From 1d72e07ffb7b349ac9a8f825eae7cce082b5e2cd Mon Sep 17 00:00:00 2001 From: dartraiden Date: Thu, 20 Jun 2019 22:31:19 +0300 Subject: iibssh2: update to 1.9.0 --- libs/libssh2/docs/NEWS | 8047 +++++++++++++++++++++++++----------------------- 1 file changed, 4225 insertions(+), 3822 deletions(-) (limited to 'libs/libssh2/docs/NEWS') diff --git a/libs/libssh2/docs/NEWS b/libs/libssh2/docs/NEWS index a9c0a3f1ba..4c7f1eacf3 100644 --- a/libs/libssh2/docs/NEWS +++ b/libs/libssh2/docs/NEWS @@ -1,5538 +1,5941 @@ Changelog for the libssh2 project. Generated with git2news.pl -Version 1.8.2 (25 Mar 2019) - -Daniel Stenberg (25 Mar 2019) -- RELEASE-NOTES: version 1.8.2 - +GitHub (19 Jun 2019) - [Will Cosgrove brought this change] - moved MAX size declarations #330 + 1.9 Formatting - [Will Cosgrove brought this change] - Fixed misapplied patch (#327) - - Fixes for user auth - -Version 1.8.1 (14 Mar 2019) + 1.9 Release notes -Will Cosgrove (14 Mar 2019) -- [Michael Buckley brought this change] +Will Cosgrove (17 May 2019) +- [Alexander Curtiss brought this change] - More 1.8.0 security fixes (#316) - - * Defend against possible integer overflows in comp_method_zlib_decomp. + libgcrypt.c : Fixed _libssh2_rsa_sha1_sign memory leak. (#370) - * Defend against writing beyond the end of the payload in _libssh2_transport_read(). + File: libgcrypt.c - * Sanitize padding_length - _libssh2_transport_read(). https://libssh2.org/CVE-2019-3861.html + Notes : Added calls to gcry_sexp_release to free memory allocated by gcry_sexp_find_token - This prevents an underflow resulting in a potential out-of-bounds read if a server sends a too-large padding_length, possibly with malicious intent. + Credit : + Reporter : beckmi + PR by: Alexander Curtiss + +- [Orivej Desh brought this change] + + libssh2_priv.h : Fix musl build warning on sys/poll.h (#346) - * Prevent zero-byte allocation in sftp_packet_read() which could lead to an out-of-bounds read. https://libssh2.org/CVE-2019-3858.html + File : libssh2_priv.h - * Check the length of data passed to sftp_packet_add() to prevent out-of-bounds reads. + Notes : + musl prints `redirecting incorrect #include to ` + http://git.musl-libc.org/cgit/musl/commit/include/sys/poll.h?id=54446d730cfb17c5f7bcf57f139458678f5066cc - * Add a required_size parameter to sftp_packet_require et. al. to require callers of these functions to handle packets that are too short. https://libssh2.org/CVE-2019-3860.html + poll is defined by POSIX to be in poll.h: + http://pubs.opengroup.org/onlinepubs/7908799/xsh/poll.html - * Additional length checks to prevent out-of-bounds reads and writes in _libssh2_packet_add(). https://libssh2.org/CVE-2019-3862.html + Credit : Orivej Desh -GitHub (14 Mar 2019) +GitHub (1 May 2019) - [Will Cosgrove brought this change] - 1.8 Security fixes (#314) + kex.c : additional bounds checks in diffie_hellman_sha1/256 (#361) - * fixed possible integer overflow in packet_length + Files : kex.c, misc.c, misc.h - CVE https://www.libssh2.org/CVE-2019-3861.html + Notes : + Fixed possible out of bounds memory access when reading malformed data in diffie_hellman_sha1() and diffie_hellman_sha256(). - * fixed possible interger overflow with userauth_keyboard_interactive + Added _libssh2_copy_string() to misc.c to return an allocated and filled char buffer from a string_buf offset. Removed no longer needed s var in kmdhgGPshakex_state_t. + +Will Cosgrove (26 Apr 2019) +- [Tseng Jun brought this change] + + sftp.c : sftp_bin2attr() Correct attrs->gid assignment (#366) - CVE https://www.libssh2.org/CVE-2019-3856.html + Regression with fix for #339 - * fixed possible out zero byte/incorrect bounds allocation + Credit : Tseng Jun + +- [Tseng Jun brought this change] + + kex.c : Correct type cast in curve25519_sha256() (#365) + +GitHub (24 Apr 2019) +- [Will Cosgrove brought this change] + + transport.c : scope local total_num var (#364) - CVE https://www.libssh2.org/CVE-2019-3857.html + file : transport.c + notes : move local `total_num` variable inside of if block to prevent scope access issues which caused #360. + +Will Cosgrove (24 Apr 2019) +- [doublex brought this change] + + transport.c : fixes bounds check if partial packet is read - * bounds checks for response packets + Files : transport.c - * fixed integer overflow in userauth_keyboard_interactive + Issue : #360 - CVE https://www.libssh2.org/CVE-2019-3863.html + Notes : + 'p->total_num' instead of local value total_num when doing bounds check. - * 1.8.1 release notes - -Version 1.8.0 (25 Oct 2016) + Credit : Doublex -Daniel Stenberg (25 Oct 2016) -- RELEASE-NOTES: adjusted for 1.8.0 +GitHub (23 Apr 2019) +- [Will Cosgrove brought this change] -Kamil Dudka (20 Oct 2016) -- Revert "aes: the init function fails when OpenSSL has AES support" - - This partially reverts commit f4f2298ef3635acd031cc2ee0e71026cdcda5864 - because it caused the compatibility code to call initialization routines - redundantly, leading to memory leakage with OpenSSL 1.1 and broken curl - test-suite in Fedora: + Editor config file for source files (#322) - 88 bytes in 1 blocks are definitely lost in loss record 5 of 8 - at 0x4C2DB8D: malloc (vg_replace_malloc.c:299) - by 0x72C607D: CRYPTO_zalloc (mem.c:100) - by 0x72A2480: EVP_CIPHER_meth_new (cmeth_lib.c:18) - by 0x4E5A550: make_ctr_evp.isra.0 (openssl.c:407) - by 0x4E5A8E8: _libssh2_init_aes_ctr (openssl.c:471) - by 0x4E5BB5A: libssh2_init (global.c:49) + Simple start to an editor config file when editing source files to make sure they are configured correctly. -Daniel Stenberg (19 Oct 2016) -- [Charles Collicutt brought this change] +- [Will Cosgrove brought this change] - libssh2_wait_socket: Fix comparison with api_timeout to use milliseconds (#134) + misc.c : String buffer API improvements (#332) - Fixes #74 - -- [Charles Collicutt brought this change] - - Set err_msg on _libssh2_wait_socket errors (#135) - -- Revert "travis: Test mbedtls too" + Files : misc.c, hostkey.c, kex.c, misc.h, openssl.c, sftp.c - This reverts commit 3e6de50a24815e72ec5597947f1831f6083b7da8. + Notes : + * updated _libssh2_get_bignum_bytes and _libssh2_get_string. Now pass in length as an argument instead of returning it to keep signedness correct. Now returns -1 for failure, 0 for success. - Travis doesn't seem to support the mbedtls-dev package - -- maketgz: support "only" to only update version number locally + _libssh2_check_length now returns 0 on success and -1 on failure to match the other string_buf functions. Added comment to _libssh2_check_length. - and fix the date output locale + Credit : Will Cosgrove -- configure: make the --with-* options override the OpenSSL default +Will Cosgrove (19 Apr 2019) +- [doublex brought this change] + + mbedtls.c : _libssh2_mbedtls_rsa_new_private_frommemory() allow private-key from memory (#359) - ... previously it would default to OpenSSL even with the --with-[crypto] - options used unless you specificly disabled OpenSSL. Now, enabling another - backend will automatically disable OpenSSL if the other one is found. + File : mbedtls.c + + Notes: _libssh2_mbedtls_rsa_new_private_frommemory() fixes private-key from memory reading to by adding NULL terminator before parsing; adds passphrase support. + + Credit: doublex -- [Keno Fischer brought this change] +- [Ryan Kelley brought this change] - docs: Add documentation on new cmake/configure options + Session.c : banner_receive() from leaking when accessing non ssh ports (#356) + + File : session.c + + Release previous banner in banner_receive() if the session is reused after a failed connection. + + Credit : Ryan Kelley -- [Keno Fischer brought this change] +GitHub (11 Apr 2019) +- [Will Cosgrove brought this change] - configure: Add support for building with mbedtls + Formatting in agent.c + + Removed whitespace. -- [wildart brought this change] +- [Will Cosgrove brought this change] - travis: Test mbedtls too + Fixed formatting in agent.c + + Quiet linter around a couple if blocks and pointer. -- [wildart brought this change] +Will Cosgrove (11 Apr 2019) +- [Zhen-Huan HWANG brought this change] - crypto: add support for the mbedTLS backend + sftp.c : discard and reset oversized packet in sftp_packet_read() (#269) - Closes #132 - -- [wildart brought this change] + file : sftp.c + + notes : when sftp_packet_read() encounters an sftp packet which exceeds SFTP max packet size it now resets the reading state so it can continue reading. + + credit : Zhen-Huan HWANG - cmake: Add CLEAR_MEMORY option, analogously to that for autoconf +GitHub (11 Apr 2019) +- [Will Cosgrove brought this change] -- README.md: fix link typo + Add agent functions libssh2_agent_get_identity_path() and libssh2_agent_set_identity_path() (#308) + + File : agent.c + + Notes : + Libssh2 uses the SSH_AUTH_SOCK env variable to read the system agent location. However, when using a custom agent path you have to set this value using setenv which is not thread-safe. The new functions allow for a way to set a custom agent socket path in a thread safe manor. -- README: markdown version to look nicer on github +- [Will Cosgrove brought this change] -Viktor Szakats (5 Sep 2016) -- [Taylor Holberton brought this change] + Simplified _libssh2_check_length (#350) + + * Simplified _libssh2_check_length + + misc.c : _libssh2_check_length() + + Removed cast and improved bounds checking and format. + + Credit : Yuriy M. Kaminskiy - openssl: add OpenSSL 1.1.0 compatibility +- [Will Cosgrove brought this change] -Daniel Stenberg (4 Sep 2016) -- [Antenore Gatta brought this change] + _libssh2_check_length() : additional bounds check (#348) + + Misc.c : _libssh2_check_length() + + Ensure the requested length is less than the total length before doing the additional bounds check - tests: HAVE_NETINET_IN_H was not defined correctly (#127) +Daniel Stenberg (25 Mar 2019) +- misc: remove 'offset' from string_buf - Fixes #125 + It isn't necessary. + + Closes #343 -- SECURITY: fix web site typo +- sftp: repair mtime from e1ead35e475 + + A regression from e1ead35e4759 broke the SFTP mtime logic in + sftp_bin2attr + + Also simplified the _libssh2_get_u32/u64 functions slightly. + + Closes #342 -- SECURITY: security process +- session_disconnect: don't zero state, just clear the right bit + + If we clear the entire field, the freeing of data in session_free() is + skipped. Instead just clear the bit that risk making the code get stuck + in the transport functions. + + Regression from 4d66f6762ca3fc45d9. + + Reported-by: dimmaq on github + Fixes #338 + Closes #340 -GitHub (14 Aug 2016) -- [Alexander Lamaison brought this change] +- libssh2_sftp.h: restore broken ABI + + Commit 41fbd44 changed variable sizes/types in a public struct which + broke the ABI, which breaks applications! + + This reverts that change. + + Closes #339 - Basic dockerised test suite. +- style: make includes and examples code style strict - This introduces a test suite for libssh2. It runs OpenSSH in a Docker - container because that works well on Windows (via docker-machine) as - well as Linux. Presumably it works on Mac too with docker-machine, but - I've not tested that. + make travis and the makefile rule verify them too - Because the test suite is docker-machine aware, you can also run it - against a cloud provider, for more realistic network testing, by setting - your cloud provider as your active docker machine. The Appveyor CI setup - in this commit does that because Appveyor doesn't support docker - locally. + Closes #334 -Kamil Dudka (3 Aug 2016) -- [Viktor Szakats brought this change] +GitHub (21 Mar 2019) +- [Daniel Stenberg brought this change] - misc.c: Delete unused static variables - - Closes #114 + create a github issue template -Daniel Stenberg (9 Apr 2016) -- [Will Cosgrove brought this change] +Daniel Stenberg (21 Mar 2019) +- stale-bot: activated + + The stale bot will automatically mark stale issues (inactive for 90 + days) and if still untouched after 21 more days, close them. + + See https://probot.github.io/apps/stale/ - Merge pull request #103 from willco007/patch-2 +- libssh2_session_supported_algs.3: fix formatting mistakes - Fix for security issue CVE-2016-0787 + Reported-by: Max Horn + Fixes #57 -Alexander Lamaison (2 Apr 2016) - [Zenju brought this change] - Fix MSVC 14 compilation errors - - For _MSC_VER == 1900 these macros are not needed and create problems: - + libssh2.h: Fix Error C2371 'ssize_t': redefinition + Closes #331 + +- travis: add code style check - 1>C:\Program Files (x86)\Windows Kits\10\Include\10.0.10240.0\ucrt\stdio.h(1925): warning C4005: 'snprintf': macro redefinition (compiling source file libssh2-files\src\mac.c) - - 1> \win32\libssh2_config.h(27): note: see previous definition of 'snprintf' (compiling source file libssh2-files\src\mac.c) + Closes #324 + +- code style: unify code style - 1>C:\Program Files (x86)\Windows Kits\10\Include\10.0.10240.0\ucrt\stdio.h(1927): fatal error C1189: #error: Macro definition of snprintf conflicts with Standard Library function declaration (compiling source file libssh2-files\src\mac.c) + Indent-level: 4 + Max columns: 79 + No spaces after if/for/while + Unified brace positions + Unified white spaces -Daniel Stenberg (26 Mar 2016) -- [Brad Harder brought this change] +- src/checksrc.pl: code style checker + + imported as-is from curl - _libssh2_channel_open: speeling error fixed in channel error message +Will Cosgrove (19 Mar 2019) +- Merge branch 'MichaelBuckley-michaelbuckley-security-fixes' -Alexander Lamaison (15 Mar 2016) -- Link with crypt32.lib on Windows. +- Silence unused var warnings (#329) - Makes linking with static OpenSSL work again. Although it's not - required for dynamic OpenSSL, it does no harm. + Silence warnings about unused variables in this test + +- Removed unneeded > 0 check - Fixes #98. + When checking `userauth_kybd_num_prompts > 100` we don't care if it's also above zero. -- [Craig A. Berry brought this change] +- [Matthew D. Fuller brought this change] - Tweak VMS help file building. + Spell OpenSS_H_ right when talking about their specific private key (#321) - Primarily this is handling cases where top-level files moved into - the docs/ directory. I also corrected a typo and removed the - claim that libssh2 is public domain. + Good catch, thanks. -- [Craig A. Berry brought this change] +GitHub (19 Mar 2019) +- [Will Cosgrove brought this change] - Build with standard stat structure on VMS. + Silence unused var warnings (#329) - This gets us large file support, is available on any VMS release - in the last decade and more, and gives stat other modern features - such as 64-bit ino_t. + Silence warnings about unused variables in this test -- [Craig A. Berry brought this change] +Michael Buckley (19 Mar 2019) +- Fix more scope and printf warning errors - Update vms/libssh2_config.h. - - VMS does have stdlib.h, gettimeofday(), and OpenSSL. The latter - is appropriate to hard-wire in the configuration because it's - installed by default as part of the base operating system and - there is currently no libgcrypt port. +- Silence unused variable warning -- [Craig A. Berry brought this change] +GitHub (19 Mar 2019) +- [Will Cosgrove brought this change] - VMS can't use %zd for off_t format. + Removed unneeded > 0 check - %z is a C99-ism that VMS doesn't currently have; even though the - compiler is C99-compliant, the library isn't quite. The off_t used - for the st_size element of the stat can be 32-bit or 64-bit, so - detect what we've got and pick a format accordingly. + When checking `userauth_kybd_num_prompts > 100` we don't care if it's also above zero. -- [Craig A. Berry brought this change] +Will Cosgrove (19 Mar 2019) +- [Matthew D. Fuller brought this change] - Normalize line endings in libssh2_sftp_get_channel.3. + Spell OpenSS_H_ right when talking about their specific private key (#321) - Somehow it got Windows-style CRLF endings so convert to just LF, - for consistency as well as not to confuse tools that will regard - the \r as content (e.g. the OpenVMS help librarian). + Good catch, thanks. -Dan Fandrich (29 Feb 2016) -- libgcrypt: Fixed a NULL pointer dereference on OOM +Michael Buckley (18 Mar 2019) +- Fix errors identified by the build process -Daniel Stenberg (24 Feb 2016) -- [Viktor Szakats brought this change] +- Fix casting errors after merge - url updates, HTTP => HTTPS - - Closes #87 +GitHub (18 Mar 2019) +- [Michael Buckley brought this change] -Dan Fandrich (23 Feb 2016) -- RELEASE-NOTES: removed some duplicated names + Merge branch 'master' into michaelbuckley-security-fixes -Version 1.7.0 (23 Feb 2016) +Michael Buckley (18 Mar 2019) +- Move fallback SIZE_MAX and UINT_MAX to libssh2_priv.h -Daniel Stenberg (23 Feb 2016) -- web: the site is now HTTPS +- Fix type and logic issues with _libssh2_get_u64 -- RELEASE-NOTES: 1.7.0 release +Daniel Stenberg (17 Mar 2019) +- examples: fix various compiler warnings -- diffie_hellman_sha256: convert bytes to bits - - As otherwise we get far too small numbers. - - Reported-by: Andreas Schneider - - CVE-2016-0787 +- lib: fix various compiler warnings -Alexander Lamaison (18 Feb 2016) -- Allow CI failures with VS 2008 x64. +- session: ignore pedantic warnings for funcpointer <=> void * + +- travis: add a build using configure - Appveyor doesn't support this combination. + Closes #320 -Daniel Stenberg (16 Feb 2016) -- [Viktor Szakats brought this change] +- configure: provide --enable-werror - GNUmakefile: list system libs after user libs +- appveyor: remove old builds that mostly cause failures - Otherwise some referenced WinSock functions will fail to - resolve when linking against LibreSSL 2.3.x static libraries - with mingw. + ... and only run on master branch. - Closes #80 + Closes #323 -- [Viktor Szakats brought this change] +- cmake: add two missing man pages to get installed too + + Both libssh2_session_handshake.3 and + libssh2_userauth_publickey_frommemory.3 were installed by the configure + build already. + + Reported-by: Arfrever on github + Fixes #278 - openssl: apply new HAVE_OPAQUE_STRUCTS macro +- include/libssh2.h: warning: "_WIN64" is not defined, evaluates to 0 - Closes #81 + We don't use #if for defines that might not be defined. -- [Viktor Szakats brought this change] +- pem: //-comments are not allowed - openssl: fix LibreSSL support after OpenSSL 1.1.0-pre1/2 support +Will Cosgrove (14 Mar 2019) +- [Daniel Stenberg brought this change] -Alexander Lamaison (14 Feb 2016) -- sftp.h: Fix non-C90 type. + userauth: fix "Function call argument is an uninitialized value" (#318) - uint64_t does not exist in C90. Use libssh2_uint64_t instead. + Detected by scan-build. -- Exclude sshd tests from AppVeyor. +- fixed unsigned/signed issue + +Daniel Stenberg (15 Mar 2019) +- session_disconnect: clear state - They fail complaining that sshd wasn't invoked with an absolute path. + If authentication is started but not completed before the application + gives up and instead wants to shut down the session, the '->state' field + might still be set and thus effectively dead-lock session_disconnect. + + This happens because both _libssh2_transport_send() and + _libssh2_transport_read() refuse to do anything as long as state is set + without the LIBSSH2_STATE_KEX_ACTIVE bit. + + Reported in curl bug https://github.com/curl/curl/issues/3650 + + Closes #310 -- Test on more versions of Visual Studio. +Will Cosgrove (14 Mar 2019) +- Release notes from 1.8.1 -- Fix Appveyor builds. +Michael Buckley (14 Mar 2019) +- Use string_buf in sftp_init(). -Daniel Stenberg (14 Feb 2016) -- [Viktor Szakats brought this change] +- Guard against out-of-bounds reads in publickey.c - openssl: add OpenSSL 1.1.0-pre3-dev compatibility - - by using API instead of accessing an internal structure. - - Closes #83 +- Guard against out-of-bounds reads in session.c -- RELEASE-NOTES: synced with 996b04ececdf +- Guard against out-of-bounds reads in userauth.c -- include/libssh2.h: next version is 1.7.0 +- Use LIBSSH2_ERROR_BUFFER_TOO_SMALL instead of LIBSSH2_ERROR_OUT_OF_BOUNDARY in sftp.c -- configure: build "silent" if possible +- Additional bounds checking in sftp.c -- sftp: re-indented some minor stuff +- Additional length checks to prevent out-of-bounds reads and writes in _libssh2_packet_add(). https://libssh2.org/CVE-2019-3862.html -- [Jakob Egger brought this change] - - sftp.c: ensure minimum read packet size - - For optimum performance we need to ensure we don't request tiny packets. - -- [Jakob Egger brought this change] +- Add a required_size parameter to sftp_packet_require et. al. to require callers of these functions to handle packets that are too short. https://libssh2.org/CVE-2019-3860.html - sftp.c: Explicit return values & sanity checks +- Check the length of data passed to sftp_packet_add() to prevent out-of-bounds reads. -- [Jakob Egger brought this change] +- Prevent zero-byte allocation in sftp_packet_read() which could lead to an out-of-bounds read. https://libssh2.org/CVE-2019-3858.html - sftp.c: Check Read Packet File Offset - - This commit adds a simple check to see if the offset of the read - request matches the expected file offset. +- Sanitize padding_length - _libssh2_transport_read(). https://libssh2.org/CVE-2019-3861.html - We could try to recover, from this condition at some point in the future. - Right now it is better to return an error instead of corrupted data. + This prevents an underflow resulting in a potential out-of-bounds read if a server sends a too-large padding_length, possibly with malicious intent. -- [Jakob Egger brought this change] +- Defend against writing beyond the end of the payload in _libssh2_transport_read(). - sftp.c: Don't return EAGAIN if data was written to buffer +- Defend against possible integer overflows in comp_method_zlib_decomp. -- [Jakob Egger brought this change] +GitHub (14 Mar 2019) +- [Will Cosgrove brought this change] - sftp.c: Send at least one read request before reading + Security fixes (#315) - This commit ensures that we have sent at least one read request before - we try to read data in sftp_read(). + * Bounds checks - Otherwise sftp_read() would return 0 bytes (indicating EOF) if the - socket is not ready for writing. - -- [Jakob Egger brought this change] - - sftp.c: stop reading when buffer is full + Fixes for CVEs + https://www.libssh2.org/CVE-2019-3863.html + https://www.libssh2.org/CVE-2019-3856.html - Since we can only store data from a single chunk in filep, - we have to stop receiving data as soon as the buffer is full. + * Packet length bounds check - This adresses the following bug report: - https://github.com/libssh2/libssh2/issues/50 - -Salvador Fandiño (21 Jan 2016) -- agent_disconnect_unix: unset the agent fd after closing it + CVE + https://www.libssh2.org/CVE-2019-3855.html - "agent_disconnect_unix", called by "libssh2_agent_disconnect", was - leaving the file descriptor in the agent structure unchanged. Later, - "libssh2_agent_free" would call again "libssh2_agent_disconnect" under - the hood and it would try to close again the same file descriptor. In - most cases that resulted in just a harmless error, but it is also - possible that the file descriptor had been reused between the two - calls resulting in the closing of an unrelated file descriptor. + * Response length check - This patch sets agent->fd to LIBSSH2_INVALID_SOCKET avoiding that - issue. + CVE + https://www.libssh2.org/CVE-2019-3859.html - Signed-off-by: Salvador Fandiño - -Daniel Stenberg (18 Jan 2016) -- [Patrick Monnerat brought this change] - - os400qc3: support encrypted private keys + * Bounds check - PKCS#8 EncryptedPrivateKeyinfo structures are recognized and decoded to get - values accepted by the Qc3 crypto library. - -- [Patrick Monnerat brought this change] - - os400qc3: New PKCS#5 decoder + CVE + https://www.libssh2.org/CVE-2019-3857.html - The Qc3 library is not able to handle PKCS#8 EncryptedPrivateKeyInfo structures - by itself. It is only capable of decrypting the (encrypted) PrivateKeyInfo - part, providing a key encryption key and an encryption algorithm are given. - Since the encryption key and algorithm description part in a PKCS#8 - EncryptedPrivateKeyInfo is a PKCS#5 structure, such a decoder is needed to - get the derived key method and hash, as well as encryption algorith and - initialisation vector. + * Bounds checking + + CVE + https://www.libssh2.org/CVE-2019-3859.html + + and additional data validation + + * Check bounds before reading into buffers + + * Bounds checking + + CVE + https://www.libssh2.org/CVE-2019-3859.html + + * declare SIZE_MAX and UINT_MAX if needed -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400qc3: force continuous update on non-final hash/hmac computation + fixed type warnings (#309) -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400qc3: Be sure hmac keys have a minimum length - - The Qc3 library requires a minimum key length depending on the target - hash algorithm. Append binary zeroes to the given key if not long enough. - This matches RFC 2104 specifications. + Bumping version number for pending 1.8.1 release -- [Patrick Monnerat brought this change] +Will Cosgrove (4 Mar 2019) +- [Daniel Stenberg brought this change] - os400qc3: Slave descriptor for key encryption key + _libssh2_string_buf_free: use correct free (#304) - The Qc3 library requires the key encryption key to exist as long as - the encrypted key is used. Its descriptor token is then kept as an - "encrypted key slave" for recursive release. + Use LIBSSH2_FREE() here, not free(). We allow memory function + replacements so free() is rarely the right choice... -- [Patrick Monnerat brought this change] +GitHub (26 Feb 2019) +- [Will Cosgrove brought this change] - os400qc3.c: comment PEM/DER decoding + Fix for building against libreSSL #302 + + Changed to use the check we use elsewhere. -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400qc3.c: improve ASN.1 header byte checks + Fix for when building against LibreSSL #302 -- [Patrick Monnerat brought this change] +Will Cosgrove (25 Feb 2019) +- [gartens brought this change] - os400qc3.c: improve OID matching + docs: update libssh2_hostkey_hash.3 [ci skip] (#301) -- [Patrick Monnerat brought this change] +GitHub (21 Feb 2019) +- [Will Cosgrove brought this change] - os400: os400qc3.c: replace malloc by LIBSSH2_ALLOC or alloca where possible + fix malloc/free mismatches #296 (#297) -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400: asn1_new_from_bytes(): use data from a single element only + Replaced malloc with calloc #295 -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400: fix an ILE/RPG prototype + Abstracted OpenSSL calls out of hostkey.c (#294) -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400: implement character encoding conversion support + Fix memory dealloc impedance mis-match #292 (#293) + + When using ed25519 host keys and a custom memory allocator. -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - os400: do not miss some external prototypes + Added call to OpenSSL_add_all_digests() #288 - Build procedure extproto() did not strip braces from header files, thus - possibly prepended them to true prototypes. This prevented the prototype to - be recognized as such. - The solution implemented here is to map braces to semicolons, effectively - considering them as potential prototype delimiters. + For OpenSSL 1.0.x we need to call OpenSSL_add_all_digests(). -- [Patrick Monnerat brought this change] +Will Cosgrove (12 Feb 2019) +- [Zhen-Huan HWANG brought this change] - os400: Really add specific README + SFTP: increase maximum packet size to 256K (#268) + + to match implementations like OpenSSH. -- [Patrick Monnerat brought this change] +- [Zenju brought this change] - os400: Add specific README and include new files in dist tarball + Fix https://github.com/libssh2/libssh2/pull/271 (#284) -- [Patrick Monnerat brought this change] +GitHub (16 Jan 2019) +- [Will Cosgrove brought this change] - os400: add compilation scripts + Agent NULL check in shutdown #281 -- [Patrick Monnerat brought this change] +Will Cosgrove (15 Jan 2019) +- [Adrian Moran brought this change] - os400: include files for ILE/RPG + mbedtls: Fix leak of 12 bytes by each key exchange. (#280) - In addition, file os400/macros.h declares all procedures originally - defined as macros. It must not be used for real inclusion and is only - intended to be used as a `database' for macro wrapping procedures generation. + Correctly free ducts by calling _libssh2_mbedtls_bignum_free() in dtor. -- [Patrick Monnerat brought this change] +- [alex-weaver brought this change] - os400: add supplementary header files/wrappers. Define configuration. + Fix error compiling on Win32 with STDCALL=ON (#275) -- [Patrick Monnerat brought this change] +GitHub (8 Nov 2018) +- [Will Cosgrove brought this change] - Protect callback function calls from macro substitution + Allow default permissions to be used in sftp_mkdir (#271) - Some structure fields holding callback addresses have the same name as the - underlying system function (connect, send, recv). Set parentheses around - their reference to suppress a possible macro substitution. - - Use a macro for connect() on OS/400 to resolve a const/nonconst parameter - problem. + Added constant LIBSSH2_SFTP_DEFAULT_MODE to use the server default permissions when making a new directory -- [Patrick Monnerat brought this change] +Will Cosgrove (13 Sep 2018) +- [Giulio Benetti brought this change] - Add interface for OS/400 crypto library QC3 + openssl: fix dereferencing ambiguity potentially causing build failure (#267) + + When dereferencing from *aes_ctr_cipher, being a pointer itself, + ambiguity can occur; fixed possible build errors. -- [Patrick Monnerat brought this change] +Viktor Szakats (12 Sep 2018) +- win32/GNUmakefile: define HAVE_WINDOWS_H + + This macro was only used in test/example code before, now it is + also used in library code, but only defined automatically by + automake/cmake, so let's do the same for the standalone win32 + make file. + + It'd be probably better to just rely on the built-in _WIN32 macro + to detect the presence of windows.h though. It's already used + in most of libssh2 library code. There is a 3rd, similar macro + named LIBSSH2_WIN32, which might also be replaced with _WIN32. + + Ref: https://github.com/libssh2/libssh2/commit/8b870ad771cbd9cd29edbb3dbb0878e950f868ab + Closes https://github.com/libssh2/libssh2/pull/266 - misc: include stdarg.h for debug code +Marc Hoersken (2 Sep 2018) +- Fix conditional check for HAVE_DECL_SECUREZEROMEMORY + + "Unlike the other `AC_CHECK_*S' macros, when a symbol is not declared, + HAVE_DECL_symbol is defined to `0' instead of leaving HAVE_DECL_symbol + undeclared. When you are sure that the check was performed, + use HAVE_DECL_symbol in #if." + + Source: autoconf documentation for AC_CHECK_DECLS. -- [Patrick Monnerat brought this change] +- Fix implicit declaration of function 'SecureZeroMemory' + + Include window.h in order to use SecureZeroMemory on Windows. - Document crypto library interface +- Fix implicit declaration of function 'free' by including stdlib.h -- [Patrick Monnerat brought this change] +GitHub (27 Aug 2018) +- [Will Cosgrove brought this change] - Feature an optional crypto-specific macro to rsa sign a data fragment vector - - OS/400 crypto library is unable to sign a precomputed SHA1 hash: however - it does support a procedure that hashes data fragments and rsa signs. - If defined, the new macro _libssh2_rsa_sha1_signv() implements this function - and disables use of _libssh2_rsa_sha1_sign(). + Use malloc abstraction function in pem parse - The function described above requires that the struct iovec unused slacks are - cleared: for this reason, macro libssh2_prepare_iovec() has been introduced. - It should be defined as empty for crypto backends that are not sensitive - to struct iovec unused slack values. + Fix warning on WinCNG build. -- [Patrick Monnerat brought this change] +- [Will Cosgrove brought this change] - Fold long lines in include files + Fixed possible junk memory read in sftp_stat #258 -- [Viktor Szakats brought this change] +- [Will Cosgrove brought this change] - kex.c: fix indentation + removed INT64_C define (#260) - Closes #71 + No longer used. -- [Viktor Szakats brought this change] +- [Will Cosgrove brought this change] - add OpenSSL-1.1.0-pre2 compatibility - - Closes #70 + Added conditional around engine.h include -- [Viktor Szakats brought this change] +Will Cosgrove (6 Aug 2018) +- [Alex Crichton brought this change] - add OpenSSL 1.1.0-pre1 compatibility + Fix OpenSSL link error with `no-engine` support (#259) - * close https://github.com/libssh2/libssh2/issues/69 - * sync a declaration with the rest of similar ones - * handle EVP_MD_CTX_new() returning NULL with OpenSSL 1.1.0 - * fix potential memory leak with OpenSSL 1.1.0 in - _libssh2_*_init() functions, when EVP_MD_CTX_new() succeeds, - but EVP_DigestInit() fails. + This commit fixes linking against an OpenSSL library that was compiled with + `no-engine` support by bypassing the initialization routines as they won't be + available anyway. -Marc Hoersken (22 Dec 2015) -- wincng.c: fixed _libssh2_wincng_hash_final return value +GitHub (2 Aug 2018) +- [Will Cosgrove brought this change] + + ED25519 Key Support #39 (#248) - _libssh2_wincng_hash_final was returning the internal BCRYPT - status code instead of a valid libssh2 return value (0 or -1). + OpenSSH Key and ED25519 support #39 + Added _libssh2_explicit_zero() to explicitly zero sensitive data in memory #120 - This also means that _libssh2_wincng_hash never returned 0. + * ED25519 Key file support - Requires OpenSSL 1.1.1 or later + * OpenSSH Key format reading support - Supports RSA/DSA/ECDSA/ED25519 types + * New string buffer reading functions - These add build-in bounds checking and convenance methods. Used for OpenSSL PEM file reading. + * Added new tests for OpenSSH formatted Keys -- wincng.c: fixed possible memory leak in _libssh2_wincng_hash +- [Will Cosgrove brought this change] + + ECDSA key types are now explicit (#251) - If _libssh2_wincng_hash_update failed _libssh2_wincng_hash_final - would never have been called before. + * ECDSA key types are now explicit - Reported by Zenju. + Issue was brough up in pull request #248 -Kamil Dudka (15 Dec 2015) -- [Paul Howarth brought this change] +Will Cosgrove (2 May 2018) +- [Jakob Egger brought this change] - libssh2.pc.in: fix the output of pkg-config --libs + Add Instructions for building from Master (#249) + +GitHub (27 Apr 2018) +- [Will Cosgrove brought this change] + + Initialize sb_intl #226 + +Will Cosgrove (19 Apr 2018) +- [doublex brought this change] + + buffer overflow (valgrind) (#159) + +- [Brendan Shanks brought this change] + + mbedTLS: Remove some C99-style intermingled variable declarations (#196) + +GitHub (18 Apr 2018) +- [Will Cosgrove brought this change] + + fix for #160 + +Will Cosgrove (18 Apr 2018) +- [doublex brought this change] + + fix memory leak when using mbedtls backend (#158) - ... such that it does not include LDFLAGS used to build libssh2 itself. - There was a similar fix in the curl project long time ago: + _libssh2_bn_init_from_bin/_libssh2_bn_free would leak bignum from mbedtls_calloc(). + +- [Brendan Shanks brought this change] + + mbedTLS: Avoid multiple definition errors for context handles (#197) + +- [Tseng Jun brought this change] + + Fix the EVP cipher meth memory leakage problem (#244) - https://github.com/bagder/curl/commit/curl-7_19_7-56-g4c8adc8 + * Fix the EVP cipher meth memory leakage problem - Bug: https://bugzilla.redhat.com/1279966 - Signed-off-by: Kamil Dudka + Looks good, thanks for the fixes. -Marc Hoersken (6 Dec 2015) -- hostkey.c: align code path of ssh_rsa_init to ssh_dss_init +Marc Hörsken (31 Mar 2018) +- [Will Cosgrove brought this change] -- hostkey.c: fix invalid memory access if libssh2_dsa_new fails + Added ECDSA defines for WinCNG (#245) - Reported by dimmaq, fixes #66 + Fixed missing defines preventing building using WinCNG -Daniel Stenberg (3 Nov 2015) +GitHub (30 Mar 2018) - [Will Cosgrove brought this change] - gcrypt: define libssh2_sha256_ctx + Fix for _libssh2_rsa_new with OpenSSL 1.0.x - Looks like it didn't make it into the latest commit for whatever reason. - - Closes #58 + missing d value assignment. -- [Salvador Fandino brought this change] +Will Cosgrove (20 Mar 2018) +- [Etienne Samson brought this change] - libssh2_session_set_last_error: Add function + A collection of small fixes (#198) - Net::SSH2, the Perl wrapping module for libssh2 implements several features* - on top of libssh2 that can fail and so need some mechanism to report the error - condition to the user. + * tests: Remove if-pyramids - Until now, besides the error state maintained internally by libssh2, another - error state was maintained at the Perl level for every session object and then - additional logic was used to merge both error states. That is a maintenance - nighmare, and actually there is no way to do it correctly and consistently. + * tests: Switch run_command arguments - In order to allow the high level language to add new features to the library - but still rely in its error reporting features the new function - libssh2_session_set_last_error (that just exposses _libssh2_error_flags) is - introduced. + * tests: Make run_command a vararg function - *) For instance, connecting to a remote SSH service giving the hostname and - port. + * tests: Xcode doesn't obey CMake's test working directory - Signed-off-by: Salvador Fandino - Signed-off-by: Salvador Fandiño + * openssl: move manual AES-CTR cipher into crypto init + + * cmake: Move our include dir before all other include paths -- [Salvador Fandino brought this change] +GitHub (15 Mar 2018) +- [Will Cosgrove brought this change] - _libssh2_error: Support allocating the error message + Fixes incorrect indexing of KEX prefs string - Before this patch "_libssh2_error" required the error message to be a - static string. + After stripping out an invalid KEX pref entry, it would incorrectly advance again leaving invalid values in the list. + +Viktor Szakats (13 Mar 2018) +- tests: fix checksrc warnings - This patch adds a new function "_libssh2_error_flags" accepting an - additional "flags" argument and specifically the flag - "LIBSSH2_ERR_FLAG_DUP" indicating that the passed string must be - duplicated into the heap. + Also: + * add 'static' qualifier to file-wide const buffers + * fix a non-ANSI C89 comment + * silence a mismatched fprintf() mask warning by adding a cast + +- cmake: recognize OpenSSL 1.1 .dll names - Then, the method "_libssh2_error" has been rewritten to use that new - function under the hood. + Also fix some comment typos and a stray tab. + +- docs: update an URL [ci skip] + +Daniel Stenberg (12 Mar 2018) +- docs/SECURITY: the max embargo is 14 days now + +Viktor Szakats (12 Mar 2018) +- docs: spelling fixes [ci skip] - Signed-off-by: Salvador Fandino - Signed-off-by: Salvador Fandiño + Closes https://github.com/libssh2/libssh2/pull/222 +GitHub (12 Mar 2018) - [Will Cosgrove brought this change] - added engine.h include to fix warning + Fixed minor tabs/spacing issues -- [sune brought this change] +- [Will Cosgrove brought this change] - kex.c: removed dupe entry from libssh2_kex_methods[] + Update kex.c + +- [Will Cosgrove brought this change] + + Added basic bounds checking #206 - Closes #51 + Basic bounds checking in ecdh_sha2_nistp() -- [Salvador Fandiño brought this change] +- [Will Cosgrove brought this change] - userauth: Fix off by one error when reading public key file + Fixed Clang warning #206 - After reading the public key from file the size was incorrectly - decremented by one. + Fixed possible garbage value for secret in an error case + +- [Will Cosgrove brought this change] + + Fixed incorrect #if to #ifdef #206 - This was usually a harmless error as the last character on the public - key file is an unimportant EOL. But if due to some error the public key - file is empty, the public key size becomes (uint)(0 - 1), resulting in - an unrecoverable out of memory error later. + When checking HAVE_OPAQUE_STRUCTS. + +Viktor Szakats (12 Mar 2018) +- src: suppress two checksrc warnings - Signed-off-by: Salvador Fandi??o + Ref: https://github.com/libssh2/libssh2/pull/235 -- [Salvador Fandino brought this change] +- src: address fopen() warnings, add missing copyright headers + + Ref: https://github.com/libssh2/libssh2/pull/235 - channel: Detect bad usage of libssh2_channel_process_startup +- src: replace sprintf() with snprintf() - A common novice programmer error (at least among those using the - wrapping Perl module Net::SSH2), is to try to reuse channels. + Ref: https://github.com/libssh2/libssh2/pull/235 + +- src: fix checksrc warnings - This patchs detects that incorrect usage and fails with a - LIBSSH2_ERROR_BAD_USE error instead of hanging. + Use checksrc.pl from the curl project, with (for now) + suppressed long line warnings and indentation set to + 4 spaces. Fixes are whitespace for the most part. - Signed-off-by: Salvador Fandino + Warning count went down from 2704 to 12. + + Also fix codespell typos, two non-ANSI C89 comments + and a stray tab in include/libssh2.h. + + Ref: https://github.com/libssh2/libssh2/pull/235 -- [Will Cosgrove brought this change] +- checksrc: add source style checker + + This is a slightly extended version of this original source + from the curl project: + https://github.com/curl/curl/blob/8b754c430b9a4c51aa606c687ee5014faf7c7b06/lib/checksrc.pl + + This version adds the following options to customize it for + libssh2 (plus some whitespace formatting): + + `-i` to override indentation spaces (2) + `-m` to override maximum line length (79) + + Command-line used to check libssh2 sources: + + $ ./checksrc.pl -i4 -m500 *.c *.h + + Closes https://github.com/libssh2/libssh2/pull/236 - kex: Added diffie-hellman-group-exchange-sha256 support +- src: add static qualifier - ... and fixed HMAC_Init depricated usage + To private, const strings. - Closes #48 + Closes https://github.com/libssh2/libssh2/pull/237 -Alexander Lamaison (21 Sep 2015) -- Prefixed new #defines to prevent collisions. +- [Will Cosgrove brought this change] + + Add support for ECDSA keys and host keys (#41) - Other libraries might have their own USE_WIN32_*FILES. + This commit lands full ECDSA key support when using the OpenSSL + backend. Which includes: + + New KEX methods: + ecdsa-sha2-nistp256, ecdsa-sha2-nistp384, ecdsa-sha2-nistp521 + + Can now read OpenSSL formatted ECDSA key files. + + Now supports known host keys of type ecdsa-sha2-nistp256. + + New curve types: + NID_X9_62_prime256v1, NID_secp384r1, NID_secp521r1 + + Default host key preferred ordering is now nistp256, nistp384, + nistp521, rsa, dss. + + Ref: https://github.com/libssh2/libssh2/issues/41 + + Closes https://github.com/libssh2/libssh2/pull/206 -- [keith-daigle brought this change] +GitHub (15 Dec 2017) +- [Will Cosgrove brought this change] - Update examples/scp.c to fix bug where large files on win32 would cause got to wrap and go negative + Fixed possible crash when decoding invalid data + + When trying to decode invalid data, it frees the buffer but doesn't nil it so the caller gets a junk memory pointer which they could potentially double free. -- [David Byron brought this change] +- [Will Cosgrove brought this change] - add libssh2_scp_recv2 to support large (> 2GB) files on windows + Remove call to OpenSSL_add_all_ciphers() + + Now lives in libssh2 init() from PR #189 -Daniel Stenberg (17 Sep 2015) -- [sune brought this change] +- [Will Cosgrove brought this change] - WinCNG: support for SHA256/512 HMAC + Fixed incorrect reference to decrypted block - Closes #47 + Fixed incorrectly copied memory from p->buf into init instead of from the decrypted buffer block. The only reason this worked was because the crypt() function decrypts the value in-place and overwrites p->buf. I'm working on a fork that no longer does this and exposed this bug. -- [brian m. carlson brought this change] +Will Cosgrove (20 Oct 2017) +- [Pan brought this change] - Add support for HMAC-SHA-256 and HMAC-SHA-512. + Fix typo in crypt.c (#218) + +Kamil Dudka (17 Oct 2017) +- session: avoid printing misleading debug messages - Implement support for these algorithms and wire them up to the libgcrypt - and OpenSSL backends. Increase the maximum MAC buffer size to 64 bytes - to prevent buffer overflows. Prefer HMAC-SHA-256 over HMAC-SHA-512, and - that over HMAC-SHA-1, as OpenSSH does. + ... while throwing LIBSSH2_ERROR_EAGAIN out of session_startup() - Closes #40 + If the session runs in blocking mode, LIBSSH2_ERROR_EAGAIN never reaches + the libssh2 API boundary and, in non-blocking mode, these messages are + suppressed by the condition in _libssh2_error_flags() anyway. + + Closes #211 -- [Zenju brought this change] +Viktor Szakats (15 Oct 2017) +- win32/GNUmakefile: allow customizing dll suffixes + + - New `LIBSSH2_DLL_SUFFIX` envvar will add a suffix to the generated + libssh2 dll name. Useful to add `-x64` to 64-bit builds so that + it can live in the same directory as the 32-bit one. By default + this is empty. + + - New `LIBSSH2_DLL_A_SUFFIX` envvar to customize the suffix of the + generated import library (implib) for libssh2 .dll. It defaults + to `dll`, and it's useful to modify that to `.dll` to have the + standard naming scheme for mingw-built .dlls, i.e. `libssh2.dll.a`. + + Ref: https://github.com/curl/curl/commit/aaa16f80256abc1463fd9374815130a165222257 + + Closes https://github.com/libssh2/libssh2/pull/215 - kex: free server host key before allocating it (again) +- makefile.m32: allow to override gcc, ar and ranlib - Fixes a memory leak when Synology server requests key exchange + Allow to ovverride certain build tools, making it possible to + use LLVM/Clang to build libssh2. The default behavior is unchanged. + To build with clang (as offered by MSYS2), these settings can + be used: - Closes #43 + LIBSSH2_CC=clang + LIBSSH2_AR=llvm-ar + LIBSSH2_RANLIB=llvm-ranlib + + Also adjust ranlib parameters to be compatible with LLVM/Clang's + ranlib tool. + + Closes https://github.com/libssh2/libssh2/pull/214 -- [Viktor Szakats brought this change] +GitHub (27 Sep 2017) +- [Will Cosgrove brought this change] - GNUmakefile: up OpenSSL version + Fixes out of bounds memory access (#210) - closes #23 + If an invalid PEM file is read and the lines are longer than 128 characters it will go out of bounds and crash on line 91. -- [Viktor Szakats brought this change] +Will Cosgrove (11 Sep 2017) +- [Kamil Dudka brought this change] - GNUmakefile: add -m64 CFLAGS when targeting mingw64, add -m32/-m64 to LDFLAGS + scp: do not NUL-terminate the command for remote exec (#208) - libssh2 equivalent of curl patch https://github.com/bagder/curl/commit/d21b66835f2af781a3c2a685abc92ef9f0cd86be + It breaks SCP download/upload from/to certain server implementations. - This allows to build for the non-default target when using a multi-target mingw distro. - Also bump default OpenSSL dependency path to 1.0.2c. + The bug does not manifest with OpenSSH, which silently drops the NUL + byte (eventually with any garbage that follows the NUL byte) before + executing it. + + Bug: https://bugzilla.redhat.com/1489736 +GitHub (21 Aug 2017) - [Viktor Szakats brought this change] - GNUmakefile: add support for LIBSSH2_LDFLAG_EXTRAS + openssl.c: remove no longer used variable (#204) - It is similar to existing LIBSSH2_CFLAG_EXTRAS, but for - extra linker options. + after e378d2e30a40bd9bcee06dc3a4250f269098e200 + +- [Will Cosgrove brought this change] + + Fix for #188 (#189) - Also delete some line/file ending whitespace. + * Update openssl.c - closes #27 + * Create openssl.h -- [nasacj brought this change] +Will Cosgrove (24 May 2017) +- [Marcel Raad brought this change] - hostkey.c: Fix compiling error when OPENSSL_NO_MD5 is defined + openssl: fix build with OpenSSL 1.1 API (#176) - Closes #32 + When building with OPENSSL_API_COMPAT=0x10100000L, OpenSSL_add_all_algorithms + and OpenSSL_add_all_ciphers don't exist. The corresponding functionality is + handled automatically with OpenSSL 1.1. -- [Mizunashi Mana brought this change] +- [Sune Bredahl brought this change] - openssl.h: adjust the rsa/dsa includes - - ... to work when built without DSA support. + Add support for SHA256 hostkey fingerprints (#180) - Closes #36 + Looks good, thanks! -Alexander Lamaison (26 Jul 2015) -- Let CMake build work as a subproject. +GitHub (12 May 2017) +- [Will Cosgrove brought this change] + + Fix memory leak of crypt_ctx->h using openSSL 1.1+ (#177) - Patch contributed by JasonHaslam. + Need to use EVP_CIPHER_CTX_free instead of EVP_CIPHER_CTX_reset. -- Fix builds with Visual Studio 2015. +Marc Hoersken (2 Mar 2017) +- tests/openssh_server/authorized_keys: add key_rsa_encrypted.pub + +- tests: add simple test for passphrase-protected PEM file support + +- os400qc3: enable passphrase-protected PEM file support using pem.c + +- pem: fix indentation and replace assert after 386e012292 + +- [Keno Fischer brought this change] + + pem: add passphrase-protected PEM file support for libgcrypt and wincng - VS2015 moved stdio functions to the header files as inline function. That means check_function_exists can't detect them because it doesn't use header files - just does a link check. Instead we need to use check_symbol_exists with the correct headers. + Since they use our own PEM parser which did not support encrypted + PEM files, trying to use such files on these backends failed. + Fix that by augmenting the PEM parser to support encrypted PEM files. -Kamil Dudka (2 Jul 2015) -- cmake: include CMake files in the release tarballs +- [Thomas brought this change] + + misc: use time constant implementation for AES CTR increment + +- [Thomas brought this change] + + wincng: add AES CTR mode support (aes128-ctr, aes192-ctr, aes256-ctr) + +- [Thomas brought this change] + + openssl: move shared AES-CTR code into misc + +Daniel Stenberg (20 Dec 2016) +- [Alex Crichton brought this change] + + kex: acknowledge error code from libssh2_dh_key_pair() - Despite we announced the CMake support in libssh2-1.6.0 release notes, - the files required by the CMake build system were not included in the - release tarballs. Hence, the only way to use CMake for build was the - upstream git repository. + Fixes a segfault using ssh-agent on Windows - This commit makes CMake actually supported in the release tarballs. - -- tests/mansyntax.sh: fix 'make distcheck' with recent autotools + This commit fixes a segfault seen dereferencing a null pointer on + Windows when using ssh-agent. The problem ended up being that errors + weren't being communicated all the way through, causing null pointers to + be used when functions should have bailed out sooner. - Do not create symbolic links off the build directory. Recent autotools - verify that out-of-source build works even if the source directory tree - is not writable. + The `_libssh2_dh_key_pair` function for WinCNG was modified to propagate + errors, and then the two callsites in kex.c of + `diffie_hellman_sha{1,256}` were updated to recognize this error and + bail out. + + Fixes #162 + Closes #163 -- openssl: fix memleak in _libssh2_dsa_sha1_verify() +Alexander Lamaison (27 Nov 2016) +- [monnerat brought this change] -Daniel Stenberg (12 Jun 2015) -- openssl: make libssh2_sha1 return error code + Implement Diffie-Hellman computations in crypto backends. (#149) - - use the internal prefix _libssh2_ for non-exported functions + Not all backends feature the low level API needed to compute a Diffie-Hellman + secret, but some of them directly implement Diffie-Hellman support with opaque + private data. The later approach is now generalized and backends are + responsible for all Diffie Hellman computations. + As a side effect, procedures/macros _libssh2_bn_rand and _libssh2_bn_mod_exp + are no longer needed outside the backends. + +Peter Stuge (16 Nov 2016) +- acinclude.m4: The mbedtls crypto backend actually requires libmbedcrypto - - removed libssh2_md5() since it wasn't used + Examples can't be linked with libmbedtls but need libmbedcrypto, and + any users of libssh2 which use libtool and libssh2.la would encounter + the same problem. - Reported-by: Kamil Dudka - -- [LarsNordin-LNdata brought this change] + This changes the mbedtls detection to search for libmbedcrypto, which + is the actual dependency for the backend. - SFTP: Increase speed and datasize in SFTP read +- acinclude.m4: Add CPPFLAGS=-I$prefix-dir/include in LIBSSH2_LIB_HAVE_LINKFLAGS - The function sftp_read never return more then 2000 bytes (as it should - when I asked Daniel). I increased the MAX_SFTP_READ_SIZE to 30000 but - didn't get the same speed as a sftp read in SecureSSH. I analyzed the - code and found that a return always was dona when a chunk has been read. - I changed it to a sliding buffer and worked on all available chunks. I - got an increase in speed and non of the test I have done has failed - (both local net and over Internet). Please review and test. I think - 30000 is still not the optimal MAX_SFTP_READ_SIZE, my next goal is to - make an API to enable changing this value (The SecureSSH sftp_read has - more complete filled packages when comparing the network traffic) + This is absolutely neccessary for header files to be found when + AC_LIB_HAVE_LINKFLAGS searches for libraries. -- bump: start working on 1.6.1 +- acinclude.m4: Make saved variables in LIBSSH2_LIB_HAVE_LINKFLAGS uniform -Version 1.6.0 (5 Jun 2015) +- docs/HACKING.CRYPTO: Improve documentation for autoconf build system -Daniel Stenberg (5 Jun 2015) -- RELEASE-NOTES: synced with 858930cae5c6a +Alexander Lamaison (16 Nov 2016) +- [Alex Arslan brought this change] -Marc Hoersken (19 May 2015) -- wincng.c: fixed indentation + Check for netinet/in.h in the tests cmake file (#148) -- [sbredahl brought this change] +- [Patrick Monnerat brought this change] - wincng.c: fixed memleak in (block) cipher destructor + Define new Diffie-Hellman context for mbedTLS -Alexander Lamaison (6 May 2015) -- [Jakob Egger brought this change] +- [monnerat brought this change] - libssh2_channel_open: more detailed error message + Make libssh2 work again on os400. (#118) - The error message returned by libssh2_channel_open in case of a server side channel open failure is now more detailed and includes the four standard error conditions in RFC 4254. + * os400: minimum supported OS version is now V6R1. + Do not log compiler informational messages. + + * Implement crypto backend specific Diffie-Hellman computation. + + This feature is now needed on os400 because the QC3 library does not + implement bn_mod_exp() natively. Up to now, this function was emulated using + an RSA encryption, but commits ca5222ea819cc5ed797860070b4c6c1aeeb28420 and + 7934c9ce2a029c43e3642a492d3b9e494d1542be (CVE-2016-0787) broke the emulation + because QC3 only supports RSA exponents up to 512 bits. + + Happily, QC3 supports a native API for Diffie-Hellman computation, with + opaque random value: this commit implements the use of this API and, as a + side effect, enables support of this feature for any other crypto backend that + would use it. + + A "generic" Diffie-Hellman computation internal API supports crypto backends + not implementing their own: this generic API uses the same functions as before. + + * Fix typos in docs/HACKING.CRYPTO. -- [Hannes Domani brought this change] +- [Peter Stuge brought this change] - kex: fix libgcrypt memory leaks of bignum - - Fixes #168. + acinclude.m4: Fixup OpenSSL EVP_aes_128_ctr() detection -Marc Hoersken (3 Apr 2015) -- configure.ac: check for SecureZeroMemory for clear memory feature +- [Peter Stuge brought this change] -- Revert "wincng.c: fix clear memory feature compilation with mingw" + configure.ac: Add --with-crypto= instead of many different --with-$backend - This reverts commit 2d2744efdd0497b72b3e1ff6e732aa4c0037fc43. + The new --with-crypto option replaces the previous backend-specific + --with-{openssl,libgcrypt,mbedtls,wincng} options and fixes some issues. - Autobuilds show that this did not solve the issue. - And it seems like RtlFillMemory is defined to memset, - which would be optimized out by some compilers. - -- wincng.c: fix clear memory feature compilation with mingw + * libgcrypt or mbedtls would previously be used whenever found, even + if configure was passed --without-libgcrypt or --without-mbedtls. + + * If --with-$backend was specified then configure would not fail even + if that library could not be found, and would instead use whichever + crypto library was found first. + + The new option defaults to `auto`, which makes configure check for all + supported crypto libraries in turn, choosing the first one found, or + exiting with an error if none can be found. -Alexander Lamaison (1 Apr 2015) -- [LarsNordin-LNdata brought this change] +- [Tony Kelman brought this change] - Enable use of OpenSSL that doesn't have DSA. + Build mbedtls from source on Travis (#133) - Added #if LIBSSH2_DSA for all DSA functions. + * Revert "Revert "travis: Test mbedtls too"" + + This reverts commit c4c60eac5ca756333034b07dd9e0b97741493ed3. + + * travis: Build mbedtls from source on Travis + + Use TOOLCHAIN_OPTION when calling cmake on mbedtls + + * tests: only run DSA tests for non-mbedtls + + crypto backends -- [LarsNordin-LNdata brought this change] +- [Peter Stuge brought this change] - Use correct no-blowfish #define with OpenSSL. + configure.ac src/Makefile.am: Remove dead AM_CONDITIONAL(OS400QC3) - The OpenSSL define is OPENSSL_NO_BF, not OPENSSL_NO_BLOWFISH. + According to os400/README400 this backend can not be built + with configure+make, and the conditional is hard coded to false. -Marc Hoersken (25 Mar 2015) -- configure: error if explicitly enabled clear-memory is not supported - - This takes 22bd8d81d8fab956085e2079bf8c29872455ce59 and - b8289b625e291bbb785ed4add31f4759241067f3 into account, - but still makes it enabled by default if it is supported - and error out in case it is unsupported and was requested. +- [Peter Stuge brought this change] -Daniel Stenberg (25 Mar 2015) -- configure: make clear-memory default but only WARN if backend unsupported + configure.ac: Add -DNDEBUG to CPPFLAGS in non-debug builds - ... instead of previous ERROR. + There are a few uses of assert() in channel.c, sftp.c and transport.c. -Marc Hoersken (24 Mar 2015) -- wincng.h: fix warning about computed return value not being used +- [Peter Stuge brought this change] -- nonblocking examples: fix warning about unused tvdiff on Mac OS X + src/global.c: Fix conditional AES-CTR support + + Most of libssh2 already has conditional support for AES-CTR according to + the LIBSSH2_AES_CTR crypto backend #define, but global.c needed fixing. -Daniel Stenberg (24 Mar 2015) -- openssl: fix compiler warnings +- [Peter Stuge brought this change] -- cofigure: fix --disable-clear-memory check + src/crypto.h src/userauth.c: Fix conditional RSA support + + Most of libssh2 already has conditional support for RSA according to + the LIBSSH2_RSA crypto backend #define, but crypto.h and userauth.c + needed a few small fixes. -Marc Hoersken (23 Mar 2015) -- scp.c: improved command length calculation +- [Peter Stuge brought this change] + + src/kex.c: Cast libssh2_sha{1,256}_update data arguments properly - Reduced number of calls to strlen, because shell_quotearg already - returns the length of the resulting string (e.q. quoted path) - which we can add to the existing and known cmd_len. - Removed obsolete call to memset again, because we can put a final - NULL-byte at the end of the string using the calculated length. + The update functions take a const unsigned char * but were called + with (const) char * in some places, causing unneccessary warnings. -- scp.c: improved and streamlined formatting +- [Peter Stuge brought this change] -- scp.c: fix that scp_recv may transmit not initialised memory + docs/HACKING.CRYPTO: Fix two type typos -- scp.c: fix that scp_send may transmit not initialised memory - - Fixes ticket 244. Thanks Torsten. +- [Sergei Trofimovich brought this change] -- kex: do not ignore failure of libssh2_sha1_init() + acinclude.m4: fix ./configure --with-libgcrypt - Based upon 43b730ce56f010e9d33573fcb020df49798c1ed8. - Fixes ticket 290. Thanks for the suggestion, mstrsn. - -- wincng.h: fix return code of libssh2_md5_init() - -- openssl.c: fix possible segfault in case EVP_DigestInit fails - -- wincng.c: fix possible use of uninitialized variables - -- wincng.c: fix unused argument warning if clear memory is not enabled - -- wincng: Added explicit clear memory feature to WinCNG backend + The change fixes passing of bogus gcrypt prefix. + Reproducible as: - This re-introduces the original feature proposed during - the development of the WinCNG crypto backend. It still needs - to be added to libssh2 itself and probably other backends. + $ ./configure --with-libgcrypt + $ make V=1 + ... + /bin/sh ../libtool --tag=CC --mode=link gcc -g -O2 -Iyes/include -version-info 1:1:0 -no-undefined -export-symbols-regex '^libssh2_.*' -lgcrypt -lz -Lyes/lib -o libssh2.la -rpath /usr/local/lib 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 libgcrypt.lo pem.lo keepalive.lo global.lo -lgcrypt + ../libtool: line 7475: cd: yes/lib: No such file or directory + libtool: error: cannot determine absolute directory name of 'yes/lib' - Memory is cleared using the function SecureZeroMemory which is - available on Windows systems, just like the WinCNG backend. - -- wincng.c: fixed mixed line-endings - -- wincng.c: fixed use of invalid parameter types in a8d14c5dcf - -- wincng.c: only try to load keys corresponding to the algorithm - -- wincng.c: moved PEM headers into definitions - -- wincng.h: fixed invalid parameter name - -- wincng: fixed mismatch with declarations in crypto.h - -- userauth.c: fixed warning C6001: using uninitialized sig and sig_len - -- pem.c: fixed warning C6269: possible incorrect order of operations - -- wincng: add support for authentication keys to be passed in memory + These + -Iyes/include + -Lyes/lib + come from libgcrypt code autodetection: + if test -n "$use_libgcrypt" && test "$use_libgcrypt" != "no"; then + LDFLAGS="$LDFLAGS -L$use_libgcrypt/lib" + CFLAGS="$CFLAGS -I$use_libgcrypt/include" - Based upon 18cfec8336e and daa2dfa2db. - -- pem.c: add _libssh2_pem_parse_memory to parse PEM from memory + I assume it's a typo to use yes/no flag as a prefix and changed + it to '$with_libgcrypt_prefix'. - Requirement to implement 18cfec8336e for Libgcrypt and WinCNG. - -- pem.c: fix copy and paste mistake from 55d030089b8 - -- userauth.c: fix another possible dereference of a null pointer - -- userauth.c: fix possible dereference of a null pointer + Reported-by: Mikhail Pukhlikov + Signed-off-by: Sergei Trofimovich -- pem.c: reduce number of calls to strlen in readline - -Alexander Lamaison (17 Mar 2015) -- [Will Cosgrove brought this change] +- [Zenju brought this change] - Initialise HMAC_CTX in more places. + libssh2_sftp_init hang: last error not set - Missed a couple more places we init ctx to avoid openssl threading crash. - -- Build build breakage in WinCNG backend caused when adding libssh2_userauth_publickey_frommemory. + The problem is that the original if statement simply returns NULL, but does not set the session last error code. The consequence is that libssh2_sftp_init() also returns NULL and libssh2_session_last_errno(sshSession) == LIBSSH2_ERROR_NONE. - The new feature isn't implemented for the WinCNG backend currently, but the WinCNG backend didn't contain any implementation of the required backend functions - even ones that returns an error. That caused link errors. + In my test the LIBSSH2_ERROR_EAGAIN is coming from sftp.c row 337: + if(4 != sftp->partial_size_len) + /* we got a short read for the length part */ + return LIBSSH2_ERROR_EAGAIN; - This change fixes the problem by providing an implementation of the backend functions that returns an error. + with "partial_size_len == 0". Not sure if this is expected. -- Fix breakage in WinCNG backend caused by introducing libssh2_hmac_ctx_init. - - The macro was defined to nothing for the libgcrypt backend, but not for WinCNG. This brings the latter into line with the former. +- [Aidan Hobson Sayers brought this change] -Daniel Stenberg (15 Mar 2015) -- userauth_publickey_frommemory.3: add AVAILABILITY + docs: correctly describe channel_wait_eof - ... it will be added in 1.6.0 + channel_wait_eof waits for channel->remote.eof, which is set on + receiving a `SSH_MSG_CHANNEL_EOF` message. This message is sent + when a party has no more data to send on a channel. -- libssh2: next version will be called 1.6.0 - - ... since we just added a new function. +- [Zenju brought this change] -- docs: add libssh2_userauth_publickey_frommemory.3 to dist + Fix MSVC 14 compilation warning (#92) - The function and man page were added in commit 18cfec8336e + 1> sftp.c + 1>libssh2-files\src\sftp.c(3393): warning C4456: declaration of 'retcode' hides previous local declaration + 1> libssh2-files\src\sftp.c(3315): note: see declaration of 'retcode' -- [Jakob Egger brought this change] +- [Salvador Fandino brought this change] - direct_tcpip: Fixed channel write + LIBSSH2_ERROR_CHANNEL_WINDOW_FULL: add new error code - There were 3 bugs in this loop: - 1) Started from beginning after partial writes - 2) Aborted when 0 bytes were sent - 3) Ignored LIBSSH2_ERROR_EAGAIN + In order to signal that the requested operation can not succeed + because the receiving window had been exhausted, the error code + LIBSSH2_ERROR_BUFFER_TOO_SMALL has been reused but I have found + that in certain context it may be ambigous. - See also: - https://trac.libssh2.org/ticket/281 - https://trac.libssh2.org/ticket/293 + This patch introduces a new error code, + LIBSSH2_ERROR_CHANNEL_WINDOW_FULL, exclusive to signal that condition. -Alexander Lamaison (15 Mar 2015) -- [Will Cosgrove brought this change] +- [Salvador Fandino brought this change] - Must init HMAC_CTX before using it. + channel_wait_eof: handle receive window exhaustion - Must init ctx before using it or openssl will reuse the hmac which is not thread safe and causes a crash. - Added libssh2_hmac_ctx_init macro. - -- Add continuous integration configurations. + Until now, in blocking mode, if the remote receiving window is + exhausted this function hangs forever as data is not read and the + remote side just keeps waiting for the window to grow before sending + more data. - Linux-based CI is done by Travis CI. Windows-based CI is done by Appveyor. + This patch, makes this function check for that condition and abort + with an error when it happens. -- [David Calavera brought this change] +- [Salvador Fandino brought this change] - Allow authentication keys to be passed in memory. + channel_wait_closed: don't fail when unread data is queued - All credits go to Joe Turpin, I'm just reaplying and cleaning his patch: - http://www.libssh2.org/mail/libssh2-devel-archive-2012-01/0015.shtml + This function was calling channel_wait_eof to ensure that the EOF + packet has already been received, but that function also checks that + the read data queue is empty before reporting the EOF. That caused + channel_wait_closed to fail with a LIBSSH2_ERROR_INVAL when some data + was queued even after a successful call to libssh2_channel_wait_eof. - * Use an unimplemented error for extracting keys from memory with libgcrypt. + This patch changes libssh2_channel_wait_closed to look directly into + channel->remote.eof so that both libssh2_channel_wait_eof and + libssh2_channel_wait_closed bahave consistently. -Daniel Stenberg (14 Mar 2015) -- docs: include the renamed INSTALL* files in dist +- [Salvador Fandino brought this change] -Alexander Lamaison (13 Mar 2015) -- Prevent collisions between CMake and Autotools in examples/ and tests/. + channel_wait_eof: fix debug message -- Avoid clash between CMake build and Autotools. - - Autotools expects a configuration template file at src/libssh2_config.h.in, which buildconf generates. But the CMake build system has its CMake-specific version of the file at this path. This means that, if you don't run buildconf, the Autotools build will fail because it configured the wrong header template. - - See https://github.com/libssh2/libssh2/pull/8. +Daniel Stenberg (25 Oct 2016) +- libssh2.h: start working on 1.8.1 -- Merge pull request #8 from alamaison/cmake - - CMake build system. +Version 1.8.0 (25 Oct 2016) -- CMake build system. +Daniel Stenberg (25 Oct 2016) +- RELEASE-NOTES: adjusted for 1.8.0 + +Kamil Dudka (20 Oct 2016) +- Revert "aes: the init function fails when OpenSSL has AES support" - Tested: - - Windows: - - Visual C++ 2005/2008/2010/2012/2013/MinGW-w64 - - static/shared - - 32/64-bit - - OpenSSL/WinCNG - - Without zlib - - Linux: - - GCC 4.6.3/Clang 3.4 - - static/shared - - 32/64-bit - - OpenSSL/Libgcrypt - - With/Without zlib - - MacOS X - - AppleClang 6.0.0 - - static - - 64-bit - - OpenSSL - - Without zlib + This partially reverts commit f4f2298ef3635acd031cc2ee0e71026cdcda5864 + because it caused the compatibility code to call initialization routines + redundantly, leading to memory leakage with OpenSSL 1.1 and broken curl + test-suite in Fedora: - Conflicts: - README + 88 bytes in 1 blocks are definitely lost in loss record 5 of 8 + at 0x4C2DB8D: malloc (vg_replace_malloc.c:299) + by 0x72C607D: CRYPTO_zalloc (mem.c:100) + by 0x72A2480: EVP_CIPHER_meth_new (cmeth_lib.c:18) + by 0x4E5A550: make_ctr_evp.isra.0 (openssl.c:407) + by 0x4E5A8E8: _libssh2_init_aes_ctr (openssl.c:471) + by 0x4E5BB5A: libssh2_init (global.c:49) -- Man man syntax tests fail gracefully if man version is not suitable. +Daniel Stenberg (19 Oct 2016) +- [Charles Collicutt brought this change] -- Return valid code from test fixture on failure. + libssh2_wait_socket: Fix comparison with api_timeout to use milliseconds (#134) - The sshd test fixture was returning -1 if an error occurred, but negative error codes aren't technically valid (google it). Bash on Windows converted them to 0 which made setup failure look as though all tests were passing. + Fixes #74 -- Let mansyntax.sh work regardless of where it is called from. +- [Charles Collicutt brought this change] -Daniel Stenberg (12 Mar 2015) -- [Viktor Szakáts brought this change] + Set err_msg on _libssh2_wait_socket errors (#135) - mingw build: allow to pass custom CFLAGS +- Revert "travis: Test mbedtls too" - Allow to pass custom `CFLAGS` options via environment variable - `LIBSSH2_CFLAG_EXTRAS`. Default and automatically added options of - `GNUmakefile` have preference over custom ones. This addition is useful - for passing f.e. custom CPU tuning or LTO optimization (`-flto - -ffat-lto-objects`) options. The only current way to do this is to edit - `GNUmakefile`. This patch makes it unnecessary. + This reverts commit 3e6de50a24815e72ec5597947f1831f6083b7da8. - This is a mirror of similar libcurl patch: - https://github.com/bagder/curl/pull/136 + Travis doesn't seem to support the mbedtls-dev package -- [Will Cosgrove brought this change] +- maketgz: support "only" to only update version number locally + + and fix the date output locale - userauth: Fixed prompt text no longer being copied to the prompts struct +- configure: make the --with-* options override the OpenSSL default - Regression from 031566f9c + ... previously it would default to OpenSSL even with the --with-[crypto] + options used unless you specificly disabled OpenSSL. Now, enabling another + backend will automatically disable OpenSSL if the other one is found. -- README: update the git repo locations +- [Keno Fischer brought this change] -- wait_socket: wrong use of difftime() - - With reversed arguments it would always return a negative value... - - Bug: https://github.com/bagder/libssh2/issues/1 + docs: Add documentation on new cmake/configure options -- bump: start working toward 1.5.1 now +- [Keno Fischer brought this change] -Version 1.5.0 (11 Mar 2015) + configure: Add support for building with mbedtls -Daniel Stenberg (11 Mar 2015) -- RELEASE-NOTES: 1.5.0 release +- [wildart brought this change] -- [Mariusz Ziulek brought this change] + travis: Test mbedtls too - kex: bail out on rubbish in the incoming packet - - CVE-2015-1782 - - Bug: http://www.libssh2.org/adv_20150311.html +- [wildart brought this change] -- docs: move INSTALL, AUTHORS, HACKING and TODO to docs/ + crypto: add support for the mbedTLS backend - And with this, cleanup README to be shorter and mention the new source - code home. + Closes #132 -- .gitignore: don't ignore INSTALL +- [wildart brought this change] -Dan Fandrich (4 Mar 2015) -- examples/x11.c: include sys/select.h for improved portability + cmake: Add CLEAR_MEMORY option, analogously to that for autoconf -Daniel Stenberg (4 Mar 2015) -- RELEASE-NOTES: synced with a8473c819bc068 - - In preparation for the upcoming 1.5.0 release. +- README.md: fix link typo -Guenter Knauf (8 Jan 2015) -- NetWare build: added some missing exports. +- README: markdown version to look nicer on github -Marc Hoersken (29 Dec 2014) -- knownhost.c: fix use of uninitialized argument variable wrote - - Detected by clang scan in line 1195, column 18. +Viktor Szakats (5 Sep 2016) +- [Taylor Holberton brought this change] -- examples/x11.c: fix result of operation is garbage or undefined - - Fix use of uninitialized structure w_size_bck. - Detected by clang scan in line 386, column 28. + openssl: add OpenSSL 1.1.0 compatibility -- examples/x11.c: remove dead assigments of some return values - - Detected by clang scan in line 212, column 9. - Detected by clang scan in line 222, column 13. - Detected by clang scan in line 410, column 13. +Daniel Stenberg (4 Sep 2016) +- [Antenore Gatta brought this change] -- examples/x11.c: fix possible memory leak if read fails + tests: HAVE_NETINET_IN_H was not defined correctly (#127) - Detected by clang scan in line 224, column 21. + Fixes #125 -- examples/x11.c: fix invalid removal of first list element - - Fix use of memory after it was being freed. - Detected by clang scan in line 56, column 12. +- SECURITY: fix web site typo -- userauth.c: make sure that sp_len is positive and avoid overflows - - ... if the pointer subtraction of sp1 - pubkey - 1 resulted in a - negative or larger value than pubkey_len, memchr would fail. - - Reported by Coverity CID 89846. +- SECURITY: security process -- channel.c: remove logically dead code, host cannot be NULL here +GitHub (14 Aug 2016) +- [Alexander Lamaison brought this change] + + Basic dockerised test suite. - ... host cannot be NULL in line 525, because it is always - valid (e.g. at least set to "0.0.0.0") after lines 430 and 431. + This introduces a test suite for libssh2. It runs OpenSSH in a Docker + container because that works well on Windows (via docker-machine) as + well as Linux. Presumably it works on Mac too with docker-machine, but + I've not tested that. - Reported by Coverity CID 89807. + Because the test suite is docker-machine aware, you can also run it + against a cloud provider, for more realistic network testing, by setting + your cloud provider as your active docker machine. The Appveyor CI setup + in this commit does that because Appveyor doesn't support docker + locally. -- session.c: check return value of session_nonblock during startup - - Reported by Coverity CID 89803. +Kamil Dudka (3 Aug 2016) +- [Viktor Szakats brought this change] -- session.c: check return value of session_nonblock in debug mode + misc.c: Delete unused static variables - Reported by Coverity CID 89805. + Closes #114 -- pem.c: fix mixed line-endings introduced with 8670f5da24 +Daniel Stenberg (9 Apr 2016) +- [Will Cosgrove brought this change] -- pem.c: make sure there's a trailing zero and b64data is not NULL + Merge pull request #103 from willco007/patch-2 - ... if there is no base64 data between PEM header and footer. - Reported by Coverity CID 89823. + Fix for security issue CVE-2016-0787 -- kex.c: make sure mlist is not set to NULL - - ... if the currently unsupported LANG methods are called. - Reported by Coverity CID 89834. +Alexander Lamaison (2 Apr 2016) +- [Zenju brought this change] -- packet.c: i < 256 was always true and i would overflow to 0 + Fix MSVC 14 compilation errors - Visualize that the 0-termination is intentional, because the array - is later passed to strlen within _libssh2_packet_askv. + For _MSC_VER == 1900 these macros are not needed and create problems: + + + + 1>C:\Program Files (x86)\Windows Kits\10\Include\10.0.10240.0\ucrt\stdio.h(1925): warning C4005: 'snprintf': macro redefinition (compiling source file libssh2-files\src\mac.c) + + 1> \win32\libssh2_config.h(27): note: see previous definition of 'snprintf' (compiling source file libssh2-files\src\mac.c) + + 1>C:\Program Files (x86)\Windows Kits\10\Include\10.0.10240.0\ucrt\stdio.h(1927): fatal error C1189: #error: Macro definition of snprintf conflicts with Standard Library function declaration (compiling source file libssh2-files\src\mac.c) -- silence multiple data conversion warnings +Daniel Stenberg (26 Mar 2016) +- [Brad Harder brought this change] -Daniel Stenberg (23 Dec 2014) -- agent_connect_unix: make sure there's a trailing zero - - ... if the path name was too long. Reported by Coverity CID 89801. + _libssh2_channel_open: speeling error fixed in channel error message -Marc Hoersken (22 Dec 2014) -- examples on Windows: use native SOCKET-type instead of int +Alexander Lamaison (15 Mar 2016) +- Link with crypt32.lib on Windows. - And check return values accordingly. + Makes linking with static OpenSSL work again. Although it's not + required for dynamic OpenSSL, it does no harm. + + Fixes #98. -- userauth.c: improve readability and clarity of for-loops +- [Craig A. Berry brought this change] -Daniel Stenberg (22 Dec 2014) -- calloc: introduce LIBSSH2_CALLOC() + Tweak VMS help file building. - A simple function using LIBSSH2_ALLOC + memset, since this pattern was - used in multiple places and this simplies code in general. + Primarily this is handling cases where top-level files moved into + the docs/ directory. I also corrected a typo and removed the + claim that libssh2 is public domain. -Marc Hoersken (15 Dec 2014) -- libssh2_priv.h: Ignore session, context and format parameters +- [Craig A. Berry brought this change] -- x11 example: check return value of socket function + Build with standard stat structure on VMS. + + This gets us large file support, is available on any VMS release + in the last decade and more, and gives stat other modern features + such as 64-bit ino_t. -- examples: fixed mixed line-endings introduced with aedfba25b8 +- [Craig A. Berry brought this change] -- wincng.c: explicitly ignore BCrypt*AlgorithmProvider return codes + Update vms/libssh2_config.h. - Fixes VS2012 code analysis warning C6031: - return value ignored: could return unexpected value + VMS does have stdlib.h, gettimeofday(), and OpenSSL. The latter + is appropriate to hard-wire in the configuration because it's + installed by default as part of the base operating system and + there is currently no libgcrypt port. -- wincng.c: fix possible invalid memory write access - - Fixes VS2012 code analysis warning C6386: - buffer overrun: accessing 'pbOutput', the writable size is - 'cbOutput' bytes, but '3' bytes may be written: libssh2 wincng.c 610 +- [Craig A. Berry brought this change] -- tests on Windows: check for WSAStartup return code + VMS can't use %zd for off_t format. - Fixes VS2012 code analysis warning C6031: - return value ignored: could return unexpected value + %z is a C99-ism that VMS doesn't currently have; even though the + compiler is C99-compliant, the library isn't quite. The off_t used + for the st_size element of the stat can be 32-bit or 64-bit, so + detect what we've got and pick a format accordingly. -- wincng.c: fix possible NULL pointer de-reference of bignum - - Fixes VS2012 code analysis warning C6011: - dereferencing NULL pointer 'bignum'. libssh2 wincng.c 1567 +- [Craig A. Berry brought this change] -- wincng.c: fix possible use of uninitialized memory + Normalize line endings in libssh2_sftp_get_channel.3. - Fixes VS2012 code analysis warning C6001: - using uninitialized memory 'cbDecoded'. libssh2 wincng.c 553 + Somehow it got Windows-style CRLF endings so convert to just LF, + for consistency as well as not to confuse tools that will regard + the \r as content (e.g. the OpenVMS help librarian). -- packet.c: fix possible NULL pointer de-reference within listen_state - - Fixes VS2012 code analysis warning C6011: - dereferencing NULL pointer 'listen_state->channel'. libssh2 packet.c 221 +Dan Fandrich (29 Feb 2016) +- libgcrypt: Fixed a NULL pointer dereference on OOM -- kex.c: fix possible NULL pointer de-reference with session->kex +Daniel Stenberg (24 Feb 2016) +- [Viktor Szakats brought this change] + + url updates, HTTP => HTTPS - Fixes VS2012 code analysis warning C6011: - dereferencing NULL pointer 'session->kex'. libssh2 kex.c 1761 + Closes #87 -- agent.c: check return code of MapViewOfFile +Dan Fandrich (23 Feb 2016) +- RELEASE-NOTES: removed some duplicated names + +Version 1.7.0 (23 Feb 2016) + +Daniel Stenberg (23 Feb 2016) +- web: the site is now HTTPS + +- RELEASE-NOTES: 1.7.0 release + +- diffie_hellman_sha256: convert bytes to bits - Fixes VS2012 code analysis warning C6387: 'p+4' may be '0': - this does not adhere to the specification for the function - 'memcpy': libssh2 agent.c 330 + As otherwise we get far too small numbers. - Fixes VS2012 code analysis warning C6387: 'p' may be '0': - this does not adhere to the specification for the function - 'UnmapViewOfFile': libssh2 agent.c 333 + Reported-by: Andreas Schneider + + CVE-2016-0787 -- examples on Windows: check for socket return code +Alexander Lamaison (18 Feb 2016) +- Allow CI failures with VS 2008 x64. - Fixes VS2012 code analysis warning C28193: - The variable holds a value that must be examined + Appveyor doesn't support this combination. -- examples on Windows: check for WSAStartup return code +Daniel Stenberg (16 Feb 2016) +- [Viktor Szakats brought this change] + + GNUmakefile: list system libs after user libs - Fixes VS2012 code analysis warning C6031: - return value ignored: could return unexpected value + Otherwise some referenced WinSock functions will fail to + resolve when linking against LibreSSL 2.3.x static libraries + with mingw. + + Closes #80 -Guenter Knauf (11 Dec 2014) -- wincng.c: silent some more gcc compiler warnings. +- [Viktor Szakats brought this change] -- wincng.c: silent gcc compiler warnings. + openssl: apply new HAVE_OPAQUE_STRUCTS macro + + Closes #81 -- Watcom build: added support for WinCNG build. +- [Viktor Szakats brought this change] -- build: updated dependencies in makefiles. + openssl: fix LibreSSL support after OpenSSL 1.1.0-pre1/2 support -Daniel Stenberg (4 Dec 2014) -- configure: change LIBS not LDFLAGS when checking for libs - - Closes #289 +Alexander Lamaison (14 Feb 2016) +- sftp.h: Fix non-C90 type. - Patch-by: maurerpe + uint64_t does not exist in C90. Use libssh2_uint64_t instead. -Guenter Knauf (3 Dec 2014) -- MinGW build: some more GNUMakefile tweaks. +- Exclude sshd tests from AppVeyor. - test/GNUmakefile: added architecture autodetection; added switches to - CFLAGS and RCFLAGS to make sure that the right architecture is used. - Added support to build with WinCNG. + They fail complaining that sshd wasn't invoked with an absolute path. -- sftpdir.c: added authentication method detection. - - Stuff copied over from ssh2.c to make testing a bit easier. +- Test on more versions of Visual Studio. -- NMake build: fixed LIBS settings. +- Fix Appveyor builds. -- NMake build: added support for WinCNG build. +Daniel Stenberg (14 Feb 2016) +- [Viktor Szakats brought this change] -- MinGW build: some GNUMakefile tweaks. + openssl: add OpenSSL 1.1.0-pre3-dev compatibility - Added architecture autodetection; added switches to CFLAGS and - RCFLAGS to make sure that the right architecture is used. - Added support to build with WinCNG. + by using API instead of accessing an internal structure. + + Closes #83 -- MinGW build: Fixed redefine warnings. +- RELEASE-NOTES: synced with 996b04ececdf -- Updated copyright year. +- include/libssh2.h: next version is 1.7.0 -Daniel Stenberg (31 Aug 2014) -- COPYING: bump the copyright year +- configure: build "silent" if possible -Dan Fandrich (28 Jul 2014) -- docs: fixed a bunch of typos +- sftp: re-indented some minor stuff -- docs: added missing libssh2_session_handshake.3 file +- [Jakob Egger brought this change] -Marc Hoersken (19 May 2014) -- wincng.c: specify the required libraries for dependencies using MSVC + sftp.c: ensure minimum read packet size - Initially reported by Bob Kast as "for MS VS builds, specify the - libraries that are required so they don't need to go into all - project files that may use this library". Thanks a lot. + For optimum performance we need to ensure we don't request tiny packets. -- [Bob Kast brought this change] +- [Jakob Egger brought this change] - windows build: do not export externals from static library + sftp.c: Explicit return values & sanity checks + +- [Jakob Egger brought this change] + + sftp.c: Check Read Packet File Offset - If you are building a DLL, then you need to explicitly export each - entry point. When building a static library, you should not. + This commit adds a simple check to see if the offset of the read + request matches the expected file offset. - libssh2 was exporting the entry points whether it was building a DLL or a - static library. To elaborate further, if libssh2 was used as a static - library, which was being linked into a DLL, the libssh2 API would be - exported from that separate DLL. + We could try to recover, from this condition at some point in the future. + Right now it is better to return an error instead of corrupted data. -Daniel Stenberg (19 May 2014) -- [Mikhail Gusarov brought this change] +- [Jakob Egger brought this change] - Fix typos in manpages + sftp.c: Don't return EAGAIN if data was written to buffer -Marc Hoersken (18 May 2014) -- wincng.c: Fixed memory leak in case of an error during ASN.1 decoding +- [Jakob Egger brought this change] -- configure: Display individual crypto backends on separate lines + sftp.c: Send at least one read request before reading - This avoids line-wrapping in between parameters and makes the - error message look like the following: + This commit ensures that we have sent at least one read request before + we try to read data in sftp_read(). - configure: error: No crypto library found! - Try --with-libssl-prefix=PATH - or --with-libgcrypt-prefix=PATH - or --with-wincng on Windows + Otherwise sftp_read() would return 0 bytes (indicating EOF) if the + socket is not ready for writing. -- [Bob Kast brought this change] +- [Jakob Egger brought this change] - libssh2_priv.h: a 1 bit bit-field should be unsigned + sftp.c: stop reading when buffer is full - some compilers may not like this + Since we can only store data from a single chunk in filep, + we have to stop receiving data as soon as the buffer is full. + + This adresses the following bug report: + https://github.com/libssh2/libssh2/issues/50 -- knownhost.c: Fixed warning that pointer targets differ in signedness +Salvador Fandiño (21 Jan 2016) +- agent_disconnect_unix: unset the agent fd after closing it + + "agent_disconnect_unix", called by "libssh2_agent_disconnect", was + leaving the file descriptor in the agent structure unchanged. Later, + "libssh2_agent_free" would call again "libssh2_agent_disconnect" under + the hood and it would try to close again the same file descriptor. In + most cases that resulted in just a harmless error, but it is also + possible that the file descriptor had been reused between the two + calls resulting in the closing of an unrelated file descriptor. + + This patch sets agent->fd to LIBSSH2_INVALID_SOCKET avoiding that + issue. + + Signed-off-by: Salvador Fandiño -- wincng.c: Fixed warning about pointer targets differing in signedness +Daniel Stenberg (18 Jan 2016) +- [Patrick Monnerat brought this change] -- tcpip-forward.c: Fixed warning that pointer targets differ in signedness + os400qc3: support encrypted private keys - libssh2_channel_forward_listen_ex uses ints instead of unsigned ints. + PKCS#8 EncryptedPrivateKeyinfo structures are recognized and decoded to get + values accepted by the Qc3 crypto library. -- misc.c: Fixed warning about mixed declarations and code +- [Patrick Monnerat brought this change] -- libgcrypt.h: Fixed warning about pointer targets differing in signedness + os400qc3: New PKCS#5 decoder + + The Qc3 library is not able to handle PKCS#8 EncryptedPrivateKeyInfo structures + by itself. It is only capable of decrypting the (encrypted) PrivateKeyInfo + part, providing a key encryption key and an encryption algorithm are given. + Since the encryption key and algorithm description part in a PKCS#8 + EncryptedPrivateKeyInfo is a PKCS#5 structure, such a decoder is needed to + get the derived key method and hash, as well as encryption algorith and + initialisation vector. -- wincng.h: Fixed warning about pointer targets differing in signedness +- [Patrick Monnerat brought this change] -- misc.c: Fixed warning about unused parameter abstract + os400qc3: force continuous update on non-final hash/hmac computation -- tcpip-forward.c: Removed unused variables shost, sport and sockopt +- [Patrick Monnerat brought this change] -- wincng.h: Added forward declarations for all WinCNG functions + os400qc3: Be sure hmac keys have a minimum length - Initially reported by Bob Kast as "Wincng - define function - prototypes for wincng routines". Thanks a lot. + The Qc3 library requires a minimum key length depending on the target + hash algorithm. Append binary zeroes to the given key if not long enough. + This matches RFC 2104 specifications. + +- [Patrick Monnerat brought this change] + + os400qc3: Slave descriptor for key encryption key - Also replaced structure definitions with type definitions. + The Qc3 library requires the key encryption key to exist as long as + the encrypted key is used. Its descriptor token is then kept as an + "encrypted key slave" for recursive release. -- [Bob Kast brought this change] +- [Patrick Monnerat brought this change] - libssh2.h: on Windows, a socket is of type SOCKET, not int + os400qc3.c: comment PEM/DER decoding -- win32: Added WinCNG targets to generated Visual Studio project - - Inspired by Bob Kast's reports, this commit enables the compilation - of libssh2 with WinCNG using the generated Visual Studio project files. - This commit adds WinCNG support to parts of the existing Win32 build - infrastructure, until new build systems, like pre-defined VS project - files or CMake files may be added. - - This commit and b20bfeb3e519119a48509a1099c06d65aa7da1d7 raise one - question: How to handle build systems, like VS project files, that - need to include all source files regardless of the desired target, - including all supported crypto backends? For now the mentioned commit - added a check for LIBSSH2_OPENSSL to openssl.c and with this commit - the supported crypto backends are hardcoded within Makefile.am. +- [Patrick Monnerat brought this change] -- libssh2_priv msvc: Removed redundant definition of inline keyword - - Initially reported by Bob Kast as "Remove redundant 'inline' define". - Thanks a lot. + os400qc3.c: improve ASN.1 header byte checks -- wincng: Made data parameter to hash update function constant - - Initially reported by Bob Kast as "formal parameter must be const - since it is used in contexts where the actual parameter may be const". - Thanks a lot. +- [Patrick Monnerat brought this change] -- wincng: fix cross-compilation against the w64 mingw-runtime package + os400qc3.c: improve OID matching -- openssl: Check for LIBSSH2_OPENSSL in order to compile with openssl +- [Patrick Monnerat brought this change] -- wincng: Fixed use of possible uninitialized variable pPaddingInfo - - Reported by Bob Kast, thanks a lot. + os400: os400qc3.c: replace malloc by LIBSSH2_ALLOC or alloca where possible -- wincng: Added cast for double to unsigned long conversion +- [Patrick Monnerat brought this change] -- wincng: Cleaned up includes and check NTSTATUS using macro - - Removed header file combination that is not supported on a real - Windows platform and can only be compiled using MinGW. Replaced - custom NTSTATUS return code checks with BCRYPT_SUCCESS macro. + os400: asn1_new_from_bytes(): use data from a single element only -Daniel Stenberg (16 Mar 2014) -- userauth_hostbased_fromfile: zero assign to avoid uninitialized use - - Detected by clang-analyze +- [Patrick Monnerat brought this change] -- channel_receive_window_adjust: store windows size always - - Avoid it sometimes returning without storing it, leaving calling - functions with unknown content! - - Detected by clang-analyzer + os400: fix an ILE/RPG prototype -- publickey_packet_receive: avoid junk in returned pointers +- [Patrick Monnerat brought this change] + + os400: implement character encoding conversion support + +- [Patrick Monnerat brought this change] + + os400: do not miss some external prototypes - clang-analyzer found this risk it would return a non-initialized pointer - in a success case + Build procedure extproto() did not strip braces from header files, thus + possibly prepended them to true prototypes. This prevented the prototype to + be recognized as such. + The solution implemented here is to map braces to semicolons, effectively + considering them as potential prototype delimiters. -Peter Stuge (16 Mar 2014) -- [Marc Hoersken brought this change] +- [Patrick Monnerat brought this change] - Added Windows Cryptography API: Next Generation based backend + os400: Really add specific README -- [Marc Hoersken brought this change] +- [Patrick Monnerat brought this change] - knownhost.c: fixed that 'key_type_len' may be used uninitialized - - ../src/knownhost.c: In function 'libssh2_knownhost_readline': - ../src/knownhost.c:651:16: warning: 'key_type_len' may be used - uninitialized in this function [-Wmaybe-uninitialized] - rc = knownhost_add(hosts, hostbuf, NULL, - ^ - ../src/knownhost.c:745:12: note: 'key_type_len' was declared here - size_t key_type_len; - ^ + os400: Add specific README and include new files in dist tarball -- [Marc Hoersken brought this change] +- [Patrick Monnerat brought this change] - pem.c: always compile pem.c independently of crypto backend + os400: add compilation scripts -- Fix non-autotools builds: Always define the LIBSSH2_OPENSSL CPP macro - - Commit d512b25f69a1b6778881f6b4b5ff9cfc6023be42 introduced a crypto - library abstraction in the autotools build system, to allow us to more - easily support new crypto libraries. In that process it was found that - all other build system which we support are hard-coded to build with - OpenSSL. Commit f5c1a0d98bd51aeb24aca3d49c7c81dcf8bd858d fixes automake - introduced into non-autotools build systems but still overlooked the - CPP macro saying that we are using OpenSSL. +- [Patrick Monnerat brought this change] + + os400: include files for ILE/RPG - Thanks to Marc Hörsken for identifying this issue and proposing a fix - for win32/{GNUmakefile,config.mk}. This commit uses a slightly different - approach but the end result is the same. + In addition, file os400/macros.h declares all procedures originally + defined as macros. It must not be used for real inclusion and is only + intended to be used as a `database' for macro wrapping procedures generation. -Dan Fandrich (15 Mar 2014) -- channel_close: Close the channel even in the case of errors +- [Patrick Monnerat brought this change] -- sftp_close_handle: ensure the handle is always closed - - Errors are reported on return, but otherwise the close path is - completed as much as possible and the handle is freed on exit. + os400: add supplementary header files/wrappers. Define configuration. -Alexander Lamaison (6 Mar 2014) -- knownhost: Restore behaviour of `libssh2_knownhost_writeline` with short buffer. - - Commit 85c6627c changed the behaviour of `libssh2_knownhost_writeline` so that it stopped returning the number of bytes needed when the given buffer was too small. Also, the function changed such that is might write to part of the buffer before realising it is too small. +- [Patrick Monnerat brought this change] + + Protect callback function calls from macro substitution - This commit restores the original behaviour, whilst keeping the unknown-key-type functionality that 85c6627c. Instead of writing to the buffer piecemeal, the length of the various parts is calculated up front and the buffer written only if there is enough space. The calculated necessary size is output in `outlen` regardless of whether the buffer was written to. + Some structure fields holding callback addresses have the same name as the + underlying system function (connect, send, recv). Set parentheses around + their reference to suppress a possible macro substitution. - The main use-case for the original behaviour that this commit restores is to allow passing in a NULL buffer to get the actual buffer size needed, before calling the function again with the buffer allocated to the exact size required. + Use a macro for connect() on OS/400 to resolve a const/nonconst parameter + problem. -- knownhost: Fix DSS keys being detected as unknown. - - I missing `else` meant ssh-dss format keys were being re-detected as unknown format. +- [Patrick Monnerat brought this change] -Dan Fandrich (6 Mar 2014) -- knownhosts: Abort if the hosts buffer is too small - - This could otherwise cause a match on the wrong host + Add interface for OS/400 crypto library QC3 -- agent_list_identities: Fixed memory leak on OOM +- [Patrick Monnerat brought this change] -- Fixed a few typos + misc: include stdarg.h for debug code -- userauth: Fixed an attempt to free from stack on error +- [Patrick Monnerat brought this change] -- Fixed a few memory leaks in error paths + Document crypto library interface -- Fixed two potential use-after-frees of the payload buffer +- [Patrick Monnerat brought this change] + + Feature an optional crypto-specific macro to rsa sign a data fragment vector - The first might occur if _libssh2_packet_add returns an error, as - fullpacket_state wasn't reset to idle so if it were possible for - fullpacket to be called again, it would return to the same state - handler and re-use the freed p->packet buffer. + OS/400 crypto library is unable to sign a precomputed SHA1 hash: however + it does support a procedure that hashes data fragments and rsa signs. + If defined, the new macro _libssh2_rsa_sha1_signv() implements this function + and disables use of _libssh2_rsa_sha1_sign(). - The second could occur if decrypt returned an error, as it freed the - packet buffer but did not clear total_num, meaning that freed buffer - could be written into again later. + The function described above requires that the struct iovec unused slacks are + cleared: for this reason, macro libssh2_prepare_iovec() has been introduced. + It should be defined as empty for crypto backends that are not sensitive + to struct iovec unused slack values. -Alexander Lamaison (28 Nov 2013) -- Fix missing `_libssh2_error` in `_libssh2_channel_write`. - - In one case, the error code from `_libssh2_transport_read` was being returned from `_libssh2_channel_write` without setting it as the last error by calling `_libssh2_error`. This commit fixes that. - - Found when using a session whose socket had been inadvertently destroyed. The calling code got confused because via `libssh2_session_last_error` it appeared no error had occurred, despite one being returned from the previous function. +- [Patrick Monnerat brought this change] -Kamil Dudka (21 Nov 2013) -- [Mark McPherson brought this change] + Fold long lines in include files - openssl: initialise the digest context before calling EVP_DigestInit() - - When using the OpenSSL libraries in FIPS mode, the function call - EVP_DigestInit() is actually #defined to FIPS_digestinit(). - Unfortunately wheres EVP_DigestInit() initialises the context and then - calls EVP_DigestInit_ex(), this function assumes that the context has - been pre-initialised and crashes when it isn't. - - Bug: https://trac.libssh2.org/ticket/279 - - Fixes #279 +- [Viktor Szakats brought this change] -- [Marc Hörsken brought this change] + kex.c: fix indentation + + Closes #71 - .gitignore: Ignore files like src/libssh2_config.h.in~ +- [Viktor Szakats brought this change] -Peter Stuge (13 Nov 2013) -- Move automake conditionals added by commit d512b25f out of Makefile.inc - - Commit d512b25f69a1b6778881f6b4b5ff9cfc6023be42 added automake - conditionals to Makefile.inc but since Makefile.inc is included - from Makefile for all other build systems that does not work. - - This commit instead adds Makefile.OpenSSL.inc and Makefile.libgcrypt.inc - and moves the automake conditional to its proper place, src/Makefile.am. - - The automake conditional includes the correct Makefile.$name.inc per - the crypto library selection/detection done by configure. + add OpenSSL-1.1.0-pre2 compatibility - All non-autotools build system files in libssh2 are hardcoded to use - OpenSSL and do not get a conditional but at least there is some reuse - because they can all include the new Makefile.OpenSSL.inc. + Closes #70 -Daniel Stenberg (27 Oct 2013) -- [Salvador Fandino brought this change] +- [Viktor Szakats brought this change] - Set default window size to 2MB - - The default channel window size used until now was 256KB. This value is - too small and results on a bottleneck on real-life networks where - round-trip delays can easily reach 300ms. - - The issue was not visible because the configured channel window size - was being ignored and a hard-coded value of ~22MB being used instead, - but that was fixed on a previous commit. - - This patch just changes the default window size - (LIBSSH2_CHANNEL_WINDOW_DEFAULT) to 2MB. It is the same value used by - OpenSSH and in our opinion represents a good compromise between memory - used and transfer speed. - - Performance tests were run to determine the optimum value. The details - and related discussion are available from the following thread on the - libssh2 mailing-list: - - http://www.libssh2.org/mail/libssh2-devel-archive-2013-10/0018.shtml - http://article.gmane.org/gmane.network.ssh.libssh2.devel/6543 - - An excerpt follows: - - "I have been running some transfer test and measuring their speed. + add OpenSSL 1.1.0-pre1 compatibility - My setup was composed of a quad-core Linux machine running Ubuntu 13.10 - x86_64 with a LXC container inside. The data transfers were performed - from the container to the host (never crossing through a physical - network device). + * close https://github.com/libssh2/libssh2/issues/69 + * sync a declaration with the rest of similar ones + * handle EVP_MD_CTX_new() returning NULL with OpenSSL 1.1.0 + * fix potential memory leak with OpenSSL 1.1.0 in + _libssh2_*_init() functions, when EVP_MD_CTX_new() succeeds, + but EVP_DigestInit() fails. + +Marc Hoersken (22 Dec 2015) +- wincng.c: fixed _libssh2_wincng_hash_final return value - Network delays were simulated using the tc tool. And ping was used to - verify that they worked as intended during the tests. + _libssh2_wincng_hash_final was returning the internal BCRYPT + status code instead of a valid libssh2 return value (0 or -1). - The operation performed was the equivalent to the following ssh command: + This also means that _libssh2_wincng_hash never returned 0. + +- wincng.c: fixed possible memory leak in _libssh2_wincng_hash - $ ssh container "dd bs=16K count=8K if=/dev/zero" >/dev/null + If _libssh2_wincng_hash_update failed _libssh2_wincng_hash_final + would never have been called before. - Though, establishment and closing of the SSH connection was excluded - from the timings. + Reported by Zenju. + +Kamil Dudka (15 Dec 2015) +- [Paul Howarth brought this change] + + libssh2.pc.in: fix the output of pkg-config --libs - I run the tests several times transferring files of sizes up to 128MB - and the results were consistent between runs. + ... such that it does not include LDFLAGS used to build libssh2 itself. + There was a similar fix in the curl project long time ago: - The results corresponding to the 128MB transfer are available here: + https://github.com/bagder/curl/commit/curl-7_19_7-56-g4c8adc8 - https://docs.google.com/spreadsheet/ccc?key=0Ao1yRmX6PQQzdG5wSFlrZl9HRWNET3ZyN0hnaGo5ZFE&usp=sharing + Bug: https://bugzilla.redhat.com/1279966 + Signed-off-by: Kamil Dudka + +Marc Hoersken (6 Dec 2015) +- hostkey.c: align code path of ssh_rsa_init to ssh_dss_init + +- hostkey.c: fix invalid memory access if libssh2_dsa_new fails - It clearly shows that 256KB is too small as the default window size. - Moving to a 512MB generates a great improvement and after the 1MB mark - the returns rapidly diminish. Other factors (TCP window size, probably) - become more limiting than the channel window size + Reported by dimmaq, fixes #66 + +Daniel Stenberg (3 Nov 2015) +- [Will Cosgrove brought this change] + + gcrypt: define libssh2_sha256_ctx - For comparison I also performed the same transfers using OpenSSH. Its - speed is usually on par with that of libssh2 using a window size of 1MB - (even if it uses a 2MB window, maybe it is less aggressive sending the - window adjust msgs)." + Looks like it didn't make it into the latest commit for whatever reason. - Signed-off-by: Salvador Fandino + Closes #58 -- [Salvador brought this change] +- [Salvador Fandino brought this change] - _libssh2_channel_read: Honour window_size_initial + libssh2_session_set_last_error: Add function - _libssh2_channel_read was using an arbitrary hard-coded limit to trigger - the window adjusting code. The adjustment used was also hard-coded and - arbitrary, 15MB actually, which would limit the usability of libssh2 on - systems with little RAM. + Net::SSH2, the Perl wrapping module for libssh2 implements several features* + on top of libssh2 that can fail and so need some mechanism to report the error + condition to the user. - This patch, uses the window_size parameter passed to - libssh2_channel_open_ex (stored as remote.window_size_initial) plus the - buflen as the base for the trigger and the adjustment calculation. + Until now, besides the error state maintained internally by libssh2, another + error state was maintained at the Perl level for every session object and then + additional logic was used to merge both error states. That is a maintenance + nighmare, and actually there is no way to do it correctly and consistently. - The memory usage when using the default window size is reduced from 22MB - to 256KB per channel (actually, if compression is used, these numbers - should be incremented by ~50% to account for the errors between the - decompressed packet sizes and the predicted sizes). + In order to allow the high level language to add new features to the library + but still rely in its error reporting features the new function + libssh2_session_set_last_error (that just exposses _libssh2_error_flags) is + introduced. - My tests indicate that this change does not impact the performance of - transfers across localhost or a LAN, being it on par with that of - OpenSSH. On the other hand, it will probably slow down transfers on - networks with high bandwidth*delay when the default window size - (LIBSSH2_CHANNEL_WINDOW_DEFAULT=256KB) is used. + *) For instance, connecting to a remote SSH service giving the hostname and + port. Signed-off-by: Salvador Fandino + Signed-off-by: Salvador Fandiño - [Salvador Fandino brought this change] - knownhosts: handle unknown key types + _libssh2_error: Support allocating the error message - Store but don't use keys of unsupported types on the known_hosts file. + Before this patch "_libssh2_error" required the error message to be a + static string. - Currently, when libssh2 parses a known_host file containing keys of some - type it doesn't natively support, it stops reading the file and returns - an error. + This patch adds a new function "_libssh2_error_flags" accepting an + additional "flags" argument and specifically the flag + "LIBSSH2_ERR_FLAG_DUP" indicating that the passed string must be + duplicated into the heap. - That means, that the known_host file can not be safely shared with other - software supporting other key types (i.e. OpenSSH). + Then, the method "_libssh2_error" has been rewritten to use that new + function under the hood. - This patch adds support for handling keys of unknown type. It can read - and write them, even if they are never going to be matched. + Signed-off-by: Salvador Fandino + Signed-off-by: Salvador Fandiño + +- [Will Cosgrove brought this change] + + added engine.h include to fix warning + +- [sune brought this change] + + kex.c: removed dupe entry from libssh2_kex_methods[] - At the source level the patch does the following things: + Closes #51 + +- [Salvador Fandiño brought this change] + + userauth: Fix off by one error when reading public key file - - add a new unknown key type LIBSSH2_KNOWNHOST_KEY_UNKNOWN + After reading the public key from file the size was incorrectly + decremented by one. - - add a new slot (key_type_name) on the known_host struct that is - used to store the key type in ascii form when it is not supported + This was usually a harmless error as the last character on the public + key file is an unimportant EOL. But if due to some error the public key + file is empty, the public key size becomes (uint)(0 - 1), resulting in + an unrecoverable out of memory error later. - - parse correctly known_hosts entries with unknown key types and - populate the key_type_name slot + Signed-off-by: Salvador Fandi??o + +- [Salvador Fandino brought this change] + + channel: Detect bad usage of libssh2_channel_process_startup - - print correctly known_hosts entries of unknown type + A common novice programmer error (at least among those using the + wrapping Perl module Net::SSH2), is to try to reuse channels. - - when checking a host key ignore keys that do not match the key + This patchs detects that incorrect usage and fails with a + LIBSSH2_ERROR_BAD_USE error instead of hanging. - Fixes #276 + Signed-off-by: Salvador Fandino -- windows build: fix build errors - - Fixes various link errors with VS2010 - - Reported-by: "kdekker" - Fixes #272 +- [Will Cosgrove brought this change] -- man page: add missing function argument + kex: Added diffie-hellman-group-exchange-sha256 support - for libssh2_userauth_publickey_fromfile_ex() + ... and fixed HMAC_Init depricated usage - Reported-by: "pastey" + Closes #48 + +Alexander Lamaison (21 Sep 2015) +- Prefixed new #defines to prevent collisions. - Fixes #262 + Other libraries might have their own USE_WIN32_*FILES. -- [Salvador brought this change] +- [keith-daigle brought this change] - Fix zlib deflate usage - - Deflate may return Z_OK even when not all data has been compressed - if the output buffer becomes full. - - In practice this is very unlikely to happen because the output buffer - size is always some KBs larger than the size of the data passed for - compression from the upper layers and I think that zlib never expands - the data so much, even on the worst cases. - - Anyway, this patch plays on the safe side checking that the output - buffer is not exhausted. - - Signed-off-by: Salvador + Update examples/scp.c to fix bug where large files on win32 would cause got to wrap and go negative -- [Salvador brought this change] +- [David Byron brought this change] - comp_method_zlib_decomp: Improve buffer growing algorithm - - The old algorithm was O(N^2), causing lots and lots of reallocations - when highly compressed data was transferred. - - This patch implements a simpler one that just doubles the buffer size - everytime it is exhausted. It results in O(N) complexity. - - Also a smaller inflate ratio is used to calculate the initial size (x4). - - Signed-off-by: Salvador + add libssh2_scp_recv2 to support large (> 2GB) files on windows -- [Salvador brought this change] +Daniel Stenberg (17 Sep 2015) +- [sune brought this change] - Fix zlib usage + WinCNG: support for SHA256/512 HMAC - Data may remain in zlib internal buffers when inflate() returns Z_OK - and avail_out == 0. In that case, inflate has to be called again. + Closes #47 + +- [brian m. carlson brought this change] + + Add support for HMAC-SHA-256 and HMAC-SHA-512. - Also, once all the data has been inflated, it returns Z_BUF_ERROR to - signal that the input buffer has been exhausted. + Implement support for these algorithms and wire them up to the libgcrypt + and OpenSSL backends. Increase the maximum MAC buffer size to 64 bytes + to prevent buffer overflows. Prefer HMAC-SHA-256 over HMAC-SHA-512, and + that over HMAC-SHA-1, as OpenSSH does. - Until now, the way to detect that a packet payload had been completely - decompressed was to check that no data remained on the input buffer - but that didn't account for the case where data remained on the internal - zlib buffers. + Closes #40 + +- [Zenju brought this change] + + kex: free server host key before allocating it (again) - That resulted in packets not being completely decompressed and the - missing data reappearing on the next packet, though the bug was masked - by the buffer allocation algorithm most of the time and only manifested - when transferring highly compressible data. + Fixes a memory leak when Synology server requests key exchange - This patch fixes the zlib usage. + Closes #43 + +- [Viktor Szakats brought this change] + + GNUmakefile: up OpenSSL version - Signed-off-by: Salvador + closes #23 -- [Salvador brought this change] +- [Viktor Szakats brought this change] - _libssh2_channel_read: fix data drop when out of window + GNUmakefile: add -m64 CFLAGS when targeting mingw64, add -m32/-m64 to LDFLAGS - After filling the read buffer with data from the read queue, when the - window size was too small, "libssh2_channel_receive_window_adjust" was - called to increase it. In non-blocking mode that function could return - EAGAIN and, in that case, the EAGAIN was propagated upwards and the data - already read on the buffer lost. + libssh2 equivalent of curl patch https://github.com/bagder/curl/commit/d21b66835f2af781a3c2a685abc92ef9f0cd86be - The function was also moving between the two read states - "libssh2_NB_state_idle" and "libssh2_NB_state_created" both of which - behave in the same way (excepting a debug statment). + This allows to build for the non-default target when using a multi-target mingw distro. + Also bump default OpenSSL dependency path to 1.0.2c. + +- [Viktor Szakats brought this change] + + GNUmakefile: add support for LIBSSH2_LDFLAG_EXTRAS - This commit modifies "_libssh2_channel_read" so that the - "libssh2_channel_receive_window_adjust" call is performed first (when - required) and if everything goes well, then it reads the data from the - queued packets into the read buffer. + It is similar to existing LIBSSH2_CFLAG_EXTRAS, but for + extra linker options. - It also removes the useless "libssh2_NB_state_created" read state. + Also delete some line/file ending whitespace. - Some rotted comments have also been updated. + closes #27 + +- [nasacj brought this change] + + hostkey.c: Fix compiling error when OPENSSL_NO_MD5 is defined - Signed-off-by: Salvador + Closes #32 -- [Salvador Fandino brought this change] +- [Mizunashi Mana brought this change] - window_size: redid window handling for flow control reasons + openssl.h: adjust the rsa/dsa includes - Until now, the window size (channel->remote.window_size) was being - updated just after receiving the packet from the transport layer. + ... to work when built without DSA support. - That behaviour is wrong because the channel queue may grow uncontrolled - when data arrives from the network faster that the upper layer consumes - it. + Closes #36 + +Alexander Lamaison (26 Jul 2015) +- Let CMake build work as a subproject. - This patch adds a new counter, read_avail, which keeps a count of the - bytes available from the packet queue for reading. Also, now the window - size is adjusted when the data is actually read by an upper layer. + Patch contributed by JasonHaslam. + +- Fix builds with Visual Studio 2015. - That way, if the upper layer stops reading data, the window will - eventually fill and the remote host will stop sending data. When the - upper layers reads enough data, a window adjust packet is delivered and - the transfer resumes. + VS2015 moved stdio functions to the header files as inline function. That means check_function_exists can't detect them because it doesn't use header files - just does a link check. Instead we need to use check_symbol_exists with the correct headers. + +Kamil Dudka (2 Jul 2015) +- cmake: include CMake files in the release tarballs - The read_avail counter is used to detect the situation when the remote - server tries to send data surpassing the window size. In that case, the - extra data is discarded. + Despite we announced the CMake support in libssh2-1.6.0 release notes, + the files required by the CMake build system were not included in the + release tarballs. Hence, the only way to use CMake for build was the + upstream git repository. - Signed-off-by: Salvador + This commit makes CMake actually supported in the release tarballs. -Peter Stuge (15 Sep 2013) -- configure.ac: Call zlib zlib and not libz in text but keep option names +- tests/mansyntax.sh: fix 'make distcheck' with recent autotools + + Do not create symbolic links off the build directory. Recent autotools + verify that out-of-source build works even if the source directory tree + is not writable. -- configure.ac: Reorder --with-* options in --help output +- openssl: fix memleak in _libssh2_dsa_sha1_verify() -- configure.ac: Rework crypto library detection +Daniel Stenberg (12 Jun 2015) +- openssl: make libssh2_sha1 return error code - This further simplifies adding new crypto libraries. - -- Clean up crypto library abstraction in build system and source code + - use the internal prefix _libssh2_ for non-exported functions - libssh2 used to explicitly check for libgcrypt and default to OpenSSL. + - removed libssh2_md5() since it wasn't used - Now all possible crypto libraries are checked for explicitly, making - the addition of further crypto libraries both simpler and cleaner. + Reported-by: Kamil Dudka -- configure.ac: Add zlib to Requires.private in libssh2.pc if using zlib +- [LarsNordin-LNdata brought this change] -- Revert "Added Windows Cryptography API: Next Generation based backend" + SFTP: Increase speed and datasize in SFTP read - This reverts commit d385230e15715e67796f16f3e65fd899f21a638b. + The function sftp_read never return more then 2000 bytes (as it should + when I asked Daniel). I increased the MAX_SFTP_READ_SIZE to 30000 but + didn't get the same speed as a sftp read in SecureSSH. I analyzed the + code and found that a return always was dona when a chunk has been read. + I changed it to a sliding buffer and worked on all available chunks. I + got an increase in speed and non of the test I have done has failed + (both local net and over Internet). Please review and test. I think + 30000 is still not the optimal MAX_SFTP_READ_SIZE, my next goal is to + make an API to enable changing this value (The SecureSSH sftp_read has + more complete filled packages when comparing the network traffic) -Daniel Stenberg (7 Sep 2013) -- [Leif Salomonsson brought this change] +- bump: start working on 1.6.1 - sftp_statvfs: fix for servers not supporting statfvs extension - - Fixes issue arising when server does not support statfvs and or fstatvfs - extensions. sftp_statvfs() and sftp_fstatvfs() after this patch will - handle the case when SSH_FXP_STATUS is returned from server. +Version 1.6.0 (5 Jun 2015) -- [Marc Hoersken brought this change] +Daniel Stenberg (5 Jun 2015) +- RELEASE-NOTES: synced with 858930cae5c6a - Added Windows Cryptography API: Next Generation based backend +Marc Hoersken (19 May 2015) +- wincng.c: fixed indentation -- [Kamil Dudka brought this change] +- [sbredahl brought this change] - partially revert "window_size: explicit adjustments only" - - This partially reverts commit 03ca9020756a4e16f0294e5b35e9826ee6af2364 - in order to fix extreme slowdown when uploading to localhost via SFTP. - - I was able to repeat the issue on RHEL-7 on localhost only. It did not - occur when uploading via network and it did not occur on a RHEL-6 box - with the same version of libssh2. - - The problem was that sftp_read() used a read-ahead logic to figure out - the window_size, but sftp_packet_read() called indirectly from - sftp_write() did not use any read-ahead logic. + wincng.c: fixed memleak in (block) cipher destructor -- _libssh2_channel_write: client spins on write when window full - - When there's no window to "write to", there's no point in waiting for - the socket to become writable since it most likely just will continue to - be. - - Patch-by: ncm - Fixes #258 +Alexander Lamaison (6 May 2015) +- [Jakob Egger brought this change] -- _libssh2_channel_forward_cancel: avoid memory leaks on error + libssh2_channel_open: more detailed error message - Fixes #257 + The error message returned by libssh2_channel_open in case of a server side channel open failure is now more detailed and includes the four standard error conditions in RFC 4254. -- _libssh2_packet_add: avoid using uninitialized memory - - In _libssh2_packet_add, called by _libssh2_packet_read, a call to - _libssh2_packet_send that is supposed to send a one-byte message - SSH_MSG_REQUEST_FAILURE would send an uninitialized byte upon re-entry - if its call to _send returns _EAGAIN. +- [Hannes Domani brought this change] + + kex: fix libgcrypt memory leaks of bignum - Fixes #259 + Fixes #168. -- _libssh2_channel_forward_cancel: accessed struct after free +Marc Hoersken (3 Apr 2015) +- configure.ac: check for SecureZeroMemory for clear memory feature + +- Revert "wincng.c: fix clear memory feature compilation with mingw" - ... and the assignment was pointless anyway since the struct was about - to be freed. Bug introduced in dde2b094. + This reverts commit 2d2744efdd0497b72b3e1ff6e732aa4c0037fc43. - Fixes #268 - -Peter Stuge (2 Jun 2013) -- [Marc Hoersken brought this change] + Autobuilds show that this did not solve the issue. + And it seems like RtlFillMemory is defined to memset, + which would be optimized out by some compilers. - Fixed compilation using mingw-w64 +- wincng.c: fix clear memory feature compilation with mingw -- [Marc Hoersken brought this change] +Alexander Lamaison (1 Apr 2015) +- [LarsNordin-LNdata brought this change] - knownhost.c: use LIBSSH2_FREE macro instead of free + Enable use of OpenSSL that doesn't have DSA. - Use LIBSSH2_FREE instead of free since - _libssh2_base64_encode uses LIBSSH2_ALLOC + Added #if LIBSSH2_DSA for all DSA functions. -Daniel Stenberg (18 May 2013) -- [Matthias Kerestesch brought this change] +- [LarsNordin-LNdata brought this change] - libssh2_agent_init: init ->fd to LIBSSH2_INVALID_SOCKET - - ... previously it was left at 0 which is a valid file descriptor! - - Bug: https://trac.libssh2.org/ticket/265 + Use correct no-blowfish #define with OpenSSL. - Fixes #265 + The OpenSSL define is OPENSSL_NO_BF, not OPENSSL_NO_BLOWFISH. -- userauth_password: pass on the underlying error code - - _libssh2_packet_requirev() may return different errors and we pass that - to the parent instead of rewriting it. +Marc Hoersken (25 Mar 2015) +- configure: error if explicitly enabled clear-memory is not supported - Bug: http://libssh2.org/mail/libssh2-devel-archive-2013-04/0029.shtml - Reported by: Cosmin + This takes 22bd8d81d8fab956085e2079bf8c29872455ce59 and + b8289b625e291bbb785ed4add31f4759241067f3 into account, + but still makes it enabled by default if it is supported + and error out in case it is unsupported and was requested. -Peter Stuge (9 May 2013) -- [Marc Hoersken brought this change] +Daniel Stenberg (25 Mar 2015) +- configure: make clear-memory default but only WARN if backend unsupported + + ... instead of previous ERROR. - libcrypt.c: Fix typo in _libssh2_rsa_sha1_sign() parameter type +Marc Hoersken (24 Mar 2015) +- wincng.h: fix warning about computed return value not being used -Kamil Dudka (4 May 2013) -- configure.ac: replace AM_CONFIG_HEADER with AC_CONFIG_HEADERS - - Reported by: Quintus - Bug: https://trac.libssh2.org/ticket/261 +- nonblocking examples: fix warning about unused tvdiff on Mac OS X -Guenter Knauf (12 Apr 2013) -- Fixed copyright string for NetWare build. +Daniel Stenberg (24 Mar 2015) +- openssl: fix compiler warnings -Daniel Stenberg (9 Apr 2013) -- [Richard W.M. Jones brought this change] +- cofigure: fix --disable-clear-memory check - sftp: Add support for fsync (OpenSSH extension). - - The new libssh2_sftp_fsync API causes data and metadata in the - currently open file to be committed to disk at the server. - - This is an OpenSSH extension to the SFTP protocol. See: +Marc Hoersken (23 Mar 2015) +- scp.c: improved command length calculation - https://bugzilla.mindrot.org/show_bug.cgi?id=1798 + Reduced number of calls to strlen, because shell_quotearg already + returns the length of the resulting string (e.q. quoted path) + which we can add to the existing and known cmd_len. + Removed obsolete call to memset again, because we can put a final + NULL-byte at the end of the string using the calculated length. -- [Richard W.M. Jones brought this change] +- scp.c: improved and streamlined formatting - sftp: statvfs: Along error path, reset the correct 'state' variable. +- scp.c: fix that scp_recv may transmit not initialised memory -- [Richard W.M. Jones brought this change] +- scp.c: fix that scp_send may transmit not initialised memory + + Fixes ticket 244. Thanks Torsten. - sftp: seek: Don't flush buffers on same offset +- kex: do not ignore failure of libssh2_sha1_init() - Signed-off-by: Richard W.M. Jones + Based upon 43b730ce56f010e9d33573fcb020df49798c1ed8. + Fixes ticket 290. Thanks for the suggestion, mstrsn. -Guenter Knauf (9 Feb 2013) -- Updated dependency libs. +- wincng.h: fix return code of libssh2_md5_init() -- Fixed tool macro names. +- openssl.c: fix possible segfault in case EVP_DigestInit fails -Daniel Stenberg (29 Nov 2012) -- [Seth Willits brought this change] +- wincng.c: fix possible use of uninitialized variables - compiler warnings: typecast strlen in macros +- wincng.c: fix unused argument warning if clear memory is not enabled + +- wincng: Added explicit clear memory feature to WinCNG backend - ... in macro parameters to avoid compiler warnings about lost precision. + This re-introduces the original feature proposed during + the development of the WinCNG crypto backend. It still needs + to be added to libssh2 itself and probably other backends. - Several macros in libssh2.h call strlen and pass the result directly to - unsigned int parameters of other functions, which warns about precision - loss because strlen returns size_t which is unsigned long on at least - some platforms (such as OS X). The fix is to simply typecast the - strlen() result to unsigned int. - -- libssh2.h: bump version to 1.4.4-DEV - -Version 1.4.3 (27 Nov 2012) + Memory is cleared using the function SecureZeroMemory which is + available on Windows systems, just like the WinCNG backend. -Daniel Stenberg (27 Nov 2012) -- RELEASE-NOTES: fixed for 1.4.3 +- wincng.c: fixed mixed line-endings -- sftp_read: return error if a too large package arrives +- wincng.c: fixed use of invalid parameter types in a8d14c5dcf -Peter Stuge (13 Nov 2012) -- Only define _libssh2_dsa_*() functions when building with DSA support +- wincng.c: only try to load keys corresponding to the algorithm -Guenter Knauf (8 Nov 2012) -- Added .def file to output. +- wincng.c: moved PEM headers into definitions -Kamil Dudka (1 Nov 2012) -- libssh2_hostkey_hash.3: update the description of return value - - The function returns NULL also if the hash algorithm is not available. +- wincng.h: fixed invalid parameter name -Guenter Knauf (24 Oct 2012) -- Fixed mode acciedently committed. +- wincng: fixed mismatch with declarations in crypto.h -- Ignore generated file. +- userauth.c: fixed warning C6001: using uninitialized sig and sig_len -- Added hack to make use of Makefile.inc. - - This should avoid further maintainance of the objects list. +- pem.c: fixed warning C6269: possible incorrect order of operations -- Fixed MSVC NMakefile. +- wincng: add support for authentication keys to be passed in memory - Added missing source files; added resource for DLL. + Based upon 18cfec8336e and daa2dfa2db. -Kamil Dudka (22 Oct 2012) -- examples: use stderr for messages, stdout for data +- pem.c: add _libssh2_pem_parse_memory to parse PEM from memory - Reported by: Karel Srot - Bug: https://bugzilla.redhat.com/867462 + Requirement to implement 18cfec8336e for Libgcrypt and WinCNG. -- openssl: do not leak memory when handling errors - - ,.. in aes_ctr_init(). Detected by Coverity. +- pem.c: fix copy and paste mistake from 55d030089b8 -- channel: fix possible NULL dereference - - ... in libssh2_channel_get_exit_signal(). Detected by Coverity. +- userauth.c: fix another possible dereference of a null pointer -- Revert "aes: the init function fails when OpenSSL has AES support" - - This partially reverts commit f4f2298ef3635acd031cc2ee0e71026cdcda5864. - - We need to use the EVP_aes_???_ctr() functions in FIPS mode. +- userauth.c: fix possible dereference of a null pointer -- crypt: use hard-wired cipher block sizes consistently +- pem.c: reduce number of calls to strlen in readline -- openssl: do not ignore failure of EVP_CipherInit() +Alexander Lamaison (17 Mar 2015) +- [Will Cosgrove brought this change] -- kex: do not ignore failure of libssh2_md5_init() + Initialise HMAC_CTX in more places. - The MD5 algorithm is disabled when running in FIPS mode. - -Daniel Stenberg (21 Aug 2012) -- [Peter Krempa brought this change] + Missed a couple more places we init ctx to avoid openssl threading crash. - known_hosts: Fail when parsing unknown keys in known_hosts file. +- Build build breakage in WinCNG backend caused when adding libssh2_userauth_publickey_frommemory. - libssh2_knownhost_readfile() silently ignored problems when reading keys - in unsupported formats from the known hosts file. When the file is - written again from the internal structures of libssh2 it gets truntcated - to the point where the first unknown key was located. + The new feature isn't implemented for the WinCNG backend currently, but the WinCNG backend didn't contain any implementation of the required backend functions - even ones that returns an error. That caused link errors. - * src/knownhost.c:libssh2_knownhost_readfile() - return error if key - parsing fails + This change fixes the problem by providing an implementation of the backend functions that returns an error. -- AUTHORS: synced with 42fec44c8a4 +- Fix breakage in WinCNG backend caused by introducing libssh2_hmac_ctx_init. - 31 recent authors added + The macro was defined to nothing for the libgcrypt backend, but not for WinCNG. This brings the latter into line with the former. -- [Dave Hayden brought this change] +Daniel Stenberg (15 Mar 2015) +- userauth_publickey_frommemory.3: add AVAILABILITY + + ... it will be added in 1.6.0 - compression: add support for zlib@openssh.com +- libssh2: next version will be called 1.6.0 - Add a "use_in_auth" flag to the LIBSSH2_COMP_METHOD struct and a - separate "zlib@openssh.com" method, along with checking session->state - for LIBSSH2_STATE_AUTHENTICATED. Appears to work on the OpenSSH servers - I've tried against, and it should work as before with normal zlib - compression. + ... since we just added a new function. -- [Dmitry Smirnov brought this change] +- docs: add libssh2_userauth_publickey_frommemory.3 to dist + + The function and man page were added in commit 18cfec8336e - configure: gcrypt doesn't come with pkg-config support +- [Jakob Egger brought this change] + + direct_tcpip: Fixed channel write - ... so use plain old -lgcrypt to the linker to link with it. + There were 3 bugs in this loop: + 1) Started from beginning after partial writes + 2) Aborted when 0 bytes were sent + 3) Ignored LIBSSH2_ERROR_EAGAIN - Fixes #225 + See also: + https://trac.libssh2.org/ticket/281 + https://trac.libssh2.org/ticket/293 -- sftp_read: Value stored to 'next' is never read +Alexander Lamaison (15 Mar 2015) +- [Will Cosgrove brought this change] + + Must init HMAC_CTX before using it. - Detected by clang-analyzer + Must init ctx before using it or openssl will reuse the hmac which is not thread safe and causes a crash. + Added libssh2_hmac_ctx_init macro. -- publickey_init: errors are negative, fix check +- Add continuous integration configurations. - Detected by clang-analyzer. + Linux-based CI is done by Travis CI. Windows-based CI is done by Appveyor. -- [Maxime Larocque brought this change] +- [David Calavera brought this change] - session_free: wrong variable used for keeping state - - If libssh2_session_free is called without the channel being freed - previously by libssh2_channel_free a memory leak could occur. - - A mismatch of states variables in session_free() prevent the call to - libssh2_channel_free function. session->state member is used instead of - session->free_state. - - It causes a leak of around 600 bytes on every connection on my systems - (Linux, x64 and PPC). + Allow authentication keys to be passed in memory. - (Debugging done under contract for Accedian Networks) + All credits go to Joe Turpin, I'm just reaplying and cleaning his patch: + http://www.libssh2.org/mail/libssh2-devel-archive-2012-01/0015.shtml - Fixes #246 - -Guenter Knauf (29 Jun 2012) -- Small NetWare makefile tweak. + * Use an unimplemented error for extracting keys from memory with libgcrypt. -- Some small Win32 makefile fixes. +Daniel Stenberg (14 Mar 2015) +- docs: include the renamed INSTALL* files in dist -Daniel Stenberg (19 Jun 2012) -- libssh2_userauth_publickey_fromfile_ex.3: mention publickey == NULL +Alexander Lamaison (13 Mar 2015) +- Prevent collisions between CMake and Autotools in examples/ and tests/. -- comp_method_zlib_decomp: handle Z_BUF_ERROR when inflating - - When using libssh2 to perform an SFTP file transfer from the "JSCAPE MFT - Server" (http://www.jscape.com) the transfer failed. The default JSCAPE - configuration is to enforce zlib compression on SSH2 sessions so the - session was compressed. The relevant part of the debug trace contained: +- Avoid clash between CMake build and Autotools. - [libssh2] 1.052750 Transport: unhandled zlib error -5 - [libssh2] 1.052750 Failure Event: -29 - decompression failure + Autotools expects a configuration template file at src/libssh2_config.h.in, which buildconf generates. But the CMake build system has its CMake-specific version of the file at this path. This means that, if you don't run buildconf, the Autotools build will fail because it configured the wrong header template. - The trace comes from comp_method_zlib_decomp() in comp.c. The "unhandled - zlib error -5" is the status returned from the zlib function - inflate(). The -5 status corresponds to "Z_BUF_ERROR". - - The inflate() function takes a pointer to a z_stream structure and - "inflates" (decompresses) as much as it can. The relevant fields of the - z_stream structure are: - - next_in - pointer to the input buffer containing compressed data - avail_in - the number of bytes available at next_in - next_out - pointer to the output buffer to be filled with uncompressed - data - avail_out - how much space available at next_out - - To decompress data you set up a z_stream struct with the relevant fields - filled in and pass it to inflate(). On return the fields will have been - updated so next_in and avail_in show how much compressed data is yet to - be processed and next_out and avail_out show how much space is left in - the output buffer. - - If the supplied output buffer is too small then on return there will be - compressed data yet to be processed (avail_in != 0) and inflate() will - return Z_OK. In this case the output buffer must be grown, avail_out - updated and inflate() called again. - - If the supplied output buffer was big enough then on return the - compressed data will have been exhausted (avail_in == 0) and inflate() - will return Z_OK, so the data has all been uncompressed. - - There is a corner case where inflate() makes no progress. That is, there - may be unprocessed compressed data and space available in the output - buffer and yet the function does nothing. In this case inflate() will - return Z_BUF_ERROR. From the zlib documentation and the source code it - is not clear under what circumstances this happens. It could be that it - needs to write multiple bytes (all in one go) from its internal state to - the output buffer before processing the next chunk of input but but - can't because there is not enough space (though my guesses as to the - cause are not really relevant). Recovery from Z_BUF_ERROR is pretty - simple - just grow the output buffer, update avail_out and call - inflate() again. - - The comp_method_zlib_decomp() function does not handle the case when - inflate() returns Z_BUF_ERROR. It treats it as a non-recoverable error - and basically aborts the session. - - Fixes #240 + See https://github.com/libssh2/libssh2/pull/8. -Guenter Knauf (12 Jun 2012) -- MinGW makefile tweaks. +- Merge pull request #8 from alamaison/cmake - Use GNU tools when compiling on Linux. - Fixed dist and dev targets. + CMake build system. -- NetWare makefile tweaks. +- CMake build system. - Changed to use Windows commandline tools instead of - GNU tools when compiling on Windows. Fixed dist and - dev targets. Enabled nlmconv error for unresolved - symbols. - -Daniel Stenberg (11 Jun 2012) -- Revert "config.rpath: generated file, no need to keep in git" + Tested: + - Windows: + - Visual C++ 2005/2008/2010/2012/2013/MinGW-w64 + - static/shared + - 32/64-bit + - OpenSSL/WinCNG + - Without zlib + - Linux: + - GCC 4.6.3/Clang 3.4 + - static/shared + - 32/64-bit + - OpenSSL/Libgcrypt + - With/Without zlib + - MacOS X + - AppleClang 6.0.0 + - static + - 64-bit + - OpenSSL + - Without zlib - This reverts commit 1ac7bd09cc685755577fb2c8829adcd081e7ab3c. + Conflicts: + README + +- Man man syntax tests fail gracefully if man version is not suitable. + +- Return valid code from test fixture on failure. - This file still used by lib/*m4 functions so we need to keep the file - around. + The sshd test fixture was returning -1 if an error occurred, but negative error codes aren't technically valid (google it). Bash on Windows converted them to 0 which made setup failure look as though all tests were passing. -- BINDINGS: added PySsh2, a Python-ctypes binding +- Let mansyntax.sh work regardless of where it is called from. -Guenter Knauf (8 Jun 2012) -- Fixed MinGW debug build. +Daniel Stenberg (12 Mar 2015) +- [Viktor Szakáts brought this change] -Daniel Stenberg (5 Jun 2012) -- BINDINGS: Added the Cocoa/Objective-C one + mingw build: allow to pass custom CFLAGS - ... and sorted the bindings after the languages, alphabetically + Allow to pass custom `CFLAGS` options via environment variable + `LIBSSH2_CFLAG_EXTRAS`. Default and automatically added options of + `GNUmakefile` have preference over custom ones. This addition is useful + for passing f.e. custom CPU tuning or LTO optimization (`-flto + -ffat-lto-objects`) options. The only current way to do this is to edit + `GNUmakefile`. This patch makes it unnecessary. - Reported by: Mike Abdullah + This is a mirror of similar libcurl patch: + https://github.com/bagder/curl/pull/136 -- BINDINGS: document the bindings we know of +- [Will Cosgrove brought this change] -Guenter Knauf (4 Jun 2012) -- Fixed LIBSSH2_INT64_T_FORMAT macro. + userauth: Fixed prompt text no longer being copied to the prompts struct - Usually a format macro should hold the whole format, otherwise - it should be named a prefix. Also fixed usage of this macro in - scp.c for a signed var where it was used as prefix for unsigned. + Regression from 031566f9c -- Removed obsolete define from makefiles. +- README: update the git repo locations -- Renamed NetWare makefiles. +- wait_socket: wrong use of difftime() + + With reversed arguments it would always return a negative value... + + Bug: https://github.com/bagder/libssh2/issues/1 -- Renamed NetWare makefiles. +- bump: start working toward 1.5.1 now -- Synced MinGW makefiles with 56c64a6..39e438f. - - Also synced MinGW test makefile with b092696..f8cb874. +Version 1.5.0 (11 Mar 2015) -Peter Stuge (30 May 2012) -- Revert "sftp: Don't send attrs.permissions on read-only SSH_FXP_OPEN" - - This reverts commit 04e79e0c798674a0796be8a55f63dd92e6877790. +Daniel Stenberg (11 Mar 2015) +- RELEASE-NOTES: 1.5.0 release -- sftp: Don't send attrs.permissions on read-only SSH_FXP_OPEN - - This works around a protocol violation in the ProFTPD 1.3.4 mod_sftp - server, as reported by Will Cosgrove in: - - http://libssh2.org/mail/libssh2-devel-archive-2012-05/0079.shtml +- [Mariusz Ziulek brought this change] + + kex: bail out on rubbish in the incoming packet - Based on a suggested fix by TJ Saunders in: + CVE-2015-1782 - http://libssh2.org/mail/libssh2-devel-archive-2012-05/0104.shtml + Bug: http://www.libssh2.org/adv_20150311.html -Guenter Knauf (28 May 2012) -- Try to detect OpenSSL build type automatically. +- docs: move INSTALL, AUTHORS, HACKING and TODO to docs/ - Also fixed recently added libgdi32 linkage which is only - required when OpenSSL libs are linked statically. - -Daniel Stenberg (25 May 2012) -- config.rpath: generated file, no need to keep in git + And with this, cleanup README to be shorter and mention the new source + code home. -Guenter Knauf (22 May 2012) -- Updated dependency libary versions. +- .gitignore: don't ignore INSTALL -Daniel Stenberg (18 May 2012) -- 1.4.3: towards the future +Dan Fandrich (4 Mar 2015) +- examples/x11.c: include sys/select.h for improved portability -Version 1.4.2 (18 May 2012) +Daniel Stenberg (4 Mar 2015) +- RELEASE-NOTES: synced with a8473c819bc068 + + In preparation for the upcoming 1.5.0 release. -Daniel Stenberg (18 May 2012) -- RELEASE-NOTES: synced with 92a9f952794 +Guenter Knauf (8 Jan 2015) +- NetWare build: added some missing exports. -Alexander Lamaison (15 May 2012) -- win32/libssh2_config.h: Remove hardcoded #define LIBSSH2_HAVE_ZLIB. - - Rationale: Everything else in this file states a fact about the win32 - platform that is unconditional for that platform. There is nothing - unconditional about the presence of zlib. It is neither included with - Windows nor with the platform SDK. Therefore, this is not an appropriate - place to assert its presence. Especially as, once asserted, it cannot be - overridden using a compiler flag. +Marc Hoersken (29 Dec 2014) +- knownhost.c: fix use of uninitialized argument variable wrote - In contrast, if it is omitted, then it can easily be reasserted by adding - a compiler flag defining LIBSSH2_HAVE_ZLIB. + Detected by clang scan in line 1195, column 18. -Daniel Stenberg (14 May 2012) -- RELEASE-NOTES: synced with 69a3354467c +- examples/x11.c: fix result of operation is garbage or undefined + + Fix use of uninitialized structure w_size_bck. + Detected by clang scan in line 386, column 28. -- _libssh2_packet_add: SSH_MSG_CHANNEL_REQUEST default to want_reply +- examples/x11.c: remove dead assigments of some return values - RFC4254 says the default 'want_reply' is TRUE but the code defaulted to - FALSE. Now changed. + Detected by clang scan in line 212, column 9. + Detected by clang scan in line 222, column 13. + Detected by clang scan in line 410, column 13. + +- examples/x11.c: fix possible memory leak if read fails - Fixes #233 + Detected by clang scan in line 224, column 21. -- gettimeofday: no need for a replacement under cygwin +- examples/x11.c: fix invalid removal of first list element - Fixes #224 + Fix use of memory after it was being freed. + Detected by clang scan in line 56, column 12. -Alexander Lamaison (13 May 2012) -- Prevent sftp_packet_read accessing freed memory. +- userauth.c: make sure that sp_len is positive and avoid overflows - sftp_packet_add takes ownership of the packet passed to it and (now that we - handle zombies) might free the packet. sftp_packet_read uses the packet type - byte as its return code but by this point sftp_packet_add might have freed - it. This change fixes the problem by caching the packet type before calling - sftp_packet_add. + ... if the pointer subtraction of sp1 - pubkey - 1 resulted in a + negative or larger value than pubkey_len, memchr would fail. - I don't understand why sftp_packet_read uses the packet type as its return - code. A future change might get rid of this entirely. + Reported by Coverity CID 89846. -Daniel Stenberg (12 May 2012) -- sftp_packet_flush: flush zombies too +- channel.c: remove logically dead code, host cannot be NULL here - As this function is called when the SFTP session is closed, it needs to - also kill all zombies left in the SFTP session to avoid leaking memory - just in case some zombie would still be in there. - -- sftp_packetlist_flush: zombies must not have responses already + ... host cannot be NULL in line 525, because it is always + valid (e.g. at least set to "0.0.0.0") after lines 430 and 431. - When flushing the packetlist, we must only add the request as a zombie - if no response has already been received. Otherwise we could wrongly - make it a zombie even though the response was already received and then - we'd get a zombie stuck there "forever"... + Reported by Coverity CID 89807. -- sftp_read: on EOF remove packet before flush +- session.c: check return value of session_nonblock during startup - Since the sftp_packetlist_flush() function will move all the existing - FXP_READ requests in this handle to the zombie list we must first remove - this just received packet as it is clearly not a zombie. + Reported by Coverity CID 89803. -- sftp_packet_require: sftp_packet_read() returning 0 is not an error +- session.c: check return value of session_nonblock in debug mode - Exactly as the comment in the code said, checking the return code from - sftp_packet_read() with <= was wrong and it should be < 0. With the new - filtering on incoming packets that are "zombies" we can now see this - getting zero returned. + Reported by Coverity CID 89805. -- sftp_packetlist_flush: only make it zombie if it was sent - - The list of outgoing packets may also contain packets that never were - sent off and we better not make them zombies too. +- pem.c: fix mixed line-endings introduced with 8670f5da24 -- [Alexander Lamaison brought this change] +- pem.c: make sure there's a trailing zero and b64data is not NULL + + ... if there is no base64 data between PEM header and footer. + Reported by Coverity CID 89823. - Mark outstanding read requests after EOF as zombies. +- kex.c: make sure mlist is not set to NULL - In order to be fast, sftp_read sends many read requests at once. With a small - file, this can mean that when EOF is received back, many of these requests are - still outstanding. Responses arriving after we close the file and abandon the - file handle are queued in the SFTP packet queue and never collected. This - causes transfer speed to drop as a progressively longer queue must be searched - for every packet. + ... if the currently unsupported LANG methods are called. + Reported by Coverity CID 89834. + +- packet.c: i < 256 was always true and i would overflow to 0 - This change introduces a zombie request-ID list in the SFTP session that is - used to recognise these outstanding requests and prevent them being added to - the queue. + Visualize that the 0-termination is intentional, because the array + is later passed to strlen within _libssh2_packet_askv. -Peter Stuge (23 Apr 2012) -- [Rafael Kitover brought this change] +- silence multiple data conversion warnings - Update win32/GNUmakefile to use OpenSSL 1.0.1a +Daniel Stenberg (23 Dec 2014) +- agent_connect_unix: make sure there's a trailing zero - libcrypto on win32 now depends on gdi32.dll, so move the OpenSSL LDLIBS - block to before the compiler definitions, so that libcrypto gets added - first, and then add -lgdi32 into the following common LDLIBS for gcc. + ... if the path name was too long. Reported by Coverity CID 89801. -Guenter Knauf (23 Apr 2012) -- Changed 'Requires' to 'Requires.private'. +Marc Hoersken (22 Dec 2014) +- examples on Windows: use native SOCKET-type instead of int - Only static builds need to link against the crypto libs. + And check return values accordingly. -- Fixed 'Requires:' names. - - The 'Requires:' line lists the names of the .pc files. +- userauth.c: improve readability and clarity of for-loops -- Added 'Requires:' line to libssh2.pc. +Daniel Stenberg (22 Dec 2014) +- calloc: introduce LIBSSH2_CALLOC() - This is necessary so that other libs which lookup libssh2 info - via pkg-config can add the right crypto lib dependencies. + A simple function using LIBSSH2_ALLOC + memset, since this pattern was + used in multiple places and this simplies code in general. -- Updated dependency lib versions. +Marc Hoersken (15 Dec 2014) +- libssh2_priv.h: Ignore session, context and format parameters -Peter Stuge (18 Apr 2012) -- configure.ac: Add option to disable build of the example applications +- x11 example: check return value of socket function + +- examples: fixed mixed line-endings introduced with aedfba25b8 + +- wincng.c: explicitly ignore BCrypt*AlgorithmProvider return codes - Examples are built by default. Any of the following options on the - configure command line will skip building them: + Fixes VS2012 code analysis warning C6031: + return value ignored: could return unexpected value + +- wincng.c: fix possible invalid memory write access - --disable-examples-build - --enable-examples-build=no - --enable-examples-build=false + Fixes VS2012 code analysis warning C6386: + buffer overrun: accessing 'pbOutput', the writable size is + 'cbOutput' bytes, but '3' bytes may be written: libssh2 wincng.c 610 -- userauth.c: fread() from public key file to correctly detect any errors +- tests on Windows: check for WSAStartup return code - If the filename parameter for file_read_publickey() was the name of a - directory instead of a file then libssh2 would spin trying to fgetc() - from the FILE * for the opened directory when trying to determine the - length of the encoded public key, since fgetc() can't report errors. + Fixes VS2012 code analysis warning C6031: + return value ignored: could return unexpected value + +- wincng.c: fix possible NULL pointer de-reference of bignum - Use fread() instead to correctly detect this error condition along - with many others. + Fixes VS2012 code analysis warning C6011: + dereferencing NULL pointer 'bignum'. libssh2 wincng.c 1567 + +- wincng.c: fix possible use of uninitialized memory - This fixes the problem reported in - http://www.libssh2.org/mail/libssh2-devel-archive-2012-04/0021.shtml + Fixes VS2012 code analysis warning C6001: + using uninitialized memory 'cbDecoded'. libssh2 wincng.c 553 + +- packet.c: fix possible NULL pointer de-reference within listen_state - Reported-by: Oleksiy Zagorskyi + Fixes VS2012 code analysis warning C6011: + dereferencing NULL pointer 'listen_state->channel'. libssh2 packet.c 221 -- Return LIBSSH2_ERROR_SOCKET_DISCONNECT on EOF when reading banner +- kex.c: fix possible NULL pointer de-reference with session->kex + + Fixes VS2012 code analysis warning C6011: + dereferencing NULL pointer 'session->kex'. libssh2 kex.c 1761 -Guenter Knauf (17 Apr 2012) -- Fixed copyright year. +- agent.c: check return code of MapViewOfFile + + Fixes VS2012 code analysis warning C6387: 'p+4' may be '0': + this does not adhere to the specification for the function + 'memcpy': libssh2 agent.c 330 + + Fixes VS2012 code analysis warning C6387: 'p' may be '0': + this does not adhere to the specification for the function + 'UnmapViewOfFile': libssh2 agent.c 333 -- Updated dependency lib versions in static makefiles. +- examples on Windows: check for socket return code + + Fixes VS2012 code analysis warning C28193: + The variable holds a value that must be examined -Daniel Stenberg (6 Apr 2012) -- version: bump to 1.4.2 +- examples on Windows: check for WSAStartup return code - We're on the 1.4.2 track now (at least) + Fixes VS2012 code analysis warning C6031: + return value ignored: could return unexpected value -Version 1.4.1 (4 Apr 2012) +Guenter Knauf (11 Dec 2014) +- wincng.c: silent some more gcc compiler warnings. -Daniel Stenberg (4 Apr 2012) -- RELEASE-NOTES: updated for 1.4.1 release +- wincng.c: silent gcc compiler warnings. -- always do "forced" window updates +- Watcom build: added support for WinCNG build. + +- build: updated dependencies in makefiles. + +Daniel Stenberg (4 Dec 2014) +- configure: change LIBS not LDFLAGS when checking for libs - When calling _libssh2_channel_receive_window_adjust() internally, we now - always use the 'force' option to prevent libssh2 to avoid sending the - update if the update isn't big enough. + Closes #289 - It isn't fully analyzed but we have seen corner cases which made a - necessary window update not get send due to this and then the other side - doesn't send data our side then sits waiting for forever. + Patch-by: maurerpe -- channel_read: force window adjusts! +Guenter Knauf (3 Dec 2014) +- MinGW build: some more GNUMakefile tweaks. - if there's not enough room to receive the data that's being requested, - the window adjustment needs to be sent to the remote and thus the force - option has to be used. _libssh2_channel_receive_window_adjust() would - otherwise "queue" small window adjustments for a later packet but that - is really terribly for the small buffer read that for example is the - final little piece of a very large file as then there is no logical next - packet! + test/GNUmakefile: added architecture autodetection; added switches to + CFLAGS and RCFLAGS to make sure that the right architecture is used. + Added support to build with WinCNG. + +- sftpdir.c: added authentication method detection. - Reported by: Armen Babakhanian - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0130.shtml + Stuff copied over from ssh2.c to make testing a bit easier. -- [Paul Howarth brought this change] +- NMake build: fixed LIBS settings. - aes: the init function fails when OpenSSL has AES support - - The internal init function only worked fine when the configure script - didn't detect the OpenSSL AES_CTR function! +- NMake build: added support for WinCNG build. + +- MinGW build: some GNUMakefile tweaks. - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0111.shtml - Reported by: Paul Howarth + Added architecture autodetection; added switches to CFLAGS and + RCFLAGS to make sure that the right architecture is used. + Added support to build with WinCNG. -- [Matthew Booth brought this change] +- MinGW build: Fixed redefine warnings. - transport_send: Finish in-progress key exchange before sending data - - _libssh2_channel_write() first reads outstanding packets before writing - new data. If it reads a key exchange request, it will immediately start - key re-exchange, which will require sending a response. If the output - socket is full, this will result in a return from - _libssh2_transport_read() of LIBSSH2_ERROR_EAGAIN. In order not to block - a write because there is no data to read, this error is explicitly - ignored and the code continues marshalling a packet for sending. When it - is sent, the remote end immediately drops the connection because it was - expecting a continuation of the key exchange, but got a data packet. - - This change adds the same check for key exchange to - _libssh2_transport_send() that is in _libssh2_transport_read(). This - ensures that key exchange is completed before any data packet is sent. +- Updated copyright year. -- channel_write: acknowledge transport errors - - When draining data off the socket with _libssh2_transport_read() (which - in turn has to be done so that we can be sure to have read any possible - window-increasing packets), this code previously ignored errors which - could lead to nasty loops. Now all error codes except EAGAIN will cause - the error to be returned at once. - - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0068.shtml - Reported by: Matthew Booth +Daniel Stenberg (31 Aug 2014) +- COPYING: bump the copyright year -- [Steven Dake brought this change] +Dan Fandrich (28 Jul 2014) +- docs: fixed a bunch of typos - In examples/x11.c, Make sure sizeof passed to read operation is correct - - sizeof(buf) expands to 8 or 4 (since its a pointer). This variable may - have been static in the past, leading to this error. +- docs: added missing libssh2_session_handshake.3 file + +Marc Hoersken (19 May 2014) +- wincng.c: specify the required libraries for dependencies using MSVC - Signed-off-by: Steven Dake + Initially reported by Bob Kast as "for MS VS builds, specify the + libraries that are required so they don't need to go into all + project files that may use this library". Thanks a lot. -- [Steven Dake brought this change] +- [Bob Kast brought this change] - Fix suspicious sizeof usage in examples/x11.c + windows build: do not export externals from static library - In the x11 example, sizeof(buf) = 8UL (on x86_64), when this should - probably represent the buffer size available. I am not sure how to - test that this change is actually correct, however. + If you are building a DLL, then you need to explicitly export each + entry point. When building a static library, you should not. - Signed-off-by: Steven Dake + libssh2 was exporting the entry points whether it was building a DLL or a + static library. To elaborate further, if libssh2 was used as a static + library, which was being linked into a DLL, the libssh2 API would be + exported from that separate DLL. -- sftp_packet_read: follow-up fix for EAGAIN/window adjust - - The commit in 7194a9bd7ba45 wasn't complete. This change makes sure - variables are initialized properly before used in the EAGAIN and window - adjust cases. +Daniel Stenberg (19 May 2014) +- [Mikhail Gusarov brought this change] -- sftp_packet_add: use named error code instead of number + Fix typos in manpages -- sftp_packet_add: verify the packet before accepting it - - In order to bail out as quickly as possible when things are wrong and - out of sync, make sure the SFTP message is one we understand. +Marc Hoersken (18 May 2014) +- wincng.c: Fixed memory leak in case of an error during ASN.1 decoding -- SFTP: preserve the original error code more +- configure: Display individual crypto backends on separate lines - Lots of places in the code translated the original error into the more - generic LIBSSH2_ERROR_SOCKET_TIMEOUT but this turns out to distort the - original error reason a lot and makes tracking down the real origin of a - problem really hard. This change makes the original error code be - preserved to a larger extent when return up to the parent function. - -- sftp_packet_read: adjust window size as necessary - - Commit 03ca9020756 tried to simplify the window sizing logic but broke - SFTP readdir as there was no window sizing code left there so large - directory listings no longer worked. - - This change introduces window sizing logic to the sftp_packet_read() - function so that it now tells the remote about the local size having a - window size that suffice when it is about to ask for directory data. + This avoids line-wrapping in between parameters and makes the + error message look like the following: - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0069.shtml - Reported by: Eric + configure: error: No crypto library found! + Try --with-libssl-prefix=PATH + or --with-libgcrypt-prefix=PATH + or --with-wincng on Windows -- [Steven Dake brought this change] +- [Bob Kast brought this change] - Tell C compiler we don't care about return code of libssh2_init - - The call of libssh2_init returns a return code, but nothing could be done - within the _libssh2_init_if_needed execution path. + libssh2_priv.h: a 1 bit bit-field should be unsigned - Signed-off-by: Steven Dake + some compilers may not like this -- [Steven Dake brought this change] +- knownhost.c: Fixed warning that pointer targets differ in signedness - Add comment indicating a resource leak is not really a resource leak - - While possibly obvious to those investigating the code, coverity complains - about this out of scope leak. +- wincng.c: Fixed warning about pointer targets differing in signedness + +- tcpip-forward.c: Fixed warning that pointer targets differ in signedness - Signed-off-by: Steven Dake + libssh2_channel_forward_listen_ex uses ints instead of unsigned ints. -- [Steven Dake brought this change] +- misc.c: Fixed warning about mixed declarations and code - Use safer snprintf rather then sprintf in scp_send() - - Signed-off-by: Steven Dake +- libgcrypt.h: Fixed warning about pointer targets differing in signedness -- [Steven Dake brought this change] +- wincng.h: Fixed warning about pointer targets differing in signedness - Use safer snprintf rather then sprintf in scp_recv() - - While the buffer is indeed allocated to a safe length, better safe then sorry. - - Signed-off-by: Steven Dake +- misc.c: Fixed warning about unused parameter abstract -- [Steven Dake brought this change] +- tcpip-forward.c: Removed unused variables shost, sport and sockopt - use snprintf in knownhost_writeline() rather then sprintf +- wincng.h: Added forward declarations for all WinCNG functions - Although the function checks the length, if the code was in error, there - could potentially be a buffer overrun with the use of sprintf. Instead replace - with snprintf. + Initially reported by Bob Kast as "Wincng - define function + prototypes for wincng routines". Thanks a lot. - Signed-off-by: Steven Dake + Also replaced structure definitions with type definitions. -- [Steven Dake brought this change] +- [Bob Kast brought this change] - Add tracing to print packets left on session at libssh2_session_free - - Signed-off-by: Steven Dake + libssh2.h: on Windows, a socket is of type SOCKET, not int -Peter Stuge (2 Mar 2012) -- Define and use LIBSSH2_INVALID_SOCKET instead of INVALID_SOCKET +- win32: Added WinCNG targets to generated Visual Studio project - INVALID_SOCKET is a special value in Windows representing a - non-valid socket identifier. We were #defining this to -1 on - non-Windows platforms, causing unneccessary namespace pollution. - Let's have our own identifier instead. + Inspired by Bob Kast's reports, this commit enables the compilation + of libssh2 with WinCNG using the generated Visual Studio project files. + This commit adds WinCNG support to parts of the existing Win32 build + infrastructure, until new build systems, like pre-defined VS project + files or CMake files may be added. - Thanks to Matt Lawson for pointing this out. - -- nw/Makefile.netware: Fix project name typo to avoid needless confusion + This commit and b20bfeb3e519119a48509a1099c06d65aa7da1d7 raise one + question: How to handle build systems, like VS project files, that + need to include all source files regardless of the desired target, + including all supported crypto backends? For now the mentioned commit + added a check for LIBSSH2_OPENSSL to openssl.c and with this commit + the supported crypto backends are hardcoded within Makefile.am. -- example/x11: Set raw terminal mode manually instead of with cfmakeraw() - - OpenSolaris has no cfmakeraw() so to make the example more portable - we simply do the equivalent operations on struct termios ourselves. +- libssh2_priv msvc: Removed redundant definition of inline keyword - Thanks to Tom Weber for reporting this problem, and finding a solution. + Initially reported by Bob Kast as "Remove redundant 'inline' define". + Thanks a lot. -Daniel Stenberg (17 Feb 2012) -- sftp_write: cannot return acked data *and* EAGAIN +- wincng: Made data parameter to hash update function constant - Whenever we have acked data and is about to call a function that *MAY* - return EAGAIN we must return the number now and wait to get called - again. Our API only allows data *or* EAGAIN and we must never try to get - both. + Initially reported by Bob Kast as "formal parameter must be const + since it is used in contexts where the actual parameter may be const". + Thanks a lot. -Peter Stuge (13 Feb 2012) -- example/x11: Build only when sys/un.h is found by configure +- wincng: fix cross-compilation against the w64 mingw-runtime package + +- openssl: Check for LIBSSH2_OPENSSL in order to compile with openssl + +- wincng: Fixed use of possible uninitialized variable pPaddingInfo - The example can't be built on systems without AF_UNIX sockets. + Reported by Bob Kast, thanks a lot. -Daniel Stenberg (10 Feb 2012) -- [Alexander Lamaison brought this change] +- wincng: Added cast for double to unsigned long conversion - Simplified sftp_read. +- wincng: Cleaned up includes and check NTSTATUS using macro - Removed the total_read variable that originally must have tracked how - much data had been written to the buffer. With non-blocking reads, we - must return straight away once we have read data into the buffer so this - variable served not purpose. + Removed header file combination that is not supported on a real + Windows platform and can only be compiled using MinGW. Replaced + custom NTSTATUS return code checks with BCRYPT_SUCCESS macro. + +Daniel Stenberg (16 Mar 2014) +- userauth_hostbased_fromfile: zero assign to avoid uninitialized use - I think it was still hanging around in case the initial processing of - 'leftover' data meant we wrote to the buffer but this case, like the - others, must return immediately. Now that it does, the last remaining - need for the variable is gone. + Detected by clang-analyze -- [Alexander Lamaison brought this change] +- channel_receive_window_adjust: store windows size always + + Avoid it sometimes returning without storing it, leaving calling + functions with unknown content! + + Detected by clang-analyzer - Cleaned up sftp_read and added more explanation. +- publickey_packet_receive: avoid junk in returned pointers - Replaced the gotos which were implementing the state machine with - a switch statement which makes the states more explicit. + clang-analyzer found this risk it would return a non-initialized pointer + in a success case -- sftp_read: avoid data *and* EAGAIN +Peter Stuge (16 Mar 2014) +- [Marc Hoersken brought this change] + + Added Windows Cryptography API: Next Generation based backend + +- [Marc Hoersken brought this change] + + knownhost.c: fixed that 'key_type_len' may be used uninitialized - Whenever we have data and is about to call a function that *MAY* return - EAGAIN we must return the data now and wait to get called again. Our API - only allows data *or* EAGAIN and we must never try to get both. + ../src/knownhost.c: In function 'libssh2_knownhost_readline': + ../src/knownhost.c:651:16: warning: 'key_type_len' may be used + uninitialized in this function [-Wmaybe-uninitialized] + rc = knownhost_add(hosts, hostbuf, NULL, + ^ + ../src/knownhost.c:745:12: note: 'key_type_len' was declared here + size_t key_type_len; + ^ -Peter Stuge (2 Feb 2012) -- Add a tcpip-forward example which demonstrates remote port forwarding +- [Marc Hoersken brought this change] -- libssh2.h: Add missing prototype for libssh2_session_banner_set() + pem.c: always compile pem.c independently of crypto backend -- example/subsystem_netconf.c: Return error when read buffer is too small +- Fix non-autotools builds: Always define the LIBSSH2_OPENSSL CPP macro - Also remove a little redundancy in the read loop condition. + Commit d512b25f69a1b6778881f6b4b5ff9cfc6023be42 introduced a crypto + library abstraction in the autotools build system, to allow us to more + easily support new crypto libraries. In that process it was found that + all other build system which we support are hard-coded to build with + OpenSSL. Commit f5c1a0d98bd51aeb24aca3d49c7c81dcf8bd858d fixes automake + introduced into non-autotools build systems but still overlooked the + CPP macro saying that we are using OpenSSL. + + Thanks to Marc Hörsken for identifying this issue and proposing a fix + for win32/{GNUmakefile,config.mk}. This commit uses a slightly different + approach but the end result is the same. -- example/subsystem_netconf.c: Add a missing newline in an error message +Dan Fandrich (15 Mar 2014) +- channel_close: Close the channel even in the case of errors -- Fix undefined reference to _libssh_error in libgcrypt backend +- sftp_close_handle: ensure the handle is always closed - Commit 209de22299b4b58e582891dfba70f57e1e0492db introduced a function - call to a non-existing function, and since then the libgcrypt backend - has not been buildable. + Errors are reported on return, but otherwise the close path is + completed as much as possible and the handle is freed on exit. -Version 1.4.0 (31 Jan 2012) +Alexander Lamaison (6 Mar 2014) +- knownhost: Restore behaviour of `libssh2_knownhost_writeline` with short buffer. + + Commit 85c6627c changed the behaviour of `libssh2_knownhost_writeline` so that it stopped returning the number of bytes needed when the given buffer was too small. Also, the function changed such that is might write to part of the buffer before realising it is too small. + + This commit restores the original behaviour, whilst keeping the unknown-key-type functionality that 85c6627c. Instead of writing to the buffer piecemeal, the length of the various parts is calculated up front and the buffer written only if there is enough space. The calculated necessary size is output in `outlen` regardless of whether the buffer was written to. + + The main use-case for the original behaviour that this commit restores is to allow passing in a NULL buffer to get the actual buffer size needed, before calling the function again with the buffer allocated to the exact size required. -Daniel Stenberg (31 Jan 2012) -- RELEASE-NOTES: synced with 6bd584d29 for 1.4.0 +- knownhost: Fix DSS keys being detected as unknown. + + I missing `else` meant ssh-dss format keys were being re-detected as unknown format. -- s/1.3.1/1.4.0 +Dan Fandrich (6 Mar 2014) +- knownhosts: Abort if the hosts buffer is too small - We're bumping the minor number + This could otherwise cause a match on the wrong host -- [Jernej Kovacic brought this change] +- agent_list_identities: Fixed memory leak on OOM - libssh2_session_supported_algs: fix compiler warning +- Fixed a few typos -- [Jernej Kovacic brought this change] +- userauth: Fixed an attempt to free from stack on error - session_supported_algs docs: added an example +- Fixed a few memory leaks in error paths -- [Gellule Xg brought this change] +- Fixed two potential use-after-frees of the payload buffer + + The first might occur if _libssh2_packet_add returns an error, as + fullpacket_state wasn't reset to idle so if it were possible for + fullpacket to be called again, it would return to the same state + handler and re-use the freed p->packet buffer. + + The second could occur if decrypt returned an error, as it freed the + packet buffer but did not clear total_num, meaning that freed buffer + could be written into again later. - sftp-seek: clear EOF flag +Alexander Lamaison (28 Nov 2013) +- Fix missing `_libssh2_error` in `_libssh2_channel_write`. - Set the EOF flag to False when calling seek64 to be able to get some - data back on a following read + In one case, the error code from `_libssh2_transport_read` was being returned from `_libssh2_channel_write` without setting it as the last error by calling `_libssh2_error`. This commit fixes that. + + Found when using a session whose socket had been inadvertently destroyed. The calling code got confused because via `libssh2_session_last_error` it appeared no error had occurred, despite one being returned from the previous function. -- [Peter Krempa brought this change] +Kamil Dudka (21 Nov 2013) +- [Mark McPherson brought this change] - userauth: Provide more informations if ssh pub key extraction fails + openssl: initialise the digest context before calling EVP_DigestInit() - If the function that extracts/computes the public key from a private key - fails the errors it reports were masked by the function calling it. This - patch modifies the key extraction function to return errors using - _libssh_error() function. The error messages are tweaked to contain - reference to the failed operaton in addition to the reason. + When using the OpenSSL libraries in FIPS mode, the function call + EVP_DigestInit() is actually #defined to FIPS_digestinit(). + Unfortunately wheres EVP_DigestInit() initialises the context and then + calls EVP_DigestInit_ex(), this function assumes that the context has + been pre-initialised and crashes when it isn't. - * AUTHORS: - add my name - * libgcrypt.c: _libssh2_pub_priv_keyfile(): - return a more verbose - error using - _libssh2_error() func. - * openssl.c: - modify call graph of _libssh2_pub_priv_keyfile() to use - _libssh2_error for error reporting(); - * userauth.c: - tweak functions calling _libssh2_pub_priv_keyfile() not - to shadow error messages + Bug: https://trac.libssh2.org/ticket/279 + + Fixes #279 -- TODO: remove issues we (sort of) did already +- [Marc Hörsken brought this change] -- ssh2_exec: skip error outputs for EAGAIN + .gitignore: Ignore files like src/libssh2_config.h.in~ + +Peter Stuge (13 Nov 2013) +- Move automake conditionals added by commit d512b25f out of Makefile.inc - Since the example uses non-blocking mode, it will just flood the output - with this "nonsense" error. + Commit d512b25f69a1b6778881f6b4b5ff9cfc6023be42 added automake + conditionals to Makefile.inc but since Makefile.inc is included + from Makefile for all other build systems that does not work. + + This commit instead adds Makefile.OpenSSL.inc and Makefile.libgcrypt.inc + and moves the automake conditional to its proper place, src/Makefile.am. + + The automake conditional includes the correct Makefile.$name.inc per + the crypto library selection/detection done by configure. + + All non-autotools build system files in libssh2 are hardcoded to use + OpenSSL and do not get a conditional but at least there is some reuse + because they can all include the new Makefile.OpenSSL.inc. -Guenter Knauf (30 Nov 2011) -- Some NetWare makefile tweaks. +Daniel Stenberg (27 Oct 2013) +- [Salvador Fandino brought this change] -Daniel Stenberg (18 Nov 2011) -- LIBSSH2_SFTP_PACKET_MAXLEN: increase to 80000 + Set default window size to 2MB - Some SFTP servers send SFTP packets larger than 40000. Since the limit - is only present to avoid insane sizes anyway, we can easily bump it. + The default channel window size used until now was 256KB. This value is + too small and results on a bottleneck on real-life networks where + round-trip delays can easily reach 300ms. - The define was formerly in the public header libssh2_sftp.h but served - no external purpose and was moved into the source dir. + The issue was not visible because the configured channel window size + was being ignored and a hard-coded value of ~22MB being used instead, + but that was fixed on a previous commit. - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-11/0004.shtml - Reported by: Michael Harris + This patch just changes the default window size + (LIBSSH2_CHANNEL_WINDOW_DEFAULT) to 2MB. It is the same value used by + OpenSSH and in our opinion represents a good compromise between memory + used and transfer speed. + + Performance tests were run to determine the optimum value. The details + and related discussion are available from the following thread on the + libssh2 mailing-list: + + http://www.libssh2.org/mail/libssh2-devel-archive-2013-10/0018.shtml + http://article.gmane.org/gmane.network.ssh.libssh2.devel/6543 + + An excerpt follows: + + "I have been running some transfer test and measuring their speed. + + My setup was composed of a quad-core Linux machine running Ubuntu 13.10 + x86_64 with a LXC container inside. The data transfers were performed + from the container to the host (never crossing through a physical + network device). + + Network delays were simulated using the tc tool. And ping was used to + verify that they worked as intended during the tests. + + The operation performed was the equivalent to the following ssh command: + + $ ssh container "dd bs=16K count=8K if=/dev/zero" >/dev/null + + Though, establishment and closing of the SSH connection was excluded + from the timings. + + I run the tests several times transferring files of sizes up to 128MB + and the results were consistent between runs. + + The results corresponding to the 128MB transfer are available here: + + https://docs.google.com/spreadsheet/ccc?key=0Ao1yRmX6PQQzdG5wSFlrZl9HRWNET3ZyN0hnaGo5ZFE&usp=sharing + + It clearly shows that 256KB is too small as the default window size. + Moving to a 512MB generates a great improvement and after the 1MB mark + the returns rapidly diminish. Other factors (TCP window size, probably) + become more limiting than the channel window size + + For comparison I also performed the same transfers using OpenSSH. Its + speed is usually on par with that of libssh2 using a window size of 1MB + (even if it uses a 2MB window, maybe it is less aggressive sending the + window adjust msgs)." + + Signed-off-by: Salvador Fandino -Alexander Lamaison (18 Nov 2011) -- [Peter Krempa brought this change] +- [Salvador brought this change] - knownhost_check(): Don't dereference ext if NULL is passed + _libssh2_channel_read: Honour window_size_initial - Documentation for libssh2_knownhost_checkp() and related functions - states that the last argument is filled with data if non-NULL. + _libssh2_channel_read was using an arbitrary hard-coded limit to trigger + the window adjusting code. The adjustment used was also hard-coded and + arbitrary, 15MB actually, which would limit the usability of libssh2 on + systems with little RAM. - "knownhost if set to non-NULL, it must be a pointer to a 'struct - libssh2_knownhost' pointer that gets filled in to point to info about a - known host that matches or partially matches." + This patch, uses the window_size parameter passed to + libssh2_channel_open_ex (stored as remote.window_size_initial) plus the + buflen as the base for the trigger and the adjustment calculation. - In this function ext is dereferenced even if set to NULL, causing - segfault in applications not needing the extra data. + The memory usage when using the default window size is reduced from 22MB + to 256KB per channel (actually, if compression is used, these numbers + should be incremented by ~50% to account for the errors between the + decompressed packet sizes and the predicted sizes). + + My tests indicate that this change does not impact the performance of + transfers across localhost or a LAN, being it on par with that of + OpenSSH. On the other hand, it will probably slow down transfers on + networks with high bandwidth*delay when the default window size + (LIBSSH2_CHANNEL_WINDOW_DEFAULT=256KB) is used. + + Signed-off-by: Salvador Fandino -Daniel Stenberg (11 Nov 2011) -- [Peter Krempa brought this change] +- [Salvador Fandino brought this change] - knownhost_add: Avoid dereferencing uninitialized memory on error path. + knownhosts: handle unknown key types - In function knownhost_add, memory is alocated for a new entry. If normal - alocation is used, memory is not initialized to 0 right after, but a - check is done to verify if correct key type is passed. This test is done - BEFORE setting the memory to null, and on the error path function - free_host() is called, that tries to dereference unititialized memory, - resulting into a glibc abort(). + Store but don't use keys of unsupported types on the known_hosts file. - * knownhost.c - knownhost_add(): - move typemask check before alloc + Currently, when libssh2 parses a known_host file containing keys of some + type it doesn't natively support, it stops reading the file and returns + an error. + + That means, that the known_host file can not be safely shared with other + software supporting other key types (i.e. OpenSSH). + + This patch adds support for handling keys of unknown type. It can read + and write them, even if they are never going to be matched. + + At the source level the patch does the following things: + + - add a new unknown key type LIBSSH2_KNOWNHOST_KEY_UNKNOWN + + - add a new slot (key_type_name) on the known_host struct that is + used to store the key type in ascii form when it is not supported + + - parse correctly known_hosts entries with unknown key types and + populate the key_type_name slot + + - print correctly known_hosts entries of unknown type + + - when checking a host key ignore keys that do not match the key + + Fixes #276 -- windows build: add define to avoid compiler warning +- windows build: fix build errors - A recent mingw compiler has started to complain on "#warning Please - include winsock2.h before windows.h" unless the magic define is set - first. + Fixes various link errors with VS2010 - Reported by: Vincent Torri - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-10/0064.shtml + Reported-by: "kdekker" + Fixes #272 -Henrik Nordstrom (31 Oct 2011) -- [Vincent Torri brought this change] +- man page: add missing function argument + + for libssh2_userauth_publickey_fromfile_ex() + + Reported-by: "pastey" + + Fixes #262 - Correct Windows include file name case, simplifying cross-compilation +- [Salvador brought this change] + + Fix zlib deflate usage - When cross compiling to Windows, libssh2.h include Windows header files - with upper case filenames : BaseTsd.h and WinSock2.h. + Deflate may return Z_OK even when not all data has been compressed + if the output buffer becomes full. - These files have lowercase names with mingw-w64 (iirc, it's the same with - mingw). And as on Windows, being lowercase or uppercase does not matter. + In practice this is very unlikely to happen because the output buffer + size is always some KBs larger than the size of the data passed for + compression from the upper layers and I think that zlib never expands + the data so much, even on the worst cases. + + Anyway, this patch plays on the safe side checking that the output + buffer is not exhausted. + + Signed-off-by: Salvador -Daniel Stenberg (25 Oct 2011) -- [Jernej Kovacic brought this change] +- [Salvador brought this change] - libssh2_session_supported_algs: added + comp_method_zlib_decomp: Improve buffer growing algorithm + + The old algorithm was O(N^2), causing lots and lots of reallocations + when highly compressed data was transferred. + + This patch implements a simpler one that just doubles the buffer size + everytime it is exhausted. It results in O(N) complexity. + + Also a smaller inflate ratio is used to calculate the initial size (x4). + + Signed-off-by: Salvador -- [Kamil Dudka brought this change] +- [Salvador brought this change] - example/sftp_RW_nonblock: do not ignore LIBSSH2_ERROR_EAGAIN + Fix zlib usage - Bug: https://bugzilla.redhat.com/745420 + Data may remain in zlib internal buffers when inflate() returns Z_OK + and avail_out == 0. In that case, inflate has to be called again. + + Also, once all the data has been inflated, it returns Z_BUF_ERROR to + signal that the input buffer has been exhausted. + + Until now, the way to detect that a packet payload had been completely + decompressed was to check that no data remained on the input buffer + but that didn't account for the case where data remained on the internal + zlib buffers. + + That resulted in packets not being completely decompressed and the + missing data reappearing on the next packet, though the bug was masked + by the buffer allocation algorithm most of the time and only manifested + when transferring highly compressible data. + + This patch fixes the zlib usage. + + Signed-off-by: Salvador -Peter Stuge (5 Oct 2011) -- example/ssh2_agent: Print host key fingerprint before authentication +- [Salvador brought this change] + + _libssh2_channel_read: fix data drop when out of window - Also moves the comment about not being authenticated to before the - agent authentication takes place, so that it better matches the code. + After filling the read buffer with data from the read queue, when the + window size was too small, "libssh2_channel_receive_window_adjust" was + called to increase it. In non-blocking mode that function could return + EAGAIN and, in that case, the EAGAIN was propagated upwards and the data + already read on the buffer lost. + + The function was also moving between the two read states + "libssh2_NB_state_idle" and "libssh2_NB_state_created" both of which + behave in the same way (excepting a debug statment). + + This commit modifies "_libssh2_channel_read" so that the + "libssh2_channel_receive_window_adjust" call is performed first (when + required) and if everything goes well, then it reads the data from the + queued packets into the read buffer. + + It also removes the useless "libssh2_NB_state_created" read state. + + Some rotted comments have also been updated. + + Signed-off-by: Salvador -Daniel Stenberg (29 Sep 2011) -- OpenSSL EVP: fix threaded use of structs +- [Salvador Fandino brought this change] + + window_size: redid window handling for flow control reasons - Make sure we don't clear or reset static structs after first init so - that they work fine even when used from multiple threads. Init the - structs in the global init. + Until now, the window size (channel->remote.window_size) was being + updated just after receiving the packet from the transport layer. - Help and assistance by: John Engstrom + That behaviour is wrong because the channel queue may grow uncontrolled + when data arrives from the network faster that the upper layer consumes + it. - Fixes #229 (again) + This patch adds a new counter, read_avail, which keeps a count of the + bytes available from the packet queue for reading. Also, now the window + size is adjusted when the data is actually read by an upper layer. + + That way, if the upper layer stops reading data, the window will + eventually fill and the remote host will stop sending data. When the + upper layers reads enough data, a window adjust packet is delivered and + the transfer resumes. + + The read_avail counter is used to detect the situation when the remote + server tries to send data surpassing the window size. In that case, the + extra data is discarded. + + Signed-off-by: Salvador -- openssl: don't init static structs differently +Peter Stuge (15 Sep 2013) +- configure.ac: Call zlib zlib and not libz in text but keep option names + +- configure.ac: Reorder --with-* options in --help output + +- configure.ac: Rework crypto library detection - make_ctr_evp() is changed to take a struct pointer, and then each - _libssh2_EVP_aes_[keylen]_ctr function is made to pass in their own - static struct + This further simplifies adding new crypto libraries. + +- Clean up crypto library abstraction in build system and source code - Reported by: John Engstrom - Fixes #229 + libssh2 used to explicitly check for libgcrypt and default to OpenSSL. + + Now all possible crypto libraries are checked for explicitly, making + the addition of further crypto libraries both simpler and cleaner. -Guenter Knauf (27 Sep 2011) -- Removed obsolete include path. +- configure.ac: Add zlib to Requires.private in libssh2.pc if using zlib -Daniel Stenberg (21 Sep 2011) -- read_state: clear the state variable better +- Revert "Added Windows Cryptography API: Next Generation based backend" - Set read_state back to idle before trying to send anything so that if - the state somehow is wrongly set. + This reverts commit d385230e15715e67796f16f3e65fd899f21a638b. + +Daniel Stenberg (7 Sep 2013) +- [Leif Salomonsson brought this change] + + sftp_statvfs: fix for servers not supporting statfvs extension - Also, avoid such a case of confusion by resetting the read_state when an - sftp handle is closed. + Fixes issue arising when server does not support statfvs and or fstatvfs + extensions. sftp_statvfs() and sftp_fstatvfs() after this patch will + handle the case when SSH_FXP_STATUS is returned from server. -- sftp_read: remove leftover fprintf +- [Marc Hoersken brought this change] + + Added Windows Cryptography API: Next Generation based backend + +- [Kamil Dudka brought this change] + + partially revert "window_size: explicit adjustments only" - Reported by: Alexander Lamaison + This partially reverts commit 03ca9020756a4e16f0294e5b35e9826ee6af2364 + in order to fix extreme slowdown when uploading to localhost via SFTP. + + I was able to repeat the issue on RHEL-7 on localhost only. It did not + occur when uploading via network and it did not occur on a RHEL-6 box + with the same version of libssh2. + + The problem was that sftp_read() used a read-ahead logic to figure out + the window_size, but sftp_packet_read() called indirectly from + sftp_write() did not use any read-ahead logic. -- sftp.h: fix the #ifdef to prevent multiple inclusions +- _libssh2_channel_write: client spins on write when window full + + When there's no window to "write to", there's no point in waiting for + the socket to become writable since it most likely just will continue to + be. + + Patch-by: ncm + Fixes #258 -- sftp_read: use a state variable to avoid bad writes +- _libssh2_channel_forward_cancel: avoid memory leaks on error - When a channel_write call has gotten an EAGAIN back, we try harder to - continue the same write in the subsequent invoke. + Fixes #257 -- window_size: explicit adjustments only +- _libssh2_packet_add: avoid using uninitialized memory - Removed the automatic window_size adjustments from - _libssh2_channel_read() and instead all channel readers must now make - sure to enlarge the window sizes properly themselves. + In _libssh2_packet_add, called by _libssh2_packet_read, a call to + _libssh2_packet_send that is supposed to send a one-byte message + SSH_MSG_REQUEST_FAILURE would send an uninitialized byte upon re-entry + if its call to _send returns _EAGAIN. - libssh2_channel_read_ex() - the public function, now grows the window - size according to the requested buffer size. Applications can still opt - to grow the window more on demand. Larger windows tend to give higher - performance. + Fixes #259 + +- _libssh2_channel_forward_cancel: accessed struct after free - sftp_read() now uses the read-ahead logic to figure out a window_size. + ... and the assignment was pointless anyway since the struct was about + to be freed. Bug introduced in dde2b094. + + Fixes #268 -- libssh2.h: bump the default window size to 256K +Peter Stuge (2 Jun 2013) +- [Marc Hoersken brought this change] -- libssh2_userauth_keyboard_interactive.3: fix man warning + Fixed compilation using mingw-w64 + +- [Marc Hoersken brought this change] + + knownhost.c: use LIBSSH2_FREE macro instead of free + + Use LIBSSH2_FREE instead of free since + _libssh2_base64_encode uses LIBSSH2_ALLOC + +Daniel Stenberg (18 May 2013) +- [Matthias Kerestesch brought this change] + + libssh2_agent_init: init ->fd to LIBSSH2_INVALID_SOCKET + + ... previously it was left at 0 which is a valid file descriptor! + + Bug: https://trac.libssh2.org/ticket/265 + + Fixes #265 + +- userauth_password: pass on the underlying error code + + _libssh2_packet_requirev() may return different errors and we pass that + to the parent instead of rewriting it. + + Bug: http://libssh2.org/mail/libssh2-devel-archive-2013-04/0029.shtml + Reported by: Cosmin + +Peter Stuge (9 May 2013) +- [Marc Hoersken brought this change] + + libcrypt.c: Fix typo in _libssh2_rsa_sha1_sign() parameter type + +Kamil Dudka (4 May 2013) +- configure.ac: replace AM_CONFIG_HEADER with AC_CONFIG_HEADERS + + Reported by: Quintus + Bug: https://trac.libssh2.org/ticket/261 + +Guenter Knauf (12 Apr 2013) +- Fixed copyright string for NetWare build. + +Daniel Stenberg (9 Apr 2013) +- [Richard W.M. Jones brought this change] + + sftp: Add support for fsync (OpenSSH extension). + + The new libssh2_sftp_fsync API causes data and metadata in the + currently open file to be committed to disk at the server. - It seemed to occur due to the excessive line length + This is an OpenSSH extension to the SFTP protocol. See: + + https://bugzilla.mindrot.org/show_bug.cgi?id=1798 -- [Mikhail Gusarov brought this change] +- [Richard W.M. Jones brought this change] - Add missing .gitignore entries + sftp: statvfs: Along error path, reset the correct 'state' variable. -- [Mikhail Gusarov brought this change] +- [Richard W.M. Jones brought this change] - Add manpage syntax checker to 'check' target + sftp: seek: Don't flush buffers on same offset - In virtually every libssh2 release Debian's lintian catches syntax errors in - manpages. Prevent it by checking manpages as a part of testsuite. + Signed-off-by: Richard W.M. Jones -- libssh2_banner_set.3: fix nroff syntax mistake +Guenter Knauf (9 Feb 2013) +- Updated dependency libs. -Guenter Knauf (10 Sep 2011) -- Use predefined resource compiler macro. +- Fixed tool macro names. -- Added casts to silent compiler warnings. +Daniel Stenberg (29 Nov 2012) +- [Seth Willits brought this change] -- Fixed uint64_t printf. + compiler warnings: typecast strlen in macros + + ... in macro parameters to avoid compiler warnings about lost precision. + + Several macros in libssh2.h call strlen and pass the result directly to + unsigned int parameters of other functions, which warns about precision + loss because strlen returns size_t which is unsigned long on at least + some platforms (such as OS X). The fix is to simply typecast the + strlen() result to unsigned int. -- Fixed macro function signatures. +- libssh2.h: bump version to 1.4.4-DEV -- NetWare makefile tweaks. +Version 1.4.3 (27 Nov 2012) -- Removed unused var. +Daniel Stenberg (27 Nov 2012) +- RELEASE-NOTES: fixed for 1.4.3 -- Added 2 samples not mentioned. +- sftp_read: return error if a too large package arrives -- Dont build x11 sample with MinGW. +Peter Stuge (13 Nov 2012) +- Only define _libssh2_dsa_*() functions when building with DSA support -- Fixed executable file description. +Guenter Knauf (8 Nov 2012) +- Added .def file to output. -- Removed unused var. +Kamil Dudka (1 Nov 2012) +- libssh2_hostkey_hash.3: update the description of return value + + The function returns NULL also if the hash algorithm is not available. -- Kill stupid gcc 3.x uninitialized warning. +Guenter Knauf (24 Oct 2012) +- Fixed mode acciedently committed. -- Build all examples. +- Ignore generated file. -- More MinGW makefile tweaks. +- Added hack to make use of Makefile.inc. - Renamed *.mingw makefiles to GNUmakefile since GNU make picks these - up automatically, and therefore win32/Makefile removed. + This should avoid further maintainance of the objects list. -- Removed forgotten WINSOCK_VERSION defines. +- Fixed MSVC NMakefile. + + Added missing source files; added resource for DLL. -Daniel Stenberg (9 Sep 2011) -- libssh2_session_startup(3) => libssh2_session_handshake(3) +Kamil Dudka (22 Oct 2012) +- examples: use stderr for messages, stdout for data - Propagate for the current function in docs and examples. - libssh2_session_startup() is deprecated. + Reported by: Karel Srot + Bug: https://bugzilla.redhat.com/867462 -- libssh2_banner_set => libssh2_session_banner_get +- openssl: do not leak memory when handling errors - Marked the old function as deprecated. Added the new name in the correct - name space with the same arguments and functionality. + ,.. in aes_ctr_init(). Detected by Coverity. -- new function: libssh2_session_banner_get +- channel: fix possible NULL dereference - Returns the banner from the server handshake + ... in libssh2_channel_get_exit_signal(). Detected by Coverity. + +- Revert "aes: the init function fails when OpenSSL has AES support" - Fixes #226 + This partially reverts commit f4f2298ef3635acd031cc2ee0e71026cdcda5864. + + We need to use the EVP_aes_???_ctr() functions in FIPS mode. -- libssh2.h: bump version to 1.4.0 for new function(s) +- crypt: use hard-wired cipher block sizes consistently -- remove embedded CVS/svn tags +- openssl: do not ignore failure of EVP_CipherInit() -- [liuzl brought this change] +- kex: do not ignore failure of libssh2_md5_init() + + The MD5 algorithm is disabled when running in FIPS mode. - API add:libssh2_sftp_get_channel +Daniel Stenberg (21 Aug 2012) +- [Peter Krempa brought this change] + + known_hosts: Fail when parsing unknown keys in known_hosts file. - Return the channel of sftp, then caller can - control the channel's behavior. + libssh2_knownhost_readfile() silently ignored problems when reading keys + in unsupported formats from the known hosts file. When the file is + written again from the internal structures of libssh2 it gets truntcated + to the point where the first unknown key was located. - Signed-off-by: liuzl + * src/knownhost.c:libssh2_knownhost_readfile() - return error if key + parsing fails -- _libssh2_channel_read: react on errors from receive_window_adjust +- AUTHORS: synced with 42fec44c8a4 - Previously the function would ignore all errors except for EAGAIN. + 31 recent authors added -- sftp_read: extend and clarify the documentation +- [Dave Hayden brought this change] -- sftp_read: cap the read ahead maximum amount + compression: add support for zlib@openssh.com - Now we only go up to LIBSSH2_CHANNEL_WINDOW_DEFAULT*30 bytes SFTP read - ahead, which currently equals 64K*30 == 1966080 bytes. + Add a "use_in_auth" flag to the LIBSSH2_COMP_METHOD struct and a + separate "zlib@openssh.com" method, along with checking session->state + for LIBSSH2_STATE_AUTHENTICATED. Appears to work on the OpenSSH servers + I've tried against, and it should work as before with normal zlib + compression. -- _libssh2_channel_read: fix non-blocking window adjusting +- [Dmitry Smirnov brought this change] + + configure: gcrypt doesn't come with pkg-config support - If EAGAIN is returned when adjusting the receive window, we must not - read from the transport directly until we've finished the adjusting. + ... so use plain old -lgcrypt to the linker to link with it. + + Fixes #225 -Guenter Knauf (8 Sep 2011) -- Fix for systems which need sys/select.h. +- sftp_read: Value stored to 'next' is never read + + Detected by clang-analyzer -- The files were not gone but renamed ... +- publickey_init: errors are negative, fix check + + Detected by clang-analyzer. -Daniel Stenberg (6 Sep 2011) -- sftp_read: added documenting comment +- [Maxime Larocque brought this change] + + session_free: wrong variable used for keeping state - Taken from some recent email conversations I added some descriptions of - the logic in sftp_read() to aid readers. + If libssh2_session_free is called without the channel being freed + previously by libssh2_channel_free a memory leak could occur. + + A mismatch of states variables in session_free() prevent the call to + libssh2_channel_free function. session->state member is used instead of + session->free_state. + + It causes a leak of around 600 bytes on every connection on my systems + (Linux, x64 and PPC). + + (Debugging done under contract for Accedian Networks) + + Fixes #246 -- 1.3.1: start the work +Guenter Knauf (29 Jun 2012) +- Small NetWare makefile tweak. -Version 1.3.0 (6 Sep 2011) +- Some small Win32 makefile fixes. -Daniel Stenberg (6 Sep 2011) -- Makefile.am: the Makefile.win32 files are gone +Daniel Stenberg (19 Jun 2012) +- libssh2_userauth_publickey_fromfile_ex.3: mention publickey == NULL -- RELEASE-NOTES: updated for 1.3.0 +- comp_method_zlib_decomp: handle Z_BUF_ERROR when inflating + + When using libssh2 to perform an SFTP file transfer from the "JSCAPE MFT + Server" (http://www.jscape.com) the transfer failed. The default JSCAPE + configuration is to enforce zlib compression on SSH2 sessions so the + session was compressed. The relevant part of the debug trace contained: + + [libssh2] 1.052750 Transport: unhandled zlib error -5 + [libssh2] 1.052750 Failure Event: -29 - decompression failure + + The trace comes from comp_method_zlib_decomp() in comp.c. The "unhandled + zlib error -5" is the status returned from the zlib function + inflate(). The -5 status corresponds to "Z_BUF_ERROR". + + The inflate() function takes a pointer to a z_stream structure and + "inflates" (decompresses) as much as it can. The relevant fields of the + z_stream structure are: + + next_in - pointer to the input buffer containing compressed data + avail_in - the number of bytes available at next_in + next_out - pointer to the output buffer to be filled with uncompressed + data + avail_out - how much space available at next_out + + To decompress data you set up a z_stream struct with the relevant fields + filled in and pass it to inflate(). On return the fields will have been + updated so next_in and avail_in show how much compressed data is yet to + be processed and next_out and avail_out show how much space is left in + the output buffer. + + If the supplied output buffer is too small then on return there will be + compressed data yet to be processed (avail_in != 0) and inflate() will + return Z_OK. In this case the output buffer must be grown, avail_out + updated and inflate() called again. + + If the supplied output buffer was big enough then on return the + compressed data will have been exhausted (avail_in == 0) and inflate() + will return Z_OK, so the data has all been uncompressed. + + There is a corner case where inflate() makes no progress. That is, there + may be unprocessed compressed data and space available in the output + buffer and yet the function does nothing. In this case inflate() will + return Z_BUF_ERROR. From the zlib documentation and the source code it + is not clear under what circumstances this happens. It could be that it + needs to write multiple bytes (all in one go) from its internal state to + the output buffer before processing the next chunk of input but but + can't because there is not enough space (though my guesses as to the + cause are not really relevant). Recovery from Z_BUF_ERROR is pretty + simple - just grow the output buffer, update avail_out and call + inflate() again. + + The comp_method_zlib_decomp() function does not handle the case when + inflate() returns Z_BUF_ERROR. It treats it as a non-recoverable error + and basically aborts the session. + + Fixes #240 + +Guenter Knauf (12 Jun 2012) +- MinGW makefile tweaks. + + Use GNU tools when compiling on Linux. + Fixed dist and dev targets. + +- NetWare makefile tweaks. + + Changed to use Windows commandline tools instead of + GNU tools when compiling on Windows. Fixed dist and + dev targets. Enabled nlmconv error for unresolved + symbols. + +Daniel Stenberg (11 Jun 2012) +- Revert "config.rpath: generated file, no need to keep in git" + + This reverts commit 1ac7bd09cc685755577fb2c8829adcd081e7ab3c. + + This file still used by lib/*m4 functions so we need to keep the file + around. + +- BINDINGS: added PySsh2, a Python-ctypes binding + +Guenter Knauf (8 Jun 2012) +- Fixed MinGW debug build. -- sftp_read: a short read is not end of file +Daniel Stenberg (5 Jun 2012) +- BINDINGS: Added the Cocoa/Objective-C one - A returned READ packet that is short will now only reduce the - offset. + ... and sorted the bindings after the languages, alphabetically - This is a temporary fix as it is slightly better than the previous - approach but still not very good. + Reported by: Mike Abdullah -- [liuzl brought this change] +- BINDINGS: document the bindings we know of - _libssh2_packet_add: adjust window size when truncating +Guenter Knauf (4 Jun 2012) +- Fixed LIBSSH2_INT64_T_FORMAT macro. - When receiving more data than what the window size allows on a - particular channel, make sure that the window size is adjusted in that - case too. Previously it would only adjust the window in the non-error - case. + Usually a format macro should hold the whole format, otherwise + it should be named a prefix. Also fixed usage of this macro in + scp.c for a signed var where it was used as prefix for unsigned. -Guenter Knauf (29 Aug 2011) -- Silent compiler warning with MinGW64. +- Removed obsolete define from makefiles. -- Fixed link to native Win32 awk tool. +- Renamed NetWare makefiles. -- Renamed MinGW makefiles. +- Renamed NetWare makefiles. -- Some MinGW makefile tweaks. +- Synced MinGW makefiles with 56c64a6..39e438f. - Enable build without GNU tools and with MinGW64 compiler. - -- Fixed aes_ctr_do_cipher() signature. - -Daniel Stenberg (26 Aug 2011) -- [liuzl brought this change] + Also synced MinGW test makefile with b092696..f8cb874. - libssh2_sftp_seek64: flush packetlist and buffered data +Peter Stuge (30 May 2012) +- Revert "sftp: Don't send attrs.permissions on read-only SSH_FXP_OPEN" - When seeking to a new position, flush the packetlist and buffered data - to prevent already received or pending data to wrongly get used when - sftp-reading from the new offset within the file. + This reverts commit 04e79e0c798674a0796be8a55f63dd92e6877790. -- sftp_read: advance offset correctly for buffered copies - - In the case where a read packet has been received from the server, but - the entire contents couldn't be copied to the user-buffer, the data is - instead buffered and copied to the user's buffer in the next invocation - of sftp_read(). When that "extra" copy is made, the 'offset' pointer was - not advanced accordingly. +- sftp: Don't send attrs.permissions on read-only SSH_FXP_OPEN - The biggest impact of this flaw was that the 'already' variable at the - top of the function that figures out how much data "ahead" that has - already been asked for would slowly go more and more out of sync, which - could lead to the file not being read all the way to the end. + This works around a protocol violation in the ProFTPD 1.3.4 mod_sftp + server, as reported by Will Cosgrove in: - This problem was most noticable in cases where the application would - only try to read the exact file size amount, like curl does. In the - examples libssh2 provides the sftp read function is most often called - with a fixed size large buffer and then the bug would not appear as - easily. + http://libssh2.org/mail/libssh2-devel-archive-2012-05/0079.shtml - This bug was introduced in the SFTP rewrite in 1.2.8. + Based on a suggested fix by TJ Saunders in: - Bug: http://curl.haxx.se/mail/lib-2011-08/0305.html - http://www.libssh2.org/mail/libssh2-devel-archive-2011-08/0085.shtml - -- wrap some long lines < 80 columns - -- LIBSSH2_RECV: fix typo, use the RECV_FD macro - -- subsystem_netconf.c: fix compiler warnings - -- [Henrik Nordstrom brought this change] - - Custom callbacks for performing low level socket I/O - -- version bump: start working towards 1.3.0 + http://libssh2.org/mail/libssh2-devel-archive-2012-05/0104.shtml -Version 1.2.9 (16 Aug 2011) +Guenter Knauf (28 May 2012) +- Try to detect OpenSSL build type automatically. + + Also fixed recently added libgdi32 linkage which is only + required when OpenSSL libs are linked statically. -Daniel Stenberg (16 Aug 2011) -- RELEASE-NOTES: synced with 95d69d3a81261 +Daniel Stenberg (25 May 2012) +- config.rpath: generated file, no need to keep in git -- [Henrik Nordstrom brought this change] +Guenter Knauf (22 May 2012) +- Updated dependency libary versions. - Document prototypes for macro defined functions +Daniel Stenberg (18 May 2012) +- 1.4.3: towards the future -- [Henrik Nordstrom brought this change] +Version 1.4.2 (18 May 2012) - Avoid reuse after free when closing X11 channels +Daniel Stenberg (18 May 2012) +- RELEASE-NOTES: synced with 92a9f952794 -- _libssh2_channel_write: handle window_size == 0 better +Alexander Lamaison (15 May 2012) +- win32/libssh2_config.h: Remove hardcoded #define LIBSSH2_HAVE_ZLIB. - When about to send data on the channel and the window size is 0, we must - not just return 0 if the transport_read() function returned EAGAIN as it - then causes a busy-loop. + Rationale: Everything else in this file states a fact about the win32 + platform that is unconditional for that platform. There is nothing + unconditional about the presence of zlib. It is neither included with + Windows nor with the platform SDK. Therefore, this is not an appropriate + place to assert its presence. Especially as, once asserted, it cannot be + overridden using a compiler flag. - Bug: http://libssh2.org/mail/libssh2-devel-archive-2011-08/0011.shtml + In contrast, if it is omitted, then it can easily be reasserted by adding + a compiler flag defining LIBSSH2_HAVE_ZLIB. -- gettimeofday: fix name space pollution +Daniel Stenberg (14 May 2012) +- RELEASE-NOTES: synced with 69a3354467c + +- _libssh2_packet_add: SSH_MSG_CHANNEL_REQUEST default to want_reply - For systems without its own gettimeofday() implementation, we still must - not provide one outside our namespace. + RFC4254 says the default 'want_reply' is TRUE but the code defaulted to + FALSE. Now changed. - Reported by: Bill Segall - -Dan Fandrich (5 Aug 2011) -- libssh2.pc.in: Fixed spelling in pkgconfig file - -Peter Stuge (17 Jul 2011) -- example/subsystem_netconf.c: Add missing #include - -- example/subsystem_netconf.c: Discard ]]>]]> and return only XML response + Fixes #233 -- example/subsystem_netconf.c: Fix uninitialized variable bug +- gettimeofday: no need for a replacement under cygwin + + Fixes #224 -- example: Add subsystem_netconf.c +Alexander Lamaison (13 May 2012) +- Prevent sftp_packet_read accessing freed memory. - This example demonstrates how to use libssh2 to send a request to - the NETCONF subsystem available e.g. in JunOS. + sftp_packet_add takes ownership of the packet passed to it and (now that we + handle zombies) might free the packet. sftp_packet_read uses the packet type + byte as its return code but by this point sftp_packet_add might have freed + it. This change fixes the problem by caching the packet type before calling + sftp_packet_add. - See also http://tools.ietf.org/html/draft-ietf-netconf-ssh-06 - -Daniel Stenberg (16 Jul 2011) -- man page cleanups: non-existing functions need no man pages + I don't understand why sftp_packet_read uses the packet type as its return + code. A future change might get rid of this entirely. -- libssh2_new_host_entry.3: removed +Daniel Stenberg (12 May 2012) +- sftp_packet_flush: flush zombies too - This is just junk leftovers. + As this function is called when the SFTP session is closed, it needs to + also kill all zombies left in the SFTP session to avoid leaking memory + just in case some zombie would still be in there. -- userauth_keyboard_interactive: fix buffer overflow - - Partly reverse 566894494b4972ae12 which was simplifying the code far too - much and ended up overflowing a buffer within the LIBSSH2_SESSION - struct. Back to allocating the buffer properly like it used to do. +- sftp_packetlist_flush: zombies must not have responses already - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-06/0032.shtml - Reported by: Alfred Gebert - -- keyboard-interactive man page: cleaned up + When flushing the packetlist, we must only add the request as a zombie + if no response has already been received. Otherwise we could wrongly + make it a zombie even though the response was already received and then + we'd get a zombie stuck there "forever"... -- [Alfred Gebert brought this change] +- sftp_read: on EOF remove packet before flush + + Since the sftp_packetlist_flush() function will move all the existing + FXP_READ requests in this handle to the zombie list we must first remove + this just received packet as it is clearly not a zombie. - _libssh2_recv(): handle ENOENT error as EAGAIN +- sftp_packet_require: sftp_packet_read() returning 0 is not an error - A sftp session failed with error "failure establishing ssh session" on - Solaris and HP-UX. Sometimes the first recv() function call sets errno - to ENOENT. In the man pages for recv of Solaris and HP-UX the error - ENOENT is not documented. + Exactly as the comment in the code said, checking the return code from + sftp_packet_read() with <= was wrong and it should be < 0. With the new + filtering on incoming packets that are "zombies" we can now see this + getting zero returned. + +- sftp_packetlist_flush: only make it zombie if it was sent - I tested Solaris SPARC and x86, HP-UX i64, AIX, Windows and Linux. + The list of outgoing packets may also contain packets that never were + sent off and we better not make them zombies too. -- agent_list_identities: fix out of scope access +- [Alexander Lamaison brought this change] + + Mark outstanding read requests after EOF as zombies. - An auto variable out of scope was being referenced and used. + In order to be fast, sftp_read sends many read requests at once. With a small + file, this can mean that when EOF is received back, many of these requests are + still outstanding. Responses arriving after we close the file and abandon the + file handle are queued in the SFTP packet queue and never collected. This + causes transfer speed to drop as a progressively longer queue must be searched + for every packet. - fixes #220 + This change introduces a zombie request-ID list in the SFTP session that is + used to recognise these outstanding requests and prevent them being added to + the queue. -- _libssh2_wait_socket: fix timeouts for poll() uses +Peter Stuge (23 Apr 2012) +- [Rafael Kitover brought this change] -- windows: inclusion fix + Update win32/GNUmakefile to use OpenSSL 1.0.1a - include winsock2.h for all windows compilers + libcrypto on win32 now depends on gdi32.dll, so move the OpenSSL LDLIBS + block to before the compiler definitions, so that libcrypto gets added + first, and then add -lgdi32 into the following common LDLIBS for gcc. -- keyb-interactive: add the fixed buffer +Guenter Knauf (23 Apr 2012) +- Changed 'Requires' to 'Requires.private'. - Belongs to commit 5668944 + Only static builds need to link against the crypto libs. -- code cleanup: don't use C99/c++ comments +- Fixed 'Requires:' names. - We aim for C89 compliance + The 'Requires:' line lists the names of the .pc files. -- keyb-interactive: allow zero length fields +- Added 'Requires:' line to libssh2.pc. - Allow zero length fields so they don't cause malloc(0) calls + This is necessary so that other libs which lookup libssh2 info + via pkg-config can add the right crypto lib dependencies. + +- Updated dependency lib versions. + +Peter Stuge (18 Apr 2012) +- configure.ac: Add option to disable build of the example applications - Avoid free()ing NULL pointers + Examples are built by default. Any of the following options on the + configure command line will skip building them: - Avoid a malloc of a fixed 5 byte buffer. + --disable-examples-build + --enable-examples-build=no + --enable-examples-build=false -- libssh2_channel_process_startup.3: clean up +- userauth.c: fread() from public key file to correctly detect any errors + + If the filename parameter for file_read_publickey() was the name of a + directory instead of a file then libssh2 would spin trying to fgetc() + from the FILE * for the opened directory when trying to determine the + length of the encoded public key, since fgetc() can't report errors. - Remove the references to the macro-fied shortcuts as they have their own - individual man pages. + Use fread() instead to correctly detect this error condition along + with many others. - Made the prototype different and more readable. - -- man page: fix .BR lines + This fixes the problem reported in + http://www.libssh2.org/mail/libssh2-devel-archive-2012-04/0021.shtml - We don't use \fI etc on .BR lines + Reported-by: Oleksiy Zagorskyi -- userauth_keyboard_interactive: skip code on zero length auth +- Return LIBSSH2_ERROR_SOCKET_DISCONNECT on EOF when reading banner -- libssh2_channel_forward_accept.3: mention how to get error - - Since this returns a pointer, libssh2_session_last_errno() must be used - to get the actual error code and it wasn't that clear before. +Guenter Knauf (17 Apr 2012) +- Fixed copyright year. -- timeout docs: mention they're added in 1.2.9 +- Updated dependency lib versions in static makefiles. -- sftp_write_sliding.c: indent fix +Daniel Stenberg (6 Apr 2012) +- version: bump to 1.4.2 - Use the standard indenting and removed CVS leftover comment + We're on the 1.4.2 track now (at least) -- [zl liu brought this change] +Version 1.4.1 (4 Apr 2012) - sftp_write_sliding: send the complete file - - When reaching the end of file there can still be data left not sent. +Daniel Stenberg (4 Apr 2012) +- RELEASE-NOTES: updated for 1.4.1 release -- [Douglas Masterson brought this change] +- always do "forced" window updates + + When calling _libssh2_channel_receive_window_adjust() internally, we now + always use the 'force' option to prevent libssh2 to avoid sending the + update if the update isn't big enough. + + It isn't fully analyzed but we have seen corner cases which made a + necessary window update not get send due to this and then the other side + doesn't send data our side then sits waiting for forever. - session_startup: init state properly +- channel_read: force window adjusts! - libssh2_session_startup() didn't set the state correctly so it could get - confused. + if there's not enough room to receive the data that's being requested, + the window adjustment needs to be sent to the remote and thus the force + option has to be used. _libssh2_channel_receive_window_adjust() would + otherwise "queue" small window adjustments for a later packet but that + is really terribly for the small buffer read that for example is the + final little piece of a very large file as then there is no logical next + packet! - Fixes #218 + Reported by: Armen Babakhanian + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0130.shtml -- timeout: added man pages +- [Paul Howarth brought this change] -- BLOCK_ADJUST_ERRNO: move rc to right level + aes: the init function fails when OpenSSL has AES support - We can't declare the variable within the block and use it in the final - do-while() expression to be properly portable C89. + The internal init function only worked fine when the configure script + didn't detect the OpenSSL AES_CTR function! + + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0111.shtml + Reported by: Paul Howarth -- [Matt Lilley brought this change] +- [Matthew Booth brought this change] - adds a timeout to blocking calls + transport_send: Finish in-progress key exchange before sending data - Fixes bug #160 as per Daniel's suggestion + _libssh2_channel_write() first reads outstanding packets before writing + new data. If it reads a key exchange request, it will immediately start + key re-exchange, which will require sending a response. If the output + socket is full, this will result in a return from + _libssh2_transport_read() of LIBSSH2_ERROR_EAGAIN. In order not to block + a write because there is no data to read, this error is explicitly + ignored and the code continues marshalling a packet for sending. When it + is sent, the remote end immediately drops the connection because it was + expecting a continuation of the key exchange, but got a data packet. - Adds libssh2_session_set_timeout() and libssh2_session_get_timeout() + This change adds the same check for key exchange to + _libssh2_transport_send() that is in _libssh2_transport_read(). This + ensures that key exchange is completed before any data packet is sent. -- SCP: fix incorrect error code - - After an error occurs in libssh2_scp_recv() or libssh2_scp_send(), the - function libssh2_session_last_error() would return - LIBSSH2_ERROR_SOCKET_NONE on error. +- channel_write: acknowledge transport errors - Bug: http://trac.libssh2.org/ticket/216 - Patch by: "littlesavage" + When draining data off the socket with _libssh2_transport_read() (which + in turn has to be done so that we can be sure to have read any possible + window-increasing packets), this code previously ignored errors which + could lead to nasty loops. Now all error codes except EAGAIN will cause + the error to be returned at once. - Fixes #216 + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0068.shtml + Reported by: Matthew Booth -Guenter Knauf (19 Apr 2011) -- Updated default (recommended) dependency versions. +- [Steven Dake brought this change] -Daniel Stenberg (17 Apr 2011) -- libssh2_session_block_directions: fix mistake - - The last LIBSSH2_SESSION_BLOCK_INBOUND should be - LIBSSH2_SESSION_BLOCK_OUTBOUND + In examples/x11.c, Make sure sizeof passed to read operation is correct - And I shortened the short description + sizeof(buf) expands to 8 or 4 (since its a pointer). This variable may + have been static in the past, leading to this error. - Reported by: "drswinghead" + Signed-off-by: Steven Dake -- msvcproj: added libs and debug stuff - - Added libraries needed to link whether using openssl dynamically or - statically +- [Steven Dake brought this change] + + Fix suspicious sizeof usage in examples/x11.c - Added LIBSSH2DEBUG define to debug versions to enable tracing + In the x11 example, sizeof(buf) = 8UL (on x86_64), when this should + probably represent the buffer size available. I am not sure how to + test that this change is actually correct, however. - URL: http://trac.libssh2.org/ticket/215 - Patch by: Mark Smith + Signed-off-by: Steven Dake -- sftp_write: clean offsets on error +- sftp_packet_read: follow-up fix for EAGAIN/window adjust - When an error has occurred on FXP_WRITE, we must make sure that the - offset, sent offset and acked counter are reset properly. + The commit in 7194a9bd7ba45 wasn't complete. This change makes sure + variables are initialized properly before used in the EAGAIN and window + adjust cases. -- example/.gitignore: ignore built binaries +- sftp_packet_add: use named error code instead of number -- sftp_write: flush the packetlist on error +- sftp_packet_add: verify the packet before accepting it - When an error occurs during write, flush the entire list of pending - outgoing SFTP packets. + In order to bail out as quickly as possible when things are wrong and + out of sync, make sure the SFTP message is one we understand. -- keepalive: add first basic man pages +- SFTP: preserve the original error code more - Someone on IRC pointed out that we don't have these documented so I - wrote up a first set based on the information in the wiki: - http://trac.libssh2.org/wiki/KeepAlive + Lots of places in the code translated the original error into the more + generic LIBSSH2_ERROR_SOCKET_TIMEOUT but this turns out to distort the + original error reason a lot and makes tracking down the real origin of a + problem really hard. This change makes the original error code be + preserved to a larger extent when return up to the parent function. -- scp_write_nonblock.c: remove pointless check +- sftp_packet_read: adjust window size as necessary - libssh2_channel_write() cannot return a value that is larger than the - input length value + Commit 03ca9020756 tried to simplify the window sizing logic but broke + SFTP readdir as there was no window sizing code left there so large + directory listings no longer worked. + + This change introduces window sizing logic to the sftp_packet_read() + function so that it now tells the remote about the local size having a + window size that suffice when it is about to ask for directory data. + + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2012-03/0069.shtml + Reported by: Eric -Mikhail Gusarov (9 Apr 2011) -- s/\.NF/.nf/ to fix wrong macro name caught by man --warnings +- [Steven Dake brought this change] -Daniel Stenberg (6 Apr 2011) -- version: bump to 1.2.9_dev + Tell C compiler we don't care about return code of libssh2_init - Also update the copyright year range to include 2011 + The call of libssh2_init returns a return code, but nothing could be done + within the _libssh2_init_if_needed execution path. + + Signed-off-by: Steven Dake -- configure: fix $VERSION +- [Steven Dake brought this change] + + Add comment indicating a resource leak is not really a resource leak - Stop using the $VERSION variable as it seems to be magically used by - autoconfig itself and thus gets set to the value set in AC_INIT() - without us wanting that. $LIBSSH2VER is now the libssh2 version as - detected. + While possibly obvious to those investigating the code, coverity complains + about this out of scope leak. - Reported by: Paul Howarth - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-04/0008.shtml + Signed-off-by: Steven Dake -- maketgz: use git2news.pl by the correct name +- [Steven Dake brought this change] -Version 1.2.8 (4 Apr 2011) + Use safer snprintf rather then sprintf in scp_send() + + Signed-off-by: Steven Dake -Daniel Stenberg (4 Apr 2011) -- RELEASE-NOTES: synced with fabf1a45ee +- [Steven Dake brought this change] -- NEWS: auto-generated from git + Use safer snprintf rather then sprintf in scp_recv() - Starting now, the NEWS file is generated from git using the git2news.pl - script. This makes it always accurate and up-to-date, even for daily - snapshots etc. - -- sftp_write: handle FXP_WRITE errors + While the buffer is indeed allocated to a safe length, better safe then sorry. - When an sftp server returns an error back on write, make sure the - function bails out and returns the proper error. + Signed-off-by: Steven Dake -- configure: stop using the deprecated AM_INIT_AUTOMAKE syntax +- [Steven Dake brought this change] -Alexander Lamaison (13 Mar 2011) -- Support unlimited number of host names in a single line of the known_hosts file. + use snprintf in knownhost_writeline() rather then sprintf - Previously the code assumed either a single host name or a hostname,ip-address pair. However, according to the spec [1], there can be any number of comma separated host names or IP addresses. + Although the function checks the length, if the code was in error, there + could potentially be a buffer overrun with the use of sprintf. Instead replace + with snprintf. - [1] http://www.openbsd.org/cgi-bin/man.cgi?query=sshd&sektion=8 + Signed-off-by: Steven Dake -Daniel Stenberg (26 Feb 2011) -- libssh2_knownhost_readfile.3: clarify return value - - This function returns the number of parsed hosts on success, not just - zero as previously documented. +- [Steven Dake brought this change] -Peter Stuge (26 Feb 2011) -- Don't save allocated packet size until it has actually been allocated + Add tracing to print packets left on session at libssh2_session_free - The allocated packet size is internal state which needs to match reality - in order to avoid problems. This commit fixes #211. - -Daniel Stenberg (21 Feb 2011) -- [Alfred Gebert brought this change] + Signed-off-by: Steven Dake - session_startup: manage server data before server identification - - Fix the bug that libssh2 could not connect if the sftp server - sends data before sending the version string. +Peter Stuge (2 Mar 2012) +- Define and use LIBSSH2_INVALID_SOCKET instead of INVALID_SOCKET - http://tools.ietf.org/html/rfc4253#section-4.2 + INVALID_SOCKET is a special value in Windows representing a + non-valid socket identifier. We were #defining this to -1 on + non-Windows platforms, causing unneccessary namespace pollution. + Let's have our own identifier instead. - "The server MAY send other lines of data before sending the version - string. Each line SHOULD be terminated by a Carriage Return and Line - Feed. Such lines MUST NOT begin with "SSH-", and SHOULD be encoded - in ISO-10646 UTF-8 [RFC3629] (language is not specified). Clients - MUST be able to process such lines." + Thanks to Matt Lawson for pointing this out. -- [Alfred Gebert brought this change] +- nw/Makefile.netware: Fix project name typo to avoid needless confusion - fullpacket: decompression only after init +- example/x11: Set raw terminal mode manually instead of with cfmakeraw() - The buffer for the decompression (remote.comp_abstract) is initialised - in time when it is needed. With this fix decompression is disabled when - the buffer (remote.comp_abstract) is not initialised. + OpenSolaris has no cfmakeraw() so to make the example more portable + we simply do the equivalent operations on struct termios ourselves. - Bug: http://trac.libssh2.org/ticket/200 + Thanks to Tom Weber for reporting this problem, and finding a solution. -- _libssh2_channel_read: store last error +Daniel Stenberg (17 Feb 2012) +- sftp_write: cannot return acked data *and* EAGAIN - When the transport layer returns EAGAIN this function didn't call - _libssh2_error() which made the last_error not get set. - -- sftp_write: clarified the comment header + Whenever we have acked data and is about to call a function that *MAY* + return EAGAIN we must return the number now and wait to get called + again. Our API only allows data *or* EAGAIN and we must never try to get + both. -- sftp_read: avoid wrapping counter to insanity - - As pointed out in bug #206, if a second invoke of libssh2_sftp_read() - would shrink the buffer size, libssh2 would go nuts and send out read - requests like crazy. This was due to an unsigned variable turning - "negative" by some wrong math, and that value would be the amount of - data attempt to pre-buffer! +Peter Stuge (13 Feb 2012) +- example/x11: Build only when sys/un.h is found by configure - Bug: http://trac.libssh2.org/ticket/206 + The example can't be built on systems without AF_UNIX sockets. -- sftp_packet_read: use 32bit variables for 32bit data +Daniel Stenberg (10 Feb 2012) +- [Alexander Lamaison brought this change] -- libssh2_sftp_stat_ex.3: cleaned up, extended + Simplified sftp_read. - Removed the macros from it as they have their own man pages. + Removed the total_read variable that originally must have tracked how + much data had been written to the buffer. With non-blocking reads, we + must return straight away once we have read data into the buffer so this + variable served not purpose. - Added the LIBSSH2_SFTP_ATTRIBUTES struct in here for easier reference. + I think it was still hanging around in case the initial processing of + 'leftover' data meant we wrote to the buffer but this case, like the + others, must return immediately. Now that it does, the last remaining + need for the variable is gone. -- sftp_readdir: return error if buffer is too small - - If asked to read data into a buffer and the buffer is too small to hold - the data, this function now returns an error instead of as previously - just copy as much as fits. +- [Alexander Lamaison brought this change] -- sftp_symlink: return error if receive buffer too small - - and clean up some variable type mismatches + Cleaned up sftp_read and added more explanation. - Discussion: http://www.libssh2.org/mail/libssh2-devel-archive-2011-01/0001.shtml + Replaced the gotos which were implementing the state machine with + a switch statement which makes the states more explicit. -- docs: clarify what happens with a too small buffer +- sftp_read: avoid data *and* EAGAIN - This flaw is subject to change, but I figured it might be valuable to - users of existing code to know how it works. + Whenever we have data and is about to call a function that *MAY* return + EAGAIN we must return the data now and wait to get called again. Our API + only allows data *or* EAGAIN and we must never try to get both. -- channel_request_pty_size: fix reqPTY_state - - The state variable isn't properly set so every other call to the - function fails! +Peter Stuge (2 Feb 2012) +- Add a tcpip-forward example which demonstrates remote port forwarding + +- libssh2.h: Add missing prototype for libssh2_session_banner_set() + +- example/subsystem_netconf.c: Return error when read buffer is too small - Bug: http://libssh2.org/mail/libssh2-devel-archive-2010-12/0096.shtml - Reported by: Steve Legg + Also remove a little redundancy in the read loop condition. -- data size: cleanup +- example/subsystem_netconf.c: Add a missing newline in an error message + +- Fix undefined reference to _libssh_error in libgcrypt backend - Fix 64bit warnings by using (s)size_t and dedicated uint32_t types more. + Commit 209de22299b4b58e582891dfba70f57e1e0492db introduced a function + call to a non-existing function, and since then the libgcrypt backend + has not been buildable. -- [Pierre Joye brought this change] +Version 1.4.0 (31 Jan 2012) - ssize_t: proper typedef with MSVC compilers +Daniel Stenberg (31 Jan 2012) +- RELEASE-NOTES: synced with 6bd584d29 for 1.4.0 + +- s/1.3.1/1.4.0 - As discussed on the mailing list, it was wrong for win64 and using the - VC-provided type is the safest approach instead of second- guessing - which one it should be. + We're bumping the minor number -Guenter Knauf (22 Dec 2010) -- Updated OpenSSL version. +- [Jernej Kovacic brought this change] -- Expanded tabs to spaces. + libssh2_session_supported_algs: fix compiler warning -Peter Stuge (21 Dec 2010) -- [Joey Degges brought this change] +- [Jernej Kovacic brought this change] - _libssh2_ntohu64: fix conversion from network bytes to uint64 + session_supported_algs docs: added an example + +- [Gellule Xg brought this change] + + sftp-seek: clear EOF flag - Cast individual bytes to uint64 to avoid overflow in arithmetic. + Set the EOF flag to False when calling seek64 to be able to get some + data back on a following read -Daniel Stenberg (20 Dec 2010) -- libssh2_userauth_list: language fix +- [Peter Krempa brought this change] + + userauth: Provide more informations if ssh pub key extraction fails - "faily" is not a good English word, and I also cleaned up some other minor - mistakes + If the function that extracts/computes the public key from a private key + fails the errors it reports were masked by the function calling it. This + patch modifies the key extraction function to return errors using + _libssh_error() function. The error messages are tweaked to contain + reference to the failed operaton in addition to the reason. + + * AUTHORS: - add my name + * libgcrypt.c: _libssh2_pub_priv_keyfile(): - return a more verbose + error using + _libssh2_error() func. + * openssl.c: - modify call graph of _libssh2_pub_priv_keyfile() to use + _libssh2_error for error reporting(); + * userauth.c: - tweak functions calling _libssh2_pub_priv_keyfile() not + to shadow error messages -- crypto: unify the generic functions +- TODO: remove issues we (sort of) did already + +- ssh2_exec: skip error outputs for EAGAIN - Added crypto.h that is the unified header to include when using crypto - functionality. It should be the only header that needs to adapt to the - underlying crypto library in use. It provides the set of prototypes that - are library agnostic. + Since the example uses non-blocking mode, it will just flood the output + with this "nonsense" error. -- [Mark Smith brought this change] +Guenter Knauf (30 Nov 2011) +- Some NetWare makefile tweaks. - userauth: derive publickey from private +Daniel Stenberg (18 Nov 2011) +- LIBSSH2_SFTP_PACKET_MAXLEN: increase to 80000 - Pass a NULL pointer for the publickey parameter of - libssh2_userauth_publickey_fromfile and - libssh2_userauth_hostbased_fromfile functions. In this case, the - functions recompute the public key from the private key file data. + Some SFTP servers send SFTP packets larger than 40000. Since the limit + is only present to avoid insane sizes anyway, we can easily bump it. - This is work done by Jean-Louis CHARTON - , then adapted by Mark Smith and - slightly edited further by me Daniel. + The define was formerly in the public header libssh2_sftp.h but served + no external purpose and was moved into the source dir. - WARNING: this does leave the feature NOT WORKING when libssh2 is built - to use libgcrypt instead of OpenSSL simply due to lack of - implementation. + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-11/0004.shtml + Reported by: Michael Harris -- ssh2_echo: Value stored to 'exitcode' is never read +Alexander Lamaison (18 Nov 2011) +- [Peter Krempa brought this change] -- _libssh2_packet_add: fix SSH_MSG_DEBUG weirdness + knownhost_check(): Don't dereference ext if NULL is passed - I believe I may have caused this weird typo style error when I cleaned - up this function a while ago. Corrected now. - -- uint32: more longs converted to proper types + Documentation for libssh2_knownhost_checkp() and related functions + states that the last argument is filled with data if non-NULL. - I also moved the MAC struct over to the mac.h header file and made sure - that the users of that struct include that file. - -- SFTP: more types to uint32_t + "knownhost if set to non-NULL, it must be a pointer to a 'struct + libssh2_knownhost' pointer that gets filled in to point to info about a + known host that matches or partially matches." - The 'num_names' field in the SSH_FXP_NAME response is an unsigned 32bit - value so we make sure to treat it like that. + In this function ext is dereferenced even if set to NULL, causing + segfault in applications not needing the extra data. -- SFTP: request_ids are uint32_t - - I went over the code and made sure we use uint32_t all over for the - request_id data. It is an unsigned 32bit value on the wire. +Daniel Stenberg (11 Nov 2011) +- [Peter Krempa brought this change] -- SFTP: store request_id separately in packets + knownhost_add: Avoid dereferencing uninitialized memory on error path. - By using a new separate struct for incoming SFTP packets and not sharing - the generic packet struct, we can get rid of an unused field and add a - new one dedicated for holding the request_id for the incoming - package. As sftp_packet_ask() is called fairly often, a "mere" integer - comparison is MUCH faster than the previous memcmp() of (typically) 5 - bytes. + In function knownhost_add, memory is alocated for a new entry. If normal + alocation is used, memory is not initialized to 0 right after, but a + check is done to verify if correct key type is passed. This test is done + BEFORE setting the memory to null, and on the error path function + free_host() is called, that tries to dereference unititialized memory, + resulting into a glibc abort(). + + * knownhost.c - knownhost_add(): - move typemask check before alloc -- libssh2_sftp_open_ex: man page extended and cleaned up +- windows build: add define to avoid compiler warning - I added the missing documentation for the 'flags' argument. + A recent mingw compiler has started to complain on "#warning Please + include winsock2.h before windows.h" unless the magic define is set + first. + + Reported by: Vincent Torri + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-10/0064.shtml -- SFTP: unify the READ/WRITE chunk structs +Henrik Nordstrom (31 Oct 2011) +- [Vincent Torri brought this change] -- SFTP: fix memory leaks + Correct Windows include file name case, simplifying cross-compilation - Make sure that we cleanup remainders when the handle is closed and when - the subsystem is shutdown. + When cross compiling to Windows, libssh2.h include Windows header files + with upper case filenames : BaseTsd.h and WinSock2.h. - Existing flaw: if a single handle sends packets that haven't been - replied to yet at the time when the handle is closed, those packets will - arrive later and end up in the generic packet brigade queue and they - will remain in there until flushed. They will use unnecessary memory, - make things slower and they will ruin the SFTP handling if the - request_id counter ever wraps (highly unlikely to every happen). + These files have lowercase names with mingw-w64 (iirc, it's the same with + mingw). And as on Windows, being lowercase or uppercase does not matter. -- sftp_close_handle: packet list is generic - - Fix comment, simplify the loop logic +Daniel Stenberg (25 Oct 2011) +- [Jernej Kovacic brought this change] -- sftp_read: pipeline reads - - The SFTP read function now does transfers the same way the SFTP write - function was made to recently: it creates a list of many outgoing - FXP_READ packets that each asks for a small data chunk. The code then - tries to keep sending read request while collecting the acks for the - previous requests and returns the received data. + libssh2_session_supported_algs: added -- sftp_write: removed unused variable +- [Kamil Dudka brought this change] -- _libssh2_channel_close: don't call transport read if disconnected - - The loop that waits for remote.close to get set may end up looping - forever since session->socket_state gets set to - LIBSSH2_SOCKET_DISCONNECTED by the packet_add() function called from the - transport_read() function and after having been set to - LIBSSH2_SOCKET_DISCONNECTED, the transport_read() function will only - return 0. + example/sftp_RW_nonblock: do not ignore LIBSSH2_ERROR_EAGAIN - Bug: http://trac.libssh2.org/ticket/198 + Bug: https://bugzilla.redhat.com/745420 -- libssh2_sftp_seek64: new man page +Peter Stuge (5 Oct 2011) +- example/ssh2_agent: Print host key fingerprint before authentication - Split off libssh2_sftp_seek64 from the libssh2_sftp_seek man page, and - mentioned that we consider the latter deprecated. Also added a mention - about the dangers of doing seek during writing or reading. + Also moves the comment about not being authenticated to before the + agent authentication takes place, so that it better matches the code. -- sftp_seek: fix - - The new SFTP write code caused a regression as the seek function no - longer worked as it didn't set the write position properly. +Daniel Stenberg (29 Sep 2011) +- OpenSSL EVP: fix threaded use of structs - It should be noted that seeking is STRONGLY PROHIBITED during upload, as - the upload magic uses two different offset positions and the multiple - outstanding packets etc make them sensitive to change in the midst of - operations. + Make sure we don't clear or reset static structs after first init so + that they work fine even when used from multiple threads. Init the + structs in the global init. - This functionality was just verified with the new example code - sftp_append. This bug was filed as bug #202: + Help and assistance by: John Engstrom - Bug: http://trac.libssh2.org/ticket/202 - -- sftp_append: new example doing SFTP append + Fixes #229 (again) -- MAX_SFTP_OUTGOING_SIZE: 30000 +- openssl: don't init static structs differently - I ran SFTP upload tests against localhost. It showed that to make the - app reach really good speeds, I needed to do a little code tweak and - change MAX_SFTP_OUTGOING_SIZE from 4000 to 30000. The tests I did before - with the high latency tests didn't show any real difference whatever I - had that size set to. + make_ctr_evp() is changed to take a struct pointer, and then each + _libssh2_EVP_aes_[keylen]_ctr function is made to pass in their own + static struct - This number is the size in bytes that libssh2 cuts off the large input - buffer and sends off as an individual sftp packet. + Reported by: John Engstrom + Fixes #229 -- sftp_write_sliding.c: new example - - This is an example that is very similar to sftp_write_nonblock.c, with - the exception that this uses - - 1 - a larger upload buffer +Guenter Knauf (27 Sep 2011) +- Removed obsolete include path. + +Daniel Stenberg (21 Sep 2011) +- read_state: clear the state variable better - 2 - a sliding buffer mechnism to allow the app to keep sending lots of - data to libssh2 without having to first drain the buffer. + Set read_state back to idle before trying to send anything so that if + the state somehow is wrongly set. - These are two key issues to make libssh2 SFTP uploads really perform - well at this point in time. + Also, avoid such a case of confusion by resetting the read_state when an + sftp handle is closed. -- cpp: s/#elsif/#elif +- sftp_read: remove leftover fprintf - This looks like a typo as #elsif is not really C... + Reported by: Alexander Lamaison -- _libssh2_channel_write: revert channel_write() use +- sftp.h: fix the #ifdef to prevent multiple inclusions + +- sftp_read: use a state variable to avoid bad writes - The attempts made to have _libssh2_channel_write() accept larger pieces - of data and split up the data by itself into 32700 byte chunks and pass - them on to channel_write() in a loop as a way to do faster operations on - larger data blocks was a failed attempt. + When a channel_write call has gotten an EAGAIN back, we try harder to + continue the same write in the subsequent invoke. + +- window_size: explicit adjustments only - The reason why it is difficult: + Removed the automatic window_size adjustments from + _libssh2_channel_read() and instead all channel readers must now make + sure to enlarge the window sizes properly themselves. - The API only allows EAGAIN or a length to be returned. When looping over - multiple blocks to get sent, one block can get sent and the next might - not. And yet: when transport_send() has returned EAGAIN we must not call - it again with new data until it has returned OK on the existing data it - is still working on. This makes it a mess and we do get a much easier - job by simply returning the bytes or EAGAIN at once, as in the EAGAIN - case we can assume that we will be called with the same arguments again - and transport_send() will be happy. + libssh2_channel_read_ex() - the public function, now grows the window + size according to the requested buffer size. Applications can still opt + to grow the window more on demand. Larger windows tend to give higher + performance. - Unfortunately, I think we take a small performance hit by not being able - to do this. + sftp_read() now uses the read-ahead logic to figure out a window_size. -- ssh2_echo: new example - - This is a new example snippet. The code is largely based on ssh2_exec, - and is written by Tommy Lindgren. I edited it into C90 compliance and to - conform to libssh2 indent style and some more. +- libssh2.h: bump the default window size to 256K -- send_existing: return after send_existing - - When a piece of data is sent from the send_existing() function we must - make the parent function return afterwards. Otherwise we risk that the - parent function tries to send more data and ends up getting an EGAIN for - that more data and since it can only return one return code it doesn't - return info for the successfully sent data. +- libssh2_userauth_keyboard_interactive.3: fix man warning - As this change is a regression I now added a larger comment explaining - why it has to work like this. + It seemed to occur due to the excessive line length -- _libssh2_channel_write: count resent data as written +- [Mikhail Gusarov brought this change] + + Add missing .gitignore entries + +- [Mikhail Gusarov brought this change] + + Add manpage syntax checker to 'check' target - In the logic that resends data that was kept for that purpose due to a - previous EAGAIN, the data was not counted as sent causing badness. + In virtually every libssh2 release Debian's lintian catches syntax errors in + manpages. Prevent it by checking manpages as a part of testsuite. -Peter Stuge (13 Nov 2010) -- Use fprintf(stderr, ) instead of write(2, ) for debugging +- libssh2_banner_set.3: fix nroff syntax mistake -- session/transport: Correctly handle when _libssh2_send() returns -EAGAIN +Guenter Knauf (10 Sep 2011) +- Use predefined resource compiler macro. -- src/agent.c: Simplify _libssh2_send() error checking ever so slightly +- Added casts to silent compiler warnings. -Daniel Stenberg (12 Nov 2010) -- send/recv: use _libssh2_recv and _libssh2_send now +- Fixed uint64_t printf. + +- Fixed macro function signatures. + +- NetWare makefile tweaks. + +- Removed unused var. + +- Added 2 samples not mentioned. + +- Dont build x11 sample with MinGW. + +- Fixed executable file description. + +- Removed unused var. + +- Kill stupid gcc 3.x uninitialized warning. + +- Build all examples. + +- More MinGW makefile tweaks. - Starting now, we unconditionally use the internal replacement functions - for send() and recv() - creatively named _libssh2_recv() and - _libssh2_send(). + Renamed *.mingw makefiles to GNUmakefile since GNU make picks these + up automatically, and therefore win32/Makefile removed. + +- Removed forgotten WINSOCK_VERSION defines. + +Daniel Stenberg (9 Sep 2011) +- libssh2_session_startup(3) => libssh2_session_handshake(3) - On errors, these functions return the negative 'errno' value instead of - the traditional -1. This design allows systems that have no "natural" - errno support to not have to invent it. It also means that no code - outside of these two transfer functions should use the errno variable. + Propagate for the current function in docs and examples. + libssh2_session_startup() is deprecated. -- channel_write: move some logic to _libssh2_channel_write +- libssh2_banner_set => libssh2_session_banner_get - Some checks are better done in _libssh2_channel_write just once per - write instead of in channel_write() since the looping will call the - latter function multiple times per _libssh2_channel_write() invoke. + Marked the old function as deprecated. Added the new name in the correct + name space with the same arguments and functionality. -- sftp_write: handle "left over" acked data +- new function: libssh2_session_banner_get - The SFTP handle struct now buffers number of acked bytes that haven't - yet been returned. The way this is used is as following: + Returns the banner from the server handshake - 1. sftp_write() gets called with a buffer of let say size 32000. We - split 32000 into 8 smaller packets and send them off one by one. One of - them gets acked before the function returns so 4000 is returned. + Fixes #226 + +- libssh2.h: bump version to 1.4.0 for new function(s) + +- remove embedded CVS/svn tags + +- [liuzl brought this change] + + API add:libssh2_sftp_get_channel - 2. sftp_write() gets called again a short while after the previous one, - now with a much smaller size passed in to the function. Lets say 8000. - In the mean-time, all of the remaining packets from the previous call - have been acked (7*4000 = 28000). This function then returns 8000 as all - data passed in are already sent and it can't return any more than what - it got passed in. But we have 28000 bytes acked. We now store the - remaining 20000 in the handle->u.file.acked struct field to add up in - the next call. + Return the channel of sftp, then caller can + control the channel's behavior. - 3. sftp_write() gets called again, and now there's a backlogged 20000 - bytes to return as fine and that will get skipped from the beginning - of the buffer that is passed in. + Signed-off-by: liuzl -- sftp_write: polished and simplified +- _libssh2_channel_read: react on errors from receive_window_adjust - Removed unnecessary struct fields and state changes within the function. + Previously the function would ignore all errors except for EAGAIN. + +- sftp_read: extend and clarify the documentation + +- sftp_read: cap the read ahead maximum amount - Made the loop that checks for ACKs only check chunks that were fully - sent. + Now we only go up to LIBSSH2_CHANNEL_WINDOW_DEFAULT*30 bytes SFTP read + ahead, which currently equals 64K*30 == 1966080 bytes. -- SCP: on failure, show the numerical error reason +- _libssh2_channel_read: fix non-blocking window adjusting - By calling libssh2_session_last_errno() + If EAGAIN is returned when adjusting the receive window, we must not + read from the transport directly until we've finished the adjusting. + +Guenter Knauf (8 Sep 2011) +- Fix for systems which need sys/select.h. + +- The files were not gone but renamed ... + +Daniel Stenberg (6 Sep 2011) +- sftp_read: added documenting comment + + Taken from some recent email conversations I added some descriptions of + the logic in sftp_read() to aid readers. + +- 1.3.1: start the work + +Version 1.3.0 (6 Sep 2011) + +Daniel Stenberg (6 Sep 2011) +- Makefile.am: the Makefile.win32 files are gone -- SFTP: provide the numerical error reason on failure +- RELEASE-NOTES: updated for 1.3.0 -- SCP: clean up failure treatment +- sftp_read: a short read is not end of file - When SCP send or recv fails, it gets a special message from the server - with a warning or error message included. We have no current API to - expose that message but the foundation is there. Removed unnecessary use - of session struct fields. + A returned READ packet that is short will now only reduce the + offset. + + This is a temporary fix as it is slightly better than the previous + approach but still not very good. -- sftp_write: enlarge buffer to perform better +- [liuzl brought this change] -- packets: code cleanup - - I added size checks in several places. I fixed the code flow to be easier - to read in some places. + _libssh2_packet_add: adjust window size when truncating - I removed unnecessary zeroing of structs. I removed unused struct fields. - -- LIBSSH2_CALLBACK_MACERROR: clarify return code use + When receiving more data than what the window size allows on a + particular channel, make sure that the window size is adjusted in that + case too. Previously it would only adjust the window in the non-error + case. -- _libssh2_userauth_publickey: avoid shadowing +Guenter Knauf (29 Aug 2011) +- Silent compiler warning with MinGW64. -- packet: avoid shadowing global symbols +- Fixed link to native Win32 awk tool. -- sftp_readdir: avoid shadowing +- Renamed MinGW makefiles. -- shadowing: don't shadow the global compress +- Some MinGW makefile tweaks. + + Enable build without GNU tools and with MinGW64 compiler. -- _libssh2_packet_add: turn ifs into a single switch +- Fixed aes_ctr_do_cipher() signature. -- _libssh2_packet_add: check SSH_MSG_GLOBAL_REQUEST packet +Daniel Stenberg (26 Aug 2011) +- [liuzl brought this change] -- _libssh2_packet_add: SSH_MSG_DEBUG length checks + libssh2_sftp_seek64: flush packetlist and buffered data - Verify lengths before using them. Read always_display from the correct - index. Don't copy stuff around just to provide zero-termination of the - strings. + When seeking to a new position, flush the packetlist and buffered data + to prevent already received or pending data to wrongly get used when + sftp-reading from the new offset within the file. -- _libssh2_packet_add: SSH_MSG_IGNORE skip memmove +- sftp_read: advance offset correctly for buffered copies - There's no promise of a zero termination of the data in the callback so - no longer perform ugly operation in order to provide it. - -- _libssh2_packet_add: SSH_MSG_DISCONNECT length checks + In the case where a read packet has been received from the server, but + the entire contents couldn't be copied to the user-buffer, the data is + instead buffered and copied to the user's buffer in the next invocation + of sftp_read(). When that "extra" copy is made, the 'offset' pointer was + not advanced accordingly. - Verify lengths before trying to read data. + The biggest impact of this flaw was that the 'already' variable at the + top of the function that figures out how much data "ahead" that has + already been asked for would slowly go more and more out of sync, which + could lead to the file not being read all the way to the end. + + This problem was most noticable in cases where the application would + only try to read the exact file size amount, like curl does. In the + examples libssh2 provides the sftp read function is most often called + with a fixed size large buffer and then the bug would not appear as + easily. + + This bug was introduced in the SFTP rewrite in 1.2.8. + + Bug: http://curl.haxx.se/mail/lib-2011-08/0305.html + http://www.libssh2.org/mail/libssh2-devel-archive-2011-08/0085.shtml -- indent: break lines at 80 columns +- wrap some long lines < 80 columns -- SSH_MSG_CHANNEL_OPEN_FAILURE: used defined values - - We don't like magic numbers in the code. Now the acceptable failure - codes sent in the SSH_MSG_CHANNEL_OPEN_FAILURE message are added as - defined values in the private header file. +- LIBSSH2_RECV: fix typo, use the RECV_FD macro -- sftp_write: don't return EAGAIN if no EAGAIN was received - - This function now only returns EAGAIN if a lower layer actually returned - EAGAIN to it. If nothing was acked and no EAGAIN was received, it will - now instead return 0. +- subsystem_netconf.c: fix compiler warnings -- _libssh2_wait_socket: detect nothing-to-wait-for - - If _libssh2_wait_socket() gets called but there's no direction set to - wait for, this causes a "hang". This code now detects this situation, - set a 1 second timeout instead and outputs a debug output about it. +- [Henrik Nordstrom brought this change] -- decomp: remove the free_dest argument - - Since the decompress function ALWAYS returns allocated memory we get a - lot simpler code by removing the ability to return data unallocated. + Custom callbacks for performing low level socket I/O -- decomp: cleaned off old compression stuff - - I cleared off legacy code from when the compression and decompression - functions were a single unified function. Makes the code easier to read - too. +- version bump: start working towards 1.3.0 -- [TJ Saunders brought this change] +Version 1.2.9 (16 Aug 2011) - decomp: increase decompression buffer sizes +Daniel Stenberg (16 Aug 2011) +- RELEASE-NOTES: synced with 95d69d3a81261 -- [TJ Saunders brought this change] +- [Henrik Nordstrom brought this change] - zlib: Add debug tracing of zlib errors + Document prototypes for macro defined functions -- sftp_packet_read: handle partial reads of the length field - - SFTP packets come as [32 bit length][payload] and the code didn't - previously handle that the initial 32 bit field was read only partially - when it was read. +- [Henrik Nordstrom brought this change] -- [Jasmeet Bagga brought this change] + Avoid reuse after free when closing X11 channels - kex_agree_hostkey: fix NULL pointer derefence +- _libssh2_channel_write: handle window_size == 0 better - While setting up the session, ssh tries to determine the type of - encryption method it can use for the session. This requires looking at - the keys offered by the remote host and comparing these with the methods - supported by libssh2 (rsa & dss). To do this there is an iteration over - the array containing the methods supported by libssh2. + When about to send data on the channel and the window size is 0, we must + not just return 0 if the transport_read() function returned EAGAIN as it + then causes a busy-loop. - If there is no agreement on the type of encryption we come to the 3rd - entry of the hostkeyp array. Here hostkeyp is valid but *hostkep is - NULL. Thus when we dereference that in (*hostkeyp)->name there is a - crash + Bug: http://libssh2.org/mail/libssh2-devel-archive-2011-08/0011.shtml -- _libssh2_transport_send: remove dead assignment +- gettimeofday: fix name space pollution - 'data' isn't accessed beyond this point so there's no need to assign it. - -- scp_recv: remove dead assignment + For systems without its own gettimeofday() implementation, we still must + not provide one outside our namespace. - Instead of assigning a variable we won't read, we now use the more - explicit (void) prefix. - -- sftp_write: removed superfluous assignment + Reported by: Bill Segall -- bugfix: avoid use of uninitialized value +Dan Fandrich (5 Aug 2011) +- libssh2.pc.in: Fixed spelling in pkgconfig file -- sftp_packet_require: propagate error codes better - - There were some chances that they would cause -1 to get returned by - public functions and as we're hunting down all such occurances and since - the underlying functions do return valuable information the code now - passes back proper return codes better. +Peter Stuge (17 Jul 2011) +- example/subsystem_netconf.c: Add missing #include -- [Alfred Gebert brought this change] +- example/subsystem_netconf.c: Discard ]]>]]> and return only XML response - fix memory leaks (two times cipher_data) for each sftp session +- example/subsystem_netconf.c: Fix uninitialized variable bug -- libssh2_userauth_authenticated: make it work as documented - - The man page clearly says it returns 1 for "already authenticated" but - the code said non-zero. I changed the code to use 1 now, as that is also - non-zero but it gets the benefit that it now matches the documentation. +- example: Add subsystem_netconf.c - Using 1 instead of non-zero is better for two reasons: + This example demonstrates how to use libssh2 to send a request to + the NETCONF subsystem available e.g. in JunOS. - 1. We have the opportunity to introduce other return codes in the future for - things like error and what not. - 2. We don't expose the internal bitmask variable value. - -- userauth_keyboard_interactive: fix indent + See also http://tools.ietf.org/html/draft-ietf-netconf-ssh-06 -- [Alfred Gebert brought this change] +Daniel Stenberg (16 Jul 2011) +- man page cleanups: non-existing functions need no man pages - fix memory leak in userauth_keyboard_interactive() +- libssh2_new_host_entry.3: removed - First I wanted to free the memory in session_free() but then - I had still memory leaks because in my test case the function - userauth_keyboard_interactive() is called twice. It is called - twice perhaps because the server has this authentication - methods available: publickey,gssapi-with-mic,keyboard-interactive - The keyboard-interactive method is successful. - -- dist: include sftp.h in dist archives - -Simon Josefsson (27 Oct 2010) -- Update header to match new function prototype, see c48840ba88. - -Daniel Stenberg (26 Oct 2010) -- bugfixes: the transport rearrange left some subtle flaws now gone - -- libssh2_userauth_publickey_fromfile_ex.3: cleaned up looks + This is just junk leftovers. -- libssh2_userauth_publickey: add man page +- userauth_keyboard_interactive: fix buffer overflow - I found an undocumented public function and we can't have it like - that. The description here is incomplete, but should serve as a template - to allow filling in... - -- libssh2_sftp_write.3: added blurb about the "write ahead" + Partly reverse 566894494b4972ae12 which was simplifying the code far too + much and ended up overflowing a buffer within the LIBSSH2_SESSION + struct. Back to allocating the buffer properly like it used to do. - Documented the new SFTP write concept + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-06/0032.shtml + Reported by: Alfred Gebert -- sftp_close_handle: free any trailing write chunks +- keyboard-interactive man page: cleaned up -- _libssh2_channel_write: fix warnings +- [Alfred Gebert brought this change] -- SFTP: bufgix, move more sftp stuff to sftp.h + _libssh2_recv(): handle ENOENT error as EAGAIN - The sftp_write function shouldn't assume that the buffer pointer will be - the same in subsequent calls, even if it assumes that the data already - passed in before haven't changed. + A sftp session failed with error "failure establishing ssh session" on + Solaris and HP-UX. Sometimes the first recv() function call sets errno + to ENOENT. In the man pages for recv of Solaris and HP-UX the error + ENOENT is not documented. - The sftp structs are now moved to sftp.h (which I forgot to add before) + I tested Solaris SPARC and x86, HP-UX i64, AIX, Windows and Linux. -- SFTP: use multiple outgoing packets when writing +- agent_list_identities: fix out of scope access - sftp_write was rewritten to split up outgoing data into multiple packets - and deal with the acks in a more asynchronous manner. This is meant to - help overcome latency and round-trip problems with the SFTP protocol. + An auto variable out of scope was being referenced and used. + + fixes #220 -- TODO: implemented a lot of the ideas now +- _libssh2_wait_socket: fix timeouts for poll() uses -- _libssh2_channel_write: removed 32500 size limit +- windows: inclusion fix - Neither _libssh2_channel_write nor sftp_write now have the 32500 size - limit anymore and instead the channel writing function now has its own - logic to send data in multiple calls until everything is sent. + include winsock2.h for all windows compilers -- send_existing: don't tell parent to return when drained +- keyb-interactive: add the fixed buffer - That will just cause unnecessary code execution. + Belongs to commit 5668944 + +- code cleanup: don't use C99/c++ comments + + We aim for C89 compliance -- _libssh2_channel_write: general code cleanup +- keyb-interactive: allow zero length fields - simplified the function and removed some unused struct fields + Allow zero length fields so they don't cause malloc(0) calls + + Avoid free()ing NULL pointers + + Avoid a malloc of a fixed 5 byte buffer. -- _libssh2_transport_send: replaces _libssh2_transport_write +- libssh2_channel_process_startup.3: clean up - The new function takes two data areas, combines them and sends them as a - single SSH packet. This allows several functions to allocate and copy - less data. + Remove the references to the macro-fied shortcuts as they have their own + individual man pages. - I also found and fixed a mixed up use of the compression function - arguments that I introduced in my rewrite in a recent commit. + Made the prototype different and more readable. -- scp_write_nonblock: use select() instead of busyloop +- man page: fix .BR lines - Make this example nicer by not busylooping. + We don't use \fI etc on .BR lines -- send_existing: clear olen when the data is sent off +- userauth_keyboard_interactive: skip code on zero length auth -- _libssh2_transport_write: allow 256 extra bytes around the packet +- libssh2_channel_forward_accept.3: mention how to get error + + Since this returns a pointer, libssh2_session_last_errno() must be used + to get the actual error code and it wasn't that clear before. -- _libssh2_transport_write: remade to send without malloc +- timeout docs: mention they're added in 1.2.9 -- compress: compression disabled by default +- sftp_write_sliding.c: indent fix - We now allow libssh2_session_flag() to enable compression with a new - flag and I added documentation for the previous LIBSSH2_FLAG_SIGPIPE - flag which I wasn't really aware of! + Use the standard indenting and removed CVS leftover comment -- comp: split the compress function +- [zl liu brought this change] + + sftp_write_sliding: send the complete file - It is now made into two separate compress and decompress functions. In - preparation for upcoming further modficications. + When reaching the end of file there can still be data left not sent. -Dan Fandrich (20 Oct 2010) -- Added header file to allow compiling in older environments +- [Douglas Masterson brought this change] -Daniel Stenberg (20 Oct 2010) -- TODO: add a possible new API for SFTP transfers + session_startup: init state properly + + libssh2_session_startup() didn't set the state correctly so it could get + confused. + + Fixes #218 -- TODO: "New Transport API" added +- timeout: added man pages -- TODO: add buffering plans +- BLOCK_ADJUST_ERRNO: move rc to right level + + We can't declare the variable within the block and use it in the final + do-while() expression to be properly portable C89. -Simon Josefsson (13 Oct 2010) -- Mention libssh2_channel_get_exit_signal and give kudos. +- [Matt Lilley brought this change] -- [Tommy Lindgren brought this change] + adds a timeout to blocking calls + + Fixes bug #160 as per Daniel's suggestion + + Adds libssh2_session_set_timeout() and libssh2_session_get_timeout() - Add libssh2_channel_get_exit_signal man page. +- SCP: fix incorrect error code - Signed-off-by: Simon Josefsson + After an error occurs in libssh2_scp_recv() or libssh2_scp_send(), the + function libssh2_session_last_error() would return + LIBSSH2_ERROR_SOCKET_NONE on error. + + Bug: http://trac.libssh2.org/ticket/216 + Patch by: "littlesavage" + + Fixes #216 -- [Tommy Lindgren brought this change] +Guenter Knauf (19 Apr 2011) +- Updated default (recommended) dependency versions. - Add libssh2_channel_get_exit_signal. +Daniel Stenberg (17 Apr 2011) +- libssh2_session_block_directions: fix mistake - Signed-off-by: Simon Josefsson - -- Add libssh2_free man page and fix typo. + The last LIBSSH2_SESSION_BLOCK_INBOUND should be + LIBSSH2_SESSION_BLOCK_OUTBOUND + + And I shortened the short description + + Reported by: "drswinghead" -- Add libssh2_free. +- msvcproj: added libs and debug stuff + + Added libraries needed to link whether using openssl dynamically or + statically + + Added LIBSSH2DEBUG define to debug versions to enable tracing + + URL: http://trac.libssh2.org/ticket/215 + Patch by: Mark Smith -Daniel Stenberg (11 Oct 2010) -- scp_recv: improved treatment of channel_read() returning zero +- sftp_write: clean offsets on error - As a zero return code from channel_read() is not an error we must make - sure that the SCP functions deal with that properly. channel_read() - always returns 0 if the channel is EOFed already so we check for EOF - after 0-reads to be able to return error properly. + When an error has occurred on FXP_WRITE, we must make sure that the + offset, sent offset and acked counter are reset properly. -- libssh2_session_methods.3: detail what can be asked for +- example/.gitignore: ignore built binaries -- compression: send zlib before none +- sftp_write: flush the packetlist on error - As the list of algorithms in a preferred order we should send zlib - before none to increase the chances that the server will let us do - compression. + When an error occurs during write, flush the entire list of pending + outgoing SFTP packets. -- compress: faster check, better return codes - - In the transport functions we avoid a strcmp() now and just check a - boolean instead. +- keepalive: add first basic man pages - The compress/decompress function's return code is now acknowledged and - used as actual return code in case of failures. + Someone on IRC pointed out that we don't have these documented so I + wrote up a first set based on the information in the wiki: + http://trac.libssh2.org/wiki/KeepAlive -- libssh2_session_handshake: replaces libssh2_session_startup() +- scp_write_nonblock.c: remove pointless check - The function libssh2_session_startup() is now considered deprecated due - to the portability issue with the socket argument. - libssh2_session_handshake() is the name of the replacement. + libssh2_channel_write() cannot return a value that is larger than the + input length value -- libssh2_socket_t: now externally visible - - In preparation for upcominig changes, the libssh2_socket_t type is now - typedef'ed in the public header. +Mikhail Gusarov (9 Apr 2011) +- s/\.NF/.nf/ to fix wrong macro name caught by man --warnings -- _libssh2_transport_drain: removed +Daniel Stenberg (6 Apr 2011) +- version: bump to 1.2.9_dev - This function proved not to be used nor useful. + Also update the copyright year range to include 2011 -- _libssh2_channel_write: don't iterate over transport writes +- configure: fix $VERSION - When a call to _libssh2_transport_write() succeeds, we must return from - _libssh2_channel_write() to allow the caller to provide the next chunk - of data. + Stop using the $VERSION variable as it seems to be magically used by + autoconfig itself and thus gets set to the value set in AC_INIT() + without us wanting that. $LIBSSH2VER is now the libssh2 version as + detected. - We cannot move on to send the next piece of data that may already have - been provided in this same function call, as we risk getting EAGAIN for - that and we can't return information both about sent data as well as - EAGAIN. So, by returning short now, the caller will call this function - again with new data to send. + Reported by: Paul Howarth + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2011-04/0008.shtml -- _libssh2_transport_write: updated documentation blurb +- maketgz: use git2news.pl by the correct name -- _libssh2_transport_write: remove fprintf remainder - - Mistake from previous debugging +Version 1.2.8 (4 Apr 2011) -- session: improved errors +Daniel Stenberg (4 Apr 2011) +- RELEASE-NOTES: synced with fabf1a45ee + +- NEWS: auto-generated from git - Replaced -1/SOCKET_NONE errors with appropriate error defines instead. + Starting now, the NEWS file is generated from git using the git2news.pl + script. This makes it always accurate and up-to-date, even for daily + snapshots etc. + +- sftp_write: handle FXP_WRITE errors - Made the verbose trace output during banner receiving less annoying for - non-blocking sessions. + When an sftp server returns an error back on write, make sure the + function bails out and returns the proper error. -- crypt_init: use correct error define +- configure: stop using the deprecated AM_INIT_AUTOMAKE syntax -- _libssh2_error: hide EAGAIN for non-blocking sessions +Alexander Lamaison (13 Mar 2011) +- Support unlimited number of host names in a single line of the known_hosts file. - In an attempt to make the trace output less cluttered for non-blocking - sessions the error function now avoids calling the debug function if the - error is the EAGAIN and the session is non-blocking. - -- agent: use better error defines + Previously the code assumed either a single host name or a hostname,ip-address pair. However, according to the spec [1], there can be any number of comma separated host names or IP addresses. + + [1] http://www.openbsd.org/cgi-bin/man.cgi?query=sshd&sektion=8 -- comp_method_zlib_init: use correct error defines +Daniel Stenberg (26 Feb 2011) +- libssh2_knownhost_readfile.3: clarify return value + + This function returns the number of parsed hosts on success, not just + zero as previously documented. -- transport: better error codes +Peter Stuge (26 Feb 2011) +- Don't save allocated packet size until it has actually been allocated - LIBSSH2_SOCKET_NONE (-1) should no longer be used as error code as it is - (too) generic and we should instead use specific and dedicated error - codes to better describe the error. + The allocated packet size is internal state which needs to match reality + in order to avoid problems. This commit fixes #211. -- channel: return code and _libssh2_error cleanup +Daniel Stenberg (21 Feb 2011) +- [Alfred Gebert brought this change] + + session_startup: manage server data before server identification - Made sure that all transport_write() call failures get _libssh2_error - called. + Fix the bug that libssh2 could not connect if the sftp server + sends data before sending the version string. + + http://tools.ietf.org/html/rfc4253#section-4.2 + + "The server MAY send other lines of data before sending the version + string. Each line SHOULD be terminated by a Carriage Return and Line + Feed. Such lines MUST NOT begin with "SSH-", and SHOULD be encoded + in ISO-10646 UTF-8 [RFC3629] (language is not specified). Clients + MUST be able to process such lines." -- _libssh2_channel_write: limit to 32700 bytes +- [Alfred Gebert brought this change] + + fullpacket: decompression only after init - The well known and used ssh server Dropbear has a maximum SSH packet - length at 32768 by default. Since the libssh2 design current have a - fixed one-to-one mapping from channel_write() to the packet size created - by transport_write() the previous limit of 32768 in the channel layer - caused the transport layer to create larger packets than 32768 at times - which Dropbear rejected forcibly (by closing the connection). + The buffer for the decompression (remote.comp_abstract) is initialised + in time when it is needed. With this fix decompression is disabled when + the buffer (remote.comp_abstract) is not initialised. - The long term fix is of course to remove the hard relation between the - outgoing SSH packet size and what the input length argument is in the - transport_write() function call. + Bug: http://trac.libssh2.org/ticket/200 + +- _libssh2_channel_read: store last error + + When the transport layer returns EAGAIN this function didn't call + _libssh2_error() which made the last_error not get set. -- libssh.h: add more dedicated error codes +- sftp_write: clarified the comment header -- SCP: allow file names with bytes > 126 +- sftp_read: avoid wrapping counter to insanity - When parsing the SCP protocol and verifying that the data looks like a - valid file name, byte values over 126 must not be consider illegal since - UTF-8 file names will use such codes. + As pointed out in bug #206, if a second invoke of libssh2_sftp_read() + would shrink the buffer size, libssh2 would go nuts and send out read + requests like crazy. This was due to an unsigned variable turning + "negative" by some wrong math, and that value would be the amount of + data attempt to pre-buffer! - Reported by: Uli Zappe - Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2010-08/0112.shtml - -Dan Fandrich (25 Aug 2010) -- Document the three sftp stat constants + Bug: http://trac.libssh2.org/ticket/206 -Guenter Knauf (18 Aug 2010) -- Fixed Win32 makefile which was now broken at resource build. +- sftp_packet_read: use 32bit variables for 32bit data -- It is sufficient to pipe stderr to NUL to get rid of the nasty messages. +- libssh2_sftp_stat_ex.3: cleaned up, extended + + Removed the macros from it as they have their own man pages. + + Added the LIBSSH2_SFTP_ATTRIBUTES struct in here for easier reference. -- [Author: Guenter Knauf brought this change] +- sftp_readdir: return error if buffer is too small + + If asked to read data into a buffer and the buffer is too small to hold + the data, this function now returns an error instead of as previously + just copy as much as fits. - Removed Win32 ifdef completely for sys/uio.h. +- sftp_symlink: return error if receive buffer too small - No idea why we had this ifdef at all but MSVC, MingW32, Watcom - and Borland all have no sys/uio.h header; so if there's another - Win32 compiler which needs it then it should be added explicitely - instead of this negative list. + and clean up some variable type mismatches + + Discussion: http://www.libssh2.org/mail/libssh2-devel-archive-2011-01/0001.shtml -- New files should also be added to Makefile.am. +- docs: clarify what happens with a too small buffer - Otherwise they will never be included with release and snapshot tarballs ... + This flaw is subject to change, but I figured it might be valuable to + users of existing code to know how it works. -Daniel Stenberg (18 Aug 2010) -- version: bump to 1.2.8_DEV +- channel_request_pty_size: fix reqPTY_state + + The state variable isn't properly set so every other call to the + function fails! + + Bug: http://libssh2.org/mail/libssh2-devel-archive-2010-12/0096.shtml + Reported by: Steve Legg -Version 1.2.7 (17 Aug 2010) +- data size: cleanup + + Fix 64bit warnings by using (s)size_t and dedicated uint32_t types more. -Daniel Stenberg (17 Aug 2010) -- release: updated to hold 1.2.7 info +- [Pierre Joye brought this change] -Guenter Knauf (17 Aug 2010) -- Use the new libssh2.rc file. + ssize_t: proper typedef with MSVC compilers + + As discussed on the mailing list, it was wrong for win64 and using the + VC-provided type is the safest approach instead of second- guessing + which one it should be. -- Added resource file for libssh2.dll (shamelessly stolen from libcurl). +Guenter Knauf (22 Dec 2010) +- Updated OpenSSL version. -- Updated Win32 MSVC dependencies versions. +- Expanded tabs to spaces. -- Added include for sys/select.h to get fd.set on some platforms. +Peter Stuge (21 Dec 2010) +- [Joey Degges brought this change] -- Added Watcom makefile borrowed from libcurl. + _libssh2_ntohu64: fix conversion from network bytes to uint64 - This makefile compiles already all files fine for static lib, but needs - final touch when I have OpenSSL fully working with shared libs and Watcom. - -- Added copyright define to libssh2.h and use it for binary builds. + Cast individual bytes to uint64 to avoid overflow in arithmetic. -- Moved version defines up in order to include from .rc file. +Daniel Stenberg (20 Dec 2010) +- libssh2_userauth_list: language fix - Blocked rest of header with ifndef so its possible to let - the rc compiler only use the version defines. - -- Some minor makefile tweaks. + "faily" is not a good English word, and I also cleaned up some other minor + mistakes -Daniel Stenberg (2 Aug 2010) -- example: treat the libssh2_channel_read() return code properly +- crypto: unify the generic functions - A short read is not an error. Only negative values are errors! + Added crypto.h that is the unified header to include when using crypto + functionality. It should be the only header that needs to adapt to the + underlying crypto library in use. It provides the set of prototypes that + are library agnostic. -- libssh2_wait_socket: reset error code to "leak" EAGAIN less - - Since libssh2 often sets LIBSSH2_ERROR_EAGAIN internally before - _libssh2_wait_socket is called, we can decrease some amount of - confusion in user programs by resetting the error code in this function - to reduce the risk of EAGAIN being stored as error when a blocking - function returns. +- [Mark Smith brought this change] -- _libssh2_wait_socket: poll needs milliseconds + userauth: derive publickey from private - As reported on the mailing list, the code path using poll() should - multiple seconds with 1000 to get milliseconds, not divide! + Pass a NULL pointer for the publickey parameter of + libssh2_userauth_publickey_fromfile and + libssh2_userauth_hostbased_fromfile functions. In this case, the + functions recompute the public key from the private key file data. - Reported by: Jan Van Boghout - -- typedef: make ssize_t get typedef without LIBSSH2_WIN32 + This is work done by Jean-Louis CHARTON + , then adapted by Mark Smith and + slightly edited further by me Daniel. - The condition around the ssize_t typedef depended on both LIBSSH2_WIN32 - *and* _MSC_VER being defined when it should be enough to depend on - _MSC_VER only. It also makes it nicer so libssh2-using code builds fine - without having custom defines. + WARNING: this does leave the feature NOT WORKING when libssh2 is built + to use libgcrypt instead of OpenSSL simply due to lack of + implementation. -- [John Little brought this change] +- ssh2_echo: Value stored to 'exitcode' is never read - session_free: free more data to avoid memory leaks +- _libssh2_packet_add: fix SSH_MSG_DEBUG weirdness + + I believe I may have caused this weird typo style error when I cleaned + up this function a while ago. Corrected now. -- channel_free: ignore problems with channel_close() +- uint32: more longs converted to proper types - As was pointed out in bug #182, we must not return failure from - _libssh2_channel_free() when _libssh2_channel_close() returns an error - that isn't EAGAIN. It can effectively cause the function to never go - through, like it did now in the case where the socket was actually - closed but socket_state still said LIBSSH2_SOCKET_CONNECTED. + I also moved the MAC struct over to the mac.h header file and made sure + that the users of that struct include that file. + +- SFTP: more types to uint32_t - I consider this fix the right thing as it now also survives other - errors, even if making sure socket_state isn't lying is also a good - idea. + The 'num_names' field in the SSH_FXP_NAME response is an unsigned 32bit + value so we make sure to treat it like that. -- publickey_list_free: no return value from a void function +- SFTP: request_ids are uint32_t - Fixed a compiler warning I introduced previously when checking input - arguments more. I also added a check for the other pointer to avoid NULL - pointer dereferences. + I went over the code and made sure we use uint32_t all over for the + request_id data. It is an unsigned 32bit value on the wire. -- [Lars Nordin brought this change] +- SFTP: store request_id separately in packets + + By using a new separate struct for incoming SFTP packets and not sharing + the generic packet struct, we can get rid of an unused field and add a + new one dedicated for holding the request_id for the incoming + package. As sftp_packet_ask() is called fairly often, a "mere" integer + comparison is MUCH faster than the previous memcmp() of (typically) 5 + bytes. - openssl: make use of the EVP interface +- libssh2_sftp_open_ex: man page extended and cleaned up - Make use of the EVP interface for the AES-funktion. Using this method - supports the use of different ENGINES in OpenSSL for the AES function - (and the direct call to the AES_encrypt should not be used according to - openssl.org) + I added the missing documentation for the 'flags' argument. -Peter Stuge (23 Jun 2010) -- [Tor Arntsen brought this change] +- SFTP: unify the READ/WRITE chunk structs - Don't overflow MD5 server hostkey +- SFTP: fix memory leaks - Use SHA_DIGEST_LENGTH and MD5_DIGEST_LENGTH in memcpy instead of hardcoded - values. An incorrect value was used for MD5. - -- Fix message length bugs in libssh2_debug() + Make sure that we cleanup remainders when the handle is closed and when + the subsystem is shutdown. - There was a buffer overflow waiting to happen when a debug message was - longer than 1536 bytes. + Existing flaw: if a single handle sends packets that haven't been + replied to yet at the time when the handle is closed, those packets will + arrive later and end up in the generic packet brigade queue and they + will remain in there until flushed. They will use unnecessary memory, + make things slower and they will ruin the SFTP handling if the + request_id counter ever wraps (highly unlikely to every happen). + +- sftp_close_handle: packet list is generic - Thanks to Daniel who spotted that there was a problem with the message - length passed to a trace handler also after commit - 0f0652a3093111fc7dac0205fdcf8d02bf16e89f. + Fix comment, simplify the loop logic -- Make libssh2_debug() create a correctly terminated string +- sftp_read: pipeline reads - Also use FILE *stderr rather than fd 2, which can very well be something - completely different. + The SFTP read function now does transfers the same way the SFTP write + function was made to recently: it creates a list of many outgoing + FXP_READ packets that each asks for a small data chunk. The code then + tries to keep sending read request while collecting the acks for the + previous requests and returns the received data. -Daniel Stenberg (23 Jun 2010) -- [TJ Saunders brought this change] +- sftp_write: removed unused variable - handshake: Compression enabled at the wrong time - - In KEXINIT messages, the client and server agree on, among other - things, whether to use compression. This method agreement occurs - in src/kex.c's kex_agree_methods() function. However, if - compression is enabled (either client->server, server->client, or - both), then the compression layer is initialized in - kex_agree_methods() -- before NEWKEYS has been received. +- _libssh2_channel_close: don't call transport read if disconnected - Instead, the initialization of the compression layer should - happen after NEWKEYS has been received. This looks to occur - insrc/kex.c's diffie_hellman_sha1(), which even has the comment: + The loop that waits for remote.close to get set may end up looping + forever since session->socket_state gets set to + LIBSSH2_SOCKET_DISCONNECTED by the packet_add() function called from the + transport_read() function and after having been set to + LIBSSH2_SOCKET_DISCONNECTED, the transport_read() function will only + return 0. - /* The first key exchange has been performed, + Bug: http://trac.libssh2.org/ticket/198 + +- libssh2_sftp_seek64: new man page - switch to active crypt/comp/mac mode */ + Split off libssh2_sftp_seek64 from the libssh2_sftp_seek man page, and + mentioned that we consider the latter deprecated. Also added a mention + about the dangers of doing seek during writing or reading. + +- sftp_seek: fix - There, after NEWKEYS is received, the cipher and mac algorithms - are initialized, and that is where the compression should be - initialized as well. + The new SFTP write code caused a regression as the seek function no + longer worked as it didn't set the write position properly. - The current implementation fails if server->client compression is - enabled because most server implementations follow OpenSSH's - lead, where compression is initialized after NEWKEYS. Since the - server initializes compression after NEWKEYS, but libssh2 - initializes compression after KEXINIT (i.e. before NEWKEYS), they - are out of sync. + It should be noted that seeking is STRONGLY PROHIBITED during upload, as + the upload magic uses two different offset positions and the multiple + outstanding packets etc make them sensitive to change in the midst of + operations. - Reported in bug report #180 + This functionality was just verified with the new example code + sftp_append. This bug was filed as bug #202: + + Bug: http://trac.libssh2.org/ticket/202 -- [TJ Saunders brought this change] +- sftp_append: new example doing SFTP append - userauth_hostbased_fromfile: packet length too short +- MAX_SFTP_OUTGOING_SIZE: 30000 - The packet length calculated in src/userauth.c's - userauth_hostbased_fromfile() function is too short by 4 bytes; - it forgets to add four bytes for the length of the hostname. - This causes hostbased authentication to fail, since the server - will read junk data. + I ran SFTP upload tests against localhost. It showed that to make the + app reach really good speeds, I needed to do a little code tweak and + change MAX_SFTP_OUTGOING_SIZE from 4000 to 30000. The tests I did before + with the high latency tests didn't show any real difference whatever I + had that size set to. - verified against proftpd's mod_sftp module + This number is the size in bytes that libssh2 cuts off the large input + buffer and sends off as an individual sftp packet. -- _libssh2_userauth_publickey: reject method names longer than the data +- sftp_write_sliding.c: new example - This functions get the method length by looking at the first 32 - bit of data, and I now made it not accept method lengths that are - longer than the whole data set is, as given in the dedicated - function argument. + This is an example that is very similar to sftp_write_nonblock.c, with + the exception that this uses - This was detected when the function was given bogus public key - data as an ascii string, which caused the first 32bits to create - a HUGE number. - -- NULL resistance: make more public functions survive NULL pointer input + 1 - a larger upload buffer - Sending in NULL as the primary pointer is now dealt with by more - public functions. I also narrowed the userauth.c code somewhat to - stay within 80 columns better. - -- agent: make libssh2_agent_userauth() work blocking properly + 2 - a sliding buffer mechnism to allow the app to keep sending lots of + data to libssh2 without having to first drain the buffer. - previously it would always work in a non-blocking manner + These are two key issues to make libssh2 SFTP uploads really perform + well at this point in time. -Peter Stuge (17 Jun 2010) -- Fix underscore typo for 64-bit printf format specifiers on Windows +- cpp: s/#elsif/#elif - Commit 49ddf447ff4bd80285f926eac0115f4e595f9425 was missing underscores. - -Daniel Stenberg (16 Jun 2010) -- libssh2_session_callback_set: extended the man page - -- [John brought this change] + This looks like a typo as #elsif is not really C... - LIBSSH2_DEBUG: macro uses incorrect function variable +- _libssh2_channel_write: revert channel_write() use - The LIBSSH2_DEBUG macro, defined in libssh2_priv.h, incorrectly uses the - function variable ssh_msg_disconnect when it should use ssh_msg_debug. + The attempts made to have _libssh2_channel_write() accept larger pieces + of data and split up the data by itself into 32700 byte chunks and pass + them on to channel_write() in a loop as a way to do faster operations on + larger data blocks was a failed attempt. - This shows that the LIBSSH2_CALLBACK_DEBUG callback never has worked... - -- warning: fix a compiler warning 'pointer differs in signedness' + The reason why it is difficult: - As reported in bug #177 - -- portability: introduce LIBSSH2_INT64_T_FORMAT for 64bit printf()s + The API only allows EAGAIN or a length to be returned. When looping over + multiple blocks to get sent, one block can get sent and the next might + not. And yet: when transport_send() has returned EAGAIN we must not call + it again with new data until it has returned OK on the existing data it + is still working on. This makes it a mess and we do get a much easier + job by simply returning the bytes or EAGAIN at once, as in the EAGAIN + case we can assume that we will be called with the same arguments again + and transport_send() will be happy. - As pointed out in bug #177, some of the Windows compilers use - %I64 to output 64 bit variables with the printf family. + Unfortunately, I think we take a small performance hit by not being able + to do this. -- debug: avoid sending NULL to sprintf %s +- ssh2_echo: new example - Via the _libssh2_debug() macro/function. Pointed out by john in bug report + This is a new example snippet. The code is largely based on ssh2_exec, + and is written by Tommy Lindgren. I edited it into C90 compliance and to + conform to libssh2 indent style and some more. -- sftp docs: show macro on macro page, only function on function page +- send_existing: return after send_existing - The individual man pages for macros now show the full convenience - macro as defined, and then the man page for the actual function - only shows the function. - -- code police: make the code use less than 80 columns + When a piece of data is sent from the send_existing() function we must + make the parent function return afterwards. Otherwise we risk that the + parent function tries to send more data and ends up getting an EGAIN for + that more data and since it can only return one return code it doesn't + return info for the successfully sent data. + + As this change is a regression I now added a larger comment explaining + why it has to work like this. -- libssh2_channel_write_ex: remove macros, added wording on buffer size +- _libssh2_channel_write: count resent data as written + + In the logic that resends data that was kept for that purpose due to a + previous EAGAIN, the data was not counted as sent causing badness. -- libssh2_sftp_write: document buffer size and changed some ordering +Peter Stuge (13 Nov 2010) +- Use fprintf(stderr, ) instead of write(2, ) for debugging -- libssh2_channel_write_stderr: show how the macro is defined +- session/transport: Correctly handle when _libssh2_send() returns -EAGAIN -- libssh2_channel_write: show how the macro is defined +- src/agent.c: Simplify _libssh2_send() error checking ever so slightly -- SFTP: limit write() to not produce overly large packets +Daniel Stenberg (12 Nov 2010) +- send/recv: use _libssh2_recv and _libssh2_send now - sftp_write() now limits how much data it gets at a time even more - than before. Since this function creates a complete outgoing - packet based on what gets passed to it, it is crucial that it - doesn't create too large packets. + Starting now, we unconditionally use the internal replacement functions + for send() and recv() - creatively named _libssh2_recv() and + _libssh2_send(). - With this method, there's also no longer any problem to use very - large buffers in your application and feed that to libssh2. I've - done numerous tests now with uploading data over SFTP using 100K - buffers and I've had no problems with that. + On errors, these functions return the negative 'errno' value instead of + the traditional -1. This design allows systems that have no "natural" + errno support to not have to invent it. It also means that no code + outside of these two transfer functions should use the errno variable. -- scp_write_nonblock: add transfer time info +- channel_write: move some logic to _libssh2_channel_write - Using the same timing logic and output format as - sftp_write_nonblock allows us to very easily run benchmarks on - SCP vs SFTP uploads using libssh2. + Some checks are better done in _libssh2_channel_write just once per + write instead of in channel_write() since the looping will call the + latter function multiple times per _libssh2_channel_write() invoke. -- sftp_write_nonblock: select() on socket, use *BIG* buffer, time transfer +- sftp_write: handle "left over" acked data - The select() is just to make it nicer so that it doesn't - crazy-loop on EAGAIN. The buffer size thing is mostly to verify - that this really work as supposed. + The SFTP handle struct now buffers number of acked bytes that haven't + yet been returned. The way this is used is as following: - Transfer timing is just a minor thing, but it can just as well be - there and help us time and work on performance easier using out - of the box examples. - -- agent: use _libssh2_error() when returning errors + 1. sftp_write() gets called with a buffer of let say size 32000. We + split 32000 into 8 smaller packets and send them off one by one. One of + them gets acked before the function returns so 4000 is returned. - As pointed out in bug report #173, this module basically never - used _libssh2_error() which made it work inconstently with other - parts of the libssh2 code base. This is my first take at making - this code more in line with the rest. - -- inputchecks: make lots of API functions check for NULL pointers + 2. sftp_write() gets called again a short while after the previous one, + now with a much smaller size passed in to the function. Lets say 8000. + In the mean-time, all of the remaining packets from the previous call + have been acked (7*4000 = 28000). This function then returns 8000 as all + data passed in are already sent and it can't return any more than what + it got passed in. But we have 28000 bytes acked. We now store the + remaining 20000 in the handle->u.file.acked struct field to add up in + the next call. - If an application accidentally provides a NULL handle pointer to - the channel or sftp public functions, they now return an error - instead of segfaulting. - -- libssh2_channel_eof: clarify that it returns negative on errors + 3. sftp_write() gets called again, and now there's a backlogged 20000 + bytes to return as fine and that will get skipped from the beginning + of the buffer that is passed in. -- SFTP: keep the sftp error code as 32 bit +- sftp_write: polished and simplified - 'last_errno' holds to the error code from the SFTP protocol and - since that is 32 bits on the wire there's no point in using a - long for this internally which is larger on some platforms. - -- agent: make the code better deal with unexpected code flows + Removed unnecessary struct fields and state changes within the function. - agent->ops gets initialized by the libssh2_agent_connect() call - but we need to make sure that we don't segfault even if a bad - sequence of function calls is used. + Made the loop that checks for ACKs only check chunks that were fully + sent. -Alexander Lamaison (10 Jun 2010) -- Better handling of invalid key files. +- SCP: on failure, show the numerical error reason - Passing an invalid public key to libssh2_userauth_publickey_fromfile_ex - triggered an assertion. Replaced this with a runtime check that rejects - obviously invalid key data. + By calling libssh2_session_last_errno() -Daniel Stenberg (10 Jun 2010) -- version: we start working on 1.2.7 now +- SFTP: provide the numerical error reason on failure -Version 1.2.6 (10 Jun 2010) +- SCP: clean up failure treatment + + When SCP send or recv fails, it gets a special message from the server + with a warning or error message included. We have no current API to + expose that message but the foundation is there. Removed unnecessary use + of session struct fields. -Daniel Stenberg (10 Jun 2010) -- NEWS: add the 1.2.6 release details +- sftp_write: enlarge buffer to perform better -- RELEASE-NOTES: 1.2.6 details added +- packets: code cleanup + + I added size checks in several places. I fixed the code flow to be easier + to read in some places. + + I removed unnecessary zeroing of structs. I removed unused struct fields. -Guenter Knauf (10 Jun 2010) -- fixed libssh2.dsw to use the generated libssh2.dsp; removed old *.dsp files. +- LIBSSH2_CALLBACK_MACERROR: clarify return code use -- moved MSVC strdup define to libssh2_config.h which we include already. +- _libssh2_userauth_publickey: avoid shadowing -- added missing source files to src/NMakefile. +- packet: avoid shadowing global symbols -Daniel Stenberg (8 Jun 2010) -- libssh2_poll: refer to poll(3) and select(3) instead +- sftp_readdir: avoid shadowing -- example: fix strdup() for MSVC compiles - - MSVC has a _strdup() that we better use. This was reported in bug +- shadowing: don't shadow the global compress -- SFTP: fail init SFTP if session isn't authenticated - - Alexander Lamaison filed bug #172 - (http://trac.libssh2.org/ticket/172), and pointed out that SFTP - init would do bad if the session isn't yet authenticated at the - time of the call, so we now check for this situation and returns - an error if detected. Calling sftp_init() at this point is bad - usage to start with. +- _libssh2_packet_add: turn ifs into a single switch -- direct_tcpip: bring back inclusion of libssh2_config.h - - In order to increase portability of this example, I'm bringing - the inclusion of libssh2_config.h back, and I also added an - require that header for this example to compile. +- _libssh2_packet_add: check SSH_MSG_GLOBAL_REQUEST packet + +- _libssh2_packet_add: SSH_MSG_DEBUG length checks - I also made all code lines fit within 80 columns. + Verify lengths before using them. Read always_display from the correct + index. Don't copy stuff around just to provide zero-termination of the + strings. -Guenter Knauf (3 Jun 2010) -- cast away a warning. +- _libssh2_packet_add: SSH_MSG_IGNORE skip memmove + + There's no promise of a zero termination of the data in the callback so + no longer perform ugly operation in order to provide it. -- moved CRT_SECURE_NO_DEPRECATE define up so its defined before the winsock headers are included. +- _libssh2_packet_add: SSH_MSG_DISCONNECT length checks + + Verify lengths before trying to read data. -- fixed platform detection for MingW32 test makefile. +- indent: break lines at 80 columns -- MingW32 has gettimeofday() implemented, so proper ifdef this function here. +- SSH_MSG_CHANNEL_OPEN_FAILURE: used defined values + + We don't like magic numbers in the code. Now the acceptable failure + codes sent in the SSH_MSG_CHANNEL_OPEN_FAILURE message are added as + defined values in the private header file. -- removed MSVC ifdef since seems we can use __int64 still with latest headers. +- sftp_write: don't return EAGAIN if no EAGAIN was received + + This function now only returns EAGAIN if a lower layer actually returned + EAGAIN to it. If nothing was acked and no EAGAIN was received, it will + now instead return 0. -- changed copyright notice for MinW32 and NetWare binaries. +- _libssh2_wait_socket: detect nothing-to-wait-for + + If _libssh2_wait_socket() gets called but there's no direction set to + wait for, this causes a "hang". This code now detects this situation, + set a 1 second timeout instead and outputs a debug output about it. -- cleaned up MSVC ifdefs which where spreaded over 3 places. +- decomp: remove the free_dest argument + + Since the decompress function ALWAYS returns allocated memory we get a + lot simpler code by removing the ability to return data unallocated. -- added uint8_t typedef for NetWare CLIB platform. +- decomp: cleaned off old compression stuff + + I cleared off legacy code from when the compression and decompression + functions were a single unified function. Makes the code easier to read + too. -- if the function declaration gets changed the header should be changed too. +- [TJ Saunders brought this change] -- this is MSVC specific and doesnt apply for all Win32 compilers; - the uint8_t typedef clashes with MingW32 headers. + decomp: increase decompression buffer sizes -- updated MingW32 makefiles for latest dependency lib versions. +- [TJ Saunders brought this change] -- updated NetWare makefiles for latest dependency lib versions. + zlib: Add debug tracing of zlib errors -Dan Fandrich (30 May 2010) -- Fixed compiling with libgcrypt +- sftp_packet_read: handle partial reads of the length field - A change of parameter types from unsigned long to size_t was - missed in the prototype in libgcrypt.h + SFTP packets come as [32 bit length][payload] and the code didn't + previously handle that the initial 32 bit field was read only partially + when it was read. -Daniel Stenberg (28 May 2010) -- statvfs: use libssh2_sftp_statvfs only, no "_ex" +- [Jasmeet Bagga brought this change] + + kex_agree_hostkey: fix NULL pointer derefence - As the long-term goal is to get rid of the extensive set of - macros from the API we can just as well start small by not adding - new macros when we add new functions. Therefore we let the - function be libssh2_sftp_statvfs() plainly without using an _ex - suffix. + While setting up the session, ssh tries to determine the type of + encryption method it can use for the session. This requires looking at + the keys offered by the remote host and comparing these with the methods + supported by libssh2 (rsa & dss). To do this there is an iteration over + the array containing the methods supported by libssh2. - I also made it use size_t instead of unsigned int for the string - length as that too is a long-term goal for the API. + If there is no agreement on the type of encryption we come to the 3rd + entry of the hostkeyp array. Here hostkeyp is valid but *hostkep is + NULL. Thus when we dereference that in (*hostkeyp)->name there is a + crash -- [Grubsky Grigory brought this change] +- _libssh2_transport_send: remove dead assignment + + 'data' isn't accessed beyond this point so there's no need to assign it. - DSP: output lib name typo +- scp_recv: remove dead assignment + + Instead of assigning a variable we won't read, we now use the more + explicit (void) prefix. -- [Grubsky Grigory brought this change] +- sftp_write: removed superfluous assignment - win32: provide a uint8_t typedef for better building on windows +- bugfix: avoid use of uninitialized value -- agent: win32: fix bad _libssh2_store_str call +- sftp_packet_require: propagate error codes better - As pointed out by Grubsky Grigory , I - made a mistake when I added the _libssh2_store_str() call before - and I made a slightly different patch than what he suggested. - Based purely on taste. - -Peter Stuge (24 May 2010) -- [Joey Degges brought this change] + There were some chances that they would cause -1 to get returned by + public functions and as we're hunting down all such occurances and since + the underlying functions do return valuable information the code now + passes back proper return codes better. - Add libssh2_sftp_statvfs() and libssh2_sftp_fstatvfs() - - These can be used to get file system statistics from servers that - support the statvfs@openssh.com and fstatvfs@openssh.com extensions. +- [Alfred Gebert brought this change] -Alexander Lamaison (22 May 2010) -- [Jose Baars brought this change] + fix memory leaks (two times cipher_data) for each sftp session - VMS specific: make sure final release can be installed over daily build +- libssh2_userauth_authenticated: make it work as documented + + The man page clearly says it returns 1 for "already authenticated" but + the code said non-zero. I changed the code to use 1 now, as that is also + non-zero but it gets the benefit that it now matches the documentation. + + Using 1 instead of non-zero is better for two reasons: + + 1. We have the opportunity to introduce other return codes in the future for + things like error and what not. + 2. We don't expose the internal bitmask variable value. -- [Jose Baars brought this change] +- userauth_keyboard_interactive: fix indent - VMS: small improvement to the man2help utilities +- [Alfred Gebert brought this change] -Peter Stuge (22 May 2010) -- [Joey Degges brought this change] + fix memory leak in userauth_keyboard_interactive() + + First I wanted to free the memory in session_free() but then + I had still memory leaks because in my test case the function + userauth_keyboard_interactive() is called twice. It is called + twice perhaps because the server has this authentication + methods available: publickey,gssapi-with-mic,keyboard-interactive + The keyboard-interactive method is successful. - libssh2_exit and libssh2_sftp_readdir man page fixes +- dist: include sftp.h in dist archives -Daniel Stenberg (21 May 2010) -- spelling: s/sue/use +Simon Josefsson (27 Oct 2010) +- Update header to match new function prototype, see c48840ba88. -Alexander Lamaison (21 May 2010) -- Change magic port number for generic knownhost check. - - libssh2_knownhost_checkp took 0 as a magic port number that indicated - a 'generic' check should be performed. However, 0 is a valid port - number in its own right so this commit changes the magic value to any - negative int. +Daniel Stenberg (26 Oct 2010) +- bugfixes: the transport rearrange left some subtle flaws now gone -Mikhail Gusarov (5 May 2010) -- Add re-discovered copyright holders to COPYING +- libssh2_userauth_publickey_fromfile_ex.3: cleaned up looks -- Restoring copyright statements from pre-git era +- libssh2_userauth_publickey: add man page - Eli Fant has contributed fragmenting SFTP requests + I found an undocumented public function and we can't have it like + that. The description here is incomplete, but should serve as a template + to allow filling in... -- Restoring my copyright statements from pre-git era +- libssh2_sftp_write.3: added blurb about the "write ahead" - keyboard_interactive, 'exit-status' information packet, non-atomic read/write - under FreeBSD, multi-channel operation bugfixes. - -Daniel Stenberg (3 May 2010) -- pedantic: make the code C90 clean + Documented the new SFTP write concept -Peter Stuge (3 May 2010) -- Do proper keyboard-interactive user dialog in the sftp.c example +- sftp_close_handle: free any trailing write chunks -Daniel Stenberg (3 May 2010) -- added to tarball: libssh2_knownhost_checkp.3 +- _libssh2_channel_write: fix warnings -- knownhost: support [host]:port in knownhost file +- SFTP: bufgix, move more sftp stuff to sftp.h - OpenSSH has ways to add hosts to the knownhosts file that include - a specific port number which makes the key associated with only - that specific host+port pair. libssh2 previously did not support - this, and I was forced to add a new function to the API to - properly expose this ability to applications: - libssh2_knownhost_checkp() + The sftp_write function shouldn't assume that the buffer pointer will be + the same in subsequent calls, even if it assumes that the data already + passed in before haven't changed. - To *add* such hosts to the knownhosts file, you make sure to pass - on the host name in that manner to the libssh2_knownhost_addc() - function. + The sftp structs are now moved to sftp.h (which I forgot to add before) -- init/exit: mention these were added in 1.2.5 +- SFTP: use multiple outgoing packets when writing + + sftp_write was rewritten to split up outgoing data into multiple packets + and deal with the acks in a more asynchronous manner. This is meant to + help overcome latency and round-trip problems with the SFTP protocol. -- libssh2_knownhost_check docs: correct the prototype +- TODO: implemented a lot of the ideas now -- examples: avoid use of uninitialized variable 'sock' +- _libssh2_channel_write: removed 32500 size limit + + Neither _libssh2_channel_write nor sftp_write now have the 32500 size + limit anymore and instead the channel writing function now has its own + logic to send data in multiple calls until everything is sent. -- KEX: stop pretending we negotiate language +- send_existing: don't tell parent to return when drained - There was some stub-like parts of an implementation for - implementing kex language negotiation that caused clang-analyzer - to warn and as it did nothing I've now removed the dead code. + That will just cause unnecessary code execution. -- Uninitialized argument +- _libssh2_channel_write: general code cleanup + + simplified the function and removed some unused struct fields -- sftpdir: removed dead assignment +- _libssh2_transport_send: replaces _libssh2_transport_write + + The new function takes two data areas, combines them and sends them as a + single SSH packet. This allows several functions to allocate and copy + less data. + + I also found and fixed a mixed up use of the compression function + arguments that I introduced in my rewrite in a recent commit. -- Makefile.am: include the VMS-specific config header as well +- scp_write_nonblock: use select() instead of busyloop + + Make this example nicer by not busylooping. -- [Jose Baars brought this change] +- send_existing: clear olen when the data is sent off - Add VMS specific libssh2_config.h +- _libssh2_transport_write: allow 256 extra bytes around the packet -- fix Value stored to 's' is never read warning - - and moved variable declaration of s to be more local +- _libssh2_transport_write: remade to send without malloc -- kexinit: simplify the code and avoid scan-build warning +- compress: compression disabled by default - Previously it would say "Value stored to 's' is never read" due - fourth increment of 's'. - -Alexander Lamaison (28 Apr 2010) -- Removed unecessary brackets. + We now allow libssh2_session_flag() to enable compression with a new + flag and I added documentation for the previous LIBSSH2_FLAG_SIGPIPE + flag which I wasn't really aware of! -- Changed sftp_attrsize macro to a static function. +- comp: split the compress function + + It is now made into two separate compress and decompress functions. In + preparation for upcoming further modficications. -Daniel Stenberg (28 Apr 2010) -- release: include the VMS-specific files +Dan Fandrich (20 Oct 2010) +- Added header file to allow compiling in older environments -- sftp_attrsize: protect the macro argument with proper parentheses +Daniel Stenberg (20 Oct 2010) +- TODO: add a possible new API for SFTP transfers -- ssh2_agent: avoid using 'session' uninitialized on failures +- TODO: "New Transport API" added -- examples: remove assignments of variable rc that's never used +- TODO: add buffering plans -- publickey_init: remove useless variable increment +Simon Josefsson (13 Oct 2010) +- Mention libssh2_channel_get_exit_signal and give kudos. -- hostkey_method_ssh_rsa_init: remove useless variable increment +- [Tommy Lindgren brought this change] -- packet_x11_open: removed useless variable increment + Add libssh2_channel_get_exit_signal man page. - and made the declaration of a variable more local + Signed-off-by: Simon Josefsson -- packet_queue_listener: removed useless variable increment - - and made the declaration of a variable more local +- [Tommy Lindgren brought this change] -- sftp_read: move a read_responses array to where its used + Add libssh2_channel_get_exit_signal. - I find that this increases readability since the array is used - only in the function call just immediately below and nowhere - else. - -- sftp_readdir: turn a small array static const and move it + Signed-off-by: Simon Josefsson -- sftp_attrsize: converted function to a macro - - This way, the macro can evaluate a static number at compile time - for two out of four uses, and it probably runs faster for the - other two cases too. +- Add libssh2_free man page and fix typo. -- sftp_open: deal with short channel_write calls - - This was an old TODO that just wasn't done before. If - channel_write returns short, that is not an error. +- Add libssh2_free. -- sftp_open: clean up, better check of input data - - The clang-analyzer report made it look into this function and - I've went through it to remove a potential use of an - uninitialized variable and I also added some validation of input - data received from the server. +Daniel Stenberg (11 Oct 2010) +- scp_recv: improved treatment of channel_read() returning zero - In general, lots of more code in this file need to validate the - input before assuming it is correct: there are servers out there - that have bugs or just have another idea of how to do the SFTP - protocol. + As a zero return code from channel_read() is not an error we must make + sure that the SCP functions deal with that properly. channel_read() + always returns 0 if the channel is EOFed already so we check for EOF + after 0-reads to be able to return error properly. -- bugfix: avoid using the socket if it failed to create one +- libssh2_session_methods.3: detail what can be asked for -- bugfix: potential use of NULL pointer +- compression: send zlib before none + + As the list of algorithms in a preferred order we should send zlib + before none to increase the chances that the server will let us do + compression. -- libssh2_userauth_password_ex: clarify errors somewhat +- compress: faster check, better return codes - The errors mentioned in this man page are possible return codes - but not necessarily the only return codes that this can return. + In the transport functions we avoid a strcmp() now and just check a + boolean instead. - Also reformatted the typ prototypes somewhat. + The compress/decompress function's return code is now acknowledged and + used as actual return code in case of failures. -- examples: fixed and made them more similar +- libssh2_session_handshake: replaces libssh2_session_startup() - The channel read/write functions can return 0 in legitimate cases - without it being an error, and we need to loop properly if they - return short. - -- [Jose Baars brought this change] + The function libssh2_session_startup() is now considered deprecated due + to the portability issue with the socket argument. + libssh2_session_handshake() is the name of the replacement. - VMS port of libssh2; changes in the libssh2 common code +- libssh2_socket_t: now externally visible + + In preparation for upcominig changes, the libssh2_socket_t type is now + typedef'ed in the public header. -- Makefile: added the two news headers userauth.h and session.h +- _libssh2_transport_drain: removed + + This function proved not to be used nor useful. -- cleanup: prefer the internal functions +- _libssh2_channel_write: don't iterate over transport writes - To get the blocking vs non-blocking to work as smooth as possible - and behave better internally, we avoid using the external - interfaces when calling functions internally. + When a call to _libssh2_transport_write() succeeds, we must return from + _libssh2_channel_write() to allow the caller to provide the next chunk + of data. - Renamed a few internal functions to use _libssh2 prefix when not - being private within a file, and removed the libssh2_ for one - that was private within the file. + We cannot move on to send the next piece of data that may already have + been provided in this same function call, as we risk getting EAGAIN for + that and we can't return information both about sent data as well as + EAGAIN. So, by returning short now, the caller will call this function + again with new data to send. -- session_free: remove dead code +- _libssh2_transport_write: updated documentation blurb -- libssh2_publickey_init: fixed to work better non-blocking - - This was triggered by a clang-analyzer complaint that turned out - to be valid, and it made me dig deeper and fix some generic non- - blocking problems I disovered in the code. +- _libssh2_transport_write: remove fprintf remainder - While cleaning this up, I moved session-specific stuff over to a - new session.h header from the libssh2_priv.h header. + Mistake from previous debugging -- channel: reduce duplicated free and returns +- session: improved errors - Simplified the code by trying to free data and return on a single - spot. + Replaced -1/SOCKET_NONE errors with appropriate error defines instead. + + Made the verbose trace output during banner receiving less annoying for + non-blocking sessions. -- channel: make variables more local +- crypt_init: use correct error define + +- _libssh2_error: hide EAGAIN for non-blocking sessions - By making 'data' and 'data_len' more local in several places in - this file it will be easier to spot how they are used and we'll - get less risks to accidentally do bad things with them. + In an attempt to make the trace output less cluttered for non-blocking + sessions the error function now avoids calling the debug function if the + error is the EAGAIN and the session is non-blocking. -Mikhail Gusarov (24 Apr 2010) -- Fix typos in manpages, catched by Lintian +- agent: use better error defines -Daniel Stenberg (24 Apr 2010) -- channel_request_pty: simplify the code +- comp_method_zlib_init: use correct error defines + +- transport: better error codes - clang-analyzer pointed out how 'data' could be accessed as a NULL - pointer if the wrong state was set, and while I don't see that - happen in real-life the code flow is easier to read and follow by - moving the LIBSSH2_FREE() call into the block that is supposed to - deal with the data pointer anyway. + LIBSSH2_SOCKET_NONE (-1) should no longer be used as error code as it is + (too) generic and we should instead use specific and dedicated error + codes to better describe the error. -- libssh2_channel_process_startup: simplify the code +- channel: return code and _libssh2_error cleanup - clang-analyzer pointed out how 'data' could be accessed as a NULL - pointer if the wrong state was set, and while I don't see that - happen in real-life the code flow is easier to read and follow by - moving the LIBSSH2_FREE() call into the block that is supposed to - deal with the data pointer anyway. + Made sure that all transport_write() call failures get _libssh2_error + called. -- sftp_close_handle: add precation to not access NULL pointer +- _libssh2_channel_write: limit to 32700 bytes - clang-analyzer pointed this out as a "Pass-by-value argument in - function call is undefined" but while I can't see exactly how - this can ever happen in reality I think a little check for safety - isn't such a bad thing here. + The well known and used ssh server Dropbear has a maximum SSH packet + length at 32768 by default. Since the libssh2 design current have a + fixed one-to-one mapping from channel_write() to the packet size created + by transport_write() the previous limit of 32768 in the channel layer + caused the transport layer to create larger packets than 32768 at times + which Dropbear rejected forcibly (by closing the connection). + + The long term fix is of course to remove the hard relation between the + outgoing SSH packet size and what the input length argument is in the + transport_write() function call. -- scp_write_nonblock: Value stored to 'nread' is never read +- libssh.h: add more dedicated error codes -- scp_write: Value stored to 'ptr' is never read +- SCP: allow file names with bytes > 126 + + When parsing the SCP protocol and verifying that the data looks like a + valid file name, byte values over 126 must not be consider illegal since + UTF-8 file names will use such codes. + + Reported by: Uli Zappe + Bug: http://www.libssh2.org/mail/libssh2-devel-archive-2010-08/0112.shtml -- scp_write_nonblock: Value stored to 'ptr' is never read +Dan Fandrich (25 Aug 2010) +- Document the three sftp stat constants -- sftp_mkdir: less silly output but show failures +Guenter Knauf (18 Aug 2010) +- Fixed Win32 makefile which was now broken at resource build. -- [Jose Baars brought this change] +- It is sufficient to pipe stderr to NUL to get rid of the nasty messages. - VMS port of libssh2 including VMS specific build procedures +- [Author: Guenter Knauf brought this change] -- two variable types changes, made lines less than 80 columns + Removed Win32 ifdef completely for sys/uio.h. - The two variable type changes are only to match type variable - fields actually read from the binary protocol. + No idea why we had this ifdef at all but MSVC, MingW32, Watcom + and Borland all have no sys/uio.h header; so if there's another + Win32 compiler which needs it then it should be added explicitely + instead of this negative list. -- remove check for negative padding_length +- New files should also be added to Makefile.am. - It was silly, since it is read as an unsigned char... + Otherwise they will never be included with release and snapshot tarballs ... -- hostkey_method_ssh_dss_init: Value stored to 's' is never read +Daniel Stenberg (18 Aug 2010) +- version: bump to 1.2.8_DEV -- libssh2_banner_set: avoid unnecessary increment and explain code +Version 1.2.7 (17 Aug 2010) -- agent_transact_unix: remove unused variable +Daniel Stenberg (17 Aug 2010) +- release: updated to hold 1.2.7 info -- remove two unnecessary increments +Guenter Knauf (17 Aug 2010) +- Use the new libssh2.rc file. -- more code converted to use _libssh2_store_*() +- Added resource file for libssh2.dll (shamelessly stolen from libcurl). -- libssh2_publickey_list_fetch: removed unused variables +- Updated Win32 MSVC dependencies versions. -- libssh2_publickey_init: remove unused variables +- Added include for sys/select.h to get fd.set on some platforms. -- libssh2_scp_send64: added to API to provide large file transfers +- Added Watcom makefile borrowed from libcurl. - The previously existing libssh2_scp_send_ex() function has no way - to send files that are larger than 'size_t' which on 32bit - systems mean 4GB. This new API uses a libssh2_int64_t type and - should thus on most modern systems be able to send enormous - files. - -- sftp_init: remove unused variables and assignments - -- libssh2_knownhost_check: Value stored to 'keylen' is never read - -- hostkey: fix compiler warning + This makefile compiles already all files fine for static lib, but needs + final touch when I have OpenSSL fully working with shared libs and Watcom. -- remove unused variable +- Added copyright define to libssh2.h and use it for binary builds. -- data types: convert more to use size_t and uint32_t +- Moved version defines up in order to include from .rc file. + + Blocked rest of header with ifndef so its possible to let + the rc compiler only use the version defines. -- channel: variable type cleanups +- Some minor makefile tweaks. -- cleanups: better binary packet gen, size_t fixes and PACKET_* removal - - I'll introduce a new internal function set named - - _libssh2_store_u32 - _libssh2_store_u64 - _libssh2_store_str - - That can be used all through the library to build binary outgoing - packets. Using these instead of the current approach removes - hundreds of lines from the library while at the same time greatly - enhances readability. I've not yet fully converted everything to - use these functions. - - I've converted LOTS of 'unsigned long' to 'size_t' where - data/string lengths are dealt with internally. This is The Right - Thing and it will help us make the transition to our - size_t-polished API later on as well. - - I'm removing the PACKET_* error codes. They were originally - introduced as a set of separate error codes from the transport - layer, but having its own set of errors turned out to be very - awkward and they were then converted into a set of #defines that - simply maps them to the global libssh2 error codes instead. Now, - I'l take the next logical step and simply replace the PACKET_* - defines with the actual LIBSSH2_ERROR_* defines. It will increase - readability and decrease confusion. +Daniel Stenberg (2 Aug 2010) +- example: treat the libssh2_channel_read() return code properly - I also separated packet stuff into its own packet.h header file. + A short read is not an error. Only negative values are errors! -- clarified the return code +- libssh2_wait_socket: reset error code to "leak" EAGAIN less + + Since libssh2 often sets LIBSSH2_ERROR_EAGAIN internally before + _libssh2_wait_socket is called, we can decrease some amount of + confusion in user programs by resetting the error code in this function + to reduce the risk of EAGAIN being stored as error when a blocking + function returns. -- rename libssh2_error to the correct _libssh2_error +- _libssh2_wait_socket: poll needs milliseconds - We reserve ^libssh2_ for public symbols and we use _libssh2 as - prefix for internal ones. I fixed the intendation of all these - edits with emacs afterwards, which then changed it slightly more - than just _libssh2_error() expressions but I didn't see any - obvious problems. + As reported on the mailing list, the code path using poll() should + multiple seconds with 1000 to get milliseconds, not divide! + + Reported by: Jan Van Boghout -- data type cleanup: made lots of code use size_t etc +- typedef: make ssize_t get typedef without LIBSSH2_WIN32 - A lot of code used 'unsigned long' and the likes when it should - rather just use plain 'int' or use size_t for data lengths. + The condition around the ssize_t typedef depended on both LIBSSH2_WIN32 + *and* _MSC_VER being defined when it should be enough to depend on + _MSC_VER only. It also makes it nicer so libssh2-using code builds fine + without having custom defines. -- wait_socket: make c89 compliant and use two fd_sets for select() +- [John Little brought this change] -- sftp_readdir: always zero terminate, detail the return code - - I also added a description for the 'longentry' field which was - previously undocumented! + session_free: free more data to avoid memory leaks -- sftp_readdir: simplified and bugfixed - - This function no longer has any special purpose code for the - single entry case, as it was pointless. - - The previous code would overflow the buffers with an off-by-one - in case the file name or longentry data fields received from the - server were exactly as long as the buffer provided to - libssh2_sftp_readdir_ex. +- channel_free: ignore problems with channel_close() - We now make sure that libssh2_sftp_readdir_ex() ALWAYS zero - terminate the buffers it fills in. + As was pointed out in bug #182, we must not return failure from + _libssh2_channel_free() when _libssh2_channel_close() returns an error + that isn't EAGAIN. It can effectively cause the function to never go + through, like it did now in the case where the socket was actually + closed but socket_state still said LIBSSH2_SOCKET_CONNECTED. - The function no longer calls the libssh2_* function again, but - properly uses the internal sftp_* instead. + I consider this fix the right thing as it now also survives other + errors, even if making sure socket_state isn't lying is also a good + idea. -- channel/transport: we now drain the outgoing send buffer when we ignore EAGAIN +- publickey_list_free: no return value from a void function - When we ignore the EAGAIN from the transport layer within channel_write, we - now drain the outgoing transport layer buffer so that remainders in that - won't cause any problems in the next invoke of _libssh2_transport_write() + Fixed a compiler warning I introduced previously when checking input + arguments more. I also added a check for the other pointer to avoid NULL + pointer dereferences. -- channel_write: if data has been sent, don't return EAGAIN - - When sending data in a loop, we must not return EAGAIN if we - managed to send data in an earlier round. This was reported in - bug #126 => http://libssh2.stuge.se/ticket/126 +- [Lars Nordin brought this change] -Simon Josefsson (14 Apr 2010) -- Fix OpenSSL AES-128-CTR detection. + openssl: make use of the EVP interface - Patch from Paul Howarth . - -Daniel Stenberg (13 Apr 2010) -- version in header file now says 1.2.6-DEV - -- 1.2.6: clean the RELEASE-NOTES for next release round - -- NEWS: add the stuff from the version 1.2.5 RELEASE-NOTES + Make use of the EVP interface for the AES-funktion. Using this method + supports the use of different ENGINES in OpenSSL for the AES function + (and the direct call to the AES_encrypt should not be used according to + openssl.org) -Version 1.2.5 (13 Apr 2010) +Peter Stuge (23 Jun 2010) +- [Tor Arntsen brought this change] -Daniel Stenberg (13 Apr 2010) -- channel_close: no longer wait for the SSH_MSG_CHANNEL_CLOSE message + Don't overflow MD5 server hostkey - As the packet may simply not arrive we cannot have the close - function wait for it unconditionally. + Use SHA_DIGEST_LENGTH and MD5_DIGEST_LENGTH in memcpy instead of hardcoded + values. An incorrect value was used for MD5. -- less code duplication in the poll vs select code flows +- Fix message length bugs in libssh2_debug() + + There was a buffer overflow waiting to happen when a debug message was + longer than 1536 bytes. - libssh2_keepalive_send and libssh2_session_block_directions are - now used outside of the #ifdef blocks. + Thanks to Daniel who spotted that there was a problem with the message + length passed to a trace handler also after commit + 0f0652a3093111fc7dac0205fdcf8d02bf16e89f. -- make it C90 compliant +- Make libssh2_debug() create a correctly terminated string + + Also use FILE *stderr rather than fd 2, which can very well be something + completely different. -- updated with all changes and bugs since 1.2.4 +Daniel Stenberg (23 Jun 2010) +- [TJ Saunders brought this change] -- Added LIBSSH2_SFTP_S_IS***() macros and updated docs + handshake: Compression enabled at the wrong time - libssh2_sftp_fstat_ex.3 is now extended quite a lot to describe a - lot of the struct and the bits it uses and how to test for them. - -- sftp_init() deal with _libssh2_channel_write() returns short + In KEXINIT messages, the client and server agree on, among other + things, whether to use compression. This method agreement occurs + in src/kex.c's kex_agree_methods() function. However, if + compression is enabled (either client->server, server->client, or + both), then the compression layer is initialized in + kex_agree_methods() -- before NEWKEYS has been received. - When _libssh2_channel_write() is asked to send off 9 bytes, the - code needs to deal with the situation where less than 9 bytes - were sent off and prepare to send the remaining piece at a later - time. - -- handle a NULL password as if it was "" + Instead, the initialization of the compression layer should + happen after NEWKEYS has been received. This looks to occur + insrc/kex.c's diffie_hellman_sha1(), which even has the comment: - libssh2_userauth_publickey_fromfile_ex() takes a "passphrase" - but didn't deal with it being set to NULL. - -- Reduce used window sizes by factor 10 + /* The first key exchange has been performed, - As reported in bug report #166 http://libssh2.stuge.se/ticket/166 - by 'ptjm', the maximum window size must be less crazy for libssh2 - to do better with more server implementations. I did not do any - testing to see how this changes raw SCP performance, but the - maximum window size is still almost 4MB. This also has the upside - that libssh2 will use less memory. - -Peter Stuge (28 Mar 2010) -- Correctly clear blocking flag after sending multipart packet + switch to active crypt/comp/mac mode */ - 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. + There, after NEWKEYS is received, the cipher and mac algorithms + are initialized, and that is where the compression should be + initialized as well. - Clear the direction flag after the final part of a packet has been sent. - -Daniel Stenberg (24 Mar 2010) -- Added man page for libssh2_knownhost_addc() + The current implementation fails if server->client compression is + enabled because most server implementations follow OpenSSH's + lead, where compression is initialized after NEWKEYS. Since the + server initializes compression after NEWKEYS, but libssh2 + initializes compression after KEXINIT (i.e. before NEWKEYS), they + are out of sync. - Added mention in libssh2_knownhost_add() docs that - libssh2_knownhost_addc() is the preferred function now. - -- at next soname bump remove libssh2_knownhost_add() + Reported in bug report #180 -- ignore TAGS ("make tags" makes them) +- [TJ Saunders brought this change] -- fix memory leak + userauth_hostbased_fromfile: packet length too short + + The packet length calculated in src/userauth.c's + userauth_hostbased_fromfile() function is too short by 4 bytes; + it forgets to add four bytes for the length of the hostname. + This causes hostbased authentication to fail, since the server + will read junk data. - we must not assign the pointer a NULL since it keeps allocated - data and at least parts of an error string + verified against proftpd's mod_sftp module -- fixed the pattern for avoiding the poll check +- _libssh2_userauth_publickey: reject method names longer than the data + + This functions get the method length by looking at the first 32 + bit of data, and I now made it not accept method lengths that are + longer than the whole data set is, as given in the dedicated + function argument. - added some comments about known problems with poll on darwin + This was detected when the function was given bogus public key + data as an ascii string, which caused the first 32bits to create + a HUGE number. -- avoid checking for poll on some systems +- NULL resistance: make more public functions survive NULL pointer input - darwin and interix are known to have broken poll implementations - so we skip the check on those and thus have them use select - unconditionally + Sending in NULL as the primary pointer is now dealt with by more + public functions. I also narrowed the userauth.c code somewhat to + stay within 80 columns better. -- ignore libssh2.dsp +- agent: make libssh2_agent_userauth() work blocking properly + + previously it would always work in a non-blocking manner -Simon Josefsson (23 Mar 2010) -- Fix logic in "on-the-fly" crypto init. +Peter Stuge (17 Jun 2010) +- Fix underscore typo for 64-bit printf format specifiers on Windows + + Commit 49ddf447ff4bd80285f926eac0115f4e595f9425 was missing underscores. -- Make sure keepalive is working even when poll is used. +Daniel Stenberg (16 Jun 2010) +- libssh2_session_callback_set: extended the man page -- [Paul Querna brought this change] +- [John brought this change] - Use poll when available on blocking API. + LIBSSH2_DEBUG: macro uses incorrect function variable - Signed-off-by: Simon Josefsson + The LIBSSH2_DEBUG macro, defined in libssh2_priv.h, incorrectly uses the + function variable ssh_msg_disconnect when it should use ssh_msg_debug. + + This shows that the LIBSSH2_CALLBACK_DEBUG callback never has worked... -Peter Stuge (20 Mar 2010) -- Fix speling +- warning: fix a compiler warning 'pointer differs in signedness' + + As reported in bug #177 -Daniel Stenberg (19 Mar 2010) -- fix NULL dereference when window adjusting a non-existing channel +- portability: introduce LIBSSH2_INT64_T_FORMAT for 64bit printf()s - Suyog Jadhav pointed out that when receiving a window adjust to - a channel not found, the code would reference a NULL pointer. - Now it will instead output a message about that fact. + As pointed out in bug #177, some of the Windows compilers use + %I64 to output 64 bit variables with the printf family. -Simon Josefsson (19 Mar 2010) -- Fix build problem. +- debug: avoid sending NULL to sprintf %s + + Via the _libssh2_debug() macro/function. Pointed out by john in bug report -- Eat our own dog food, call libssh2_init and libssh2_exit in the examples. +- sftp docs: show macro on macro page, only function on function page + + The individual man pages for macros now show the full convenience + macro as defined, and then the man page for the actual function + only shows the function. -- Fix init/exit logic. Add self-test of it. +- code police: make the code use less than 80 columns -Daniel Stenberg (19 Mar 2010) -- fix typo +- libssh2_channel_write_ex: remove macros, added wording on buffer size -Simon Josefsson (19 Mar 2010) -- Add man page for libssh2_init and libssh2_exit. Fix libssh2_exit prototype. +- libssh2_sftp_write: document buffer size and changed some ordering -- Shorten constant a bit. More documentation. +- libssh2_channel_write_stderr: show how the macro is defined -- Fix namespace pollution. +- libssh2_channel_write: show how the macro is defined -- Add global init/exit points, to do crypto initialization in one place. +- SFTP: limit write() to not produce overly large packets + + sftp_write() now limits how much data it gets at a time even more + than before. Since this function creates a complete outgoing + packet based on what gets passed to it, it is crucial that it + doesn't create too large packets. - By Lars Nordin. + With this method, there's also no longer any problem to use very + large buffers in your application and feed that to libssh2. I've + done numerous tests now with uploading data over SFTP using 100K + buffers and I've had no problems with that. -Daniel Stenberg (14 Mar 2010) -- libssh2 is released under the Modifed BSD license, not GPL +- scp_write_nonblock: add transfer time info + + Using the same timing logic and output format as + sftp_write_nonblock allows us to very easily run benchmarks on + SCP vs SFTP uploads using libssh2. -Alexander Lamaison (14 Mar 2010) -- Add libssh2_knownhost_addc to handle comments. +- sftp_write_nonblock: select() on socket, use *BIG* buffer, time transfer - Comments in known_hosts file were not handle properly. They were parsed as - part of the key causing key matching to return a mismatch if the entry had a - comment. This adds a new API function that takes an optional comment and - changes libssh2_knownhost_readline to parse the comment as pass it to the - new function. + The select() is just to make it nicer so that it doesn't + crazy-loop on EAGAIN. The buffer size thing is mostly to verify + that this really work as supposed. - Fixes #164. + Transfer timing is just a minor thing, but it can just as well be + there and help us time and work on performance easier using out + of the box examples. -- Fix gettimeofday to compile with Visual C++ 6. +- agent: use _libssh2_error() when returning errors - Reported by Steven Van Ingelgem. - -Simon Josefsson (10 Mar 2010) -- Add. - -- keepalive.c: Fix libssh2_error usage. - -- Fix typo in last commit. - -- Tidy up build option notice. - -- Add entry about keep alive stuff. - -- Add keep-alive support. - -Alexander Lamaison (7 Mar 2010) -- Untabify. - -- Fix memory leak in libssh2_knownhost_add. - -Daniel Stenberg (6 Mar 2010) -- change 'int' to 'libssh2_socket_t' in the public API for sockets - -- reduce code duplication and return underlying error better + As pointed out in bug report #173, this module basically never + used _libssh2_error() which made it work inconstently with other + parts of the libssh2 code base. This is my first take at making + this code more in line with the rest. -- acknowledge when _libssh2_packet_requirev() returns error +- inputchecks: make lots of API functions check for NULL pointers - when _libssh2_packet_requirev() returns an error when waiting for - SSH_MSG_USERAUTH_SUCCESS or SSH_MSG_USERAUTH_FAILURE, it is an - error and it should be treated as such - -- wrap long lines - -- polished the phrasing in two error strings - -- silence picky compiler warnings + If an application accidentally provides a NULL handle pointer to + the channel or sftp public functions, they now return an error + instead of segfaulting. -- silence picky compiler warnings +- libssh2_channel_eof: clarify that it returns negative on errors -- removed libssh2_error()'s forth argument +- SFTP: keep the sftp error code as 32 bit - libssh2_error() no longer allocates a string and only accepts a const - error string. I also made a lot of functions use the construct of - return libssh2_error(...) instead of having one call to - libssh2_error() and then a separate return call. In several of those - cases I then also changed the former -1 return code to a more - detailed one - something that I think will not change behaviors - anywhere but it's worth keeping an eye open for any such. + 'last_errno' holds to the error code from the SFTP protocol and + since that is 32 bits on the wire there's no point in using a + long for this internally which is larger on some platforms. -- repaired --enable-debug +- agent: make the code better deal with unexpected code flows + + agent->ops gets initialized by the libssh2_agent_connect() call + but we need to make sure that we don't segfault even if a bad + sequence of function calls is used. -Simon Josefsson (1 Mar 2010) -- Make ./configure output a summary of build options. +Alexander Lamaison (10 Jun 2010) +- Better handling of invalid key files. + + Passing an invalid public key to libssh2_userauth_publickey_fromfile_ex + triggered an assertion. Replaced this with a runtime check that rejects + obviously invalid key data. -Daniel Stenberg (1 Mar 2010) -- let the err_msg in the session struct be const too +Daniel Stenberg (10 Jun 2010) +- version: we start working on 1.2.7 now -Simon Josefsson (1 Mar 2010) -- Revert #ifdef change that pulled in AES-CTR code when explicitly disabled. +Version 1.2.6 (10 Jun 2010) -Daniel Stenberg (1 Mar 2010) -- fix #ifdefs +Daniel Stenberg (10 Jun 2010) +- NEWS: add the 1.2.6 release details -- make function match the new proto +- RELEASE-NOTES: 1.2.6 details added -Simon Josefsson (1 Mar 2010) -- Improve AES-CTR check. +Guenter Knauf (10 Jun 2010) +- fixed libssh2.dsw to use the generated libssh2.dsp; removed old *.dsp files. -Daniel Stenberg (1 Mar 2010) -- use const to silence a bazillion warnings +- moved MSVC strdup define to libssh2_config.h which we include already. -Simon Josefsson (1 Mar 2010) -- Use AES-CTR from OpenSSL when available. - - Reported by Lars Nordin . +- added missing source files to src/NMakefile. -- Make it possible to disable DSA. - - Patch from Lars Nordin . +Daniel Stenberg (8 Jun 2010) +- libssh2_poll: refer to poll(3) and select(3) instead -Peter Stuge (1 Mar 2010) -- Send and receive channel EOF before sending SSH_MSG_CHANNEL_CLOSE - - Sending SSH_MSG_CHANNEL_CLOSE without channel EOF is explicitly allowed - in RFC 4254, but some non-conforming servers will hang or time out when - the channel is closed before EOF. +- example: fix strdup() for MSVC compiles - Other common clients send and receive EOF before closing, there are no - drawbacks, and some servers need it to work correctly. + MSVC has a _strdup() that we better use. This was reported in bug -Alexander Lamaison (26 Feb 2010) -- Style improvements to knownhost error handling. +- SFTP: fail init SFTP if session isn't authenticated - Made improvements as suggested by Peter Stuge: http://www.libssh2.org/mail/libssh2-devel-archive-2010-02/0161.shtml. + Alexander Lamaison filed bug #172 + (http://trac.libssh2.org/ticket/172), and pointed out that SFTP + init would do bad if the session isn't yet authenticated at the + time of the call, so we now check for this situation and returns + an error if detected. Calling sftp_init() at this point is bad + usage to start with. -- Call libssh2_error for every knownhost API failure. - - The libssh2 API calls should set the last error code and a message when - returning a failure by calling libssh2_error. This changeset adds these - calls to the libssh2_knownhost_* API as well as libssh2_base64_decode. +- direct_tcpip: bring back inclusion of libssh2_config.h - This change also makes libssh2_error into a function rather than a macro. - Its implementation is moved to misc.c. This function returns the error - code passed to it allowing callers to return the error value directly - without duplicating the error code. - -- Fix LIBSSH2_ALLOC checks. + In order to increase portability of this example, I'm bringing + the inclusion of libssh2_config.h back, and I also added an + require that header for this example to compile. - These appear to be cut-and paste errors where the wrong variable is checked - for NULLness after calling LIBSSH2_ALLOC. - -Simon Josefsson (23 Feb 2010) -- Silence compiler warning. + I also made all code lines fit within 80 columns. -- Make it portable; test uses = for string comparison (not ==). Indent. +Guenter Knauf (3 Jun 2010) +- cast away a warning. -Alexander Lamaison (22 Feb 2010) -- libssh2_knownhost_del: fix write to freed memory. - - When removing a known host, libssh2_knownhost_del would remove the node from the linked list, free its memory and then overwrite the struct parameter (which indicated which node to remove) with 0. However, this struct is actually allocated within the just-freed node meaning we're writing to freed memory. This made Windows very upset. - - The fix is simply to overwrite the struct first before freeing the memory. +- moved CRT_SECURE_NO_DEPRECATE define up so its defined before the winsock headers are included. -Daniel Stenberg (21 Feb 2010) -- show more verbose error when SCP send fails +- fixed platform detection for MingW32 test makefile. -- libssh2_socket_t is done, a library-free function is needed +- MingW32 has gettimeofday() implemented, so proper ifdef this function here. -- clarify that this frees all data associated with a session +- removed MSVC ifdef since seems we can use __int64 still with latest headers. -- improved error handling +- changed copyright notice for MinW32 and NetWare binaries. -- add missing libssh2_error() calls - - To make sure the public API is functional and that the - BLOCK_ADJUST_ERRNO() macro works correctly we MUST make sure to - call libssh2_error() when we return errors. +- cleaned up MSVC ifdefs which where spreaded over 3 places. -- fix memory leak in userauth_keyboard_interactive() - - Mr anonymous in bug #125 pointed out that the userauth_keyboard_interactive() - function does in fact assign the same pointer a second time to a new allocated - buffer without properly freeing the previous one, which caused a memory leak. +- added uint8_t typedef for NetWare CLIB platform. -- added missing error codes - - To allow the libssh2_session_last_error() function to work as - documented, userauth_password() now better makes sure to call - libssh2_error() everywhere before it returns error. - - Pointed out by mr anonymous in bug #128 +- if the function declaration gets changed the header should be changed too. -Peter Stuge (16 Feb 2010) -- Fix resource and memory leaks in examples as reported by cppcheck - - Thanks to Ettl Martin for the report and patch. This fixes #132 +- this is MSVC specific and doesnt apply for all Win32 compilers; + the uint8_t typedef clashes with MingW32 headers. -Daniel Stenberg (15 Feb 2010) -- mention the new man pages for macros +- updated MingW32 makefiles for latest dependency lib versions. -- added man pages for API macros - - all #defined macros in the public headers are considered to be part - of the API and I've generated individual man pages for each of them - to A) make it easier to figure out what each function/macro actually - is for so that automated lookups work better and for B) make sure we - have all public functions document (both macros and functions) to - make it easier for us to work away from all the macros in a future - release. +- updated NetWare makefiles for latest dependency lib versions. -- Committed the patch by Yoichi Iwaki in bug #2929647 +Dan Fandrich (30 May 2010) +- Fixed compiling with libgcrypt - Committed the patch by Yoichi Iwaki in bug #2929647, which fixed a memory - leak when an 'outbuf' was still allocated when a session was freed. + A change of parameter types from unsigned long to size_t was + missed in the prototype in libgcrypt.h -- free "outbuf" when killