i.e. src/openssl.c didn't read src/openssl.h... Good point -- I agree for LIBSSH2_AES_CTR, and fixed that. The change actually made the code behave incorrectly. The second (HAVE_EVP_AES128_CTR) is an aesthetic issue. While my preference is to not have unnecessary defined checks, I don't care strongly about it. (While I'm at it, another aesthetic discussion is on the unnecessary parenthesis after a 'defined' keyword...) /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 13:19:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CIoJs022122; Mon, 1 Mar 2010 13:18:52 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CImmk022117 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 13:18:48 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21CIkfb023505 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 13:18:48 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> <87vddgxnfg.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::cCEacAa53h6tdRgc:MzOG Date: Mon, 01 Mar 2010 13:18:46 +0100 In-Reply-To: (Daniel Stenberg's message of "Mon, 1 Mar 2010 13:08:18 +0100 (CET)") Message-ID: <87k4twxmjd.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Mon, 1 Mar 2010, Simon Josefsson wrote: > >> I couldn't reproduce a compiler failure, but the warning was likely >> due to -Wundef (which IMHO is pointless -- undeclared CPP constants >> must evaluate to 0). Just curious; I don't care strongly. > > Ah, yes it was a warning only, the error must've been some mistake of > mine. I like -Wundef because it makes you write stricter code that > leaves less mistakes around. Like when you do #if [misspelled define] > which always will evaluate false but with -Wundef will give you a > clear warning. That specific mistake is also very hard to spot and > track down (and yes I've seen it happen more than once). I understand, and I've seen that too. There are disadvantages though: if someone declares the constant to 0 and expects the feature to be disabled. Further, it encourages adding defined-checks to all CPP symbols regardless of how they are intended to be used. I've seen these two patterns cause problems more than once as well. Anyway, this is an aesthetic discussion. I believe discussing it helps to make us aware of the underlying trade-offs though. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 13:21:25 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CLMs0024768; Mon, 1 Mar 2010 13:21:24 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CLLQo024745 for ; Mon, 1 Mar 2010 13:21:21 +0100 Date: Mon, 1 Mar 2010 13:21:21 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code In-Reply-To: <87k4twxmjd.fsf@mocca.josefsson.org> Message-ID: References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> <87vddgxnfg.fsf@mocca.josefsson.org> <87k4twxmjd.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, Simon Josefsson wrote: >> Ah, yes it was a warning only, the error must've been some mistake of mine. >> I like -Wundef because it makes you write stricter code that leaves less >> mistakes around. Like when you do #if [misspelled define] which always will >> evaluate false but with -Wundef will give you a clear warning. That >> specific mistake is also very hard to spot and track down (and yes I've >> seen it happen more than once). > > I understand, and I've seen that too. There are disadvantages though: if > someone declares the constant to 0 and expects the feature to be disabled. > Further, it encourages adding defined-checks to all CPP symbols regardless > of how they are intended to be used. I've seen these two patterns cause > problems more than once as well. Yes, as a consequence of the above I think enabling and disabling features in code should be done with #ifdef and #define/#undef, and not #if and #define [variable] 0/1. > Anyway, this is an aesthetic discussion. I believe discussing it helps > to make us aware of the underlying trade-offs though. Agreed! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 13:44:27 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CiET9013853; Mon, 1 Mar 2010 13:44:23 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21CiCg7013796 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 13:44:12 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21CiAA7023954 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 13:44:12 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> <87vddgxnfg.fsf@mocca.josefsson.org> <87k4twxmjd.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::AnUSj1ohT3kG5XFr:5MKy Date: Mon, 01 Mar 2010 13:44:10 +0100 In-Reply-To: (Daniel Stenberg's message of "Mon, 1 Mar 2010 13:21:21 +0100 (CET)") Message-ID: <87tyt0w6sl.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Mon, 1 Mar 2010, Simon Josefsson wrote: > >>> Ah, yes it was a warning only, the error must've been some mistake >>> of mine. I like -Wundef because it makes you write stricter code >>> that leaves less mistakes around. Like when you do #if [misspelled >>> define] which always will evaluate false but with -Wundef will give >>> you a clear warning. That specific mistake is also very hard to >>> spot and track down (and yes I've seen it happen more than once). >> >> I understand, and I've seen that too. There are disadvantages >> though: if someone declares the constant to 0 and expects the >> feature to be disabled. Further, it encourages adding defined-checks >> to all CPP symbols regardless of how they are intended to be used. >> I've seen these two patterns cause problems more than once as well. > > Yes, as a consequence of the above I think enabling and disabling > features in code should be done with #ifdef and #define/#undef, and > not #if and #define [variable] 0/1. Consistency is good; there is plenty of legacy code in crypt.c, openssl.c, mac.c etc that we could clean up. Right now the code seems to use a mix of these two patterns, which IMHO is worse then picking any of them and sticking to it. Yes, I am probably responsible for some of this, but hopefully not all of it. ;-) To little time to fix this right now though. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 14:52:30 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21DqIpk025858; Mon, 1 Mar 2010 14:52:27 +0100 Received: from av9-1-sn2.hy.skanova.net (av9-1-sn2.hy.skanova.net [81.228.8.179]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21DqH5u025799 for ; Mon, 1 Mar 2010 14:52:17 +0100 Received: by av9-1-sn2.hy.skanova.net (Postfix, from userid 502) id 2DB1238143; Mon, 1 Mar 2010 14:52:13 +0100 (CET) Received: from smtp4-1-sn2.hy.skanova.net (smtp4-1-sn2.hy.skanova.net [81.228.8.92]) by av9-1-sn2.hy.skanova.net (Postfix) with ESMTP id 8C13F3859E for ; Mon, 1 Mar 2010 14:34:01 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp4-1-sn2.hy.skanova.net (Postfix) with ESMTP id 81E6337E56 for ; Mon, 1 Mar 2010 14:34:01 +0100 (CET) Message-ID: <4B8BC246.2030706@SDlabs.se> Date: Mon, 01 Mar 2010 14:33:58 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> In-Reply-To: <87ljec1iwt.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 2010-03-01 10:38, Simon Josefsson wrote: > Lars Nordin writes: > > >> This is a patch to use OpenSSL's version of aes_*_ctr functions and to >> not use DSA-code if not supported by OpenSSL >> > Thanks. I applied your DSA-code patch as-is, but modified the AES-CTR > patch somewhat... > > >> From version 0.9.8h, full support for aes_*_ctr was added. I have >> tested with 0.9.8h, 0.9.8i (using the openssl version of ctr-functions >> and 0.9.7f (with builtin ctr function). 0.9.8g is not working, but >> it's the same with a non-patched libssh2. >> > ...but did you test that AES-CTR worked and not just built? > > As far as I can tell, your patch would disable AES-CTR completely with > newer OpenSSLs since it sets LIBSSH2_AES_CTR to 0 which causes crypt.c > to not register it. > > >> +#if OPENSSL_VERSION_NUMBER>= 0x0090808fL >> +# define LIBSSH2_AES_CTR 0 >> +#else >> # define LIBSSH2_AES_CTR 1 >> +#endif >> > The internal LIBSSH2_AES_CTR shall NOT be used if openSSL supports aes_ctr, there is nothing wrong in the patch. I have testcompiled and check (using nm) if the LIBSSH2_AES_ctr function was used. I have also tested both versions using example/ssh2 to a openssh server on localhost. > This is the part that seems wrong to me. > > Please test the code in git now. > > /Simon > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 14:55:22 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21DtJpP027182; Mon, 1 Mar 2010 14:55:20 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21DtH0S027173 for ; Mon, 1 Mar 2010 14:55:18 +0100 Received: (qmail 14475 invoked by uid 501); 1 Mar 2010 13:55:18 -0000 Message-ID: <20100301135518.14474.qmail@stuge.se> Date: Mon, 1 Mar 2010 14:55:18 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <87ljec1iwt.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson wrote: > > This is a patch to use OpenSSL's version of aes_*_ctr functions and to > > not use DSA-code if not supported by OpenSSL > > Thanks. I applied your DSA-code patch as-is, Don't do this. DSA is required by RFC. Please have a look. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 14:56:55 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Duq1e027506; Mon, 1 Mar 2010 14:56:53 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21DupXJ027502 for ; Mon, 1 Mar 2010 14:56:51 +0100 Received: (qmail 14707 invoked by uid 501); 1 Mar 2010 13:56:51 -0000 Message-ID: <20100301135651.14706.qmail@stuge.se> Date: Mon, 1 Mar 2010 14:56:51 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <87sk8kz5mu.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson wrote: > It seems some OpenSSL have AES-CTR and some do not, I expressed an idea a while back for libssh2 to autodetect what crypto is available and run with that. No need to decide at build time, right? //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 14:59:57 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21DxrE4029908; Mon, 1 Mar 2010 14:59:55 +0100 Received: from dottedmag.net (burger.dottedmag.net [212.75.37.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Dxpud029904 for ; Mon, 1 Mar 2010 14:59:52 +0100 Received: from vertex.dottedmag (unknown [91.197.127.125]) by dottedmag.net (Postfix) with ESMTPSA id 483018C74C for ; Mon, 1 Mar 2010 14:59:51 +0100 (CET) Received: from dottedmag by vertex.dottedmag with local (Exim 4.69) (envelope-from ) id 1Nm6A0-000285-11 for libssh2-devel@cool.haxx.se; Mon, 01 Mar 2010 19:59:48 +0600 From: Mikhail Gusarov To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <20100301135651.14706.qmail@stuge.se> Date: Mon, 01 Mar 2010 19:59:47 +0600 In-Reply-To: <20100301135651.14706.qmail@stuge.se> (Peter Stuge's message of "Mon, 1 Mar 2010 14:56:51 +0100") Message-ID: <87tyt0b0rw.fsf@vertex.dottedmag> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1150211997==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============1150211997== Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --=-=-= Content-Transfer-Encoding: quoted-printable Twas brillig at 14:56:51 01.03.2010 UTC+01 when peter@stuge.se did gyre and= gimble: >> It seems some OpenSSL have AES-CTR and some do not, PS> I expressed an idea a while back for libssh2 to autodetect what PS> crypto is available and run with that. In the next version libssh2 would detect all available crypto, test it for speed and select the fastest. Next version would include downloading crypto library and compiling it if nothing found. And then the subsequent version would bring bootstrapping toolchain for building libraries if necessary :) =2D-=20 http://fossarchy.blogspot.com/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iQIcBAEBAgAGBQJLi8hTAAoJEJ0g9lA+M4iIkdoQALuzWWyi5pp/MzbpJFKa/ib1 TKayk4P/0wleINWY8c0pitOTx8i10MiSTXtGRzgosT1mExuzjCiQ1geSUXDiJLJe O70jxIjF66kBa7VjLndTm9bkesb9AbEOCB8mOPowI5P+0CoUJwS3NF/BN2mEoiUk 3AptVbXRbcSN+I+rBbEWLViX9qNWOMQJ54myh8YKr6m3C4His/74UCdftL0kSThd nsUxIow1mxy3O7ezz6n3gl0dCrUClNpAyNxsTA8a+JxRHIsf8tetc5QDERs7IbEm bqiuZqdy3ZojKt9pZGj4yJzgI3uxxfnr6cVyvl/WdYiqE22OvMDtNQ7/56mKMSn+ +DgmRF3IYIARRoD5aWo0BrSamem4LhG+yKmcGU20BwdhVYdmN4pggxRD08W2utcw +x7ZcSmP59y0V70TUl7eLHQHesFILe9TyFFEfCcKeaneyxSc386etSQ/4K+0pXjL sUlLbLQC4xIJoJ4TWf/mUAqOGzqZRRMbQyHUb+sr5EfhOrBsWlk17oZ46q3GLpd6 Z15fhEGcuA+t2ASkgs2+4RAXB8QKHU/XeeRetuYdU35xE1eP2n2RyaOhz8H0/c97 6z7ZT4hWeXLClxXzHrdzDoWLMdXVwy+uDePpw8OpNF7Vz+pl1uOS0SQJH8hB9t1j 9T3e3i7Py4gx4AFZ9qZB =WaXR -----END PGP SIGNATURE----- --=-=-=-- --===============1150211997== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1150211997==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:00:08 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21E03L2030296; Mon, 1 Mar 2010 15:00:04 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21E02ie030134 for ; Mon, 1 Mar 2010 15:00:02 +0100 Received: (qmail 15350 invoked by uid 501); 1 Mar 2010 14:00:02 -0000 Message-ID: <20100301140002.15349.qmail@stuge.se> Date: Mon, 1 Mar 2010 15:00:02 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <87ljecz3ho.fsf@mocca.josefsson.org> <87vddgxnfg.fsf@mocca.josefsson.org> <87k4twxmjd.fsf@mocca.josefsson.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <87k4twxmjd.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson wrote: > if someone declares the constant to 0 and expects the feature to be > disabled. The way to deal with this is usually #if FEATURE == 1 //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:09:18 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21E9BTZ008930; Mon, 1 Mar 2010 15:09:15 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21E99mA008925 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 15:09:09 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21E96ut026149 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 15:09:08 +0100 From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <20100301135518.14474.qmail@stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::Y1hSUC94my2y5EKa:80fs Date: Mon, 01 Mar 2010 15:09:06 +0100 In-Reply-To: <20100301135518.14474.qmail@stuge.se> (Peter Stuge's message of "Mon, 1 Mar 2010 14:55:18 +0100") Message-ID: <878wacuoal.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge writes: > Simon Josefsson wrote: >> > This is a patch to use OpenSSL's version of aes_*_ctr functions and to >> > not use DSA-code if not supported by OpenSSL >> >> Thanks. I applied your DSA-code patch as-is, > > Don't do this. DSA is required by RFC. Please have a look. What should libssh2 do if OpenSSL doesn't implement DSA? Refuse to build? /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:11:24 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21EBLVb010065; Mon, 1 Mar 2010 15:11:23 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21EBK2E010059 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 15:11:20 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21EBJfw026186 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 15:11:20 +0100 From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <87sk8kz5mu.fsf@mocca.josefsson.org> <20100301135651.14706.qmail@stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::dbdiSKs1N1H5gJv/:yv2 Date: Mon, 01 Mar 2010 15:11:19 +0100 In-Reply-To: <20100301135651.14706.qmail@stuge.se> (Peter Stuge's message of "Mon, 1 Mar 2010 14:56:51 +0100") Message-ID: <874ol0uo6w.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge writes: > Simon Josefsson wrote: >> It seems some OpenSSL have AES-CTR and some do not, > > I expressed an idea a while back for libssh2 to autodetect what > crypto is available and run with that. No need to decide at build > time, right? That should work, and sounds great. Patches welcome. :) However, I recall that OpenSSL can be built in ABI-incompatible ways depending on its build flags. So libssh2 would have to auto-detect this too, and use the right struct's depending on which libssl is available on the system. That may get messy. On the other hand, it may be that libssh2's usage of OpenSSL is so simple that we aren't affected by the ABI-incompatible stuff. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:14:30 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21EEPl8011107; Mon, 1 Mar 2010 15:14:28 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21EEO8i011102 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 15:14:24 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21EEMJx026284 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 15:14:24 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <4B8BC246.2030706@SDlabs.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::6U1y+v//wa2QoN2E:0QMRG Date: Mon, 01 Mar 2010 15:14:22 +0100 In-Reply-To: <4B8BC246.2030706@SDlabs.se> (Lars Nordin's message of "Mon, 01 Mar 2010 14:33:58 +0100") Message-ID: <87zl2st9hd.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Lars Nordin writes: > The internal LIBSSH2_AES_CTR shall NOT be used if openSSL supports > aes_ctr, there is nothing wrong in the patch. As far as I can tell, your patch would set LIBSSH2_AES_CTR to 0 if a new OpenSSL is used. In crypt.c, that would disable AES-CTR because of this code: static const LIBSSH2_CRYPT_METHOD *_libssh2_crypt_methods[] = { #if LIBSSH2_AES_CTR &libssh2_crypt_method_aes128_ctr, &libssh2_crypt_method_aes192_ctr, &libssh2_crypt_method_aes256_ctr, #endif /* LIBSSH2_AES */ > I have testcompiled and check (using nm) if the LIBSSH2_AES_ctr > function was used. I have also tested both versions using example/ssh2 > to a openssh server on localhost. Did you verify that AES-CTR was negotiated, instead of just AES? Anyway, please test libssh2 from git! /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:29:08 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21ET3mA017815; Mon, 1 Mar 2010 15:29:06 +0100 Received: from av9-2-sn2.hy.skanova.net (av9-2-sn2.hy.skanova.net [81.228.8.180]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21ET1cI017737 for ; Mon, 1 Mar 2010 15:29:01 +0100 Received: by av9-2-sn2.hy.skanova.net (Postfix, from userid 502) id 6DAD1384A1; Mon, 1 Mar 2010 15:28:57 +0100 (CET) Received: from smtp4-2-sn2.hy.skanova.net (smtp4-2-sn2.hy.skanova.net [81.228.8.93]) by av9-2-sn2.hy.skanova.net (Postfix) with ESMTP id 016DE384A0 for ; Mon, 1 Mar 2010 15:28:57 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp4-2-sn2.hy.skanova.net (Postfix) with ESMTP id D8FD737E4E for ; Mon, 1 Mar 2010 15:28:54 +0100 (CET) Message-ID: <4B8BCF27.7030902@SDlabs.se> Date: Mon, 01 Mar 2010 15:28:55 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <20100301135518.14474.qmail@stuge.se> In-Reply-To: <20100301135518.14474.qmail@stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 2010-03-01 14:55, Peter Stuge wrote: > Simon Josefsson wrote: > >>> This is a patch to use OpenSSL's version of aes_*_ctr functions and to >>> not use DSA-code if not supported by OpenSSL >>> >> Thanks. I applied your DSA-code patch as-is, >> > Don't do this. DSA is required by RFC. Please have a look. > > > //Peter > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > LIBSSH2 checks if DSA is supported by OpenSSL. If not LIBSSH2_DSA is set to 0. My patch was to have this #if LIBSSH2_DSA around the actual dsa-codem, in order to compile if used with an openssl-version without DSA SSH can be used without DSA, but will break the RFC. It's up to the enduser if he wants an SSH2 that break's the RFC (for me, the client i', building will only talk with a openssh-server compiled using the same OpenSSL-version (without DSA) /Lars _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:34:21 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21EYG7F020094; Mon, 1 Mar 2010 15:34:20 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21EYEnT020083 for ; Mon, 1 Mar 2010 15:34:14 +0100 Received: (qmail 21265 invoked by uid 501); 1 Mar 2010 14:34:14 -0000 Message-ID: <20100301143414.21264.qmail@stuge.se> Date: Mon, 1 Mar 2010 15:34:14 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: ./configure summary? Mail-Followup-To: libssh2-devel@cool.haxx.se References: <87hbp0z31f.fsf@mocca.josefsson.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <87hbp0z31f.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson wrote: > Any objections to pushing the patch below, so that when ./configure > has completed it will print something like this: I like it. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 15:54:38 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21EsTeI002891; Mon, 1 Mar 2010 15:54:36 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21EsTP8002883 for ; Mon, 1 Mar 2010 15:54:29 +0100 Received: (qmail 24950 invoked by uid 501); 1 Mar 2010 14:54:29 -0000 Message-ID: <20100301145429.24949.qmail@stuge.se> Date: Mon, 1 Mar 2010 15:54:29 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <20100301135518.14474.qmail@stuge.se> <878wacuoal.fsf@mocca.josefsson.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <878wacuoal.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson wrote: > >> > This is a patch to use OpenSSL's version of aes_*_ctr functions and to > >> > not use DSA-code if not supported by OpenSSL > >> > >> Thanks. I applied your DSA-code patch as-is, > > > > Don't do this. DSA is required by RFC. Please have a look. > > What should libssh2 do if OpenSSL doesn't implement DSA? Refuse to > build? Yep, I feel pretty strongly that it's the only sane thing to do. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 16:02:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21F1w4s008963; Mon, 1 Mar 2010 16:02:02 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21F1uLN008904 for ; Mon, 1 Mar 2010 16:01:57 +0100 Received: (qmail 26401 invoked by uid 501); 1 Mar 2010 15:01:57 -0000 Message-ID: <20100301150157.26400.qmail@stuge.se> Date: Mon, 1 Mar 2010 16:01:57 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <20100301135518.14474.qmail@stuge.se> <4B8BCF27.7030902@SDlabs.se> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <4B8BCF27.7030902@SDlabs.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Lars Nordin wrote: > SSH can be used without DSA, but will break the RFC. It's up to the > enduser if he wants an SSH2 that break's the RFC (for me, the > client i', building will only talk with a openssh-server compiled > using the same OpenSSL-version (without DSA) RFC compliance is really important for interoperability. In a case such as yours, where you explicitly do not want wide interoperability, non-compliance is of course fine. But I think we must not silently build a non-compliant library, so at the very least this needs to become a configure option; --disable-rfc-compliance maybe? The same knob might be used also for other things in the code. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 16:28:11 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21FRuNx031741; Mon, 1 Mar 2010 16:28:01 +0100 Received: from av9-2-sn3.vrr.skanova.net (av9-2-sn3.vrr.skanova.net [81.228.9.186]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21FRs6m031720 for ; Mon, 1 Mar 2010 16:27:54 +0100 Received: by av9-2-sn3.vrr.skanova.net (Postfix, from userid 502) id 4D5BC3856D; Mon, 1 Mar 2010 16:27:50 +0100 (CET) Received: from smtp3-1-sn3.vrr.skanova.net (smtp3-1-sn3.vrr.skanova.net [81.228.9.101]) by av9-2-sn3.vrr.skanova.net (Postfix) with ESMTP id 6348E38AFD for ; Mon, 1 Mar 2010 16:15:17 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp3-1-sn3.vrr.skanova.net (Postfix) with ESMTP id 963EA37E47 for ; Mon, 1 Mar 2010 16:15:09 +0100 (CET) Message-ID: <4B8BD9FA.9090305@SDlabs.se> Date: Mon, 01 Mar 2010 16:15:06 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <20100301135518.14474.qmail@stuge.se> <4B8BCF27.7030902@SDlabs.se> <20100301150157.26400.qmail@stuge.se> In-Reply-To: <20100301150157.26400.qmail@stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 2010-03-01 16:01, Peter Stuge wrote: > Lars Nordin wrote: > >> SSH can be used without DSA, but will break the RFC. It's up to the >> enduser if he wants an SSH2 that break's the RFC (for me, the >> client i', building will only talk with a openssh-server compiled >> using the same OpenSSL-version (without DSA) >> > RFC compliance is really important for interoperability. > > In a case such as yours, where you explicitly do not want wide > interoperability, non-compliance is of course fine. > > But I think we must not silently build a non-compliant library, so at > the very least this needs to become a configure option; > > --disable-rfc-compliance > > maybe? The same knob might be used also for other things in the code. > > > //Peter > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > It can be a better way, my usage is special and can have as much configure options as needed :-). My point was that LIBSSH2 already contained RFC-non-compliant code, what I did was to extend the usage to fully compile the code without references to DSA. So the patch should be applied and then the next discussion, should the non-DSA usage for LIBSSH2 have an extra configure option. We should discuss the right things, not patches to fully use an compile option, but the compile option. /Lars _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 16:28:37 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21FSZ27032059; Mon, 1 Mar 2010 16:28:36 +0100 Received: from av6-2-sn3.vrr.skanova.net (av6-2-sn3.vrr.skanova.net [81.228.9.180]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21FSYOA032022 for ; Mon, 1 Mar 2010 16:28:34 +0100 Received: by av6-2-sn3.vrr.skanova.net (Postfix, from userid 502) id 98E5538360; Mon, 1 Mar 2010 16:28:29 +0100 (CET) Received: from smtp3-1-sn3.vrr.skanova.net (smtp3-1-sn3.vrr.skanova.net [81.228.9.101]) by av6-2-sn3.vrr.skanova.net (Postfix) with ESMTP id 7522C3835B for ; Mon, 1 Mar 2010 16:28:29 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp3-1-sn3.vrr.skanova.net (Postfix) with ESMTP id 41DA037E65 for ; Mon, 1 Mar 2010 16:27:27 +0100 (CET) Message-ID: <4B8BDCDC.5020101@SDlabs.se> Date: Mon, 01 Mar 2010 16:27:24 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Missing functions like libssh2_[init|exit] X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se When looking into libssh2 and my intended usage, I'm missing a general place to initialize the OpenSSL-library. Today libssh2_crypto_init() is implicit called from libssh_session_init(). (and libssh2_crypto_init() is defined to an empty string in openssl.h, load of the crypto-function is implicit done in openssl.c when needed. I think there should exist a libssh2_init(), where all necessary initialized are done. Maybe to be backward compatible libssh2_init can be called in session_init() if sone configurations-option is used (--with implicit_init??) I don't know right now what shall be done in libssh2_exit, but an init-call should be ended with an exit-call. /Lars _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 16:40:51 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21FekDO005399; Mon, 1 Mar 2010 16:40:49 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Fei8V005332 for ; Mon, 1 Mar 2010 16:40:44 +0100 Date: Mon, 1 Mar 2010 16:40:44 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Missing functions like libssh2_[init|exit] In-Reply-To: <4B8BDCDC.5020101@SDlabs.se> Message-ID: References: <4B8BDCDC.5020101@SDlabs.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, Lars Nordin wrote: > I think there should exist a libssh2_init(), where all necessary initialized > are done. Maybe to be backward compatible libssh2_init can be called in > session_init() if sone configurations-option is used (--with > implicit_init??) > > I don't know right now what shall be done in libssh2_exit, but an init-call > should be ended with an exit-call. I too think we need to have an init function - as the crypto libs need global inits. It might even so so that the init functions that init OpenSSL or gcrypt are not even thread safe (iirc) so we need to document that properly - if that is indeed the case. The exit/cleanup function would do the reverse of the init function, as OpenSSL for example allocates data in their init calls that can be cleaned up afterwards when nothing else is using it anymore. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 17:03:30 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21G3Ipj023506; Mon, 1 Mar 2010 17:03:26 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21G3F4h023450 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 17:03:16 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21G3BeE029149 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 17:03:14 +0100 From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: Re: ./configure summary? References: <87hbp0z31f.fsf@mocca.josefsson.org> <20100301143414.21264.qmail@stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::5sENLvYQRpr5kMgs:FeoY Date: Mon, 01 Mar 2010 17:03:11 +0100 In-Reply-To: <20100301143414.21264.qmail@stuge.se> (Peter Stuge's message of "Mon, 1 Mar 2010 15:34:14 +0100") Message-ID: <87sk8km3ls.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge writes: > Simon Josefsson wrote: >> Any objections to pushing the patch below, so that when ./configure >> has completed it will print something like this: > > I like it. Applied. The output helps me quickly see if libssh2 is using OpenSSL or libgcrypt. If Daniel or someone else dislikes it strongly, we can revert it. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 17:08:48 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21G8imc030256; Mon, 1 Mar 2010 17:08:46 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21G8g8C030251 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 17:08:43 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21G8f51029256 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 17:08:42 +0100 From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: Re: Patch to use aes_*_ctr in newer versions of OpenSSL and #if LIBSSH2_DSA for more DSA-code References: <4B8B8323.80606@SDlabs.se> <87ljec1iwt.fsf@mocca.josefsson.org> <20100301135518.14474.qmail@stuge.se> <4B8BCF27.7030902@SDlabs.se> <20100301150157.26400.qmail@stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::7OOykof3C0wZZHpF:1Yp5 Date: Mon, 01 Mar 2010 17:08:41 +0100 In-Reply-To: <20100301150157.26400.qmail@stuge.se> (Peter Stuge's message of "Mon, 1 Mar 2010 16:01:57 +0100") Message-ID: <87ocj8m3cm.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge writes: > Lars Nordin wrote: >> SSH can be used without DSA, but will break the RFC. It's up to the >> enduser if he wants an SSH2 that break's the RFC (for me, the >> client i', building will only talk with a openssh-server compiled >> using the same OpenSSL-version (without DSA) > > RFC compliance is really important for interoperability. > > In a case such as yours, where you explicitly do not want wide > interoperability, non-compliance is of course fine. > > But I think we must not silently build a non-compliant library, so at > the very least this needs to become a configure option; > > --disable-rfc-compliance > > maybe? The same knob might be used also for other things in the code. I like this, although prefer one configure option per software configuration, i.e., --disable-rfc-compliancy-dsa. Just because one wants to disable DSA one might not want to automatically disable any other RFC-compliancy behaviour. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 17:11:54 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21GBo04032019; Mon, 1 Mar 2010 17:11:52 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21GBnF5032015 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 17:11:49 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21GBlcG029334 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 17:11:49 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Missing functions like libssh2_[init|exit] References: <4B8BDCDC.5020101@SDlabs.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::M6OBWwPMg11VMwls:qXda Date: Mon, 01 Mar 2010 17:11:48 +0100 In-Reply-To: (Daniel Stenberg's message of "Mon, 1 Mar 2010 16:40:44 +0100 (CET)") Message-ID: <87k4twm37f.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Mon, 1 Mar 2010, Lars Nordin wrote: > >> I think there should exist a libssh2_init(), where all necessary >> initialized are done. Maybe to be backward compatible libssh2_init >> can be called in session_init() if sone configurations-option is >> used (--with implicit_init??) >> >> I don't know right now what shall be done in libssh2_exit, but an >> init-call should be ended with an exit-call. > > I too think we need to have an init function - as the crypto libs need > global inits. It might even so so that the init functions that init > OpenSSL or gcrypt are not even thread safe (iirc) so we need to > document that properly - if that is indeed the case. > > The exit/cleanup function would do the reverse of the init function, > as OpenSSL for example allocates data in their init calls that can be > cleaned up afterwards when nothing else is using it anymore. +1. It is typical for shared libraries to have one init/done function for global (de-)initialization. This won't solve the libgcrypt thread mutex issue completely, though, because only the application knows what kind of thread library is in use, and it is the application that needs to configure libgcrypt properly. But this is the same as for any other application linking with anything that ultimately links to libgcrypt. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 17:18:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21GIOgJ002514; Mon, 1 Mar 2010 17:18:27 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21GINub002483 for ; Mon, 1 Mar 2010 17:18:23 +0100 Date: Mon, 1 Mar 2010 17:18:23 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: ./configure summary? In-Reply-To: <87sk8km3ls.fsf@mocca.josefsson.org> Message-ID: References: <87hbp0z31f.fsf@mocca.josefsson.org> <20100301143414.21264.qmail@stuge.se> <87sk8km3ls.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, Simon Josefsson wrote: > Applied. The output helps me quickly see if libssh2 is using OpenSSL or > libgcrypt. If Daniel or someone else dislikes it strongly, we can revert > it. me likes! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 20:57:34 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21Jv3Xh008424; Mon, 1 Mar 2010 20:57:13 +0100 Received: from mail.gmx.net (mail.gmx.net [213.165.64.20]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21Juvur008345 for ; Mon, 1 Mar 2010 20:56:58 +0100 Received: (qmail invoked by alias); 01 Mar 2010 19:51:08 -0000 Received: from 91-115-26-172.adsl.highway.telekom.at (EHLO [192.168.1.106]) [91.115.26.172] by mail.gmx.net (mp072) with SMTP; 01 Mar 2010 20:51:08 +0100 X-Authenticated: #1568940 X-Provags-ID: V01U2FsdGVkX1+l2vR8e1Ol+teLRiV5Zx3Ip+6U8r6oMFVXRRDicz 2rPfykP92d530x Message-ID: <4B8C1A83.9050205@gmx.at> Date: Mon, 01 Mar 2010 20:50:27 +0100 From: double User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: libssh2 development Subject: Preprocessor... X-Y-GMX-Trusted: 0 X-FuHaFi: 0.58999999999999997 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hello, In order to define the buffer-size, I would like to check the libssh2-version. What I did (simplifed): #if LIBSSH2_VERSION_MAJOR == 1 && LIBSSH2_VERSION_MINOR <= 2 && LIBSSH2_VERSION_PATCH <= 2 #define LIBSSH2_BUFFERSIZE 1024 #endif The problem, libssh 1.2: #define LIBSSH2_VERSION_MAJOR 1 #define LIBSSH2_VERSION_MINOR 2 #define LIBSSH2_VERSION_PATCH Of course, GCC reports: ssh.cpp:30:89: error: operator '&&' has no right operand Is there a chance to check, if LIBSSH2_VERSION_PATCH is empty? Thanks Marcus _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 21:22:59 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21KMrf0028957; Mon, 1 Mar 2010 21:22:57 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21KMqIL028943 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 1 Mar 2010 21:22:52 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o21KMoYE004478 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 1 Mar 2010 21:22:52 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Preprocessor... References: <4B8C1A83.9050205@gmx.at> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100301:libssh2-devel@cool.haxx.se::4sVhUWSndV+QcAFM:6/sK Date: Mon, 01 Mar 2010 21:22:50 +0100 In-Reply-To: <4B8C1A83.9050205@gmx.at> (double's message of "Mon, 01 Mar 2010 20:50:27 +0100") Message-ID: <873a0jn65h.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se double writes: > Hello, > > In order to define the buffer-size, I would like to check the > libssh2-version. > > What I did (simplifed): > > #if LIBSSH2_VERSION_MAJOR == 1 && LIBSSH2_VERSION_MINOR <= 2 && > LIBSSH2_VERSION_PATCH <= 2 > #define LIBSSH2_BUFFERSIZE 1024 > #endif > > The problem, libssh 1.2: > #define LIBSSH2_VERSION_MAJOR 1 > #define LIBSSH2_VERSION_MINOR 2 > #define LIBSSH2_VERSION_PATCH > > Of course, GCC reports: > ssh.cpp:30:89: error: operator '&&' has no right operand > > Is there a chance to check, if LIBSSH2_VERSION_PATCH is empty? Try: #if (LIBSSH2_VERSION_PATCH + LIBSSH2_VERSION_MINOR * 10 + LIBSSH2_VERSION_MAJOR * 100) >= 120 #define LIBSSH2_BUFFERSIZE 1024 #endif Or something like it. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 21:39:17 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21KdAsD006197; Mon, 1 Mar 2010 21:39:15 +0100 Received: from mail.gmx.net (mail.gmx.net [213.165.64.20]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o21Kd4xf006083 for ; Mon, 1 Mar 2010 21:39:05 +0100 Received: (qmail invoked by alias); 01 Mar 2010 20:31:27 -0000 Received: from 91-115-26-172.adsl.highway.telekom.at (EHLO [192.168.1.106]) [91.115.26.172] by mail.gmx.net (mp069) with SMTP; 01 Mar 2010 21:31:27 +0100 X-Authenticated: #1568940 X-Provags-ID: V01U2FsdGVkX1/Dvn5KynXDHtHv09vm6OMF8BOocvmLl75Hl7/B23 zqKwP5IWrIT3Kf Message-ID: <4B8C23F5.8070701@gmx.at> Date: Mon, 01 Mar 2010 21:30:45 +0100 From: double User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: libssh2 development Subject: Re: Preprocessor... References: <4B8C1A83.9050205@gmx.at> <873a0jn65h.fsf@mocca.josefsson.org> In-Reply-To: <873a0jn65h.fsf@mocca.josefsson.org> X-Y-GMX-Trusted: 0 X-FuHaFi: 0.68000000000000005 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson schrieb: > Try: > > #if (LIBSSH2_VERSION_PATCH + LIBSSH2_VERSION_MINOR * 10 + LIBSSH2_VERSION_MAJOR * 100) >= 120 > #define LIBSSH2_BUFFERSIZE 1024 > #endif > > Or something like it. > > /Simon > Hi Simon, Good idea! Thank you, it works. Marcus _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 1 22:20:15 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21LK1LC005961; Mon, 1 Mar 2010 22:20:13 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o21LJxPU005947 for ; Mon, 1 Mar 2010 22:19:59 +0100 Date: Mon, 1 Mar 2010 22:19:59 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Preprocessor... In-Reply-To: <4B8C1A83.9050205@gmx.at> Message-ID: References: <4B8C1A83.9050205@gmx.at> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 1 Mar 2010, double wrote: > #if LIBSSH2_VERSION_MAJOR == 1 && LIBSSH2_VERSION_MINOR <= 2 && > LIBSSH2_VERSION_PATCH <= 2 > #define LIBSSH2_BUFFERSIZE 1024 > #endif There's a much easier way. We have the LIBSSH2_VERSION_NUM define that lets you do simple comparisons like this: #if LIBSSH2_VERSION_NUM >= 0x010203 /* later or equal than 1.2.3 */ -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 2 15:43:49 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22EhZCP011293; Tue, 2 Mar 2010 15:43:45 +0100 Received: from mail-ew0-f224.google.com (mail-ew0-f224.google.com [209.85.219.224]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22EhYtj011183 for ; Tue, 2 Mar 2010 15:43:34 +0100 Received: by ewy24 with SMTP id 24so177810ewy.7 for ; Tue, 02 Mar 2010 06:43:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=+ckWAG7FTzYji8ZZ5D9u/UuBm+OVq9O3d64ODiP+d9U=; b=DEPtYUeuIiPHgg3QFsCCHj39KEvnzuxdTyiC799n/yoSePGCbbwPq7z+QcineRhG2K t49+obTClozpiLq04aCvT9TpM9uPKRtJyDiG+7DFGRUD97itMOsnXssJvqIrp6qiVqqv WW9P/8SAPWgKOyrWXAt/oOVYnaPVniUW2KddM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=CVnlGNed4oZKj2pIrAbI0hH0bx5PAL19u6GTfUxVLTqx4++Nc9B72+T1B9SIbH7Fp/ RUtg9haE9zYYF6C0zu9VbGEH0q4kPdM8piYWsOHmnhlGozqKOduxsIH+sshcmkiFttne JGmIOopZS7nzCe7+EhCM78rBM8g2tmyqHIh38= MIME-Version: 1.0 Received: by 10.213.1.151 with SMTP id 23mr1011145ebf.34.1267541010071; Tue, 02 Mar 2010 06:43:30 -0800 (PST) In-Reply-To: References: <20100301100559.2197.qmail@earth.stuge.se> Date: Tue, 2 Mar 2010 14:43:30 +0000 X-Google-Sender-Auth: 0be5374a84a0e303 Message-ID: <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 1 March 2010 10:08, Daniel Stenberg wrote: > On Mon, 1 Mar 2010, libssh2@git.stuge.se wrote: > >> =A0 use const to silence a bazillion warnings I left the errmsg parameter non-const as the string may be freed when the should_free parameter is true. Now that it's const, I get warnings in all the places that try to free it. Basically, whichever way we do this we're screwed as this function can't take both a const and a non-const parameter. Even providing two versions, one that frees and one that doesn't wont help as unless the session itself has one pointer for the const string and one for the freeable non-const string. Do we ever actually call libssh2_error with a string we want freed? I've not come across any but its tricky to search for. > My advice to committers: use configure --enable-debug to get a wide range= of > picky compiler options enabled. I'm developing on Windows but I'll try to build on Linux before pushing in future. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 2 19:37:47 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22IbXEH016386; Tue, 2 Mar 2010 19:37:43 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22IbWMo016376 for ; Tue, 2 Mar 2010 19:37:32 +0100 Date: Tue, 2 Mar 2010 19:37:32 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings In-Reply-To: <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> Message-ID: References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 2 Mar 2010, Alexander Lamaison wrote: > I left the errmsg parameter non-const as the string may be freed when the > should_free parameter is true. Now that it's const, I get warnings in all > the places that try to free it. Haha, I got the reverse. I got warnings all over when passing const strings to arguments not using const... > Do we ever actually call libssh2_error with a string we want freed? I've not > come across any but its tricky to search for. I did a search for it and I find two or three cases. I'm not at all convinced those rare cases are worth fighting for. Or perhaps I would even prefer having a static char array and then allow sprintf()-style arguments to make the error message. It would then of course be limited to a fixed length, like 80 or 120 characters or so. How would others feel about that? -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 2 21:58:22 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22KwAlw019097; Tue, 2 Mar 2010 21:58:19 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o22Kw8Nj019092 for ; Tue, 2 Mar 2010 21:58:08 +0100 Received: (qmail 23464 invoked by uid 501); 2 Mar 2010 20:58:08 -0000 Message-ID: <20100302205808.23463.qmail@stuge.se> Date: Tue, 2 Mar 2010 21:58:08 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings Mail-Followup-To: libssh2-devel@cool.haxx.se References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg wrote: > static char array .. > How would others feel about that? I like static, but since that is not reentrant it should take an optional pointer to a static array in the caller. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 2 22:01:04 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22L113v021808; Tue, 2 Mar 2010 22:01:02 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22L0x28021747 for ; Tue, 2 Mar 2010 22:00:59 +0100 Date: Tue, 2 Mar 2010 22:00:59 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings In-Reply-To: <20100302205808.23463.qmail@stuge.se> Message-ID: References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> <20100302205808.23463.qmail@stuge.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 2 Mar 2010, Peter Stuge wrote: >> How would others feel about that? > > I like static, but since that is not reentrant it should take an optional > pointer to a static array in the caller. Well, I meant static in the sense that it would be an array within the session struct, and we already before only held a single error message string per session so it wouldn't introduce any new restrictions really (other than the length restriction due to a fixed buffer size). -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 2 22:26:44 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o22LQcH8013049; Tue, 2 Mar 2010 22:26:40 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o22LQbdQ013045 for ; Tue, 2 Mar 2010 22:26:37 +0100 Received: (qmail 27353 invoked by uid 501); 2 Mar 2010 21:26:38 -0000 Message-ID: <20100302212638.27352.qmail@stuge.se> Date: Tue, 2 Mar 2010 22:26:38 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings Mail-Followup-To: libssh2-devel@cool.haxx.se References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> <20100302205808.23463.qmail@stuge.se> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg wrote: >> I like static, but since that is not reentrant it should take an >> optional pointer to a static array in the caller. > > Well, I meant static in the sense that it would be an array within > the session struct, If we want to move to thread safe API I don't think that's enough, but it would be simple to optionally push it onto the caller. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 3 09:46:33 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o238kJD7028141; Wed, 3 Mar 2010 09:46:27 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o238kIU7028125 for ; Wed, 3 Mar 2010 09:46:18 +0100 Date: Wed, 3 Mar 2010 09:46:18 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings In-Reply-To: <20100302212638.27352.qmail@stuge.se> Message-ID: References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> <20100302205808.23463.qmail@stuge.se> <20100302212638.27352.qmail@stuge.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 2 Mar 2010, Peter Stuge wrote: > If we want to move to thread safe API I don't think that's enough, but it > would be simple to optionally push it onto the caller. Yes, and I say we can care about those details once we've defined exactly what the thread safe API would be. I suggest we change libssh2_error() to become something like this: int libssh2_error(LIBSSH2_SESSION* session, int errcode, const char* msg, ...) { va_list ap; va_start(ap, msg); vsnprintf(session->err_msg_buffer, sizeof(session->err_msg_buffer), msg, ap); session->err_code = errcode; return errcode; } -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 3 10:02:20 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2392EAD004688; Wed, 3 Mar 2010 10:02:17 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2392CsP004602 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 3 Mar 2010 10:02:12 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o23927UT020366 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 3 Mar 2010 10:02:09 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100303:libssh2-devel@cool.haxx.se::Vcum6Os+dtz+CDl4:1dND Date: Wed, 03 Mar 2010 10:02:08 +0100 In-Reply-To: (Daniel Stenberg's message of "Tue, 2 Mar 2010 19:37:32 +0100 (CET)") Message-ID: <878wa9n5gv.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: >> Do we ever actually call libssh2_error with a string we want freed? >> I've not come across any but its tricky to search for. > > I did a search for it and I find two or three cases. I'm not at all > convinced those rare cases are worth fighting for. Or perhaps I would > even prefer having a static char array and then allow sprintf()-style > arguments to make the error message. It would then of course be > limited to a fixed length, like 80 or 120 characters or so. > > How would others feel about that? Are those 2-3 cases really that important to motive this work? I prefer to avoid dynamic error messages. In my experience dynamic error messages tend to be mostly useful to developers, and they can use debuggers to find out what the problem is. Normal users are not helped significantly by having some more details from libssh2, it is typically beyond what they can fix anyway. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 3 14:13:08 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23DCsAY030166; Wed, 3 Mar 2010 14:13:06 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23DCs1q030158 for ; Wed, 3 Mar 2010 14:12:54 +0100 Date: Wed, 3 Mar 2010 14:12:54 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings In-Reply-To: <878wa9n5gv.fsf@mocca.josefsson.org> Message-ID: References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> <878wa9n5gv.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 3 Mar 2010, Simon Josefsson wrote: > Are those 2-3 cases really that important to motive this work? I prefer to > avoid dynamic error messages. You're right. Let's just change the rare few places with fixed strings instead. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 3 14:16:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23DG8tj031746; Wed, 3 Mar 2010 14:16:10 +0100 Received: from mail-ew0-f224.google.com (mail-ew0-f224.google.com [209.85.219.224]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23DG7Gw031650 for ; Wed, 3 Mar 2010 14:16:07 +0100 Received: by ewy24 with SMTP id 24so12654ewy.7 for ; Wed, 03 Mar 2010 05:16:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=AAdRl/YynREV7aC1LFL7pTbaKCX2G+y2DnF/HT9+s3I=; b=v19lHo1GgJgsMakz/Nhk/lY6z7T8iw44AN5T5H6vZSabDwMISv0lDSGoRQYCO/OTx1 SeTIS+Q/FGwkOmUMPdWPBip1cGFT/rDc8KylGntJInTIFUL2RpbaJqagACJeaOPNn8wj cryIF6Yuizgv9g16moZTsCY7tjwHxG5e4NsXc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=CuFIpqluRsIccaOM3wk1JSzkb7bHtTUDzZFRJyCARRaLiH7el5PcsYtVDYRM84OBjS EHDcighJLJsIRP3GjjuYYHGObv0g0wsXG2qYDmPj+SFz8ZlEHSoBt0cHsRxA96kLi4Ff oqlE/WQ55VHvMAzSSQa3nqfVFfeEuvBx4JS+g= MIME-Version: 1.0 Received: by 10.213.1.15 with SMTP id 15mr2176928ebd.42.1267622163223; Wed, 03 Mar 2010 05:16:03 -0800 (PST) In-Reply-To: References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> <878wa9n5gv.fsf@mocca.josefsson.org> Date: Wed, 3 Mar 2010 13:16:03 +0000 X-Google-Sender-Auth: 15a0b3a23ecfe089 Message-ID: <9c3275a91003030516w4ca83529i9109bdc9c0d863d8@mail.gmail.com> Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 3 March 2010 13:12, Daniel Stenberg wrote: > On Wed, 3 Mar 2010, Simon Josefsson wrote: > >> Are those 2-3 cases really that important to motive this work? =A0I pref= er >> to avoid dynamic error messages. > > You're right. Let's just change the rare few places with fixed strings > instead. This gets my vote too. Very handy for converting to C++ exceptions which should always hold their message as a const char*. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 3 23:50:43 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23MoT5f018188; Wed, 3 Mar 2010 23:50:38 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23MoRZM018180 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 3 Mar 2010 23:50:27 +0100 Received: (qmail 24092 invoked from network); 3 Mar 2010 22:50:28 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 3 Mar 2010 22:50:28 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 03 Mar 2010 22:50:28 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163 Message-ID: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #163: use libssh2_socket_t for socket type --------------------+------------------------------------------------------- Reporter: sldr | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.5 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- In libssh2.h (line 250) in struct _LIBSSH2_POLLFD a socket is declared as int instead of libssh2_socket_t. I don't see an include for libssh2_priv.h where libssh2_socket_t is defined but that could be fixed too. The int instead of libssh2_socket_t causes a signed unsigned comparison in session.c in function libssh2_poll (code is: "if (fds[i].fd.socket > maxfd)"). -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 4 00:01:25 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23N1Kce028413; Thu, 4 Mar 2010 00:01:23 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23N1Ipn028400 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 4 Mar 2010 00:01:18 +0100 Received: (qmail 24771 invoked from network); 3 Mar 2010 23:01:18 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 3 Mar 2010 23:01:18 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 03 Mar 2010 23:01:18 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163#comment:1 Message-ID: <049.c177e6db33331bb7eab81d8af42258a1@libssh2.stuge.se> References: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 In-Reply-To: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se IzE2MzogdXNlIGxpYnNzaDJfc29ja2V0X3QgZm9yIHNvY2tldCB0eXBlCi0tLS0tLS0tLS0tLS0t LS0tLS0tKy0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0NCiBSZXBvcnRlcjogIHNsZHIgICAgfCAgICAgICBPd25lcjogICAgICAgDQogICAgIFR5 cGU6ICBkZWZlY3QgIHwgICAgICBTdGF0dXM6ICBuZXcgIA0KIFByaW9yaXR5OiAgbm9ybWFsICB8 ICAgTWlsZXN0b25lOiAgICAgICANCkNvbXBvbmVudDogIEFQSSAgICAgfCAgICAgVmVyc2lvbjog IDEuMi41DQogS2V5d29yZHM6ICAgICAgICAgIHwgICAgQmxvY2tpbmc6ICAgICAgIA0KQmxvY2tl ZGJ5OiAgICAgICAgICB8ICANCi0tLS0tLS0tLS0tLS0tLS0tLS0tKy0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0KQ2hhbmdlcyAoYnkgc2xkcik6 CgogKiBjYzogc2xkcjIyQOKApiAoYWRkZWQpDQoKCkNvbW1lbnQ6CgogVGhpcyBoYXBwZW5zIG9u IHRoZSBXaW5kb3dzIHBsYXRmb3JtIHNpbmNlIGEgc29ja2V0IGlzIGRlZmluZWQgYXMgYW4NCiB1 bnNpZ25lZCBpbnQgKG9yIHVuc2lnbmVkIF9faW50NjQgZm9yIDY0IGJpdCB3aW5kb3dzKS4KCi0t IApUaWNrZXQgVVJMOiA8aHR0cDovL2xpYnNzaDIuc3R1Z2Uuc2UvdGlja2V0LzE2MyNjb21tZW50 OjE+CmxpYnNzaDIgPGh0dHA6Ly9saWJzc2gyLnN0dWdlLnNlLz4KQyBsaWJyYXJ5IGZvciB3cml0 aW5nIHBvcnRhYmxlIFNTSDIgY2xpZW50cwpfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fXwpsaWJzc2gyLWRldmVsIGh0dHA6Ly9jb29sLmhheHguc2UvY2dpLWJp bi9tYWlsbWFuL2xpc3RpbmZvL2xpYnNzaDItZGV2ZWwK From libssh2-devel-bounces@cool.haxx.se Thu Mar 4 00:15:43 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23NFcIk005103; Thu, 4 Mar 2010 00:15:41 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o23NFaav005088 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 4 Mar 2010 00:15:36 +0100 Received: (qmail 25201 invoked from network); 3 Mar 2010 23:15:37 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 3 Mar 2010 23:15:37 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 03 Mar 2010 23:15:37 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163#comment:2 Message-ID: <049.2824dc81c7036aea226e3adba6cd878b@libssh2.stuge.se> References: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 In-Reply-To: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #163: use libssh2_socket_t for socket type --------------------+------------------------------------------------------- Reporter: sldr | Owner: bagder Type: defect | Status: assigned Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- Changes (by bagder): * owner: => bagder * status: new => assigned * version: 1.2.5 => 1.2.4 Comment: It can't include libssh2_priv.h since that is private and libssh2.h is a public header file. In general I think we've left the sockets as ints in the public prototypes just by being conservative in what we change. libssh2_poll() is marked "deprecated" anyway so fixing it isn't really worth it IMHO. libssh2_session_startup() however, is also using int for a socket and that is a function I think could be considered for fixing. We can't change it to 64 bits though since that would break the ABI. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 4 01:57:33 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o240vLTq000779; Thu, 4 Mar 2010 01:57:30 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o240vKEQ000774 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 4 Mar 2010 01:57:20 +0100 Received: (qmail 1415 invoked from network); 4 Mar 2010 00:57:21 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 4 Mar 2010 00:57:21 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Thu, 04 Mar 2010 00:57:21 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163#comment:3 Message-ID: <049.533ca264ef5733aca55468eb080a3c96@libssh2.stuge.se> References: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 In-Reply-To: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #163: use libssh2_socket_t for socket type --------------------+------------------------------------------------------- Reporter: sldr | Owner: bagder Type: defect | Status: assigned Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- Comment(by sldr): What if the following (or something like it) was put in libssh2.h near top: #ifdef WIN32 typedef SOCKET libssh2_socket_t; #else /* !WIN32 */ typedef int libssh2_socket_t; #define INVALID_SOCKET -1 #endif /* WIN32 */ Change libssh2_session_startup to: LIBSSH2_API int libssh2_session_startup(LIBSSH2_SESSION *session, libssh2_socket_t sock) Also change the struct _LIBSSH2_POLLFD to: typedef struct _LIBSSH2_POLLFD { unsigned char type; /* LIBSSH2_POLLFD_* below */ union { libssh2_socket_t socket; /* File descriptors -- examined with system select() call */ The only problems I see are: 1. If non-windows platforms define WIN32 for some reason but I don't thing this is a issue. 2. On windows you will need to include WinSock2.h (or windows.h, etc.) so that SOCKET is defined or define it manually before libssh2.h is included. I think any code using the libssh2.h include file is going to need to do some type of include to get a socket type defined anyway on Windows platforms. The use of SOCKET also means that libssh2 has a better chance of working on a 64bit compile on Windows (I know that LOTS of other things would need to be checked too). Thanks for your time and consideration, SLDR (Stephen L. De Rudder) -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 4 02:07:55 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2417ndw013449; Thu, 4 Mar 2010 02:07:53 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2417lnK013422 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 4 Mar 2010 02:07:47 +0100 Received: (qmail 2047 invoked from network); 4 Mar 2010 01:07:48 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 4 Mar 2010 01:07:48 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Thu, 04 Mar 2010 01:07:48 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163#comment:4 Message-ID: <049.f1fd0fe86f4401862a5dfaee3d918d16@libssh2.stuge.se> References: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 In-Reply-To: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #163: use libssh2_socket_t for socket type --------------------+------------------------------------------------------- Reporter: sldr | Owner: bagder Type: defect | Status: assigned Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- Comment(by roadrunn): I may not be doing things correctly, but I cast my socket type to int when calling libssh2_session_startup. This works in all (Windows x86/IA64, Linux i386/ia64, and HP/UX). -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 4 02:27:55 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o241RmcF028138; Thu, 4 Mar 2010 02:27:53 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o241Rkgn028130 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 4 Mar 2010 02:27:46 +0100 Received: (qmail 2928 invoked from network); 4 Mar 2010 01:27:47 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 4 Mar 2010 01:27:47 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Thu, 04 Mar 2010 01:27:47 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163#comment:5 Message-ID: <049.0be930fe4e8b879218643578c4b14ba7@libssh2.stuge.se> References: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 In-Reply-To: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #163: use libssh2_socket_t for socket type --------------------+------------------------------------------------------- Reporter: sldr | Owner: bagder Type: defect | Status: assigned Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- Comment(by sldr): Replying to [comment:4 roadrunn]: > I may not be doing things correctly, but I cast my socket type to int when calling libssh2_session_startup. This works in all (Windows x86/IA64, Linux i386/ia64, and HP/UX). With a 64bit program if you open enough sockets and/or handles (since a socket is a handle on windows), you could get a value that wouldn't fit in 32bits so cast would cause failure in libssh2 when it was finally passed it to a OS API (ie: Win32 API). This doesn't happen very often since most programs don't open lots of handles and sockets that might cause the value to get large enough that it doesn't fit in 32bits. FYI: Ignore the 32 in Win32 because the 64bit APIs are also called Win32 APIs except when you are running in WOW64 witch means they are 32bit APIs. :-) (bad windows joke) -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 4 09:06:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2485vnt015113; Thu, 4 Mar 2010 09:06:08 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2485tcX015067 for ; Thu, 4 Mar 2010 09:05:55 +0100 Date: Thu, 4 Mar 2010 09:05:55 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master 3420e98... use const to silence a bazillion warnings In-Reply-To: <9c3275a91003030516w4ca83529i9109bdc9c0d863d8@mail.gmail.com> Message-ID: References: <20100301100559.2197.qmail@earth.stuge.se> <9c3275a91003020643r7b90e038s3fa5f3ed5b234ff0@mail.gmail.com> <878wa9n5gv.fsf@mocca.josefsson.org> <9c3275a91003030516w4ca83529i9109bdc9c0d863d8@mail.gmail.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 3 Mar 2010, Alexander Lamaison wrote: >> You're right. Let's just change the rare few places with fixed strings >> instead. > > This gets my vote too. Very handy for converting to C++ exceptions which > should always hold their message as a const char*. In case someone missed it: I've pushed a rather big fix that implemented this simpler approach. I hope I didn't mess up anywhere, but I ran a few basic tests before I sent it off and it worked fine for me. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sat Mar 6 03:12:23 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o262C715021327; Sat, 6 Mar 2010 03:12:18 +0100 Received: from mail-fx0-f225.google.com (mail-fx0-f225.google.com [209.85.220.225]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o262C5Bp021249 for ; Sat, 6 Mar 2010 03:12:05 +0100 Received: by fxm25 with SMTP id 25so1312036fxm.7 for ; Fri, 05 Mar 2010 18:12:02 -0800 (PST) MIME-Version: 1.0 Received: by 10.239.157.10 with SMTP id o10mr114065hbc.84.1267841521740; Fri, 05 Mar 2010 18:12:01 -0800 (PST) In-Reply-To: <4B8867C7.6050807@qsl.net> References: <4B87E2F0.8010401@qsl.net> <50cb0cb61002250615q1cb5684pd7d6ab0f9b64e6cb@mail.gmail.com> <0A050DAD-DF9D-458F-BE7D-831D8C3528B2@intersystems.com> <4B8740A0.2070409@qsl.net> <50cb0cb61002260625m139ce81dpc36194c52fdaf7e@mail.gmail.com> <20100226163657.20114.qmail@stuge.se> <4B8867C7.6050807@qsl.net> Date: Sat, 6 Mar 2010 03:12:01 +0100 Message-ID: <50cb0cb61003051812m4526dbfi47f7f2a5578a8a41@mail.gmail.com> Subject: Re: Port to VMS; how to get changes into the main code stream? From: Jose Baars To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hi, I have made some build procedures, and tested all examples. I decided not to go the GNV way, as John suggested, the main challenge to be solved are building a shareable library with both upper case and mixed (in the case of libssh2 lower) case procedure entries, GNU doesn't really help there. Two questions there : - Are the _libssh2_xxxx routines (the ones starting with underscore) supposed to be callable from programs linked against libssh2? Are they not supposed to, but is it offered as an option for the programmer, or are they strictly for internal use in libssh2 itself? If they are for internal use only, as I expect, I will not make them callable in the shared image library (the VMS dll, for anyone still reading this, but without VMS background). - Will there be more functions defined in the libssh2 API? Will the parameters of functions be subject to change? In other words: To what extent will libssh2 be upward compatible? I tested all examples. SCP is ssh1 so I didn't test that extensively, but it worked. The number of systems actually supporting SSH 1 is getting less and less, and on the standard SSH server on VMS it is not supported at all. The only example failing is the sftp_rw_nonblock.c, but after examining the code that seemed logical, as the non-blocking socket simply doesn't have data when it is read, and a select() call like in sftp_nonblock.c is not in place. To me, it fails by design. Am I wrong there? Thanks Jose _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sat Mar 6 03:34:45 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o262YdW8010127; Sat, 6 Mar 2010 03:34:42 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o262Yc4s010121 for ; Sat, 6 Mar 2010 03:34:38 +0100 Received: (qmail 26801 invoked by uid 501); 6 Mar 2010 02:34:39 -0000 Message-ID: <20100306023439.26800.qmail@stuge.se> Date: Sat, 6 Mar 2010 03:34:39 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Port to VMS; how to get changes into the main code stream? Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B87E2F0.8010401@qsl.net> <50cb0cb61002250615q1cb5684pd7d6ab0f9b64e6cb@mail.gmail.com> <0A050DAD-DF9D-458F-BE7D-831D8C3528B2@intersystems.com> <4B8740A0.2070409@qsl.net> <50cb0cb61002260625m139ce81dpc36194c52fdaf7e@mail.gmail.com> <20100226163657.20114.qmail@stuge.se> <4B8867C7.6050807@qsl.net> <50cb0cb61003051812m4526dbfi47f7f2a5578a8a41@mail.gmail.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <50cb0cb61003051812m4526dbfi47f7f2a5578a8a41@mail.gmail.com> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Jose Baars wrote: > I have made some build procedures, and tested all examples. Sounds great! > - Are the _libssh2_xxxx routines (the ones starting with underscore) .. > strictly for internal use in libssh2 itself? Yes. Strictly for internal use. Everything that should be in the public API is in the public include files in the source tree: include/libssh2.h, include/libssh2_publickey.h and include/libssh2_sftp.h. Everything outside those headers is off limits for applications. > If they are for internal use only, as I expect, I will not > make them callable in the shared image library (the VMS dll, for > anyone still reading this, but without VMS background). Sounds good! > - Will there be more functions defined in the libssh2 API? Yes, new functions can and will be added. > Will the parameters of functions be subject to change? There are a number of incompatible API changes coming, but there is no plan made for them. In Linux and other Unix-like systems dynamic libraries have a version called soname, and the linker knows which version of the API each application requires, so it's possible to have more than one library installed without conflict, and the linker will always do the right thing. The API changes remain unplanned because even though there is versioning, the changes will still create hassle for application developers, so we try to gather up some fair number of changes and do them all at once, and we want to do them only seldom, to not annoy application developers too much. > In other words: To what extent will libssh2 be upward compatible? The API will change at some point, and backwards compatibility may break. I don't know if and how that can be dealt with for VMS? > SCP is ssh1 so I didn't test that extensively, but it worked. SCP can run also on SSH 2, it's a separate program running on the server, and doesn't even care which version SSH it's running on. SFTP is only SSH 2 though. > The only example failing is the sftp_rw_nonblock.c, but after > examining the code that seemed logical, as the non-blocking socket > simply doesn't have data when it is read, and a select() call like > in sftp_nonblock.c is not in place. > To me, it fails by design. Am I wrong there? I have't looked now but I'm sure you are right. The examples are not by any means spectacular quality code. All patches are welcome. :) Thanks! //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sat Mar 6 19:53:45 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o26IrWgJ000808; Sat, 6 Mar 2010 19:53:41 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o26IrU4r000804 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sat, 6 Mar 2010 19:53:31 +0100 Received: (qmail 12555 invoked from network); 6 Mar 2010 18:53:31 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 6 Mar 2010 18:53:31 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Sat, 06 Mar 2010 18:53:31 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #163: use libssh2_socket_t for socket type X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/163#comment:6 Message-ID: <049.91ce6421b3c9ff9e633dfe5827344504@libssh2.stuge.se> References: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-Trac-Ticket-ID: 163 In-Reply-To: <040.95d30fa745e3202b7013b4cef3185cc7@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #163: use libssh2_socket_t for socket type ----------------------+----------------------------------------------------- Reporter: sldr | Owner: bagder Type: defect | Status: closed Priority: normal | Milestone: Component: API | Version: 1.2.4 Resolution: wontfix | Keywords: Blocking: | Blockedby: ----------------------+----------------------------------------------------- Changes (by bagder): * status: assigned => closed * resolution: => wontfix Comment: Since changing the type will change the ABI, that will not happen until we bump the soname (and break the ABI). This is now mentioned in the TODO. I close this bug as 'wontfix', meaning it will not be done at present time but it will happen at some point in the future. Thanks. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sat Mar 6 21:17:59 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o26KHlQ6026538; Sat, 6 Mar 2010 21:17:56 +0100 Received: from mail-ew0-f226.google.com (mail-ew0-f226.google.com [209.85.219.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o26KHjUl026514 for ; Sat, 6 Mar 2010 21:17:46 +0100 Received: by ewy26 with SMTP id 26so2927497ewy.3 for ; Sat, 06 Mar 2010 12:17:42 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=m6j2DpUSky8lrtLKXMp3WsYvch+JP7EYm43AJyqstIE=; b=sX/KFEtUSQ/wtEWoD3c6E0/6y825HWjPHqihdO8CjDgg2JBIsQTCqej65nnt6E1uR5 qhyd85PtaPRKnG+LUlVHMQlFey8WVc/SXRXVnjqKIVfKxTBoe8WEd//mfxPa4jqLKZTH eEUqV8SV+QIFP+GpABQdy6VwtpIqPPv2+EAZ4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=ULC4z1n3/TcR08l8BPN/L2dNjwixhGBTFpDY92tSoCcqQLK+La6PxRBe/dtIWLiKG/ hzwPAnXpYVDFnD/F2LQJflBc1AVYFGQoBhn3EFWpE3VWmA5VR5UK+h+oSlsPshpS5i7f C5C5GzeoeoZ/LMvb09ar0wuUL6f3aQx+Cev68= MIME-Version: 1.0 Received: by 10.213.109.217 with SMTP id k25mr1780465ebp.88.1267906661749; Sat, 06 Mar 2010 12:17:41 -0800 (PST) In-Reply-To: <9c3275a91003061200v282ad60jd9b08c3a9dc13b4d@mail.gmail.com> References: <9c3275a91003061200v282ad60jd9b08c3a9dc13b4d@mail.gmail.com> Date: Sat, 6 Mar 2010 20:17:41 +0000 X-Google-Sender-Auth: 6d5ff807cded3c22 Message-ID: <9c3275a91003061217x7cb96753y6b233889a749dadf@mail.gmail.com> Subject: Fwd: libssh2_knownhost_check key format From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Am I misinterpreting the libssh2_knownhost_check API? I'm trying to display the old and new host-keys to my users when there is a mismatch and so I looked at what curl does. =A0However, I'm not convinced it's using the API correctly. =A0It appears to try to use binary data as NULL-terminated string: =A0 =A0 =A0keycheck =3D libssh2_knownhost_check(sshc->kh, =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 conn->host.name, =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 remotekey, keylen, =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 LIBSSH2_KNOWNHOST_TYPE_PLAIN| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 LIBSSH2_KNOWNHOST_KEYENC_RAW| =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 keybit, =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 &host); The key held by 'host' here should be in binary format because LIBSSH2_KNOWNHOST_KEYENC_RAW was used. =A0But now curl tries to treat it like a NULL-terminated string: =A0 =A0 =A0infof(data, "SSH host check: %d, key: %s\n", keycheck, =A0 =A0 =A0 =A0 =A0 =A0(keycheck <=3D LIBSSH2_KNOWNHOST_CHECK_MISMATCH)? =A0 =A0 =A0 =A0 =A0 =A0host->key:""); As the key will always start with '\0' this would lead to no key showing up in the final string. =A0Then it stores the key in a struct whose specification says "points to a zero-terminated string encoded with base64 if len is zero, otherwise to the raw data": =A0 =A0 =A0/* setup 'knownkey' */ =A0 =A0 =A0if(keycheck <=3D LIBSSH2_KNOWNHOST_CHECK_MISMATCH) { =A0 =A0 =A0 =A0knownkey.key =3D host->key; =A0 =A0 =A0 =A0knownkey.len =3D 0; =A0 =A0 =A0 =A0knownkey.keytype =3D (keytype =3D=3D LIBSSH2_HOSTKEY_TYPE_RS= A)? =A0 =A0 =A0 =A0 =A0CURLKHTYPE_RSA : CURLKHTYPE_DSS; =A0 =A0 =A0 =A0knownkeyp =3D &knownkey; =A0 =A0 =A0} So clearly this code is assuming again that the key is a NULL-terminated base64-encoded string which AFAICS it isn't. This may be a curl bug but I'm posting it here first in case I'm misinterpreting the knownhost API. =A0Am I? Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sat Mar 6 22:30:40 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o26LUTrb013961; Sat, 6 Mar 2010 22:30:38 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o26LURp8013954 for ; Sat, 6 Mar 2010 22:30:27 +0100 Date: Sat, 6 Mar 2010 22:30:27 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Fwd: libssh2_knownhost_check key format In-Reply-To: <9c3275a91003061217x7cb96753y6b233889a749dadf@mail.gmail.com> Message-ID: References: <9c3275a91003061200v282ad60jd9b08c3a9dc13b4d@mail.gmail.com> <9c3275a91003061217x7cb96753y6b233889a749dadf@mail.gmail.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-1470944514-1157623280-1267911027=:389" X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---1470944514-1157623280-1267911027=:389 Content-Type: TEXT/PLAIN; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: 8BIT On Sat, 6 Mar 2010, Alexander Lamaison wrote: First, let me just say that it was a while since I did these things so I'm not completely up to speed with them. But I did write both the libssh2 side and the libcurl code using it so whatever mistakes are there, I made them! ;-) >      keycheck = libssh2_knownhost_check(sshc->kh, >                                         conn->host.name, >                                         remotekey, keylen, >                                         LIBSSH2_KNOWNHOST_TYPE_PLAIN| >                                         LIBSSH2_KNOWNHOST_KEYENC_RAW| >                                         keybit, >                                         &host); > > The key held by 'host' here should be in binary format because > LIBSSH2_KNOWNHOST_KEYENC_RAW was used. I believe the 'typemask' as is set by the function above is for the key passed in to the function. That's the 'remotekey' argument. It doesn't specify what format the info about the existing known host key is using. 'host' is a pointer to a struct that gets returned by libssh2_knownhost_check() and that returns a 'struct libssh2_knownhost' which in the public libssh2.h header has its 'key' field documented as "key in base64/printable format". > As the key will always start with '\0' this would lead to no key showing up > in the final string. I'd say this is a bug on the libssh2 side. It is documented and intended to only "export" the key as base64-encoded or an otherwise prinatable format, which should be fine to print as zero-terminated. I see how that the knownhost_to_external() function makes no effort to make sure that is the case, and that seems wrong. Or maybe the problem is earlier. How does it end up kept in memory raw in the first place? Iirc, the code keeps the data base64-encoded in memory. (FYI: I'll be away the upcoming week so I'll be slower to respond to mails.) -- / daniel.haxx.se ---1470944514-1157623280-1267911027=:389 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel ---1470944514-1157623280-1267911027=:389-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 7 03:20:03 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o272Jj4T004328; Sun, 7 Mar 2010 03:19:57 +0100 Received: from Encompasserve.org (Eisner.encompasserve.org [67.53.90.116]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o272Jfx7004259 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 7 Mar 2010 03:19:43 +0100 Received: from CONVERSION-DAEMON.Encompasserve.org by Encompasserve.org (PMDF V6.3-x13 #31732) id <01NKFWP01E28006PLY@Encompasserve.org> for libssh2-devel@cool.haxx.se; Sat, 06 Mar 2010 20:19:41 -0600 (CST) Received: from [192.168.0.31] (173-23-187-137.client.mchsi.com [173.23.187.137]) by Encompasserve.org (PMDF V6.3-x13 #31732) with ESMTPSA id <01NKFWOXUVJS007MKZ@Encompasserve.org> for libssh2-devel@cool.haxx.se; Sat, 06 Mar 2010 20:19:40 -0600 (CST) Date: Sat, 06 Mar 2010 20:19:37 -0600 From: "John E. Malmberg" Subject: Re: Port to VMS; how to get changes into the main code stream? In-reply-to: <20100306023439.26800.qmail@stuge.se> To: libssh2-devel@cool.haxx.se Message-id: <4B930D39.7040002@qsl.net> MIME-version: 1.0 User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) References: <4B87E2F0.8010401@qsl.net> <50cb0cb61002250615q1cb5684pd7d6ab0f9b64e6cb@mail.gmail.com> <0A050DAD-DF9D-458F-BE7D-831D8C3528B2@intersystems.com> <4B8740A0.2070409@qsl.net> <50cb0cb61002260625m139ce81dpc36194c52fdaf7e@mail.gmail.com> <20100226163657.20114.qmail@stuge.se> <4B8867C7.6050807@qsl.net> <50cb0cb61003051812m4526dbfi47f7f2a5578a8a41@mail.gmail.com> <20100306023439.26800.qmail@stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge wrote: > Jose Baars wrote: > >> - Will there be more functions defined in the libssh2 API? > > Yes, new functions can and will be added. You can leave spare symbol vector entries at the bottom of the list. Also generally you can always add symbol vectors. >> Will the parameters of functions be subject to change? > > There are a number of incompatible API changes coming, but there is > no plan made for them. In Linux and other Unix-like systems dynamic > libraries have a version called soname, and the linker knows which > version of the API each application requires, so it's possible to > have more than one library installed without conflict, and the linker > will always do the right thing. With VMS, you use a logical name to designate where the shared image is for any libraries not stored in SYS$SHARE:. The general convention is that third party libraries like this are not put in SYS$SHARE: A different image name would be used for different incompatible revisions of a shared image. A startup script is added to the system startup that installs the shared image and defines the logical name in the SYSTEM EXECUTIVE table. System Executive names can be accessed by privileged programs. I have registered the prefix GNV$ for the logical names installed in the gnv directory tree. Per HP conventions, all system logical names used by layered products are required to have registered prefixes. If a product is setting an un-prefixed logical name, that is a bug. Also the xxx$ in front of filenames also need to have the xxx registered. Again, I have gnv$ registered for use with ported programs. By HP rules, the GNV directories should be in GNV$GNU: not GNU: which the current GNV kit uses. The startup script needs to look for both names until HP fixes the GNV kit. A program does not have to be built with GNV to use the GNV directories to hold it. The GNV directory tree is a convenient one to use to match Unix conventions. I would recommend having a PCSI kit install and if needed create the GNV$GNU: directory tree, and to put everything under GNV$GNU:[usr] which will be seen by Unix programs as "/usr" Also please include a VMSified xxxx.pc file if the libssh2 project has one, and any of the documentation files in the same directories that they would be on a Unix system. One benefit of using the GNV style build is that you can run the UNIX install procedure in a test mode to identify what files should be in the binary kit and where they should be. Otherwise, you have to figure that out some other way. >> In other words: To what extent will libssh2 be upward compatible? > > The API will change at some point, and backwards compatibility may > break. I don't know if and how that can be dealt with for VMS? One way is to use different names for the shared image for the newer incompatible version. The way that most shared images on VMS use is to use a different "Universal" symbol name for the new image, typically by adding a prefix to it. A Universal symbol is one that is visible outside the shared image. The old entry point remains the same for existing programs to use, but the header files will set a prefix to the new routines for newly built images. If new parameters are added, then the va args method can be used. VMS calling standard allows a routine to be changed from a fixed number of arguments to one that takes additional optional arguments, as long as the routine understands that. Direct references to data in a VMS shared image is a pain, because it is difficult to maintain backwards compatibility if the size of the data changes. The trick that I have seen to do this is to only use functions that return a pointer to that data. Then the header files use macros that will change direct references to the data to call the routines. Please start a libssh2 note in the PORTING_TO_VMS conference about your project. Once you get a PCSI kit built, I can get it added to the FTP repository there. I have not been signing the PCSI kits with certificates. I have not seen if I could qualify for one of the free ones from godaddy, and they only last a year, so I would have to re-sign them. I have been considering using a self-signed one, but I do not see how that would really add more security. Regards, -John _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 7 17:47:52 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27GlZTu018082; Sun, 7 Mar 2010 17:47:48 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.26]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27GlY95018079 for ; Sun, 7 Mar 2010 17:47:34 +0100 Received: by ey-out-2122.google.com with SMTP id 9so736190eyd.31 for ; Sun, 07 Mar 2010 08:47:35 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=fYm7YBUEtYQA3bj031J+AoV16MJpSWYdGRPaHGgHACQ=; b=Jvop0xtgaZAzN6WeIi1+cgCttTa0lGk1aSEqN9r+6jp88t/Q7brNcEh3g9okrqIJ2V H5VnPCp9Nya928DoPh40rIzVhQQIIMCeDdr/Rn1sDnGMj76eOD2N7gXtJBGBAnh15cY5 gyVha/uLANsg5zgFoKCpzmWZrcsqvG0VVr86E= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=cV7QBmxSTSpBIN1n41DB4QPGNe+nkxTaK1d5J1uR0TGEsi04KPhQmm+pugPKNArOVK 5GbtwHUWTrYhjQxeZ09y0k9f7QFBhAztKl3yD1CBmAgplX2xel4RE7XBJC2YQiHsCZy5 pEBmxCwwpPfVe+bbYnE8QnL37/RHJZZ7qMBYM= MIME-Version: 1.0 Received: by 10.213.39.199 with SMTP id h7mr2234535ebe.45.1267980453789; Sun, 07 Mar 2010 08:47:33 -0800 (PST) In-Reply-To: References: <9c3275a91003061200v282ad60jd9b08c3a9dc13b4d@mail.gmail.com> <9c3275a91003061217x7cb96753y6b233889a749dadf@mail.gmail.com> Date: Sun, 7 Mar 2010 16:47:33 +0000 X-Google-Sender-Auth: 344e381d7c525e18 Message-ID: <9c3275a91003070847h413a3313obf0a95b40ea83bf3@mail.gmail.com> Subject: Re: Fwd: libssh2_knownhost_check key format From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 6 March 2010 21:30, Daniel Stenberg wrote: > On Sat, 6 Mar 2010, Alexander Lamaison wrote: >> =A0 =A0 =A0keycheck =3D libssh2_knownhost_check(sshc->kh, >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 conn->host.name, >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 remotekey, keylen, >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 LIBSSH2_KNOWNHOST_TYPE_PLAIN| >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 LIBSSH2_KNOWNHOST_KEYENC_RAW| >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 keybit, >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 &host); >> >> The key held by 'host' here should be in binary format because >> LIBSSH2_KNOWNHOST_KEYENC_RAW was used. > > I believe the 'typemask' as is set by the function above is for the key > passed in to the function. That's the 'remotekey' argument. It doesn't > specify what format the info about the existing known host key is using. Ah, you're quite right. > 'host' is a pointer to a struct that gets returned by > libssh2_knownhost_check() and that returns a 'struct libssh2_knownhost' > which in the public libssh2.h header has its 'key' field documented as "k= ey > in base64/printable format". I was getting confused between the keys coming from the knownhost_* API which are base64-encoded and the keys coming from session_hostkey() which are not. > I'd say this is a bug on the libssh2 side. It is documented and intended = to > only "export" the key as base64-encoded or an otherwise prinatable format, > which should be fine to print as zero-terminated. I see how that the > knownhost_to_external() function makes no effort to make sure that is the > case, and that seems wrong. False alarm. It works as documented. Thanks for the help. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 7 18:28:44 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27HSc0g016288; Sun, 7 Mar 2010 18:28:42 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27HSZEB016284 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 7 Mar 2010 18:28:36 +0100 Received: (qmail 19542 invoked from network); 7 Mar 2010 17:28:36 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 7 Mar 2010 17:28:36 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Sun, 07 Mar 2010 17:28:36 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164 Message-ID: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- OpenSSH-format known_hosts lines can include a comment after the key, e.g.: {{{ host2.example.com,10.0.0.1 ssh-rsa AAAAB3NzsnipAfglyt5/w== comment }}} libssh2 treats this comment as part of the key rather than a separate entry. The most annoying consequence of this is that `libssh2_knownhost_check` always finds a mismatch if the key has a comment (as server hostkeys don't have comments). Fixing this properly for the knownhost API would require an ABI change but I propose the following fix that takes care of `libssh2_knownhost_check`: {{{ #!diff src/knownhost.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/src/knownhost.c b/src/knownhost.c index fb0bff5..61ea441 100644 --- a/src/knownhost.c +++ b/src/knownhost.c @@ -327,8 +327,11 @@ libssh2_knownhost_check(LIBSSH2_KNOWNHOSTS *hosts, break; } if(match) { + char* comment = strchr(node->key, ' '); + /* host name match, now compare the keys */ - if(!strcmp(key, node->key)) { + if((!comment && !strcmp(key, node->key)) || + (comment && !strncmp(key, node->key, comment - node->key))) { /* they match! */ *ext = knownhost_to_external(node); badkey = NULL; }}} This requires, however, that the key passed to `libssh2_knownhost_check` doesn't have a comment. Would this break anyone's code? I imagine it's unlikely as typically the key would have come from `libssh2_session_hostkey`. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 7 22:47:20 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27Ll9xc023037; Sun, 7 Mar 2010 22:47:18 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27Ll7uY023034 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 7 Mar 2010 22:47:07 +0100 Received: (qmail 828 invoked from network); 7 Mar 2010 21:47:07 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 7 Mar 2010 21:47:07 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Sun, 07 Mar 2010 21:47:07 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:1 Message-ID: <054.0531938935b13d6c812172bfd406b0c6@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- Comment(by bagder): Surely libssh2_knownhost_readline() can be fixed to also parse and store the comment internally, separately from the key and that won't break any ABI. knownhost_writeline() could then be similarly adjusted to write the comment when such keys are written again. Or did I miss something? -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 7 22:51:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27LpaSc024714; Sun, 7 Mar 2010 22:51:38 +0100 Received: from mail-ew0-f218.google.com (mail-ew0-f218.google.com [209.85.219.218]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27LpZOt024696 for ; Sun, 7 Mar 2010 22:51:35 +0100 Received: by ewy10 with SMTP id 10so555093ewy.12 for ; Sun, 07 Mar 2010 13:51:31 -0800 (PST) Received: by 10.213.107.143 with SMTP id b15mr2662156ebp.72.1267998686863; Sun, 07 Mar 2010 13:51:26 -0800 (PST) Received: from [192.168.0.35] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 5sm6285302eyf.43.2010.03.07.13.51.25 (version=SSLv3 cipher=RC4-MD5); Sun, 07 Mar 2010 13:51:26 -0800 (PST) Message-ID: <4B941FDB.4090209@peut.org> Date: Sun, 07 Mar 2010 22:51:23 +0100 From: Jose Baars User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: libssh2 development Subject: Re: Port to VMS; how to get changes into the main code stream? References: <4B87E2F0.8010401@qsl.net> <50cb0cb61002250615q1cb5684pd7d6ab0f9b64e6cb@mail.gmail.com> <0A050DAD-DF9D-458F-BE7D-831D8C3528B2@intersystems.com> <4B8740A0.2070409@qsl.net> <50cb0cb61002260625m139ce81dpc36194c52fdaf7e@mail.gmail.com> <20100226163657.20114.qmail@stuge.se> <4B8867C7.6050807@qsl.net> <50cb0cb61003051812m4526dbfi47f7f2a5578a8a41@mail.gmail.com> <20100306023439.26800.qmail@stuge.se> <4B930D39.7040002@qsl.net> In-Reply-To: <4B930D39.7040002@qsl.net> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1155129763==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a cryptographically signed message in MIME format. --===============1155129763== Content-Type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary="------------ms040007020001030806080604" This is a cryptographically signed message in MIME format. --------------ms040007020001030806080604 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable John E. Malmberg schreef: > A different image name would be used for different incompatible > revisions of a shared image. > This is probably the easiest way to go. > A program does not have to be built with GNV to use the GNV directories= > to hold it. The GNV directory tree is a convenient one to use to match= > Unix conventions. Yes, why not. I'm less convinced of your insistence of using Unix=20 conventions or build procedures, but it's good idea to try to start a common place=20 to put things, regardless of the way how they end up there. > I would recommend having a PCSI kit install and if needed create the > GNV$GNU: directory tree, and to put everything under GNV$GNU:[usr] whic= h > will be seen by Unix programs as "/usr" > Yes, I like that idea too, anyone on VMS knows that way of installing. If I put the build procedure in the vms subdirectory, this would not be pose a major problem to anyone else having a go at doing this for libssh2. Never made a PCSI install kit, but I know which door to knock if I run into problems. > Also please include a VMSified xxxx.pc file if the libssh2 project has > one, and any of the documentation files in the same directories that > they would be on a Unix system. No .pc file. Lots of docs, I'll have a go at converting the man stuff=20 automatically to a help library. The HTML could stay as is. > One benefit of using the GNV style build is that you can run the UNIX > install procedure in a test mode to identify what files should be in th= e > binary kit and where they should be. Otherwise, you have to figure tha= t > out some other way. Yes, but a lot of people are not running GNV on their system, and they usually have good reasons, or are not allowed to by company policy. > > If new parameters are added, then the va args method can be used. VMS > calling standard allows a routine to be changed from a fixed number of > arguments to one that takes additional optional arguments, as long as > the routine understands that. > I think that would be too intrusive in the libssh2 code, and/or a lot of = work creating jacket routines. Also, if parameters change in meaning ( for instance Session* becomes Channel*) you would end up with programs producing weird run time errors. If something is incompatible, then it is=20 incompatible. Bump the (major) version and put the version in the imagename, allowing f= or dozens of libssh2's to be alive. I'm working on a build procedure that adds new procedure entry points to = the the end of the shared image by a balanced read aginst the major version=20 vectors, so new image libraries would maintain compatibity over minor versions.=20 > Direct references to data in a VMS shared image is a pain, because it i= s > difficult to maintain backwards compatibility if the size of the data > changes. The trick that I have seen to do this is to only use function= s > that return a pointer to that data. Then the header files use macros > that will change direct references to the data to call the routines. > No such thing in libssh2 as far as I can tell. > Please start a libssh2 note in the PORTING_TO_VMS conference about your= > project. I will. > Once you get a PCSI kit built, I can get it added to the FTP repository= > there. Sounds good. --=20 peut.org is the private domain of Jose Baars*Jos=E9 Baars* peut@peut.org +31653668905=20 State 56,5509 NX Veldhoven,Nederland --------------ms040007020001030806080604 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIPWjCC BMwwggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1 MTAyNzIzNTk1OVowgd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEf MB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNl IGF0IGh0dHBzOi8vd3d3LnZlcmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNv bmEgTm90IFZhbGlkYXRlZDE3MDUGA1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFs IFN1YnNjcmliZXIgQ0EgLSBHMjCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnf rOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyVzm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs +Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zplYu//EHuiVrvFTnAt1qIfPO2wQuhejVch rKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFBL2OyOj++pRpu9MlKWz2VphW7NQIZ +dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5gJ925rXXOL3OVekA6hXVJsLjf aLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUCAwEAAaOCAYQwggGAMBIG A1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXATAqMCgGCCsGAQUF BwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIBBjARBglghkgB hvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJlbDMtMjA0 OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAkoCKG IGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEg UHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+Iqyz cqpVMA0GCSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESf D0b3+qD+0x0Yo9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Pr v4NZmP1m3umGMpqSKTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFQTCCBCmgAwIB AgIQQkuBnab/mKbwbYynoi4GkzANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAV BgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3Jr MTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJp U2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyMB4XDTA5MTExMjAw MDAwMFoXDTEwMTExMjIzNTk1OVowggEKMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0G A1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNv bS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElBQi5MVEQoYyk5ODEeMBwGA1UE CxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdpdGFsIElEIENsYXNzIDEg LSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxEzARBgNVBAMUCkpvc2UgQmFhcnMxHDAaBgkqhkiG 9w0BCQEWDXBldXRAcGV1dC5vcmcwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCi x8yI3kFLmfYElzZSqsup782ajeuv3TASyn8mUfGrnr2aW1M7jN4vlM8/7/sVL883rH+kWPzV Q4ITYrfW5QFxH+CBqtX0F1xsS1A7AR8xZIs8o7/sMxV2xOSXYEJXlKP4XVWY8oU+TKH5dWYZ 9BJ9pkJ840ayiqqI/P3MHqmu7vS2nPDu8YZFa+46EtWkGbJTjzFdnrU5Jmdudp37fd1btNss RIkqztDOeQOTaZ6pS9RqjhmNuWVtgZucKho6c9Qf2htGz8AL95MqttF2OZYL9AlXZhjJcIMO aZBVjuwLowX/kYL2lo1vyUGt3eZxSkZR/ICfkQwj1BKpFZGFtM5XAgMBAAGjgcwwgckwCQYD VR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQt Y3JsLnZlcmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEB ABcuOD4QwMrXJgD8zvCWBRjBqsQLXn0ybqrLZHlcic7+lstAl06fi7fGD3+1Nrcp4VImkyUv JQEvFSvkB+/Mt8kOVSdt+eiYIInKt/lBkSzy0sardsh0u7Ifcgt+nmLZYW+azviXnD4VNMqe Q9QIZPHSUhWDe3IA5NSYE4hfqwXX5c13HllqpUcXRmWIGA9MPr9SQxZ0jg7MFUsZVgG9f5Ts A3yIEmvUgutyX6GRzZQL+wlWymRzh/XA7j2QnYk+IPlNQHJ62MHHlAI48y3ssSXwR8hfbH4e 0sw+8ezuh785OuFMBI3qhxlQX4wT8+v8rKHtiLcddoIohWM0ID+IdOEwggVBMIIEKaADAgEC AhBCS4Gdpv+YpvBtjKeiLgaTMA0GCSqGSIb3DQEBBQUAMIHdMQswCQYDVQQGEwJVUzEXMBUG A1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdvcmsx OzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBh IChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZlcmlT aWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzIwHhcNMDkxMTEyMDAw MDAwWhcNMTAxMTEyMjM1OTU5WjCCAQoxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxMzAxBgNVBAsTKkRpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlIEZ1bGwgU2VydmljZTETMBEGA1UEAxQKSm9zZSBCYWFyczEcMBoGCSqGSIb3 DQEJARYNcGV1dEBwZXV0Lm9yZzCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKLH zIjeQUuZ9gSXNlKqy6nvzZqN66/dMBLKfyZR8auevZpbUzuM3i+Uzz/v+xUvzzesf6RY/NVD ghNit9blAXEf4IGq1fQXXGxLUDsBHzFkizyjv+wzFXbE5JdgQleUo/hdVZjyhT5Mofl1Zhn0 En2mQnzjRrKKqoj8/cweqa7u9Lac8O7xhkVr7joS1aQZslOPMV2etTkmZ252nft93Vu02yxE iSrO0M55A5NpnqlL1GqOGY25ZW2Bm5wqGjpz1B/aG0bPwAv3kyq20XY5lgv0CVdmGMlwgw5p kFWO7AujBf+RgvaWjW/JQa3d5nFKRlH8gJ+RDCPUEqkVkYW0zlcCAwEAAaOBzDCByTAJBgNV HRMEAjAAMEQGA1UdIAQ9MDswOQYLYIZIAYb4RQEHFwEwKjAoBggrBgEFBQcCARYcaHR0cHM6 Ly93d3cudmVyaXNpZ24uY29tL3JwYTALBgNVHQ8EBAMCBaAwHQYDVR0lBBYwFAYIKwYBBQUH AwQGCCsGAQUFBwMCMEoGA1UdHwRDMEEwP6A9oDuGOWh0dHA6Ly9JbmRDMURpZ2l0YWxJRC1j cmwudmVyaXNpZ24uY29tL0luZEMxRGlnaXRhbElELmNybDANBgkqhkiG9w0BAQUFAAOCAQEA Fy44PhDAytcmAPzO8JYFGMGqxAtefTJuqstkeVyJzv6Wy0CXTp+Lt8YPf7U2tynhUiaTJS8l AS8VK+QH78y3yQ5VJ2356Jggicq3+UGRLPLSxqt2yHS7sh9yC36eYtlhb5rO+JecPhU0yp5D 1Ahk8dJSFYN7cgDk1JgTiF+rBdflzXceWWqlRxdGZYgYD0w+v1JDFnSODswVSxlWAb1/lOwD fIgSa9SC63JfoZHNlAv7CVbKZHOH9cDuPZCdiT4g+U1AcnrYwceUAjjzLeyxJfBHyF9sfh7S zD7x7O6Hvzk64UwEjeqHGVBfjBPz6/ysoe2Itx12giiFYzQgP4h04TGCBOwwggToAgEBMIHy MIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZl cmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxp ZGF0ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVy IENBIC0gRzICEEJLgZ2m/5im8G2Mp6IuBpMwCQYFKw4DAhoFAKCCAs4wGAYJKoZIhvcNAQkD MQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMTAwMzA3MjE1MTIzWjAjBgkqhkiG9w0B CQQxFgQUrlHSksjw+bmj0G5+2xwqpFZB07IwXwYJKoZIhvcNAQkPMVIwUDALBglghkgBZQME AQIwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIH MA0GCCqGSIb3DQMCAgEoMIIBAwYJKwYBBAGCNxAEMYH1MIHyMIHdMQswCQYDVQQGEwJVUzEX MBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdv cmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20v cnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZl cmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzICEEJLgZ2m/5im 8G2Mp6IuBpMwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTsw OQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAo YykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhBCS4Gdpv+YpvBtjKei LgaTMA0GCSqGSIb3DQEBAQUABIIBAJlLAjqeSNgqQe5v9UTxHtk5IHas3L7xUKjFajiepvw0 /iLj/oFpgcifp1XYVvvqTeunT610l4XUOAycHVfzX0fXTKYNC4Cl1RQ1j6pbqTs+mqBYd2Ip 5q49YQ22gWuSrD5/LAohnaqWdJCn0uxiuyDWq0hdjSkod2hBOa+DO/9hK/TI6gy/CTqaygau 15aSkpfFHFYaNWk+SARe9e+Tb45vlXaGx0C2PXhYuGD+4Sw1kfKL2maLWQpMsl6tK8K6KUSy GJHgjl/OSb97dE+A2FRKBWC1vwlE0H4Bo+226vjwNpdrkGzueEBzP5owL/UcFu42It4JKouh zQc2TbNt79IAAAAAAAA= --------------ms040007020001030806080604-- --===============1155129763== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1155129763==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 7 23:41:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27Mexjm020251; Sun, 7 Mar 2010 23:41:04 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o27MeviI020230 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 7 Mar 2010 23:40:57 +0100 Received: (qmail 4112 invoked from network); 7 Mar 2010 22:40:58 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 7 Mar 2010 22:40:58 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Sun, 07 Mar 2010 22:40:58 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:2 Message-ID: <054.d1920c25b6d9df2a42fea11dc2da7bd8@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- Comment(by alamaison): Replying to [comment:1 bagder]: > Surely libssh2_knownhost_readline() can be fixed to also parse and store the comment internally, separately from the key and that won't break any ABI. `libssh2_knownhost_readline` eventually delegates to `libssh2_knownhost_add`. I suppose we could replace that with a call to an internal version of `add` that take a comment parameter or we could change `add` so that it parses the key into two pieces. But we're still left with `libssh2_knownhost_get` that returns the key and the comment as one. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 8 17:42:26 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28Gg0x4014637; Mon, 8 Mar 2010 17:42:12 +0100 Received: from mail-bw0-f226.google.com (mail-bw0-f226.google.com [209.85.218.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28Gfw5R014473 for ; Mon, 8 Mar 2010 17:41:58 +0100 Received: by bwz26 with SMTP id 26so1413220bwz.7 for ; Mon, 08 Mar 2010 08:41:54 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:from:date :x-google-sender-auth:message-id:subject:to:content-type; bh=giF1PgpL76ozqpu6KZJRKHMWfMNSjByeB5hCA181jsw=; b=oSFww/HuuCAkBuwHR2EIVLJP5hsDsp3h6eph+AhEfDp0JY9jeFfRyRR1S/E5KMgEpv vxAS2KZgKK8ZvjB2N3hb9PMrn+sOSNVomsYu9Z+N+fHvd+3mqcncFp3Kg6a0lfEOEm0v QJMWNOCrJbD2m09Wi0gUfAIdTsbWTtjLJtm1k= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:from:date:x-google-sender-auth:message-id :subject:to:content-type; b=pvxrWhIIOPBQZh1jD9IEPcVmfZ8O9KOWnVnYOzhYWR/iTpY0bDmAOOja6M7esssnqd 97Rt1ELTUs6D8CEROQW5qVtfaxd8zrRQL0zPI4OmFS3jZanScvVsLPqAngg/N25Rhmj8 Lqw/TFi66LoB4NppmqpHAuMOQK8qqSItiaBsU= MIME-Version: 1.0 Received: by 10.204.48.209 with SMTP id s17mr5564953bkf.88.1268066514345; Mon, 08 Mar 2010 08:41:54 -0800 (PST) From: Steven Van Ingelgem Date: Mon, 8 Mar 2010 17:41:33 +0100 X-Google-Sender-Auth: d8f2ef53b75813cf Message-ID: <5862ce3f1003080841r2e4e07b8ge6b75ac74dd673@mail.gmail.com> Subject: libssh2_publickey_* To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0838258879==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0838258879== Content-Type: multipart/alternative; boundary=00032555875a239ffe04814cbe63 --00032555875a239ffe04814cbe63 Content-Type: text/plain; charset=UTF-8 Am I correct with saying that by using "libssh2_publickey_add/remove" you can add/remove keys from .ssh/authorized_keys? Thx, Steven --00032555875a239ffe04814cbe63 Content-Type: text/html; charset=UTF-8 Am I correct with saying that by using "libssh2_publickey_add/remove" you can add/remove keys from .ssh/authorized_keys?


Thx,
Steven
--00032555875a239ffe04814cbe63-- --===============0838258879== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0838258879==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 8 18:29:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28HSvm2024447; Mon, 8 Mar 2010 18:29:02 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.27]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28HStNb024427 for ; Mon, 8 Mar 2010 18:28:55 +0100 Received: by ey-out-2122.google.com with SMTP id 9so836645eyd.31 for ; Mon, 08 Mar 2010 09:28:56 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=1AgffgtbuFSD5XMb30tQfBUFj9qh50loI5d4cMPpUKg=; b=l4K2MNYjweYpblnzUSO7PqRkAILSoFbli9Lg9dvTIzigJv3NhAONA2HUoBu8G/Or3S Bc9oyaML93CFvhLAmS0kXP07n0iyuVZRuhbPB1Di28n1rT6HW8RkeWPh+PpqB1wgfNtu E+y7ydEfJipiAzuS9gRGdrXCdR5Ekw8FPXhFw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=EMWQTr/Na6QQ5aN+OXxAvFrVdrQIgFsbo76RDLJW4gwK1ZtI1nRw3sUpJEjscV8iWs On4wk1t0NY75b3liGnZTVwNSklOZpKkLFPcZmnCRnLf8QYPuVxuQZRDmpGtUIT7/3Oit VnacUc+/kb7lOt0ikWM6DWRVYV74Ohoepy2Cs= MIME-Version: 1.0 Received: by 10.213.57.10 with SMTP id a10mr3294886ebh.92.1268069336635; Mon, 08 Mar 2010 09:28:56 -0800 (PST) In-Reply-To: <5862ce3f1003080841r2e4e07b8ge6b75ac74dd673@mail.gmail.com> References: <5862ce3f1003080841r2e4e07b8ge6b75ac74dd673@mail.gmail.com> Date: Mon, 8 Mar 2010 17:28:56 +0000 X-Google-Sender-Auth: 301c8f143907e24d Message-ID: <9c3275a91003080928k16012069ne5e5b9a2aac82a43@mail.gmail.com> Subject: Re: libssh2_publickey_* From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 8 March 2010 16:41, Steven Van Ingelgem wrote: > Am I correct with saying that by using "libssh2_publickey_add/remove" you > can add/remove keys from .ssh/authorized_keys? In theory. But it relies on server support and OpenSSH doesn't. Shame. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 8 20:28:25 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28JSC7G013288; Mon, 8 Mar 2010 20:28:23 +0100 Received: from mail-bw0-f226.google.com (mail-bw0-f226.google.com [209.85.218.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28JSA3W013224 for ; Mon, 8 Mar 2010 20:28:10 +0100 Received: by bwz26 with SMTP id 26so1594621bwz.7 for ; Mon, 08 Mar 2010 11:28:07 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:from:date:x-google-sender-auth:message-id:subject:to :content-type; bh=VdUxnmNdALt1gjadnd7hkh1vADlOSTCeoCo0Zo9oyGs=; b=yCaYIpNrOQggVgymQuCaZRyu4NOy+Yi2hkNd1cHw7vNNOvkQh3Tn15Nj/WxhTet9Mr 9mm7kE+k7laffG3oIHufiUWGjGKnJZWn1eCtcpikkz+3SNp5QKZH+dMCIS+dn09/59uM Pjs6IWN2NY2CYIzIyF0uND6OZZlnQnLCzlwLE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:content-type; b=ixjU95Yh0nMxNY/cp1YkFc9PoR6amm9mrxXT2p1CHogmoFiRIe/jNd+8xD1HW0RVaH Vom4DROC+Y5MD8wyQGYOo/XALlNoOW3MQ3MGIQkTGxtkLAHRU4Ez9rmBLLDfc7t7o9mX j+YRmeIKTr+CkIEDhLG7juiLLaZ2mGmBDGSwg= MIME-Version: 1.0 Received: by 10.204.34.15 with SMTP id j15mr217281bkd.72.1268076487129; Mon, 08 Mar 2010 11:28:07 -0800 (PST) In-Reply-To: <9c3275a91003080928k16012069ne5e5b9a2aac82a43@mail.gmail.com> References: <5862ce3f1003080841r2e4e07b8ge6b75ac74dd673@mail.gmail.com> <9c3275a91003080928k16012069ne5e5b9a2aac82a43@mail.gmail.com> From: Steven Van Ingelgem Date: Mon, 8 Mar 2010 20:27:47 +0100 X-Google-Sender-Auth: f753f58f8b0c95db Message-ID: <5862ce3f1003081127h5deb27edm8ad2f0fc58af8aa9@mail.gmail.com> Subject: Re: libssh2_publickey_* To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0116645682==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0116645682== Content-Type: multipart/alternative; boundary=000325557a9a9039df04814f10cc --000325557a9a9039df04814f10cc Content-Type: text/plain; charset=UTF-8 That's a pity :( On 8 March 2010 18:28, Alexander Lamaison wrote: > On 8 March 2010 16:41, Steven Van Ingelgem wrote: > > Am I correct with saying that by using "libssh2_publickey_add/remove" you > > can add/remove keys from .ssh/authorized_keys? > > In theory. But it relies on server support and OpenSSH doesn't. Shame. > > Alex > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > --000325557a9a9039df04814f10cc Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable That's a pity :(

On 8 March 2010 18:2= 8, Alexander Lamaison <swish@lammy.co.uk> wrote:
On 8 March 2010 16:41, Steven Van Ingelgem <steven@vaningelgem.be> wrote:
> Am I correct with saying that by using "libssh2_publickey_add/rem= ove" you
> can add/remove keys from .ssh/authorized_keys?

In theory. =C2=A0But it relies on server support and OpenSSH doesn= 9;t. =C2=A0Shame.

Alex
_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/li= bssh2-devel

--000325557a9a9039df04814f10cc-- --===============0116645682== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0116645682==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 8 21:49:04 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28KmqF7011013; Mon, 8 Mar 2010 21:49:01 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o28KmpZT011009 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 8 Mar 2010 21:48:51 +0100 Received: (qmail 23821 invoked from network); 8 Mar 2010 20:48:51 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 8 Mar 2010 20:48:51 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 08 Mar 2010 20:48:51 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:3 Message-ID: <054.aa7a19a841f0b4793d7655dc9b4986b6@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- Comment(by bagder): Right, we would use a new version of libssh2_knownhost_add() that supports the comment part. Since we internally would store the key and comment separately, we would probably just make libssh2_knownhost_get not return any comment at all since the public struct has no such field. Not ideal, but it would make the file parsing know comments and it could make the key comparison stuff to work. And it would be the same ABI. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 11:01:25 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29A16Qf024550; Tue, 9 Mar 2010 11:01:20 +0100 Received: from f1wall.ipetronik.com (f1wall.ipetronik.com [217.25.129.2]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29A14Vq024404 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 9 Mar 2010 11:01:04 +0100 Received: from [10.1.1.105] (port=23761 helo=IPEMAIL.ipetronik.de) by f1wall.ipetronik.com with esmtp (Exim 4.69) (envelope-from ) id 1NowFG-00024T-0Z for libssh2-devel@cool.haxx.se; Tue, 09 Mar 2010 11:00:58 +0100 Received: from IPE_DOM-MTA by IPEMAIL.ipetronik.de with Novell_GroupWise; Tue, 09 Mar 2010 11:00:58 +0100 Message-Id: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> X-Mailer: Novell GroupWise Internet Agent 8.0.1 Date: Tue, 09 Mar 2010 11:00:52 +0100 From: "Sebastien Fricker" To: Subject: Setting timeout in blocking mode Mime-Version: 1.0 Content-Disposition: inline X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se SGksCkknbSBjdXJyZW50bHkgd29uZGVyaW5nIHdoeSB0aGVyZSBpcyBub3cgd2FpdCBzbyBzcGVj aWZ5IGEgdGltZW91dCB3aGVuCndvcmtpbmcgb24gYmxvY2tpbmcgbW9kZS4KCklmIHdlIGhhdmUg YSBsb29rIG9uIF9saWJzc2gyX3dhaXRfc29ja2V0KCkgd2UgY2FuIHNlZSB0aGF0IHRoZSBzZWxl Y3QKYmxvY2tzIGluZmluaXRlbHksIGJ1dCB0aGVyZSBpcyBhIApjb2RlIHdoaWNoIHNob3VsZCBo YW5kbGUgdGhlIHRpbWVvdXQuCgpDb2RlOgogICAgLyogTm90ZSB0aGF0IHRoaXMgQ09VTEQgYmUg bWFkZSB0byB1c2UgYSB0aW1lb3V0IHRoYXQgcGVyaGFwcyBjb3VsZApiZQogICAgICAgY3VzdG9t aXphYmxlIGJ5IHRoZSBhcHAgb3Igc29tZXRoaW5nLi4uICovCiAgICByYyA9IHNlbGVjdChzZXNz aW9uLT5zb2NrZXRfZmQgKyAxLCByZWFkZmQsIHdyaXRlZmQsIE5VTEwsIE5VTEwpOwoKICAgIGlm KHJjIDw9IDApIHsKICAgICAgICAvKiB0aW1lb3V0IChvciBlcnJvciksIGJhaWwgb3V0IHdpdGgg YSB0aW1lb3V0IGVycm9yICovCiAgICAgICAgc2Vzc2lvbi0+ZXJyX2NvZGUgPSBMSUJTU0gyX0VS Uk9SX1RJTUVPVVQ7CiAgICAgICAgcmV0dXJuIExJQlNTSDJfRVJST1JfVElNRU9VVDsKICAgIH0K CklzIHRoZXJlIGFueSByZWFzb24gbm90IHRvIGFkZCBhIHRpbWVvdXQgcGFyYW1ldGVyIGFzIGZv bGxvd3M/OgogICAgcmMgPSBzZWxlY3Qoc2Vzc2lvbi0+c29ja2V0X2ZkICsgMSwgcmVhZGZkLCB3 cml0ZWZkLCBOVUxMLApzZXNzaW9uLT50aW1lb3V0KTsKClRoaXMgd291bGQgb25seSByZXF1aXJl IHRvIGFkZCAndGltZW91dCcgaW50byB0aGUgc2Vzc2lvbiBzdHJ1Y3QgYW5kIHRvCmNyZWF0ZSBh IGZ1bmN0aW9uIEFQSSB3aGljaCBwZXJtaXRzIHRvIHNldCBpdHMgdmFsdWUuClJlZ2FyZHMKU8Op YmFzdGllbgoKLS0gCgotLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0KSXBldHJvbmlrIEdtYkggJiBDby5LRwpKYWVnZXJ3ZWcgMQpELTc2 NTMyIEJhZGVuLUJhZGVuCiAKUGhvbmUtTm8uICs0OSAtICgwKTcyMjEvOTkyMi00NjcKRmF4LU5v LiArNDkgLSAoMCk3MjIxLzk5MjItMTUzCiAKbWFpbHRvOiBzZWJhc3RpZW4uZnJpY2tlckBpcGV0 cm9uaWsuY29tCndlYjogaHR0cDovL3d3dy5pcGV0cm9uaWsuY29tIAogCi0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQpLb21tYW5kaXRn ZXNlbGxzY2hhZnQgbWl0IFNpdHogaW4gQmFkZW4tQmFkZW4sIFJlZ2lzdGVyZ2VyaWNodCBIUkEK TnIuMjAxMzEzClBlcnNvZW5saWNoIGhhZnRlbmRlIEdlc2VsbHNjaGFmdCBpc3QgZGllIElQRVRS T05JSyBWZXJ3YWx0dW5ncyBHbWJIIAptaXQgU2l0eiBpbiBCYWRlbi1CYWRlbiwgUmVnaXN0ZXJn ZXJpY2h0IE1hbm5oZWltIEhSQiBOci4yMDIwODksIApHZXNjaMOkZnRzZsO8aHJlcjogSi5BYnJv bWVpdCwgRS5SdWRvbGYsIEEuIFdvY2tlCi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0KIAogCgpfX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fXwpsaWJzc2gyLWRldmVsIGh0dHA6Ly9jb29sLmhheHgu c2UvY2dpLWJpbi9tYWlsbWFuL2xpc3RpbmZvL2xpYnNzaDItZGV2ZWwK From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 12:30:04 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29BTpOF020936; Tue, 9 Mar 2010 12:30:00 +0100 Received: from web24404.mail.ird.yahoo.com (web24404.mail.ird.yahoo.com [87.248.114.216]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o29BTn21020930 for ; Tue, 9 Mar 2010 12:29:49 +0100 Received: (qmail 70149 invoked by uid 60001); 9 Mar 2010 11:29:44 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.fr; s=s1024; t=1268134184; bh=xArNK+cuBdUqmztHrwZZfob2uWgeje4duPhViX8zw64=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=tfHdbwaTBP0K4sOE9FXENm6dLHbkJRs/KXDFm7xuCAo+bJElwxO+9NJeHVARpHVyAloUwmc5sjlN0cbUHvjjGUpYwIUfBLvD4O9JlOzmZfkzBrcJTxoP8XKu9k6oGW740GLd/YMnVOj4icKtjwU4/MfC3X69XcDVqUjUQ5h1nQQ= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.fr; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=nocPeM3pqa2ycTJvocUrbLoybW2G1e7iCDVDYUrz5f1mlPELEImwp4KmQdM2esWt/xnObm0H07q7Qby8OyCupTTkp3HbPnhEbWSFFm+ySQUU0NWTuW3FXFO3557AhB4+qGahQa4YDW+g7DMlbLITKLzjDNS7HYBZLzBXPytqK+4=; Message-ID: <832690.68608.qm@web24404.mail.ird.yahoo.com> X-YMail-OSG: BBDBlqgVM1mfTRpS_RoFyLNMXBjKqZLgRRjF7IPmVCkegJ0 dyuJfSHLo9p4bFcVa9XxDPVD_zwXzsQFDCrJ.3L7U1xEGcCgR2YAPZ8LfBks __YijmBCR8yqaYqn6yMgYNHgKixLAdxnV7ISonswmaftfhWUxUG4G4km7YjE ey3uGh04eOwaaK.iWcz3y0Lnr.tkSAiQ7NkvqLUv0pi9HluG1SwVWgZwX6yh Wqkhp.JL.QHeraK1X3YY9P3VwiC06mcCSo4GDCSRJDw-- Received: from [193.251.74.9] by web24404.mail.ird.yahoo.com via HTTP; Tue, 09 Mar 2010 11:29:44 GMT X-Mailer: YahooMailRC/300.3 YahooMailWebService/0.8.100.260964 Date: Tue, 9 Mar 2010 11:29:44 +0000 (GMT) From: sebastien isquierdo Subject: About libssh2_channel_read To: libssh2-devel@cool.haxx.se MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hello, = I got an application which listen on a port and use libssh2 to connect to a= SSH Server and make port forwarding to MySQL. It works fine with 1 connection. I'm trying now to make 2 or more connections to MySQL through SSH. Actually, i do this : = - create a non-blocking session (libssh2_session_init, ) - start the session (libssh2_session_startup) - Listen every connections to my local port (create new socket) - and for each,create a new channel (libssh2_channel_direct_tcpip_ex) - for each socket write on the good channel (libssh2_channel_write) - and listen to the good channel to (libssh2_channel_read) = I make a new thread for reading on every channel. My problem is libssh2_channel_read returns informations only from on the la= st created channel and nothing from the others. Thanks for any advice. S=E9bastien. = _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 13:18:11 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29CI3lA030709; Tue, 9 Mar 2010 13:18:09 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29CI1Hl030641 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 9 Mar 2010 13:18:01 +0100 Received: (qmail 22979 invoked from network); 9 Mar 2010 12:18:01 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 9 Mar 2010 12:18:01 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Tue, 09 Mar 2010 12:18:01 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:4 Message-ID: <054.047df3176584f97663b77d01a11ab4f6@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- Comment(by alamaison): Replying to [comment:3 bagder]: > Right, we would use a new version of libssh2_knownhost_add() that supports the comment part. Do you mean adding an extra parameter to `add` (this wouldn't change the ABI, right?) or adding a new version of `add`, say `libssh2_knownhost_add_with_comment` that takes a separate comment parameter. I must admit I'm not entirely clear on which changes are ABI preserving and which aren't. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 13:32:10 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29CW00C010725; Tue, 9 Mar 2010 13:32:08 +0100 Received: from mail-bw0-f216.google.com (mail-bw0-f216.google.com [209.85.218.216]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29CVwGx010706 for ; Tue, 9 Mar 2010 13:31:58 +0100 Received: by bwz8 with SMTP id 8so3065628bwz.3 for ; Tue, 09 Mar 2010 04:31:54 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:from:date :x-google-sender-auth:message-id:subject:to:content-type; bh=hpFip+jUny2320djvnhNKwyvVIMW227N75izUVL1Txs=; b=ow+8f/e0juK3rmUZaV5LImnPD0Z7Z/IiXMBezrtSFTu5gJNqpB0fiELK6QckdZCbfK ZVZuJ0nle6kITwl9ibltM3kXDMW7KkfAPYg07VvxkT3jXChXzR1qkUrDaCbGHy1nYiV5 +ni1KYBJNdZfzot/GlQTYzATqbyfpcjpJGVAw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:from:date:x-google-sender-auth:message-id :subject:to:content-type; b=b+MIBB33Zk+DhtMd3LGO6Do3Uo6aKUqNPzmSP0nAQwaeXPWSkV8iG9pYnYOBsfaQpY T6blpvUCxab0B7+5PVNkR126iv6OdBHDLh9pa4qSAgls3+XODEnDKor4TJYIyNkb8B4M SVEQYsW+swrDEFLfnk85Kask1dpB12tvaeghM= MIME-Version: 1.0 Received: by 10.204.32.15 with SMTP id a15mr7081203bkd.142.1268137914099; Tue, 09 Mar 2010 04:31:54 -0800 (PST) From: Steven Van Ingelgem Date: Tue, 9 Mar 2010 13:31:34 +0100 X-Google-Sender-Auth: b89009b386ee012b Message-ID: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> Subject: Visual Studio 6 compilation issues (libssh2 1.2.4) To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0400798795==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0400798795== Content-Type: multipart/alternative; boundary=000325553116e5696804815d5d1b --000325553116e5696804815d5d1b Content-Type: text/plain; charset=UTF-8 Please find hereby errors (and warnings) I receive by compiling the php extension php_ssh2 with v1.2.4 of libssh2. Grtz, Steven ext\ssh2\libssh2\src\misc.c(524) : error C2632: 'long' followed by 'long' is illegal ext\ssh2\libssh2\src\misc.c(531) : error C2059: syntax error : 'bad suffix on number' ext\ssh2\libssh2\src\misc.c(531) : error C2146: syntax error : missing ')' before identifier 'L' ext\ssh2\libssh2\src\misc.c(531) : error C2059: syntax error : 'bad suffix on number' ext\ssh2\libssh2\src\misc.c(532) : error C2059: syntax error : 'bad suffix on number' ext\ssh2\libssh2\src\misc.c(532) : error C2146: syntax error : missing ')' before identifier 'L' ext\ssh2\libssh2\src\misc.c(532) : error C2059: syntax error : 'bad suffix on number' openssl.c packet.c ext\ssh2\libssh2\src\packet.c(1131) : warning C4761: integral size mismatch in argument; conversion supplied pem.c publickey.c scp.c ext\ssh2\libssh2\src\scp.c(753) : warning C4244: '=' : conversion from 'long ' to 'unsigned short ', possible loss of data session.c ext\ssh2\libssh2\src\session.c(352) : error C2065: 'socklen_t' : undeclared identifier ext\ssh2\libssh2\src\session.c(352) : error C2146: syntax error : missing ';' before identifier 'option_len' ext\ssh2\libssh2\src\session.c(352) : error C2065: 'option_len' : undeclared identifier ext\ssh2\libssh2\src\session.c(1359) : warning C4018: '==' : signed/unsigned mismatch ext\ssh2\libssh2\src\session.c(1360) : warning C4018: '>' : signed/unsigned mismatch ext\ssh2\libssh2\src\session.c(1364) : warning C4018: '==' : signed/unsigned mismatch ext\ssh2\libssh2\src\session.c(1365) : warning C4018: '>' : signed/unsigned mismatch sftp.c transport.c userauth.c NMAKE : fatal error U1077: '"cl.exe"' : return code '0x2' --000325553116e5696804815d5d1b Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Please find hereby errors (and warnings) I receive by compiling the php ext= ension php_ssh2 with v1.2.4 of libssh2.


Grtz,
Steven

<= br>

ext\ssh2\libssh2\src\misc.c(524) : error C2632: 'long' f= ollowed by 'long' is illegal
ext\ssh2\libssh2\src\misc.c(531) : error C2059: syntax error : 'bad suf= fix on number'
ext\ssh2\libssh2\src\misc.c(531) : error C2146: synta= x error : missing ')' before identifier 'L'
ext\ssh2\lib= ssh2\src\misc.c(531) : error C2059: syntax error : 'bad suffix on numbe= r'
ext\ssh2\libssh2\src\misc.c(532) : error C2059: syntax error : 'bad suf= fix on number'
ext\ssh2\libssh2\src\misc.c(532) : error C2146: synta= x error : missing ')' before identifier 'L'
ext\ssh2\lib= ssh2\src\misc.c(532) : error C2059: syntax error : 'bad suffix on numbe= r'
openssl.c
packet.c
ext\ssh2\libssh2\src\packet.c(1131) : warning C476= 1: integral size mismatch in argument; conversion supplied
pem.c
publ= ickey.c
scp.c
ext\ssh2\libssh2\src\scp.c(753) : warning C4244: '= =3D' : conversion from 'long ' to 'unsigned short ', po= ssible loss of data
session.c
ext\ssh2\libssh2\src\session.c(352) : error C2065: 'sockle= n_t' : undeclared identifier
ext\ssh2\libssh2\src\session.c(352) : e= rror C2146: syntax error : missing ';' before identifier 'optio= n_len'
ext\ssh2\libssh2\src\session.c(352) : error C2065: 'option_len' : u= ndeclared identifier
ext\ssh2\libssh2\src\session.c(1359) : warning C401= 8: '=3D=3D' : signed/unsigned mismatch
ext\ssh2\libssh2\src\sess= ion.c(1360) : warning C4018: '>' : signed/unsigned mismatch
ext\ssh2\libssh2\src\session.c(1364) : warning C4018: '=3D=3D' : si= gned/unsigned mismatch
ext\ssh2\libssh2\src\session.c(1365) : warning C4= 018: '>' : signed/unsigned mismatch
sftp.c
transport.c
userauth.c
NMAKE : fatal error U1077: '"cl.exe"' : ret= urn code '0x2'
--000325553116e5696804815d5d1b-- --===============0400798795== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0400798795==-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 14:17:19 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DHBhe012283; Tue, 9 Mar 2010 14:17:16 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DHADX012261 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 9 Mar 2010 14:17:10 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o29DH9K1005210 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 9 Mar 2010 14:17:10 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Setting timeout in blocking mode References: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100309:libssh2-devel@cool.haxx.se::n52LpL6USBcKt1Ea:IyVR Date: Tue, 09 Mar 2010 14:17:09 +0100 In-Reply-To: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> (Sebastien Fricker's message of "Tue, 09 Mar 2010 11:00:52 +0100") Message-ID: <87tyspprca.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se "Sebastien Fricker" writes: > Hi, > I'm currently wondering why there is now wait so specify a timeout when > working on blocking mode. > > If we have a look on _libssh2_wait_socket() we can see that the select > blocks infinitely, but there is a > code which should handle the timeout. > > Code: > /* Note that this COULD be made to use a timeout that perhaps could > be > customizable by the app or something... */ > rc = select(session->socket_fd + 1, readfd, writefd, NULL, NULL); > > if(rc <= 0) { > /* timeout (or error), bail out with a timeout error */ > session->err_code = LIBSSH2_ERROR_TIMEOUT; > return LIBSSH2_ERROR_TIMEOUT; > } > > Is there any reason not to add a timeout parameter as follows?: > rc = select(session->socket_fd + 1, readfd, writefd, NULL, > session->timeout); > > This would only require to add 'timeout' into the session struct and to > create a function API which permits to set its value. The simplest solution is to do the select call in your application and use libssh2 in non-blocking mode. If you don't want to risk libssh2 blocking forever in some situations, I believe you need to do this anyway. This way your application has full control. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 14:18:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DIbXO012992; Tue, 9 Mar 2010 14:18:39 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DIavD012984 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 9 Mar 2010 14:18:36 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o29DIZFg005241 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 9 Mar 2010 14:18:37 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100309:libssh2-devel@cool.haxx.se::il9rjzzFYLPzHS1U:CLaA Date: Tue, 09 Mar 2010 14:18:35 +0100 In-Reply-To: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> (Steven Van Ingelgem's message of "Tue, 9 Mar 2010 13:31:34 +0100") Message-ID: <87pr3dpr9w.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Steven Van Ingelgem writes: > Please find hereby errors (and warnings) I receive by compiling the php > extension php_ssh2 with v1.2.4 of libssh2. > > > Grtz, > Steven > > > > > ext\ssh2\libssh2\src\misc.c(524) : error C2632: 'long' followed by 'long' is > illegal This means your compiler doesn't have 'long long', which is curious because this code is only enabled for Windows. Maybe you could write a better 'gettimeofday' implementation that works on Windows and without 'long long'? /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 14:19:50 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DJkcK013488; Tue, 9 Mar 2010 14:19:48 +0100 Received: from dottedmag.net (burger.dottedmag.net [212.75.37.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DJjAt013485 for ; Tue, 9 Mar 2010 14:19:45 +0100 Received: from vertex.dottedmag (unknown [91.197.127.125]) by dottedmag.net (Postfix) with ESMTPSA id 984508C072 for ; Tue, 9 Mar 2010 14:19:45 +0100 (CET) Received: from dottedmag by vertex.dottedmag with local (Exim 4.69) (envelope-from ) id 1NozLa-0004Ld-Mv for libssh2-devel@cool.haxx.se; Tue, 09 Mar 2010 19:19:42 +0600 From: Mikhail Gusarov To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> Date: Tue, 09 Mar 2010 19:19:41 +0600 In-Reply-To: <87pr3dpr9w.fsf@mocca.josefsson.org> (Simon Josefsson's message of "Tue, 09 Mar 2010 14:18:35 +0100") Message-ID: <87y6i1bpjm.fsf@vertex.dottedmag> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0127174595==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0127174595== Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --=-=-= Content-Transfer-Encoding: quoted-printable Twas brillig at 14:18:35 09.03.2010 UTC+01 when simon@josefsson.org did gyr= e and gimble: SJ> This means your compiler doesn't have 'long long', which is curious SJ> because this code is only enabled for Windows. Maybe you could SJ> write a better 'gettimeofday' implementation that works on Windows SJ> and without 'long long'? s/long long/__int64/g =2D-=20 http://fossarchy.blogspot.com/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iQIcBAEBAgAGBQJLlkruAAoJEJ0g9lA+M4iIuhoP/jLRH8yw3tssq0yAh/JZNtFq rXs27sYuqYKMGm0Kzpi3uVa2deR4d5/w8pEXBMzIuiRig6Mtb1R0ldt3KifI5nsK azbEUsPIF6uWVaEjL2oq58jIaKMg6vhnNPkT9U1D6ARl95EfabYHICLW8xucqjD0 yCZRCvdIs0nSRnf5mQ0/zLy9UQn2x5ME7P0descW9jC0+qqwJQzOu/YzmGSe+qSY GrqroSSir+kCICWLspmtXed77Gq7bDzwSkN+wspDwyjhIhQX8g6FWYztRd0/KTsf +VlF7cy2v7zJjyet/gBjEz6G1Q3sHf6GX5+n1Vyqhq7Cp19lsv4drUrd0VlmIlTO kvl8xqegAqRQYLUTl9W7dHVr7ghSKuOjOr890Rt3owjmToirMWVMOKO7xMvIZBnL wnHNotPn2WfDlbXm9AILGOLcf+G2UJTDAlxEuRlrzpWmnqEvLtxi47MEDlzdtAAu cy3CAYWY645mOkn/CwAuHAkc1FhDMpwegL10MrRFoM3vLdkl7XnkbLGT82ZBzeUs mGJvFoT9vyvEtBu0EX0H6B2oYGwjXFw89HxEWzYJpxj5HGwqST6Zq0P0GY6VQ8dF TvKaJE0SU8JmXUfnCr07KRvJmBoapc+HgQzJzlmwajnceuCC0hMPcN0TbEpUTu2m +FOV9FKYZw1Kf92GhZ34 =4YX+ -----END PGP SIGNATURE----- --=-=-=-- --===============0127174595== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0127174595==-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 14:30:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29DU7cQ025107; Tue, 9 Mar 2010 14:30:10 +0100 Received: from mail.gmx.net (mail.gmx.net [213.165.64.20]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o29DU6kO024755 for ; Tue, 9 Mar 2010 14:30:06 +0100 Received: (qmail invoked by alias); 09 Mar 2010 13:30:02 -0000 Received: from xdsl-188-155-90-113.adslplus.ch (EHLO [192.168.1.71]) [188.155.90.113] by mail.gmx.net (mp055) with SMTP; 09 Mar 2010 14:30:02 +0100 X-Authenticated: #1568940 X-Provags-ID: V01U2FsdGVkX18I9scCgfAIPia5xpUPRgxazE7rUr5Cnp+Vw4A167 +WOBzPjR1yewXx Message-ID: <4B964D58.1070309@gmx.at> Date: Tue, 09 Mar 2010 14:30:00 +0100 From: double User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: libssh2 development Subject: Re: Setting timeout in blocking mode References: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> In-Reply-To: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> X-Y-GMX-Trusted: 0 X-FuHaFi: 0.70999999999999996 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Sebastien Fricker schrieb: > Hi, > I'm currently wondering why there is now wait so specify a timeout when > working on blocking mode. Maybe setsockopt(SO_RCVTIMEO + SO_SNDTIMEO) ? Marcus _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 15:02:24 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29E2DYx009502; Tue, 9 Mar 2010 15:02:21 +0100 Received: from mail-bw0-f216.google.com (mail-bw0-f216.google.com [209.85.218.216]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29E2BXp009338 for ; Tue, 9 Mar 2010 15:02:11 +0100 Received: by bwz8 with SMTP id 8so3154034bwz.3 for ; Tue, 09 Mar 2010 06:02:07 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:from:date:x-google-sender-auth:message-id:subject:to :content-type; bh=y9d8ovO+5ZZis3esr+AWUMTlwafel2/y6IrFkWFmNnI=; b=qfvEL4vdbqPCwBgggTOwG7fgOylLRcYIeYKQWtEjN4o5fXCgv5OCqxQKVupv4Zi+kn ZutxJ4tGLdQbI3Iy+QR4TEeq7rk+60yDhAPgLW+Wfzb7IgwsxKWfOWLYL2C3/jaJn2/u qAGMhMQ3CFIkriCO4wSUQ8NsMCq7/GdJe4hxc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:content-type; b=r0tUon105xusv9GymCd9+sjq7U5aSlDDHlBTabK/lY//MiCRAMTkegkYTFEqeOljSr 3oHGC7sgtCFJWyRJW4RPDcTzgvJeE28B8cG8ioDY2S6K90g7OlE0of4BIVbLsGcFCeFf IpeUqq6+Q0HPvgIDC6Vr8/aldHYKNqEjD9lOA= MIME-Version: 1.0 Received: by 10.204.9.6 with SMTP id j6mr1607643bkj.37.1268143327366; Tue, 09 Mar 2010 06:02:07 -0800 (PST) In-Reply-To: <87y6i1bpjm.fsf@vertex.dottedmag> References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> From: Steven Van Ingelgem Date: Tue, 9 Mar 2010 15:01:47 +0100 X-Google-Sender-Auth: 3736a973afa9c68a Message-ID: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0084084516==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0084084516== Content-Type: multipart/alternative; boundary=0015175884ae8d4f1704815ea0bf --0015175884ae8d4f1704815ea0bf Content-Type: text/plain; charset=UTF-8 what I did was using the ULONGLONG instead + removing the "ULL" after the numbers. Just wanted to let everyone know that there are issues on VS6 On 9 March 2010 14:19, Mikhail Gusarov wrote: > > Twas brillig at 14:18:35 09.03.2010 UTC+01 when simon@josefsson.org did > gyre and gimble: > > SJ> This means your compiler doesn't have 'long long', which is curious > SJ> because this code is only enabled for Windows. Maybe you could > SJ> write a better 'gettimeofday' implementation that works on Windows > SJ> and without 'long long'? > > s/long long/__int64/g > > -- > http://fossarchy.blogspot.com/ > > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > --0015175884ae8d4f1704815ea0bf Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable what I did was using the ULONGLONG instead + removing the "ULL" a= fter the numbers.

Just wanted to let everyone know that there are is= sues on VS6

On 9 March 2010 14:19, Mikhai= l Gusarov <= dottedmag@dottedmag.net> wrote:

Twas brillig at 14:18:35 09.03.2010 UTC+01 when simon@josefsson.org did gyre and gimble:

=C2=A0SJ> This means your compiler doesn't have 'long long',= which is curious
=C2=A0SJ> because this code is only enabled for Windows. =C2=A0Maybe you= could
=C2=A0SJ> write a better 'gettimeofday' implementation that work= s on Windows
=C2=A0SJ> and without 'long long'?

s/long long/__int64/g

--
=C2=A0http://= fossarchy.blogspot.com/

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/li= bssh2-devel


--0015175884ae8d4f1704815ea0bf-- --===============0084084516== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0084084516==-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 15:10:28 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29EANVw018172; Tue, 9 Mar 2010 15:10:27 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o29EALr9018168 for ; Tue, 9 Mar 2010 15:10:21 +0100 Received: (qmail 22760 invoked by uid 501); 9 Mar 2010 14:10:22 -0000 Message-ID: <20100309141022.22759.qmail@stuge.se> Date: Tue, 9 Mar 2010 15:10:22 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: About libssh2_channel_read Mail-Followup-To: libssh2-devel@cool.haxx.se References: <832690.68608.qm@web24404.mail.ird.yahoo.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <832690.68608.qm@web24404.mail.ird.yahoo.com> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se sebastien isquierdo wrote: > I'm trying now to make 2 or more connections to MySQL through SSH. .. > - create a non-blocking session (libssh2_session_init, ) Good, this is important. > - for each socket write on the good channel (libssh2_channel_write) > - and listen to the good channel to (libssh2_channel_read) > > I make a new thread for reading on every channel. Note that libssh2 is not at all thread safe. You need a global mutex for the entire session until libssh2_channel_write() or libssh2_channel_read() has transfered the full data you want to transfer. > My problem is libssh2_channel_read returns informations only from > on the last created channel and nothing from the others. Please send a libssh2 debug log from establishing one connection, passing data, then establishing a second connection, and trying to pass data, in that order. Let's see if we can spot the problem. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 15:16:44 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29EGcl2021102; Tue, 9 Mar 2010 15:16:42 +0100 Received: from mail-ew0-f226.google.com (mail-ew0-f226.google.com [209.85.219.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29EGac4021075 for ; Tue, 9 Mar 2010 15:16:37 +0100 Received: by ewy26 with SMTP id 26so3996871ewy.3 for ; Tue, 09 Mar 2010 06:16:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=QuqZw9Xm0j4sBNDcLmXmyBP/OSwYHejDfzAd1cYTs54=; b=ae5So2EbV/PIuB7vDY7u+Vr3eMc6aTQp4nCgxx8lzu40fm7tS4DsTNDNyt2jykM0zV hHrmdlx0F/xc8QvwigXPo5H4cX+l2AMgez0uY7SJTrZMRvCOOtt2FBC70gX0+8HcL4Mo GgJyFPNUvOPow9S82sNXDB2S603wzIm3hD60Q= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=HefyqvPEkZJMbyRdx5jxBQGYUsfMkTNtj42mIr6KRgee4N69pmrzU1CnQzdGj4E4lm Q0vRr72hlgm8j4wvgebWc/T5xLLoaYVlje92IurioYA6a/DzeIP7u3nfG+ikbleEDm64 /ps+h4dwSz2hr+hNwJO5aM7IsCuGzQ7H9FL/4= MIME-Version: 1.0 Received: by 10.213.24.21 with SMTP id t21mr3911001ebb.99.1268144192644; Tue, 09 Mar 2010 06:16:32 -0800 (PST) In-Reply-To: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> Date: Tue, 9 Mar 2010 14:16:32 +0000 X-Google-Sender-Auth: aa78219ff897ce03 Message-ID: <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 9 March 2010 14:01, Steven Van Ingelgem wrote: > what I did was using the ULONGLONG instead + removing the "ULL" after the > numbers. Could removing the ULL cause any problems with other versions of MSVC? If not, I'll commit a fix based on this. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 15:54:56 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29Esnn7012528; Tue, 9 Mar 2010 15:54:53 +0100 Received: from hrndva-omtalb.mail.rr.com (hrndva-omtalb.mail.rr.com [71.74.56.125]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29EskKL012480 for ; Tue, 9 Mar 2010 15:54:46 +0100 X-Authority-Analysis: v=1.0 c=1 a=0O5caf2HW_cA:10 a=O-pu0537hIgA:10 a=kj9zAlcOel0A:10 a=UJ5Y5Z__AAAA:8 a=Kuz3bnEbIxOXsbb0faEA:9 a=hNl0nHuvpbPCMrfXpjwO_wcXX74A:4 a=CjuIK1q_8ugA:10 X-Cloudmark-Score: 0 X-Originating-IP: 70.116.71.237 Received: from [70.116.71.237] ([70.116.71.237:46779] helo=dc7100.roadrunn.com) by hrndva-oedge03.mail.rr.com (envelope-from ) (ecelerity 2.2.2.39 r()) with ESMTP id E9/2A-22310-131669B4; Tue, 09 Mar 2010 14:54:41 +0000 Received: by dc7100.roadrunn.com (Postfix, from userid 501) id 60054830A2D; Tue, 9 Mar 2010 09:48:42 -0600 (CST) X-Spam-Checker-Version: SpamAssassin 3.2.4 (2008-01-01) on dc7100.roadrunn.com X-Spam-Level: X-Spam-Status: No, score=-4.3 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.4 Received: from ngierman3 (dc7100 [192.168.1.6]) by dc7100.roadrunn.com (Postfix) with ESMTP id 9EAFB830A2A for ; Tue, 9 Mar 2010 09:48:37 -0600 (CST) From: "Neil Gierman" To: "'libssh2 development'" References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> In-Reply-To: <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> Subject: RE: Visual Studio 6 compilation issues (libssh2 1.2.4) Date: Tue, 9 Mar 2010 08:54:37 -0600 Message-ID: <001501cabf98$702b2390$50816ab0$@com> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Acq/nAOeh9KjHOupRJ+2J2FxMIpRFQAA68TA Content-Language: en-us X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se If someone could send a patch, I can test against Visual Studio 2008 (Win32/x64/Itanium). > -----Original Message----- > From: libssh2-devel-bounces@cool.haxx.se [mailto:libssh2-devel- > bounces@cool.haxx.se] On Behalf Of Alexander Lamaison > Sent: Tuesday, March 09, 2010 8:17 AM > To: libssh2 development > Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) > > On 9 March 2010 14:01, Steven Van Ingelgem > wrote: > > what I did was using the ULONGLONG instead + removing the "ULL" after > the > > numbers. > > Could removing the ULL cause any problems with other versions of MSVC? > If not, I'll commit a fix based on this. > > Alex > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2- > devel _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 16:14:16 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29FE4hx028886; Tue, 9 Mar 2010 16:14:12 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.27]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29FE2Bn028868 for ; Tue, 9 Mar 2010 16:14:02 +0100 Received: by ey-out-2122.google.com with SMTP id 9so928985eyd.31 for ; Tue, 09 Mar 2010 07:14:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=3pH3JeIrUa0Q2gCj/hMGVxUiGu91KCbY9XkICkYVM5c=; b=dB+N4BwumBH4kM2RZaqcthEPyNqP5YM7XPnLIq0V8xlTiozK3SGALhfcySf187Tx4b Wl1isM5YmEsEM+rBVqKMakYwU6SoIhd0Pjztt0OE/c/Wsh1S/Qly5YHfYdCbsbyTgq98 23eGiePl52GlW7Y6DK2048xY6D0eVfVeZ1dOM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=tHU8g9VlzDtWg8ONJdYvZQNGie99H03yvykWgL2a6VRLQC3gPYkuzRBL+7Wp27eOkq JAVznwq/ZtRPoHtUcqHQ2jPCu22p/Wn++Fr7Jn/MUvyjRi5+7+0zgiz22oxETiSCizaH ugPtZhYGJQg9bupBtZW/XyqWI+XLGy/r9SuFQ= MIME-Version: 1.0 Received: by 10.213.96.206 with SMTP id i14mr4214685ebn.74.1268147642686; Tue, 09 Mar 2010 07:14:02 -0800 (PST) In-Reply-To: <001501cabf98$702b2390$50816ab0$@com> References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> Date: Tue, 9 Mar 2010 15:14:02 +0000 X-Google-Sender-Auth: ab84c94953735dbb Message-ID: <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) From: Alexander Lamaison To: libssh2 development Content-Type: multipart/mixed; boundary=001636c5bd9bc3dc9104815fa1d8 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --001636c5bd9bc3dc9104815fa1d8 Content-Type: text/plain; charset=ISO-8859-1 On 9 March 2010 14:54, Neil Gierman wrote: > If someone could send a patch, I can test against Visual Studio 2008 > (Win32/x64/Itanium). Try this. Seems ok on VS 2005. --001636c5bd9bc3dc9104815fa1d8 Content-Type: application/octet-stream; name="gettimofday_vc6_fix.patch" Content-Disposition: attachment; filename="gettimofday_vc6_fix.patch" Content-Transfer-Encoding: base64 X-Attachment-Id: f_g6kudj050 ZGlmZiAtLWdpdCBhL3NyYy9taXNjLmMgYi9zcmMvbWlzYy5jCmluZGV4IDc5OGIyZTYuLjE5MzU1 NTMgMTAwNjQ0Ci0tLSBhL3NyYy9taXNjLmMKKysrIGIvc3JjL21pc2MuYwpAQCAtNTI4LDIyICs1 MjgsMjIgQEAgdm9pZCBfbGlic3NoMl9saXN0X2luc2VydChzdHJ1Y3QgbGlzdF9ub2RlICphZnRl ciwgLyogaW5zZXJ0IGJlZm9yZSB0aGlzICovCiAgKi8KIAogLyogT2Zmc2V0IGJldHdlZW4gMS8x LzE2MDEgYW5kIDEvMS8xOTcwIGluIDEwMCBuYW5vc2VjIHVuaXRzICovCi0jZGVmaW5lIF9XMzJf RlRfT0ZGU0VUICgxMTY0NDQ3MzYwMDAwMDAwMDBVTEwpCisjZGVmaW5lIF9XMzJfRlRfT0ZGU0VU ICgxMTY0NDQ3MzYwMDAwMDAwMDApCiAKIAogaW50IF9fY2RlY2wgZ2V0dGltZW9mZGF5KHN0cnVj dCB0aW1ldmFsICp0cCwKICAgICAgICAgICAgICAgICAgICAgICAgICB2b2lkICp0enApCiAgewog ICB1bmlvbiB7Ci0gICAgdW5zaWduZWQgbG9uZyBsb25nIG5zMTAwOyAvKnRpbWUgc2luY2UgMSBK YW4gMTYwMSBpbiAxMDBucyB1bml0cyAqLworICAgIHVuc2lnbmVkIF9faW50NjQgbnMxMDA7IC8q dGltZSBzaW5jZSAxIEphbiAxNjAxIGluIDEwMG5zIHVuaXRzICovCiAgICAgRklMRVRJTUUgZnQ7 CiAgIH0gIF9ub3c7CiAKICAgaWYodHApCiAgICAgewogICAgICAgR2V0U3lzdGVtVGltZUFzRmls ZVRpbWUgKCZfbm93LmZ0KTsKLSAgICAgIHRwLT50dl91c2VjPShsb25nKSgoX25vdy5uczEwMCAv IDEwVUxMKSAlIDEwMDAwMDBVTEwgKTsKLSAgICAgIHRwLT50dl9zZWM9IChsb25nKSgoX25vdy5u czEwMCAtIF9XMzJfRlRfT0ZGU0VUKSAvIDEwMDAwMDAwVUxMKTsKKyAgICAgIHRwLT50dl91c2Vj PShsb25nKSgoX25vdy5uczEwMCAvIDEwKSAlIDEwMDAwMDAgKTsKKyAgICAgIHRwLT50dl9zZWM9 IChsb25nKSgoX25vdy5uczEwMCAtIF9XMzJfRlRfT0ZGU0VUKSAvIDEwMDAwMDAwKTsKICAgICB9 CiAgIC8qIEFsd2F5cyByZXR1cm4gMCBhcyBwZXIgT3BlbiBHcm91cCBCYXNlIFNwZWNpZmljYXRp b25zIElzc3VlIDYuCiAgICAgIERvIG5vdCBzZXQgZXJybm8gb24gZXJyb3IuICAqLwo= --001636c5bd9bc3dc9104815fa1d8 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --001636c5bd9bc3dc9104815fa1d8-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 17:41:21 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29Geolk026829; Tue, 9 Mar 2010 17:41:00 +0100 Received: from hrndva-omtalb.mail.rr.com (hrndva-omtalb.mail.rr.com [71.74.56.123]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29Gell3026680 for ; Tue, 9 Mar 2010 17:40:48 +0100 X-Authority-Analysis: v=1.0 c=1 a=0O5caf2HW_cA:10 a=O-pu0537hIgA:10 a=kj9zAlcOel0A:10 a=yMhMjlubAAAA:8 a=tvAvMhc3U6kudSdo-NQA:9 a=YYQ7Qwktk7OOYn-Qq8EA:7 a=mGsKqi64qEaFDP5anZNZbe_hKQsA:4 a=CjuIK1q_8ugA:10 a=V0VK7F0muM3rOWE1:21 a=M_SYuPgKGrTHanXT:21 X-Cloudmark-Score: 0 X-Originating-IP: 70.116.71.237 Received: from [70.116.71.237] ([70.116.71.237:46865] helo=dc7100.roadrunn.com) by hrndva-oedge03.mail.rr.com (envelope-from ) (ecelerity 2.2.2.39 r()) with ESMTP id 7B/C6-22310-B0A769B4; Tue, 09 Mar 2010 16:40:43 +0000 Received: by dc7100.roadrunn.com (Postfix, from userid 501) id B7516830A2D; Tue, 9 Mar 2010 11:34:43 -0600 (CST) X-Spam-Checker-Version: SpamAssassin 3.2.4 (2008-01-01) on dc7100.roadrunn.com X-Spam-Level: X-Spam-Status: No, score=-4.3 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.4 Received: from ngierman3 (dc7100 [192.168.1.6]) by dc7100.roadrunn.com (Postfix) with ESMTP id 54CCF830A2A for ; Tue, 9 Mar 2010 11:34:39 -0600 (CST) From: "Neil Gierman" To: "'libssh2 development'" References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> In-Reply-To: <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> Subject: RE: Visual Studio 6 compilation issues (libssh2 1.2.4) Date: Tue, 9 Mar 2010 10:40:38 -0600 Message-ID: <000601cabfa7$3fd640d0$bf82c270$@com> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Acq/ov6QMc12zGqJT+eyT7uvYJyhBQAACpPA Content-Language: en-us X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se > > If someone could send a patch, I can test against Visual Studio 2008 > > (Win32/x64/Itanium). > > Try this. Seems ok on VS 2005. Well that patch didn't break things although there is another issue. I noticed 446 compiler warnings related to data type conversions when compiling in VS2008 Itanium mode. They are mostly size_t to unsigned long/int conversion warnings. We have defined that in most places the length of messages and packets to be unsigned long or unsigned int. In most places we are storing the return of strlen() and other related function in to our length variables. An example is the definition of libssh2_error() where we are storing the return of strlen(errmsg) into session->err_msglen. Strlen returns size_t and in Win64 builds, size_t = unsigned __int64. Long = 4 bytes, __int64 = 8bytes (http://msdn.microsoft.com/en-us/library/s3f49ktz(VS.80).aspx). There are quite a number of changes to sync up data types to match size_t and I don't know if some of those would break the API/ABI. Or we could go the easy way and cast the returns of strlen, etc to unsigned long (or whatever we want it as) however I think that is masking a future problem. _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 19:08:19 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29I81Ew001638; Tue, 9 Mar 2010 19:08:14 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o29I801U001594 for ; Tue, 9 Mar 2010 19:08:00 +0100 Received: (qmail 26879 invoked by uid 501); 9 Mar 2010 18:08:01 -0000 Message-ID: <20100309180801.26878.qmail@stuge.se> Date: Tue, 9 Mar 2010 19:08:01 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) Mail-Followup-To: libssh2-devel@cool.haxx.se References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <000601cabfa7$3fd640d0$bf82c270$@com> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Neil Gierman wrote: > in Win64 builds, size_t = unsigned __int64. Long = 4 bytes, > __int64 = 8bytes > (http://msdn.microsoft.com/en-us/library/s3f49ktz(VS.80).aspx). unsigned long used to be *the* way to get the maximum length CPU-native data type, until 64-bit Windows.. Thanks, Microsoft. :( I'm not sure if we can deal with it without changing the API? //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 20:28:21 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29JS97X026060; Tue, 9 Mar 2010 20:28:18 +0100 Received: from mail.intersystems.com (mail.intersys.COM [198.133.74.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29JS5VZ025866 for ; Tue, 9 Mar 2010 20:28:06 +0100 X-InterSystems: Sent from InterSystems X-InterSystems: Sent from InterSystems From: Dave McCaldon Mime-Version: 1.0 (Apple Message framework v1077) Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) Date: Tue, 9 Mar 2010 14:27:59 -0500 In-Reply-To: <20100309180801.26878.qmail@stuge.se> To: libssh2 development References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <20100309180801.26878.qmail@stuge.se> Message-Id: <113AAF8F-E2E0-45AB-B32B-BFB7035B8D13@intersystems.com> X-Mailer: Apple Mail (2.1077) X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0305915527==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0305915527== Content-Type: multipart/alternative; boundary=Apple-Mail-136--54613089 --Apple-Mail-136--54613089 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii On Mar 9, 2010, at 1:08 PM, Peter Stuge wrote: > Neil Gierman wrote: >> in Win64 builds, size_t =3D unsigned __int64. Long =3D 4 bytes, >> __int64 =3D 8bytes >> (http://msdn.microsoft.com/en-us/library/s3f49ktz(VS.80).aspx). >=20 > unsigned long used to be *the* way to get the maximum length > CPU-native data type, until 64-bit Windows.. Thanks, Microsoft. :( __int64 has been around for a while (even on 32-bit systems), I don't = know why they didn't just use "long long" like everyone else. I guess = when they went to 64-bit, they didn't want to break all of the badly = written code that presumed a long was 32-bits. I guess that makes it a = LLP64 platform: http://www.unix.org/version2/whatsnew/lp64_wp.html > I'm not sure if we can deal with it without changing the API? Most sane 64-bit systems are LP64, so it's likely that size_t =3D=3D = unsigned long. The same should be true for most 32-bit systems, size_t = =3D=3D unsigned long. --Apple-Mail-136--54613089 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=us-ascii
Neil = Gierman wrote:
in Win64 builds, size_t =3D = unsigned __int64. Long =3D 4 bytes,
__int64 =3D 8bytes
(http= ://msdn.microsoft.com/en-us/library/s3f49ktz(VS.80).aspx).

unsigned long used to be *the* way to get the maximum = length
CPU-native data type, until 64-bit Windows.. Thanks, = Microsoft. :(

__int64 has been = around for a while (even on 32-bit systems), I don't know why they = didn't just use "long long" like everyone else.  I guess when they = went to 64-bit, they didn't want to break all of the badly written code = that presumed a long was 32-bits.  I guess that makes it a LLP64 = platform:


I'm not sure if we can deal with it without changing = the API?

Most sane 64-bit systems = are LP64, so it's likely that size_t =3D=3D unsigned long.  The = same should be true for most 32-bit systems, size_t =3D=3D unsigned = long.


= --Apple-Mail-136--54613089-- --===============0305915527== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0305915527==-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 22:25:22 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29LPAM2023605; Tue, 9 Mar 2010 22:25:18 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29LP8qi023525 for ; Tue, 9 Mar 2010 22:25:08 +0100 Date: Tue, 9 Mar 2010 22:25:08 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Setting timeout in blocking mode In-Reply-To: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> Message-ID: References: <4B962A640200003F00008B2F@IPEMAIL.ipetronik.de> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 9 Mar 2010, Sebastien Fricker wrote: > I'm currently wondering why there is now wait so specify a timeout when > working on blocking mode. See http://www.libssh2.org/mail/libssh2-devel-archive-2010-02/0073.shtml and http://libssh2.stuge.se/ticket/160 for details around how it can be done and why it hasn't been done yet. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 9 22:43:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29LhOxE004936; Tue, 9 Mar 2010 22:43:27 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o29LhNC1004911 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 9 Mar 2010 22:43:24 +0100 Received: (qmail 20247 invoked from network); 9 Mar 2010 21:43:24 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 9 Mar 2010 21:43:24 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Tue, 09 Mar 2010 21:43:24 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:5 Message-ID: <054.c006cba1e26e46bab2bcd88b36cf6314@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- Comment(by bagder): Replying to [comment:4 alamaison]: > Replying to [comment:3 bagder]: > > Right, we would use a new version of libssh2_knownhost_add() that supports the comment part. > > Do you mean adding an extra parameter to `add` (this wouldn't change the ABI, right?) I'm quite sure we can't safely add arguments without changing the ABI. So I meant: > or adding a new version of `add`, say `libssh2_knownhost_add_with_comment` that > takes a separate comment parameter. That's what I meant. And when we add that, we mark the first one as deprecated and we add a mention in the TODO to remove it completely at the next soname bump. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 09:13:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A8BPbh015447; Wed, 10 Mar 2010 09:11:37 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A8BLR8015435 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 09:11:21 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2A8BJvA027756 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 09:11:20 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::fzsUAPp6+qAp/E/0:3rRr Date: Wed, 10 Mar 2010 09:11:19 +0100 In-Reply-To: <000601cabfa7$3fd640d0$bf82c270$@com> (Neil Gierman's message of "Tue, 9 Mar 2010 10:40:38 -0600") Message-ID: <87tysolhp4.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se "Neil Gierman" writes: > I noticed 446 compiler warnings related to data type conversions when > compiling in VS2008 Itanium mode. They are mostly size_t to unsigned > long/int conversion warnings. We have defined that in most places the length > of messages and packets to be unsigned long or unsigned int. In most places > we are storing the return of strlen() and other related function in to our > length variables. An example is the definition of libssh2_error() where we > are storing the return of strlen(errmsg) into session->err_msglen. Strlen > returns size_t and in Win64 builds, size_t = unsigned __int64. Long = 4 > bytes, __int64 = 8bytes > (http://msdn.microsoft.com/en-us/library/s3f49ktz(VS.80).aspx). There are > quite a number of changes to sync up data types to match size_t and I don't > know if some of those would break the API/ABI. Or we could go the easy way > and cast the returns of strlen, etc to unsigned long (or whatever we want it > as) however I think that is masking a future problem. Agreed. Libssh2 is broken here, the Right Thing is to always use size_t for string lengths. This change will break the ABI if we don't rename all the functions though... /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 10:31:38 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A9VJMF015084; Wed, 10 Mar 2010 10:31:27 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A9VHgu015077 for ; Wed, 10 Mar 2010 10:31:17 +0100 Date: Wed, 10 Mar 2010 10:31:17 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) In-Reply-To: <87tysolhp4.fsf@mocca.josefsson.org> Message-ID: References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 10 Mar 2010, Simon Josefsson wrote: > Agreed. Libssh2 is broken here, the Right Thing is to always use size_t for > string lengths. This change will break the ABI if we don't rename all the > functions though... I agree. But we can probably clean up a good deal of internals first to use size_t and only do the translations from int or longs to size_t until we fix the APIs to use size_t properly (which should be added to the TODO). -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 10:57:07 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A9tHV2000683; Wed, 10 Mar 2010 10:55:21 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A9tFIM000677 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 10:55:15 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2A9tD8b030131 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 10:55:15 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::M/A2aUqMfAQannea:MWWu Date: Wed, 10 Mar 2010 10:55:14 +0100 In-Reply-To: (Daniel Stenberg's message of "Wed, 10 Mar 2010 10:31:17 +0100 (CET)") Message-ID: <87sk88jybh.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Wed, 10 Mar 2010, Simon Josefsson wrote: > >> Agreed. Libssh2 is broken here, the Right Thing is to always use >> size_t for string lengths. This change will break the ABI if we >> don't rename all the functions though... > > I agree. But we can probably clean up a good deal of internals first > to use size_t and only do the translations from int or longs to size_t > until we fix the APIs to use size_t properly (which should be added to > the TODO). Yes. Once that is done, we can replace all API entry points with small interface functions that behave like this: int libssh2_session_last_error(LIBSSH2_SESSION *session, char **errmsg, int *errmsg_len, int want_buf) { size_t tmp = *errmsg_len int ret = libssh2_session_last_error_real(session, errmsg, &tmp, want_buf); *errmsg_len = tmp; return ret; } int libssh2_session_last_error_real(LIBSSH2_SESSION *session, char **errmsg, size_t *errmsg_len, int want_buf); ... Or whatever, I'm not even sure the casting is correct, I'm just thinking generally. Then we could isolate all the broken non-size_t using code into just one file. The rest of libssh2 could then use size_t consistently. Btw, maybe we can add an libssh2_session_error_last with size_t in the next version, and deprecate libssh2_session_last_error? Then at least one interface will be fixed... /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 10:58:50 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A9wlGf002142; Wed, 10 Mar 2010 10:58:48 +0100 Received: from dottedmag.net (burger.dottedmag.net [212.75.37.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2A9wYgl002014 for ; Wed, 10 Mar 2010 10:58:34 +0100 Received: from vertex.dottedmag (unknown [91.197.127.125]) by dottedmag.net (Postfix) with ESMTPSA id C8EDE8C761 for ; Wed, 10 Mar 2010 10:58:34 +0100 (CET) Received: from dottedmag by vertex.dottedmag with local (Exim 4.69) (envelope-from ) id 1NpIgT-0001EA-AS for libssh2-devel@cool.haxx.se; Wed, 10 Mar 2010 15:58:33 +0600 From: Mikhail Gusarov To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> Date: Wed, 10 Mar 2010 15:58:32 +0600 In-Reply-To: <87tysolhp4.fsf@mocca.josefsson.org> (Simon Josefsson's message of "Wed, 10 Mar 2010 09:11:19 +0100") Message-ID: <87wrxk8pmf.fsf@vertex.dottedmag> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0751443713==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0751443713== Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --=-=-= Content-Transfer-Encoding: quoted-printable Twas brillig at 09:11:19 10.03.2010 UTC+01 when simon@josefsson.org did gyr= e and gimble: SJ> Agreed. Libssh2 is broken here, the Right Thing is to always use SJ> size_t for string lengths. This change will break the ABI if we SJ> don't rename all the functions though... It will break ABI only for architectures where size_t !=3D unsigned long, r= ight? =2D-=20 http://fossarchy.blogspot.com/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iQIcBAEBAgAGBQJLl21JAAoJEJ0g9lA+M4iIQjIP/3jhTDldNDu1boPq2cj0NkZG muQbnYjXrq4WQhxI6qIc3Hu5bXPEsR2lO1nh23AmLiLPJ4PI8O7GSIGsy1AMLnrN SvrkFvNR+wGDJw+/I+J2v2puNRvxQ6pDbL6adkdLwIFeoe7xAIcMa2VvfGjqdsLA PTyqKSD8XWSQgenp2G4Iz8PmWB4+6Iz2rwZhbsFgBGzzK3i/NAgSICKYYMin4uo5 cGqb2eJ5+xtSSW6oDFHDw69dVhbIDY5B91Q2MRAIESl+SojJrGAQhJXCsdikTIZt qLdFCHZUiKLQlDIb5ijl7fR0+hUowyJ/0/f2T/kfF+29NQtgX2rF/Dv3D4oVAMsy LQmJKlAwPhNIIQtfM7KylokNcjGOuml06zzdbo9AWby8xrnzZ0r0OliqVOIj65Ob PEoXr38jAjgkKpuI2ok2N9jTB1oXX3lHgxK5d8VR7aeGNH3xLJqtEgzU+qEHzf3c MfMNWIn8JLYW0v7uNwS6fLYe9swJ5NncEAZtdbS01LQYf8CCcowcaFB9eiQ6ao6j OvszXpJvLTV+cb8hyOGm8vK8EZ1Zkc/sA6MxPobZdKDp7RDRAiY0am4geEWSBnqj yIXcctGPIHxzfnF+fJ5HMcUZ0uSXzT6z7M/hK59WWq7+B0xZPKRxyj636TAVEtu1 ySWh9kDSvX260En6HpwV =DNjE -----END PGP SIGNATURE----- --=-=-=-- --===============0751443713== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0751443713==-- From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 11:01:42 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AA1ajC005491; Wed, 10 Mar 2010 11:01:40 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AA1Ziv005450 for ; Wed, 10 Mar 2010 11:01:35 +0100 Date: Wed, 10 Mar 2010 11:01:35 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) In-Reply-To: <87sk88jybh.fsf@mocca.josefsson.org> Message-ID: References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 10 Mar 2010, Simon Josefsson wrote: > Then we could isolate all the broken non-size_t using code into just one > file. The rest of libssh2 could then use size_t consistently. That's a great idea! > Btw, maybe we can add an libssh2_session_error_last with size_t in the next > version, and deprecate libssh2_session_last_error? Then at least one > interface will be fixed... Yes indeed. Once we have the size_t-flaws separated in to its own file, we can also easier mark functions that are deprecated and those that still haven't been properly fixed. When all the size_t functions have fixed alternatives, we can seriously consider doing the dreaded soname bump. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 11:07:13 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AA77JD013839; Wed, 10 Mar 2010 11:07:11 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AA76Om013833 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 11:07:06 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2AA74Kj030341 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 11:07:06 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87wrxk8pmf.fsf@vertex.dottedmag> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::eeoErreDoI1iwEpo:7BTD Date: Wed, 10 Mar 2010 11:07:04 +0100 In-Reply-To: <87wrxk8pmf.fsf@vertex.dottedmag> (Mikhail Gusarov's message of "Wed, 10 Mar 2010 15:58:32 +0600") Message-ID: <87bpewjxrr.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Mikhail Gusarov writes: > Twas brillig at 09:11:19 10.03.2010 UTC+01 when simon@josefsson.org did gyre and gimble: > > SJ> Agreed. Libssh2 is broken here, the Right Thing is to always use > SJ> size_t for string lengths. This change will break the ABI if we > SJ> don't rename all the functions though... > > It will break ABI only for architectures where size_t != unsigned long, right? Yes... However I think libssh2 is deployed on many such systems already that we need to care about them. We _could_ bump the ABI only on the affected platforms, but IMHO maintaining that will be a pain and the risk for build mistakes are rather high. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 11:11:01 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AAAwWf015943; Wed, 10 Mar 2010 11:11:00 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AAAvct015935 for ; Wed, 10 Mar 2010 11:10:57 +0100 Date: Wed, 10 Mar 2010 11:10:57 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) In-Reply-To: <87bpewjxrr.fsf@mocca.josefsson.org> Message-ID: References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87pr3dpr9w.fsf@mocca.josefsson.org> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87wrxk8pmf.fsf@vertex.dottedmag> <87bpewjxrr.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 10 Mar 2010, Simon Josefsson wrote: > We _could_ bump the ABI only on the affected platforms, but IMHO maintaining > that will be a pain and the risk for build mistakes are rather high. Yes, that'll quickly turn into a nightmare. I think we should consider any ABI breakage on any platform libssh2 runs on to be "an ABI breakage". It's also better to play it safe and conservative as then we reduce the risk of accidentally breaking the ABI on a platform we didn't consider or even know about. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 12:26:52 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ABQUH5004865; Wed, 10 Mar 2010 12:26:40 +0100 Received: from av11-2-sn2.hy.skanova.net (av11-2-sn2.hy.skanova.net [81.228.8.184]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ABQTZG004791 for ; Wed, 10 Mar 2010 12:26:29 +0100 Received: by av11-2-sn2.hy.skanova.net (Postfix, from userid 502) id AA17438253; Wed, 10 Mar 2010 12:26:24 +0100 (CET) Received: from smtp4-2-sn2.hy.skanova.net (smtp4-2-sn2.hy.skanova.net [81.228.8.93]) by av11-2-sn2.hy.skanova.net (Postfix) with ESMTP id 7C78E3820C for ; Wed, 10 Mar 2010 12:26:24 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp4-2-sn2.hy.skanova.net (Postfix) with ESMTP id 6962237E4A for ; Wed, 10 Mar 2010 12:26:24 +0100 (CET) Message-ID: <4B9781DE.2060707@SDlabs.se> Date: Wed, 10 Mar 2010 12:26:22 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2 development Subject: patch to add libssh2_init() and libssh2_exit() Content-Type: multipart/mixed; boundary="------------010904040308050007040008" X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --------------010904040308050007040008 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Please find attached patch to make libssh2 have a clean way of initialize the library. The patch adds a new file, global.c. The makefile.in is changes, but not the WIN32 files. session.c is changed to call libssh2_init only if the global variable libssh2_initialized is zero. The implicit load och ciphers in openssl.c has been removed. /Lars From libssh2.h: /* * libssh2_init() * * Initialize the libssh2 functions. flags can be: * LIBSSH2_INIT_ALL: Initialize everything * LIBSSH2_INIT_NO_CRYPTO_INIT: Do not initialize the crypto library (ie. * OPENSSL_add_cipher_algoritms() for OpenSSL * * Returns 0 if succeeded, or a negative value for error. */ LIBSSH2_API int libssh2_init(int flags); /* * libssh2_exit() * * Exit the libssh2 functions and free's all memory used internal. */ LIBSSH2_API void libssh2_exit(); --------------010904040308050007040008 Content-Type: text/plain; name="libssh2_initexit.diff" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="libssh2_initexit.diff" Only in libssh2-1.2.5-20100309_orig/: ChangeLog diff -ru libssh2-1.2.5-20100309_orig/include/libssh2.h libssh2-1.2.5-20100309/include/libssh2.h --- libssh2-1.2.5-20100309_orig/include/libssh2.h 2010-03-09 06:06:36.000000000 +0100 +++ libssh2-1.2.5-20100309/include/libssh2.h 2010-03-10 11:58:23.000000000 +0100 @@ -996,6 +996,26 @@ void* context, libssh2_trace_handler_func callback); +#define LIBSSH2_INIT_ALL 0x0000 +#define LIBSSH2_INIT_NO_CRYPTO_INIT 0x0001 +/* + * libssh2_init() + * + * Initialize the libssh2 functions. flags can be: + * LIBSSH2_INIT_ALL: Initialize everything + * LIBSSH2_INIT_NO_CRYPTO_INIT: Do not initialize the crypto library (ie. + * OPENSSL_add_cipher_algoritms() for OpenSSL + * + * Returns 0 if succeeded, or a negative value for error. + */ +LIBSSH2_API int libssh2_init(int flags); +/* + * libssh2_exit() + * + * Exits the libssh2 function and free's all memory used internal. + */ +LIBSSH2_API void libssh2_exit(); + #ifdef __cplusplus } /* extern "C" */ #endif Only in libssh2-1.2.5-20100309/src: global.c diff -ru libssh2-1.2.5-20100309_orig/src/libgcrypt.h libssh2-1.2.5-20100309/src/libgcrypt.h --- libssh2-1.2.5-20100309_orig/src/libgcrypt.h 2009-11-17 06:06:05.000000000 +0100 +++ libssh2-1.2.5-20100309/src/libgcrypt.h 2010-03-10 12:03:22.000000000 +0100 @@ -93,6 +93,7 @@ #define libssh2_hmac_cleanup(ctx) gcry_md_close (*ctx); #define libssh2_crypto_init() gcry_control (GCRYCTL_DISABLE_SECMEM) +#define libssh2_crypto_exit() #define libssh2_rsa_ctx struct gcry_sexp diff -ru libssh2-1.2.5-20100309_orig/src/libssh2_priv.h libssh2-1.2.5-20100309/src/libssh2_priv.h --- libssh2-1.2.5-20100309_orig/src/libssh2_priv.h 2010-03-04 06:06:04.000000000 +0100 +++ libssh2-1.2.5-20100309/src/libssh2_priv.h 2010-03-10 12:06:06.000000000 +0100 @@ -1242,5 +1242,6 @@ #define ARRAY_SIZE(a) (sizeof ((a)) / sizeof ((a)[0])) +extern int libssh2_initialized; #endif /* LIBSSH2_H */ diff -ru libssh2-1.2.5-20100309_orig/src/Makefile.in libssh2-1.2.5-20100309/src/Makefile.in --- libssh2-1.2.5-20100309_orig/src/Makefile.in 2010-03-09 06:06:16.000000000 +0100 +++ libssh2-1.2.5-20100309/src/Makefile.in 2010-03-10 11:59:28.000000000 +0100 @@ -65,7 +65,7 @@ am__objects_1 = channel.lo comp.lo crypt.lo hostkey.lo kex.lo mac.lo \ misc.lo packet.lo publickey.lo scp.lo session.lo sftp.lo \ userauth.lo transport.lo version.lo knownhost.lo agent.lo \ - openssl.lo libgcrypt.lo pem.lo + openssl.lo libgcrypt.lo pem.lo global.lo am__objects_2 = am_libssh2_la_OBJECTS = $(am__objects_1) $(am__objects_2) libssh2_la_OBJECTS = $(am_libssh2_la_OBJECTS) @@ -217,7 +217,7 @@ AUTOMAKE_OPTIONS = foreign nostdinc CSOURCES = channel.c comp.c crypt.c hostkey.c kex.c mac.c misc.c \ packet.c publickey.c scp.c session.c sftp.c userauth.c transport.c \ - version.c knownhost.c agent.c openssl.c libgcrypt.c pem.c + version.c knownhost.c agent.c openssl.c libgcrypt.c pem.c global.c HHEADERS = libssh2_priv.h openssl.h libgcrypt.h transport.h channel.h comp.h mac.h misc.h diff -ru libssh2-1.2.5-20100309_orig/src/openssl.c libssh2-1.2.5-20100309/src/openssl.c --- libssh2-1.2.5-20100309_orig/src/openssl.c 2010-03-02 06:06:07.000000000 +0100 +++ libssh2-1.2.5-20100309/src/openssl.c 2010-03-10 12:02:17.000000000 +0100 @@ -358,14 +358,6 @@ (pem_read_bio_func) &PEM_read_bio_RSAPrivateKey; (void) session; - if (!EVP_get_cipherbyname("des")) { -/* If this cipher isn't loaded it's a pretty good indication that none are. - * I have *NO DOUBT* that there's a better way to deal with this ($#&%#$(%$#( - * Someone buy me an OpenSSL manual and I'll read up on it. - */ - OpenSSL_add_all_ciphers(); - } - return read_private_key_from_file((void **) rsa, read_rsa, filename, passphrase); } @@ -380,14 +372,6 @@ (pem_read_bio_func) &PEM_read_bio_DSAPrivateKey; (void) session; - if (!EVP_get_cipherbyname("des")) { -/* If this cipher isn't loaded it's a pretty good indication that none are. - * I have *NO DOUBT* that there's a better way to deal with this ($#&%#$(%$#( - * Someone buy me an OpenSSL manual and I'll read up on it. - */ - OpenSSL_add_all_ciphers(); - } - return read_private_key_from_file((void **) dsa, read_dsa, filename, passphrase); } @@ -461,4 +445,15 @@ } #endif /* LIBSSH_DSA */ +void libssh2_crypto_init() +{ + ERR_load_crypto_strings(); + OpenSSL_add_all_algorithms(); + ENGINE_load_builtin_engines(); +} + +void libssh2_crypto_exit() +{ +} + #endif /* !LIBSSH2_LIBGCRYPT */ diff -ru libssh2-1.2.5-20100309_orig/src/openssl.h libssh2-1.2.5-20100309/src/openssl.h --- libssh2-1.2.5-20100309_orig/src/openssl.h 2010-03-02 06:06:07.000000000 +0100 +++ libssh2-1.2.5-20100309/src/openssl.h 2010-03-10 12:02:43.000000000 +0100 @@ -130,8 +130,6 @@ #define libssh2_hmac_final(ctx, data) HMAC_Final(&(ctx), data, NULL) #define libssh2_hmac_cleanup(ctx) HMAC_cleanup(ctx) -#define libssh2_crypto_init() - #define libssh2_rsa_ctx RSA int _libssh2_rsa_new(libssh2_rsa_ctx ** rsa, diff -ru libssh2-1.2.5-20100309_orig/src/session.c libssh2-1.2.5-20100309/src/session.c --- libssh2-1.2.5-20100309_orig/src/session.c 2010-03-04 06:06:04.000000000 +0100 +++ libssh2-1.2.5-20100309/src/session.c 2010-03-10 12:00:34.000000000 +0100 @@ -456,7 +456,8 @@ session->api_block_mode = 1; /* blocking API by default */ _libssh2_debug(session, LIBSSH2_TRACE_TRANS, "New session resource allocated"); - libssh2_crypto_init(); + if (libssh2_initialized == 0) + libssh2_init(LIBSSH2_INIT_ALL); } return session; } Only in libssh2-1.2.5-20100309_orig/win32: libssh2.dsp --------------010904040308050007040008 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --------------010904040308050007040008-- From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 12:43:03 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ABgw37016002; Wed, 10 Mar 2010 12:43:01 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ABguHW015940 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 12:42:57 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2ABgt00000392 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 12:42:57 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: patch to add libssh2_init() and libssh2_exit() References: <4B9781DE.2060707@SDlabs.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::V5supUeBVmf0h4em:7NFg Date: Wed, 10 Mar 2010 12:42:55 +0100 In-Reply-To: <4B9781DE.2060707@SDlabs.se> (Lars Nordin's message of "Wed, 10 Mar 2010 12:26:22 +0100") Message-ID: <87y6i0ierk.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Lars Nordin writes: > Please find attached patch to make libssh2 have a clean way of > initialize the library. Hi. Great, I agree libssh2 should have proper global init/exit functions. > The patch adds a new file, global.c. The makefile.in is changes, but > not the WIN32 files. The Makefile.in file is automatically generated, the file to modify is Makefile.inc. You forgot to attach the new globals.c? > + * LIBSSH2_INIT_ALL: Initialize everything Maybe we could remove this one? I think it will restrict us too much in the future, we could conceivable have init-flags that are not related to disable some initialization, and then this flag becomes confusing. Just document that callers should pass 0 if they don't want to use any flag. > + ERR_load_crypto_strings(); > + OpenSSL_add_all_algorithms(); > + ENGINE_load_builtin_engines(); Are the first and last calls really needed? /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 13:02:46 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AC2dX4031136; Wed, 10 Mar 2010 13:02:43 +0100 Received: from mail-ew0-f226.google.com (mail-ew0-f226.google.com [209.85.219.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AC2bt9030991 for ; Wed, 10 Mar 2010 13:02:38 +0100 Received: by ewy26 with SMTP id 26so4494325ewy.3 for ; Wed, 10 Mar 2010 04:02:33 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=6qXMXcg1UcgTA0x4OGI3z/zfvYGQHU9EjPeJsiN0ebo=; b=Keq3ai/yZH6ysljYA7u7db8fU92DpIQyHoBt/NjYqTi5kWKXKjlIGPpIP+24Pkjc07 +hC2VVcH1o699ntufjYRuYP4aGZty1fUFG+2KU2u43ms3WENagHWta1x8EmQQ3neEFrz AV7ZhlRJq1UZF1I8Xz9f4fjRGe2bjUmwD92UQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=aERqJPBj3WtoseT59+7mXA37k+5zKVNpI5r2Ys6xOTwegib7QMFAtHE09bS0AC+O/n +uyHGvIhrvGSK9NRGHY0/MsACbqLjxTT5mZAlVSEucAVjXejRLOtMha+vu/EuOJvWJK2 KfXo2N0ujoY8nN9wTs0wL6azJjv6t8cT6/ffc= MIME-Version: 1.0 Received: by 10.213.100.151 with SMTP id y23mr821656ebn.78.1268222553309; Wed, 10 Mar 2010 04:02:33 -0800 (PST) In-Reply-To: <87sk88jybh.fsf@mocca.josefsson.org> References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> Date: Wed, 10 Mar 2010 12:02:33 +0000 X-Google-Sender-Auth: fd8f75b97d3df024 Message-ID: <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 10 March 2010 09:55, Simon Josefsson wrote: > > Btw, maybe we can add an libssh2_session_error_last with size_t in the > next version, and deprecate libssh2_session_last_error? =A0Then at least > one interface will be fixed... Why bother? Fixing one call isn't going to solve the overall problems so why make extra work for clients when we could just wait till the soname bump? Also session_last_error is the perfect name for this function; session_error_last sounds awful. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 13:26:51 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ACQbgD019743; Wed, 10 Mar 2010 13:26:47 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ACQZbJ019705 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 13:26:35 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2ACQXcC001570 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 13:26:35 +0100 X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::iIUKsNGCNpvfPtYS:7NvJ From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: more keep-alive stuff OpenPGP: id=B565716F; url=http://josefsson.org/key.txt Date: Wed, 10 Mar 2010 13:26:34 +0100 Message-ID: <87d3zcicqt.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Folks, I have pushed support for sending keepalive requests. (Recall, my earlier work on keepalive support was to _receive_ keepalive requests.) This code hasn't been tested a lot, so I'd really appreciate if people with applications using libssh2 can test it. Sending keepalive's is still off by default. I've made sure that libssh2 is essentially not doing anything else when keepalive's are off, so this should cause a minimum of intrusion for already working code. The two new APIs are: /* * libssh2_keepalive_config() * * Set how often keepalive messages should be sent. WANT_REPLY * indicates whether the keepalive messages should request a response * from the server. INTERVAL is number of seconds that can pass * without any I/O, use 0 (the default) to disable keepalives. To * avoid some busy-loop corner-cases, if you specify an interval of 1 * it will be treated as 2. * * Note that non-blocking applications are responsible for sending the * keepalive messages using libssh2_keepalive_send(). */ LIBSSH2_API void libssh2_keepalive_config (LIBSSH2_SESSION *session, int want_reply, unsigned interval); /* * libssh2_keepalive_send() * * Send a keepalive message if needed. SECONDS_TO_NEXT indicates how * many seconds you can sleep after this call before you need to call * it again. Returns 0 on success, or LIBSSH2_ERROR_SOCKET_SEND on * I/O errors. */ LIBSSH2_API int libssh2_keepalive_send (LIBSSH2_SESSION *session, int *seconds_to_next); If your application is blocking, to enable sending of keepalive requests, you just add a call like this: libssh2_keepalive_config (session, 1, 5); This will send keepalive messages every 5 seconds, and the message that will be sent requests the server to respond to the message (because want_reply=1). That's it! If your application is non-blocking, you need to first call libssh2_keepalive_config as above, but you also need to make sure you call libssh2_keepalive_send in your select loop and use its output as the maximum time to sleep in your select call. For example: while (1) { struct timeval tmout; int err = libssh2_keepalive_send(sess, &tmout.tv_sec); if (err) { // ... } tmout.tv_usec = 0; // set up fd_sets ev = select(..., &tmout); // handle events } That's it. I hope it works. Let me know if you find anything weird... I'm aware of one issue with the code, but I'm hoping it is just theoretical. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 13:33:38 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ACXWfm023405; Wed, 10 Mar 2010 13:33:35 +0100 Received: from av7-2-sn3.vrr.skanova.net (av7-2-sn3.vrr.skanova.net [81.228.9.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ACXUbi023402 for ; Wed, 10 Mar 2010 13:33:30 +0100 Received: by av7-2-sn3.vrr.skanova.net (Postfix, from userid 502) id 92BC638417; Wed, 10 Mar 2010 13:29:20 +0100 (CET) Received: from smtp3-2-sn3.vrr.skanova.net (smtp3-2-sn3.vrr.skanova.net [81.228.9.102]) by av7-2-sn3.vrr.skanova.net (Postfix) with ESMTP id 5F4F038411 for ; Wed, 10 Mar 2010 13:29:20 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp3-2-sn3.vrr.skanova.net (Postfix) with ESMTP id 0B67337E82 for ; Wed, 10 Mar 2010 13:32:55 +0100 (CET) Message-ID: <4B979175.6050403@SDlabs.se> Date: Wed, 10 Mar 2010 13:32:53 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2 development Subject: Re: patch to add libssh2_init() and libssh2_exit() References: <4B9781DE.2060707@SDlabs.se> <87y6i0ierk.fsf@mocca.josefsson.org> In-Reply-To: <87y6i0ierk.fsf@mocca.josefsson.org> Content-Type: multipart/mixed; boundary="------------030406030407030303000700" X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --------------030406030407030303000700 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit On 2010-03-10 12:42, Simon Josefsson wrote: > Lars Nordin writes: > > >> Please find attached patch to make libssh2 have a clean way of >> initialize the library. >> > Hi. Great, I agree libssh2 should have proper global init/exit > functions. > > >> The patch adds a new file, global.c. The makefile.in is changes, but >> not the WIN32 files. >> > The Makefile.in file is automatically generated, the file to modify is > Makefile.inc. > > Can you modify or should a resend the patch? > You forgot to attach the new globals.c? > Attached... > >> + * LIBSSH2_INIT_ALL: Initialize everything >> > Maybe we could remove this one? I think it will restrict us too much in > the future, we could conceivable have init-flags that are not related to > disable some initialization, and then this flag becomes confusing. Just > document that callers should pass 0 if they don't want to use any flag. > > libssh2_init() needs a parameter and I think it should be a name rather then just 0. Maybe change the name or just accept that libssh2_init(0); is what normally will be done (if you don't handle the init of te used cryptolibrary yourself) >> + ERR_load_crypto_strings(); >> + OpenSSL_add_all_algorithms(); >> + ENGINE_load_builtin_engines(); >> > Are the first and last calls really needed? > > The first call is needed to get errormessages from OpenSSL with names instead of numbers for error-funtions, but can be removed until vi display OpenSSL error-messages for DEBUG-reasons. The last one should be removed, usage of OpenSSL-engines can now be handled outside libssh2 (and then call libssh2_init with LIBSSH2_INIT_NO_CRYPTO_INIT) The same for entropy-handling, can be handled before libssh2_init is called (ie load/update of .rnd as openssh does it to improve random-generation > /Simon > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > /Lars --------------030406030407030303000700 Content-Type: text/plain; name="global.c" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="global.c" /* Copyright (c) 2004-2007, Sara Golemon * Copyright (c) 2009 by Daniel Stenberg * All rights reserved. * * Redistribution and use in source and binary forms, * with or without modification, are permitted provided * that the following conditions are met: * * Redistributions of source code must retain the above * copyright notice, this list of conditions and the * following disclaimer. * * Redistributions in binary form must reproduce the above * copyright notice, this list of conditions and the following * disclaimer in the documentation and/or other materials * provided with the distribution. * * Neither the name of the copyright holder nor the names * of any other contributors may be used to endorse or * promote products derived from this software without * specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY * OF SUCH DAMAGE. */ #include "libssh2_priv.h" int libssh2_initialized = 0; int libssh2_init_flags = 0; LIBSSH2_API int libssh2_init(int flags) { if (!(flags & LIBSSH2_INIT_NO_CRYPTO_INIT)) { libssh2_crypto_init(); } libssh2_initialized++; libssh2_init_flags &= flags; return 0; } LIBSSH2_API void libssh2_exit() { if (libssh2_initialized == 0) return; if (!(libssh2_init_flags & LIBSSH2_INIT_NO_CRYPTO_INIT)) { libssh2_crypto_exit(); } libssh2_initialized--; return; } --------------030406030407030303000700 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --------------030406030407030303000700-- From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 13:39:48 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ACdhBs025928; Wed, 10 Mar 2010 13:39:46 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ACdfaN025923 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 13:39:41 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2ACdc8i001812 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 13:39:41 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090431u20ac3d0ap12365247f0da7c19@mail.gmail.com> <87y6i1bpjm.fsf@vertex.dottedmag> <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::r9xJx7FmqNFQBbYR:7j8w Date: Wed, 10 Mar 2010 13:39:39 +0100 In-Reply-To: <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> (Alexander Lamaison's message of "Wed, 10 Mar 2010 12:02:33 +0000") Message-ID: <878wa0ic50.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Alexander Lamaison writes: > On 10 March 2010 09:55, Simon Josefsson wrote: >> >> Btw, maybe we can add an libssh2_session_error_last with size_t in the >> next version, and deprecate libssh2_session_last_error? =A0Then at least >> one interface will be fixed... > > Why bother? > > Fixing one call isn't going to solve the overall problems so why make > extra work for clients when we could just wait till the soname bump? > Also session_last_error is the perfect name for this function; > session_error_last sounds awful. I was just thinking if we can start this process without a soname bump, by replacing each and every API call with a new one that has size_t in it. Then we could officially deprecate the old interfaces, and after some time make the old APIs call syslog and warn about obsolete interface, and after some more time, do the soname bump and remove the old interfaces. But maybe biting the dust and fixing this problem for real, and do a soname bump, is better. We could also rename the header file, to make it possible to have both API/ABI versions installed at the same time. I don't feel strongly about what to do, but I feel something should be done so that we aren't stuck with a sub-optimal design. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 18:03:42 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AH3QIN028416; Wed, 10 Mar 2010 18:03:36 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2AH3Pwj028362 for ; Wed, 10 Mar 2010 18:03:25 +0100 Received: (qmail 17031 invoked by uid 501); 10 Mar 2010 17:03:25 -0000 Message-ID: <20100310170325.17030.qmail@stuge.se> Date: Wed, 10 Mar 2010 18:03:25 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) Mail-Followup-To: libssh2-devel@cool.haxx.se References: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <9c3275a91003090616y2ca466b5y2ab1055ff631285f@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> <878wa0ic50.fsf@mocca.josefsson.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <878wa0ic50.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson wrote: > But maybe biting the dust and fixing this problem for real, and do > a soname bump, is better. I think so. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 18:40:47 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AHdgH7023988; Wed, 10 Mar 2010 18:39:48 +0100 Received: from mail-ew0-f226.google.com (mail-ew0-f226.google.com [209.85.219.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AHdfLF023942 for ; Wed, 10 Mar 2010 18:39:41 +0100 Received: by ewy26 with SMTP id 26so4671970ewy.3 for ; Wed, 10 Mar 2010 09:39:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=wYF45jFN/jWA6Ysihwhz05QTZjuStLs7SqZAFMAJa9g=; b=VikBrDYxjQ7gJHRgo+ScBtJ4rxa8ojjo8CWgxrq+2TI3NMOvk21txhCPHzF4QdKH1P sLTgAROj7L3mev8CimFsqSSRyfoE6uIzqdUfpb1egRHC/QSgBZ61J8n9DUaB23knjZbF 9Xsv5qgJKSu823Q+4EnhsCpdCANyXBTCQBrxE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=SagDEKUeNspMV7YWV3YNOB51m3JWIMxKZGKqfcpD4hRC1SVLkdXVVfeGUlvhjtkxCE Yitk3lPP15N2BJ0qkQkHUOL6nNUfkPLzOCv764hS4mbGsFTPt55YSNJBp5VNJeNtc/A4 plUsVqcd9CP6wlybc91Cvmzg4qA6X9Gk48RqQ= MIME-Version: 1.0 Received: by 10.213.106.199 with SMTP id y7mr1131817ebo.55.1268242777067; Wed, 10 Mar 2010 09:39:37 -0800 (PST) In-Reply-To: <20100310170325.17030.qmail@stuge.se> References: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> <878wa0ic50.fsf@mocca.josefsson.org> <20100310170325.17030.qmail@stuge.se> Date: Wed, 10 Mar 2010 17:39:37 +0000 X-Google-Sender-Auth: 8dc915427fc9b34e Message-ID: <9c3275a91003100939w17cf37f7w2cf4e695bee2cc9b@mail.gmail.com> Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 10 March 2010 17:03, Peter Stuge wrote: > Simon Josefsson wrote: >> But maybe biting the dust and fixing this problem for real, and do >> a soname bump, is better. > > I think so. If we truly are considering a soname bump, would now be the time to really nail down those write problems that Daniel and Peter have talked about for so long. Though I understand very little about the issue, I imagine they might require API changes. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 18:51:52 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AHplvG004362; Wed, 10 Mar 2010 18:51:49 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AHpitu004336 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 10 Mar 2010 18:51:45 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2AHpdHV010331 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 10 Mar 2010 18:51:42 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) References: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> <878wa0ic50.fsf@mocca.josefsson.org> <20100310170325.17030.qmail@stuge.se> <9c3275a91003100939w17cf37f7w2cf4e695bee2cc9b@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100310:libssh2-devel@cool.haxx.se::P28WvDAM2FfaT+I2:GU1W Date: Wed, 10 Mar 2010 18:51:38 +0100 In-Reply-To: <9c3275a91003100939w17cf37f7w2cf4e695bee2cc9b@mail.gmail.com> (Alexander Lamaison's message of "Wed, 10 Mar 2010 17:39:37 +0000") Message-ID: <87r5nskqtx.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Alexander Lamaison writes: > On 10 March 2010 17:03, Peter Stuge wrote: >> Simon Josefsson wrote: >>> But maybe biting the dust and fixing this problem for real, and do >>> a soname bump, is better. >> >> I think so. > > If we truly are considering a soname bump, would now be the time to > really nail down those write problems that Daniel and Peter have > talked about for so long. Though I understand very little about the > issue, I imagine they might require API changes. Let's add all issues to the 'At next SONAME bump' section in the TODO. I added the size_t issue. The bigger question is who will write the patch for this... /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 10 21:13:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AKDSbW006929; Wed, 10 Mar 2010 21:13:36 +0100 Received: from hrndva-omtalb.mail.rr.com (hrndva-omtalb.mail.rr.com [71.74.56.122]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2AKDQVP006914 for ; Wed, 10 Mar 2010 21:13:26 +0100 X-Authority-Analysis: v=1.0 c=1 a=0O5caf2HW_cA:10 a=O-pu0537hIgA:10 a=kj9zAlcOel0A:10 a=EJN1104aAAAA:8 a=UJ5Y5Z__AAAA:8 a=XNiDR7TIWshVPxZ7684A:9 a=94uausaNSq1q47n2W9XKungfVggA:4 a=CjuIK1q_8ugA:10 a=GXkJ9VDv-M4A:10 a=z1r5ifv-IIqt9uKK:21 a=iBz6pejNL8g-sTdI:21 X-Cloudmark-Score: 0 X-Originating-IP: 70.116.71.237 Received: from [70.116.71.237] ([70.116.71.237:32944] helo=dc7100.roadrunn.com) by hrndva-oedge04.mail.rr.com (envelope-from ) (ecelerity 2.2.2.39 r()) with ESMTP id F7/51-20938-06DF79B4; Wed, 10 Mar 2010 20:13:21 +0000 Received: by dc7100.roadrunn.com (Postfix, from userid 501) id 35A04830A2D; Wed, 10 Mar 2010 15:07:15 -0600 (CST) X-Spam-Checker-Version: SpamAssassin 3.2.4 (2008-01-01) on dc7100.roadrunn.com X-Spam-Level: X-Spam-Status: No, score=-4.3 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.4 Received: from ngierman3 (dc7100 [192.168.1.6]) by dc7100.roadrunn.com (Postfix) with ESMTP id 8BE22830A2A for ; Wed, 10 Mar 2010 15:06:57 -0600 (CST) From: "Neil Gierman" To: "'libssh2 development'" References: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <001501cabf98$702b2390$50816ab0$@com> <9c3275a91003090714x3be1afd4kd1bcfecf2a81cc24@mail.gmail.com> <000601cabfa7$3fd640d0$bf82c270$@com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> <878wa0ic50.fsf@mocca.josefsson.org> <20100310170325.17030.qmail@stuge.se> <9c3275a91003100939w17cf37f7w2cf4e695bee2cc9b@mail.gmail.com> <87r5nskqtx.fsf@mocca.josefsson.org> In-Reply-To: <87r5nskqtx.fsf@mocca.josefsson.org> Subject: RE: Visual Studio 6 compilation issues (libssh2 1.2.4) Date: Wed, 10 Mar 2010 14:13:02 -0600 Message-ID: <000301cac08e$16ac2570$44047050$@com> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: AcrAgoOfuvmbmIAaQJyFcfLRZSPn+AAC3bkg Content-Language: en-us X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se I can start writing the patch since I can get the compiler warnings every time on VS2008/Itanium. Might take a while thou, and will be quite a bit of changes. > -----Original Message----- > From: libssh2-devel-bounces@cool.haxx.se [mailto:libssh2-devel- > bounces@cool.haxx.se] On Behalf Of Simon Josefsson > Sent: Wednesday, March 10, 2010 11:52 AM > To: libssh2 development > Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) > > Alexander Lamaison writes: > > > On 10 March 2010 17:03, Peter Stuge wrote: > >> Simon Josefsson wrote: > >>> But maybe biting the dust and fixing this problem for real, and do > >>> a soname bump, is better. > >> > >> I think so. > > > > If we truly are considering a soname bump, would now be the time to > > really nail down those write problems that Daniel and Peter have > > talked about for so long. Though I understand very little about the > > issue, I imagine they might require API changes. > > Let's add all issues to the 'At next SONAME bump' section in the TODO. > I added the size_t issue. > > The bigger question is who will write the patch for this... > > /Simon > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2- > devel _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 11 19:52:42 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BIqOvD007554; Thu, 11 Mar 2010 19:52:35 +0100 Received: from mail-fx0-f223.google.com (mail-fx0-f223.google.com [209.85.220.223]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BIqMin007465 for ; Thu, 11 Mar 2010 19:52:22 +0100 Received: by fxm23 with SMTP id 23so390640fxm.3 for ; Thu, 11 Mar 2010 10:51:18 -0800 (PST) MIME-Version: 1.0 Received: by 10.239.188.210 with SMTP id q18mr359368hbh.143.1268333477210; Thu, 11 Mar 2010 10:51:17 -0800 (PST) Date: Thu, 11 Mar 2010 19:51:17 +0100 Message-ID: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> Subject: so name bump; impact on major, minor and patch version From: Jose Baars To: libssh2-devel@cool.haxx.se X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0637491910==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0637491910== Content-Type: multipart/alternative; boundary=001485f44f9e5da9b104818ae6d9 --001485f44f9e5da9b104818ae6d9 Content-Type: text/plain; charset=ISO-8859-1 Hi, I noticed the discussion on size_t, and saw an so name bump was considered, to implement changes to would break the ABI. Excuse my ignorance, what is the impact on the major version, minor version and patch version when an so name bump is done? On VMS usually the major version is increased, to make it impossible for calling programs developed against a lower version to even open the shared image. Jose. --001485f44f9e5da9b104818ae6d9 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi,

I noticed the discussion on size_t, and saw an so name bump was = considered,
to implement changes to would break the ABI.
=A0
Excus= e my ignorance, what is the impact on the major version, minor version
and patch version when an so name bump is done?

On VMS usually the m= ajor version is increased, to make it impossible for
calling programs d= eveloped against a lower version to even open the shared
image.
=A0
Jose.
--001485f44f9e5da9b104818ae6d9-- --===============0637491910== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0637491910==-- From libssh2-devel-bounces@cool.haxx.se Thu Mar 11 20:51:59 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BJphqF026035; Thu, 11 Mar 2010 20:51:47 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BJpgBB026030 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 11 Mar 2010 20:51:42 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2BJoeos014114 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Thu, 11 Mar 2010 20:50:43 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: so name bump; impact on major, minor and patch version References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100311:libssh2-devel@cool.haxx.se::aYf0irL8d/nq1iGT:PMlj Date: Thu, 11 Mar 2010 20:50:40 +0100 In-Reply-To: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> (Jose Baars's message of "Thu, 11 Mar 2010 19:51:17 +0100") Message-ID: <87vdd2hc33.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Jose Baars writes: > Hi, > > I noticed the discussion on size_t, and saw an so name bump was considered, > to implement changes to would break the ABI. > > Excuse my ignorance, what is the impact on the major version, minor version > and patch version when an so name bump is done? No correlation can be assumed, but I suspect we'll go to libssh2 1.4.x or even 2.x.y from 1.2.x to reflect that we've done major changes. Daniel? > On VMS usually the major version is increased, to make it impossible for > calling programs developed against a lower version to even open the shared > image. You should not use the libssh2 release version number for the library version on VMS -- please compute the shared library version from the '-version-info 1:1:0' information in src/Makefile.am. The release version number of libssh2 has nothing to do with shared library versioning, strictly speaking. See the comments in src/Makefile.am for more background, or read the GNU Libtool manual for the entire story. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 11 21:42:09 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BKfxSO002632; Thu, 11 Mar 2010 21:42:07 +0100 Received: from web22.pochta.ru (web22.pochta.ru [82.204.219.122]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BGZ8X2000388 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 11 Mar 2010 17:35:09 +0100 Received: from [127.0.0.1] (port=33915 helo=localhost) by web22.pochta.ru ( sendmail 8.13.3/8.13.1) with esmtp id 1NplKd-0002rs-63 for libssh2-devel@cool.haxx.se; Thu, 11 Mar 2010 19:34:03 +0300 From: andpar@pisem.net To: libssh2-devel@cool.haxx.se Subject: Working with sudo Date: Thu, 11 Mar 2010 19:33:55 +0300 Message-Id: <3d9dcfba01e85f6ec153220714e854293efd30c1@mail.qip.ru> X-Priority: 3 Content-Disposition: inline X-SpamTest-Envelope-From: andpar@pisem.net X-SpamTest-Group-ID: 00000000 X-SpamTest-Info: Profiles 12611 [Mar 11 2010] X-SpamTest-Method: none X-SpamTest-Rate: 0 X-SpamTest-Status: Not detected X-SpamTest-Status-Extended: not_detected X-SpamTest-Version: SMTP-Filter Version 3.0.0 [0278], KAS30/Release X-Mailman-Approved-At: Thu, 11 Mar 2010 21:41:58 +0100 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se I want to exec my command like "sudo pwd" though libssh2_channel_exec. Sometimes I must enter password in response to this command. How can I undertand that I need to send password and how can I do it? Best regards,Andrey _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 11 21:53:20 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BKrEsf012324; Thu, 11 Mar 2010 21:53:17 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BKrCc2012314 for ; Thu, 11 Mar 2010 21:53:12 +0100 Date: Thu, 11 Mar 2010 21:53:12 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: so name bump; impact on major, minor and patch version In-Reply-To: <87vdd2hc33.fsf@mocca.josefsson.org> Message-ID: References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Thu, 11 Mar 2010, Simon Josefsson wrote: > No correlation can be assumed, but I suspect we'll go to libssh2 1.4.x or > even 2.x.y from 1.2.x to reflect that we've done major changes. Daniel? In my view it depends on what exact changes we do and not only if we bump the soname. If we truly make all the changes mentioned in the TODO, then I think we should go for a full 2.0.0 release but if we "only" clean up things such as the size_t problems etc then I think 1.3.0 is good enough. Ideally, I would like us to gently with the bump. Possibly start off by adding #ifdefs or similar so that we can build the existing or the new ABI, so that we can allow apps to try it out easier and get a feel for what's new and what changes their apps require. We can then introduce the new ABI but not built by default in a release or two before we take the jump and cut out the old and run with only the new. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 11 22:55:31 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BLtHVB021129; Thu, 11 Mar 2010 22:55:27 +0100 Received: from mail.intersystems.com (mail.intersys.COM [198.133.74.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2BLtF4X021086 for ; Thu, 11 Mar 2010 22:55:15 +0100 X-InterSystems: Sent from InterSystems X-InterSystems: Sent from InterSystems From: Dave McCaldon Subject: dispatch_protocol_error: type 90 seq 4 Date: Thu, 11 Mar 2010 16:54:04 -0500 Message-Id: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> To: libssh2 development Mime-Version: 1.0 (Apple Message framework v1077) X-Mailer: Apple Mail (2.1077) X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0067678458==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0067678458== Content-Type: multipart/alternative; boundary=Apple-Mail-41-126952331 --Apple-Mail-41-126952331 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii I have a Windows client app using libssh2-1.2.3 that seems to be unable = to connect to a remote server (OpenSSH 5.2) to do SFTP, it fails on the = server side with: dispatch_protocol_error: type 90 seq 4 The client side reports: [libssh2] 44.093943 Transport: Disconnect(7): Service not requested or = authenticated(en) [libssh2] 44.093943 Failure Event: -13 - socket disconnect Does anyone have any more details on what this means? (configuration = perhaps?) --- Here's the sshd log: debug3: fd 5 is not O_NONBLOCK debug1: Server will not fork when running in debugging mode. debug3: send_rexec_state: entering fd =3D 8 config len 414 debug3: ssh_msg_send: type 0 debug3: send_rexec_state: done debug1: rexec start in 5 out 5 newsock 5 pipe -1 sock 8 debug1: inetd sockets after dupping: 3, 3 Connection from 10.1.72.50 port 53813 debug1: Client protocol version 2.0; client software version = libssh2_1.2.3 debug1: no match: libssh2_1.2.3 debug1: Enabling compatibility mode for protocol 2.0 debug1: Local version string SSH-2.0-OpenSSH_5.2 debug2: fd 3 setting O_NONBLOCK debug3: privsep user:group 71:65 debug1: permanently_set_uid: 71/65 debug1: list_hostkey_types: ssh-rsa,ssh-dss debug1: SSH2_MSG_KEXINIT sent debug2: Network child is on pid 29175 debug3: preauth child monitor started debug3: mm_request_receive entering debug1: SSH2_MSG_KEXINIT received debug2: kex_parse_kexinit: = diffie-hellman-group-exchange-sha256,diffie-hellman-group-exchange-sha1,di= ffie-hellman-group14-sha1, diffie-hellman-group1-sha1 debug2: kex_parse_kexinit: ssh-rsa,ssh-dss debug2: kex_parse_kexinit: = aes128-ctr,aes192-ctr,aes256-ctr,arcfour256,arcfour128,aes128-cbc,3des-cbc= ,blowfish-cbc,cast128-cbc, aes192-cbc,aes256-cbc,arcfour,rijndael-cbc@lysator.liu.se debug2: kex_parse_kexinit: = aes128-ctr,aes192-ctr,aes256-ctr,arcfour256,arcfour128,aes128-cbc,3des-cbc= ,blowfish-cbc,cast128-cbc, aes192-cbc,aes256-cbc,arcfour,rijndael-cbc@lysator.liu.se debug2: kex_parse_kexinit: = hmac-md5,hmac-sha1,umac-64@openssh.com,hmac-ripemd160,hmac-ripemd160@opens= sh.com,hmac-sha1-96,hmac-m d5-96 debug2: kex_parse_kexinit: = hmac-md5,hmac-sha1,umac-64@openssh.com,hmac-ripemd160,hmac-ripemd160@opens= sh.com,hmac-sha1-96,hmac-m d5-96 debug2: kex_parse_kexinit: none,zlib@openssh.com debug2: kex_parse_kexinit: none,zlib@openssh.com debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: kex_parse_kexinit: = diffie-hellman-group14-sha1,diffie-hellman-group-exchange-sha1,diffie-hell= man-group1-sha1 debug2: kex_parse_kexinit: ssh-rsa,ssh-dss debug2: kex_parse_kexinit: = aes128-ctr,aes192-ctr,aes256-ctr,aes256-cbc,rijndael-cbc@lysator.liu.se,ae= s192-cbc,aes128-cbc,blowfi sh-cbc,arcfour128,arcfour,cast128-cbc,3des-cbc debug2: kex_parse_kexinit: = aes128-ctr,aes192-ctr,aes256-ctr,aes256-cbc,rijndael-cbc@lysator.liu.se,ae= s192-cbc,aes128-cbc,blowfi sh-cbc,arcfour128,arcfour,cast128-cbc,3des-cbc debug2: kex_parse_kexinit: = hmac-sha1,hmac-sha1-96,hmac-md5,hmac-md5-96,hmac-ripemd160,hmac-ripemd160@= openssh.com debug2: kex_parse_kexinit: = hmac-sha1,hmac-sha1-96,hmac-md5,hmac-md5-96,hmac-ripemd160,hmac-ripemd160@= openssh.com debug2: kex_parse_kexinit: none debug2: kex_parse_kexinit: none debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: debug2: kex_parse_kexinit: first_kex_follows 0 debug2: kex_parse_kexinit: reserved 0 debug2: mac_setup: found hmac-sha1 debug1: kex: client->server aes128-ctr hmac-sha1 none debug2: mac_setup: found hmac-sha1 debug1: kex: server->client aes128-ctr hmac-sha1 none debug2: dh_gen_key: priv key bits set: 157/320 debug2: bits set: 1040/2048 debug1: expecting SSH2_MSG_KEXDH_INIT debug2: bits set: 1016/2048 debug3: mm_key_sign entering debug3: mm_request_send entering: type 4 debug3: mm_key_sign: waiting for MONITOR_ANS_SIGN debug3: mm_request_receive_expect entering: type 5 debug3: mm_request_receive entering debug3: monitor_read: checking request 4 debug3: mm_answer_sign debug3: mm_answer_sign: signature 0x7f5746305010(271) debug3: mm_request_send entering: type 5 debug2: monitor_read: 4 used once, disabling now debug3: mm_request_receive entering debug2: kex_derive_keys debug2: set_newkeys: mode 1 debug1: SSH2_MSG_NEWKEYS sent debug1: expecting SSH2_MSG_NEWKEYS debug2: set_newkeys: mode 0 debug1: SSH2_MSG_NEWKEYS received debug1: KEX done dispatch_protocol_error: type 90 seq 4 The client trace looks like this: [libssh2] 44.057943 SFTP: Initializing SFTP subsystem [libssh2] 44.057943 Conn: Allocated new channel ID#0 [libssh2] 44.057943 Conn: Opening Channel - win 65536 pack 32768 =3D> libssh2_transport_write plain (24 bytes) 0000: 5A 00 00 00 07 73 65 73 73 69 6F 6E 00 00 00 00 : = Z....session.... 0010: 00 01 00 00 00 00 80 00 : ........ =3D> libssh2_transport_write send() (68 bytes) 0000: C7 83 17 24 26 30 30 D2 B6 95 56 2E 4F 82 46 BC : = ...$&00...V.O.F. 0010: 96 46 66 5E 9A 5A 17 38 98 6C A1 B3 7A 05 AF 91 : = .Ff^.Z.8.l..z... 0020: 33 57 50 EC 0F 73 51 CE B5 EA 31 51 C2 7F C6 2D : = 3WP..sQ...1Q...- 0030: DB 59 68 2F D4 6C 21 E8 AF CC 58 1A 5A E3 39 71 : = .Yh/.l!...X.Z.9q 0040: FD EB AC 50 : ...P [libssh2] 44.058943 Transport: Looking for packet of type: 91 [libssh2] 44.058943 Transport: Looking for packet of type: 92 [libssh2] 44.058943 Failure Event: -37 - Would block [libssh2] 44.058943 Failure Event: -37 - Would block starting up channel [libssh2] 44.093943 Transport: Looking for packet of type: 91 [libssh2] 44.093943 Transport: Looking for packet of type: 92 =3D> libssh2_transport_read() raw (116 bytes) 0000: 08 0A B0 5C E3 B6 7C DD 97 2E E3 6F C2 8C A9 DC : = ...\..|....o.... 0010: 1F A7 3C 24 FA 52 03 5A 69 CC 74 F5 2E 4C AA 32 : = ..<$.R.Zi.t..L.2 0020: 13 4D 70 9E AF C8 D8 5C 33 8E 74 E4 FC CD 6D E5 : = .Mp....\3.t...m. 0030: 5D A2 C5 8B 51 EF 26 E1 CE BE BB EE 0C 7C 14 D4 : = ]...Q.&......|.. 0040: 80 85 BE 5E 1C DB 69 ED D6 6A 8A A1 97 8D DA EC : = ...^..i..j...... 0050: 8E 11 F0 A0 E2 A3 68 FF F1 60 8C 5B 7C 30 6E A7 : = ......h..`.[|0n. 0060: DE DE 68 55 31 91 47 CF 8A 38 44 30 4B 8A FB 76 : = ..hU1.G..8D0K..v 0070: 09 92 2B 01 : ..+. =3D> libssh2_transport_read() plain (53 bytes) 0000: 01 00 00 00 07 00 00 00 26 53 65 72 76 69 63 65 : = ........&Service 0010: 20 6E 6F 74 20 72 65 71 75 65 73 74 65 64 20 6F : not requested = o 0020: 72 20 61 75 74 68 65 6E 74 69 63 61 74 65 64 00 : r = authenticated. 0030: 00 00 02 65 6E : ...en [libssh2] 44.093943 Transport: Packet type 1 received, length=3D53 [libssh2] 44.093943 Transport: Disconnect(7): Service not requested or = authenticated(en) [libssh2] 44.093943 Failure Event: -13 - socket disconnect [libssh2] 44.093943 Transport: Looking for packet of type: 94 [libssh2] 44.093943 Transport: Looking for packet of type: 95 [libssh2] 44.093943 Failure Event: -21 - Unable to startup channel [libssh2] 44.572943 Transport: Disconnecting: reason=3D11, desc=3DNormal = Shutdown, Thank you for playing, lang=3D =3D> libssh2_transport_write plain (51 bytes) 0000: 01 00 00 00 0B 00 00 00 26 4E 6F 72 6D 61 6C 20 : ........&Normal=20= 0010: 53 68 75 74 64 6F 77 6E 2C 20 54 68 61 6E 6B 20 : Shutdown, Thank=20= 0020: 79 6F 75 20 66 6F 72 20 70 6C 61 79 69 6E 67 00 : you for = playing. 0030: 00 00 00 : ... =3D> libssh2_transport_write send() (100 bytes) 0000: C1 FD 85 D0 FC 6C 6C B0 2A 1C 0C C7 0C 81 4A 19 : = .....ll.*.....J. 0010: 88 EA 7F 41 02 A6 EF FA B8 38 34 07 B9 11 D9 83 : = ...A.....84..... 0020: 73 C7 82 5F B7 03 BD 9C F9 CF 8E C4 5E 78 1C FB : = s.._........^x.. 0030: 9C 72 DF 23 E9 F2 0F 32 47 C9 55 F2 A6 66 87 E1 : = .r.#...2G.U..f.. 0040: FF 94 07 04 3A CE FF F9 40 3D F5 98 96 C1 EB 58 : = ....:...@=3D.....X 0050: 34 65 4B 12 38 34 F8 A8 CC 39 22 BA 9D 30 57 BC : = 4eK.84...9"..0W. 0060: 8B 38 77 02 : .8w. [libssh2] 44.572943 Transport: Freeing session resource --Apple-Mail-41-126952331 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=us-ascii dispatch_protocol_error: type 90 seq = 4

The client side = reports:

[libssh2] 44.093943 Transport: Disconnect(7): = Service not requested or authenticated(en)
[libssh2] = 44.093943 Failure Event: -13 - socket = disconnect

Does anyone have any = more details on what this means?  (configuration = perhaps?)


---
Here= 's the sshd log:

debug3: fd 5 is = not O_NONBLOCK
debug1: Server will not fork when running in = debugging mode.
debug3: send_rexec_state: entering fd =3D 8 = config len 414
debug3: ssh_msg_send: type = 0
debug3: send_rexec_state: = done
debug1: rexec start in 5 out 5 newsock 5 pipe = -1 sock 8
debug1: inetd sockets after dupping: 3, = 3
Connection from 10.1.72.50 port = 53813
debug1: Client protocol version 2.0; client = software version libssh2_1.2.3
debug1: no match: = libssh2_1.2.3
debug1: Enabling compatibility mode for = protocol 2.0
debug1: Local version string = SSH-2.0-OpenSSH_5.2
debug2: fd 3 = setting O_NONBLOCK
debug3: privsep = user:group 71:65
debug1: permanently_set_uid: = 71/65
debug1: list_hostkey_types: = ssh-rsa,ssh-dss
debug1: SSH2_MSG_KEXINIT = sent
debug2: Network child is on pid = 29175
debug3: preauth child monitor = started
debug3: mm_request_receive = entering
debug1: SSH2_MSG_KEXINIT = received
debug2: = kex_parse_kexinit:
diffie-hellman-group-exchange-sha256,diffie-hellman-group-exchange-= sha1,diffie-hellman-group14-sha1,
diffie-hellman-group1-sha1
debug2: = kex_parse_kexinit: ssh-rsa,ssh-dss
debug2: = kex_parse_kexinit:
aes128-ctr,aes192-ctr,aes256-ctr,arcfour256,arcfour128,aes128-cbc,3= des-cbc,blowfish-cbc,cast128-cbc,
aes192-cbc,aes256-cbc,arcfour,rijndael-cbc@lysator.liu.se
d5-96
debug2: = kex_parse_kexinit:
hmac-md5,hmac-sha1,umac-64@openssh.com,hmac-ripemd160= ,hmac-ripemd160@openssh.com,= hmac-sha1-96,hmac-m
d5-96
debug2: kex_parse_kexinit: none,zlib@openssh.com
<= div>debug2: = kex_parse_kexinit: none,zlib@openssh.com
<= div>debug2: = kex_parse_kexinit:
debug2: = kex_parse_kexinit:
debug2: = kex_parse_kexinit: first_kex_follows 0
debug2: = kex_parse_kexinit: reserved 0
debug2: = kex_parse_kexinit:
diffie-hellman-group14-sha1,diffie-hellman-group-exchange-sha1,diff= ie-hellman-group1-sha1
debug2: = kex_parse_kexinit: ssh-rsa,ssh-dss
debug2: = kex_parse_kexinit:
aes128-ctr,aes192-ctr,aes256-ctr,aes256-cbc,rijndael-cbc@lysator.liu.se,aes192-cbc,aes128-cbc,blowfi
<= div>debug2: = kex_parse_kexinit:<= div>debug2: = kex_parse_kexinit:
debug2: kex_parse_kexinit:
hmac-sha1,hmac-sha1-96,hmac-md5,hmac-md5-96,hmac-ripemd160,hmac-ripemd160@openssh.com<= /span>
debug2: kex_parse_kexinit: none
debug2: = kex_parse_kexinit: none
debug2: = kex_parse_kexinit:
debug2: = kex_parse_kexinit:
debug2: = kex_parse_kexinit: first_kex_follows 0
debug2: = kex_parse_kexinit: reserved 0
debug2: mac_setup: = found hmac-sha1
debug1: kex: client->server aes128-ctr = hmac-sha1 none
debug2: mac_setup: found = hmac-sha1
debug1: kex: server->client aes128-ctr = hmac-sha1 none
debug2: dh_gen_key: priv key bits set: = 157/320
debug2: bits set: = 1040/2048
debug1: expecting = SSH2_MSG_KEXDH_INIT
debug2: bits set: = 1016/2048
debug3: mm_key_sign = entering
debug3: mm_request_send entering: type = 4
debug3: mm_key_sign: waiting for = MONITOR_ANS_SIGN
debug3: mm_request_receive_expect entering: = type 5
debug3: mm_request_receive = entering
debug3: monitor_read: checking request = 4
debug3: = mm_answer_sign
debug3: mm_answer_sign: signature = 0x7f5746305010(271)
debug3: = mm_request_send entering: type 5
debug2: = monitor_read: 4 used once, disabling now
debug3: = mm_request_receive entering
debug2: = kex_derive_keys
debug2: set_newkeys: mode = 1
debug1: SSH2_MSG_NEWKEYS = sent
debug1: expecting = SSH2_MSG_NEWKEYS
debug2: set_newkeys: mode = 0
debug1: SSH2_MSG_NEWKEYS = received
debug1: KEX = done
dispatch_protocol_error: type 90 seq = 4


The client = trace looks like = this:


[libssh2] = 44.057943 SFTP: Initializing SFTP = subsystem
[libssh2] 44.057943 Conn: Allocated new = channel ID#0
[libssh2] 44.057943 Conn: Opening Channel - = win 65536 pack 32768
=3D> = libssh2_transport_write plain (24 bytes)
0000: 5A 00 00 00 = 07 73 65 73  73 69 6F 6E 00 00 00 00 : = Z....session....
0010: 00 01 00 00 00 00 80 00     =                     =  : ........
=3D> libssh2_transport_write send() (68 = bytes)
0000: C7 83 17 24 26 30 30 D2  B6 95 56 = 2E 4F 82 46 BC : ...$&00...V.O.F.
0010: 96 46 66 5E = 9A 5A 17 38  98 6C A1 B3 7A 05 AF 91 : = .Ff^.Z.8.l..z...
0020: 33 57 50 EC 0F 73 51 CE  B5 EA 31 = 51 C2 7F C6 2D : 3WP..sQ...1Q...-
0030: DB 59 68 2F = D4 6C 21 E8  AF CC 58 1A 5A E3 39 71 : = .Yh/.l!...X.Z.9q
0040: FD EB AC 50         =                     =          : ...P
[libssh2] = 44.058943 Transport: Looking for packet of type: = 91
[libssh2] 44.058943 Transport: Looking for = packet of type: 92
[libssh2] = 44.058943 Failure Event: -37 - Would block
[libssh2] = 44.058943 Failure Event: -37 - Would block starting up = channel
[libssh2] 44.093943 Transport: Looking for = packet of type: 91
[libssh2] = 44.093943 Transport: Looking for packet of type: = 92
=3D> libssh2_transport_read() raw (116 = bytes)
0000: 08 0A B0 5C E3 B6 7C DD  97 2E E3 = 6F C2 8C A9 DC : ...\..|....o....
0010: 1F A7 3C 24 = FA 52 03 5A  69 CC 74 F5 2E 4C AA 32 : = ..<$.R.Zi.t..L.2
0020: 13 4D 70 9E = AF C8 D8 5C  33 8E 74 E4 FC CD 6D E5 : = .Mp....\3.t...m.
0030: 5D A2 C5 8B 51 EF 26 E1  CE BE BB = EE 0C 7C 14 D4 : ]...Q.&......|..
0040: 80 85 BE 5E = 1C DB 69 ED  D6 6A 8A A1 97 8D DA EC : = ...^..i..j......
0050: 8E 11 F0 A0 E2 A3 68 FF  F1 60 8C = 5B 7C 30 6E A7 : ......h..`.[|0n.
0060: DE DE 68 55 = 31 91 47 CF  8A 38 44 30 4B 8A FB 76 : = ..hU1.G..8D0K..v
0070: 09 92 2B 01         =                     =          : ..+.
=3D> = libssh2_transport_read() plain (53 bytes)
0000: 01 00 00 00 = 07 00 00 00  26 53 65 72 76 69 63 65 : = ........&Service
0010: 20 6E 6F 74 = 20 72 65 71  75 65 73 74 65 64 20 6F :  not requested = o
0020: 72 20 61 75 74 68 65 6E  74 69 63 = 61 74 65 64 00 : r authenticated.
0030: 00 00 02 65 = 6E                     =               : = ...en
[libssh2] 44.093943 Transport: Packet type 1 = received, length=3D53
[libssh2] = 44.093943 Transport: Disconnect(7): Service not requested or = authenticated(en)
[libssh2] 44.093943 Failure Event: -13 - = socket disconnect
[libssh2] 44.093943 Transport: Looking for = packet of type: 94
[libssh2] = 44.093943 Transport: Looking for packet of type: = 95
[libssh2] 44.093943 Failure Event: -21 - = Unable to startup channel
[libssh2] = 44.572943 Transport: Disconnecting: reason=3D11, desc=3DNormal Shutdown, = Thank you for playing, lang=3D
=3D> = libssh2_transport_write plain (51 bytes)
0000: 01 00 00 00 = 0B 00 00 00  26 4E 6F 72 6D 61 6C 20 : = ........&Normal 
0010: 53 68 75 74 = 64 6F 77 6E  2C 20 54 68 61 6E 6B 20 : Shutdown, = Thank 
0020: 79 6F 75 20 66 6F 72 20  70 6C 61 = 79 69 6E 67 00 : you for playing.
0030: 00 00 00 =                     =                     : = ...
=3D> libssh2_transport_write send() (100 = bytes)
0000: C1 FD 85 D0 FC 6C 6C B0  2A 1C 0C = C7 0C 81 4A 19 : .....ll.*.....J.
0010: 88 EA 7F 41 = 02 A6 EF FA  B8 38 34 07 B9 11 D9 83 : = ...A.....84.....
0020: 73 C7 82 5F B7 03 BD 9C  F9 CF 8E = C4 5E 78 1C FB : s.._........^x..
0030: 9C 72 DF 23 = E9 F2 0F 32  47 C9 55 F2 A6 66 87 E1 : = .r.#...2G.U..f..
0040: FF 94 07 04 3A CE FF F9  40 3D F5 = 98 96 C1 EB 58 : ....:...@=3D.....X
0050: 34 65 4B 12 = 38 34 F8 A8  CC 39 22 BA 9D 30 57 BC : = 4eK.84...9"..0W.
0060: 8B 38 77 02         =                     =          : .8w.
[libssh2] = 44.572943 Transport: Freeing session = resource

= --Apple-Mail-41-126952331-- --===============0067678458== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0067678458==-- From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 10:48:28 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2C9mCjw009901; Fri, 12 Mar 2010 10:48:22 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.26]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2C9mAZa009898 for ; Fri, 12 Mar 2010 10:48:10 +0100 Received: by ey-out-2122.google.com with SMTP id 9so104553eyd.31 for ; Fri, 12 Mar 2010 01:48:11 -0800 (PST) Received: by 10.213.59.145 with SMTP id l17mr298474ebh.56.1268387291653; Fri, 12 Mar 2010 01:48:11 -0800 (PST) Received: from [192.168.0.35] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 14sm777014ewy.10.2010.03.12.01.48.10 (version=SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 01:48:10 -0800 (PST) Message-ID: <4B9A0DD6.3020403@peut.org> Date: Fri, 12 Mar 2010 10:48:06 +0100 From: Jose Baars User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; nl; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: dispatch_protocol_error: type 90 seq 4 References: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> In-Reply-To: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0608376838==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a cryptographically signed message in MIME format. --===============0608376838== Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms050301040509070000070203" This is a cryptographically signed message in MIME format. --------------ms050301040509070000070203 Content-Type: multipart/alternative; boundary="------------040606030506070308070509" This is a multi-part message in MIME format. --------------040606030506070308070509 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Op 3/11/2010 10:54 PM, Dave McCaldon schreef: > [libssh2] 44.058943 Failure Event: -37 - Would block > [libssh2] 44.058943 Failure Event: -37 - Would block starting up channe= l Isn't that the real problem? It looks a lot like the errors I got when testing non-blocking without th= e waitforsocket (select() before a read of the receiving socket). --------------040606030506070308070509 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Op 3/11/2010 10:54 PM, Dave McCaldon schreef:
[libssh2] 44.05894= 3 Failure Event: -37 - Would block
[libssh2] 44.05894= 3 Failure Event: -37 - Would block starting up channel
Isn't that the real problem?

It looks a lot like the errors I got when testing non-blocking without the
waitforsocket (select() before a read of the receiving socket).
--------------040606030506070308070509-- --------------ms050301040509070000070203 Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIPWjCC BMwwggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1 MTAyNzIzNTk1OVowgd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEf MB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNl IGF0IGh0dHBzOi8vd3d3LnZlcmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNv bmEgTm90IFZhbGlkYXRlZDE3MDUGA1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFs IFN1YnNjcmliZXIgQ0EgLSBHMjCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnf rOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyVzm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs +Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zplYu//EHuiVrvFTnAt1qIfPO2wQuhejVch rKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFBL2OyOj++pRpu9MlKWz2VphW7NQIZ +dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5gJ925rXXOL3OVekA6hXVJsLjf aLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUCAwEAAaOCAYQwggGAMBIG A1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXATAqMCgGCCsGAQUF BwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIBBjARBglghkgB hvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJlbDMtMjA0 OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAkoCKG IGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEg UHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+Iqyz cqpVMA0GCSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESf D0b3+qD+0x0Yo9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Pr v4NZmP1m3umGMpqSKTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFQTCCBCmgAwIB AgIQQkuBnab/mKbwbYynoi4GkzANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAV BgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3Jr MTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJp U2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyMB4XDTA5MTExMjAw MDAwMFoXDTEwMTExMjIzNTk1OVowggEKMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0G A1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNv bS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElBQi5MVEQoYyk5ODEeMBwGA1UE CxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdpdGFsIElEIENsYXNzIDEg LSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxEzARBgNVBAMUCkpvc2UgQmFhcnMxHDAaBgkqhkiG 9w0BCQEWDXBldXRAcGV1dC5vcmcwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCi x8yI3kFLmfYElzZSqsup782ajeuv3TASyn8mUfGrnr2aW1M7jN4vlM8/7/sVL883rH+kWPzV Q4ITYrfW5QFxH+CBqtX0F1xsS1A7AR8xZIs8o7/sMxV2xOSXYEJXlKP4XVWY8oU+TKH5dWYZ 9BJ9pkJ840ayiqqI/P3MHqmu7vS2nPDu8YZFa+46EtWkGbJTjzFdnrU5Jmdudp37fd1btNss RIkqztDOeQOTaZ6pS9RqjhmNuWVtgZucKho6c9Qf2htGz8AL95MqttF2OZYL9AlXZhjJcIMO aZBVjuwLowX/kYL2lo1vyUGt3eZxSkZR/ICfkQwj1BKpFZGFtM5XAgMBAAGjgcwwgckwCQYD VR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQt Y3JsLnZlcmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEB ABcuOD4QwMrXJgD8zvCWBRjBqsQLXn0ybqrLZHlcic7+lstAl06fi7fGD3+1Nrcp4VImkyUv JQEvFSvkB+/Mt8kOVSdt+eiYIInKt/lBkSzy0sardsh0u7Ifcgt+nmLZYW+azviXnD4VNMqe Q9QIZPHSUhWDe3IA5NSYE4hfqwXX5c13HllqpUcXRmWIGA9MPr9SQxZ0jg7MFUsZVgG9f5Ts A3yIEmvUgutyX6GRzZQL+wlWymRzh/XA7j2QnYk+IPlNQHJ62MHHlAI48y3ssSXwR8hfbH4e 0sw+8ezuh785OuFMBI3qhxlQX4wT8+v8rKHtiLcddoIohWM0ID+IdOEwggVBMIIEKaADAgEC AhBCS4Gdpv+YpvBtjKeiLgaTMA0GCSqGSIb3DQEBBQUAMIHdMQswCQYDVQQGEwJVUzEXMBUG A1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdvcmsx OzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBh IChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZlcmlT aWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzIwHhcNMDkxMTEyMDAw MDAwWhcNMTAxMTEyMjM1OTU5WjCCAQoxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxMzAxBgNVBAsTKkRpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlIEZ1bGwgU2VydmljZTETMBEGA1UEAxQKSm9zZSBCYWFyczEcMBoGCSqGSIb3 DQEJARYNcGV1dEBwZXV0Lm9yZzCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKLH zIjeQUuZ9gSXNlKqy6nvzZqN66/dMBLKfyZR8auevZpbUzuM3i+Uzz/v+xUvzzesf6RY/NVD ghNit9blAXEf4IGq1fQXXGxLUDsBHzFkizyjv+wzFXbE5JdgQleUo/hdVZjyhT5Mofl1Zhn0 En2mQnzjRrKKqoj8/cweqa7u9Lac8O7xhkVr7joS1aQZslOPMV2etTkmZ252nft93Vu02yxE iSrO0M55A5NpnqlL1GqOGY25ZW2Bm5wqGjpz1B/aG0bPwAv3kyq20XY5lgv0CVdmGMlwgw5p kFWO7AujBf+RgvaWjW/JQa3d5nFKRlH8gJ+RDCPUEqkVkYW0zlcCAwEAAaOBzDCByTAJBgNV HRMEAjAAMEQGA1UdIAQ9MDswOQYLYIZIAYb4RQEHFwEwKjAoBggrBgEFBQcCARYcaHR0cHM6 Ly93d3cudmVyaXNpZ24uY29tL3JwYTALBgNVHQ8EBAMCBaAwHQYDVR0lBBYwFAYIKwYBBQUH AwQGCCsGAQUFBwMCMEoGA1UdHwRDMEEwP6A9oDuGOWh0dHA6Ly9JbmRDMURpZ2l0YWxJRC1j cmwudmVyaXNpZ24uY29tL0luZEMxRGlnaXRhbElELmNybDANBgkqhkiG9w0BAQUFAAOCAQEA Fy44PhDAytcmAPzO8JYFGMGqxAtefTJuqstkeVyJzv6Wy0CXTp+Lt8YPf7U2tynhUiaTJS8l AS8VK+QH78y3yQ5VJ2356Jggicq3+UGRLPLSxqt2yHS7sh9yC36eYtlhb5rO+JecPhU0yp5D 1Ahk8dJSFYN7cgDk1JgTiF+rBdflzXceWWqlRxdGZYgYD0w+v1JDFnSODswVSxlWAb1/lOwD fIgSa9SC63JfoZHNlAv7CVbKZHOH9cDuPZCdiT4g+U1AcnrYwceUAjjzLeyxJfBHyF9sfh7S zD7x7O6Hvzk64UwEjeqHGVBfjBPz6/ysoe2Itx12giiFYzQgP4h04TGCBOwwggToAgEBMIHy MIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZl cmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxp ZGF0ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVy IENBIC0gRzICEEJLgZ2m/5im8G2Mp6IuBpMwCQYFKw4DAhoFAKCCAs4wGAYJKoZIhvcNAQkD MQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMTAwMzEyMDk0ODA2WjAjBgkqhkiG9w0B CQQxFgQUNeR6nmY6kG3oEqzzu8oyTN96n64wXwYJKoZIhvcNAQkPMVIwUDALBglghkgBZQME AQIwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIH MA0GCCqGSIb3DQMCAgEoMIIBAwYJKwYBBAGCNxAEMYH1MIHyMIHdMQswCQYDVQQGEwJVUzEX MBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdv cmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20v cnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZl cmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzICEEJLgZ2m/5im 8G2Mp6IuBpMwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTsw OQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAo YykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhBCS4Gdpv+YpvBtjKei LgaTMA0GCSqGSIb3DQEBAQUABIIBAIX//wChgxAhN2e4ono3nHYuuoBCqiLWXHKJyaJCAjqj OAt1S4NMzguPCtcluAQh0bjsHS8P7mHdYIpYJJNBa/0RF9CDUt6dxiUb4A2GiLfUrbW55fw4 qx4ToxS+2XWWIATlvFg57ZCzY8EzOgFw/7Rt19R8IOMa8Q1nN4ibqIEtTZb6seCGdaE9Ar7a p2Yy4LEHfDLazvCh7cXT/PVxqwjawLyU+L3LcdYtGei8EOdL1idnPWiL2NxOtiDeyEdWziQ4 uorMI/95gcwTBlhJHCzTjPV7b/D98ioUjLxv6Rg2dhnATZkSA1jY7mHj1952KZbI5RBzWURZ ZgmU9Gi+D7cAAAAAAAA= --------------ms050301040509070000070203-- --===============0608376838== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0608376838==-- From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 11:12:05 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CAC0iM025957; Fri, 12 Mar 2010 11:12:03 +0100 Received: from mail.gmx.net (mail.gmx.net [213.165.64.20]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2CABwj5025951 for ; Fri, 12 Mar 2010 11:11:58 +0100 Received: (qmail invoked by alias); 12 Mar 2010 10:11:54 -0000 Received: from xdsl-188-155-90-113.adslplus.ch (EHLO [192.168.1.71]) [188.155.90.113] by mail.gmx.net (mp044) with SMTP; 12 Mar 2010 11:11:54 +0100 X-Authenticated: #1568940 X-Provags-ID: V01U2FsdGVkX1/3PwsI0bn3u3zLPyHvI+owo0YdjqRRc9ynYycuX9 YLhzPpsNSvf1VZ Message-ID: <4B9A1366.2050709@gmx.at> Date: Fri, 12 Mar 2010 11:11:50 +0100 From: double User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: libssh2 development Subject: Transfer files X-Y-GMX-Trusted: 0 X-FuHaFi: 0.73999999999999999 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hello, Is there an easy way to transfer a file from server "B" to server "C", if libssh2 runs on server "A"? To login on server "B" and put a file to server "C" via sftp? Thank you very much Marcus _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 12:02:46 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CB2Qmq021889; Fri, 12 Mar 2010 12:02:35 +0100 Received: from mail-ww0-f54.google.com (mail-ww0-f54.google.com [74.125.82.54]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CB2OL2021829 for ; Fri, 12 Mar 2010 12:02:25 +0100 Received: by wwb24 with SMTP id 24so84395wwb.41 for ; Fri, 12 Mar 2010 03:02:21 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=k6oVQwS1Ev8T3puVUvZ669Owitt2d50S1GTYnL5Ihik=; b=DPkVvsZcQ/gX7e9oPWoRVMVtHrwSdX96aCj+oyQ6cSiW5qaf5fZQ62GuCVkd+neNrD /IdAEIzcZOfse+iQ2BtxZC565QRq6qBia6FIdIg+EORDzYg6ItFDoa6VWA0hD9Ftgqac R1zEieJjDHJWuH/8Z5DLCaAjbFucvpyDxDRCU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=G0rM700qmWMVsC493WJh7Qmic61sNgKUS0iYzzYJg2st7lPVcLFA91s8iZRw3hQYWQ MZw2PamyDBe5lZkOXrRmx3VzMuKE7XD3Ikikv1h4xpdowcTrbJ7d+UjDRF3AekjWNB0o vqEYNekS4BFta44WZSvzVR1vty9NIMgEem9Wo= MIME-Version: 1.0 Received: by 10.216.90.6 with SMTP id d6mr1460846wef.89.1268391741478; Fri, 12 Mar 2010 03:02:21 -0800 (PST) In-Reply-To: References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> Date: Fri, 12 Mar 2010 11:02:21 +0000 X-Google-Sender-Auth: 4c1f7f87b15ccf55 Message-ID: <9c3275a91003120302k5263afblce4882830ffb466d@mail.gmail.com> Subject: Re: so name bump; impact on major, minor and patch version From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 11 March 2010 20:53, Daniel Stenberg wrote: > > Ideally, I would like us to gently with the bump. Possibly start off by > adding #ifdefs or similar so that we can build the existing or the new ABI, > so that we can allow apps to try it out easier and get a feel for what's new > and what changes their apps require. We can then introduce the new ABI but > not built by default in a release or two before we take the jump and cut out > the old and run with only the new. This sounds great. Those of us for whom ABI doesn't matter (Windows users shipping custom-built DLLs) can test out the new API before everyone else takes the leap. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 12:34:50 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBYidf014096; Fri, 12 Mar 2010 12:34:48 +0100 Received: from mail-ww0-f54.google.com (mail-ww0-f54.google.com [74.125.82.54]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBYg9V014045 for ; Fri, 12 Mar 2010 12:34:42 +0100 Received: by wwb24 with SMTP id 24so100002wwb.41 for ; Fri, 12 Mar 2010 03:34:39 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=u5EC98CBP4faFojbkVbOlOQ1uYL25HT1rWaq9ggJ1X4=; b=GiGCzRokh+/fOHDjIHaKEETecdmIy4EXMRkoDDob+dUtX0krugyBrlNsGyxIkK2qkk YrWgpsdyxYaRtY1wcaApmyrfUBHCa9u4kT1/7ZiBY1FIXJnlpC+14PC4x/dHkCrNO+rc 4l2n7socid6DwFTgz7Ixsk4vMI3VjaRpa7TNE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=S2sP3UOyNAVtbeAEM91FQHq8qp+iAZuuCjYsYnLeHF0XR7LJdp+qOaKLouUfKM4Iou S154ailLne0e3eUyW8iWP8lKzMVNesoSRwpKv6jx3n/UpSVKB/SPgHxaHhdM7f7l+dTZ 9eqAF5EO6cvjElOzqUergr5nwEkobT4+uGHLI= MIME-Version: 1.0 Received: by 10.216.89.135 with SMTP id c7mr616861wef.62.1268393679343; Fri, 12 Mar 2010 03:34:39 -0800 (PST) In-Reply-To: <000301cac08e$16ac2570$44047050$@com> References: <5862ce3f1003090601o15a20823w471e0ef363f7fe62@mail.gmail.com> <87tysolhp4.fsf@mocca.josefsson.org> <87sk88jybh.fsf@mocca.josefsson.org> <9c3275a91003100402i54f6041dt77f05d6da640111f@mail.gmail.com> <878wa0ic50.fsf@mocca.josefsson.org> <20100310170325.17030.qmail@stuge.se> <9c3275a91003100939w17cf37f7w2cf4e695bee2cc9b@mail.gmail.com> <87r5nskqtx.fsf@mocca.josefsson.org> <000301cac08e$16ac2570$44047050$@com> Date: Fri, 12 Mar 2010 11:34:39 +0000 X-Google-Sender-Auth: 560ebe7a2e398ad3 Message-ID: <9c3275a91003120334m1fb78c2fr276fae0ec0d3ed9a@mail.gmail.com> Subject: Re: Visual Studio 6 compilation issues (libssh2 1.2.4) From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Just briefly getting back to the original purpose of this thread :-P I've committed the VC6 fix. Let me know if you have any problems. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 12:37:57 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBbrJK015894; Fri, 12 Mar 2010 12:37:55 +0100 Received: from dottedmag.net (burger.dottedmag.net [212.75.37.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBbqHG015882 for ; Fri, 12 Mar 2010 12:37:52 +0100 Received: from vertex.dottedmag (unknown [91.197.127.125]) by dottedmag.net (Postfix) with ESMTPSA id C0EA08C072 for ; Fri, 12 Mar 2010 12:37:53 +0100 (CET) Received: from dottedmag by vertex.dottedmag with local (Exim 4.69) (envelope-from ) id 1Nq3Be-0003eq-LK for libssh2-devel@cool.haxx.se; Fri, 12 Mar 2010 17:37:50 +0600 From: Mikhail Gusarov To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master c585828... Fix gettimeofday to compile with Visual C++ 6. References: <20100312113244.16168.qmail@earth.stuge.se> Date: Fri, 12 Mar 2010 17:37:49 +0600 In-Reply-To: <20100312113244.16168.qmail@earth.stuge.se> (libssh2@git.stuge.se's message of "12 Mar 2010 11:32:44 -0000") Message-ID: <87y6hx6a9e.fsf@vertex.dottedmag> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0880605492==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0880605492== Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --=-=-= Content-Transfer-Encoding: quoted-printable Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.se did gyre and gimble: l> - unsigned long long ns100; /*time since 1 Jan 1601 in 100ns units */ l> + unsigned __int64 ns100; /*time since 1 Jan 1601 in 100ns units */ Does it work with gcc now? =2D-=20 http://fossarchy.blogspot.com/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iQIcBAEBAgAGBQJLmieOAAoJEJ0g9lA+M4iIJ60P/iktAhH7Hscsf8F30mSHO52J 0bZPacrMNRwYAVDasOPRoYzKurWRPIobH/voSEt8culONx48K13RNlORI0LFGNih TzNm9UWyRoN66ZBBukEtoSMK85zz4FUc977/ZXBFpzgTaFFmZjQ2cNTMYmn4a2Yz AukXpbbsyvTzh8R+hK6ZJ+AumBLS0kiN6SaOWGCLgfIVyzauRruQoxOxxbw2yaNX WKr51ls1pApO5eIFrLgQnzJxxVYKLZ077Rtc9Lwbrxdit4aUfGi1yu04a1JM/ftk /sYRt5jbv6kQVhWmVkJbZduc9Zl7o7F+ksy5KP4pUhJzOufFdJRMiCZhv5K/1dWk J7E2iUr92dwFJSK8khYEoHw35ddVpPaIFjFcdvx4Ak750pl/JNYy523ezV3bhVT8 AjBGA+Gx1nSLl1KN9e3u9zWoW4vqu00l2wvZ29n/NMRTC6CWpaBTcoDXIfLIwVrx 1uajkUsOEAqU7asL+XbKJYKxidR23Hjtjul7key1fk+5Z3KqMpq+eHtlqjzLUHRm 95GO22nAHAUq43UMrZFLbncvY05LHQvn1cb1/NCXpQjGcdYmAukBWOWqUiX+VPVU YuzpXzfSuY+zSm2bZ8jQG8GIYF8aUQLC+eAfrvgL7W4Z9URR5lnAnFwlxrhYxVFk YhHuBrP0gWSR4ijPuEyy =7TUw -----END PGP SIGNATURE----- --=-=-=-- --===============0880605492== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0880605492==-- From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 12:52:16 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBpsCM027857; Fri, 12 Mar 2010 12:51:58 +0100 Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBprQH027818 for ; Fri, 12 Mar 2010 12:51:53 +0100 Received: by wyb33 with SMTP id 33so486612wyb.41 for ; Fri, 12 Mar 2010 03:51:49 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=DrX0E58n9SunpseEh0UCkZcSxnRZ95areLGCo47pqDM=; b=c85XON774nWpbMtQRDeX6JPxmPqXxae0denfT9b6uKRvsDGwPRJ1ROHTEGOp3QrXjD m6MZsUnYOEnU1KOXWPCYymMdZhGwdmsVbdlvq4Mk27CxgQwT5uH1DK3OmSBAwFmAnHs3 ABJbDnw/uL1WtemMaBAYGKosDidH4ftH//qMc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=CXV4Xd6rfknbtvEFMBymbjQ7Hd5wq+eqW5zl26nU911gTuwVdLQ1CI9svlKMrR3Gy1 U6ZlukszZJWNksbzkcHnjY/D/5aj+nY3jOXWWaJYdQQn7/mWaPUC1wqKhX/VoeK8X/OD LNY7P5Vico6XzJ/0RPHed8qqpk0AlPOeEqdII= MIME-Version: 1.0 Received: by 10.216.88.7 with SMTP id z7mr626107wee.19.1268394707540; Fri, 12 Mar 2010 03:51:47 -0800 (PST) In-Reply-To: <87d3zcicqt.fsf@mocca.josefsson.org> References: <87d3zcicqt.fsf@mocca.josefsson.org> Date: Fri, 12 Mar 2010 11:51:47 +0000 X-Google-Sender-Auth: 3a78121344217631 Message-ID: <9c3275a91003120351y79622d39w1655980b5548bbbe@mail.gmail.com> Subject: Re: more keep-alive stuff From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 10 March 2010 12:26, Simon Josefsson wrote: > Folks, > > I have pushed support for sending keepalive requests. =A0(Recall, my > earlier work on keepalive support was to _receive_ keepalive requests.) ... snip Simon, I hope you don't mind, I've taken the text you posted here and made a HOWTO article out of it on the Trac wiki: http://libssh2.stuge.se/wiki/KeepAlive. I've noticed that lots of excellent explanations are being lost here on the mailing list. These explanations tend to be longer and from a different perspective than official API documentation or man pages so until now they've not had a home. This seems a shame. So, from now on, I will try to add them to a HOWTOs section on Trac. If anyone can point me towards other useful mailing-list posts I'll add them too. Of course, feel free to write your own on the Wiki :) Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 12:54:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBs2dC028246; Fri, 12 Mar 2010 12:54:03 +0100 Received: from mail-ww0-f54.google.com (mail-ww0-f54.google.com [74.125.82.54]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CBs0PO028190 for ; Fri, 12 Mar 2010 12:54:00 +0100 Received: by wwb24 with SMTP id 24so109929wwb.41 for ; Fri, 12 Mar 2010 03:53:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=PG0UsIX/8rqTWFJxOIdXa1UDc2DlR/SIFy09cM+uFRk=; b=rwAb4b4KTmkdXAN1ZUv0liXEaenwtZmdb3kjCjmYvmBY1pwKJR4s5/RZ04eTts409E mkkKz7sN++Go0nX2GmAvGhR6fzFYObFm7AMYtDHOTCBA2ZvTDtKFj2QQislYya6TKotu EJqIxKpxDV9keVPHcjE3c1Nv/QSGK8Mogri4A= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=ro3+AGRNaPka0Hdn3t4lkhbowDgdxjXKGRx4mF8P4/+VFss2OpFL6RytPXm6WRCrGL lg21nzQBM3OiaeJHi8mjjBD4+yYuP8vttXymdhOg8e+/S0zod4GjzseNjG8OAY5NZF0O htlK9MBooUrqbczVg1n57glk0lleQgIMTErZk= MIME-Version: 1.0 Received: by 10.216.90.6 with SMTP id d6mr1501597wef.89.1268394837813; Fri, 12 Mar 2010 03:53:57 -0800 (PST) In-Reply-To: <87y6hx6a9e.fsf@vertex.dottedmag> References: <20100312113244.16168.qmail@earth.stuge.se> <87y6hx6a9e.fsf@vertex.dottedmag> Date: Fri, 12 Mar 2010 11:53:57 +0000 X-Google-Sender-Auth: 39e695f1149d784c Message-ID: <9c3275a91003120353v7438fd95x1d6ef5fa1131be01@mail.gmail.com> Subject: Re: libssh2 master c585828... Fix gettimeofday to compile with Visual C++ 6. From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 12 March 2010 11:37, Mikhail Gusarov wrote: > > Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.se did > gyre and gimble: > > =A0l> - =A0 =A0unsigned long long ns100; /*time since 1 Jan 1601 in 100ns= units */ > =A0l> + =A0 =A0unsigned __int64 ns100; /*time since 1 Jan 1601 in 100ns u= nits */ > > Does it work with gcc now? I assume you mean when compiling with gcc on/for Windows? Good point. Does Mingw etc. support __int64? _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 13:51:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CCpGr9008221; Fri, 12 Mar 2010 13:51:26 +0100 Received: from mail-bw0-f226.google.com (mail-bw0-f226.google.com [209.85.218.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CCpF23008217 for ; Fri, 12 Mar 2010 13:51:15 +0100 Received: by bwz26 with SMTP id 26so93158bwz.7 for ; Fri, 12 Mar 2010 04:51:12 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:from:date:x-google-sender-auth:message-id:subject:to :content-type; bh=pQE+nqm5FvEXbdIPBhvC9tgPNFii/hauk3+yDX30PqE=; b=rS7TDo7x3KShDju/8zrneN3ChkXm6WT2ta/MHekZ1TKQ9For6URLA8lN/ow+teohbp C+J2bMFBqHRF0n277rO9OoKCwsyoAyyUK+H6Im7KnjeM9II24fTzFQQJqjYcL4+oyXBo VUgeDXVQQUMAXW/p/NdZOD1EWpcTnXbB58Dno= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:content-type; b=ki3fgQjUJa7ZwuKpemNWbYMy7cuWVybHZr43NNTJc10DtL2kHUYX889O1/n6az59bn kbOPFJKV9lfkApJ6d3DAO+w/fyzlMp9pRt164KTp8Xsr0Pm76BN49gAbpLRn7tTq64rV jvY3SiTd7iWwxNh9Cqf5qxEeIHILxDoSAbdRQ= MIME-Version: 1.0 Received: by 10.204.25.130 with SMTP id z2mr5417970bkb.211.1268398272257; Fri, 12 Mar 2010 04:51:12 -0800 (PST) In-Reply-To: <9c3275a91003120353v7438fd95x1d6ef5fa1131be01@mail.gmail.com> References: <20100312113244.16168.qmail@earth.stuge.se> <87y6hx6a9e.fsf@vertex.dottedmag> <9c3275a91003120353v7438fd95x1d6ef5fa1131be01@mail.gmail.com> From: Steven Van Ingelgem Date: Fri, 12 Mar 2010 13:50:51 +0100 X-Google-Sender-Auth: b0545fc715565a5c Message-ID: <5862ce3f1003120450g7e81077fv859e84d4f74f4628@mail.gmail.com> Subject: Re: libssh2 master c585828... Fix gettimeofday to compile with Visual C++ 6. To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0295598081==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0295598081== Content-Type: multipart/alternative; boundary=00032555d23273a05d048199fc41 --00032555d23273a05d048199fc41 Content-Type: text/plain; charset=UTF-8 _mingw.h: __int64 define to be long long. Using a typedef _mingw.h: doesn't work for "unsigned __int64" _mingw.h:# ifndef __int64 _mingw.h:# define __int64 long long On 12 March 2010 12:53, Alexander Lamaison wrote: > On 12 March 2010 11:37, Mikhail Gusarov wrote: > > > > Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.se did > > gyre and gimble: > > > > l> - unsigned long long ns100; /*time since 1 Jan 1601 in 100ns units > */ > > l> + unsigned __int64 ns100; /*time since 1 Jan 1601 in 100ns units > */ > > > > Does it work with gcc now? > > I assume you mean when compiling with gcc on/for Windows? Good point. > Does Mingw etc. support __int64? > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > --00032555d23273a05d048199fc41 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable _mingw.h:=C2=A0=C2=A0 __int64=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0 define to be long long.=C2=A0 Using a typedef
_= mingw.h:=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 doesn't= work for "unsigned __int64"
_mingw.h:# ifndef __int64
_min= gw.h:#=C2=A0 define __int64 long long

On 12 March 2010 12:53, Alexander Lamaison <= span dir=3D"ltr"><swish@lammy.co.uk= > wrote:
On 12 March 2010 11:37, Mikhail Gusarov <dottedmag@dottedmag.net> wrote:
>
> Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.se did
> gyre and gimble:
>
> =C2=A0l> - =C2=A0 =C2=A0unsigned long long ns100; /*time since 1 Ja= n 1601 in 100ns units */
> =C2=A0l> + =C2=A0 =C2=A0unsigned __int64 ns100; /*time since 1 Jan = 1601 in 100ns units */
>
> Does it work with gcc now?

I assume you mean when compiling with gcc on/for Windows? =C2=A0Good = point.
=C2=A0Does Mingw etc. support __int64?
_________________________________________= ______
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/li= bssh2-devel

--00032555d23273a05d048199fc41-- --===============0295598081== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0295598081==-- From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 13:54:27 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CCsOjH009002; Fri, 12 Mar 2010 13:54:25 +0100 Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CCsNXj008977 for ; Fri, 12 Mar 2010 13:54:23 +0100 Received: by wyb33 with SMTP id 33so509471wyb.41 for ; Fri, 12 Mar 2010 04:54:20 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=8GIUe52HOJ1CWQL9tWFle892hQOQCMsZKJ7Ct5nhUyo=; b=OnuSrkIfoGrQby9hR8hLtoTff+CI1boGFc97ExwZfN2lwW4+TWrbSGTNVL+KFD15/6 DRjceBaezsyY5yfy7Ddizx82fOQKOzp+OjSIZWpyjFQRZviQ0e83gYiMKe65nulcBkVu j+FT9qwq7TH26O2ZunOjrq3FVMV7AALbpTna8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=VyxfLnykXqAhauQXOfmLAwqlXnf3wzYg+8X1v5UAi6Q/gpdrtWMQoMK6JJS5czA4l7 fFI9YIICCYAeMFD1Jj+buNR5bL/fNm1rSkltHAt0P/InV7CruF/tz+MApQbrurGGx8ug zHqcdvT8ECny9hFXURe7S5iT8ZTSzp3ZrcB+Q= MIME-Version: 1.0 Received: by 10.216.157.129 with SMTP id o1mr759266wek.197.1268398460219; Fri, 12 Mar 2010 04:54:20 -0800 (PST) In-Reply-To: <5862ce3f1003120450g7e81077fv859e84d4f74f4628@mail.gmail.com> References: <20100312113244.16168.qmail@earth.stuge.se> <87y6hx6a9e.fsf@vertex.dottedmag> <9c3275a91003120353v7438fd95x1d6ef5fa1131be01@mail.gmail.com> <5862ce3f1003120450g7e81077fv859e84d4f74f4628@mail.gmail.com> Date: Fri, 12 Mar 2010 12:54:20 +0000 X-Google-Sender-Auth: e8cfe3555b293933 Message-ID: <9c3275a91003120454i6e1dc326m43bb314ece3d09f9@mail.gmail.com> Subject: Re: libssh2 master c585828... Fix gettimeofday to compile with Visual C++ 6. From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 12 March 2010 12:50, Steven Van Ingelgem wrote: > > On 12 March 2010 12:53, Alexander Lamaison wrote: >> >> On 12 March 2010 11:37, Mikhail Gusarov wrote: >> > >> > Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.se d= id >> > gyre and gimble: >> > >> > =A0l> - =A0 =A0unsigned long long ns100; /*time since 1 Jan 1601 in 10= 0ns >> > units */ >> > =A0l> + =A0 =A0unsigned __int64 ns100; /*time since 1 Jan 1601 in 100n= s units >> > */ >> > >> > Does it work with gcc now? >> >> I assume you mean when compiling with gcc on/for Windows? =A0Good point. >> =A0Does Mingw etc. support __int64? > > _mingw.h: __int64 define to be long long. Usin= g a > typedef > _mingw.h: doesn't work for "unsigned > __int64" > _mingw.h:# ifndef __int64 > _mingw.h:# define __int64 long long Is _mingw.h included by default or do we have to do it explicitly? _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 14:29:46 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CDTdbb005037; Fri, 12 Mar 2010 14:29:44 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CDTcKP005034 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 12 Mar 2010 14:29:38 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2CDTcWK009012 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Fri, 12 Mar 2010 14:29:40 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: more keep-alive stuff References: <87d3zcicqt.fsf@mocca.josefsson.org> <9c3275a91003120351y79622d39w1655980b5548bbbe@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100312:libssh2-devel@cool.haxx.se::vALQw0uMRwuPBTwU:1NU/ Date: Fri, 12 Mar 2010 14:29:38 +0100 In-Reply-To: <9c3275a91003120351y79622d39w1655980b5548bbbe@mail.gmail.com> (Alexander Lamaison's message of "Fri, 12 Mar 2010 11:51:47 +0000") Message-ID: <87ocitacsd.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Alexander Lamaison writes: > On 10 March 2010 12:26, Simon Josefsson wrote: >> Folks, >> >> I have pushed support for sending keepalive requests. =A0(Recall, my >> earlier work on keepalive support was to _receive_ keepalive requests.) > ... snip > > Simon, I hope you don't mind, I've taken the text you posted here and > made a HOWTO article out of it on the Trac wiki: > http://libssh2.stuge.se/wiki/KeepAlive. > > I've noticed that lots of excellent explanations are being lost here > on the mailing list. These explanations tend to be longer and from a > different perspective than official API documentation or man pages so > until now they've not had a home. This seems a shame. > > So, from now on, I will try to add them to a HOWTOs section on Trac. > If anyone can point me towards other useful mailing-list posts I'll > add them too. Of course, feel free to write your own on the Wiki :) Great, thank you! /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 12 15:09:17 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CE94iV030478; Fri, 12 Mar 2010 15:09:14 +0100 Received: from mail-bw0-f226.google.com (mail-bw0-f226.google.com [209.85.218.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2CE8wVG030300 for ; Fri, 12 Mar 2010 15:09:02 +0100 Received: by bwz26 with SMTP id 26so159987bwz.7 for ; Fri, 12 Mar 2010 06:08:56 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:from:date:x-google-sender-auth:message-id:subject:to :content-type; bh=hunF7ghqreikeKMwB47RBcR5hR8FG+zzkManA/qWmX0=; b=lAvBCSlZTRkIXlhQY2FdtOyHbocx46y8Tl+ErhFWjRlzINOOzzZhPCT75lauDQQnuG 5u/IuTN5Mz1cWx5UGmddTLNZfr5LwyZ4ViECUkiEPacB1k1Jh9cDaiL7TW7uEqwcQg06 d1s4XpQuVpcGkwRRzgicM55dbT0+Aj8ROkx8U= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:content-type; b=UMXDEaSTqDtkK13rnq7FcbMkty+AqANY2tLFqA3g2V7dO4+1CCjcmQ1MtI77VZRgGl yWroLp0q6TSF0t16hpYVayta1JLPJOBjI39yRrrDY/2HSOZVJPi5E01vbpvMavR9itBD 1rKv9hSdLrUrNUWJhiZW0JRfFcnT08MAZA1R8= MIME-Version: 1.0 Received: by 10.204.141.69 with SMTP id l5mr886410bku.64.1268402936139; Fri, 12 Mar 2010 06:08:56 -0800 (PST) In-Reply-To: <9c3275a91003120454i6e1dc326m43bb314ece3d09f9@mail.gmail.com> References: <20100312113244.16168.qmail@earth.stuge.se> <87y6hx6a9e.fsf@vertex.dottedmag> <9c3275a91003120353v7438fd95x1d6ef5fa1131be01@mail.gmail.com> <5862ce3f1003120450g7e81077fv859e84d4f74f4628@mail.gmail.com> <9c3275a91003120454i6e1dc326m43bb314ece3d09f9@mail.gmail.com> From: Steven Van Ingelgem Date: Fri, 12 Mar 2010 15:08:36 +0100 X-Google-Sender-Auth: 59bc1c17c9d7ea1a Message-ID: <5862ce3f1003120608h6bc9bd2fn7db6de02b44d6a65@mail.gmail.com> Subject: Re: libssh2 master c585828... Fix gettimeofday to compile with Visual C++ 6. To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1962338418==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============1962338418== Content-Type: multipart/alternative; boundary=00151747927a70d02d04819b129d --00151747927a70d02d04819b129d Content-Type: text/plain; charset=UTF-8 I assume it is included by default: C:\tmp\MinGW 3.4.5\include>g -n _mingw.h * _mingw.h:3: * _mingw.h assert.h:18:#include <_mingw.h> c++/3.4.5/mingw32/bits/gthr-default.h:68:#include <_mingw.h> c++/3.4.5/mingw32/bits/gthr-default.h:357:extern int __mingwthr_key_dtor (unsigned long, void (*) (void *)); c++/3.4.5/mingw32/bits/gthr-default.h:524: status = __mingwthr_key_dtor (*key, dtor); complex.h:25:#include <_mingw.h> conio.h:16:#include <_mingw.h> ctype.h:15:#include <_mingw.h> direct.h:15:#include <_mingw.h> dirent.h:12:#include <_mingw.h> dos.h:15:#include <_mingw.h> errno.h:15:#include <_mingw.h> excpt.h:18:#include <_mingw.h> fcntl.h:15:#include <_mingw.h> fenv.h:4:#include <_mingw.h> float.h:25:#include <_mingw.h> getopt.h:33:#include <_mingw.h> inttypes.h:6:#include <_mingw.h> io.h:14:#include <_mingw.h> libgen.h:17:#include <_mingw.h> limits.h:21:#include <_mingw.h> locale.h:16:#include <_mingw.h> malloc.h:18:#include <_mingw.h> math.h:20:#include <_mingw.h> mbctype.h:25:#include <_mingw.h> mbstring.h:25:#include <_mingw.h> process.h:15:#include <_mingw.h> search.h:27:#include <_mingw.h> setjmp.h:16:#include <_mingw.h> share.h:15:#include <_mingw.h> signal.h:15:#include <_mingw.h> stdio.h:19:#include <_mingw.h> stdlib.h:15:#include <_mingw.h> string.h:15:#include <_mingw.h> sys/locking.h:15:#include <_mingw.h> sys/stat.h:16:#include <_mingw.h> sys/timeb.h:15:#include <_mingw.h> sys/types.h:15:#include <_mingw.h> sys/utime.h:14:#include <_mingw.h> tchar.h:26:#include <_mingw.h> time.h:15:#include <_mingw.h> wchar.h:18:#include <_mingw.h> wctype.h:27:#include <_mingw.h> On 12 March 2010 13:54, Alexander Lamaison wrote: > On 12 March 2010 12:50, Steven Van Ingelgem wrote: > > > > On 12 March 2010 12:53, Alexander Lamaison wrote: > >> > >> On 12 March 2010 11:37, Mikhail Gusarov > wrote: > >> > > >> > Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.sedid > >> > gyre and gimble: > >> > > >> > l> - unsigned long long ns100; /*time since 1 Jan 1601 in 100ns > >> > units */ > >> > l> + unsigned __int64 ns100; /*time since 1 Jan 1601 in 100ns > units > >> > */ > >> > > >> > Does it work with gcc now? > >> > >> I assume you mean when compiling with gcc on/for Windows? Good point. > >> Does Mingw etc. support __int64? > > > > _mingw.h: __int64 define to be long long. > Using a > > typedef > > _mingw.h: doesn't work for "unsigned > > __int64" > > _mingw.h:# ifndef __int64 > > _mingw.h:# define __int64 long long > > Is _mingw.h included by default or do we have to do it explicitly? > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > --00151747927a70d02d04819b129d Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable I assume it is included by default:

C:\tmp\MinGW 3.4.5\include>g = -n _mingw.h *
_mingw.h:3: * _mingw.h
assert.h:18:#include <_mingw.= h>
c++/3.4.5/mingw32/bits/gthr-default.h:68:#include <_mingw.h>=
c++/3.4.5/mingw32/bits/gthr-default.h:357:extern int __mingwthr_key_dtor (u= nsigned long, void (*) (void *));
c++/3.4.5/mingw32/bits/gthr-default.h:= 524:=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 status =3D __mingwthr_key_dtor (*key, dt= or);
complex.h:25:#include <_mingw.h>
conio.h:16:#include <_mingw.h>
ctype.h:15:#include <_mingw.h>= ;
direct.h:15:#include <_mingw.h>
dirent.h:12:#include <_min= gw.h>
dos.h:15:#include <_mingw.h>
errno.h:15:#include <_= mingw.h>
excpt.h:18:#include <_mingw.h>
fcntl.h:15:#include <_mingw.h>= ;
fenv.h:4:#include <_mingw.h>
float.h:25:#include <_mingw.h= >
getopt.h:33:#include <_mingw.h>
inttypes.h:6:#include <= _mingw.h>
io.h:14:#include <_mingw.h>
libgen.h:17:#include <_mingw.h><= br>limits.h:21:#include <_mingw.h>
locale.h:16:#include <_mingw= .h>
malloc.h:18:#include <_mingw.h>
math.h:20:#include <_= mingw.h>
mbctype.h:25:#include <_mingw.h>
mbstring.h:25:#include <_mingw= .h>
process.h:15:#include <_mingw.h>
search.h:27:#include &l= t;_mingw.h>
setjmp.h:16:#include <_mingw.h>
share.h:15:#incl= ude <_mingw.h>
signal.h:15:#include <_mingw.h>
stdio.h:19:#include <_mingw.h&g= t;
stdlib.h:15:#include <_mingw.h>
string.h:15:#include <_mi= ngw.h>
sys/locking.h:15:#include <_mingw.h>
sys/stat.h:16:#i= nclude <_mingw.h>
sys/timeb.h:15:#include <_mingw.h>
sys/types.h:15:#include <_mi= ngw.h>
sys/utime.h:14:#include <_mingw.h>
tchar.h:26:#includ= e <_mingw.h>
time.h:15:#include <_mingw.h>
wchar.h:18:#in= clude <_mingw.h>
wctype.h:27:#include <_mingw.h>


On 12 March 2010 13:54, Alexander Lamaison <swish@lammy.co.uk> wrote:
On 12 March 2010 12:50, Steven Van Ingelgem <steven@vaningelgem.be> wrote:
>
> On 12 March 2010 12:53, Alexander Lamaison <swish@lammy.co.uk> wrote:
>>
>> On 12 March 2010 11:37, Mikhail Gusarov <dottedmag@dottedmag.net> wrote:
>> >
>> > Twas brillig at 11:32:44 12.03.2010 UTC+00 when libssh2@git.stuge.se did
>> > gyre and gimble:
>> >
>> > =C2=A0l> - =C2=A0 =C2=A0unsigned long long ns100; /*time s= ince 1 Jan 1601 in 100ns
>> > units */
>> > =C2=A0l> + =C2=A0 =C2=A0unsigned __int64 ns100; /*time sin= ce 1 Jan 1601 in 100ns units
>> > */
>> >
>> > Does it work with gcc now?
>>
>> I assume you mean when compiling with gcc on/for Windows? =C2=A0Go= od point.
>> =C2=A0Does Mingw etc. support __int64?
>
> _mingw.h: =C2=A0 __int64 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 define to be long long. =C2=A0Using = a
> typedef
> _mingw.h: =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 doesn't wor= k for "unsigned
> __int64"
> _mingw.h:# ifndef __int64
> _mingw.h:# =C2=A0define __int64 long long

Is _mingw.h included by default or do we have to do it explicitly?
_________________________________________= ______
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/li= bssh2-devel

--00151747927a70d02d04819b129d-- --===============1962338418== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1962338418==-- From libssh2-devel-bounces@cool.haxx.se Sat Mar 13 22:07:26 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2DL76IB024119; Sat, 13 Mar 2010 22:07:20 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2DL74iX024100 for ; Sat, 13 Mar 2010 22:07:04 +0100 Date: Sat, 13 Mar 2010 22:07:04 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: dispatch_protocol_error: type 90 seq 4 In-Reply-To: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> Message-ID: References: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Thu, 11 Mar 2010, Dave McCaldon wrote: > I have a Windows client app using libssh2-1.2.3 that seems to be unable to > connect to a remote server (OpenSSH 5.2) to do SFTP, it fails on the server > side with: And other tools have no problems using SFTP against this server? > dispatch_protocol_error: type 90 seq 4 > [libssh2] 44.093943 Transport: Disconnect(7): Service not requested or > authenticated(en) > Does anyone have any more details on what this means? (configuration > perhaps?) That's the error message from the server so I figure more research into what makes the server say so is required to better understand this. > => libssh2_transport_write plain (24 bytes) > 0000: 5A 00 00 00 07 73 65 73 73 69 6F 6E 00 00 00 00 : Z....session.... > 0010: 00 01 00 00 00 00 80 00 : ........ There's the type 90 I believe (0x5A). > => libssh2_transport_read() plain (53 bytes) > 0000: 01 00 00 00 07 00 00 00 26 53 65 72 76 69 63 65 : ........&Service > 0010: 20 6E 6F 74 20 72 65 71 75 65 73 74 65 64 20 6F : not requested o > 0020: 72 20 61 75 74 68 65 6E 74 69 63 61 74 65 64 00 : r authenticated. > 0030: 00 00 02 65 6E : ...en And that's the server denying the client to play. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 03:01:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2E21R46007418; Sun, 14 Mar 2010 03:01:37 +0100 Received: from mail.intersystems.com (mail.intersys.COM [198.133.74.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2E21PvT007184 for ; Sun, 14 Mar 2010 03:01:26 +0100 X-InterSystems: Sent from InterSystems X-InterSystems: Sent from InterSystems Mime-Version: 1.0 (Apple Message framework v1077) Subject: Re: dispatch_protocol_error: type 90 seq 4 From: Dave McCaldon In-Reply-To: Date: Sat, 13 Mar 2010 21:01:28 -0500 Message-Id: References: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> To: libssh2 development X-Mailer: Apple Mail (2.1077) X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mar 13, 2010, at 4:07 PM, Daniel Stenberg wrote: > On Thu, 11 Mar 2010, Dave McCaldon wrote: > >> I have a Windows client app using libssh2-1.2.3 that seems to be unable to connect to a remote server (OpenSSH 5.2) to do SFTP, it fails on the server side with: > > And other tools have no problems using SFTP against this server? I don't have direct access to the server, but I believe that other SFTP tools are ok. I will check into it. >> dispatch_protocol_error: type 90 seq 4 > >> [libssh2] 44.093943 Transport: Disconnect(7): Service not requested or authenticated(en) > >> Does anyone have any more details on what this means? (configuration perhaps?) > > That's the error message from the server so I figure more research into what makes the server say so is required to better understand this. >> => libssh2_transport_write plain (24 bytes) >> 0000: 5A 00 00 00 07 73 65 73 73 69 6F 6E 00 00 00 00 : Z....session.... >> 0010: 00 01 00 00 00 00 80 00 : ........ > > There's the type 90 I believe (0x5A). Does anyone know what type 90 is/means? > >> => libssh2_transport_read() plain (53 bytes) >> 0000: 01 00 00 00 07 00 00 00 26 53 65 72 76 69 63 65 : ........&Service >> 0010: 20 6E 6F 74 20 72 65 71 75 65 73 74 65 64 20 6F : not requested o >> 0020: 72 20 61 75 74 68 65 6E 74 69 63 61 74 65 64 00 : r authenticated. >> 0030: 00 00 02 65 6E : ...en > > And that's the server denying the client to play. Right, I almost wonder if it's a configuration issue, but I'd expect a more meaningful error in this case on the server side. Thanks! _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 05:09:59 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2E49lUc011770; Sun, 14 Mar 2010 05:09:56 +0100 Received: from mail-ew0-f226.google.com (mail-ew0-f226.google.com [209.85.219.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2E49kNZ011746 for ; Sun, 14 Mar 2010 05:09:46 +0100 Received: by ewy26 with SMTP id 26so766398ewy.3 for ; Sat, 13 Mar 2010 20:09:41 -0800 (PST) Received: by 10.213.66.1 with SMTP id l1mr1361893ebi.91.1268539780205; Sat, 13 Mar 2010 20:09:40 -0800 (PST) Received: from [192.168.0.35] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 15sm1858258ewy.12.2010.03.13.20.09.39 (version=SSLv3 cipher=RC4-MD5); Sat, 13 Mar 2010 20:09:39 -0800 (PST) Message-ID: <4B9C6182.7090107@peut.org> Date: Sun, 14 Mar 2010 05:09:38 +0100 From: Jose Baars User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; nl; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: so name bump; impact on major, minor and patch version References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> In-Reply-To: <87vdd2hc33.fsf@mocca.josefsson.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0846865009==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a cryptographically signed message in MIME format. --===============0846865009== Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms000204010205020101030609" This is a cryptographically signed message in MIME format. --------------ms000204010205020101030609 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Op 3/11/2010 8:50 PM, Simon Josefsson schreef: > > You should not use the libssh2 release version number for the library > version on VMS -- please compute the shared library version from the > '-version-info 1:1:0' information in src/Makefile.am. The release > version number of libssh2 has nothing to do with shared library > versioning, strictly speaking. See the comments in src/Makefile.am for= > more background, or read the GNU Libtool manual for the entire story. > > > =20 I had a discussion with some VMS people about this, and that is not very = practical or even feasable on VMS due to the different way .so is implemented on unixes and VMS. If you are curious, see the discussion at http://encompasserve.org/anon/htnotes/range?f1=3DPORTING_TO_VMS&f2=3D46.*= &f4=3Dt Jose. --------------ms000204010205020101030609 Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIPWjCC BMwwggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1 MTAyNzIzNTk1OVowgd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEf MB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNl IGF0IGh0dHBzOi8vd3d3LnZlcmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNv bmEgTm90IFZhbGlkYXRlZDE3MDUGA1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFs IFN1YnNjcmliZXIgQ0EgLSBHMjCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnf rOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyVzm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs +Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zplYu//EHuiVrvFTnAt1qIfPO2wQuhejVch rKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFBL2OyOj++pRpu9MlKWz2VphW7NQIZ +dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5gJ925rXXOL3OVekA6hXVJsLjf aLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUCAwEAAaOCAYQwggGAMBIG A1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXATAqMCgGCCsGAQUF BwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIBBjARBglghkgB hvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJlbDMtMjA0 OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAkoCKG IGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEg UHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+Iqyz cqpVMA0GCSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESf D0b3+qD+0x0Yo9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Pr v4NZmP1m3umGMpqSKTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFQTCCBCmgAwIB AgIQQkuBnab/mKbwbYynoi4GkzANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAV BgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3Jr MTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJp U2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyMB4XDTA5MTExMjAw MDAwMFoXDTEwMTExMjIzNTk1OVowggEKMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0G A1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNv bS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElBQi5MVEQoYyk5ODEeMBwGA1UE CxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdpdGFsIElEIENsYXNzIDEg LSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxEzARBgNVBAMUCkpvc2UgQmFhcnMxHDAaBgkqhkiG 9w0BCQEWDXBldXRAcGV1dC5vcmcwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCi x8yI3kFLmfYElzZSqsup782ajeuv3TASyn8mUfGrnr2aW1M7jN4vlM8/7/sVL883rH+kWPzV Q4ITYrfW5QFxH+CBqtX0F1xsS1A7AR8xZIs8o7/sMxV2xOSXYEJXlKP4XVWY8oU+TKH5dWYZ 9BJ9pkJ840ayiqqI/P3MHqmu7vS2nPDu8YZFa+46EtWkGbJTjzFdnrU5Jmdudp37fd1btNss RIkqztDOeQOTaZ6pS9RqjhmNuWVtgZucKho6c9Qf2htGz8AL95MqttF2OZYL9AlXZhjJcIMO aZBVjuwLowX/kYL2lo1vyUGt3eZxSkZR/ICfkQwj1BKpFZGFtM5XAgMBAAGjgcwwgckwCQYD VR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQt Y3JsLnZlcmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEB ABcuOD4QwMrXJgD8zvCWBRjBqsQLXn0ybqrLZHlcic7+lstAl06fi7fGD3+1Nrcp4VImkyUv JQEvFSvkB+/Mt8kOVSdt+eiYIInKt/lBkSzy0sardsh0u7Ifcgt+nmLZYW+azviXnD4VNMqe Q9QIZPHSUhWDe3IA5NSYE4hfqwXX5c13HllqpUcXRmWIGA9MPr9SQxZ0jg7MFUsZVgG9f5Ts A3yIEmvUgutyX6GRzZQL+wlWymRzh/XA7j2QnYk+IPlNQHJ62MHHlAI48y3ssSXwR8hfbH4e 0sw+8ezuh785OuFMBI3qhxlQX4wT8+v8rKHtiLcddoIohWM0ID+IdOEwggVBMIIEKaADAgEC AhBCS4Gdpv+YpvBtjKeiLgaTMA0GCSqGSIb3DQEBBQUAMIHdMQswCQYDVQQGEwJVUzEXMBUG A1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdvcmsx OzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBh IChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZlcmlT aWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzIwHhcNMDkxMTEyMDAw MDAwWhcNMTAxMTEyMjM1OTU5WjCCAQoxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxMzAxBgNVBAsTKkRpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlIEZ1bGwgU2VydmljZTETMBEGA1UEAxQKSm9zZSBCYWFyczEcMBoGCSqGSIb3 DQEJARYNcGV1dEBwZXV0Lm9yZzCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKLH zIjeQUuZ9gSXNlKqy6nvzZqN66/dMBLKfyZR8auevZpbUzuM3i+Uzz/v+xUvzzesf6RY/NVD ghNit9blAXEf4IGq1fQXXGxLUDsBHzFkizyjv+wzFXbE5JdgQleUo/hdVZjyhT5Mofl1Zhn0 En2mQnzjRrKKqoj8/cweqa7u9Lac8O7xhkVr7joS1aQZslOPMV2etTkmZ252nft93Vu02yxE iSrO0M55A5NpnqlL1GqOGY25ZW2Bm5wqGjpz1B/aG0bPwAv3kyq20XY5lgv0CVdmGMlwgw5p kFWO7AujBf+RgvaWjW/JQa3d5nFKRlH8gJ+RDCPUEqkVkYW0zlcCAwEAAaOBzDCByTAJBgNV HRMEAjAAMEQGA1UdIAQ9MDswOQYLYIZIAYb4RQEHFwEwKjAoBggrBgEFBQcCARYcaHR0cHM6 Ly93d3cudmVyaXNpZ24uY29tL3JwYTALBgNVHQ8EBAMCBaAwHQYDVR0lBBYwFAYIKwYBBQUH AwQGCCsGAQUFBwMCMEoGA1UdHwRDMEEwP6A9oDuGOWh0dHA6Ly9JbmRDMURpZ2l0YWxJRC1j cmwudmVyaXNpZ24uY29tL0luZEMxRGlnaXRhbElELmNybDANBgkqhkiG9w0BAQUFAAOCAQEA Fy44PhDAytcmAPzO8JYFGMGqxAtefTJuqstkeVyJzv6Wy0CXTp+Lt8YPf7U2tynhUiaTJS8l AS8VK+QH78y3yQ5VJ2356Jggicq3+UGRLPLSxqt2yHS7sh9yC36eYtlhb5rO+JecPhU0yp5D 1Ahk8dJSFYN7cgDk1JgTiF+rBdflzXceWWqlRxdGZYgYD0w+v1JDFnSODswVSxlWAb1/lOwD fIgSa9SC63JfoZHNlAv7CVbKZHOH9cDuPZCdiT4g+U1AcnrYwceUAjjzLeyxJfBHyF9sfh7S zD7x7O6Hvzk64UwEjeqHGVBfjBPz6/ysoe2Itx12giiFYzQgP4h04TGCBOwwggToAgEBMIHy MIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZl cmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxp ZGF0ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVy IENBIC0gRzICEEJLgZ2m/5im8G2Mp6IuBpMwCQYFKw4DAhoFAKCCAs4wGAYJKoZIhvcNAQkD MQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMTAwMzE0MDQwOTM4WjAjBgkqhkiG9w0B CQQxFgQU8mpJIs0Qy3XjVuFqM+ZD98y/cDgwXwYJKoZIhvcNAQkPMVIwUDALBglghkgBZQME AQIwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIH MA0GCCqGSIb3DQMCAgEoMIIBAwYJKwYBBAGCNxAEMYH1MIHyMIHdMQswCQYDVQQGEwJVUzEX MBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdv cmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20v cnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZl cmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzICEEJLgZ2m/5im 8G2Mp6IuBpMwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTsw OQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAo YykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhBCS4Gdpv+YpvBtjKei LgaTMA0GCSqGSIb3DQEBAQUABIIBABVZsRDdQmDKKpNDw0VdHtAt7OcmU2rX1o7GTfpm/6z8 wsqA8JziBlc06VxfXi1gwT72SKbxo0TEryqQ0PJlB36EcvPqqFMwt+EHMiZjSu5ax8i9WN2B gXLT5GrZ4voqe7ecmVvC3/qwU6p1YZOwtwKVE2BY3SsCmGKwQsEB8iTt0XHJehqW8th97yZE qKYwTrragdzwaFaxel64wW5QUsiKHl9OMXILQxYW2deRS3UlyRQrQEp4iwYyZGGdAgVpkhk8 kyrC4F4lRm8vldVZyagyAxFwKF7K8ISuXsWeo4tV7vslsgyTW3mEoLxv1PdtjYkeUDwdFTwh fwjA+ZkLUI0AAAAAAAA= --------------ms000204010205020101030609-- --===============0846865009== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0846865009==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 09:31:08 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2E8UqqU020618; Sun, 14 Mar 2010 09:31:04 +0100 Received: from idcmail-mo1so.shaw.ca (idcmail-mo1so.shaw.ca [24.71.223.10]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2E8Unxj020614 for ; Sun, 14 Mar 2010 09:30:50 +0100 Received: from pd3ml3so-ssvc.prod.shaw.ca ([10.0.141.149]) by pd3mo1so-svcs.prod.shaw.ca with ESMTP; 14 Mar 2010 01:30:45 -0700 X-Cloudmark-SP-Filtered: true X-Cloudmark-SP-Result: v=1.0 c=1 a=MsizEjY045YA:10 a=VphdPIyG4kEA:10 a=_4SF6Xvc910A:10 a=y9ASP8Sd+Sa59kG1URKZZw==:17 a=d2ew4BzoDWDCulBsfi8A:9 a=s-fDS-VSwJ0yveksNTUA:7 a=GF9_ql87Bdy2Kpn5Q6V2MeXspMgA:4 a=CjuIK1q_8ugA:10 a=yZaBySXCDdIwJFmjaXAA:9 a=qghWhIejDThtFt5rgAYA:7 a=45uaMijHFtUuEoB8oYMG1s4kbu0A:4 Received: from unknown (HELO mail.xepol.com) ([68.146.216.64]) by pd3ml3so-dmz.prod.shaw.ca with ESMTP; 14 Mar 2010 01:30:45 -0700 Received: from quadx64 [192.168.1.3] by mail.xepol.com with SMTP (EHLO Quadx64) (ArGoSoft Mail Server Pro for WinNT/2000/XP, Version 1.8 (1.8.9.2)); Sun, 14 Mar 2010 01:30:36 -0700 From: "C Johnson" To: Subject: Greetings and a quick note about licening issuse in the Makefile.win32 in 1.2.4 Date: Sun, 14 Mar 2010 01:30:34 -0700 Message-ID: MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 11 Thread-Index: AcrDUJ3ZKcxnzxJrRAG7rlhRaTDTkQ== X-MimeOLE: Produced By Microsoft MimeOLE V6.1.7600.16385 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0409306598==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --===============0409306598== Content-Type: multipart/alternative; boundary="----=_NextPart_000_0003_01CAC315.F24729C0" This is a multi-part message in MIME format. ------=_NextPart_000_0003_01CAC315.F24729C0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Greetings all. I'm C Johnson. I have recently gotten involved with libssh2 in the Win32 arena - specifically porting the headers to Delphi. I'm not sure if the Libssh2 group itself will want to include the translated headers as part of the official distro, but once I'm done porting all the demos apps and finished my testing, I would definitely be willing to make them available - increasing the reach of libssh2 even further. I did, however, find one problem with the win32\Makefile.win32 - it lists the license as GPL, and the website lists the license as modified BSD. Specifically line: @echo $(DL) VALUE "License","Released under GPL.\0"$(DL) >> $@ I assume should actually read @echo $(DL) VALUE "License","Released under Modified BSD.\0"$(DL) >> $@ I'm relatively new to the whole world of open source C development (I've been a Delphi developer for a long time now), so if this isn't correct way to raise the issue, I apologize. While I am on the topic of licensing, I found it to be very difficult to track down clear information about the license in either the source tarball or the website itself. Normally I have seen a link on project's websites and a license or license.txt file in the source distro (I've not been an active participant, but I have tried to look in from time to time) - has this step perhaps been considered in past, or was it perhaps just overlooked thus far? For many of us in the Win32 world, licensing terms are VERY important. Again, greetings to all, and I hope to become a more active member of the community as time grows. - C Johnson ------=_NextPart_000_0003_01CAC315.F24729C0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable
Greetings all. =20 I'm C Johnson.  I have recently gotten involved with libssh2 in the = Win32=20 arena - specifically porting the headers to Delphi.  =
 
I'm = not sure if the=20 Libssh2 group itself will want to include the translated headers as part = of the=20 official distro, but once I'm done porting all the demos apps and = finished my=20 testing, I would definitely be willing to make them available - = increasing the=20 reach of libssh2 even further.
 
I did, = however, find=20 one problem with the win32\Makefile.win32 - it lists the license as GPL, = and the=20 website lists the license as modified BSD.
 
Specifically=20 line:
 
 @echo=20 $(DL)      VALUE "License","Released under = GPL.\0"$(DL)=20 >> $@
 
I assume should actually read
 
 @echo=20 $(DL)      VALUE "License","Released under = Modified=20 BSD.\0"$(DL) >> $@
I'm relatively=20 new to the whole world of open source C development (I've been a Delphi=20 developer for a long time now), so if this isn't correct way to raise = the issue,=20 I apologize.
 
While = I am on the=20 topic of licensing, I found it to be very difficult to track down clear=20 information about the license in either the source tarball or the = website=20 itself.  Normally I have seen a link on project's websites and a = license or=20 license.txt file in the source distro (I've not been an active = participant, but=20 I have tried to look in from time to time) - has this step perhaps been=20 considered in past, or was it perhaps just overlooked thus far?  = For many=20 of us in the Win32 world, licensing terms are VERY=20 important.
 
Again, = greetings to=20 all, and I hope to become a more active member of the community as time=20 grows.
 
- C=20 Johnson
------=_NextPart_000_0003_01CAC315.F24729C0-- --===============0409306598== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0409306598==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 19:57:48 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EIvWVN001432; Sun, 14 Mar 2010 19:57:44 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EIvVAM001425 for ; Sun, 14 Mar 2010 19:57:31 +0100 Date: Sun, 14 Mar 2010 19:57:31 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: so name bump; impact on major, minor and patch version In-Reply-To: <4B9C6182.7090107@peut.org> Message-ID: References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Sun, 14 Mar 2010, Jose Baars wrote: >> You should not use the libssh2 release version number for the library >> version on VMS -- please compute the shared library version from the >> '-version-info 1:1:0' information in src/Makefile.am. The release version >> number of libssh2 has nothing to do with shared library versioning, >> strictly speaking. See the comments in src/Makefile.am for more >> background, or read the GNU Libtool manual for the entire story. >> >> > I had a discussion with some VMS people about this, and that is not very > practical or even feasable on VMS due to the different way .so is > implemented on unixes and VMS. It is feasable, even if not optimal, on VMS since I know we do exactly like that in other projects that also build on VMS... If not using libtool and their suggested way of numbering the libs, what exactly are you suggesting? And isn't the problem you refer to rather a problem with libtool rather than with libssh2? -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 20:21:55 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EJLn0m020988; Sun, 14 Mar 2010 20:21:53 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2EJLlOt020970 for ; Sun, 14 Mar 2010 20:21:47 +0100 Received: (qmail 21310 invoked by uid 501); 14 Mar 2010 19:21:53 -0000 Message-ID: <20100314192153.21309.qmail@stuge.se> Date: Sun, 14 Mar 2010 20:21:53 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: so name bump; impact on major, minor and patch version Mail-Followup-To: libssh2-devel@cool.haxx.se References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> MIME-Version: 1.0 In-Reply-To: <4B9C6182.7090107@peut.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0324036379==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0324036379== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="6TrnltStXW4iwmi0" Content-Disposition: inline --6TrnltStXW4iwmi0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Jose Baars wrote: > I had a discussion with some VMS people about this, > http://encompasserve.org/anon/htnotes/range?f1=PORTING_TO_VMS&f2=46.*&f4=t .. > I am not going to bother the libssh2 people with that problem, this > is a VMS specific problem. Even if this problem has not been seen by us before, at least I am quite open to making libssh2 a better VMS citizen. However, it will require some education of us unix guys without strong VMS background. I know next to nothing about VMS so far, but if we can make changes to libssh2 code or development practice to fit better into VMS then I think those changes should be considered very carefully. The overall goal is that libssh2 should be very portable. That means we love VMS too. :) //Peter --6TrnltStXW4iwmi0 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQFLnTdRhR3Q0dhIfEgRAkwQAKCrAAiGTODlg8lf1qFRpGgMsQ1YgwCdG0ht mwVzHcej6hE+05qTqAa2b3s= =SZgg -----END PGP SIGNATURE----- --6TrnltStXW4iwmi0-- --===============0324036379== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0324036379==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 21:03:35 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EK3Lt4015635; Sun, 14 Mar 2010 21:03:32 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EK3Jcn015556 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 14 Mar 2010 21:03:19 +0100 Received: (qmail 1777 invoked from network); 14 Mar 2010 20:03:25 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 14 Mar 2010 20:03:25 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Sun, 14 Mar 2010 20:03:25 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:6 Message-ID: <054.047eaee46f1c7e3516e38175d3bab8c0@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly -----------------------+---------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: new Priority: normal | Milestone: 1.2.5 Component: API | Version: 1.2.4 Keywords: knownhost | Blocking: Blockedby: | -----------------------+---------------------------------------------------- Changes (by alamaison): * milestone: => 1.2.5 Comment: Committed a fix for this: [9abf81de97f5496e904eb855170026dd9fd1a4b8]. It leaves most of the API unchanged but adds a new function `libssh2_knowhost_addc` that takes an optional comment and changes `readline` to use it. The fix isn't perfect as parsing the comment is harder than it seems. `ssh-rsa` and `ssh-dsa` keys can't contain spaces but `rsa1` blocks can. As we don't support checking these anyway it shouldn't matter. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 21:18:17 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EKICpw024707; Sun, 14 Mar 2010 21:18:16 +0100 Received: from mail-ew0-f226.google.com (mail-ew0-f226.google.com [209.85.219.226]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EKIAHk024697 for ; Sun, 14 Mar 2010 21:18:10 +0100 Received: by ewy26 with SMTP id 26so891675ewy.3 for ; Sun, 14 Mar 2010 13:18:11 -0700 (PDT) Received: by 10.213.37.209 with SMTP id y17mr4144360ebd.60.1268597891327; Sun, 14 Mar 2010 13:18:11 -0700 (PDT) Received: from [192.168.0.35] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 15sm2165372ewy.8.2010.03.14.13.18.10 (version=SSLv3 cipher=RC4-MD5); Sun, 14 Mar 2010 13:18:10 -0700 (PDT) Message-ID: <4B9D4481.1040200@peut.org> Date: Sun, 14 Mar 2010 21:18:09 +0100 From: Jose Baars User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; nl; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: so name bump; impact on major, minor and patch version References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1176010058==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a cryptographically signed message in MIME format. --===============1176010058== Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms060202010905040304080206" This is a cryptographically signed message in MIME format. --------------ms060202010905040304080206 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Op 3/14/2010 7:57 PM, Daniel Stenberg schreef: > If not using libtool and their suggested way of numbering the libs,=20 > what exactly are you suggesting? And isn't the problem you refer to=20 > rather a problem with libtool rather than with libssh2? > For the foreseeable future I won't be using libtool, it solves problems=20 that do not exist on VMS, and introduces problems due to the non-native nature of bash and = the gnu tools on VMS, although John Malmberg has been pretty successful in getting it=20 to work for curl and zlib amongst others. My main argument not to use it is that it doesn't really help solving a=20 lot of VMS specifics, handling the 4 configure dependencies and generating a makefile is=20 peanuts compared to that. My personal priority is in getting the SSH functionality available on=20 VMS, and device a sustainable way to keep up with the quick pace in which libssh2 evolves. = At the moment I can't see a lot of benefit from using libtool. --------------ms060202010905040304080206 Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIPWjCC BMwwggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1 MTAyNzIzNTk1OVowgd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEf MB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNl IGF0IGh0dHBzOi8vd3d3LnZlcmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNv bmEgTm90IFZhbGlkYXRlZDE3MDUGA1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFs IFN1YnNjcmliZXIgQ0EgLSBHMjCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnf rOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyVzm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs +Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zplYu//EHuiVrvFTnAt1qIfPO2wQuhejVch rKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFBL2OyOj++pRpu9MlKWz2VphW7NQIZ +dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5gJ925rXXOL3OVekA6hXVJsLjf aLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUCAwEAAaOCAYQwggGAMBIG A1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXATAqMCgGCCsGAQUF BwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIBBjARBglghkgB hvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJlbDMtMjA0 OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAkoCKG IGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEg UHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+Iqyz cqpVMA0GCSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESf D0b3+qD+0x0Yo9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Pr v4NZmP1m3umGMpqSKTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFQTCCBCmgAwIB AgIQQkuBnab/mKbwbYynoi4GkzANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAV BgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3Jr MTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJp U2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyMB4XDTA5MTExMjAw MDAwMFoXDTEwMTExMjIzNTk1OVowggEKMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0G A1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNv bS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElBQi5MVEQoYyk5ODEeMBwGA1UE CxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdpdGFsIElEIENsYXNzIDEg LSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxEzARBgNVBAMUCkpvc2UgQmFhcnMxHDAaBgkqhkiG 9w0BCQEWDXBldXRAcGV1dC5vcmcwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCi x8yI3kFLmfYElzZSqsup782ajeuv3TASyn8mUfGrnr2aW1M7jN4vlM8/7/sVL883rH+kWPzV Q4ITYrfW5QFxH+CBqtX0F1xsS1A7AR8xZIs8o7/sMxV2xOSXYEJXlKP4XVWY8oU+TKH5dWYZ 9BJ9pkJ840ayiqqI/P3MHqmu7vS2nPDu8YZFa+46EtWkGbJTjzFdnrU5Jmdudp37fd1btNss RIkqztDOeQOTaZ6pS9RqjhmNuWVtgZucKho6c9Qf2htGz8AL95MqttF2OZYL9AlXZhjJcIMO aZBVjuwLowX/kYL2lo1vyUGt3eZxSkZR/ICfkQwj1BKpFZGFtM5XAgMBAAGjgcwwgckwCQYD VR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQt Y3JsLnZlcmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEB ABcuOD4QwMrXJgD8zvCWBRjBqsQLXn0ybqrLZHlcic7+lstAl06fi7fGD3+1Nrcp4VImkyUv JQEvFSvkB+/Mt8kOVSdt+eiYIInKt/lBkSzy0sardsh0u7Ifcgt+nmLZYW+azviXnD4VNMqe Q9QIZPHSUhWDe3IA5NSYE4hfqwXX5c13HllqpUcXRmWIGA9MPr9SQxZ0jg7MFUsZVgG9f5Ts A3yIEmvUgutyX6GRzZQL+wlWymRzh/XA7j2QnYk+IPlNQHJ62MHHlAI48y3ssSXwR8hfbH4e 0sw+8ezuh785OuFMBI3qhxlQX4wT8+v8rKHtiLcddoIohWM0ID+IdOEwggVBMIIEKaADAgEC AhBCS4Gdpv+YpvBtjKeiLgaTMA0GCSqGSIb3DQEBBQUAMIHdMQswCQYDVQQGEwJVUzEXMBUG A1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdvcmsx OzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBh IChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZlcmlT aWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzIwHhcNMDkxMTEyMDAw MDAwWhcNMTAxMTEyMjM1OTU5WjCCAQoxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxMzAxBgNVBAsTKkRpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlIEZ1bGwgU2VydmljZTETMBEGA1UEAxQKSm9zZSBCYWFyczEcMBoGCSqGSIb3 DQEJARYNcGV1dEBwZXV0Lm9yZzCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKLH zIjeQUuZ9gSXNlKqy6nvzZqN66/dMBLKfyZR8auevZpbUzuM3i+Uzz/v+xUvzzesf6RY/NVD ghNit9blAXEf4IGq1fQXXGxLUDsBHzFkizyjv+wzFXbE5JdgQleUo/hdVZjyhT5Mofl1Zhn0 En2mQnzjRrKKqoj8/cweqa7u9Lac8O7xhkVr7joS1aQZslOPMV2etTkmZ252nft93Vu02yxE iSrO0M55A5NpnqlL1GqOGY25ZW2Bm5wqGjpz1B/aG0bPwAv3kyq20XY5lgv0CVdmGMlwgw5p kFWO7AujBf+RgvaWjW/JQa3d5nFKRlH8gJ+RDCPUEqkVkYW0zlcCAwEAAaOBzDCByTAJBgNV HRMEAjAAMEQGA1UdIAQ9MDswOQYLYIZIAYb4RQEHFwEwKjAoBggrBgEFBQcCARYcaHR0cHM6 Ly93d3cudmVyaXNpZ24uY29tL3JwYTALBgNVHQ8EBAMCBaAwHQYDVR0lBBYwFAYIKwYBBQUH AwQGCCsGAQUFBwMCMEoGA1UdHwRDMEEwP6A9oDuGOWh0dHA6Ly9JbmRDMURpZ2l0YWxJRC1j cmwudmVyaXNpZ24uY29tL0luZEMxRGlnaXRhbElELmNybDANBgkqhkiG9w0BAQUFAAOCAQEA Fy44PhDAytcmAPzO8JYFGMGqxAtefTJuqstkeVyJzv6Wy0CXTp+Lt8YPf7U2tynhUiaTJS8l AS8VK+QH78y3yQ5VJ2356Jggicq3+UGRLPLSxqt2yHS7sh9yC36eYtlhb5rO+JecPhU0yp5D 1Ahk8dJSFYN7cgDk1JgTiF+rBdflzXceWWqlRxdGZYgYD0w+v1JDFnSODswVSxlWAb1/lOwD fIgSa9SC63JfoZHNlAv7CVbKZHOH9cDuPZCdiT4g+U1AcnrYwceUAjjzLeyxJfBHyF9sfh7S zD7x7O6Hvzk64UwEjeqHGVBfjBPz6/ysoe2Itx12giiFYzQgP4h04TGCBOwwggToAgEBMIHy MIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZl cmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxp ZGF0ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVy IENBIC0gRzICEEJLgZ2m/5im8G2Mp6IuBpMwCQYFKw4DAhoFAKCCAs4wGAYJKoZIhvcNAQkD MQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMTAwMzE0MjAxODA5WjAjBgkqhkiG9w0B CQQxFgQUaNld/bTlFWcmOrFOK/P/9zPqCi8wXwYJKoZIhvcNAQkPMVIwUDALBglghkgBZQME AQIwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIH MA0GCCqGSIb3DQMCAgEoMIIBAwYJKwYBBAGCNxAEMYH1MIHyMIHdMQswCQYDVQQGEwJVUzEX MBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdv cmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20v cnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZl cmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzICEEJLgZ2m/5im 8G2Mp6IuBpMwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTsw OQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAo YykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhBCS4Gdpv+YpvBtjKei LgaTMA0GCSqGSIb3DQEBAQUABIIBABT0Ft2VXxNtegCUF5kbW79duDnAy46XxzMYH7GAcDap QZrvV3QAUWzly7LIljSIJS0Vwx2fe79d5yGDRFCCQ5MHGCxovrQfKx6WDXmAvyfczHqiZHGb PYRorCFZtlzYKhYzHeJUgZC1xHEKHpdLoQP8d+E2dR0h2bZ2WX1GJuk3jM78xtCoxKImWHNU rAevCn3URZYlC4JVhAnMW7FzUFtbgnIaRMxWRZprTZQQvZLUGKdYPiJ+MGTDKGdYvZR4NOSg B/XugQLbE848xllqXKsxUoFnY88iorg1i+saStczVMY3iirpRpRQx5sUuibzQCC35McLBiar yMta+Q9uwwIAAAAAAAA= --------------ms060202010905040304080206-- --===============1176010058== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1176010058==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 22:31:17 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ELV2uI009529; Sun, 14 Mar 2010 22:31:13 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2ELV1L7009480 for ; Sun, 14 Mar 2010 22:31:01 +0100 Received: (qmail 11591 invoked by uid 501); 14 Mar 2010 21:31:08 -0000 Message-ID: <20100314213108.11590.qmail@stuge.se> Date: Sun, 14 Mar 2010 22:31:07 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: so name bump; impact on major, minor and patch version Mail-Followup-To: libssh2-devel@cool.haxx.se References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> <4B9D4481.1040200@peut.org> MIME-Version: 1.0 In-Reply-To: <4B9D4481.1040200@peut.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1671588097==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============1671588097== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="wzJLGUyc3ArbnUjN" Content-Disposition: inline --wzJLGUyc3ArbnUjN Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Jose Baars wrote: > For the foreseeable future I won't be using libtool, it solves > problems that do not exist on VMS, and introduces problems due to > the non-native nature of bash and the gnu tools on VMS, Absolutely no reason to force libtool into the mix then. I'd be fine with adding VMS build files to libssh2 same as there are win32 build files. However, maybe VMS can still use the soname (the 1.1 API version) as an indicator of compatibility? Even though libssh2 changes constantly, the API has only been expanded and not altered for a pretty long time, which is also why we're facing a compatibility break - by now we've gathered enough issues/reasons to actually motivate breaking compatibility. That said the new API will be introduced in a nice way somehow, details under discussion already. > My personal priority is in getting the SSH functionality available > on VMS, and device a sustainable way to keep up with the quick pace > in which libssh2 evolves. I'm now reading=20 http://h71000.www7.hp.com/openvms/journal/v7/faking_it_with_openvms_shareab= le_images.pdf to learn more about shared images. Let's see if we can find something that works well. :) //Peter --wzJLGUyc3ArbnUjN Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQFLnVWbhR3Q0dhIfEgRAuxqAJ4gMWhqvYzrcLmGW7egVZ7H04ldDgCg1+y3 /QOL92/yafNDe8h0BsZCq8g= =hawA -----END PGP SIGNATURE----- --wzJLGUyc3ArbnUjN-- --===============1671588097== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1671588097==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 22:55:20 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ELtFDK028561; Sun, 14 Mar 2010 22:55:18 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.25]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ELtDwZ028556 for ; Sun, 14 Mar 2010 22:55:13 +0100 Received: by ey-out-2122.google.com with SMTP id 9so283308eyd.31 for ; Sun, 14 Mar 2010 14:55:20 -0700 (PDT) Received: by 10.213.42.198 with SMTP id t6mr4412545ebe.38.1268603718920; Sun, 14 Mar 2010 14:55:18 -0700 (PDT) Received: from [192.168.0.35] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 14sm2245489ewy.6.2010.03.14.14.55.16 (version=SSLv3 cipher=RC4-MD5); Sun, 14 Mar 2010 14:55:16 -0700 (PDT) Message-ID: <4B9D5B44.1050102@peut.org> Date: Sun, 14 Mar 2010 22:55:16 +0100 From: Jose Baars User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; nl; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: so name bump; impact on major, minor and patch version References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> <20100314192153.21309.qmail@stuge.se> In-Reply-To: <20100314192153.21309.qmail@stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0808291170==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a cryptographically signed message in MIME format. --===============0808291170== Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms030505090305050006060303" This is a cryptographically signed message in MIME format. --------------ms030505090305050006060303 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Op 3/14/2010 8:21 PM, Peter Stuge schreef: > > However, it will require some education of us unix guys without > strong VMS background. I know next to nothing about VMS so far, but > if we can make changes to libssh2 code or development practice to fit > better into VMS then I think those changes should be considered very > carefully. > =20 Ok, I'll try. ( I'm going for Olympic Geek medal here :-)) see also : http://labs.hoffmanlabs.com/node/163 if you really want know: =20 http://h71000.www7.hp.com/doc/83final/4548/4548pro_011.html#itn_share_ch if you don't get scared quickly, for a really geeky use of shared images = : http://h71000.www7.hp.com/openvms/journal/v7/faking_it_with_openvms_share= able_images.pdf In short, there are 2 features for shared images on VMS that are=20 different to .dll's and .so's: - Normally, when developing applications using a shared image, you link=20 against the shared image on the system. At link time, the jump vectors in the shared=20 image are read and coded into the resulting executable ( the application you wrote). This is fundamentally different on Unix and Windows, where at run=20 time the shared image (excuse my French) after an extensive search is opened, and the =20 application is directed to the proper routine in the shared image based on the name=20 of the routine. This means that on VMS if a new routine is added to the shared image=20 without further ado, the vector table after this new routine has shifted one place, and=20 executables linked against the the old version will bomb, or worse, do something really horrible, =20 if any of these routines is accessed. The standard solution is of course to make sure new routines are=20 added at the end of the vector table, so all old routines keep the same vector.=20 And as a safeguard there is the global section match. ( there is way to use shared images on VMS with a dynamic lookup,=20 lib$find_image_symbol but that is not a common way to offer access to a complete API, although it does exist for=20 Apache apparently.) - To make sure the available shared image on a system and the shared=20 image at link time of an application are compatible, every shared image is linked with=20 a Major version and a Minor version, and a match parameter, usually Less or Equal, meaning the major=20 version of the shared image against which the application has been linked must match, and the Minor=20 version at runtime has to be at least the Minor version at linktime. When an application is started, VMS will open the shared image=20 available, and will immediately bomb out if this global section match as it is called doesn't meet the=20 conditions in the shared image. This is the version that will need to be maintained. It roughly=20 matches the ld version, but needs extra care and attention, and maybe more 'bumping'. Pro's and con's: This sounds all a bit restrictive and complicated, but there are=20 some advantages too: - as a developer, it is possible to guard against mismatches of=20 applications and shared libraries. - dll hell, is an almost non-existent phenomenon on VMS, and if it=20 is, it's a bug introduced by the developer of the shared image, noone else. That is a mixed blessin= g. - there is no such thing as opendl. That problem has already been=20 solved at link time. Anyway, these are design decisions made in the 70's and a discussion=20 on the pro's and con's is pretty useless: it won't change. Plan: So, I will not be following the ld versioning in the global section=20 match id,, but the major ld version and major * 1000000 + minor*1000 + patch of the build version = to start with, as this will probably be most close to what the glocal section id is=20 intended for. It is a number that actually doesn't mean anything, it's just a=20 safeguard against DDL hell. In the actual identification of the image ( the string that describes = the shared library ) I'll use both the ld version and the build version to exactly identify the image. To make sure several versions of libssh2 can exist on a VMS system=20 I'll also make sure the Major ( and if necessary Minor version) are in the filename of shared image = in the binary kit if it is an incompatible version to the last one. To make this transparent the logical name=20 pointing to the image library has to reflect this version as well. I don't think it would be fair for the libssh2 maintainers to take=20 these restrictions into account, as it is so VMS specific. On the other hand, it would be nice to=20 have a routine vs. date-added/deleted/modified cross reference. > The overall goal is that libssh2 should be very portable. That means > we love VMS too. :) > > > //Peter > =20 > =20 LOL! The actual VMS specific changes in the code are very small, I am=20 trying keeping up with the daily builds now, my agenda is now making a VMS PCSI ( binary) kit and automating the=20 process of producing this as automatically as possible. Jose --------------ms030505090305050006060303 Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIPWjCC BMwwggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1 MTAyNzIzNTk1OVowgd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEf MB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNl IGF0IGh0dHBzOi8vd3d3LnZlcmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNv bmEgTm90IFZhbGlkYXRlZDE3MDUGA1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFs IFN1YnNjcmliZXIgQ0EgLSBHMjCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnf rOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyVzm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs +Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zplYu//EHuiVrvFTnAt1qIfPO2wQuhejVch rKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFBL2OyOj++pRpu9MlKWz2VphW7NQIZ +dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5gJ925rXXOL3OVekA6hXVJsLjf aLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUCAwEAAaOCAYQwggGAMBIG A1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXATAqMCgGCCsGAQUF BwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIBBjARBglghkgB hvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJlbDMtMjA0 OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAkoCKG IGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEg UHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+Iqyz cqpVMA0GCSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESf D0b3+qD+0x0Yo9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Pr v4NZmP1m3umGMpqSKTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFQTCCBCmgAwIB AgIQQkuBnab/mKbwbYynoi4GkzANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAV BgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3Jr MTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJp U2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyMB4XDTA5MTExMjAw MDAwMFoXDTEwMTExMjIzNTk1OVowggEKMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0G A1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNv bS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElBQi5MVEQoYyk5ODEeMBwGA1UE CxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdpdGFsIElEIENsYXNzIDEg LSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxEzARBgNVBAMUCkpvc2UgQmFhcnMxHDAaBgkqhkiG 9w0BCQEWDXBldXRAcGV1dC5vcmcwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCi x8yI3kFLmfYElzZSqsup782ajeuv3TASyn8mUfGrnr2aW1M7jN4vlM8/7/sVL883rH+kWPzV Q4ITYrfW5QFxH+CBqtX0F1xsS1A7AR8xZIs8o7/sMxV2xOSXYEJXlKP4XVWY8oU+TKH5dWYZ 9BJ9pkJ840ayiqqI/P3MHqmu7vS2nPDu8YZFa+46EtWkGbJTjzFdnrU5Jmdudp37fd1btNss RIkqztDOeQOTaZ6pS9RqjhmNuWVtgZucKho6c9Qf2htGz8AL95MqttF2OZYL9AlXZhjJcIMO aZBVjuwLowX/kYL2lo1vyUGt3eZxSkZR/ICfkQwj1BKpFZGFtM5XAgMBAAGjgcwwgckwCQYD VR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQt Y3JsLnZlcmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEB ABcuOD4QwMrXJgD8zvCWBRjBqsQLXn0ybqrLZHlcic7+lstAl06fi7fGD3+1Nrcp4VImkyUv JQEvFSvkB+/Mt8kOVSdt+eiYIInKt/lBkSzy0sardsh0u7Ifcgt+nmLZYW+azviXnD4VNMqe Q9QIZPHSUhWDe3IA5NSYE4hfqwXX5c13HllqpUcXRmWIGA9MPr9SQxZ0jg7MFUsZVgG9f5Ts A3yIEmvUgutyX6GRzZQL+wlWymRzh/XA7j2QnYk+IPlNQHJ62MHHlAI48y3ssSXwR8hfbH4e 0sw+8ezuh785OuFMBI3qhxlQX4wT8+v8rKHtiLcddoIohWM0ID+IdOEwggVBMIIEKaADAgEC AhBCS4Gdpv+YpvBtjKeiLgaTMA0GCSqGSIb3DQEBBQUAMIHdMQswCQYDVQQGEwJVUzEXMBUG A1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdvcmsx OzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBh IChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZlcmlT aWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzIwHhcNMDkxMTEyMDAw MDAwWhcNMTAxMTEyMjM1OTU5WjCCAQoxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxMzAxBgNVBAsTKkRpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlIEZ1bGwgU2VydmljZTETMBEGA1UEAxQKSm9zZSBCYWFyczEcMBoGCSqGSIb3 DQEJARYNcGV1dEBwZXV0Lm9yZzCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKLH zIjeQUuZ9gSXNlKqy6nvzZqN66/dMBLKfyZR8auevZpbUzuM3i+Uzz/v+xUvzzesf6RY/NVD ghNit9blAXEf4IGq1fQXXGxLUDsBHzFkizyjv+wzFXbE5JdgQleUo/hdVZjyhT5Mofl1Zhn0 En2mQnzjRrKKqoj8/cweqa7u9Lac8O7xhkVr7joS1aQZslOPMV2etTkmZ252nft93Vu02yxE iSrO0M55A5NpnqlL1GqOGY25ZW2Bm5wqGjpz1B/aG0bPwAv3kyq20XY5lgv0CVdmGMlwgw5p kFWO7AujBf+RgvaWjW/JQa3d5nFKRlH8gJ+RDCPUEqkVkYW0zlcCAwEAAaOBzDCByTAJBgNV HRMEAjAAMEQGA1UdIAQ9MDswOQYLYIZIAYb4RQEHFwEwKjAoBggrBgEFBQcCARYcaHR0cHM6 Ly93d3cudmVyaXNpZ24uY29tL3JwYTALBgNVHQ8EBAMCBaAwHQYDVR0lBBYwFAYIKwYBBQUH AwQGCCsGAQUFBwMCMEoGA1UdHwRDMEEwP6A9oDuGOWh0dHA6Ly9JbmRDMURpZ2l0YWxJRC1j cmwudmVyaXNpZ24uY29tL0luZEMxRGlnaXRhbElELmNybDANBgkqhkiG9w0BAQUFAAOCAQEA Fy44PhDAytcmAPzO8JYFGMGqxAtefTJuqstkeVyJzv6Wy0CXTp+Lt8YPf7U2tynhUiaTJS8l AS8VK+QH78y3yQ5VJ2356Jggicq3+UGRLPLSxqt2yHS7sh9yC36eYtlhb5rO+JecPhU0yp5D 1Ahk8dJSFYN7cgDk1JgTiF+rBdflzXceWWqlRxdGZYgYD0w+v1JDFnSODswVSxlWAb1/lOwD fIgSa9SC63JfoZHNlAv7CVbKZHOH9cDuPZCdiT4g+U1AcnrYwceUAjjzLeyxJfBHyF9sfh7S zD7x7O6Hvzk64UwEjeqHGVBfjBPz6/ysoe2Itx12giiFYzQgP4h04TGCBOwwggToAgEBMIHy MIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZl cmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxp ZGF0ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVy IENBIC0gRzICEEJLgZ2m/5im8G2Mp6IuBpMwCQYFKw4DAhoFAKCCAs4wGAYJKoZIhvcNAQkD MQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMTAwMzE0MjE1NTE2WjAjBgkqhkiG9w0B CQQxFgQUxjnykTJf/36uC5OuMgRvjo8GEPYwXwYJKoZIhvcNAQkPMVIwUDALBglghkgBZQME AQIwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIH MA0GCCqGSIb3DQMCAgEoMIIBAwYJKwYBBAGCNxAEMYH1MIHyMIHdMQswCQYDVQQGEwJVUzEX MBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdv cmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20v cnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZl cmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzICEEJLgZ2m/5im 8G2Mp6IuBpMwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTsw OQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAo YykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhBCS4Gdpv+YpvBtjKei LgaTMA0GCSqGSIb3DQEBAQUABIIBAEU8pJKXT/Z1dx2QjhqF+X/9WXrk6dr/PEGX9Z7jU62B CfzVSP0CmZAJxWIttjMDv1Cn0z4N52jmC9VzqNyBLv2eNqbAkDVgHi/UQmdesoxnDacf567v wjwOH1vjBhN0ChmJm/cZCQxGUfyPHEuQHSVfTTXx0g72z8EACo53fr8QJOqOBDP4dTTYz9yf NQV2/9njZ4RNLsvvxDx8hDjB8NZ/CJn53mMqJVOrFG2dufoPsgKgP9XEyNrmGUysSCL9xA3R qcJ/z9tOUKI0tbE78s/fIBrS04+zW4KU5Ngw8cWBpvOk7C/dLB4PFlLM0fTj6aodqanIOlOo /TT9oSU/xhwAAAAAAAA= --------------ms030505090305050006060303-- --===============0808291170== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0808291170==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 23:18:02 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMHtPE012094; Sun, 14 Mar 2010 23:17:59 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMHsKt012091 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 14 Mar 2010 23:17:54 +0100 Received: (qmail 8240 invoked from network); 14 Mar 2010 22:18:01 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 14 Mar 2010 22:18:01 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Sun, 14 Mar 2010 22:18:01 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #164: Knownhost API handles comments badly X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/164#comment:7 Message-ID: <054.0a6b3ebab90c9b99df6ad317a58809e3@libssh2.stuge.se> References: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-Trac-Ticket-ID: 164 In-Reply-To: <045.09358f07c267780e8128767eda710f67@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #164: Knownhost API handles comments badly ------------------------+--------------------------------------------------- Reporter: alamaison | Owner: Type: defect | Status: closed Priority: normal | Milestone: 1.2.5 Component: API | Version: 1.2.4 Resolution: fixed | Keywords: knownhost Blocking: | Blockedby: ------------------------+--------------------------------------------------- Changes (by bagder): * status: new => closed * resolution: => fixed Comment: Thanks, I'm setting this entry to "resolve as fixed" then! -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 23:36:48 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMaehE021384; Sun, 14 Mar 2010 23:36:47 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMacEH021368 for ; Sun, 14 Mar 2010 23:36:38 +0100 Date: Sun, 14 Mar 2010 23:36:38 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Greetings and a quick note about licening issuse in the Makefile.win32 in 1.2.4 In-Reply-To: Message-ID: References: User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Sun, 14 Mar 2010, C Johnson wrote: > Greetings all. I'm C Johnson. I have recently gotten involved with libssh2 > in the Win32 arena - specifically porting the headers to Delphi. > > I'm not sure if the Libssh2 group itself will want to include the translated > headers as part of the official distro, but once I'm done porting all the > demos apps and finished my testing, I would definitely be willing to make > them available - increasing the reach of libssh2 even further. Surely that qualifies as a "library binding" ? I don't think adding bindings into the core libssh2 source tree is necessary or even a very good idea. I rather think bindings are rather suitable for being separate from the core libssh2 project. > I did, however, find one problem with the win32\Makefile.win32 - it lists > the license as GPL, and the website lists the license as modified BSD. Oh right, I've pushed a fix for this now. > While I am on the topic of licensing, I found it to be very difficult to > track down clear information about the license in either the source tarball > or the website itself. In the tarballs it is mentioned in the README and the full license is in the COPYING file, both in the root dir. The info is indeed missing on the site. I'll add a prominent link somewhere on the front page! > Again, greetings to all, and I hope to become a more active member of the > community as time grows. Nice, you're most welcome! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 23:48:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMmY5P026581; Sun, 14 Mar 2010 23:48:38 +0100 Received: from dottedmag.net (burger.dottedmag.net [212.75.37.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMmX6j026557 for ; Sun, 14 Mar 2010 23:48:34 +0100 Received: from vertex.dottedmag (unknown [91.197.127.125]) by dottedmag.net (Postfix) with ESMTPSA id A49F18C758 for ; Sun, 14 Mar 2010 23:48:40 +0100 (CET) Received: from dottedmag by vertex.dottedmag with local (Exim 4.69) (envelope-from ) id 1Nqwbu-0000v6-AP for libssh2-devel@cool.haxx.se; Mon, 15 Mar 2010 04:48:38 +0600 From: Mikhail Gusarov To: libssh2 development Subject: Re: so name bump; impact on major, minor and patch version References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> <20100314192153.21309.qmail@stuge.se> <4B9D5B44.1050102@peut.org> Date: Mon, 15 Mar 2010 04:48:37 +0600 In-Reply-To: <4B9D5B44.1050102@peut.org> (Jose Baars's message of "Sun, 14 Mar 2010 22:55:16 +0100") Message-ID: <87pr36ecze.fsf@vertex.dottedmag> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0125512190==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0125512190== Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --=-=-= Content-Transfer-Encoding: quoted-printable Twas brillig at 22:55:16 14.03.2010 UTC+01 when peut@peut.org did gyre and = gimble: [] Sounds really like sonames (semantically, not implementation-wide). patchlevel increase: no ABI changes, no problem to upgrade library in-place. minor version increase: conservative ABI extension, not a problem for VMS if new entries are added to the end of list (which is sane thing to do, I presume). New version will serve all applications linked with with library with the same major version just fine. major version increase: ABI breakage. Another version of library should be installed alongside the older one. =2D-=20 http://fossarchy.blogspot.com/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iQIcBAEBAgAGBQJLnWfFAAoJEJ0g9lA+M4iIW3oQAI2FH8yPoL19fQHjIFhIxx+b JxAMmmVdTY6NtlrWppFwR3nD3gpApKxnng9FijUd3XeK6Xe5HTGik4YPlBl+qbDc u/PX4QykLTDWSJRq2xXN6dQ39reBm12mAJNWcZNJeKgnr7cHsTC1kbsYC3Sh22eg 6MVFPJ0eFNMKdy4jIQ4cf9vWvxe8xgZdQ6UqyTJ8V3LUZCqi0grJkMWc9XQtgnY2 3jHbDTdCZhC38Jwmm4QggH6qT8tZNkZrYxOXhwh1AjldUNwI0klg7MsYlqJwWWMU scL9k59ll0IYLgSAt629Gcxi+LSEe/Oqlz76nVjEx2HJnNMZcNcZiWPKe1O/nFMG 6MMMKLyLhIcIvse3Et3CR0nQAmyqIqFYe02/LsmpX7e1DYdm6/tsJJ9smr2gJlYY VyjDgHCK2oC8OeX2OsrJKt9vGoHgIs9nxxKyMpiRlk+6XQS/2BWMaHWHBrG3+YNZ Qp3nVLkwsMGjSkPUh/sO3Yr2nf3DNqOFD54wbqjy1605rxbKF3u1qYRatU0fYhXV mJRfdkuKU/FGbuPuyAmFpFsMKv8t4Te7Vhc9Oass0krwjs0sbB8jrYv9ozd1ir37 r7Q44D4QghlZ+MDhJAQmF2CfQdBS5rTknva8nBoIT6hn5Q1KtMbXikRSVMQHLLQt F+yMNtzPDdvBxsiuXKkP =eDtS -----END PGP SIGNATURE----- --=-=-=-- --===============0125512190== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0125512190==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 14 23:55:24 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2EMtKp2029430; Sun, 14 Mar 2010 23:55:22 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2EMtINQ029427 for ; Sun, 14 Mar 2010 23:55:18 +0100 Received: (qmail 23412 invoked by uid 501); 14 Mar 2010 22:55:26 -0000 Message-ID: <20100314225526.23411.qmail@stuge.se> Date: Sun, 14 Mar 2010 23:55:26 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Greetings and a quick note about licening issuse in the Makefile.win32 in 1.2.4 Mail-Followup-To: libssh2-devel@cool.haxx.se References: MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg wrote: >> I'm not sure if the Libssh2 group itself will want to include the >> translated headers as part of the official distro, but once I'm >> done porting all the demos apps and finished my testing, I would >> definitely be willing to make them available - increasing the >> reach of libssh2 even further. > > Surely that qualifies as a "library binding" ? I'm not sure. C Johnson, could you please explain a little bit more about those translated headers and demos? An example would be great! >> Again, greetings to all, and I hope to become a more active member >> of the community as time grows. > > Nice, you're most welcome! Yes indeed. As far as bringing up issues there are several approaches; inquire on mailing list (as you did), inquire on IRC channel (less formal and highly variable response time, from immediate to days) and finally there's the option to create a new ticket in Trac at http://libssh2.stuge.se/ Which option fits best depends a bit on the actual issue, and sometimes the true nature of the issue is only discovered after some investigation, so if something comes up in one forum that would actually probably be better dealt with in one of the others then we usually hint about that. Trac also posts to the mailing list, so that's a two-in-one. :) Trac tickets are especially good for things that may take a little longer to investigate or fix. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 00:42:59 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ENglhF000369; Mon, 15 Mar 2010 00:42:55 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.26]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ENgjVn000326 for ; Mon, 15 Mar 2010 00:42:45 +0100 Received: by ey-out-2122.google.com with SMTP id 9so288101eyd.31 for ; Sun, 14 Mar 2010 16:42:53 -0700 (PDT) Received: by 10.213.97.4 with SMTP id j4mr4433715ebn.8.1268610172744; Sun, 14 Mar 2010 16:42:52 -0700 (PDT) Received: from [192.168.0.34] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 14sm2289253ewy.2.2010.03.14.16.42.51 (version=TLSv1/SSLv3 cipher=RC4-MD5); Sun, 14 Mar 2010 16:42:52 -0700 (PDT) From: peut@peut.org To: libssh2 development Subject: Re: so name bump; impact on major, minor and patch version Date: Mon, 15 Mar 2010 00:42:14 +0100 Message-ID: X-Mailer: EPOC Email Version 2.10 MIME-Version: 1.0 Content-Language: i-default X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: peut@peut.org, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Yes it's similar, but on VMS the minor version of the shared image must be bumped as well in case of an addition, which is why I cannot use the ld version. Jose -origineel bericht- Onderwerp: Re: so name bump; impact on major, minor and patch version Van: Mikhail Gusarov Datum: 14-03-2010 23:48 Twas brillig at 22:55:16 14.03.2010 UTC+01 when peut@peut.org did gyre and gimble: [] Sounds really like sonames (semantically, not implementation-wide). patchlevel increase: no ABI changes, no problem to upgrade library in-place. minor version increase: conservative ABI extension, not a problem for VMS if new entries are added to the end of list (which is sane thing to do, I presume). New version will serve all applications linked with with library with the same major version just fine. major version increase: ABI breakage. Another version of library should be installed alongside the older one. -- http://fossarchy.blogspot.com/ _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel <> _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 00:49:30 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ENnPml006636; Mon, 15 Mar 2010 00:49:28 +0100 Received: from Encompasserve.org (Eisner.encompasserve.org [67.53.90.116]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ENnNU6006632 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 00:49:23 +0100 Received: from CONVERSION-DAEMON.Encompasserve.org by Encompasserve.org (PMDF V6.3-x13 #31732) id <01NKQZUXEP6O000AQP@Encompasserve.org> for libssh2-devel@cool.haxx.se; Sun, 14 Mar 2010 18:49:28 -0500 (CDT) Received: from [192.168.0.31] (173-23-189-28.client.mchsi.com [173.23.189.28]) by Encompasserve.org (PMDF V6.3-x13 #31732) with ESMTPSA id <01NKQZUWXBS8000A4Z@Encompasserve.org> for libssh2-devel@cool.haxx.se; Sun, 14 Mar 2010 18:49:27 -0500 (CDT) Date: Sun, 14 Mar 2010 18:49:26 -0500 From: "John E. Malmberg" Subject: Re: Greetings and a quick note about licening issuse in the Makefile.win32 in 1.2.4 In-reply-to: To: libssh2 development Message-id: <4B9D7606.9080008@qsl.net> MIME-version: 1.0 User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) References: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se As a side note, I was looking at the home page for libssh2 to determine what license it is under and could only find a reference in the comparison between libssh2 and libssh. Could the license description be made more prominent? Thanks, -John _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 01:09:35 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2F09UkS026873; Mon, 15 Mar 2010 01:09:33 +0100 Received: from Encompasserve.org (Eisner.encompasserve.org [67.53.90.116]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2F09S1P026868 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 01:09:29 +0100 Received: from CONVERSION-DAEMON.Encompasserve.org by Encompasserve.org (PMDF V6.3-x13 #31732) id <01NKR0KUPJ5C0006E4@Encompasserve.org> for libssh2-devel@cool.haxx.se; Sun, 14 Mar 2010 19:09:35 -0500 (CDT) Received: from [192.168.0.31] (173-23-189-28.client.mchsi.com [173.23.189.28]) by Encompasserve.org (PMDF V6.3-x13 #31732) with ESMTPSA id <01NKR0KU7EFY000A4Z@Encompasserve.org> for libssh2-devel@cool.haxx.se; Sun, 14 Mar 2010 19:09:35 -0500 (CDT) Date: Sun, 14 Mar 2010 19:09:33 -0500 From: "John E. Malmberg" Subject: Re: so name bump; impact on major, minor and patch version In-reply-to: <4B9D5B44.1050102@peut.org> To: libssh2 development Message-id: <4B9D7ABD.4040208@qsl.net> MIME-version: 1.0 User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> <20100314192153.21309.qmail@stuge.se> <4B9D5B44.1050102@peut.org> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Jose Baars wrote: > Op 3/14/2010 8:21 PM, Peter Stuge schreef: > > In short, there are 2 features for shared images on VMS that are > different to .dll's and .so's: > - Normally, when developing applications using a shared image, you link > against the shared image on the system. At link time, the jump vectors > in the shared image are read and coded into the resulting executable > ( the application you wrote). An additional restriction is that the once the order of the jump vectors are established they must not change. Libtool is not being used to build shared images with curl, it is just being used to compile the module. For libtool to be useful for building VMS shared images, there has to be a way for it to know what the previous list contained and order it. > This is fundamentally different on Unix and Windows, where at run time > the shared image(excuse my French) after an extensive search is opened, > and the application is directed to the proper routine in the shared image > based on the name of the routine. Actually Windows DLLs have the same restriction as VMS for full compatibility that the order of the symbols never changes for future versions. This is because Windows DLLs support calling by ordinal, which is the routine number. The last time I corresponded with the libtool maintainers, it could not handle this. Neither they or myself know of any examples of this actually causing a problem. Right now, linking a VMS shared image is so foreign to the existing builds for any other platform it is best to leave it as a set of VMS specific scripts. For the Curl package that I produced, all the binaries in the kit were linked using VMS specific command procedures. -John wb8tyw@qsl.net Personal Opinion Only _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 09:27:48 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2F8RaNQ030022; Mon, 15 Mar 2010 09:27:45 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2F8RYPO030015 for ; Mon, 15 Mar 2010 09:27:34 +0100 Date: Mon, 15 Mar 2010 09:27:34 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Greetings and a quick note about licening issuse in the Makefile.win32 in 1.2.4 In-Reply-To: <4B9D7606.9080008@qsl.net> Message-ID: References: <4B9D7606.9080008@qsl.net> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Sun, 14 Mar 2010, John E. Malmberg wrote: > As a side note, I was looking at the home page for libssh2 to determine what > license it is under and could only find a reference in the comparison > between libssh2 and libssh. > > Could the license description be made more prominent? I added a link under Documentation on the front page that links to a page that shows the COPYING file in a web page: http://www.libssh2.org/license.html Good enough? One of these days I hope we'll get the git repo for the web site working and then more people should be able to help out with the web content edits. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 09:52:26 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2F8qK6r014015; Mon, 15 Mar 2010 09:52:23 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2F8qIqT013993 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 09:52:18 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2F8q6IV022380 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Mon, 15 Mar 2010 09:52:08 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: so name bump; impact on major, minor and patch version References: <50cb0cb61003111051o64e0d891w85fe164c30acdb93@mail.gmail.com> <87vdd2hc33.fsf@mocca.josefsson.org> <4B9C6182.7090107@peut.org> <4B9D4481.1040200@peut.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100315:libssh2-devel@cool.haxx.se::yCY3XhwmOJePUR/R:0Ybk Date: Mon, 15 Mar 2010 09:52:07 +0100 In-Reply-To: <4B9D4481.1040200@peut.org> (Jose Baars's message of "Sun, 14 Mar 2010 21:18:09 +0100") Message-ID: <87ljdu55mw.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Jose Baars writes: > For the foreseeable future I won't be using libtool, it solves > problems that do not exist on VMS, and introduces problems due to the > non-native nature of bash and the gnu tools on VMS, although John > Malmberg has been pretty successful in getting it to work for curl and > zlib amongst others. Just a general point. Libtool doesn't solve a problem on any particular platform, instead, it's point is to allow maintainers to avoid having to think about platform specific peculiarities. However if you maintain the VMS port for libssh2, and don't want to use libtool, that's fine, but then you need to stay on top of API/ABI changes etc manually so the same problem doesn't hit VMS users. I think a VMS directory, like the win32 directory, is fine. But it requires someone to maintain it, but with your generous work I think we have that covered! /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 14:07:20 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FD737t018053; Mon, 15 Mar 2010 14:07:15 +0100 Received: from Encompasserve.org (Eisner.encompasserve.org [67.53.90.116]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FD70Dr017988 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 14:07:02 +0100 Received: from CONVERSION-DAEMON.Encompasserve.org by Encompasserve.org (PMDF V6.3-x13 #31732) id <01NKRRPSWA9S000CJL@Encompasserve.org> for libssh2-devel@cool.haxx.se; Mon, 15 Mar 2010 08:07:03 -0500 (CDT) Received: from [10.37.80.130] (204-11-77-50.ip.ctc.biz [204.11.77.50]) by Encompasserve.org (PMDF V6.3-x13 #31732) with ESMTPSA id <01NKRRPSNPCK000A4Z@Encompasserve.org> for libssh2-devel@cool.haxx.se; Mon, 15 Mar 2010 08:07:03 -0500 (CDT) Date: Mon, 15 Mar 2010 08:07:03 -0500 From: "John E. Malmberg" Subject: Re: Greetings and a quick note about licening issuse in the Makefile.win32 in 1.2.4 In-reply-to: To: libssh2 development Message-id: <4B9E30F7.5040404@qsl.net> MIME-version: 1.0 User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) References: <4B9D7606.9080008@qsl.net> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg wrote: > > I added a link under Documentation on the front page that links to a > page that shows the COPYING file in a web page: > http://www.libssh2.org/license.html > > Good enough? Looks good. Thanks, -John _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 17:20:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGJox6030731; Mon, 15 Mar 2010 17:20:01 +0100 Received: from hrndva-omtalb.mail.rr.com (hrndva-omtalb.mail.rr.com [71.74.56.124]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGJmmf030723 for ; Mon, 15 Mar 2010 17:19:48 +0100 X-Authority-Analysis: v=1.0 c=1 a=O-pu0537hIgA:10 a=UyuVG8ZPJOt52AkBbC8A:9 a=aonG8QLQ8yGmJ8wqXDZhHZOPcDEA:4 a=CjuIK1q_8ugA:10 a=SSmOFEACAAAA:8 a=Y2VNeNrzAAAA:8 a=yMhMjlubAAAA:8 a=TW66zc2HAAAA:8 a=HQ31llbKAAAA:8 a=NT_Dqa3x-nndH1USW_YA:9 a=6mw9lRLjRpbyKIsNajsA:7 a=i-J0ZMjFNmyK1a2qevuRmBJyj_sA:4 X-Cloudmark-Score: 0 X-Originating-IP: 70.116.71.237 Received: from [70.116.71.237] ([70.116.71.237:37268] helo=dc7100.roadrunn.com) by hrndva-oedge03.mail.rr.com (envelope-from ) (ecelerity 2.2.2.39 r()) with ESMTP id D7/A5-01194-42E5E9B4; Mon, 15 Mar 2010 16:19:48 +0000 Received: by dc7100.roadrunn.com (Postfix, from userid 501) id 6B909830A2B; Mon, 15 Mar 2010 11:19:15 -0500 (CDT) X-Spam-Checker-Version: SpamAssassin 3.2.4 (2008-01-01) on dc7100.roadrunn.com X-Spam-Level: X-Spam-Status: No, score=-4.3 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, HTML_MESSAGE autolearn=ham version=3.2.4 Received: from ngierman3 (dc7100 [192.168.1.6]) by dc7100.roadrunn.com (Postfix) with ESMTP id 5F2CF830A2B for ; Mon, 15 Mar 2010 11:19:11 -0500 (CDT) From: "Neil Gierman" To: "'libssh2 development'" Subject: size_t conversion Date: Mon, 15 Mar 2010 11:19:43 -0500 Message-ID: <000001cac45b$527a8670$f76f9350$@com> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: AcrEW1DzoLAmOJwVRNOJIeiJPwt+Rg== Content-Language: en-us x-cr-hashedpuzzle: Czn+ C75B DrcY D9Qp EQCw ErcY EzJe E/6J FFLK FdFX GPl3 GksQ IAzM IRK3 IyNN I7va; 1; bABpAGIAcwBzAGgAMgAtAGQAZQB2AGUAbABAAGMAbwBvAGwALgBoAGEAeAB4AC4AcwBlAA==; Sosha1_v1; 7; {DA245768-3DBA-442F-B59F-5A846250EE4D}; bgBnAGkAZQByAG0AYQBuAEAAcgBvAGEAZAByAHUAbgBuAC4AYwBvAG0A; Mon, 15 Mar 2010 16:19:41 GMT; cwBpAHoAZQBfAHQAIABjAG8AbgB2AGUAcgBzAGkAbwBuAA== x-cr-puzzleid: {DA245768-3DBA-442F-B59F-5A846250EE4D} X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0104100979==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --===============0104100979== Content-Type: multipart/alternative; boundary="----=_NextPart_000_0001_01CAC431.69A47E70" Content-Language: en-us This is a multi-part message in MIME format. ------=_NextPart_000_0001_01CAC431.69A47E70 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit I have an initial patch and some questions about the conversion from unsigned long to size to (and also found int to ssize_t). Would it be better to discuss it here, or open a Trac ticket? ------=_NextPart_000_0001_01CAC431.69A47E70 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

I have an initial patch and some questions about = the conversion from unsigned long to size to (and also found int to ssize_t). Would it = be better to discuss it here, or open a Trac ticket?

------=_NextPart_000_0001_01CAC431.69A47E70-- --===============0104100979== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0104100979==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 17:24:27 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGOO7H032767; Mon, 15 Mar 2010 17:24:25 +0100 Received: from mail-ww0-f54.google.com (mail-ww0-f54.google.com [74.125.82.54]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGOLAo032728 for ; Mon, 15 Mar 2010 17:24:22 +0100 Received: by wwb18 with SMTP id 18so167361wwb.41 for ; Mon, 15 Mar 2010 09:24:22 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=uHDjAPWyMwrgC/1hA66L5BT1xdqR7vmB0rdAFMdRbFA=; b=C3jOiMB6fLfrdGW7ijIMvGHLsq4onJi5d0uUViPxLPiOtfinruATGtH49cSL1aSOui hEV993qAzk6a4DWKJ/5ZEvOwOAOXn6UP2Av1R0kR1xmf2Bvbil00yzDJeMcFmfectVns YxewityFPliecFiBqGhE+64y/jekPLEuqME9s= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=o0835PHUKlbMDlL+dXVBN0jWewDNI1sF0NC99fSDGRigbKz4mwYew1U1mPF50oFok5 q33TvlhqL/rIrLzmKRR6My+yK4QAoP9dw198wVSZ1ROOwzZBmv72k8uDii9KvAwhCesg CwWEKKFa5GGOTCIncbZ50zIeIHcYPzmapSsE0= MIME-Version: 1.0 Received: by 10.216.168.9 with SMTP id j9mr2219477wel.152.1268670262246; Mon, 15 Mar 2010 09:24:22 -0700 (PDT) In-Reply-To: <000001cac45b$527a8670$f76f9350$@com> References: <000001cac45b$527a8670$f76f9350$@com> Date: Mon, 15 Mar 2010 16:24:22 +0000 X-Google-Sender-Auth: ada17f3bcc8886bb Message-ID: <9c3275a91003150924w2bf94a8u93e0184b6a099b77@mail.gmail.com> Subject: Re: size_t conversion From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 15 March 2010 16:19, Neil Gierman wrote: > > I have an initial patch and some questions about the conversion from unsigned long to size > to (and also found int to ssize_t). Would it be better to discuss it here, or open a Trac ticket? I'd favour a Trac ticket: makes it harder to forget patches. Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 17:31:22 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGVFhY003162; Mon, 15 Mar 2010 17:31:20 +0100 Received: from idcmail-mo2no.shaw.ca (idcmail-mo2no.shaw.ca [64.59.134.9]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGVCvl003115 for ; Mon, 15 Mar 2010 17:31:13 +0100 Received: from pd7ml1no-ssvc.prod.shaw.ca ([10.0.153.161]) by pd6mo1no-svcs.prod.shaw.ca with ESMTP; 15 Mar 2010 10:31:12 -0600 X-Cloudmark-SP-Filtered: true X-Cloudmark-SP-Result: v=1.0 c=1 a=Gd3-4gBQwxUA:10 a=VphdPIyG4kEA:10 a=_4SF6Xvc910A:10 a=kj9zAlcOel0A:10 a=y9ASP8Sd+Sa59kG1URKZZw==:17 a=oeppcwj8AAAA:8 a=mjb21GjqPjas3ZnBzmQA:9 a=JeDYPxmh82bbBFjfx---cHtyT8MA:4 a=CjuIK1q_8ugA:10 a=ie-UTDoLF60A:10 Received: from unknown (HELO mail.xepol.com) ([68.146.216.64]) by pd7ml1no-dmz.prod.shaw.ca with ESMTP; 15 Mar 2010 10:31:12 -0600 Received: from quadx64 [192.168.1.3] by mail.xepol.com with SMTP (EHLO Quadx64) (ArGoSoft Mail Server Pro for WinNT/2000/XP, Version 1.8 (1.8.9.2)); Mon, 15 Mar 2010 10:30:54 -0600 From: "C Johnson" To: "'libssh2 development'" References: <4B9D7606.9080008@qsl.net> In-Reply-To: Subject: RE: Greetings and a quick note about licening issuse in theMakefile.win32 in 1.2.4 Date: Mon, 15 Mar 2010 10:30:53 -0600 Message-ID: <8237C98D37F14F489B55D2C46F06F492@Quadx64> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 11 Thread-Index: AcrEGcmo4MULN023TK2g3cHWAGTsdAAQq1ow X-MimeOLE: Produced By Microsoft MimeOLE V6.1.7600.16385 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se > I added a link under Documentation on the front page that > links to a page that shows the COPYING file in a web page: > http://www.libssh2.org/license.html > > Good enough? Looks good! Easy to find, prominently placed. - C Johnson _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 17:38:24 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGcIsW005637; Mon, 15 Mar 2010 17:38:22 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGcG4R005633 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 17:38:16 +0100 Received: (qmail 22676 invoked from network); 15 Mar 2010 16:38:21 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 15 Mar 2010 16:38:21 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 15 Mar 2010 16:38:21 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #165: size_t conversion X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/165 Message-ID: <044.e1ec2089c65b3d0192e00f84b5cdb6b5@libssh2.stuge.se> X-Trac-Ticket-ID: 165 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #165: size_t conversion -------------------------+-------------------------------------------------- Reporter: roadrunn | Owner: Type: enhancement | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | -------------------------+-------------------------------------------------- Attached is the initial patch against 1.2.4 for converting unsigned long to size_t. Many functions had to be touched since we bubble up returns from lower functions. I am down from 446 warnings before, to 7 warnings now. The remaining warnings are below that are not easy fixes. 1>..\src\misc.c(83) : warning C4267: 'function' : conversion from 'size_t' to 'int', possible loss of data 1>..\src\misc.c(102) : warning C4267: 'function' : conversion from 'size_t' to 'int', possible loss of data 1>..\src\packet.c(1131) : warning C4244: 'function' : conversion from 'SSIZE_T' to 'unsigned char', possible loss of data 1>..\src\scp.c(689) : warning C4244: '=' : conversion from 'long' to 'unsigned short', possible loss of data 1>..\src\session.c(541) : warning C4244: 'function' : conversion from 'libssh2_socket_t' to 'int', possible loss of data 1>..\src\session.c(569) : warning C4244: 'function' : conversion from 'libssh2_socket_t' to 'int', possible loss of data 1>..\src\session.c(1557) : warning C4244: 'function' : conversion from 'libssh2_socket_t' to 'int', possible loss of data -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 17:40:25 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGeKKS007032; Mon, 15 Mar 2010 17:40:22 +0100 Received: from hrndva-omtalb.mail.rr.com (hrndva-omtalb.mail.rr.com [71.74.56.125]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FGeIdP006889 for ; Mon, 15 Mar 2010 17:40:19 +0100 X-Authority-Analysis: v=1.0 c=1 a=O-pu0537hIgA:10 a=kj9zAlcOel0A:10 a=lLgI09EQqg17OLhugrgA:9 a=k-4V2JQB9BFtTHDRR53MttPoCUsA:4 a=CjuIK1q_8ugA:10 X-Cloudmark-Score: 0 X-Originating-IP: 70.116.71.237 Received: from [70.116.71.237] ([70.116.71.237:37284] helo=dc7100.roadrunn.com) by hrndva-oedge03.mail.rr.com (envelope-from ) (ecelerity 2.2.2.39 r()) with ESMTP id DE/1E-01194-3F26E9B4; Mon, 15 Mar 2010 16:40:19 +0000 Received: by dc7100.roadrunn.com (Postfix, from userid 501) id 72F40830A2B; Mon, 15 Mar 2010 11:39:46 -0500 (CDT) X-Spam-Checker-Version: SpamAssassin 3.2.4 (2008-01-01) on dc7100.roadrunn.com X-Spam-Level: X-Spam-Status: No, score=-4.3 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.2.4 Received: from ngierman3 (dc7100 [192.168.1.6]) by dc7100.roadrunn.com (Postfix) with ESMTP id 5714D830A2B for ; Mon, 15 Mar 2010 11:39:42 -0500 (CDT) From: "Neil Gierman" To: "'libssh2 development'" References: <000001cac45b$527a8670$f76f9350$@com> <9c3275a91003150924w2bf94a8u93e0184b6a099b77@mail.gmail.com> In-Reply-To: <9c3275a91003150924w2bf94a8u93e0184b6a099b77@mail.gmail.com> Subject: RE: size_t conversion Date: Mon, 15 Mar 2010 11:40:14 -0500 Message-ID: <001401cac45e$301f0670$905d1350$@com> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: AcrEXMjlpldZq7J5RtCOs8WrcWuNMwAAVmiw Content-Language: en-us X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se > > I have an initial patch and some questions about the conversion from > unsigned long to size > > to (and also found int to ssize_t). Would it be better to discuss it > here, or open a Trac ticket? > > I'd favour a Trac ticket: makes it harder to forget patches. > > Alex Submitted. _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 18:08:33 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FH8AKL000424; Mon, 15 Mar 2010 18:08:15 +0100 Received: from idcmail-mo2no.shaw.ca (idcmail-mo2no.shaw.ca [64.59.134.9]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FH88AA000331 for ; Mon, 15 Mar 2010 18:08:08 +0100 Received: from pd7ml2no-ssvc.prod.shaw.ca ([10.0.153.162]) by pd7mo1no-svcs.prod.shaw.ca with ESMTP; 15 Mar 2010 11:08:08 -0600 X-Cloudmark-SP-Filtered: true X-Cloudmark-SP-Result: v=1.0 c=1 a=Gd3-4gBQwxUA:10 a=VphdPIyG4kEA:10 a=_4SF6Xvc910A:10 a=kj9zAlcOel0A:10 a=y9ASP8Sd+Sa59kG1URKZZw==:17 a=hEddX2fDoi5qWcdJyPUA:9 a=hQhPJfU2Hso3TdsBO_cA:7 a=fX_nn3l5wfjErAwh7JlIXr9TTJIA:4 a=CjuIK1q_8ugA:10 Received: from unknown (HELO mail.xepol.com) ([68.146.216.64]) by pd7ml2no-dmz.prod.shaw.ca with ESMTP; 15 Mar 2010 11:08:08 -0600 Received: from quadx64 [192.168.1.3] by mail.xepol.com with SMTP (EHLO Quadx64) (ArGoSoft Mail Server Pro for WinNT/2000/XP, Version 1.8 (1.8.9.2)); Mon, 15 Mar 2010 11:07:50 -0600 From: "C Johnson" To: "'libssh2 development'" References: <20100314225526.23411.qmail@stuge.se> In-Reply-To: <20100314225526.23411.qmail@stuge.se> Subject: RE: Greetings and a quick note about licening issuse in theMakefile.win32 in 1.2.4 Date: Mon, 15 Mar 2010 11:07:49 -0600 Message-ID: MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 11 Thread-Index: AcrDyZBI+xaLrIL/QkCBaCF7lFaEsgAk7egw X-MimeOLE: Produced By Microsoft MimeOLE V6.1.7600.16385 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se > [mailto:libssh2-devel-bounces@cool.haxx.se] On Behalf Of Peter Stuge > > I'm not sure. C Johnson, could you please explain a little > bit more about those translated headers and demos? An example > would be great! If I understand correctly, libssh2 is meant to be used as a shared library instead of linked directly into projects. This is definitely how it would be accessed from Delphi. However, because Delphi is Object Pascal it can not use any of the current files to build the linkage to the shared library. Instead, the header files have to be translated. Amoung other problems, Delphi (and pascal in general) doesn't have a pre-processor so many of the #define wedges need to be recoded as full functions/procedures. Simple conversions can be: #define LIBSSH2_VERSION_MAJOR 1 #define LIBSSH2_VERSION_MINOR 2 #define LIBSSH2_VERSION_PATCH 5 To Const LIBSSH2_VERSION_MAJOR = 1; LIBSSH2_VERSION_MINOR = 2; LIBSSH2_VERSION_PATCH = 5; And more complicated conversions such as #define LIBSSH2_ALLOC_FUNC(name) void *name(size_t count, void **abstract) #define LIBSSH2_REALLOC_FUNC(name) void *name(void *ptr, size_t count, \ void **abstract) #define LIBSSH2_FREE_FUNC(name) void name(void *ptr, void **abstract) Which are function call types convert to Type LIBSSH2_ALLOC_FUNC = Procedure(count : size_t; Var abstract : Pointer ); cdecl; LIBSSH2_REALLOC_FUNC = Procedure(ptr : Pointer; count : size_t; Var abstract : Pointer); cdecl; LIBSSH2_FREE_FUNC = Procedure (ptr : Pointer; Var abstract : Pointer); cdecl; And actual library calls such as libssh2_session_init_ex(LIBSSH2_ALLOC_FUNC((*my_alloc)), LIBSSH2_FREE_FUNC((*my_free)), LIBSSH2_REALLOC_FUNC((*my_realloc)), void *abstract); Convert to Function libssh2_session_init_ex(my_alloc : LIBSSH2_ALLOC_FUNC; my_free : LIBSSH2_FREE_FUNC; my_realloc : LIBSSH2_REALLOC_FUNC; abstract : Pointer) : PLIBSSH2_SESSION; cdecl; external 'libssh2.dll' name 'libssh2_session_init_ex' And helper macros like #define libssh2_session_init() libssh2_session_init_ex(NULL, NULL, NULL, NULL) Become Function libssh2_session_init : PLIBSSH2_SESSION; Begin Result := libssh2_session_init_ex(nil, nil, nil, nil); End; There are a few additional issues that arise from translation. For example, Pascal isn't a case sensitive language and source files called units are considered a name space. This means that the LIBSSH2_PUBLICKEY.h translation as Unit LIBSSH2_PUBLICKEY.pas inherently causes a name collision conflict with LIBSSH2_PUBLICKEY (typedef'd from the _LIBSSH2_PUBLICKEY struct). In Pascal, all structs (called records) are already types, and do not need typedef aliasing, so the solution is either to rename the file (in the pascal translation - I'm not a fan), or just dropping use of LIBSSH2_PUBLICKEY as a type infavor of _LIBSSH2_PUBLICKEY (again, only in Pascal code). A few additional cases of case sensitivity conflict also happen with As an alternate solution it would also be possible to fold all libssh2.h, libssh2_publickey.h and libssh2_sftp.h into a single libssh2.pas import file to avoid the name conflicts - in fact, this might be the simplest method in the long run. One translation issue I am still struggling with is that pascal does not consider pointers and arrays to be interchangable. This definitely affects how items like _libssh2_publickey_attribute gets treated. My goal is to produce a translation of the headers that is as pure as possible to minimize any actual code differences as possible - so that someone trying to use libssh2 in Delphi can look at code written in C and expect to write nearly idenitical code (language syntax changes not withstanding) - and vice vera. In translating the C Demo programs to Dephi, the goal would be as close to a line for line translation as possible with the possible exception of more stringent resource management for when errors do occur. The concept of creating higher level objects that encapuslate and hide libssh2 itself fall outside of what I was suggesting to provide. I have an initial translation of the header files already converted and a single demo app translated - they are by no means final, but definitely work in all the areas I have tested so far. I would be willing to email them to anyone interested in looking at them in their entirty - they compress in to a 17K .7z file. Just drop me an email and I will happilly fire off a copy. - C Johnson _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 22:06:46 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FL6W45014017; Mon, 15 Mar 2010 22:06:43 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FL6Ux0013966 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 22:06:30 +0100 Received: (qmail 5534 invoked from network); 15 Mar 2010 21:06:37 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 15 Mar 2010 21:06:37 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 15 Mar 2010 21:06:37 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #165: size_t conversion X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/165#comment:1 Message-ID: <053.af5ec2a9814d57ac1c6b536a29c2874b@libssh2.stuge.se> References: <044.e1ec2089c65b3d0192e00f84b5cdb6b5@libssh2.stuge.se> X-Trac-Ticket-ID: 165 In-Reply-To: <044.e1ec2089c65b3d0192e00f84b5cdb6b5@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #165: size_t conversion -------------------------+-------------------------------------------------- Reporter: roadrunn | Owner: Type: enhancement | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | -------------------------+-------------------------------------------------- Comment(by bagder): Please make the patch with diff -u, or just use git diff... -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 15 22:08:30 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FL8PLI014539; Mon, 15 Mar 2010 22:08:28 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2FL8OLH014536 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 15 Mar 2010 22:08:24 +0100 Received: (qmail 5733 invoked from network); 15 Mar 2010 21:08:31 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 15 Mar 2010 21:08:31 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 15 Mar 2010 21:08:31 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #123: libssh2_channel_close() hangs in non blocking mode X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/123#comment:3 Message-ID: <054.68f38fcc78b59356b06e20940fc5b184@libssh2.stuge.se> References: <045.44caf8b8215f44f6756e43fff7ed8a8c@libssh2.stuge.se> X-Trac-Ticket-ID: 123 In-Reply-To: <045.44caf8b8215f44f6756e43fff7ed8a8c@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #123: libssh2_channel_close() hangs in non blocking mode ------------------------+--------------------------------------------------- Reporter: anonymous | Owner: bagder Type: defect | Status: closed Priority: normal | Milestone: Component: | Version: 1.2.1 Resolution: invalid | Keywords: Blocking: | Blockedby: ------------------------+--------------------------------------------------- Changes (by bagder): * status: assigned => closed * resolution: => invalid Comment: It is really a strangely put problem and there's no feedback on the problem even after a very long time. I'm closing this as invalid. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 17 19:47:53 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2HIkqSV007989; Wed, 17 Mar 2010 19:47:05 +0100 Received: from mail.intersystems.com (mail.intersys.COM [198.133.74.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2HIko7W007978 for ; Wed, 17 Mar 2010 19:46:51 +0100 X-InterSystems: Sent from InterSystems X-InterSystems: Sent from InterSystems Mime-Version: 1.0 (Apple Message framework v1077) Subject: Re: dispatch_protocol_error: type 90 seq 4 From: Dave McCaldon In-Reply-To: Date: Wed, 17 Mar 2010 14:46:50 -0400 Message-Id: <9D13D144-C1A7-422C-9CBA-80FE4D900C93@intersystems.com> References: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> To: libssh2 development X-Mailer: Apple Mail (2.1077) X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mar 13, 2010, at 4:07 PM, Daniel Stenberg wrote: > On Thu, 11 Mar 2010, Dave McCaldon wrote: > >> I have a Windows client app using libssh2-1.2.3 that seems to be unable to connect to a remote server (OpenSSH 5.2) to do SFTP, it fails on the server side with: > > And other tools have no problems using SFTP against this server? No, putty was able to connect. >> dispatch_protocol_error: type 90 seq 4 > >> [libssh2] 44.093943 Transport: Disconnect(7): Service not requested or authenticated(en) > >> Does anyone have any more details on what this means? (configuration perhaps?) > > That's the error message from the server so I figure more research into what makes the server say so is required to better understand this. In the putty client log I see this (below) right after KEX done, which is just before the dispatch_protocol error: debug3: Wrote 52 bytes for a total of 2497 debug1: userauth-request for user USERNAME service ssh-connection method none debug1: attempt 0 failures 0 ... It then goes on to try to connect using password which seems to succeed. >> => libssh2_transport_write plain (24 bytes) >> 0000: 5A 00 00 00 07 73 65 73 73 69 6F 6E 00 00 00 00 : Z....session.... >> 0010: 00 01 00 00 00 00 80 00 : ........ > > There's the type 90 I believe (0x5A). Does anyone know what we're sending to the server here? Thanks! _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 17 19:55:02 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2HIsfsK011228; Wed, 17 Mar 2010 19:54:44 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2HIseAF011221 for ; Wed, 17 Mar 2010 19:54:40 +0100 Date: Wed, 17 Mar 2010 19:54:40 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: dispatch_protocol_error: type 90 seq 4 In-Reply-To: <9D13D144-C1A7-422C-9CBA-80FE4D900C93@intersystems.com> Message-ID: References: <396758DF-1D31-41F0-AA85-6EDDE6077374@intersystems.com> <9D13D144-C1A7-422C-9CBA-80FE4D900C93@intersystems.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 17 Mar 2010, Dave McCaldon wrote: >>> => libssh2_transport_write plain (24 bytes) >>> 0000: 5A 00 00 00 07 73 65 73 73 69 6F 6E 00 00 00 00 : Z....session.... >>> 0010: 00 01 00 00 00 00 80 00 : ........ >> >> There's the type 90 I believe (0x5A). > > Does anyone know what we're sending to the server here? libssh2/src/channel.c:_libssh2_channel_open() line 194 shows how it sets the first byte in the packet to SSH_MSG_CHANNEL_OPEN which is 90/0x5a. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 07:48:01 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J6llSx028970; Fri, 19 Mar 2010 07:47:57 +0100 Received: from web8703.mail.in.yahoo.com (web8703.mail.in.yahoo.com [203.84.221.124]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2J6li6i028947 for ; Fri, 19 Mar 2010 07:47:45 +0100 Received: (qmail 15965 invoked by uid 60001); 19 Mar 2010 06:47:40 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1268981260; bh=STAdzsfXNyf3XhPuV0Aowiftj/i5GW9K+BsKbf4+RfM=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=sI1MdPiyE0qFibL/+BmHsaEClnAY+w5zlBB6KZ44l0dHoI3HbzcxYCZH5ZceFm0kZX4YNg+i/kZtitwrhH9r2HG7AfglUWdzhemnUWCVq62ieXhX6akOecVU6UOvjuzTPMtic/q4VuDk5kDaKcG03ZIeFnYI0RpHpNaKThK4Q/Y= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=CCI1OpCPWBEBvovYOkxieMGK4DqsijEjrjy6wcUTjsI9m31QJfY0J4lIyggwH9i8hc3Q5bMs0EyZHq3iN21gNIpwPYSWcZH81CKU/TAZLbJkJYtQXdHI7LoMD03Os99pLhB7ORXiYxxy+EfFSLivVDzWvYNXMwdTT/JHs4p2qIA=; Message-ID: <868336.15735.qm@web8703.mail.in.yahoo.com> X-YMail-OSG: 91CR.QcVM1mUxp_B98rEZN67SrO_3CC4VQCPaJEuRVY1hbC nhKBs9Ll186OevTM5ETZZtNrQ11.4eq5g8VsPiRnDlovKteydu5eQsjHahpj Y8Lal2NK4jG5k4OQY.x8e1Z4OYx4GBi3PYvEs5ziOq8TBSmfcrtGYYroxuUE R3cOqHdr_XxxQlil.lrd7U28oNXNRfQxaMADE_V_rvIQ3wS2JGhLW34gKUv9 uaSSspzCJck5q4_elPbgE0ojC5Daq8HCRXcL2DLckHit16AGdSdkW9z1Dp2O 0Izpr0HLRAw0Gfoy5oHtLIo7o75e8ZizNl9u69Za. Received: from [220.225.69.166] by web8703.mail.in.yahoo.com via HTTP; Fri, 19 Mar 2010 12:17:40 IST X-Mailer: YahooMailRC/324.3 YahooMailWebService/0.8.100.260964 Date: Fri, 19 Mar 2010 12:17:40 +0530 (IST) From: suyog jadhav Subject: SIGSEGV in packet.c To: libssh2-devel@cool.haxx.se MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0414892884==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0414892884== Content-Type: multipart/alternative; boundary="0-1389942342-1268981260=:15735" --0-1389942342-1268981260=:15735 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi All,=0A=0AI am posting for first time ,so don't know how to go about thi= s.=0A=0AI found a bug in code which results in SIGSEGV in my application fo= r a particular condition.=0ADebugging revealed the culprit to be in packet.= c, =0A=0A case SSH_MSG_CHANNEL_WINDOW_ADJUST:=0A {=0A = unsigned long bytestoadd =3D _libssh2_ntohu32(data + 5);=0A = session->packAdd_channel =3D=0A _libssh2_cha= nnel_locate(session,_libssh2_ntohu32(data + 1));=0A=0A if (s= ession->packAdd_channel && bytestoadd) {=0A session->pac= kAdd_channel->local.window_size +=3D bytestoadd;=0A }=0A = _libssh2_debug(session, LIBSSH2_TRACE_CONN,=0A = "Window adjust received for channel %lu/%lu, adding %lu bytes= , new window_size=3D%lu",=0A session->packAdd= _channel->local.id,=0A session->packAdd_chann= el->remote.id,=0A bytestoadd,=0A = session->packAdd_channel->local.window_size);=0A=0A = LIBSSH2_FREE(session, data);=0A session->packAdd_s= tate =3D libssh2_NB_state_idle;=0A return 0;=0A }= =0A=0Aif _libssh2_channel_locate returns NULL to session->packAdd_channel, = and then if it is accessed in _libssh2_debug and it segfaults, so the _libs= sh2_debug call should be inside the preceding IF block.=0AI understand the = underlying cause has to be researched more i.e. why is SSH_MSG_CHANNEL_WIND= OW_ADJUST received for non-existent channel, but the above check has to be= done for correctness any way.any ideas?=0A=0AKindly let me know how this c= hange can be added in libssh2?=0A=0ARegards=0ASuyog Jadhav=0A=0A=0A T= he INTERNET now has a personality. YOURS! See your Yahoo! Homepage. http://= in.yahoo.com/ --0-1389942342-1268981260=:15735 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
Hi All,

I am posting for first time ,so don't know how to go= about this.

I found a bug in code which results in SIGSEGV in my ap= plication for a particular condition.
Debugging revealed the culprit to = be in packet.c,

        case SSH= _MSG_CHANNEL_WINDOW_ADJUST:
       &n= bsp;    {
        = ;        unsigned long bytestoadd =3D _l= ibssh2_ntohu32(data + 5);
       &nbs= p;        session->packAdd_channel = =3D
           &n= bsp;        _libssh2_channel_locate(session,_libssh2_ntohu32(data + 1));

 =             &nb= sp;  if (session->packAdd_channel && bytestoadd) {
 = ;            &n= bsp;      session->packAdd_channel->local.wi= ndow_size +=3D bytestoadd;
       &nb= sp;        }
    =             _libssh2= _debug(session, LIBSSH2_TRACE_CONN,
      =             &nb= sp;            "Wind= ow adjust received for channel %lu/%lu, adding %lu bytes, new window_size=3D%lu",
        &nb= sp;            =           session->packAdd_= channel->local.id,
        &n= bsp;            = ;          session->packAdd= _channel->remote.id,
        =             &nb= sp;          bytestoadd,
&n= bsp;            = ;            &n= bsp;     session->packAdd_channel->local.window_size);

  &nb= sp;            = LIBSSH2_FREE(session, data);
       =          session->packAdd_state = =3D libssh2_NB_state_idle;
       &nb= sp;        return 0;
  &nbs= p;         }

if _libssh2_cha= nnel_locate returns NULL to session->packAdd_channel, and then if it is = accessed in _libssh2_debug and it segfaults, so the _libssh2_debug call sho= uld be inside the preceding IF block.
I understand the underlying cause = has to be researched more i.e. why is SSH_MSG_CHANNEL_WINDOW_ADJUST  r= eceived for non-existent channel, but the above check has to be done for co= rrectness any way.any ideas?

Kindly let me know how this change  can be added in libssh2?

Regards
Suyog Jadhav
=0A

=0A=0A=0A=0A
=0AThe IN= TERNET now has a personality. YOURS! See your Yahoo! Homepage= . --0-1389942342-1268981260=:15735-- --===============0414892884== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0414892884==-- From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 09:08:56 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J88kTn024694; Fri, 19 Mar 2010 09:08:54 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J88j3M024690 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 19 Mar 2010 09:08:45 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2J88jeP018593 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT); Fri, 19 Mar 2010 09:08:47 +0100 X-Hashcash: 1:22:100319:lars.nordin@sdlabs.se::A0UcVWGERR3WN6Z0:CfVl From: Simon Josefsson To: libssh2 development Subject: Re: patch to add libssh2_init() and libssh2_exit() References: <4B9781DE.2060707@SDlabs.se> <87y6i0ierk.fsf@mocca.josefsson.org> <4B979175.6050403@SDlabs.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100319:libssh2-devel@cool.haxx.se::1Vdiv6VMJGKgFF7M:E7/L Date: Fri, 19 Mar 2010 09:08:45 +0100 In-Reply-To: <4B979175.6050403@SDlabs.se> (Lars Nordin's message of "Wed, 10 Mar 2010 13:32:53 +0100") Message-ID: <87eijg4tte.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean Cc: Lars Nordin X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Lars, I have applied this patch now, with a few cleanups. Could you test it? Others, we now global init/exit functions for libssh2: libssh2_init and libssh2_exit. This is really normal practice for shared libraries, and we need it to do proper crypto library initialization. They may be used for other things in the future, e.g. to initialize a GSS-API library or similar if we add GSS-API authentication. Or other things. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 09:18:01 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J8HvWJ027909; Fri, 19 Mar 2010 09:17:59 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J8HucI027897 for ; Fri, 19 Mar 2010 09:17:56 +0100 Date: Fri, 19 Mar 2010 09:17:56 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. In-Reply-To: <20100319080615.14727.qmail@earth.stuge.se> Message-ID: References: <20100319080615.14727.qmail@earth.stuge.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 19 Mar 2010, libssh2@git.stuge.se wrote: > Author: Simon Josefsson > Commit: Simon Josefsson > > Add global init/exit points, to do crypto initialization in one place. > > By Lars Nordin. Just a reminder: use the --author switch to git commit when you commit someone else's patch as then git will keep that meta data correctly for us! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 09:38:44 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J8cdrR010383; Fri, 19 Mar 2010 09:38:42 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J8cbCe010377 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 19 Mar 2010 09:38:37 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2J8cbos019805 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Fri, 19 Mar 2010 09:38:38 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. References: <20100319080615.14727.qmail@earth.stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100319:libssh2-devel@cool.haxx.se::IechF5SyXElXkfIB:8N2m Date: Fri, 19 Mar 2010 09:38:37 +0100 In-Reply-To: (Daniel Stenberg's message of "Fri, 19 Mar 2010 09:17:56 +0100 (CET)") Message-ID: <877hp84sfm.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Fri, 19 Mar 2010, libssh2@git.stuge.se wrote: > >> Author: Simon Josefsson >> Commit: Simon Josefsson >> >> Add global init/exit points, to do crypto initialization in one place. >> >> By Lars Nordin. > > Just a reminder: use the --author switch to git commit when you commit > someone else's patch as then git will keep that meta data correctly > for us! Oops! Will try to remember next time. I was a bit unsure whether to commit his work directly, and then make it work in later commits, or make it work directly and commit in one go. Now the result was a mix of both approaches, I committed a modified version of Lars patch and then made more changes to it... I guess the preferred approach would be to commit his patch as-is with --author and the fix it in later commits? /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 09:48:58 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J8mpdW021375; Fri, 19 Mar 2010 09:48:56 +0100 Received: from mail-fx0-f225.google.com (mail-fx0-f225.google.com [209.85.220.225]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J8mofX021369 for ; Fri, 19 Mar 2010 09:48:50 +0100 Received: by fxm25 with SMTP id 25so320162fxm.7 for ; Fri, 19 Mar 2010 01:48:48 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type:content-transfer-encoding; bh=mq/K53YM/zjrTBf/xUSl06o/Z7dnN2VBG4osXLHIWYw=; b=r05Xs+MBk7ILZ4egPsBgUBUb/CmdkkAQKEFrsm/JSmcG1XnkdspvCE91GVWsQp4Mfr 37A58obxZR1UGs7YzyV8N/JVgJ/UzX9QV+Y2v682Ia2d89VoXiQg6zoSEphZ0cCufrgQ A0Nswfs6cR6t6b+4VKL5b7X5z0okGI4ZAHQ3Q= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=HLik8uLWjZaLQKn2LuERwijI74Tnbks1k+h7jmTFb+M0/zJe42g63lt4ZmcSWIMxJ9 0254j7pBoQNvUZ9ZsyzKLBsZUQJRZ2g2hRz9cC6nOgVvdZ/muSgxtFWMunmVVkfDJGzW ILeID8M44dauVJZ8wUpLEVjAtSadC+FCIMph8= MIME-Version: 1.0 Received: by 10.239.186.79 with SMTP id f15mr613275hbh.103.1268988528013; Fri, 19 Mar 2010 01:48:48 -0700 (PDT) In-Reply-To: <877hp84sfm.fsf@mocca.josefsson.org> References: <20100319080615.14727.qmail@earth.stuge.se> <877hp84sfm.fsf@mocca.josefsson.org> Date: Fri, 19 Mar 2010 09:48:47 +0100 X-Google-Sender-Auth: 2ce71b1ed98708eb Message-ID: Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. From: Tor Arntsen To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, Mar 19, 2010 at 09:38, Simon Josefsson wrote: >> Just a reminder: use the --author switch to git commit when you commit >> someone else's patch as then git will keep that meta data correctly >> for us! > > Oops! =A0Will try to remember next time. =A0I was a bit unsure whether to > commit his work directly, and then make it work in later commits, or > make it work directly and commit in one go. =A0Now the result was a mix of > both approaches, I committed a modified version of Lars patch and then > made more changes to it... > > I guess the preferred approach would be to commit his patch as-is with > --author and the fix it in later commits? You can also use --signoff=3D"you yourself " in addition to --author (at least with any reasonably new git version). And you can also fix it after the fact, e.g. git commit -m"Commit message" oops, forgot.. git commit --amend --author=3D....etc which will fix the last commit. Just like any amend it should only be done before it's been pushed (if you do changes to your own clone, and then push), or at least before anyone has got a chance to pull from it. -Tor _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 10:00:57 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J90nAM027369; Fri, 19 Mar 2010 10:00:53 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2J90lex027332 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 19 Mar 2010 10:00:48 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2J90mqX020433 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Fri, 19 Mar 2010 10:00:50 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. References: <20100319080615.14727.qmail@earth.stuge.se> <877hp84sfm.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100319:libssh2-devel@cool.haxx.se::9HKGJmmq3+65ozB1:IGhB Date: Fri, 19 Mar 2010 10:00:48 +0100 In-Reply-To: (Tor Arntsen's message of "Fri, 19 Mar 2010 09:48:47 +0100") Message-ID: <87hboc3cu7.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Tor Arntsen writes: > On Fri, Mar 19, 2010 at 09:38, Simon Josefsson wrot= e: > >>> Just a reminder: use the --author switch to git commit when you commit >>> someone else's patch as then git will keep that meta data correctly >>> for us! >> >> Oops! =A0Will try to remember next time. =A0I was a bit unsure whether to >> commit his work directly, and then make it work in later commits, or >> make it work directly and commit in one go. =A0Now the result was a mix = of >> both approaches, I committed a modified version of Lars patch and then >> made more changes to it... >> >> I guess the preferred approach would be to commit his patch as-is with >> --author and the fix it in later commits? > > You can also use --signoff=3D"you yourself " in addition to > --author (at least with any reasonably new git version). > > And you can also fix it after the fact, e.g. > > git commit -m"Commit message" > oops, forgot.. > git commit --amend --author=3D....etc > > which will fix the last commit. Just like any amend it should only be > done before it's been pushed (if you do changes to your own clone, and > then push), or at least before anyone has got a chance to pull from > it. Thanks! I'm a bit too quick with pushing the changes, I guess, but will try to remember these hints for the future. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 13:13:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JCDFw5007445; Fri, 19 Mar 2010 13:13:26 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2JCDDZV007425 for ; Fri, 19 Mar 2010 13:13:13 +0100 Received: (qmail 4192 invoked by uid 501); 19 Mar 2010 12:13:17 -0000 Message-ID: <20100319121317.4191.qmail@stuge.se> Date: Fri, 19 Mar 2010 13:13:17 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master d002839... Fix namespace pollution. Mail-Followup-To: libssh2-devel@cool.haxx.se MIME-Version: 1.0 Content-Disposition: inline X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se libssh2@git.stuge.se wrote: > +++ b/src/global.c > @@ -37,12 +37,12 @@ > > #include "libssh2_priv.h" > > -int libssh2_initialized = 0; > -int libssh2_init_flags = 0; > +static int _libssh2_initialized = 0; > +static int _libssh2_init_flags = 0; > > LIBSSH2_API int libssh2_init(int flags) > { > - if (!(flags & LIBSSH2_INIT_NO_CRYPTO_INIT)) { > + if (libssh2_initialized == 0 && !(flags & LIBSSH2_INIT_NO_CRYPTO_INIT)) { > libssh2_crypto_init(); > } > > @@ -57,11 +57,18 @@ LIBSSH2_API void libssh2_exit() > if (libssh2_initialized == 0) > return; > > + libssh2_initialized--; Shouldn't these two uses of the variable be changed? Maybe there are other places too? //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 13:30:50 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JCUiKQ014410; Fri, 19 Mar 2010 13:30:48 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2JCUhJE014399 for ; Fri, 19 Mar 2010 13:30:43 +0100 Received: (qmail 7202 invoked by uid 501); 19 Mar 2010 12:30:47 -0000 Message-ID: <20100319123047.7201.qmail@stuge.se> Date: Fri, 19 Mar 2010 13:30:47 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. Mail-Followup-To: libssh2-devel@cool.haxx.se References: <20100319080615.14727.qmail@earth.stuge.se> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20100319080615.14727.qmail@earth.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se libssh2@git.stuge.se wrote: > +void > +libssh2_md5(const unsigned char *message, unsigned long len, > + unsigned char *out) > +{ > + EVP_MD_CTX ctx; > + > + EVP_DigestInit(&ctx, EVP_get_digestbyname("md5")); > + EVP_DigestUpdate(&ctx, message, len); > + EVP_DigestFinal(&ctx, out, NULL); > +} Guys, MD5 is not required for SSH so it MUST NOT be required for libssh2! I'm quite annoyed with some of these changes from Lars that have been committed in a hurry. Don't get me wrong - we must make libssh2 work also for Lars, but we should remember that his requirements seem to be fairly specialized and I do not think it is acceptable for libssh2 to be optimized for, or assume, such situations. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 14:15:03 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDErkP007332; Fri, 19 Mar 2010 14:15:01 +0100 Received: from av11-1-sn2.hy.skanova.net (av11-1-sn2.hy.skanova.net [81.228.8.183]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDEqxF007321 for ; Fri, 19 Mar 2010 14:14:52 +0100 Received: by av11-1-sn2.hy.skanova.net (Postfix, from userid 502) id EE73F37F72; Fri, 19 Mar 2010 14:14:51 +0100 (CET) Received: from smtp4-2-sn2.hy.skanova.net (smtp4-2-sn2.hy.skanova.net [81.228.8.93]) by av11-1-sn2.hy.skanova.net (Postfix) with ESMTP id C013237E69 for ; Fri, 19 Mar 2010 14:14:51 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp4-2-sn2.hy.skanova.net (Postfix) with ESMTP id AA09837E4C for ; Fri, 19 Mar 2010 14:14:51 +0100 (CET) Message-ID: <4BA378CB.60405@SDlabs.se> Date: Fri, 19 Mar 2010 14:14:51 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. References: <20100319080615.14727.qmail@earth.stuge.se> <20100319123047.7201.qmail@stuge.se> In-Reply-To: <20100319123047.7201.qmail@stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0169613307==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --===============0169613307== Content-Type: multipart/alternative; boundary="------------070509030206010408020408" This is a multi-part message in MIME format. --------------070509030206010408020408 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit On 2010-03-19 13:30, Peter Stuge wrote: > libssh2@git.stuge.se wrote: > >> +void >> +libssh2_md5(const unsigned char *message, unsigned long len, >> + unsigned char *out) >> +{ >> + EVP_MD_CTX ctx; >> + >> + EVP_DigestInit(&ctx, EVP_get_digestbyname("md5")); >> + EVP_DigestUpdate(&ctx, message, len); >> + EVP_DigestFinal(&ctx, out, NULL); >> +} >> > Guys, MD5 is not required for SSH so it MUST NOT be required for > libssh2! > > Peter, A lot of the code in libssh2 is missing #ifdef's around definitions for MD5. and Yes, I missed that to. But the complain should not be pointed to me, please. > I'm quite annoyed with some of these changes from Lars that have been > committed in a hurry. > > Don't get me wrong - we must make libssh2 work also for Lars, but we > should remember that his requirements seem to be fairly specialized > and I do not think it is acceptable for libssh2 to be optimized for, > or assume, such situations. > > The openssl part of libssh2 is using part (like AES, MD5 and SHA) that is using the direct calls to the internal function. This is not working for my use BUT it's also recommended by openssl to use the EVP-functions from www.openssl.org "Applications should use the higher level functions EVP_DigestInit(3) etc. instead of calling the hash functions directly." > //Peter > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > /Lars --------------070509030206010408020408 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit On 2010-03-19 13:30, Peter Stuge wrote:
libssh2@git.stuge.se wrote:
  
+void
+libssh2_md5(const unsigned char *message, unsigned long len,
+            unsigned char *out)
+{
+    EVP_MD_CTX ctx;
+
+    EVP_DigestInit(&ctx, EVP_get_digestbyname("md5"));
+    EVP_DigestUpdate(&ctx, message, len);
+    EVP_DigestFinal(&ctx, out, NULL);
+}
    
Guys, MD5 is not required for SSH so it MUST NOT be required for
libssh2!

  
Peter,

A lot of the code in libssh2 is missing #ifdef's around definitions for MD5. and Yes, I missed that to. But the complain should not be pointed to me, please.

I'm quite annoyed with some of these changes from Lars that have been
committed in a hurry.

Don't get me wrong - we must make libssh2 work also for Lars, but we
should remember that his requirements seem to be fairly specialized
and I do not think it is acceptable for libssh2 to be optimized for,
or assume, such situations.

  
The openssl part of libssh2 is using part (like AES, MD5 and SHA) that is using the direct calls to the internal function. This is not working for my use BUT it's also recommended by openssl to use the EVP-functions from www.openssl.org
"Applications should use the higher level functions EVP_DigestInit(3) etc. instead of calling the hash functions directly."


//Peter
_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel

  
/Lars
--------------070509030206010408020408-- --===============0169613307== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0169613307==-- From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 14:25:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDP2tp016600; Fri, 19 Mar 2010 14:25:05 +0100 Received: from av9-2-sn2.hy.skanova.net (av9-2-sn2.hy.skanova.net [81.228.8.180]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDP0Jk016447 for ; Fri, 19 Mar 2010 14:25:00 +0100 Received: by av9-2-sn2.hy.skanova.net (Postfix, from userid 502) id DACA8383FE; Fri, 19 Mar 2010 14:24:59 +0100 (CET) Received: from smtp4-1-sn2.hy.skanova.net (smtp4-1-sn2.hy.skanova.net [81.228.8.92]) by av9-2-sn2.hy.skanova.net (Postfix) with ESMTP id 93D64383F3 for ; Fri, 19 Mar 2010 14:24:58 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp4-1-sn2.hy.skanova.net (Postfix) with ESMTP id 7D53437E49 for ; Fri, 19 Mar 2010 14:24:58 +0100 (CET) Message-ID: <4BA37B2E.8040001@SDlabs.se> Date: Fri, 19 Mar 2010 14:25:02 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master d002839... Fix namespace pollution. References: <20100319121317.4191.qmail@stuge.se> In-Reply-To: <20100319121317.4191.qmail@stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 2010-03-19 13:13, Peter Stuge wrote: > libssh2@git.stuge.se wrote: > >> +++ b/src/global.c >> @@ -37,12 +37,12 @@ >> >> #include "libssh2_priv.h" >> >> -int libssh2_initialized = 0; >> -int libssh2_init_flags = 0; >> +static int _libssh2_initialized = 0; >> +static int _libssh2_init_flags = 0; >> >> LIBSSH2_API int libssh2_init(int flags) >> { >> - if (!(flags& LIBSSH2_INIT_NO_CRYPTO_INIT)) { >> + if (libssh2_initialized == 0&& !(flags& LIBSSH2_INIT_NO_CRYPTO_INIT)) { >> libssh2_crypto_init(); >> } >> >> @@ -57,11 +57,18 @@ LIBSSH2_API void libssh2_exit() >> if (libssh2_initialized == 0) >> return; >> >> + libssh2_initialized--; >> > Shouldn't these two uses of the variable be changed? Maybe there are > other places too? > > Making those to static would require a new function to check the value of libssh2_initialized (used in session.c to preserve the methode if doing init in session_init. It's a question on how to handle global variables, should they be handled only by calls to functions in global.c? Or is it OK to check and set the variable direct in other parts of libssh2. BTW, is _ the libssh2 standard for static variables? I trying to understand the code-standard. > //Peter > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel > > _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 14:29:09 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDT5Z9020482; Fri, 19 Mar 2010 14:29:07 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDT3Yx020443 for ; Fri, 19 Mar 2010 14:29:03 +0100 Date: Fri, 19 Mar 2010 14:29:03 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: libssh2 master d002839... Fix namespace pollution. In-Reply-To: <4BA37B2E.8040001@SDlabs.se> Message-ID: References: <20100319121317.4191.qmail@stuge.se> <4BA37B2E.8040001@SDlabs.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 19 Mar 2010, Lars Nordin wrote: > BTW, is _ the libssh2 standard for static variables? I trying to understand > the code-standard. No. We use _libssh2 as prefix for library-wide non-static symbols that aren't part of the public API. static symbols can be called anything with no prefix requirements. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 14:43:40 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDhZ4X028086; Fri, 19 Mar 2010 14:43:38 +0100 Received: from av10-2-sn2.hy.skanova.net (av10-2-sn2.hy.skanova.net [81.228.8.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JDhYTn028062 for ; Fri, 19 Mar 2010 14:43:34 +0100 Received: by av10-2-sn2.hy.skanova.net (Postfix, from userid 502) id BC1A137F18; Fri, 19 Mar 2010 14:43:33 +0100 (CET) Received: from smtp4-2-sn2.hy.skanova.net (smtp4-2-sn2.hy.skanova.net [81.228.8.93]) by av10-2-sn2.hy.skanova.net (Postfix) with ESMTP id B5053382EC for ; Fri, 19 Mar 2010 14:43:24 +0100 (CET) Received: from [192.168.20.161] (90-227-111-200-no53.business.telia.com [90.227.111.200]) by smtp4-2-sn2.hy.skanova.net (Postfix) with ESMTP id 9FAE837E4A for ; Fri, 19 Mar 2010 14:43:24 +0100 (CET) Message-ID: <4BA37F7C.6060803@SDlabs.se> Date: Fri, 19 Mar 2010 14:43:24 +0100 From: Lars Nordin User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7) Gecko/20100125 Shredder/3.0.1 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master d002839... Fix namespace pollution. References: <20100319121317.4191.qmail@stuge.se> <4BA37B2E.8040001@SDlabs.se> In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 2010-03-19 14:29, Daniel Stenberg wrote: > On Fri, 19 Mar 2010, Lars Nordin wrote: > >> BTW, is _ the libssh2 standard for static variables? I trying to >> understand the code-standard. > > No. > > We use _libssh2 as prefix for library-wide non-static symbols that > aren't part of the public API. static symbols can be called anything > with no prefix requirements. > OK, then the name should be changed (my fault), but not being static. I think we should allow global variables, instead of having to hide them and making read and set functions. _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 15:01:41 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JE1bPu005628; Fri, 19 Mar 2010 15:01:40 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JE1aqJ005587 for ; Fri, 19 Mar 2010 15:01:36 +0100 Date: Fri, 19 Mar 2010 15:01:36 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: SIGSEGV in packet.c In-Reply-To: <868336.15735.qm@web8703.mail.in.yahoo.com> Message-ID: References: <868336.15735.qm@web8703.mail.in.yahoo.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 19 Mar 2010, suyog jadhav wrote: > I found a bug in code which results in SIGSEGV in my application for a > particular condition. Debugging revealed the culprit to be in packet.c, Thanks! I've pushed a fix for this now that also outputs a little extra debug message when in fact the channel to which window adjust is received isn't found. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 15:16:23 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JEG5vH018919; Fri, 19 Mar 2010 15:16:13 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JEG4gp018912 for ; Fri, 19 Mar 2010 15:16:04 +0100 Date: Fri, 19 Mar 2010 15:16:04 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Transfer files In-Reply-To: <4B9A1366.2050709@gmx.at> Message-ID: References: <4B9A1366.2050709@gmx.at> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 12 Mar 2010, double wrote: > Is there an easy way to transfer a file from server "B" to server "C", if > libssh2 runs on server "A"? To login on server "B" and put a file to server > "C" via sftp? Well, you can send a command from A to server B: "scp file remote@C:file" but other than so, SSH has no particular provisions for third party transfers the way you describe. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 21:56:13 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JKtvYX019495; Fri, 19 Mar 2010 21:56:10 +0100 Received: from mail.gmx.net (mail.gmx.net [213.165.64.20]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2JKtuXi019491 for ; Fri, 19 Mar 2010 21:55:56 +0100 Received: (qmail invoked by alias); 19 Mar 2010 20:55:58 -0000 Received: from xdsl-188-155-149-216.adslplus.ch (EHLO [192.168.1.65]) [188.155.149.216] by mail.gmx.net (mp072) with SMTP; 19 Mar 2010 21:55:58 +0100 X-Authenticated: #1568940 X-Provags-ID: V01U2FsdGVkX19e+ohj1QBePke5Q5VwhYJ/xZC0zUActpTgIS1mE6 wan1yDtTUX4Zeq Message-ID: <4BA3E4D6.60307@gmx.at> Date: Fri, 19 Mar 2010 21:55:50 +0100 From: double User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: libssh2 development Subject: Re: Transfer files References: <4B9A1366.2050709@gmx.at> In-Reply-To: X-Y-GMX-Trusted: 0 X-FuHaFi: 0.70999999999999996 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hi Daniel, Daniel Stenberg schrieb: > Well, you can send a command from A to server B: "scp file > remote@C:file" but > other than so, SSH has no particular provisions for third party > transfers the way > you describe. Good idea! Quite simple. Thanks! Marcus _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 19 22:49:10 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JLn45Z030495; Fri, 19 Mar 2010 22:49:08 +0100 Received: from fg-out-1718.google.com (fg-out-1718.google.com [72.14.220.159]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2JLn3JX030488 for ; Fri, 19 Mar 2010 22:49:03 +0100 Received: by fg-out-1718.google.com with SMTP id 22so190774fge.11 for ; Fri, 19 Mar 2010 14:49:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:content-type; bh=icvJm7yMAqGuEQHF5Xjah0mIpMjdJnfx3f16MnQ1xoA=; b=rzR56U4ajc8i+rVbWZK0RZPPUCrde6E5e8Rcx5Is0dfbk3DioWm1+bJhA1vQzKz9jK +g8EQuVH730fXLRvpGLz5niv6H2/0Ub7rImF8OT73+sGJLYClywpkBHJFud7ZUjCjUzJ VQQNczWg1QFzHQAsO1LCs+QCws6wndY1lkjBo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=Vb8zk1OyxKuyJiTTlKr0Zk8x2NWCPW6w7s4mc+fVrAywYbX4k9c5IZLT+1EVaTtOsQ 6w+oyApJUMjRrjs5CtAwmQmp0zKePLcfFlwphN+EcUjJUmqBx66cjD5TIlfjPCe6dyPp DSLDDlnIw/raJp158GS7pdPRJ4sPxrw3ltu+I= MIME-Version: 1.0 Received: by 10.103.67.20 with SMTP id u20mr3136522muk.97.1269035350349; Fri, 19 Mar 2010 14:49:10 -0700 (PDT) Date: Fri, 19 Mar 2010 22:49:10 +0100 Message-ID: <65800b2a1003191449y4a82f0d5v249ff647427a20cb@mail.gmail.com> Subject: scp example problem From: vicent roca daniel To: libssh2-devel@cool.haxx.se X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0294912318==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0294912318== Content-Type: multipart/alternative; boundary=0016e659fb4443da0304822e5127 --0016e659fb4443da0304822e5127 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi guys, I'm having some problems with libssh2 sending files using scp. The problem is that I can't send files > 1MB. The transfer seems to stop without error. I tryed my example based on the scp_write.c provided on the examples directory and same result as the scp_write.c it self. =BFDoes anyone know what can I try? I've tryed wih Mac os and linux and differents target servers. Same result. I'm using release: libssh2 1.2.4from February 13 2010 Thanks in advance. :) --0016e659fb4443da0304822e5127 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi guys,

I'm having some problems with libssh2 sending files usi= ng scp.
The problem is that I can't send files > 1MB.
The tran= sfer seems to stop without error.
I tryed my example based on the scp_wr= ite.c provided on the examples directory and same result as the scp_write.c= it self.

=BFDoes anyone know what can I try?
I've tryed wih Mac os and li= nux and differents target servers. Same result.
I'm using=A0 releas= e: libssh2 1.2.4 from February 13 2010

Thanks in advance. :)
--0016e659fb4443da0304822e5127-- --===============0294912318== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0294912318==-- From libssh2-devel-bounces@cool.haxx.se Sat Mar 20 03:33:28 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2K2XEN2015549; Sat, 20 Mar 2010 03:33:24 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2K2XDt2015546 for ; Sat, 20 Mar 2010 03:33:13 +0100 Received: (qmail 26996 invoked by uid 501); 20 Mar 2010 02:33:22 -0000 Message-ID: <20100320023322.26995.qmail@stuge.se> Date: Sat, 20 Mar 2010 03:33:22 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Transfer files Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4B9A1366.2050709@gmx.at> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <4B9A1366.2050709@gmx.at> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se double wrote: > Is there an easy way to transfer a file from server "B" to server > "C", if libssh2 runs on server "A"? To login on server "B" and put > a file to server "C" via sftp? You can use direct-tcpip to tunnel a second SSH connection between B and C back to A. Use SFTP within the tunneled connection. This way you have API control in your application of everything that goes on, but pumping the data requires more work by you. See the direct-tcpip example and feel free to ask any questions. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sat Mar 20 13:01:33 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2KC1GHf011776; Sat, 20 Mar 2010 13:01:30 +0100 Received: from mail-pw0-f54.google.com (mail-pw0-f54.google.com [209.85.160.54]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2KC1EQS011590 for ; Sat, 20 Mar 2010 13:01:15 +0100 Received: by pwj4 with SMTP id 4so3166706pwj.41 for ; Sat, 20 Mar 2010 05:01:12 -0700 (PDT) MIME-Version: 1.0 Received: by 10.142.67.22 with SMTP id p22mr2459550wfa.154.1269086471412; Sat, 20 Mar 2010 05:01:11 -0700 (PDT) Date: Sat, 20 Mar 2010 08:01:11 -0400 Message-ID: <509c0c331003200501u2b57b108g13fbf33918234ef3@mail.gmail.com> Subject: sftp shutdown error with MessageWay From: Matt Wixson To: libssh2-devel@cool.haxx.se X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1555762000==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============1555762000== Content-Type: multipart/alternative; boundary=001636e907d0515d3904823a381d --001636e907d0515d3904823a381d Content-Type: text/plain; charset=ISO-8859-1 Hello, I am attempting to download a file (any size, small or large) with an SFTP server by MessageWay.com and cannot get a clean shutdown after transfer. I have tried several versions of libssh2 including libssh2-1.2.5-20100318. The same transfer works using OpenSSH/sftp, but never with curl/libssh2. Unfortunately, I do not have access to the MessageWay SFTP server (version etc.) as it is hosted by a third party. The file is transferred completely, the file size on the client matches the file size on the server, but the process hangs for several minutes and then I receive an error from close. I know this isn't a curl list but the curl output happens to be: libssh2_sftp_close: Resource temporarily unavailable (this is output from perror after the call to libssh2_sftp_close()) curl: (56) Failure when receiving data from the peer Here is the debug information from SSH2: trace output here...I see the complete file transfer PROCESS HANGS FOR SEVERAL SECONDS == Info: SSH DISCONNECT starts now == Info: SFTP 0x9714090 state change from SSH_STOP to SSH_SFTP_SHUTDOWN == Info: Failed to close libssh2 file during shutdown: -30 == Info: SFTP 0x9714090 state change from SSH_SFTP_SHUTDOWN to SSH_SESSION_DISCONNECT == Info: SFTP 0x9714090 state change from SSH_SESSION_DISCONNECT to SSH_SESSION_FREE == Info: SFTP 0x9714090 state change from SSH_SESSION_FREE to SSH_STOP == Info: SSH DISCONNECT is done == Info: Closing connection #0 == Info: Failure when receiving data from the peer Any thoughts on how I can correct or avoid this, via setsockopt or configure options? Why would OpenSSH/sftp succeed but libssh2 fail during the close? Any help is appreciated! Thanks, Matt --001636e907d0515d3904823a381d Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello,

I am attempting to download a file (any size, small or large)= with an SFTP server by MessageWay.com and cannot get a clean shutdown afte= r transfer.=A0=A0 I have tried several versions of libssh2 including libssh= 2-1.2.5-20100318.=A0=A0 The same transfer works using OpenSSH/sftp, but nev= er with curl/libssh2.=A0=A0 Unfortunately, I do not have access to the Mess= ageWay SFTP server (version etc.) as it is hosted by a third party.

The file is transferred completely, the file size on the client matches= the file size on the server, but the process hangs for several minutes and= then I receive an error from close.=A0 I know this isn't a curl list b= ut the curl output happens to be:
libssh2_sftp_close: Resource temporarily unavailable (this is output from p= error after the call to libssh2_sftp_close())
curl: (56)=20 Failure when receiving data from the peer

Here is the debug informat= ion from SSH2:
trace output here...I see the complete file transfer
<= br>=A0=A0=A0 PROCESS HANGS FOR SEVERAL SECONDS

=3D=3D Info: SSH DISC= ONNECT starts now
=3D=3D Info: SFTP 0x9714090 state change from SSH_STOP to SSH_SFTP_SHUTDOWN=
=3D=3D Info: Failed to close libssh2 file during shutdown: -30
=3D= =3D Info: SFTP 0x9714090 state change from SSH_SFTP_SHUTDOWN to SSH_SESSION= _DISCONNECT
=3D=3D Info: SFTP 0x9714090 state change from SSH_SESSION_DISCONNECT to SSH= _SESSION_FREE
=3D=3D Info: SFTP 0x9714090 state change from SSH_SESSION_= FREE to SSH_STOP
=3D=3D Info: SSH DISCONNECT is done
=3D=3D Info: Clo= sing connection #0
=3D=3D Info: Failure when receiving data from the peer

Any thoughts = on how I can correct or avoid this, via setsockopt or configure options?=A0= =A0 Why would OpenSSH/sftp succeed but libssh2 fail during the close?=A0=A0= =A0 Any help is appreciated!

Thanks,
Matt


--001636e907d0515d3904823a381d-- --===============1555762000== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1555762000==-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 21 19:41:00 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2LIej8W013887; Sun, 21 Mar 2010 19:40:56 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2LIehVQ013835 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 21 Mar 2010 19:40:44 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2LIelGg027794 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Sun, 21 Mar 2010 19:40:48 +0100 From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master d002839... Fix namespace pollution. References: <20100319121317.4191.qmail@stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100321:libssh2-devel@cool.haxx.se::Gx3iFP9o8rCnlrbE:OYE0 Date: Sun, 21 Mar 2010 19:40:47 +0100 In-Reply-To: <20100319121317.4191.qmail@stuge.se> (Peter Stuge's message of "Fri, 19 Mar 2010 13:13:17 +0100") Message-ID: <87y6hlsekw.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge writes: > libssh2@git.stuge.se wrote: >> +++ b/src/global.c >> @@ -37,12 +37,12 @@ >> >> #include "libssh2_priv.h" >> >> -int libssh2_initialized = 0; >> -int libssh2_init_flags = 0; >> +static int _libssh2_initialized = 0; >> +static int _libssh2_init_flags = 0; >> >> LIBSSH2_API int libssh2_init(int flags) >> { >> - if (!(flags & LIBSSH2_INIT_NO_CRYPTO_INIT)) { >> + if (libssh2_initialized == 0 && !(flags & LIBSSH2_INIT_NO_CRYPTO_INIT)) { >> libssh2_crypto_init(); >> } >> >> @@ -57,11 +57,18 @@ LIBSSH2_API void libssh2_exit() >> if (libssh2_initialized == 0) >> return; >> >> + libssh2_initialized--; > > Shouldn't these two uses of the variable be changed? Maybe there are > other places too? I fixed it shortly after committing that, I think. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 21 19:42:14 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2LIgBp8015884; Sun, 21 Mar 2010 19:42:13 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2LIgANA015856 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 21 Mar 2010 19:42:10 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2LIgEJK027823 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Sun, 21 Mar 2010 19:42:16 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: libssh2 master d002839... Fix namespace pollution. References: <20100319121317.4191.qmail@stuge.se> <4BA37B2E.8040001@SDlabs.se> <4BA37F7C.6060803@SDlabs.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100321:libssh2-devel@cool.haxx.se::DlnaM6Ox8eEuh5M+:2QTz Date: Sun, 21 Mar 2010 19:42:13 +0100 In-Reply-To: <4BA37F7C.6060803@SDlabs.se> (Lars Nordin's message of "Fri, 19 Mar 2010 14:43:24 +0100") Message-ID: <87tys9seii.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Lars Nordin writes: > On 2010-03-19 14:29, Daniel Stenberg wrote: >> On Fri, 19 Mar 2010, Lars Nordin wrote: >> >>> BTW, is _ the libssh2 standard for static variables? I trying to >>> understand the code-standard. >> >> No. >> >> We use _libssh2 as prefix for library-wide non-static symbols that >> aren't part of the public API. static symbols can be called anything >> with no prefix requirements. >> > OK, then the name should be changed (my fault), but not being > static. I think we should allow global variables, instead of having to > hide them and making read and set functions. Check out how I solved this in latest git without any global variables or read/set functions. IMHO having cross-file global variables is a significant pain, and it was rather simple to avoid it here. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 21 19:49:58 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2LInt6J023731; Sun, 21 Mar 2010 19:49:58 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2LInsUL023728 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 21 Mar 2010 19:49:54 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2LInvdw027923 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Sun, 21 Mar 2010 19:49:59 +0100 From: Simon Josefsson To: libssh2-devel@cool.haxx.se Subject: Re: libssh2 master 9ea5b06... Add global init/exit points, to do crypto initialization in one place. References: <20100319080615.14727.qmail@earth.stuge.se> <20100319123047.7201.qmail@stuge.se> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100321:libssh2-devel@cool.haxx.se::zdGjU3XWXmsxAlYP:2ZSu Date: Sun, 21 Mar 2010 19:49:57 +0100 In-Reply-To: <20100319123047.7201.qmail@stuge.se> (Peter Stuge's message of "Fri, 19 Mar 2010 13:30:47 +0100") Message-ID: <87mxy1se5m.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Peter Stuge writes: > libssh2@git.stuge.se wrote: >> +void >> +libssh2_md5(const unsigned char *message, unsigned long len, >> + unsigned char *out) >> +{ >> + EVP_MD_CTX ctx; >> + >> + EVP_DigestInit(&ctx, EVP_get_digestbyname("md5")); >> + EVP_DigestUpdate(&ctx, message, len); >> + EVP_DigestFinal(&ctx, out, NULL); >> +} > > Guys, MD5 is not required for SSH so it MUST NOT be required for > libssh2! How does the patch change anything compared to before? That function is not used if md5 support is disabled. We can wrap the libssh2_md5 function in a '#if !LIBSSH2_MD5' if you want? It won't change anything as far as I can tell though. If the patch DID change anything in libssh2's behaviour, I'd like to understand what, because then I have misunderstood some logic in how the code worked. > I'm quite annoyed with some of these changes from Lars that have been > committed in a hurry. > > Don't get me wrong - we must make libssh2 work also for Lars, but we > should remember that his requirements seem to be fairly specialized > and I do not think it is acceptable for libssh2 to be optimized for, > or assume, such situations. I agree fully. Any such change is accidental. Libssh2 should be a standards-compliant implementation, and it should ideally warn the user if she tries to build it in any other way. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 22 23:44:22 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2MMi9RD027509; Mon, 22 Mar 2010 23:44:20 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2MMi8bh027488 for ; Mon, 22 Mar 2010 23:44:08 +0100 Date: Mon, 22 Mar 2010 23:44:08 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: sftp shutdown error with MessageWay In-Reply-To: <509c0c331003200501u2b57b108g13fbf33918234ef3@mail.gmail.com> Message-ID: References: <509c0c331003200501u2b57b108g13fbf33918234ef3@mail.gmail.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Sat, 20 Mar 2010, Matt Wixson wrote: > The file is transferred completely, the file size on the client matches the > file size on the server, but the process hangs for several minutes and then > I receive an error from close. I know this isn't a curl list but the curl > output happens to be: > libssh2_sftp_close: Resource temporarily unavailable (this is output from > perror after the call to libssh2_sftp_close()) > curl: (56) Failure when receiving data from the peer > > Here is the debug information from SSH2: > trace output here...I see the complete file transfer > > PROCESS HANGS FOR SEVERAL SECONDS What happens immediately before this "hang"? > == Info: SSH DISCONNECT starts now Instead of using libcurl for this, can you build a libssh2 sftp example program instead, and enable full debugging and show us one of those? That's the best kind of debug output we can get and we may actually spot something using such a log. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 18:52:33 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NHqFqM018261; Tue, 23 Mar 2010 18:52:30 +0100 Received: from minotaur.apache.org (minotaur-2.apache.org [140.211.11.10]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2NHqCQt018256 for ; Tue, 23 Mar 2010 18:52:13 +0100 Received: (qmail 30574 invoked by uid 99); 23 Mar 2010 17:52:11 -0000 Received: from localhost.apache.org (HELO mail-yw0-f201.google.com) (127.0.0.1) (smtp-auth username pquerna, mechanism plain) by minotaur.apache.org (qpsmtpd/0.29) with ESMTP; Tue, 23 Mar 2010 17:52:11 +0000 Received: by ywh39 with SMTP id 39so3371716ywh.21 for ; Tue, 23 Mar 2010 10:52:10 -0700 (PDT) MIME-Version: 1.0 Received: by 10.90.177.5 with SMTP id z5mr6145386age.64.1269366730243; Tue, 23 Mar 2010 10:52:10 -0700 (PDT) Date: Tue, 23 Mar 2010 17:52:10 +0000 Message-ID: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> Subject: [patch] use poll when available on blocking API From: Paul Querna To: libssh2-devel@cool.haxx.se Content-Type: multipart/mixed; boundary=00163630f01b0bc99804827b7981 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --00163630f01b0bc99804827b7981 Content-Type: text/plain; charset=UTF-8 Hello, We use libssh2 as part of Reconnoiter: Recently, we kept seeing libssh2 crashing once we passed a certain level of concurrency inside noitd, and eventually we traced it down to libssh2 using select, with sockets > FD_SETSIZE. The attached patch uses poll inside _libssh2_wait_socket, and fixes the issue for us. Ideally Reconnoiter will switch to the async API methods, but fixing this was quicker than rewriting the noitd module. Thanks, Paul --00163630f01b0bc99804827b7981 Content-Type: application/octet-stream; name="wait_socket.patch" Content-Disposition: attachment; filename="wait_socket.patch" Content-Transfer-Encoding: base64 X-Attachment-Id: f_g7505uao0 LS0tIHNlc3Npb24uYy5vcmlnaW5hbCAgMjAxMC0wMy0yMiAxODoyMTozMy4wMDAwMDAwMDAgLTA3 MDAKKysrIHNlc3Npb24uYyAgIDIwMTAtMDMtMjIgMTg6MjA6NTMuMDAwMDAwMDAwIC0wNzAwCkBA IC01MTgsMTEgKzUxOCwyOCBAQAogICovCiBpbnQgX2xpYnNzaDJfd2FpdF9zb2NrZXQoTElCU1NI Ml9TRVNTSU9OICpzZXNzaW9uKQogeworICAgIGludCByYzsKKyAgICBpbnQgZGlyOworI2lmIEhB VkVfUE9MTAorICAgIHN0cnVjdCBwb2xsZmQgc29ja2V0c1sxXTsKKyAgICBzb2NrZXRzWzBdLmZk ID0gc2Vzc2lvbi0+c29ja2V0X2ZkOworICAgIHNvY2tldHNbMF0uZXZlbnRzID0gMDsKKyAgICBz b2NrZXRzWzBdLnJldmVudHMgPSAwOworCisgICAgLyogbm93IG1ha2Ugc3VyZSB3ZSB3YWl0IGlu IHRoZSBjb3JyZWN0IGRpcmVjdGlvbiAqLworICAgIGRpciA9IGxpYnNzaDJfc2Vzc2lvbl9ibG9j a19kaXJlY3Rpb25zKHNlc3Npb24pOworICAgIAorICAgIGlmKGRpciAmIExJQlNTSDJfU0VTU0lP Tl9CTE9DS19JTkJPVU5EKQorICAgICAgc29ja2V0c1swXS5ldmVudHMgfD0gUE9MTElOOworICAg IAorICAgIGlmKGRpciAmIExJQlNTSDJfU0VTU0lPTl9CTE9DS19PVVRCT1VORCkKKyAgICAgIHNv Y2tldHNbMF0uZXZlbnRzIHw9IFBPTExPVVQ7CisgIAorICAgIHJjID0gcG9sbChzb2NrZXRzLCAx LCAtMSk7CisjZWxzZQogICAgIGZkX3NldCBmZDsKICAgICBmZF9zZXQgKndyaXRlZmQgPSBOVUxM OwogICAgIGZkX3NldCAqcmVhZGZkID0gTlVMTDsKLSAgICBpbnQgZGlyOwotICAgIGludCByYzsK IAogICAgIEZEX1pFUk8oJmZkKTsKICAgICBGRF9TRVQoc2Vzc2lvbi0+c29ja2V0X2ZkLCAmZmQp OwpAQCAtNTM5LDYgKzU1Niw3IEBACiAgICAgLyogTm90ZSB0aGF0IHRoaXMgQ09VTEQgYmUgbWFk ZSB0byB1c2UgYSB0aW1lb3V0IHRoYXQgcGVyaGFwcyBjb3VsZCBiZQogICAgICAgIGN1c3RvbWl6 YWJsZSBieSB0aGUgYXBwIG9yIHNvbWV0aGluZy4uLiAqLwogICAgIHJjID0gc2VsZWN0KHNlc3Np b24tPnNvY2tldF9mZCArIDEsIHJlYWRmZCwgd3JpdGVmZCwgTlVMTCwgTlVMTCk7CisjZW5kaWYK IAogICAgIGlmKHJjIDw9IDApIHsKICAgICAgICAgLyogdGltZW91dCAob3IgZXJyb3IpLCBiYWls IG91dCB3aXRoIGEgdGltZW91dCBlcnJvciAqLwo= --00163630f01b0bc99804827b7981 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --00163630f01b0bc99804827b7981-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 19:12:57 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NICrGC001704; Tue, 23 Mar 2010 19:12:57 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NICqBH001699 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 23 Mar 2010 19:12:52 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2NICtdI024768 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 23 Mar 2010 19:12:57 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100323:libssh2-devel@cool.haxx.se::oVH8riY2CnuSTkUV:LlHE Date: Tue, 23 Mar 2010 19:12:55 +0100 In-Reply-To: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> (Paul Querna's message of "Tue, 23 Mar 2010 17:52:10 +0000") Message-ID: <87sk7qsy8o.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Paul Querna writes: > Hello, > > We use libssh2 as part of Reconnoiter: > > > Recently, we kept seeing libssh2 crashing once we passed a certain > level of concurrency inside noitd, and eventually we traced it down to > libssh2 using select, with sockets > FD_SETSIZE. > > The attached patch uses poll inside _libssh2_wait_socket, and fixes > the issue for us. > > Ideally Reconnoiter will switch to the async API methods, but fixing > this was quicker than rewriting the noitd module. Thanks, I think the patch looks good. Objections from anyone else? /Simon > Thanks, > > Paul > > --- session.c.original 2010-03-22 18:21:33.000000000 -0700 > +++ session.c 2010-03-22 18:20:53.000000000 -0700 > @@ -518,11 +518,28 @@ > */ > int _libssh2_wait_socket(LIBSSH2_SESSION *session) > { > + int rc; > + int dir; > +#if HAVE_POLL > + struct pollfd sockets[1]; > + sockets[0].fd = session->socket_fd; > + sockets[0].events = 0; > + sockets[0].revents = 0; > + > + /* now make sure we wait in the correct direction */ > + dir = libssh2_session_block_directions(session); > + > + if(dir & LIBSSH2_SESSION_BLOCK_INBOUND) > + sockets[0].events |= POLLIN; > + > + if(dir & LIBSSH2_SESSION_BLOCK_OUTBOUND) > + sockets[0].events |= POLLOUT; > + > + rc = poll(sockets, 1, -1); > +#else > fd_set fd; > fd_set *writefd = NULL; > fd_set *readfd = NULL; > - int dir; > - int rc; > > FD_ZERO(&fd); > FD_SET(session->socket_fd, &fd); > @@ -539,6 +556,7 @@ > /* Note that this COULD be made to use a timeout that perhaps could be > customizable by the app or something... */ > rc = select(session->socket_fd + 1, readfd, writefd, NULL, NULL); > +#endif > > if(rc <= 0) { > /* timeout (or error), bail out with a timeout error */ > _______________________________________________ > libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 19:37:36 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NIbVdc010664; Tue, 23 Mar 2010 19:37:35 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NIbUtE010657 for ; Tue, 23 Mar 2010 19:37:30 +0100 Date: Tue, 23 Mar 2010 19:37:30 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [patch] use poll when available on blocking API In-Reply-To: <87sk7qsy8o.fsf@mocca.josefsson.org> Message-ID: References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 23 Mar 2010, Simon Josefsson wrote: >> The attached patch uses poll inside _libssh2_wait_socket, and fixes >> the issue for us. > > Thanks, I think the patch looks good. Objections from anyone else? Looks fine to me. There's of course a missing piece for configure to do the poll check - which reminds me that some systems have bad poll() implementations that shouldn't be used so the configure check can't be too simple. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 20:07:58 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NJ7ogp004860; Tue, 23 Mar 2010 20:07:57 +0100 Received: from mail-gx0-f210.google.com (mail-gx0-f210.google.com [209.85.217.210]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NJ6CEb003296 for ; Tue, 23 Mar 2010 20:06:13 +0100 Received: by gxk2 with SMTP id 2so1715847gxk.3 for ; Tue, 23 Mar 2010 12:06:13 -0700 (PDT) MIME-Version: 1.0 Received: by 10.91.152.12 with SMTP id e12mr738845ago.80.1269371173096; Tue, 23 Mar 2010 12:06:13 -0700 (PDT) In-Reply-To: References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> Date: Tue, 23 Mar 2010 19:06:13 +0000 Message-ID: <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> Subject: Re: [patch] use poll when available on blocking API From: Paul Querna To: libssh2 development X-Mailman-Approved-At: Tue, 23 Mar 2010 20:07:49 +0100 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se T24gVHVlLCBNYXIgMjMsIDIwMTAgYXQgNjozNyBQTSwgRGFuaWVsIFN0ZW5iZXJnIDxkYW5pZWxA aGF4eC5zZT4gd3JvdGU6Cj4gT24gVHVlLCAyMyBNYXIgMjAxMCwgU2ltb24gSm9zZWZzc29uIHdy b3RlOgo+Cj4+PiBUaGUgYXR0YWNoZWQgcGF0Y2ggdXNlcyBwb2xsIGluc2lkZSBfbGlic3NoMl93 YWl0X3NvY2tldCwgYW5kIGZpeGVzCj4+PiB0aGUgaXNzdWUgZm9yIHVzLgo+Pgo+PiBUaGFua3Ms IEkgdGhpbmsgdGhlIHBhdGNoIGxvb2tzIGdvb2QuIMKgT2JqZWN0aW9ucyBmcm9tIGFueW9uZSBl bHNlPwo+Cj4gTG9va3MgZmluZSB0byBtZS4gVGhlcmUncyBvZiBjb3Vyc2UgYSBtaXNzaW5nIHBp ZWNlIGZvciBjb25maWd1cmUgdG8gZG8gdGhlCj4gcG9sbCBjaGVjayAtIHdoaWNoIHJlbWluZHMg bWUgdGhhdCBzb21lIHN5c3RlbXMgaGF2ZSBiYWQgcG9sbCgpCj4gaW1wbGVtZW50YXRpb25zIHRo YXQgc2hvdWxkbid0IGJlIHVzZWQgc28gdGhlIGNvbmZpZ3VyZSBjaGVjayBjYW4ndCBiZSB0b28K PiBzaW1wbGUuCgpUaGUgY29uZmlndXJlLmFjIGFscmVhZHkgaGFzOgpBQ19DSEVDS19GVU5DUyhw b2xsIGdldHRpbWVvZmRheSBzZWxlY3Qgc3RydG9sbCkKClNvLCBIQVZFX1BPTEwgaXMgZGVmaW5l ZCBpZiBpdCBpcyBhdmFpbGFibGUsIGJ1dCB5ZXMgaXQgbGlrZWx5IG5lZWRzCnRvIGNvcHkgYSBi bGFja2xpc3QgZnJvbSBhbm90aGVyIHByb2plY3QuLi4uLi4uCl9fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fCmxpYnNzaDItZGV2ZWwgaHR0cDovL2Nvb2wuaGF4 eC5zZS9jZ2ktYmluL21haWxtYW4vbGlzdGluZm8vbGlic3NoMi1kZXZlbAo= From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 20:53:04 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NJqwOj030220; Tue, 23 Mar 2010 20:53:03 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NJqu7d030205 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 23 Mar 2010 20:52:56 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2NJqt2s027064 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 23 Mar 2010 20:52:57 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100323:libssh2-devel@cool.haxx.se::IlulvzVpDH5sF08B:1bMy Date: Tue, 23 Mar 2010 20:52:55 +0100 In-Reply-To: <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> (Paul Querna's message of "Tue, 23 Mar 2010 19:06:13 +0000") Message-ID: <87fx3qrf1k.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Paul Querna writes: > On Tue, Mar 23, 2010 at 6:37 PM, Daniel Stenberg wrote: >> On Tue, 23 Mar 2010, Simon Josefsson wrote: >> >>>> The attached patch uses poll inside _libssh2_wait_socket, and fixes >>>> the issue for us. >>> >>> Thanks, I think the patch looks good. =A0Objections from anyone else? >> >> Looks fine to me. There's of course a missing piece for configure to do = the >> poll check - which reminds me that some systems have bad poll() >> implementations that shouldn't be used so the configure check can't be t= oo >> simple. > > The configure.ac already has: > AC_CHECK_FUNCS(poll gettimeofday select strtoll) Right, I pushed your patch. > So, HAVE_POLL is defined if it is available, but yes it likely needs > to copy a blacklist from another project....... Let's hope it doesn't make the libssh2 code too ugly... In other projects I use gnulib to provide replacement functions for broken systems, so the core code remains nice and clean, but libssh2's BSD license makes it difficult to use here. Still, if we need something like that, maybe we can adapt a similar approach, if possible. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 21:12:39 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NKCVGX013980; Tue, 23 Mar 2010 21:12:38 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NKCTCm013977 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 23 Mar 2010 21:12:29 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2NKCXkc027712 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 23 Mar 2010 21:12:35 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100323:libssh2-devel@cool.haxx.se::bs00HzVfxmr84UXL:0U1p Date: Tue, 23 Mar 2010 21:12:33 +0100 In-Reply-To: <87fx3qrf1k.fsf@mocca.josefsson.org> (Simon Josefsson's message of "Tue, 23 Mar 2010 20:52:55 +0100") Message-ID: <878w9ire4u.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Simon Josefsson writes: > Paul Querna writes: > >> On Tue, Mar 23, 2010 at 6:37 PM, Daniel Stenberg wrote: >>> On Tue, 23 Mar 2010, Simon Josefsson wrote: >>> >>>>> The attached patch uses poll inside _libssh2_wait_socket, and fixes >>>>> the issue for us. >>>> >>>> Thanks, I think the patch looks good. =A0Objections from anyone else? >>> >>> Looks fine to me. There's of course a missing piece for configure to do= the >>> poll check - which reminds me that some systems have bad poll() >>> implementations that shouldn't be used so the configure check can't be = too >>> simple. >> >> The configure.ac already has: >> AC_CHECK_FUNCS(poll gettimeofday select strtoll) > > Right, I pushed your patch. Hmm, the self test is hanging in poll when tearing down the channel now. Is there some bad interaction between poll and a closed socket? Paul, can you take a look? Try 'make check'. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:06:39 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NL6UFO016384; Tue, 23 Mar 2010 22:06:36 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NL6S5H016331 for ; Tue, 23 Mar 2010 22:06:28 +0100 Date: Tue, 23 Mar 2010 22:06:28 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [patch] use poll when available on blocking API In-Reply-To: <87fx3qrf1k.fsf@mocca.josefsson.org> Message-ID: References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 23 Mar 2010, Simon Josefsson wrote: >> So, HAVE_POLL is defined if it is available, but yes it likely needs >> to copy a blacklist from another project....... > > Let's hope it doesn't make the libssh2 code too ugly... I just pushed a fix for configure that skips the poll check on two systems with known problematic poll implementations. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:14:21 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLEDBw019717; Tue, 23 Mar 2010 22:14:20 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLEBuQ019712 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 23 Mar 2010 22:14:11 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2NLEFRm028842 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 23 Mar 2010 22:14:17 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100323:libssh2-devel@cool.haxx.se::rppY7v+dj+/mSAtD:7vxS Date: Tue, 23 Mar 2010 22:14:15 +0100 In-Reply-To: (Daniel Stenberg's message of "Tue, 23 Mar 2010 22:06:28 +0100 (CET)") Message-ID: <878w9ipwpk.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Tue, 23 Mar 2010, Simon Josefsson wrote: > >>> So, HAVE_POLL is defined if it is available, but yes it likely needs >>> to copy a blacklist from another project....... >> >> Let's hope it doesn't make the libssh2 code too ugly... > > I just pushed a fix for configure that skips the poll check on two > systems with known problematic poll implementations. Great. I take it that 'darwin' doesn't refer to modern Mac OS X systems? The pattern 'darwin*' doesn't match a Mac I have access to, which is called powerpc-apple-darwin8.11.0 /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:16:22 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLGLO6020653; Tue, 23 Mar 2010 22:16:22 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLGJBr020632 for ; Tue, 23 Mar 2010 22:16:19 +0100 Date: Tue, 23 Mar 2010 22:16:19 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [patch] use poll when available on blocking API In-Reply-To: <878w9ipwpk.fsf@mocca.josefsson.org> Message-ID: References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 23 Mar 2010, Simon Josefsson wrote: >> I just pushed a fix for configure that skips the poll check on two systems >> with known problematic poll implementations. > > Great. I take it that 'darwin' doesn't refer to modern Mac OS X systems? > The pattern 'darwin*' doesn't match a Mac I have access to, which is called > powerpc-apple-darwin8.11.0 Even fairly modern Mac OS X machines have broken poll(), and I'm not even sure the most modern ones have that fixed. We did have an Apple guy educate us on these matters within the curl project which made us use that pattern... -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:22:23 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLMIOi025617; Tue, 23 Mar 2010 22:22:22 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLMH2k025552 for ; Tue, 23 Mar 2010 22:22:17 +0100 Date: Tue, 23 Mar 2010 22:22:17 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [patch] use poll when available on blocking API In-Reply-To: Message-ID: References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 23 Mar 2010, Daniel Stenberg wrote: > Even fairly modern Mac OS X machines have broken poll(), and I'm not even > sure the most modern ones have that fixed. We did have an Apple guy educate > us on these matters within the curl project which made us use that > pattern... In other words, I think I did the pattern matching wrong. I copied it from curl's configure.ac but apparently that has a different variable checked. Maybe we should just use "*darwin*|*interix*)" to play safe? BTW http://lists.apple.com/archives/Darwin-userlevel/2009/Dec/msg00004.html seems to indicate that Mac OS X still has a broken poll as of 10.5.8 which I believe is rather recent. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:29:17 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLTEc0001762; Tue, 23 Mar 2010 22:29:17 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLTDT7001745 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 23 Mar 2010 22:29:13 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2NLTIWs029165 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 23 Mar 2010 22:29:19 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100323:libssh2-devel@cool.haxx.se::tlerWOhqOFxRG9GT:5xRp Date: Tue, 23 Mar 2010 22:29:18 +0100 In-Reply-To: (Daniel Stenberg's message of "Tue, 23 Mar 2010 22:16:19 +0100 (CET)") Message-ID: <87tys6ohg1.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Tue, 23 Mar 2010, Simon Josefsson wrote: > >>> I just pushed a fix for configure that skips the poll check on two >>> systems with known problematic poll implementations. >> >> Great. I take it that 'darwin' doesn't refer to modern Mac OS X >> systems? The pattern 'darwin*' doesn't match a Mac I have access to, >> which is called powerpc-apple-darwin8.11.0 > > Even fairly modern Mac OS X machines have broken poll(), and I'm not > even sure the most modern ones have that fixed. We did have an Apple > guy educate us on these matters within the curl project which made us > use that pattern... The pattern used by libssh2 now (i.e., 'darwin*') will not match modern Mac OS X systems, I believe. Do you use 'darwin*' or '*darwin*' in curl? I'm only aware of these problems on Mac OS X: http://www.gnu.org/software/gnulib/manual/html_node/poll.html /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:35:49 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLZkY6005255; Tue, 23 Mar 2010 22:35:48 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLZi1P005249 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 23 Mar 2010 22:35:44 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2NLZne9029264 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Tue, 23 Mar 2010 22:35:50 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100323:libssh2-devel@cool.haxx.se::aXoXm229RL+JZW2n:BuNy Date: Tue, 23 Mar 2010 22:35:49 +0100 In-Reply-To: (Daniel Stenberg's message of "Tue, 23 Mar 2010 22:22:17 +0100 (CET)") Message-ID: <87pr2uoh56.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > BTW > http://lists.apple.com/archives/Darwin-userlevel/2009/Dec/msg00004.html > seems to indicate that Mac OS X still has a broken poll as of 10.5.8 > which I believe is rather recent. Hm, as far as I can tell, Daniel Mack says that poll ignores the timeout argument? His bug reports are not publicly visible, or at least I cannot figure out how to view them without logging in. O well, anyway, I don't care particulary about Mac OS X, so I don't care what we do here, and instead I expect that Mac OS X people will send us bug reports and hopefully patches eventually. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 23 22:48:54 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLmnXK017319; Tue, 23 Mar 2010 22:48:53 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2NLmlBp017314 for ; Tue, 23 Mar 2010 22:48:47 +0100 Date: Tue, 23 Mar 2010 22:48:47 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [patch] use poll when available on blocking API In-Reply-To: <87pr2uoh56.fsf@mocca.josefsson.org> Message-ID: References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> <87pr2uoh56.fsf@mocca.josefsson.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 23 Mar 2010, Simon Josefsson wrote: I dug around in old mail archives to get some specifics to what we're trying to avoid. It might be so that modern Mac OS X has a poll() that is good enough to be used even though poll() still has serious quirks on that OS even though we may never touch those. However, older versions of poll() was worse and in general the function has a rather bad track record so I think it's safer to just avoid it on macs. > Hm, as far as I can tell, Daniel Mack says that poll ignores the timeout > argument? Yes, and that goes with older problems we've seen in curl when Mac OS ignored the timeout argument. Some versions of poll() on Mac OS failed hard when we used it with no file descriptors (basically doing a sub-second sleep). It doesn't work on files in /dev. Earlier versions have an "inconsistent return of POLLHUP where other implementations give POLLIN". -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 24 08:39:19 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2O7d76h019408; Wed, 24 Mar 2010 08:39:17 +0100 Received: from yxa-v.extundo.com (yxa-v.extundo.com [83.241.177.39]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2O7d5mA019388 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 24 Mar 2010 08:39:06 +0100 Received: from mocca (c80-216-24-99.bredband.comhem.se [80.216.24.99]) (authenticated bits=0) by yxa-v.extundo.com (8.14.3/8.14.3/Debian-5+lenny1) with ESMTP id o2O7d4RZ009141 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT) for ; Wed, 24 Mar 2010 08:39:06 +0100 From: Simon Josefsson To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> <87pr2uoh56.fsf@mocca.josefsson.org> OpenPGP: id=B565716F; url=http://josefsson.org/key.txt X-Hashcash: 1:22:100324:libssh2-devel@cool.haxx.se::e6WQTDH63wIFe5Ou:4JR Date: Wed, 24 Mar 2010 08:39:04 +0100 In-Reply-To: (Daniel Stenberg's message of "Tue, 23 Mar 2010 22:48:47 +0100 (CET)") Message-ID: <87r5namanb.fsf@mocca.josefsson.org> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-Spam-Status: No, score=-0.9 required=5.0 tests=AWL,BAYES_00, DATE_IN_FUTURE_96_XX,RDNS_DYNAMIC,SPF_FAIL autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on yxa-v.extundo.com X-Virus-Scanned: clamav-milter 0.95.3 at yxa-v X-Virus-Status: Clean X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Daniel Stenberg writes: > On Tue, 23 Mar 2010, Simon Josefsson wrote: > > I dug around in old mail archives to get some specifics to what we're > trying to avoid. It might be so that modern Mac OS X has a poll() that > is good enough to be used even though poll() still has serious quirks > on that OS even though we may never touch those. However, older > versions of poll() was worse and in general the function has a rather > bad track record so I think it's safer to just avoid it on macs. > >> Hm, as far as I can tell, Daniel Mack says that poll ignores the timeout >> argument? > > Yes, and that goes with older problems we've seen in curl when Mac OS > ignored the timeout argument. Some versions of poll() on Mac OS failed > hard when we used it with no file descriptors (basically doing a > sub-second sleep). It doesn't work on files in /dev. Earlier versions > have an "inconsistent return of POLLHUP where other implementations > give POLLIN". Thanks for info -- I'll see if I can reproduce any of it on my system and improve the gnulib documentation about system status. /Simon _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 24 08:59:16 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2O7xDai001405; Wed, 24 Mar 2010 08:59:16 +0100 Received: from susano.oderland.com (susano.oderland.com [91.201.63.143]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2O7xBme001372 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 24 Mar 2010 08:59:11 +0100 Received: from h215n1fls309o1003.telia.com ([81.224.184.215] helo=stajlis.springfield.se) by susano.oderland.com with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.69) (envelope-from ) id 1NuLUa-0000Yz-Tm for libssh2-devel@cool.haxx.se; Wed, 24 Mar 2010 08:59:09 +0100 Message-ID: <4BA9C64A.90609@Strombergson.com> Date: Wed, 24 Mar 2010 08:59:06 +0100 From: =?UTF-8?B?Sm9hY2hpbSBTdHLDtm1iZXJnc29u?= Organization: Kryptologik User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2 development Subject: Re: [patch] use poll when available on blocking API References: <4239a4321003231052n5b217439l8428c34434b59302@mail.gmail.com> <87sk7qsy8o.fsf@mocca.josefsson.org> <4239a4321003231206p29bf6f75jda6de110714f4d26@mail.gmail.com> <87fx3qrf1k.fsf@mocca.josefsson.org> <878w9ipwpk.fsf@mocca.josefsson.org> In-Reply-To: X-Enigmail-Version: 1.0.1 X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - susano.oderland.com X-AntiAbuse: Original Domain - cool.haxx.se X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - Strombergson.com X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: Joachim@strombergson.com, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se LS0tLS1CRUdJTiBQR1AgU0lHTkVEIE1FU1NBR0UtLS0tLQpIYXNoOiBTSEExCgpBbG9oYSEKCk9s ZCBsdXJrZXIgaGVyZSBqdW1waW5nIGluLgoKT24gMjAxMC0wMy0yMyAyMi4yMiwgRGFuaWVsIFN0 ZW5iZXJnIHdyb3RlOgo+IE9uIFR1ZSwgMjMgTWFyIDIwMTAsIERhbmllbCBTdGVuYmVyZyB3cm90 ZToKPiAKPj4gRXZlbiBmYWlybHkgbW9kZXJuIE1hYyBPUyBYIG1hY2hpbmVzIGhhdmUgYnJva2Vu IHBvbGwoKSwgYW5kIEknbSBub3QKPj4gZXZlbiBzdXJlIHRoZSBtb3N0IG1vZGVybiBvbmVzIGhh dmUgdGhhdCBmaXhlZC4gV2UgZGlkIGhhdmUgYW4gQXBwbGUKPj4gZ3V5IGVkdWNhdGUgdXMgb24g dGhlc2UgbWF0dGVycyB3aXRoaW4gdGhlIGN1cmwgcHJvamVjdCB3aGljaCBtYWRlIHVzCj4+IHVz ZSB0aGF0IHBhdHRlcm4uLi4KPiAKPiBJbiBvdGhlciB3b3JkcywgSSB0aGluayBJIGRpZCB0aGUg cGF0dGVybiBtYXRjaGluZyB3cm9uZy4gSSBjb3BpZWQgaXQKPiBmcm9tIGN1cmwncyBjb25maWd1 cmUuYWMgYnV0IGFwcGFyZW50bHkgdGhhdCBoYXMgYSBkaWZmZXJlbnQgdmFyaWFibGUKPiBjaGVj a2VkLiBNYXliZSB3ZSBzaG91bGQganVzdCB1c2UgIipkYXJ3aW4qfCppbnRlcml4KikiIHRvIHBs YXkgc2FmZT8KPiAKPiBCVFcKPiBodHRwOi8vbGlzdHMuYXBwbGUuY29tL2FyY2hpdmVzL0Rhcndp bi11c2VybGV2ZWwvMjAwOS9EZWMvbXNnMDAwMDQuaHRtbAo+IHNlZW1zIHRvIGluZGljYXRlIHRo YXQgTWFjIE9TIFggc3RpbGwgaGFzIGEgYnJva2VuIHBvbGwgYXMgb2YgMTAuNS44Cj4gd2hpY2gg SSBiZWxpZXZlIGlzIHJhdGhlciByZWNlbnQuCgpJIHRyaWVkIHRoZSBzbmlwcGV0IGluIHRoZSBi dWcgcmVwb3J0IGJ5IERhbmllbCBNYWNrIGFuZCBvbiBteSBtYWNoaW5lCmF0IGxlYXN0IGl0IGRv ZXMgbm90IGhhbmcsIGJ1dCBjb21wbGV0ZXMgZGlyZWN0bHkuCgpNeSBzeXN0ZW0gaXMgYToKanNA c3Rhamxpcy5zcHJpbmdmaWVsZC5zZTovVXNlcnMvanMvdG1wPnVuYW1lIC1hCkRhcndpbiBzdGFq bGlzLnNwcmluZ2ZpZWxkLnNlIDEwLjIuMCBEYXJ3aW4gS2VybmVsIFZlcnNpb24gMTAuMi4wLi4u CgpBbmQgdGhlICJBYm91dCB0aGlzIE1hYyIgc3RhdGVzIHRoYXQgaXQgaXMgU25vdyBMZW9wYXJk IDEwLjYuMgoKSWYgRGFuaWVscyB0ZXN0IGNhc2UgaXMgYW55IHByb29mLCBpdCBzZWVtcyBwb2xs KCkgaXMgZml4ZWQgaW4gMTAuNiBhbmQKb253YXJkcy4gQXBwbGUsIHVuZm9ydHVuYXRlbHkgaXMg bm90IHJlYWxseSBhIGZvdW50YWluIG9mIGluZm9ybWF0aW9uCnJlZ2FyZGluZyB3aGF0IGhhcyBi ZWVuIGZpeGVkIGluIGEgcmVsZWFzZS4KCklzIHRoZXJlIHNvbWUgdGVzdCB5b3UgbmVlZCB0byBn ZXQgcGVyZm9ybWVkIHRvIGNsZWFyIHRoaXMgSSdsbCBiZSBoYXBweQp0byB0cnkgYW5kIGhlbHAu CgotIC0tIApNZWQgdsOkbmxpZyBow6Rsc25pbmcsIFlvdXJzCgpKb2FjaGltIFN0csO2bWJlcmdz b24gLSBBbGx0aWQgaSBoYXJtb25pc2sgc3bDpG5nbmluZy4KPT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09CktyeXB0 b2Jsb2cgLSBJVC1zw6RrZXJoZXQgcMOlIHN2ZW5za2EKaHR0cDovL3d3dy5zdHJvbWJlcmdzb24u Y29tL2tyeXB0b2Jsb2cKPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLS0tQkVHSU4gUEdQIFNJR05BVFVSRS0t LS0tClZlcnNpb246IEdudVBHIHYxLjQuOSAoRGFyd2luKQpDb21tZW50OiBVc2luZyBHbnVQRyB3 aXRoIE1vemlsbGEgLSBodHRwOi8vZW5pZ21haWwubW96ZGV2Lm9yZy8KCmlFWUVBUkVDQUFZRkFr dXB4a29BQ2drUVpvUHI4SFQzMFFIUXB3Q2dzdmNMbmw4YjFsc29rSjR5cFhCVk51QmsKWFE0QW5S bWxxZHk4V0R0TExja2ozYnRjUjVCTXJ2VGoKPVFXWU4KLS0tLS1FTkQgUEdQIFNJR05BVFVSRS0t LS0tCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fCmxpYnNz aDItZGV2ZWwgaHR0cDovL2Nvb2wuaGF4eC5zZS9jZ2ktYmluL21haWxtYW4vbGlzdGluZm8vbGli c3NoMi1kZXZlbAo= From libssh2-devel-bounces@cool.haxx.se Wed Mar 24 17:54:39 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OGsNCj005820; Wed, 24 Mar 2010 17:54:35 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OGsMf0005817 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 24 Mar 2010 17:54:23 +0100 Received: (qmail 13306 invoked from network); 24 Mar 2010 16:54:28 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 24 Mar 2010 16:54:28 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 24 Mar 2010 16:54:28 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #166: Some servers terminate connection due to large window adjustment X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/166 Message-ID: <040.e4d863057078a2a001d9e278040bea57@libssh2.stuge.se> X-Trac-Ticket-ID: 166 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #166: Some servers terminate connection due to large window adjustment ---------------------------+------------------------------------------------ Reporter: ptjm | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: protocol | Version: 1.2.4 Keywords: window adjust | Blocking: Blockedby: | ---------------------------+------------------------------------------------ I've seen this problem connecting to a few different servers. When libssh2 sends a window adjust message, the server shuts down the connection. Here's an excerpt from a trace: [libssh2] 0.784404 Transport: Received Banner: SSH-2.0-3.2.9 SSH Secure Shell [...] [libssh2] 2.129966 SFTP: Asking for 102 packet [libssh2] 2.129983 SFTP: Asking for 101 packet [libssh2] 2.129997 SFTP: recv packet [libssh2] 2.130011 Conn: channel_read() wants 4 bytes from channel 0/0 stream #0 => libssh2_transport_read() raw (120 bytes) 0000: f8 05 11 5d 9f e1 71 35 d9 20 64 de 47 c4 10 5f ...]..q5. d.G.._ 0010: 62 43 89 c2 81 04 78 59 76 d3 34 2e a0 27 f1 47 bC....xYv.4..'.G 0020: 3b ef b1 84 0b 40 3e b3 8f ef de 2b 02 58 f8 f5 ;....@>....+.X.. 0030: 37 6e 1c ff fe ed 53 a3 c5 60 61 8a 30 65 88 e5 7n....S..`a.0e.. 0040: 48 2d 0d c6 6d 2a 91 a5 8d fa 8a d9 6c c5 02 09 H-..m*......l... 0050: 67 22 4d d1 11 8c 80 e9 63 e0 84 fc b2 ea 26 73 g"M.....c.....&s 0060: 69 4d 72 7c 30 0c ec d1 80 85 bf b0 41 37 b8 d7 iMr|0.......A7.. 0070: 9a 58 b8 4a 4c f3 7d b2 .X.JL.}. => libssh2_transport_read() plain (5 bytes) 0000: 02 00 00 00 00 ..... [libssh2] 2.132312 Transport: Packet type 2 received, length=5 => libssh2_transport_read() plain (48 bytes) 0000: 01 00 00 00 02 00 00 00 21 42 61 64 20 77 69 6e ........!Bad win 0010: 64 6f 77 20 73 69 7a 65 20 69 6e 20 57 49 4e 44 dow size in WIND 0020: 4f 57 5f 41 44 4a 55 53 54 2e 00 00 00 02 65 6e OW_ADJUST.....en [libssh2] 2.133175 Transport: Packet type 1 received, length=48 [libssh2] 2.133190 Transport: Disconnect(2): Bad window size in WINDOW_ADJUST.(e n) [libssh2] 2.133206 Failure Event: -13 - socket disconnect [libssh2] 2.133221 Failure Event: -13 - tranport read [libssh2] 2.133235 Failure Event: -21 - Read part of packet [libssh2] 2.133248 Failure Event: -1 - Timeout waiting for status message I found that decreasing the window size by a factor of 10 avoids this problem. It's not clear to me whether this change has any performance impact, but, e.g., openssh gets by with a maximum adjustment size of 128kb. What is clear to me is that with this change, libssh2 can connect to several servers, while without it, it can't. I'm attaching a diff against channel.c from 1.2.4 -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 24 18:25:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OHP0CW006369; Wed, 24 Mar 2010 18:25:05 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OHOwmx006097 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 24 Mar 2010 18:24:58 +0100 Received: (qmail 14857 invoked from network); 24 Mar 2010 17:25:04 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 24 Mar 2010 17:25:04 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 24 Mar 2010 17:25:04 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #167: Unable to init SFTP session vs SSH-2.0-SSHD X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/167 Message-ID: <040.b3944a7034e0353dd801d2508d9c873d@libssh2.stuge.se> X-Trac-Ticket-ID: 167 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #167: Unable to init SFTP session vs SSH-2.0-SSHD --------------------+------------------------------------------------------- Reporter: ptjm | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: SFTP | Version: 1.2.5 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- I found that libssh2 is able to authenticate versus one particular sftp server, but that it fails to initiate the sftp session. The attached file sftpfailure.log shows the trace of sftpdir vs a server which returns the banner "SSH-2.0-SSHD". The trace was run using version 1.2.5-20100317. There were no modifications to the library but sftpdir.c was modified to use public key authentication and to call libssh2_trace(session, LIBSSH2_TRACE_TRANS | LIBSSH2_TRACE_KEX | LIBSSH2_TRACE_AUTH | LIBSSH2_TRACE_CONN | LIBSSH2_TRACE_SFTP | LIBSSH2_TRACE_ERROR); The only other thing I can think to add is that the machine I'm connecting from is rather old & the openssl version is 0.9.7d. I am able to connect from this machine and perform sftp transfers using openssh, though. I know nothing about the server machine. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 24 23:16:09 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OMFtSF029544; Wed, 24 Mar 2010 23:16:07 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OMFr0x029539 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 24 Mar 2010 23:15:53 +0100 Received: (qmail 28829 invoked from network); 24 Mar 2010 22:16:00 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 24 Mar 2010 22:16:00 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 24 Mar 2010 22:16:00 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #166: Some servers terminate connection due to large window adjustment X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/166#comment:1 Message-ID: <049.5b1106677bd85b0cc8c74f1f3432cbb3@libssh2.stuge.se> References: <040.e4d863057078a2a001d9e278040bea57@libssh2.stuge.se> X-Trac-Ticket-ID: 166 In-Reply-To: <040.e4d863057078a2a001d9e278040bea57@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #166: Some servers terminate connection due to large window adjustment -----------------------+---------------------------------------------------- Reporter: ptjm | Owner: Type: defect | Status: closed Priority: normal | Milestone: 1.2.5 Component: protocol | Version: 1.2.4 Resolution: fixed | Keywords: window adjust Blocking: | Blockedby: -----------------------+---------------------------------------------------- Changes (by bagder): * status: new => closed * resolution: => fixed * milestone: => 1.2.5 Comment: Thanks, as we've got other reports too on problems with that large maximum window size I think it is only fair that we lower it as you suggest. Change committed just now! -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 24 23:28:47 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OMSf4T002839; Wed, 24 Mar 2010 23:28:45 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2OMSeYB002835 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 24 Mar 2010 23:28:40 +0100 Received: (qmail 29358 invoked from network); 24 Mar 2010 22:28:47 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 24 Mar 2010 22:28:47 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Wed, 24 Mar 2010 22:28:47 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #167: Unable to init SFTP session vs SSH-2.0-SSHD X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/167#comment:1 Message-ID: <049.ea00207dc50fd4f66c39fca55508abbb@libssh2.stuge.se> References: <040.b3944a7034e0353dd801d2508d9c873d@libssh2.stuge.se> X-Trac-Ticket-ID: 167 In-Reply-To: <040.b3944a7034e0353dd801d2508d9c873d@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #167: Unable to init SFTP session vs SSH-2.0-SSHD --------------------+------------------------------------------------------- Reporter: ptjm | Owner: bagder Type: defect | Status: assigned Priority: normal | Milestone: Component: SFTP | Version: 1.2.5 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- Changes (by bagder): * owner: => bagder * status: new => assigned Comment: I don't think it has anything to do with the openssl version. At line 460 in your log file, we can see how the error at code line src/sftp.c:629 is reached. The question is what 'rc' contains at that point! Also, as can be seen in that function there's a bad assumption that _libssh2_channel_write() will either return PACKET_EAGAIN, send the entire packet or fail. There's also the chance that it simply sent a piece of the packet and thus have to send the rest of the packet in a second try. I suspect that is what happening for you. Seeing the error code as I mention above, might help us verify this. If that is indeed the case (and really, even so if it isn't), then we should fix this function by keeping "number of bytes already sent" in a variable and allow the state machine to get back and send the rest of the packet at next oppurtunity. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 01:19:51 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2P0Jgpn013660; Thu, 25 Mar 2010 01:19:50 +0100 Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.27]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2P0Jeoe013597 for ; Thu, 25 Mar 2010 01:19:41 +0100 Received: by ey-out-2122.google.com with SMTP id d26so494665eyd.7 for ; Wed, 24 Mar 2010 17:19:49 -0700 (PDT) Received: by 10.213.50.84 with SMTP id y20mr6127194ebf.71.1269476388807; Wed, 24 Mar 2010 17:19:48 -0700 (PDT) Received: from [192.168.0.35] (96-160.ftth.onsbrabantnet.nl [88.159.160.96]) by mx.google.com with ESMTPS id 14sm423431ewy.2.2010.03.24.17.19.47 (version=SSLv3 cipher=RC4-MD5); Wed, 24 Mar 2010 17:19:48 -0700 (PDT) Message-ID: <4BAAAC22.6010307@peut.org> Date: Thu, 25 Mar 2010 01:19:46 +0100 From: Jose Baars User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; nl; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: [libssh2] #166: Some servers terminate connection due to large window adjustment References: <040.e4d863057078a2a001d9e278040bea57@libssh2.stuge.se> <049.5b1106677bd85b0cc8c74f1f3432cbb3@libssh2.stuge.se> In-Reply-To: <049.5b1106677bd85b0cc8c74f1f3432cbb3@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1646659952==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a cryptographically signed message in MIME format. --===============1646659952== Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms000200010702050303010703" This is a cryptographically signed message in MIME format. --------------ms000200010702050303010703 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Op 3/24/2010 11:16 PM, libssh2 Trac schreef: Thanks, I have been struggling with that too. According to several sources on the web, for instance : http://www.psc.edu/networking/projects/hpn-ssh/theory.php there is no use to enlarge the SSH protocol window above the TCP/IP window size of the socket. Problem is that I found it is not easy to find out what the socket=20 window size is for a given socket, getsockopt has no SO_WINDOWSIZE or whatever. Apparently, according to windows docs at=20 http://msdn.microsoft.com/en-us/library/ms819736.aspx , the maximum window size reachable is 1 Gb. On the other hand RFC 4254 (SSH connection protocol) states in paragraph = 5.2: =2E.. After receiving this message, the recipient MAY send the given number of bytes more than it was previously allowed to send; the window size= is incremented. Implementations MUST correctly handle window sizes of up to 2^32 - 1 bytes. The window MUST NOT be increased above 2^32 - 1 bytes. =2E.. The TCP/IP services VMS ssh server has a hard coded maximum window size of 16 Mb, which is obviously not in accordance with RFC4254, but then aga= in, there is no specific guarantee or promise of it adhering to this or any o= ther RFC. The quick fix is the division by 10 and wouldn't break anything, although= performance on really fast connections with RFC compliant SSH servers (any out there?= ) might suffer a performance impact. Jose --------------ms000200010702050303010703 Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIPWjCC BMwwggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1 MTAyNzIzNTk1OVowgd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEf MB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNl IGF0IGh0dHBzOi8vd3d3LnZlcmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNv bmEgTm90IFZhbGlkYXRlZDE3MDUGA1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFs IFN1YnNjcmliZXIgQ0EgLSBHMjCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnf rOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyVzm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs +Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zplYu//EHuiVrvFTnAt1qIfPO2wQuhejVch rKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFBL2OyOj++pRpu9MlKWz2VphW7NQIZ +dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5gJ925rXXOL3OVekA6hXVJsLjf aLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUCAwEAAaOCAYQwggGAMBIG A1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXATAqMCgGCCsGAQUF BwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIBBjARBglghkgB hvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJlbDMtMjA0 OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAkoCKG IGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEg UHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+Iqyz cqpVMA0GCSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESf D0b3+qD+0x0Yo9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Pr v4NZmP1m3umGMpqSKTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFQTCCBCmgAwIB AgIQQkuBnab/mKbwbYynoi4GkzANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAV BgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3Jr MTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJp U2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyMB4XDTA5MTExMjAw MDAwMFoXDTEwMTExMjIzNTk1OVowggEKMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0G A1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNv bS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElBQi5MVEQoYyk5ODEeMBwGA1UE CxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdpdGFsIElEIENsYXNzIDEg LSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxEzARBgNVBAMUCkpvc2UgQmFhcnMxHDAaBgkqhkiG 9w0BCQEWDXBldXRAcGV1dC5vcmcwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCi x8yI3kFLmfYElzZSqsup782ajeuv3TASyn8mUfGrnr2aW1M7jN4vlM8/7/sVL883rH+kWPzV Q4ITYrfW5QFxH+CBqtX0F1xsS1A7AR8xZIs8o7/sMxV2xOSXYEJXlKP4XVWY8oU+TKH5dWYZ 9BJ9pkJ840ayiqqI/P3MHqmu7vS2nPDu8YZFa+46EtWkGbJTjzFdnrU5Jmdudp37fd1btNss RIkqztDOeQOTaZ6pS9RqjhmNuWVtgZucKho6c9Qf2htGz8AL95MqttF2OZYL9AlXZhjJcIMO aZBVjuwLowX/kYL2lo1vyUGt3eZxSkZR/ICfkQwj1BKpFZGFtM5XAgMBAAGjgcwwgckwCQYD VR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUF BwMEBggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQt Y3JsLnZlcmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEB ABcuOD4QwMrXJgD8zvCWBRjBqsQLXn0ybqrLZHlcic7+lstAl06fi7fGD3+1Nrcp4VImkyUv JQEvFSvkB+/Mt8kOVSdt+eiYIInKt/lBkSzy0sardsh0u7Ifcgt+nmLZYW+azviXnD4VNMqe Q9QIZPHSUhWDe3IA5NSYE4hfqwXX5c13HllqpUcXRmWIGA9MPr9SQxZ0jg7MFUsZVgG9f5Ts A3yIEmvUgutyX6GRzZQL+wlWymRzh/XA7j2QnYk+IPlNQHJ62MHHlAI48y3ssSXwR8hfbH4e 0sw+8ezuh785OuFMBI3qhxlQX4wT8+v8rKHtiLcddoIohWM0ID+IdOEwggVBMIIEKaADAgEC AhBCS4Gdpv+YpvBtjKeiLgaTMA0GCSqGSIb3DQEBBQUAMIHdMQswCQYDVQQGEwJVUzEXMBUG A1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdvcmsx OzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBh IChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZlcmlT aWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzIwHhcNMDkxMTEyMDAw MDAwWhcNMTAxMTEyMjM1OTU5WjCCAQoxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxMzAxBgNVBAsTKkRpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlIEZ1bGwgU2VydmljZTETMBEGA1UEAxQKSm9zZSBCYWFyczEcMBoGCSqGSIb3 DQEJARYNcGV1dEBwZXV0Lm9yZzCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKLH zIjeQUuZ9gSXNlKqy6nvzZqN66/dMBLKfyZR8auevZpbUzuM3i+Uzz/v+xUvzzesf6RY/NVD ghNit9blAXEf4IGq1fQXXGxLUDsBHzFkizyjv+wzFXbE5JdgQleUo/hdVZjyhT5Mofl1Zhn0 En2mQnzjRrKKqoj8/cweqa7u9Lac8O7xhkVr7joS1aQZslOPMV2etTkmZ252nft93Vu02yxE iSrO0M55A5NpnqlL1GqOGY25ZW2Bm5wqGjpz1B/aG0bPwAv3kyq20XY5lgv0CVdmGMlwgw5p kFWO7AujBf+RgvaWjW/JQa3d5nFKRlH8gJ+RDCPUEqkVkYW0zlcCAwEAAaOBzDCByTAJBgNV HRMEAjAAMEQGA1UdIAQ9MDswOQYLYIZIAYb4RQEHFwEwKjAoBggrBgEFBQcCARYcaHR0cHM6 Ly93d3cudmVyaXNpZ24uY29tL3JwYTALBgNVHQ8EBAMCBaAwHQYDVR0lBBYwFAYIKwYBBQUH AwQGCCsGAQUFBwMCMEoGA1UdHwRDMEEwP6A9oDuGOWh0dHA6Ly9JbmRDMURpZ2l0YWxJRC1j cmwudmVyaXNpZ24uY29tL0luZEMxRGlnaXRhbElELmNybDANBgkqhkiG9w0BAQUFAAOCAQEA Fy44PhDAytcmAPzO8JYFGMGqxAtefTJuqstkeVyJzv6Wy0CXTp+Lt8YPf7U2tynhUiaTJS8l AS8VK+QH78y3yQ5VJ2356Jggicq3+UGRLPLSxqt2yHS7sh9yC36eYtlhb5rO+JecPhU0yp5D 1Ahk8dJSFYN7cgDk1JgTiF+rBdflzXceWWqlRxdGZYgYD0w+v1JDFnSODswVSxlWAb1/lOwD fIgSa9SC63JfoZHNlAv7CVbKZHOH9cDuPZCdiT4g+U1AcnrYwceUAjjzLeyxJfBHyF9sfh7S zD7x7O6Hvzk64UwEjeqHGVBfjBPz6/ysoe2Itx12giiFYzQgP4h04TGCBOwwggToAgEBMIHy MIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZl cmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxp ZGF0ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVy IENBIC0gRzICEEJLgZ2m/5im8G2Mp6IuBpMwCQYFKw4DAhoFAKCCAs4wGAYJKoZIhvcNAQkD MQsGCSqGSIb3DQEHATAcBgkqhkiG9w0BCQUxDxcNMTAwMzI1MDAxOTQ2WjAjBgkqhkiG9w0B CQQxFgQUZ8AVBB2cyAxZVkaFnq7o9pY7nIMwXwYJKoZIhvcNAQkPMVIwUDALBglghkgBZQME AQIwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIH MA0GCCqGSIb3DQMCAgEoMIIBAwYJKwYBBAGCNxAEMYH1MIHyMIHdMQswCQYDVQQGEwJVUzEX MBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsTFlZlcmlTaWduIFRydXN0IE5ldHdv cmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczovL3d3dy52ZXJpc2lnbi5jb20v cnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxNzA1BgNVBAMTLlZl cmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0gRzICEEJLgZ2m/5im 8G2Mp6IuBpMwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTsw OQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAo YykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhBCS4Gdpv+YpvBtjKei LgaTMA0GCSqGSIb3DQEBAQUABIIBAGzhUvlPUZlWdwwjXk+L5BvLl60GUm9N1TtnOxUhH3O3 V6YMQ8BZ1Ws/ULZq/qHYYuPUFjysJSL9mTNXFxmJli1EnkGGv4RN1caywVib74fo4kfw7lXb a/UjZCYjc/tuJxQfG/nK/fdZsxr7ljzekMhnHDR9FXSBhTbWw/gwiL4am/r7hPvwcM7CswBl QytgZ9kXNl3U0hBCJ5RfEesmEnRVH1NcL52PBncS/DkfFj13HnZZ8HfzTZyQbTE26ddKx7T/ Y80jU5xKgCryCQd6Bn5oFOnfWlVeTPi6ZfZ6COVrC+7sOvouPjPIPul3RhTF3K6r4kRV1wmv L4YjEcnqgMUAAAAAAAA= --------------ms000200010702050303010703-- --===============1646659952== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1646659952==-- From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 05:31:18 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2P4V33X007828; Thu, 25 Mar 2010 05:31:16 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2P4V1Ke007778 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 25 Mar 2010 05:31:02 +0100 Received: (qmail 17856 invoked from network); 25 Mar 2010 04:31:01 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 25 Mar 2010 04:31:01 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Thu, 25 Mar 2010 04:31:01 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #167: Unable to init SFTP session vs SSH-2.0-SSHD X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/167#comment:2 Message-ID: <049.5ab50690c68f381116e02d84044a1370@libssh2.stuge.se> References: <040.b3944a7034e0353dd801d2508d9c873d@libssh2.stuge.se> X-Trac-Ticket-ID: 167 In-Reply-To: <040.b3944a7034e0353dd801d2508d9c873d@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #167: Unable to init SFTP session vs SSH-2.0-SSHD --------------------+------------------------------------------------------- Reporter: ptjm | Owner: bagder Type: defect | Status: assigned Priority: normal | Milestone: Component: SFTP | Version: 1.2.5 Keywords: | Blocking: Blockedby: | --------------------+------------------------------------------------------- Comment(by ptjm): I ran the trace again, this time with LIBSSH2_TRACE_SOCKET, and printing out the value of rc. 35 is EAGAIN on this system. I added one more bit of tracing at line 2001 of src/channel.c: [libssh2] 1.736138 Conn: channel->local.window_size == 0 -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 09:42:43 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2P8gU1g020817; Thu, 25 Mar 2010 09:42:41 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2P8gTvI020786 for ; Thu, 25 Mar 2010 09:42:29 +0100 Date: Thu, 25 Mar 2010 09:42:29 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [libssh2] #166: Some servers terminate connection due to large window adjustment In-Reply-To: <4BAAAC22.6010307@peut.org> Message-ID: References: <040.e4d863057078a2a001d9e278040bea57@libssh2.stuge.se> <049.5b1106677bd85b0cc8c74f1f3432cbb3@libssh2.stuge.se> <4BAAAC22.6010307@peut.org> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Thu, 25 Mar 2010, Jose Baars wrote: > there is no use to enlarge the SSH protocol window above the TCP/IP window > size of the socket. Well, first I'm not so sure about that and secondly we don't know/check the TCP window size so we're not in synch... > The quick fix is the division by 10 and wouldn't break anything, although > performance on really fast connections with RFC compliant SSH servers (any > out there?) might suffer a performance impact. I am the one who bumped the window sizes that much and yes I have successfully tranfered many gigabytes using both SCP and SFTP with libssh2 to and from openssh and it deals with the window sizes just fine and as the spec says. I'm the one who bumped the sizes that much as I saw an increased performance in doing so, but of course performance isn't worth sacrificed functionality. It should be easy enough for anyone to measure what impact this smaller maximum window will have, but I don't personally feel that I have the time and energy to play around with that at the moment. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 14:28:37 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PDSMs5010298; Thu, 25 Mar 2010 14:28:35 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PDSLEa010292 for ; Thu, 25 Mar 2010 14:28:21 +0100 Date: Thu, 25 Mar 2010 14:28:21 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: scp example problem In-Reply-To: <65800b2a1003191449y4a82f0d5v249ff647427a20cb@mail.gmail.com> Message-ID: References: <65800b2a1003191449y4a82f0d5v249ff647427a20cb@mail.gmail.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-1470944514-607835820-1269523701=:16475" X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---1470944514-607835820-1269523701=:16475 Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8BIT On Fri, 19 Mar 2010, vicent roca daniel wrote: > I'm having some problems with libssh2 sending files using scp. > The problem is that I can't send files > 1MB. > The transfer seems to stop without error. > > ¿Does anyone know what can I try? Enable full debugging and show us the last 100 lines before the stop. > I'm using release: libssh2 > 1.2.4from > February 13 2010 It'd be interesting if you could try the latest git version and see if you possibly suffer from the windowing problem. -- / daniel.haxx.se ---1470944514-607835820-1269523701=:16475 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel ---1470944514-607835820-1269523701=:16475-- From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 21:42:51 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PKgYZj002349; Thu, 25 Mar 2010 21:42:48 +0100 Received: from enki.site5.com (enki.site5.com [74.53.3.197]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PKgVrx002172 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 25 Mar 2010 21:42:32 +0100 Received: from [199.247.241.239] (helo=[192.168.1.58]) by enki.site5.com with esmtpa (Exim 4.69) (envelope-from ) id 1Nutsu-0005ea-6k for libssh2-devel@cool.haxx.se; Thu, 25 Mar 2010 15:42:32 -0500 Message-ID: <4BABCAB4.3010906@tsco.ca> Date: Thu, 25 Mar 2010 13:42:28 -0700 From: Joseph Collins User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Memory leak in libssh2_session_startup? X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - enki.site5.com X-AntiAbuse: Original Domain - cool.haxx.se X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - tsco.ca X-Source: X-Source-Args: X-Source-Dir: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se I'm using libssh2 to do some automated file transfers, and there's a memory leak that's causing me some grief. This is the code I'm running (error checks removed for brevity): LIBSSH2_SESSION* ssh_session = libssh2_session_init(); int sock = socket(AF_INET, SOCK_STREAM, 0); struct sockaddr_in addr; addr.sin_family = AF_INET; addr.sin_port = htons(22); inet_pton(AF_INET, "127.0.0.1", &addr.sin_addr); connect(sock, (struct sockaddr*)&addr, sizeof(addr)); libssh2_session_startup(ssh_session, sock); libssh2_userauth_password(ssh_session, USERNAME, PASSWORD); LIBSSH2_SFTP* sftp_session = libssh2_sftp_init(ssh_session); libssh2_sftp_shutdown(sftp_session); libssh2_session_disconnect(ssh_session, "File transfer complete"); libssh2_session_free(ssh_session); close(sock); Valgrind is reporting the following leaks: ==7236== HEAP SUMMARY: ==7236== in use at exit: 2,376 bytes in 40 blocks ==7236== total heap usage: 265 allocs, 225 frees, 91,929 bytes allocated ==7236== ==7236== 20 bytes in 1 blocks are definitely lost in loss record 1 of 7 ==7236== at 0x4024C1C: malloc (vg_replace_malloc.c:195) ==7236== by 0x41A850F: _gcry_private_malloc (stdmem.c:108) ==7236== by 0x41A4747: do_malloc (global.c:737) ==7236== by 0x41A493C: _gcry_malloc (global.c:759) ==7236== by 0x41A499D: _gcry_xmalloc (global.c:903) ==7236== by 0x41F1A18: _gcry_mpi_alloc (mpiutil.c:52) ==7236== by 0x41A357C: gcry_mpi_new (visibility.c:243) ==7236== by 0x403B0CB: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:755) ==7236== by 0x403B259: libssh2_kex_exchange (kex.c:1746) ==7236== by 0x404392E: libssh2_session_startup (session.c:601) ==7236== by 0x80488D8: main (main.c:22) ==7236== ==7236== 20 bytes in 1 blocks are definitely lost in loss record 2 of 7 ==7236== at 0x4024C1C: malloc (vg_replace_malloc.c:195) ==7236== by 0x41A850F: _gcry_private_malloc (stdmem.c:108) ==7236== by 0x41A4747: do_malloc (global.c:737) ==7236== by 0x41A493C: _gcry_malloc (global.c:759) ==7236== by 0x41A499D: _gcry_xmalloc (global.c:903) ==7236== by 0x41F1A18: _gcry_mpi_alloc (mpiutil.c:52) ==7236== by 0x41A357C: gcry_mpi_new (visibility.c:243) ==7236== by 0x4039308: diffie_hellman_sha1 (kex.c:98) ==7236== by 0x403B075: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:768) ==7236== by 0x403B259: libssh2_kex_exchange (kex.c:1746) ==7236== by 0x404392E: libssh2_session_startup (session.c:601) ==7236== by 0x80488D8: main (main.c:22) ==7236== ==7236== LEAK SUMMARY: ==7236== definitely lost: 40 bytes in 2 blocks ==7236== indirectly lost: 0 bytes in 0 blocks ==7236== possibly lost: 0 bytes in 0 blocks ==7236== still reachable: 2,336 bytes in 38 blocks ==7236== suppressed: 0 bytes in 0 blocks ==7236== Reachable blocks (those to which a pointer was found) are not shown. ==7236== To see them, rerun with: --leak-check=full --show-reachable=yes Is there another method I should be calling to free whatever libssh2_session_startup is allocating? Thanks, --Joseph _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 23:17:12 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PMH3NB010552; Thu, 25 Mar 2010 23:17:11 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PMH1U0010523 for ; Thu, 25 Mar 2010 23:17:02 +0100 Date: Thu, 25 Mar 2010 23:17:01 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Memory leak in libssh2_session_startup? In-Reply-To: <4BABCAB4.3010906@tsco.ca> Message-ID: References: <4BABCAB4.3010906@tsco.ca> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Thu, 25 Mar 2010, Joseph Collins wrote: > I'm using libssh2 to do some automated file transfers, and there's a memory > leak that's causing me some grief. Does this leak always happen for you? What libssh2 version is this? > Is there another method I should be calling to free whatever > libssh2_session_startup is allocating? libssh2_session_free() is supposed to free all that memory and if it doesn't it bugs! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 23:38:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PMcNxn018174; Thu, 25 Mar 2010 23:38:29 +0100 Received: from enki.site5.com (enki.site5.com [74.53.3.197]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PMcIbx018142 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 25 Mar 2010 23:38:21 +0100 Received: from [199.247.241.239] (helo=[192.168.1.58]) by enki.site5.com with esmtpa (Exim 4.69) (envelope-from ) id 1Nuvgx-000093-VO for libssh2-devel@cool.haxx.se; Thu, 25 Mar 2010 17:38:20 -0500 Message-ID: <4BABE5D8.6080602@tsco.ca> Date: Thu, 25 Mar 2010 15:38:16 -0700 From: Joseph Collins User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3 MIME-Version: 1.0 To: libssh2-devel@cool.haxx.se Subject: Re: Memory leak in libssh2_session_startup? X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - enki.site5.com X-AntiAbuse: Original Domain - cool.haxx.se X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - tsco.ca X-Source: X-Source-Args: X-Source-Dir: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1167133452==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part message in MIME format. --===============1167133452== Content-Type: multipart/alternative; boundary="------------090707060805040004010709" This is a multi-part message in MIME format. --------------090707060805040004010709 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Yes, this leak is always happening. I am using the latest release (1.2.4). Since this is indeed a bug, I will file a bug report on Trac. Thanks, Daniel. --Joseph On 3/25/2010 15:17 PM, Daniel Stenberg wrote: On Thu, 25 Mar 2010, Joseph Collins wrote: /> I'm using libssh2 to do some automated file transfers, and there's a memory / /> leak that's causing me some grief. / Does this leak always happen for you? What libssh2 version is this? /> Is there another method I should be calling to free whatever / /> libssh2_session_startup is allocating? / libssh2_session_free() is supposed to free all that memory and if it doesn't it bugs! --------------090707060805040004010709 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Yes, this leak is always happening.  I am using the latest release (1.2.4).

Since this is indeed a bug, I will file a bug report on Trac.

Thanks, Daniel.
--Joseph


On 3/25/2010 15:17 PM, Daniel Stenberg wrote:

On Thu, 25 Mar 2010, Joseph Collins wrote:

> I'm using libssh2 to do some automated file transfers, and there's a memory
> leak that's causing me some grief.

Does this leak always happen for you? What libssh2 version is this?

> Is there another method I should be calling to free whatever
> libssh2_session_startup is allocating?

libssh2_session_free() is supposed to free all that memory and if it doesn't
it bugs!
--------------090707060805040004010709-- --===============1167133452== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1167133452==-- From libssh2-devel-bounces@cool.haxx.se Thu Mar 25 23:46:38 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PMkEdF021865; Thu, 25 Mar 2010 23:46:37 +0100 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2PMkDGJ021857 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Thu, 25 Mar 2010 23:46:13 +0100 Received: (qmail 6360 invoked from network); 25 Mar 2010 22:46:20 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 25 Mar 2010 22:46:20 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Thu, 25 Mar 2010 22:46:20 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #168: Memory leak in libssh2_session_startup X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/168 Message-ID: <044.2a48d5b8cb5be2e3469617435c23e879@libssh2.stuge.se> X-Trac-Ticket-ID: 168 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #168: Memory leak in libssh2_session_startup ----------------------+----------------------------------------------------- Reporter: jcollins | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | ----------------------+----------------------------------------------------- Calling libssh2_session_startup results in a memory leak which is not rectified by libssh_sesion_free. Steps to reproduce: - Change the attached "main.c" to the username and password for your server - Compile using gcc, linking with -lssh2 - Run under valgrind --tool=memcheck --leak-check=full Expected result: no errors or memory leaks Actual result: 2 memory leaks found (see attached valgrind.txt) This bug is present in the latest Ubuntu repo version (1.1) as well as the latest release (1.2.4). -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 26 11:55:00 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QAsiQ3025797; Fri, 26 Mar 2010 11:54:58 +0100 Received: from f1wall.ipetronik.com (f1wall.ipetronik.com [217.25.129.2]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QAsgN8025789 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 26 Mar 2010 11:54:42 +0100 Received: from [10.1.1.105] (port=5185 helo=IPEMAIL.ipetronik.de) by f1wall.ipetronik.com with esmtp (Exim 4.69) (envelope-from ) id 1Nv7BP-0002mH-2L for libssh2-devel@cool.haxx.se; Fri, 26 Mar 2010 11:54:31 +0100 Received: from IPE_DOM-MTA by IPEMAIL.ipetronik.de with Novell_GroupWise; Fri, 26 Mar 2010 11:54:31 +0100 Message-Id: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> X-Mailer: Novell GroupWise Internet Agent 8.0.1 Date: Fri, 26 Mar 2010 11:54:25 +0100 From: "Sebastien Fricker" To: Subject: Select (function _libssh2_wait_socket) issue on some on an embedded platform Mime-Version: 1.0 Content-Disposition: inline X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se SGksCndlIGhhdmUgZm91bmQgYW4gaXNzdWUgb24gYSByZWFsLXRpbWUgT1MgKGhlcmUgUlRPUy0z MiBmcm9tIE9uVGltZSkKYWJvdXQgdGhlIHVzYWdlIG9mIHNlbGVjdC4KCkluIHRoZSBmdW5jdGlv biBfbGlic3NoX3dhaXRfc29ja2V0IHNlbGVjdCBpcyB1c2VkIHdpdGggMiBwb2ludGVyIG9mCmRl c2NyaXB0b3JzOiByZWFkZmQgYW5kIHdyaXRlZmQuClRoaXMgcG9pbnRlcnMgYXJlIHBvaW50aW5n IHRvIHRoZSBzYW1lIGZkX3NldC4gCk9uIFJUT1MzMiwgdGhpcyBjYXVzZXMgYW4gZXJyb3IuCgpJ ZiByZWFkZmQgYW5kIHdyaXRlZmQgYXJlIHBvaW50aW5nIHRvIDIgZGlmZmVyZW50IHZhcmlhYmxl cywgc2VsZWN0CndvcmtzIHdlbGwuCgpJbiB0aGlzIGNhc2UgdGhlIG1vZGlmaWVkIGNvZGUgd291 bGQgYmU6CmludCBfbGlic3NoMl93YWl0X3NvY2tldChMSUJTU0gyX1NFU1NJT04gKnNlc3Npb24p CnsKICAgIGZkX3NldCBmZF9yZWFkOwogICAgZmRfc2V0IGZkX3dyaXRlOwogICAgZmRfc2V0ICp3 cml0ZWZkID0gTlVMTDsKICAgIGZkX3NldCAqcmVhZGZkID0gTlVMTDsKICAgIGludCBkaXI7CiAg ICBpbnQgcmM7CgogICAgRkRfWkVSTygmZmRfcmVhZCk7CiAgICBGRF9TRVQoc2Vzc2lvbi0+c29j a2V0X2ZkLCAmZmRfcmVhZCk7CiAgICBGRF9aRVJPKCZmZF93cml0ZSk7CiAgICBGRF9TRVQoc2Vz c2lvbi0+c29ja2V0X2ZkLCAmZmRfd3JpdGUpOwoKICAgIC8qIG5vdyBtYWtlIHN1cmUgd2Ugd2Fp dCBpbiB0aGUgY29ycmVjdCBkaXJlY3Rpb24gKi8KICAgIGRpciA9IGxpYnNzaDJfc2Vzc2lvbl9i bG9ja19kaXJlY3Rpb25zKHNlc3Npb24pOwoKICAgIGlmKGRpciAmIExJQlNTSDJfU0VTU0lPTl9C TE9DS19JTkJPVU5EKQogICAgICAgIHJlYWRmZCA9ICZmZF9yZWFkOwoKICAgIGlmKGRpciAmIExJ QlNTSDJfU0VTU0lPTl9CTE9DS19PVVRCT1VORCkKICAgICAgICB3cml0ZWZkID0gJmZkX3dyaXRl OwoKICAuLi4uLgoKVGhlIG1vZGlmaWNhdGlvbiBzaG91bGQgbm90IGNyZWF0ZSBhbnkgcmVncmVz c2lvbiB0byBsaWJzc2gyLCBhbmQgaWYKbm9ib2R5IGFyZSBzZWVpbmcgYW55IHByb2JsZW0gYWJv dXQgdGhpcyBjb2RlLCBJIHdpbGwgcHJvdmlkZSBhIHBhdGNoLgpSZWdhcmRzIApTw6liYXN0aWVu CgoKLS0gCgotLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0KSXBldHJvbmlrIEdtYkggJiBDby5LRwpKYWVnZXJ3ZWcgMQpELTc2NTMyIEJh ZGVuLUJhZGVuCiAKUGhvbmUtTm8uICs0OSAtICgwKTcyMjEvOTkyMi00NjcKRmF4LU5vLiArNDkg LSAoMCk3MjIxLzk5MjItMTUzCiAKbWFpbHRvOiBzZWJhc3RpZW4uZnJpY2tlckBpcGV0cm9uaWsu Y29tCndlYjogaHR0cDovL3d3dy5pcGV0cm9uaWsuY29tIAogCi0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQpLb21tYW5kaXRnZXNlbGxz Y2hhZnQgbWl0IFNpdHogaW4gQmFkZW4tQmFkZW4sIFJlZ2lzdGVyZ2VyaWNodCBIUkEKTnIuMjAx MzEzClBlcnNvZW5saWNoIGhhZnRlbmRlIEdlc2VsbHNjaGFmdCBpc3QgZGllIElQRVRST05JSyBW ZXJ3YWx0dW5ncyBHbWJIIAptaXQgU2l0eiBpbiBCYWRlbi1CYWRlbiwgUmVnaXN0ZXJnZXJpY2h0 IE1hbm5oZWltIEhSQiBOci4yMDIwODksIApHZXNjaMOkZnRzZsO8aHJlcjogSi5BYnJvbWVpdCwg RS5SdWRvbGYsIEEuIFdvY2tlCi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0KIAogCgpfX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fXwpsaWJzc2gyLWRldmVsIGh0dHA6Ly9jb29sLmhheHguc2UvY2dp LWJpbi9tYWlsbWFuL2xpc3RpbmZvL2xpYnNzaDItZGV2ZWwK From libssh2-devel-bounces@cool.haxx.se Fri Mar 26 14:13:58 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QDDnfs020392; Fri, 26 Mar 2010 14:13:57 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QDDm3X020385 for ; Fri, 26 Mar 2010 14:13:48 +0100 Date: Fri, 26 Mar 2010 14:13:48 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Select (function _libssh2_wait_socket) issue on some on an embedded platform In-Reply-To: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> Message-ID: References: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 26 Mar 2010, Sebastien Fricker wrote: > If readfd and writefd are pointing to 2 different variables, select works > well. ... > The modification should not create any regression to libssh2, and if nobody > are seeing any problem about this code, I will provide a patch. I wouldn't mind that minor work-around to make it run better. But please, add an explanatory comment about the reason for this weirdness, as otherwise someone will just optimize it away again in the future... -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 26 14:29:55 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QDTqSo001634; Fri, 26 Mar 2010 14:29:54 +0100 Received: from f1wall.ipetronik.com (f1wall.ipetronik.com [217.25.129.2]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QDTpLM001622 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 26 Mar 2010 14:29:51 +0100 Received: from [10.1.1.105] (port=11289 helo=IPEMAIL.ipetronik.de) by f1wall.ipetronik.com with esmtp (Exim 4.69) (envelope-from ) id 1Nv9bZ-0001mW-2c for libssh2-devel@cool.haxx.se; Fri, 26 Mar 2010 14:29:41 +0100 Received: from IPE_DOM-MTA by IPEMAIL.ipetronik.de with Novell_GroupWise; Fri, 26 Mar 2010 14:29:41 +0100 Message-Id: <4BACC4D20200003F00008EE1@IPEMAIL.ipetronik.de> X-Mailer: Novell GroupWise Internet Agent 8.0.1 Date: Fri, 26 Mar 2010 14:29:38 +0100 From: "Sebastien Fricker" To: "libssh2 development" Subject: Antw: Re: Select (function _libssh2_wait_socket) issue on some on an embedded platform References: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> In-Reply-To: Mime-Version: 1.0 Content-Disposition: inline X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se RGFuaWVsLApPbiBSVE9TIHNlbGVjdCBhbHdheXMgcmV0dXJuIDAgaWYgcmVhZGZkIGFuZCB3cml0 ZWZkIGFyZSBwb2ludGluZyB0bwp0aGUgc2FtZSB2YXJpYWJsZS4KCmlmIHlvdSBoYXZlIGEgbG9v ayBvbiB0aGUgcHJvdG90eXBlIG9mIHRoZSBzZWxlY3QgZnVuY3Rpb24geW91IGNhbiBzZWUKdGhh dCB0aGUgZGZfc2V0IHBvaW50ZXIgYXJlIG5vdCBjb25zdCBwb2ludGVyczoKCmludCBzZWxlY3Qo aW50IG5mZHMsIGZkX3NldCAqcmVhZGZkcywgZmRfc2V0ICp3cml0ZWZkcywKICAgICAgICAgICAg ICAgICAgZmRfc2V0ICpleGNlcHRmZHMsIHN0cnVjdCB0aW1ldmFsICp0aW1lb3V0KTsKClRoaXMg bWVhbiB0aGF0IHNlbGVjdCBjYW4gbW9kaWZ5IHRoZSBjb250YWluIG9mIHJlYWRmZCBhbmQgd3Jp dGVmZC4gSW4KdGhlIG1hbnVhbCBvZiBzZWxlY3Qgb24gTGludXgsIEkgYWxzbyBmb3VuZDoKIiBP biBleGl0LCB0aGUgc2V0cyBhcmUgIG1vZGlmaWVkICBpbiBwbGFjZSB0byBpbmRpY2F0ZSB3aGlj aCBmaWxlCmRlc2NyaXB0b3JzIGFjdHVhbGx5IGNoYW5nZWQgc3RhdHVzLiAiCgpJIGd1ZXNzIHRo YXQgdGhpcyBpcyB0aGUgcHJvYmxlbTogJmZkIGlzIG1vZGlmaWVkIHR3aWNlLCBhbmQgdGhpcyBj b3VsZApiZSBhIHByb2JsZW0gb24gc29tZSBPUy4KUmVnYXJkcwpTw6liYXN0aWVuCgoKCj4+PiBE YW5pZWwgU3RlbmJlcmcgPGRhbmllbEBoYXh4LnNlPiBzY2hyaWViIGFtIDI2LjAzLjIwMTAgdW0g MTQ6MTMgaW4KTmFjaHJpY2h0CjxhbHBpbmUuREVCLjIuMDAuMTAwMzI2MTQxMjIwMC4yNzgwOEB0 dm5hZy51bmtrLmZyPjoKPiBPbiBGcmksIDI2IE1hciAyMDEwLCBTZWJhc3RpZW4gRnJpY2tlciB3 cm90ZToKPiAKPj4gSWYgcmVhZGZkIGFuZCB3cml0ZWZkIGFyZSBwb2ludGluZyB0byAyIGRpZmZl cmVudCB2YXJpYWJsZXMsIHNlbGVjdAp3b3JrcyAKPj4gd2VsbC4KPiAKPiAuLi4KPiAKPj4gVGhl IG1vZGlmaWNhdGlvbiBzaG91bGQgbm90IGNyZWF0ZSBhbnkgcmVncmVzc2lvbiB0byBsaWJzc2gy LCBhbmQgaWYKbm9ib2R5IAo+PiBhcmUgc2VlaW5nIGFueSBwcm9ibGVtIGFib3V0IHRoaXMgY29k ZSwgSSB3aWxsIHByb3ZpZGUgYSBwYXRjaC4KPiAKPiBJIHdvdWxkbid0IG1pbmQgdGhhdCBtaW5v ciB3b3JrLWFyb3VuZCB0byBtYWtlIGl0IHJ1biBiZXR0ZXIuIEJ1dApwbGVhc2UsIGFkZCAKPiAK PiBhbiBleHBsYW5hdG9yeSBjb21tZW50IGFib3V0IHRoZSByZWFzb24gZm9yIHRoaXMgd2VpcmRu ZXNzLCBhcwpvdGhlcndpc2UgCj4gc29tZW9uZSB3aWxsIGp1c3Qgb3B0aW1pemUgaXQgYXdheSBh Z2FpbiBpbiB0aGUgZnV0dXJlLi4uCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fCmxpYnNzaDItZGV2ZWwgaHR0cDovL2Nvb2wuaGF4eC5zZS9jZ2ktYmluL21h aWxtYW4vbGlzdGluZm8vbGlic3NoMi1kZXZlbAo= From libssh2-devel-bounces@cool.haxx.se Fri Mar 26 14:43:19 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QDhFhl007215; Fri, 26 Mar 2010 14:43:19 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QDhD2c007195 for ; Fri, 26 Mar 2010 14:43:13 +0100 Date: Fri, 26 Mar 2010 14:43:13 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Antw: Re: Select (function _libssh2_wait_socket) issue on some on an embedded platform In-Reply-To: <4BACC4D20200003F00008EE1@IPEMAIL.ipetronik.de> Message-ID: References: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> <4BACC4D20200003F00008EE1@IPEMAIL.ipetronik.de> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 26 Mar 2010, Sebastien Fricker wrote: Please don't top-post, it makes it very hard to follow the discussion. > On RTOS select always return 0 if readfd and writefd are pointing to > the same variable. That is crazy behavior and you don't need to explain it to us here, I want it explained with a comment in the patch! > if you have a look on the prototype of the select function you can see that > the df_set pointer are not const pointers: Yes, but that only mean that we cannot use the contents of the variables for anything sensible, and if you read the code you'll see that we don't. > I guess that this is the problem: &fd is modified twice, and this could be a > problem on some OS. Only in bad implementations, but I already explained that I'm fine with adapting to them. I would argue that a select() implementation should have the values stored in variables of its own and copy those values to the out-parameters (instead of assuming that the pointers point to two different areas), as then it would work perfectly fine even when apps do what we do. Given that we know this code works on quite a few platforms, I would assume most select()-implmentations work that way. If I were you, I'd file a bug report to the OS manufacturer. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Fri Mar 26 18:33:57 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QHXgU1008570; Fri, 26 Mar 2010 18:33:54 +0100 Received: from f1wall.ipetronik.com (f1wall.ipetronik.com [217.25.129.2]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2QHXer5008559 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 26 Mar 2010 18:33:41 +0100 Received: from [10.1.1.105] (port=21134 helo=IPEMAIL.ipetronik.de) by f1wall.ipetronik.com with esmtp (Exim 4.69) (envelope-from ) id 1NvDPW-0004uc-01 for libssh2-devel@cool.haxx.se; Fri, 26 Mar 2010 18:33:30 +0100 Received: from IPE_DOM-MTA by IPEMAIL.ipetronik.de with Novell_GroupWise; Fri, 26 Mar 2010 18:33:29 +0100 Message-Id: <4BACFDF60200003F00008F48@IPEMAIL.ipetronik.de> X-Mailer: Novell GroupWise Internet Agent 8.0.1 Date: Fri, 26 Mar 2010 18:33:26 +0100 From: "Sebastien Fricker" To: Subject: send_existing does not clear the flag LIBSSH2_SESSION_BLOCK_OUTBOUND Mime-Version: 1.0 Content-Disposition: inline X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se SGksCndoZW4gc2VuZF9leGlzdGluZygpIGhhcyBmdWxseSB0cmFuc2ZlcnJlZCB0aGUgcmVtYWlu aW5nIG91dGdvaW5nIGRhdGEsCnRoZSBmbGFnIExJQlNTSDJfU0VTU0lPTl9CTE9DS19PVVRCT1VO RCBzaG91bGQgYmUgY2xlYXIuCgpIZXJlIHRoZSBzb3VyY2Ugb3JpZ2luYWwgc291cmNlIGNvZGU6 CnRyYW5zcG9ydC5jIGxpbmUgNjQ2OgooaHR0cDovL2dpdC5saWJzc2gyLm9yZy8/cD1saWJzc2gy LmdpdDthPWJsb2I7Zj1zcmMvdHJhbnNwb3J0LmM7aD00M2RmNzZjMTNjM2Y1NjAwNDA5NjE4OTgx MTViOGU2ZTZiZmQzODJkO2hiPUhFQUQjbDY0OCkKIDY0NiAgICAgaWYgKHJjID09IGxlbmd0aCkg ewogNjQ3ICAgICAgICAgLyogdGhlIHJlbWFpbmRlciBvZiB0aGUgcGFja2FnZSB3YXMgc2VudCAq LwogNjQ4ICAgICAgICAgTElCU1NIMl9GUkVFKHNlc3Npb24sIHAtPm91dGJ1Zik7CiA2NDkgICAg ICAgICBwLT5vdXRidWYgPSBOVUxMOwogNjUwICAgICAgICAgcC0+b3RvdGFsX251bSA9IDA7CiA2 NTEgICAgIH0KCldlIHNob3VsZCBtb2RpZnkgaXQgYXMgZm9sbG93czoKCiA2NDYgICAgIGlmIChy YyA9PSBsZW5ndGgpIHsKIDY0NyAgICAgICAgIC8qIHRoZSByZW1haW5kZXIgb2YgdGhlIHBhY2th Z2Ugd2FzIHNlbnQgKi8KICAgICAgICAgICAgIHNlc3Npb24tPnNvY2tldF9ibG9ja19kaXJlY3Rp b25zICY9Cn5MSUJTU0gyX1NFU1NJT05fQkxPQ0tfT1VUQk9VTkQ7CiA2NDggICAgICAgICBMSUJT U0gyX0ZSRUUoc2Vzc2lvbiwgcC0+b3V0YnVmKTsKIDY0OSAgICAgICAgIHAtPm91dGJ1ZiA9IE5V TEw7CiA2NTAgICAgICAgICBwLT5vdG90YWxfbnVtID0gMDsKIDY1MSAgICAgfQoKVGhpcyBhdm9p ZHMgdGhhdCBsaWJzc2gyIHVzZXMgMTAwJSBDUFUsIHdoZW4gYW4gb3V0Z29pbmcgdHJhbnNmZXIg dG8gYQpzZXJ2ZXIgaXMgZmluaXNoZWQgYW5kIHdoZW4gbGlic3NoIGlzIHdhaXRpbmcgZm9yIHRo ZSBhbnN3ZXIgZnJvbQpzZXJ2ZXIuCgpQbGVhc2UgZmVlbCBmcmVlIHRvIGNvbW1lbnQgdGhpcyBt b2RpZmljYXRpb24uClJlZ2FyZHMKU8OpYmFzdGllbgpfX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fXwpsaWJzc2gyLWRldmVsIGh0dHA6Ly9jb29sLmhheHguc2Uv Y2dpLWJpbi9tYWlsbWFuL2xpc3RpbmZvL2xpYnNzaDItZGV2ZWwK From libssh2-devel-bounces@cool.haxx.se Sat Mar 27 23:28:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2RMRmxV031514; Sat, 27 Mar 2010 23:28:02 +0100 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2RMRlnQ031509 for ; Sat, 27 Mar 2010 23:27:47 +0100 Date: Sat, 27 Mar 2010 23:27:47 +0100 (CET) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: send_existing does not clear the flag LIBSSH2_SESSION_BLOCK_OUTBOUND In-Reply-To: <4BACFDF60200003F00008F48@IPEMAIL.ipetronik.de> Message-ID: References: <4BACFDF60200003F00008F48@IPEMAIL.ipetronik.de> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Fri, 26 Mar 2010, Sebastien Fricker wrote: > when send_existing() has fully transferred the remaining outgoing data, the > flag LIBSSH2_SESSION_BLOCK_OUTBOUND should be clear. Hm, yes. I agree and that's why the code originally did that! However, it seems there was some problem with that approach and the commit 7317edab61d2179febc38a2c2c4da0b951d74cbc by Peter Stuge changed that behavior. I'm trying to read the commit message and think through the code to understand why that change was made, but I can't quite get it. My analysis is instead that send_existing() should set the LIBSSH2_SESSION_BLOCK_OUTBOUND bit for the case when only a piece of the buffer was sent and it returns PACKET_EAGAIN - in the last line of the function. As otherwise, it will return EAGAIN without a direction bitmask set and that's not good. It signals to the caller that it wants to be called again and it sets for what direction it wants the socket to be ready to get called. Comments anyone? -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 28 01:36:29 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2S0aHPC019521; Sun, 28 Mar 2010 01:36:28 +0100 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2S0aFIf019518 for ; Sun, 28 Mar 2010 01:36:15 +0100 Received: (qmail 7493 invoked by uid 501); 28 Mar 2010 00:36:23 -0000 Message-ID: <20100328003623.7492.qmail@stuge.se> Date: Sun, 28 Mar 2010 01:36:23 +0100 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: send_existing does not clear the flag LIBSSH2_SESSION_BLOCK_OUTBOUND Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4BACFDF60200003F00008F48@IPEMAIL.ipetronik.de> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="/WwmFnJnmDyWGHa4" Content-Disposition: inline In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --/WwmFnJnmDyWGHa4 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Daniel Stenberg wrote: >> when send_existing() has fully transferred the remaining outgoing data, >> the flag LIBSSH2_SESSION_BLOCK_OUTBOUND should be clear. > > Hm, yes. I agree and that's why the code originally did that! > > However, it seems there was some problem with that approach and the > commit 7317edab61d2179febc38a2c2c4da0b951d74cbc by Peter Stuge > changed that behavior. That commit improved the previous state, but wasn't a full fix. Thanks to Sebastien for reporting this problem! Please try the attached patch. I'll push it if it solves the problem you are seeing. (Please also describe the circumstance that exposed this problem for you.) Maybe this can help those curious about what is going on here: See the code in _transport_write() after the call to send_existing(). Note the return value for send_existing() and how *ret is used. send_existing() returns PACKET_NONE (0) when it has sent the final part of a packet, and before that it always returns PACKET_EAGAIN or an actual error. *ret is set to 1 when there are parts of a packet remaining in (=before) this call to send_existing(). The if() immediately following the call to send_existing() in current git makes _transport_write() return right away either on error in send_existing() (good!) or when send_existing() finished sending a packet (bad, must clear flag first!). The only situation in which the flag is cleared, is if this call to _transport_write() is the first call for a new packet. The attached patch should split this up properly. > I'm trying to read the commit message and think through the code to > understand why that change was made, but I can't quite get it. Read _transport_write() and send_existing() over and over. It's pretty convoluted, again because of how the API for _transport_write() works, but I'm sure you'll get it. Feel free to ask questions if we can help. > My analysis is instead that send_existing() should set the > LIBSSH2_SESSION_BLOCK_OUTBOUND bit for the case when only a piece of > the buffer was sent and it returns PACKET_EAGAIN - in the last line > of the function. As otherwise, it will return EAGAIN without a > direction bitmask set and that's not good. > > It signals to the caller that it wants to be called again and it > sets for what direction it wants the socket to be ready to get > called. > > Comments anyone? Explicitly setting the flag not only on send() error but also on send() success, instead of clearing the flag after finishing, would work just as well but may not make a significant difference? The question is if the bit can be cleared by someone else, between calls to libssh2_transport_write(). If it can be cleared, then send_existing() should always set the flag as you say. If noone else will clear the bit, then it's fine for send_existing() to just not touch the bit, and the setting of the bit in the error code path could then be removed. Keep in mind that send_existing() only sends the second part of a packet, and since _transport_write() already has set the flag on line 827 (when it sends out the first part) there's no point in setting it again as long as the flag isn't cleared until the whole packet has been sent out. //Peter --/WwmFnJnmDyWGHa4 Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="0001-Correctly-clear-blocking-flag-after-sending-multipar.patch" From b6ed60d2ccb32452e4727f7b2da1f3052f0e63b5 Mon Sep 17 00:00:00 2001 From: Peter Stuge Date: Sun, 28 Mar 2010 00:50:49 +0100 Subject: [PATCH] Correctly clear blocking flag after sending multipart packet commit 7317edab61d2179febc38a2c2c4da0b951d74cbc cleared the outbound blocking bit when send_existing() returned PACKET_NONE and *ret=0, as opposed to before even calling send_existing(), but because *ret=1 when sending parts 2..n of an existing packet, the bit would only be cleared when calling libssh2_transport_write() for a new packet. Clear the direction flag after the final part of a packet has been sent. --- src/transport.c | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/src/transport.c b/src/transport.c index 43df76c..605e002 100644 --- a/src/transport.c +++ b/src/transport.c @@ -708,13 +708,14 @@ _libssh2_transport_write(LIBSSH2_SESSION * session, unsigned char *data, /* FIRST, check if we have a pending write to complete */ rc = send_existing(session, data, data_len, &ret); - if (rc || ret) { + if (rc) return rc; - } - /* default clear the bit */ session->socket_block_directions &= ~LIBSSH2_SESSION_BLOCK_OUTBOUND; + if (ret) + return rc; + encrypted = (session->state & LIBSSH2_STATE_NEWKEYS) ? 1 : 0; /* check if we should compress */ -- 1.6.3.3 --/WwmFnJnmDyWGHa4 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --/WwmFnJnmDyWGHa4-- From libssh2-devel-bounces@cool.haxx.se Sun Mar 28 03:17:45 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2S1Hd8v007906; Sun, 28 Mar 2010 03:17:44 +0200 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2S1HcGS007903 for ; Sun, 28 Mar 2010 03:17:38 +0200 Received: (qmail 16566 invoked by uid 501); 28 Mar 2010 01:17:46 -0000 Message-ID: <20100328011746.16565.qmail@stuge.se> Date: Sun, 28 Mar 2010 03:17:46 +0200 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Select (function _libssh2_wait_socket) issue on some on an embedded platform Mail-Followup-To: libssh2-devel@cool.haxx.se References: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <4BACA0710200003F00008ECE@IPEMAIL.ipetronik.de> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Sebastien Fricker wrote: > In the function _libssh_wait_socket select is used with 2 pointer > of descriptors: readfd and writefd. > This pointers are pointing to the same fd_set. > On RTOS32, this causes an error. I just pushed this fix. > if nobody are seeing any problem about this code, I will provide a > patch. **PLEASE** just send the patch immediately! It is less work for you to create than copypasting a bunch of code, it is MUCH easier to read, and if everyone approves then the patch can be committed directly and the case is closed. So it would save time for everyone! //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Sun Mar 28 15:40:11 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2SDduuU001612; Sun, 28 Mar 2010 15:40:08 +0200 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2SDdsB0001600 for ; Sun, 28 Mar 2010 15:39:54 +0200 Date: Sun, 28 Mar 2010 15:39:54 +0200 (CEST) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: send_existing does not clear the flag LIBSSH2_SESSION_BLOCK_OUTBOUND In-Reply-To: <20100328003623.7492.qmail@stuge.se> Message-ID: References: <4BACFDF60200003F00008F48@IPEMAIL.ipetronik.de> <20100328003623.7492.qmail@stuge.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Sun, 28 Mar 2010, Peter Stuge wrote: > Please try the attached patch. > > I'll push it if it solves the problem you are seeing. (Please also describe > the circumstance that exposed this problem for you.) This fix appears to me to be a good one! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 06:20:09 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2T4Ju1f023568; Mon, 29 Mar 2010 06:20:07 +0200 Received: from mail-pv0-f182.google.com (mail-pv0-f182.google.com [74.125.83.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2T4IqQW023371; Mon, 29 Mar 2010 06:18:52 +0200 Received: by pvc7 with SMTP id 7so4723532pvc.41 for ; Sun, 28 Mar 2010 21:18:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:received:message-id :subject:from:to:content-type; bh=W/uzOx1aUlvb8yGyzELDRwa6HiELMe6GYLwLtZ6+m3w=; b=YQaVJxUJsqEGsaqdNAWaiNdTkIi5p0M768e+CnYjrIITAa2yjftZoi75cmiS6/frvp y+8z5hIe3CBfhwmNTzi1vknnAtAz5IlVcPkrkeg+nohFzeFaHvca6gJdy2LVujDN7aHD toN/e0DdC3k5ceAfS2Gj48nWop/N7IIeEjw5Q= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=rAV7A64WJgshx1gnoEmnjTllIwpVJSoxza+o/6LTnxrJ3da9do1FBd+y4WXDMObtbK CBFVHekGGvSeWxujSooRkySWXy0cKAHCPIWqJnpWRtC9ACDj3dnNP0kgWqNze9PsfM4Q XmB+wl574xPUrBXoaNrnMzp9sLdxNqGq3iPbQ= MIME-Version: 1.0 Received: by 10.142.213.13 with HTTP; Sun, 28 Mar 2010 21:18:47 -0700 (PDT) Date: Mon, 29 Mar 2010 09:48:47 +0530 Received: by 10.143.27.1 with SMTP id e1mr1734814wfj.343.1269836327816; Sun, 28 Mar 2010 21:18:47 -0700 (PDT) Message-ID: Subject: Character conversion for sftp From: NAGASEKHAR BHUVANAGIRI To: libcurl development , libssh2-devel@cool.haxx.se X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1821329053==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============1821329053== Content-Type: multipart/alternative; boundary=0050450294e33e10470482e8cfe6 --0050450294e33e10470482e8cfe6 Content-Type: text/plain; charset=ISO-8859-1 Hi, I am using libcurl 7.18.1 to do ftp and sftp (libssh2-1.1) on z/OS. By defining the CURL_ICONV_CODESET_OF_HOST and CURL_ICONV_CODESET_OF_NETWORK, ftp is done correctly. But, I still see the problem with sftp. I see that the "ftp_readresp" does data conversion using Curl_convert_from_network and "Curl_nbftpsendf" does it through Curl_convert_to_network (/lib/ftp.c). Similar conversion is not done for sftp's Curl_sftp_recv and Curl_sftp_send (src/curl/curl-7.18.1/lib/ssh.c). Could someone advice me how the character conversion can be handled for sftp? Thanks in advance, Nagasekhar --0050450294e33e10470482e8cfe6 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Hi,
I am using libcurl 7.18.1=A0to do ftp and sftp (libssh2-1.1) on z/OS. = By defining the CURL_ICONV_CODESET_OF_HOST and CURL_ICONV_CODESET_OF_NETWORK, ftp is done correctly.
But, I stil= l see the problem with sftp.=A0
I=A0see that the "ftp_readresp" does data conversion using Curl_con= vert_from_network and "Curl_nbftpsendf" does it = through Curl_convert_to_network (/lib/ftp.c). Similar conversion is not don= e for sftp's Curl_sftp_recv and Curl_= sftp_send=A0 (src/curl/curl-7.18.1/lib/ssh.c).=A0
Could someone advice me how the character conversion can be handled for sft= p?

Thanks in ad= vance,
Nagasekhar
--0050450294e33e10470482e8cfe6-- --===============1821329053== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1821329053==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 14:18:18 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TCI3K0001517; Mon, 29 Mar 2010 14:18:16 +0200 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TCI2HJ001475 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 29 Mar 2010 14:18:02 +0200 Received: (qmail 7374 invoked from network); 29 Mar 2010 12:18:05 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 29 Mar 2010 12:18:05 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 29 Mar 2010 12:18:05 -0000 X-URL: http://libssh2.stuge.se/ Subject: [libssh2] #169: segmentation fault in pubkey authentication X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/169 Message-ID: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-Trac-Ticket-ID: 169 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #169: segmentation fault in pubkey authentication -------------------------+-------------------------------------------------- Reporter: suyogjadhav | Owner: Type: defect | Status: new Priority: high | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | -------------------------+-------------------------------------------------- libssh2 encounter SIGSEGV when null passphrase is passed to libssh2_userauth_publickey_fromfile while the passed key requires a valid passphrase. My application takes these configuration from user,and when passed along they cause app to crash. The error is traced to passphrase_cb in openssl.c. This function does not check for null passphrase and tries to find strlen causing the segmentation fault. A simple check such as this would suffice, passphrase_len = 0; if(passphrase != NULL) passphrase_len = strlen(passphrase); -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 14:24:06 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TCO3er008199; Mon, 29 Mar 2010 14:24:05 +0200 Received: from dottedmag.net (burger.dottedmag.net [212.75.37.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TCO08B008094 for ; Mon, 29 Mar 2010 14:24:01 +0200 Received: from vertex.dottedmag (unknown [91.197.127.125]) by dottedmag.net (Postfix) with ESMTPSA id 492DB8C00C for ; Mon, 29 Mar 2010 14:24:04 +0200 (CEST) Received: from dottedmag by vertex.dottedmag with local (Exim 4.71) (envelope-from ) id 1NwE0f-0006sg-Ic for libssh2-devel@cool.haxx.se; Mon, 29 Mar 2010 19:24:01 +0700 From: Mikhail Gusarov To: libssh2 development Subject: Re: [libssh2] #169: segmentation fault in pubkey authentication References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> Date: Mon, 29 Mar 2010 19:24:00 +0700 In-Reply-To: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> (libssh's message of "Mon, 29 Mar 2010 12:18:05 -0000") Message-ID: <87zl1r9uzj.fsf@vertex.dottedmag> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1981204706==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============1981204706== Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --=-=-= Content-Transfer-Encoding: quoted-printable Twas brillig at 12:18:05 29.03.2010 UTC+00 when trac@libssh2.stuge.se did g= yre and gimble: lT> libssh2 encounter SIGSEGV when null passphrase is passed to lT> libssh2_userauth_publickey_fromfile while the passed key requires a lT> valid passphrase. Looks like GIGO for me. =2D-=20 http://fossarchy.blogspot.com/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) iQIcBAEBAgAGBQJLsJvhAAoJEJ0g9lA+M4iIMUkQALICV++BHx1ANUnFoGaWpbdB NT9uDtp3RwMSc8v4A7fJiT/maBnb0I+igOdFWMWer6HW02NbIgKVbHH+8AW93FiP T9Jt6F+BQxx+M0rEX/iluMHmmlHts/s1QijTmOScjgx5W4F8aTyiHHKZa+VFyXpP TAo12xCPIYsDW/DSwPFh0+EDLYCvTF/m8E5kULyeU2MfwMDe81YK38ESeNPsnH4P HJss5040gwQOWYC/jCV3oJ5toU/s2bU+Y9IUY0ieUutmwihcYx18T+m4Twx1q92z wZioINEw3rFzw/v5UDfgmF2HT7ThlcBmzgGO/ULvI7jYKoDEb6olcOsv0uozXoP1 7b/IykIFJ6ABuFqZ7iukWIv1H4hUo5GJKnPc8JSJlnRPobUMtD2lLqWyeHLZza3r bqz/cqmJ/Cf/6JZz761XVXkjPrYHabGNYU8YWnjm8KTDxpWQPI9s8TDwk4hk8E/y 6lFHa0/zOIL09HXFFrGy8W+cD0o5tOH9SlBvmJsVg3wFIO6goR0kew0xAsCVfybZ teXO14KDmcPBBCP4NfPkYrdeGwalaiCLQaCBLg+ZWNGxXSTMLO0vBUX+PDdR1u/e VT804DXqeY7CnoS50VSGCGQfdqL+Oslz+gaiFanRx8E2mTjEDO4e+KeMj9kNH2B0 6KbPZ5lIIIs8dgreprB/ =KnUE -----END PGP SIGNATURE----- --=-=-=-- --===============1981204706== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1981204706==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 14:35:08 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TCZ13I016308; Mon, 29 Mar 2010 14:35:07 +0200 Received: from web8701.mail.in.yahoo.com (web8701.mail.in.yahoo.com [203.84.221.122]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2TCYxQi016262 for ; Mon, 29 Mar 2010 14:34:59 +0200 Received: (qmail 75577 invoked by uid 60001); 29 Mar 2010 12:34:57 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1269866097; bh=VOE7JOdSrcLzL4oTbFIRfXcitW/6pYKwxcjsQnZg3iM=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=L6j4jK0vjch8RIuOI5dnnF0pwq+TADQeyGj6/4q78+8ECOPB1Vv2EqqNyyf4/2giaMVyDQofdJesEUPreuhGZtGLC/kx8KXvgCpPdUgLCijkXTufEDzWI/TW/pN+DQDHzdB7pPU2HLFXfRidyoz6tiPhMsKiLm+1wwcvYosV6/M= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=iU6BNh1+2ipOFLLGO/fhA8nHo9e/v+rVYLvdQqL7POLbaU7O0R68FU5dPzBI+uArjhzqGNifpUglPImpD0BfnrldZIoc4pSHgQQlCXs581wXRueJs+BSvgqsTsZAeIZNrUCw5hRVXAIHun4bt2CT/Wo736Ai1HJ84nQa07uOO0g=; Message-ID: <666991.75542.qm@web8701.mail.in.yahoo.com> X-YMail-OSG: HCIjtg8VM1mmwubajwWjveZ7VdF1mI5KATT_uK1UNwv32h1 PlDnCl9Tw7iuK3qATiHq.Dw1XBfQY4TakNr1GxtZ0YTW1iSl6hnHt4mGyeya sjUGBDKVo1pQY910a.UT2RXFl7DdQVlBqPO6OMd2VQThxRFU9GsQxLNK1Td_ sNPL3uVVl.JU33Be8PYk90zJ1z66Ab3JVMcW1MjQHXV02oytdEw4zhDJinps 4osATe02kDp86OuOKi2hOL8RZ9kZberehXoFd_lipGt9lVdcqgJE_ITdwX23 Yme001wa.xLuhUBF3lGd8zmunTZLrUNoxiz8MzrYRP9Sv5Q-- Received: from [220.225.69.166] by web8701.mail.in.yahoo.com via HTTP; Mon, 29 Mar 2010 18:04:57 IST X-Mailer: YahooMailRC/324.3 YahooMailWebService/0.8.100.260964 Date: Mon, 29 Mar 2010 18:04:57 +0530 (IST) From: suyog jadhav Subject: pubkey auth needs to verify supplied passphrase To: libssh2-devel@cool.haxx.se MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0473372391==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0473372391== Content-Type: multipart/alternative; boundary="0-109855144-1269866097=:75542" --0-109855144-1269866097=:75542 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi,=0A=0AThe scenario is like the follows,=0A=0AI have a key which has pass= phrase as NULL(no passphrase).=0ANow I give it as parameter to libssh2_user= auth_publickey_fromfile along with NOT-NULL(anything) passphrase.=0AThe pub= key auth succeeds in this case,as there is nowhere check for this scenario.= =0A=0AI understand that default ssh client doesn't ask for passphrase for s= uch key,which was the requirement for such case,I think.=0A=0AAnyway,my app= lication asks user to configure these values (pubkey path,passphrase).Now i= f user has given SOME passphrase for a key which doesn't require passphrase= , all the operations such as login and command execution succeeds contrary = to wrong configuration.=0AThere is no way to detect this and notify user.= =0A=0AI am trying to find a way around this,but could anybody please sugges= t where is the appropriate place to put such check and transmit that error = back?=0A=0AI have also raised a ticket http://libssh2.stuge.se/ticket/169 f= or a fix similar to this problem.=0A=0ARegards=0ASuyog=0A=0A=0A=0A The= INTERNET now has a personality. YOURS! See your Yahoo! Homepage. http://in= .yahoo.com/ --0-109855144-1269866097=:75542 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
Hi,

The scenario is like the follows,

I have a key wh= ich has passphrase as NULL(no passphrase).
Now I give it as parameter to= libssh2_userauth_publickey_fromfile along with NOT-NULL(anything) passphra= se.
The pubkey auth succeeds in this case,as there is nowhere check for = this scenario.

I understand that default ssh client doesn't ask for = passphrase for such key,which was the requirement for such case,I think.
Anyway,my application asks user to configure these values (pubkey path= ,passphrase).Now if user has given SOME passphrase for a key which doesn't = require passphrase, all the operations such as login and command execution = succeeds contrary to wrong configuration.
There is no way to detect this= and notify user.

I am trying to find a way around this,but could anybody please suggest where is the appropriate place to put such check an= d transmit that error back?

I have also raised a ticket http://libssh2.s= tuge.se/ticket/169 for a fix similar to this problem.

Reg= ards
Suyog
=0A

=0A
=0A= Your Mail works best with the New Yahoo Optimized IE8. Get it NOW!. --0-109855144-1269866097=:75542-- --===============0473372391== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0473372391==-- From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 15:25:40 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TDPSYX031736; Mon, 29 Mar 2010 15:25:39 +0200 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TDPRiu031702 for ; Mon, 29 Mar 2010 15:25:27 +0200 Date: Mon, 29 Mar 2010 15:25:27 +0200 (CEST) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: pubkey auth needs to verify supplied passphrase In-Reply-To: <666991.75542.qm@web8701.mail.in.yahoo.com> Message-ID: References: <666991.75542.qm@web8701.mail.in.yahoo.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 29 Mar 2010, suyog jadhav wrote: > I have a key which has passphrase as NULL(no passphrase). > Now I give it as parameter to libssh2_userauth_publickey_fromfile along with > NOT-NULL(anything) passphrase. The pubkey auth succeeds in this case,as > there is nowhere check for this scenario. You could easily first try with a blank passphrase to see if this is indeed the case. > I understand that default ssh client doesn't ask for passphrase for such > key,which was the requirement for such case,I think. The "default ssh client" ? Are you referring to the openssh tool? > I have also raised a ticket http://libssh2.stuge.se/ticket/169 for a fix > similar to this problem. Sorry, but ticket 169 is not a fix. It simply says you crash libssh2 by passing in a NULL pointer instead of a pointer to a passphrase. I agree that the man page doesn't say explicitly what is allowed there but it also doesn't say that NULL is a legitimate input. I still think we should check for a NULL pointer to make it more robust. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 17:21:30 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TFLGvI004135; Mon, 29 Mar 2010 17:21:28 +0200 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TFLEub004112 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 29 Mar 2010 17:21:14 +0200 Received: (qmail 18709 invoked from network); 29 Mar 2010 15:21:19 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 29 Mar 2010 15:21:19 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 29 Mar 2010 15:21:19 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #169: segmentation fault in pubkey authentication X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/169#comment:1 Message-ID: <056.da5199a699a8a4069637ab45ca8b0844@libssh2.stuge.se> References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-Trac-Ticket-ID: 169 In-Reply-To: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #169: segmentation fault in pubkey authentication --------------------------+------------------------------------------------- Reporter: suyogjadhav | Owner: Type: defect | Status: closed Priority: high | Milestone: 1.2.5 Component: API | Version: 1.2.4 Resolution: fixed | Keywords: Blocking: | Blockedby: --------------------------+------------------------------------------------- Changes (by bagder): * status: new => closed * resolution: => fixed * milestone: => 1.2.5 Comment: Thanks, this is now fixed in git! -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 18:43:27 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TGhEFa011193; Mon, 29 Mar 2010 18:43:26 +0200 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TGhDSA011176 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 29 Mar 2010 18:43:13 +0200 Received: (qmail 23386 invoked from network); 29 Mar 2010 16:43:18 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 29 Mar 2010 16:43:18 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 29 Mar 2010 16:43:18 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #169: segmentation fault in pubkey authentication X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/169#comment:2 Message-ID: <056.206a4f6e79a8d8a1e0d8b09ac1811c34@libssh2.stuge.se> References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-Trac-Ticket-ID: 169 In-Reply-To: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #169: segmentation fault in pubkey authentication --------------------------+------------------------------------------------- Reporter: suyogjadhav | Owner: Type: defect | Status: closed Priority: high | Milestone: 1.2.5 Component: API | Version: 1.2.4 Resolution: fixed | Keywords: Blocking: | Blockedby: --------------------------+------------------------------------------------- Comment(by stuge): Replying to [comment:1 bagder]: > Thanks, this is now fixed in git! I'm with Mikhail on this one: garbage in => garbage out. I don't think it is the job of libssh2 to validate every string pointer that applications pass in to the library. I'd moderately prefer the commit to be reverted. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 18:46:56 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TGks5P016640; Mon, 29 Mar 2010 18:46:56 +0200 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TGkqEu016605 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Mon, 29 Mar 2010 18:46:52 +0200 Received: (qmail 23876 invoked from network); 29 Mar 2010 16:46:58 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 29 Mar 2010 16:46:58 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Mon, 29 Mar 2010 16:46:58 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #169: segmentation fault in pubkey authentication X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/169#comment:3 Message-ID: <056.47f16ecfda0cfb8c783b62ccc1d8cf2e@libssh2.stuge.se> References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-Trac-Ticket-ID: 169 In-Reply-To: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #169: segmentation fault in pubkey authentication --------------------------+------------------------------------------------- Reporter: suyogjadhav | Owner: Type: defect | Status: closed Priority: high | Milestone: 1.2.5 Component: API | Version: 1.2.4 Resolution: fixed | Keywords: Blocking: | Blockedby: --------------------------+------------------------------------------------- Comment(by alamaison): I've got to say I would find it very strange to revert a fix that makes using our library ''safer''. Ideally, it should never be possible to crash no matter what inputs a component receives. -1 for reverting. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 19:43:48 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2THhYDH031110; Mon, 29 Mar 2010 19:43:46 +0200 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2THhXme031097 for ; Mon, 29 Mar 2010 19:43:33 +0200 Date: Mon, 29 Mar 2010 19:43:33 +0200 (CEST) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: [libssh2] #169: segmentation fault in pubkey authentication In-Reply-To: <056.47f16ecfda0cfb8c783b62ccc1d8cf2e@libssh2.stuge.se> Message-ID: References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> <056.47f16ecfda0cfb8c783b62ccc1d8cf2e@libssh2.stuge.se> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 29 Mar 2010, libssh2 Trac wrote: > Comment(by alamaison): > > I've got to say I would find it very strange to revert a fix that makes > using our library ''safer''. Ideally, it should never be possible to crash > no matter what inputs a component receives. > > -1 for reverting. I don't quite understand that resistance either. As I said before: this function is not documented clearly to *not* accept a NULL for a blank passphrase even though it isn't documented to accept it either. Given that small uncertainty in API funcionality I think it is quite easy to check against this specific case. It is not the same as to say that we always should check all arguments for junk or whatever. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Mon Mar 29 22:08:25 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TK8A4h023021; Mon, 29 Mar 2010 22:08:22 +0200 Received: from idcmail-mo2no.shaw.ca (idcmail-mo2no.shaw.ca [64.59.134.9]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2TK86U1022889 for ; Mon, 29 Mar 2010 22:08:07 +0200 Received: from pd6ml1no-ssvc.prod.shaw.ca ([10.0.153.160]) by pd7mo1no-svcs.prod.shaw.ca with ESMTP; 29 Mar 2010 14:08:07 -0600 X-Cloudmark-SP-Filtered: true X-Cloudmark-SP-Result: v=1.0 c=1 a=ht8ysHsmaUMA:10 a=VphdPIyG4kEA:10 a=_4SF6Xvc910A:10 a=kj9zAlcOel0A:10 a=y9ASP8Sd+Sa59kG1URKZZw==:17 a=xjc_6C1fz1zZj-I1jxQA:9 a=7xFtrR-sxFHqt-dqrFI7SnfhCVMA:4 a=CjuIK1q_8ugA:10 Received: from unknown (HELO mail.xepol.com) ([68.146.216.64]) by pd6ml1no-dmz.prod.shaw.ca with ESMTP; 29 Mar 2010 14:08:07 -0600 Received: from quadx64 [192.168.1.3] by mail.xepol.com with SMTP (EHLO Quadx64) (ArGoSoft Mail Server Pro for WinNT/2000/XP, Version 1.8 (1.8.9.2)); Mon, 29 Mar 2010 14:08:01 -0600 From: "C Johnson" To: "'libssh2 development'" References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se><056.47f16ecfda0cfb8c783b62ccc1d8cf2e@libssh2.stuge.se> In-Reply-To: Subject: RE: [libssh2] #169: segmentation fault in pubkey authentication Date: Mon, 29 Mar 2010 14:07:59 -0600 Message-ID: <613465C4B81A4A52833D08389F8CBDA0@Quadx64> MIME-Version: 1.0 X-Mailer: Microsoft Office Outlook 11 Thread-Index: AcrPaBuYmPQloWh/RxyMM4CT6Pc30wAEo86A X-MimeOLE: Produced By Microsoft MimeOLE V6.1.7600.16385 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Mon, 29 Mar 2010, Daniel Stenberg wrote: > I don't quite understand that resistance either. As I said > before: this function is not documented clearly to *not* > accept a NULL for a blank passphrase even though it isn't > documented to accept it either. Given that small uncertainty > in API funcionality I think it is quite easy to check against > this specific case. It would seem best to always clarify any points of uncertainty when ever possible. > It is not the same as to say that we always should check all > arguments for junk or whatever. I would not rule this out. Obviously it can be taken to absurd extremes, but a reasonable step to sanitize inputs can be very important. As a library, you can not be sure of the source of the inputs being sent to your library - did the application sanitize them enough or does doing nothing leave open a potential exploit? There has to be a balance somewhere between so sanitizing so insanly it drags everything to a griding halt and just leaving the barn door open and hoping no one notices or cares. - C Johnson _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 07:30:00 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2U5Tkif013012; Tue, 30 Mar 2010 07:29:56 +0200 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2U5TigV013006 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 30 Mar 2010 07:29:44 +0200 Received: (qmail 7323 invoked from network); 30 Mar 2010 05:29:44 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 30 Mar 2010 05:29:44 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Tue, 30 Mar 2010 05:29:44 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #169: segmentation fault in pubkey authentication X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/169#comment:4 Message-ID: <056.2e0823cea00086bca60e4cc919fd5419@libssh2.stuge.se> References: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-Trac-Ticket-ID: 169 In-Reply-To: <047.32f58e398ecd88b31a936adc4483abb9@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #169: segmentation fault in pubkey authentication --------------------------+------------------------------------------------- Reporter: suyogjadhav | Owner: Type: defect | Status: closed Priority: high | Milestone: 1.2.5 Component: API | Version: 1.2.4 Resolution: fixed | Keywords: Blocking: | Blockedby: --------------------------+------------------------------------------------- Comment(by suyogjadhav): just to point out that...these values are taken from user...and there is no way to know beforehand if the key requires a valid passphrase or null.If there is some way to do this please let me know. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 08:19:55 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2U6JnIv014870; Tue, 30 Mar 2010 08:19:54 +0200 Received: from mail-pv0-f182.google.com (mail-pv0-f182.google.com [74.125.83.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2U6JlXb014833 for ; Tue, 30 Mar 2010 08:19:47 +0200 Received: by pvc7 with SMTP id 7so5061006pvc.41 for ; Mon, 29 Mar 2010 23:19:43 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:received:message-id:subject:from:to:content-type; bh=SIKNnzirOgJY7m1tulz/yhZKar7b4dFfl3cuVJy+TcQ=; b=YHb0T5LWoSS3BWxwB0ke9lAHMSaMgPDolJPB7d5H1uZ7KMOSfVS0cV9XhB1BTYCpdf 7TeQJXZGP4mlNSfNbbo8SY37fMO1zQKw2UdIFSfQ8iG2h/FRdqEhOQHw0FvkvCyY21sc 02M785Of1B+t4EPD2slDii+j8wmvm/tLOwP/I= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; b=XADxN/Qs2Wzezk5myEGQ4znQeUlvuyQzINsP/WwqbLZ3duP3hzJaHeAewvjYwCdSD5 smW/E4bRXm0ASpyd0CEBQuGOL2lw/hoxoFOY8zmDL4n/rDAd2iZTUOp85vJy3/oiItAY eeeEw8srpom45HPlDJj+wwHBQJ9rNaETTPmJY= MIME-Version: 1.0 Received: by 10.142.213.13 with HTTP; Mon, 29 Mar 2010 23:19:43 -0700 (PDT) In-Reply-To: References: Date: Tue, 30 Mar 2010 11:49:43 +0530 Received: by 10.142.119.33 with SMTP id r33mr2325334wfc.213.1269929983238; Mon, 29 Mar 2010 23:19:43 -0700 (PDT) Message-ID: Subject: Re: Character conversion for sftp From: NAGASEKHAR BHUVANAGIRI To: libssh2-devel@cool.haxx.se X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============0669100897==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se --===============0669100897== Content-Type: multipart/alternative; boundary=001636e8ff918a638d0482fe9d09 --001636e8ff918a638d0482fe9d09 Content-Type: text/plain; charset=ISO-8859-1 On Mon, Mar 29, 2010 at 1:21 PM, Daniel Stenberg wrote: On Mon, 29 Mar 2010, NAGASEKHAR BHUVANAGIRI wrote: Could someone advice me how the character conversion can be handled for sftp? (I removed the libssh2 list as a receiver, this is a libcurl question) Why would libcurl do any conversion of data when doing SFTP? SFTP is defined as a pure binary transfer and you get the data exactly the way the server sends it! If you want to convert it somehow, then that responsibility is yours. -- / daniel.haxx.se ------------------------------------------------------------------- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html The conversion is for control responses and not for data. The reponse received by banner_receive method in (libssh2's session.c) is garbled and gived PACKET_FAIL error. This issue is specific to z/OS. Thanks, Nagasekhar Aha, but then this issue is *competely* about libssh2 internals and the libssh2-devel is the proper place for this discussion. I don't think we have any support for such converions in libssh2 so it will probably take some effort to get it added. -- / daniel.haxx.se ------------------------------------------------------------------- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html Is anyone having similar problem with conversion of characters for control responses with sftp? Thanks, --001636e8ff918a638d0482fe9d09 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

=A0

On Mon, Mar 29, 2010 at 1:21 PM, Daniel Stenberg <daniel@haxx.se> wrote:

On Mon, 29 Mar 2010, NAGASEKHAR BHUVANAGIRI wrote:


Could someone advice me how the character conversion can be handled = for sftp?

=A0

(I removed the libssh2 list as a receiver, this is a libcurl question)

Why would libcurl do any conversion of data when doing SFTP? SFTP is def= ined as a pure binary transfer and you get the data exactly the way the ser= ver sends it!

If you want to convert it somehow, then that responsibility is yours.

--

=A0/ daniel.haxx.se
-----------= --------------------------------------------------------
List admin: http://cool.haxx.se= /list/listinfo/curl-library
Etiquette:=A0 http://cu= rl.haxx.se/mail/etiquette.html

=A0
=A0
The conversion is for control responses and not for data. = The reponse received by banner_receive method in (libssh2's session.c) = is garbled and gived PACKET_FAIL error. This issue is specific to z/OS. Thanks,
Nagasekhar

=A0
Aha, but then this issue is *competely* about libssh2 internals and th= e libssh2-devel is the proper place for this discussion.
I don't think we have any support for such converions in libssh2 s= o it will probably take some effort to get it added.
=A0
--
=A0/ daniel.haxx.se
---------= ----------------------------------------------------------
List admin: <= a href=3D"http://cool.haxx.se/list/listinfo/curl-library">http://cool.haxx.= se/list/listinfo/curl-library
Etiquette:=A0 http://cu= rl.haxx.se/mail/etiquette.html


Is anyone having similar prob= lem with conversion of characters for control responses with sftp?
Thank= s,
--001636e8ff918a638d0482fe9d09-- --===============0669100897== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============0669100897==-- From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 09:01:16 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2U717hT013328; Tue, 30 Mar 2010 09:01:15 +0200 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2U715k5013288 for ; Tue, 30 Mar 2010 09:01:05 +0200 Date: Tue, 30 Mar 2010 09:01:05 +0200 (CEST) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Character conversion for sftp In-Reply-To: Message-ID: References: User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 30 Mar 2010, NAGASEKHAR BHUVANAGIRI wrote: > Could someone advice me how the character conversion can be handled for > sftp? How about you making a suggestion on how you think we should do it? Also, I'm sure not everyone here even knows what character conversion you speak about so please try to explain a bit more! -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 16:24:32 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UEOH4e029064; Tue, 30 Mar 2010 16:24:29 +0200 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2UEOFLV028986 for ; Tue, 30 Mar 2010 16:24:15 +0200 Received: (qmail 18259 invoked by uid 501); 30 Mar 2010 14:24:19 -0000 Message-ID: <20100330142419.18258.qmail@stuge.se> Date: Tue, 30 Mar 2010 16:24:18 +0200 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Character conversion for sftp Mail-Followup-To: libssh2-devel@cool.haxx.se References: MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hi Nagasekhar, NAGASEKHAR BHUVANAGIRI wrote: > The conversion is for control responses and not for data. The > reponse received by banner_receive method in (libssh2's session.c) > is garbled and gived PACKET_FAIL error. This issue is specific to > z/OS. Can you please provide a libssh2 trace of a failed connection, using one of the example programs which are included with libssh2? Note that this may require you to build libssh2 on your own, since trace output is sometimes not enabled in libssh2 packages made by others. Ideally I would like you to create a ticket for this issue at http://libssh2.stuge.se/ but at the very least please try to send trace output, so that we can try to identify the problem and then fix it. Thanks //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 18:01:19 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UG18Gx008029; Tue, 30 Mar 2010 18:01:17 +0200 Received: from listserv.isdn4linux.de (listserv.isdn4linux.de [87.106.208.227]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UG17Ih008011 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 30 Mar 2010 18:01:08 +0200 Received: from fsun.fe.think (dslb-084-057-170-205.pools.arcor-ip.net [84.57.170.205]) (authenticated bits=128) by listserv.isdn4linux.de (8.14.3/8.14.3) with ESMTP id o2UG1Au7030542 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Tue, 30 Mar 2010 18:01:11 +0200 X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.95.3 at listserv.isdn4linux.de Received: from [192.168.2.25] (fritz.fe.think [192.168.2.25]) (authenticated bits=0) by fsun.fe.think (8.14.1/8.14.1) with ESMTP id o2UG1AHl019697 for ; Tue, 30 Mar 2010 18:01:10 +0200 Message-ID: <4BB22045.5000803@fritz-elfert.de> Date: Tue, 30 Mar 2010 18:01:09 +0200 From: Fritz Elfert User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: libssh2 development Subject: Re: Character conversion for sftp References: <20100330142419.18258.qmail@stuge.se> In-Reply-To: <20100330142419.18258.qmail@stuge.se> X-Enigmail-Version: 0.96.0 OpenPGP: id=0332A6B0 X-Virus-Scanned: ClamAV 0.92.1/10665/Tue Mar 30 14:20:33 2010 on fsun.fe.think X-Virus-Status: Clean X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.3 (listserv.isdn4linux.de [87.106.208.227]); Tue, 30 Mar 2010 18:01:11 +0200 (CEST) X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,UNPARSEABLE_RELAY autolearn=ham version=3.3.0-i4l X-Spam-Checker-Version: SpamAssassin 3.3.0-i4l (2010-01-18) on listserv.isdn4linux.de X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi guys, Apologies for my intrusion here (im lurking on libbs2-devel for some time now) ... Peter Stuge schrieb: > Hi Nagasekhar, > > NAGASEKHAR BHUVANAGIRI wrote: >> The conversion is for control responses and not for data. The >> reponse received by banner_receive method in (libssh2's session.c) >> is garbled and gived PACKET_FAIL error. To me, this looks more like an EBCDIC vs. ASCII problem rather than a problem with libssh2 code itself. What rings a bell here, is this: >> This issue is specific to >> z/OS. On z/OS, source code is treated as EBCDIC by default, so if you compile something like const char *str = "foo"; the resulting string constant is *not* 0x66 0x6f 0x6f and if you transfer that over the wire to some ASCII-based system it *looks* like garbage. IMHO, the easiest way of porting to z/OS is to treat the whole source code as ASCII and then use special ioctls (provided by IBM's libc) to enable auto-conversion back to EBCDIC for stdin, stdout and stderr (if those are connected to a terminal). My last z/OS porting work is almost a year ago, so I would have too look up the exact cc options and those ioctls. There is extensive documentation on that topic available at: http://www-03.ibm.com/systems/z/os/zos/features/unix/bpxa1p03.html Cheers -Fritz -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iD8DBQFLsiBFboM4mAMyprARAhLOAKCctO3GEOYZE20vRRuAFLaqNXeUsACePELa KFbi5rFFlt3fArLzVyy1NRI= =Vftm -----END PGP SIGNATURE----- _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 18:17:23 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UGHIIf021103; Tue, 30 Mar 2010 18:17:22 +0200 Received: from foo.birdnet.se (qmailr@foo.birdnet.se [213.88.146.6]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with SMTP id o2UGHHEi021093 for ; Tue, 30 Mar 2010 18:17:17 +0200 Received: (qmail 4125 invoked by uid 501); 30 Mar 2010 16:17:22 -0000 Message-ID: <20100330161722.4124.qmail@stuge.se> Date: Tue, 30 Mar 2010 18:17:22 +0200 From: Peter Stuge To: libssh2-devel@cool.haxx.se Subject: Re: Character conversion for sftp Mail-Followup-To: libssh2-devel@cool.haxx.se References: <20100330142419.18258.qmail@stuge.se> <4BB22045.5000803@fritz-elfert.de> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <4BB22045.5000803@fritz-elfert.de> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Fritz Elfert wrote: > IMHO, the easiest way of porting to z/OS is to treat the whole source > code as ASCII and then use special ioctls (provided by IBM's libc) to > enable auto-conversion back to EBCDIC for stdin, stdout and stderr (if > those are connected to a terminal). Can we somehow make this easier by changing libssh2 or the build system? //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 18:45:32 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UGjQkw017490; Tue, 30 Mar 2010 18:45:31 +0200 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UGjP8K017469 for ; Tue, 30 Mar 2010 18:45:25 +0200 Date: Tue, 30 Mar 2010 18:45:25 +0200 (CEST) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Character conversion for sftp In-Reply-To: <4BB22045.5000803@fritz-elfert.de> Message-ID: References: <20100330142419.18258.qmail@stuge.se> <4BB22045.5000803@fritz-elfert.de> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Tue, 30 Mar 2010, Fritz Elfert wrote: > To me, this looks more like an EBCDIC vs. ASCII problem rather than a > problem with libssh2 code itself. What rings a bell here, is this: Yes, that's what he's talking about and that's why I asked him to clarify, since it was far from obvious in his first mail here. He posted to the libcurl list first where I read it (and directed him here) so I knew somewhat more what he was coming from. libcurl offers a callback for charset translations, but as I'm not a EBCDIC guy myself I'd much rather hear from someone who uses such systems on how a suitable solution for libssh2 would look like. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 19:33:00 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UHWmCx018969; Tue, 30 Mar 2010 19:32:58 +0200 Received: from listserv.isdn4linux.de (listserv.isdn4linux.de [87.106.208.227]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2UHWGLM018911 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 30 Mar 2010 19:32:17 +0200 Received: from fsun.fe.think (dslb-084-057-170-205.pools.arcor-ip.net [84.57.170.205]) (authenticated bits=128) by listserv.isdn4linux.de (8.14.3/8.14.3) with ESMTP id o2UHWK3J001063 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Tue, 30 Mar 2010 19:32:21 +0200 X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.95.3 at listserv.isdn4linux.de Received: from [192.168.2.25] (fritz.fe.think [192.168.2.25]) (authenticated bits=0) by fsun.fe.think (8.14.1/8.14.1) with ESMTP id o2UHWK9h021504 for ; Tue, 30 Mar 2010 19:32:20 +0200 Message-ID: <4BB235A3.3070308@fritz-elfert.de> Date: Tue, 30 Mar 2010 19:32:19 +0200 From: Fritz Elfert User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: libssh2 development Subject: Re: Character conversion for sftp References: <20100330142419.18258.qmail@stuge.se> <4BB22045.5000803@fritz-elfert.de> <20100330161722.4124.qmail@stuge.se> In-Reply-To: <20100330161722.4124.qmail@stuge.se> X-Enigmail-Version: 0.96.0 OpenPGP: id=0332A6B0 X-Virus-Scanned: ClamAV 0.92.1/10669/Tue Mar 30 17:45:31 2010 on fsun.fe.think X-Virus-Status: Clean X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.3 (listserv.isdn4linux.de [87.106.208.227]); Tue, 30 Mar 2010 19:32:21 +0200 (CEST) X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,UNPARSEABLE_RELAY autolearn=ham version=3.3.0-i4l X-Spam-Checker-Version: SpamAssassin 3.3.0-i4l (2010-01-18) on listserv.isdn4linux.de X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Stuge schrieb: > Fritz Elfert wrote: >> IMHO, the easiest way of porting to z/OS is to treat the whole source >> code as ASCII and then use special ioctls (provided by IBM's libc) to >> enable auto-conversion back to EBCDIC for stdin, stdout and stderr (if >> those are connected to a terminal). > > Can we somehow make this easier by changing libssh2 or the build > system? I'm afraid there's no really simple solution - especially when dealing with a library. At the moment, i don't have a z/OS system available. I can start digging in some backups of old porting projects though ... just give me some time ... Here's what i remember right now: 1. There are #pragma directives for switching encoding of the compiler: #pragma convert(819) const char *str1 = "hello"; /* ASCII string const. */ #pragma convert(37) const char *str1 = "hello"; /* EBCDIC string const. */ 2. On the filesystem level, you can tag source files to be interpreted as ASCII using the command chtag -tc ISO8859-1 mysource.c But then, you would need a special ASCII editor like viascii to edit them. 3. The stdio functions use EBCDIC by default, e.g: if fgets() stumbles over an ASCII-'\n', it would *not* see an end-of-line. 4. There are ioctl's which can switch a file descriptor into auto-convert mode (sorry - can't remember the exact names/syntax). Cheers -Fritz -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iD8DBQFLsjWjboM4mAMyprARAgPLAJ9+MgwazBVaQI1WSVO++aRF0UOt5QCcCo7I 5YpCnmME4kynxStZR0/qavw= =j9k8 -----END PGP SIGNATURE----- _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Tue Mar 30 23:55:21 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ULt7VW016270; Tue, 30 Mar 2010 23:55:19 +0200 Received: from earth.stuge.se (earth.stuge.se [213.88.146.15]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2ULt55F016259 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Tue, 30 Mar 2010 23:55:06 +0200 Received: (qmail 15497 invoked from network); 30 Mar 2010 21:55:12 -0000 Received: from unknown (HELO earth.stuge.se) (127.0.0.1) by localhost with SMTP; 30 Mar 2010 21:55:12 -0000 MIME-Version: 1.0 From: "libssh2 Trac" X-Trac-Version: 0.11.6 Precedence: bulk Cc: libssh2-devel@cool.haxx.se Auto-Submitted: auto-generated X-Mailer: Trac 0.11.6, by Edgewall Software X-Trac-Project: libssh2 Date: Tue, 30 Mar 2010 21:55:12 -0000 X-URL: http://libssh2.stuge.se/ Subject: Re: [libssh2] #168: Memory leak in libssh2_session_startup X-Trac-Ticket-URL: http://libssh2.stuge.se/ticket/168#comment:1 Message-ID: <053.e634b0f6161a2ab4223be96ea0b7626b@libssh2.stuge.se> References: <044.2a48d5b8cb5be2e3469617435c23e879@libssh2.stuge.se> X-Trac-Ticket-ID: 168 In-Reply-To: <044.2a48d5b8cb5be2e3469617435c23e879@libssh2.stuge.se> X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Reply-To: trac@libssh2.stuge.se, libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se #168: Memory leak in libssh2_session_startup ----------------------+----------------------------------------------------- Reporter: jcollins | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: API | Version: 1.2.4 Keywords: | Blocking: Blockedby: | ----------------------+----------------------------------------------------- Comment(by bagder): I can confirm this. It seems to be a bug in gcrypt (I use 1.4.5-2 on debian) since I get this easily with ./example/scp against localhost (control-c after the fingerprint) but doing the exact same operation with libssh2 built to use OpenSSL show no leak at all. Another reason could be there's something missing in the gcrypt glue layer. -- Ticket URL: libssh2 C library for writing portable SSH2 clients _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 31 17:27:37 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VFROVt019332; Wed, 31 Mar 2010 17:27:34 +0200 Received: from p02c11o149.mxlogic.net (p02c11o149.mxlogic.net [208.65.144.82]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VFRKKq019317 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 31 Mar 2010 17:27:21 +0200 Received: from unknown [64.62.199.48] (EHLO mail02.netenrich.com) by p02c11o149.mxlogic.net(mxl_mta-6.5.0-2) over TLS secured channel with ESMTP id dd963bb4.0.297.00-014.646.p02c11o149.mxlogic.net (envelope-from ); Wed, 31 Mar 2010 09:27:27 -0600 (MDT) X-MXL-Hash: 4bb369df64229bf3-02768e73adf80e4b169ad57ce9a72c900751f485 Received: from EX22.netenrich.com (172.22.2.30) by EX02.netenrich.com (172.20.1.31) with Microsoft SMTP Server (TLS) id 8.1.393.1; Wed, 31 Mar 2010 08:27:22 -0700 Received: from EX22.netenrich.com ([::1]) by EX22.netenrich.com ([::1]) with mapi; Wed, 31 Mar 2010 08:27:19 -0700 From: Kalyan Polasi To: "libssh2-devel@cool.haxx.se" Date: Wed, 31 Mar 2010 08:27:13 -0700 Subject: Issues with using LibSSH2 Thread-Topic: Issues with using LibSSH2 Thread-Index: AcrQ5qOBG2DBSX4qRUmbSKuVBg+Z1g== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-Processed-By: Rebuild v2.0-0 X-Spam: [F=0.2000000000; CM=0.500; S=0.200(2010032501)] X-MAIL-FROM: X-SOURCE-IP: [64.62.199.48] X-AnalysisOut: [v=1.0 c=1 a=VphdPIyG4kEA:10 a=ybBoJlpP49CkjQP4YFKjpg==:17 ] X-AnalysisOut: [a=adCLvFn2mXbKYi-CvlIA:9 a=BDbi9nXxX7b1bz9DiAgA:7 a=_wwtZ3] X-AnalysisOut: [TBY4sn3JIqHrJUxw_3fiAA:4 a=CjuIK1q_8ugA:10 a=yMhMjlubAAAA:] X-AnalysisOut: [8 a=SSmOFEACAAAA:8 a=h2RiTz9kBJn2_jQXfVkA:9 a=9TeZ7yNyjUSr] X-AnalysisOut: [PLZx4kkA:7 a=mvdJmiQTXk--iSWCwG8BwEPDCpUA:4] X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============1016381836==" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se This is a multi-part MIME message. --===============1016381836== Content-Language: en-US Content-Type: multipart/alternative; boundary="=_reb-r4569251E-t4BB369DF" This is a multi-part MIME message. --=_reb-r4569251E-t4BB369DF Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hello Team, We have a requirement to do Port-forwarding. For that we thought of using L= ibSSH2. We tried to compile the LIBBSSH2 source code of the version 2-0.9, which we= are able to do it successfully after adding the external libraries from Op= enSSL and zLib. Now we have release version libssh2.lb file. After that we tried to start with the example "direct_tcpip.c" which is av= ailable in the site. But whenever we tried to call any of the libssh2 defi= nitions, we are getting into linker errors, which we were not getting the s= olution to fix this. I am pasting the code and the errors below for your reference: /*Didn't pasted the header files information here, which are same as the ex= ample in the site*/ int main(int argc, char *argv[]) { int rc; LIBSSH2_SESSION *session; #ifdef WIN32 char sockopt; WSADATA wsadata; WSAStartup(MAKEWORD(2,0), &wsadata); #else int sockopt; #endif rc =3D libssh2_init (0); /* Create a session instance */ session =3D libssh2_session_init(); } 1. If I try to call the function "libssh2_init(0);", I am getting the belo= w error "Error 2 error C3861: 'libssh2_init': identifier not found" 2. And if I try to call the function "libssh2_session_init()", I am g= etting linker error as shown below. "Error 4 error LNK2001: unresolved external symbol "struct = _LIBSSH2_SESSION * __cdecl libssh2_session_init_ex(void * (__cdecl*)(unsign= ed int,void * *),void (__cdecl*)(void *,void * *),void * (__cdecl*)(voi= d *,unsigned int,void * *),void *)" (?libssh2_session_init_ex@@YAPAU_LIBSSH= 2_SESSION@@P6APAXIPAPAX@ZP6AXPAX0@ZP6APAX2I0@Z2@Z)" This is the case for whatever function I call from my example program. Please help us in getting this issue fixed. We are using VS2005 on Windows XP Professional. Regards, Kalyan.Polasi P Please consider the environment before printing this e-mail. --------------------------------------------------=20 Confidentiality And Disclaimer Notice=20 Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may cont= ain proprietary, confidential or privileged information. If you are not the int= ended recipient, you should not disseminate, distribute or copy this e-mail. Plea= se notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. ……www.netenrich.com=20 ------------------------------------------------- --=_reb-r4569251E-t4BB369DF Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hello Team,

 

We have a requirement to do Port-forwarding. For that = we thought of using LibSSH2.

 

We tried to compile the LIBBSSH2 source code of the ve= rsion 2-0.9, which we are able to do it successfully after adding the external libra= ries from OpenSSL and zLib.

 

Now we have release version libssh2.lb file.

 

After that we tried to start with the example “d= irect_tcpip.c”  which is available in the site.  But whenever we tried to call a= ny of the libssh2 definitions, we are getting into linker errors, which we were not getting the solution to fix this.

 

I am pasting the code and the errors below for your reference:

 

/*Didn’t pasted the header files information here, which are same as the example in the site*/

int main(int argc= , char *argv[])

{

    in= t rc;

    LIBSSH2_SESSION *session;

#ifdef WIN32

    ch= ar sockopt;

    WSADATA wsadata; <= /span>

    WSAStartup(MAKEWORD(2,0), &wsadata);

#else

    int sockopt;<= /o:p>

#endif 

    rc =3D libssh2_init (0);   

    /* Create a session instance */

    session =3D libssh2_session_i= nit();

}

 

1.  If I try to call the function “libssh2_init(0);”, I am getting the below error<= /o:p>

 

      “Error 2 error C3861: 'libssh2_init': identifier not found“

 

2.&n= bsp;      And if I try to call the function “libssh2_session_init()= ”, I am getting linker error as shown below.

 

         =       “Error 4  error LNK2001: unresolved external symbol "struct _LIBSSH2_SESSION * __cdecl libssh2_session_init_ex(void * (__cdecl*)(unsign= ed int,void *     *),void (__cdecl*)(void *,void * *),void= * (__cdecl*)(void *,unsigned int,void * *),void *)" (?libssh2_session_init_ex@@YAPAU_LIBSSH2_SESSION@@P6APAXIPAPAX@ZP6AXPAX0@ZP= 6APAX2I0@Z2@Z)”

 

 

This is the case for whatever function I call from my = example program.

Please help us in getting this issue fixed.=

 

We are using VS2005 on Windows XP Professional.

 

Regards,

Kalyan.Polasi 

P Please consider the environment before printing this e-m= ail.

 


--------------------------------------------------=20
Confidentiality And Disclaimer Notice=20
Please do not print this email unless it is absolutely necessary. The
information contained in this electronic message and any attachments to this
message are intended for the exclusive use of the addressee(s) and may cont=
ain
proprietary, confidential or privileged information. If you are not the int=
ended
recipient, you should not disseminate, distribute or copy this e-mail. Plea=
se
notify the sender immediately and destroy all copies of this message and any
attachments. WARNING: Computer viruses can be transmitted via email. The
recipient should check this email and any attachments for the presence of
viruses. The company accepts no liability for any damage caused by any virus
transmitted by this email. ……www.netenrich.com=20
-------------------------------------------------
--=_reb-r4569251E-t4BB369DF-- --===============1016381836== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel --===============1016381836==-- From libssh2-devel-bounces@cool.haxx.se Wed Mar 31 17:32:18 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VFWFfW021507; Wed, 31 Mar 2010 17:32:17 +0200 Received: from giant.haxx.se (giant.haxx.se [83.168.254.42]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VFWE8u021499 for ; Wed, 31 Mar 2010 17:32:14 +0200 Date: Wed, 31 Mar 2010 17:32:14 +0200 (CEST) From: Daniel Stenberg X-X-Sender: dast@giant.haxx.se To: libssh2 development Subject: Re: Issues with using LibSSH2 In-Reply-To: Message-ID: References: User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) X-fromdanielhimself: yes MIME-Version: 1.0 X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On Wed, 31 Mar 2010, Kalyan Polasi wrote: > We tried to compile the LIBBSSH2 source code of the version 2-0.9, which we > are able to do it successfully after adding the external libraries from > OpenSSL and zLib. I assume you mean libssh2 version 0.9 ? That's a very old version we really cannot recommend anyone to use. > Now we have release version libssh2.lb file. I don't know what version that is. I recommend using version 1.2.4 or trying the git version. > After that we tried to start with the example "direct_tcpip.c" which is > available in the site. You're better off using the example from the release archive for your version. The version on the site is the current development version and right now that means the examples don't quite work with (older) releases. > 1. If I try to call the function "libssh2_init(0);", I am getting the below > error This function was recently added and does not exist in libssh2 version 1.2.4 or earlier, but will exist in 1.2.5 and future releases. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 31 20:22:26 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VIMFr0027418; Wed, 31 Mar 2010 20:22:24 +0200 Received: from p02c11o148.mxlogic.net (p02c11o148.mxlogic.net [208.65.144.81]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VIMC8M027358 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Wed, 31 Mar 2010 20:22:13 +0200 Received: from unknown [64.62.199.48] (EHLO mail02.netenrich.com) by p02c11o148.mxlogic.net(mxl_mta-6.5.0-2) over TLS secured channel with ESMTP id ad293bb4.0.5647.00-016.13238.p02c11o148.mxlogic.net (envelope-from ); Wed, 31 Mar 2010 12:22:20 -0600 (MDT) X-MXL-Hash: 4bb392dc2e67d927-20cda34c949e7b42b507ad168376e637b07cf03b Received: from EX22.netenrich.com (172.22.2.30) by EX02.netenrich.com (172.20.1.31) with Microsoft SMTP Server (TLS) id 8.1.393.1; Wed, 31 Mar 2010 11:22:16 -0700 Received: from EX22.netenrich.com ([::1]) by EX22.netenrich.com ([::1]) with mapi; Wed, 31 Mar 2010 11:22:13 -0700 From: Kalyan Polasi To: libssh2 development Date: Wed, 31 Mar 2010 11:22:13 -0700 Subject: RE: Issues with using LibSSH2 Thread-Topic: Issues with using LibSSH2 Thread-Index: AcrQ55wFI6rGYUzFSjyadTJ+sXI4WAAFpIHN Message-ID: References: , In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-Processed-By: Rebuild v2.0-0 X-Spam: [F=0.2000000000; CM=0.500; S=0.200(2010032501)] X-MAIL-FROM: X-SOURCE-IP: [64.62.199.48] X-AnalysisOut: [v=1.0 c=1 a=VphdPIyG4kEA:10 a=kj9zAlcOel0A:10 a=ybBoJlpP49] X-AnalysisOut: [CkjQP4YFKjpg==:17 a=UJ5Y5Z__AAAA:8 a=GXxshfd9Ape_swufC7QA:] X-AnalysisOut: [9 a=wTHsNufv4Y-cSE7XFgadxKV5ZggA:4 a=CjuIK1q_8ugA:10] X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se Hello daniel/team, Could you please suggest me the current working version for port-forwarding? or can you help me with the simple example to connect to other machine using Libssh2? I was also not able to make use of libssh2_session_init(). Can you please tell me why I does I am getting that linker error? Any help on this is highly appreciable. Regards Kalyan Polasi ________________________________________ From: libssh2-devel-bounces@cool.haxx.se [libssh2-devel-bounces@cool.haxx.se] On Behalf Of Daniel Stenberg [daniel@haxx.se] Sent: Wednesday, March 31, 2010 9:02 PM To: libssh2 development Subject: Re: Issues with using LibSSH2 On Wed, 31 Mar 2010, Kalyan Polasi wrote: > We tried to compile the LIBBSSH2 source code of the version 2-0.9, which we > are able to do it successfully after adding the external libraries from > OpenSSL and zLib. I assume you mean libssh2 version 0.9 ? That's a very old version we really cannot recommend anyone to use. > Now we have release version libssh2.lb file. I don't know what version that is. I recommend using version 1.2.4 or trying the git version. > After that we tried to start with the example "direct_tcpip.c" which is > available in the site. You're better off using the example from the release archive for your version. The version on the site is the current development version and right now that means the examples don't quite work with (older) releases. > 1. If I try to call the function "libssh2_init(0);", I am getting the below > error This function was recently added and does not exist in libssh2 version 1.2.4 or earlier, but will exist in 1.2.5 and future releases. -- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel -------------------------------------------------- Confidentiality And Disclaimer Notice Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. ……www.netenrich.com ------------------------------------------------- _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel From libssh2-devel-bounces@cool.haxx.se Wed Mar 31 20:46:16 2010 Return-Path: Received: from giant.haxx.se (localhost.localdomain [127.0.0.1]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VIk9EG008768; Wed, 31 Mar 2010 20:46:14 +0200 Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by giant.haxx.se (8.14.3/8.14.3/Debian-9) with ESMTP id o2VIk8YY008720 for ; Wed, 31 Mar 2010 20:46:08 +0200 Received: by wyb33 with SMTP id 33so192016wyb.41 for ; Wed, 31 Mar 2010 11:46:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:received:message-id:subject :from:to:content-type; bh=TMV77UjStLoIY2vVYfdNi+zgH0ktjJLiIl7TK7rD3XM=; b=NQPgP1QDSOMcr5HKmjf02+wx8rtRrE8rpO9krbOonDb54UEux88pQolUV8OEnZ5R+n PpQL3WU9nvn44nRiSyLrsMED1Tj8EGt202g+JziKC81D18HldEWGV91pCPyioBFERYBG fwKnwsNd6ItWpGbH6NJ2w0z/l6OgP332tAc9c= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=BVwlDrpg4qRHRfq34Bi/0ZtitGNAjIZScc93oAaUegQbKhUSoAxElLIcWPlKzlV76z CETOZn7Vxm2abdP3JtIWnvH+baTeO4Ns47/H/Bgg9RNWGfisq/0sPJYOKLoT6TqiMROo pWm9csyDP+y2fwrQnRc1ASAFIykYOT8F6afMs= MIME-Version: 1.0 Received: by 10.216.49.81 with HTTP; Wed, 31 Mar 2010 11:46:09 -0700 (PDT) In-Reply-To: References: Date: Wed, 31 Mar 2010 19:46:09 +0100 X-Google-Sender-Auth: 64e33f1bea65eacc Received: by 10.216.179.18 with SMTP id g18mr1218688wem.52.1270061170003; Wed, 31 Mar 2010 11:46:10 -0700 (PDT) Message-ID: Subject: Re: Issues with using LibSSH2 From: Alexander Lamaison To: libssh2 development X-BeenThere: libssh2-devel@cool.haxx.se X-Mailman-Version: 2.1.12 Precedence: list Reply-To: libssh2 development List-Id: libssh2 development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: libssh2-devel-bounces@cool.haxx.se Errors-To: libssh2-devel-bounces@cool.haxx.se On 31 March 2010 19:22, Kalyan Polasi wrote: > Hello daniel/team, > > Could you please suggest me the current working version for port-forwarding? > > or can you help me with the simple example to connect to other machine using Libssh2? > > I was also not able to make use of libssh2_session_init(). Can you please tell me why I does I am getting that linker error? Dear Kalyan, Daniel answered all these questions in his previous response. In what way did it leave you unsatisfied? In summary, you are using code from the latest samples with and ancient version of libssh2. Don't. > Confidentiality And Disclaimer Notice > Please do not print this email unless it is absolutely necessary. The > information contained in this electronic message and any attachments to this > message are intended for the exclusive use of the addressee(s) and may contain > proprietary, confidential or privileged information. If you are not the intended > recipient, you should not disseminate, distribute or copy this e-mail. Please > notify the sender immediately and destroy all copies of this message and any > attachments. Let's think about this a second. You've just posted to a public mailing list. This can't possibly be confidential. Please respect the intelligence of everyone here and save this legal mumbo jumbo for appropriate situations (if any exist). Alex _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel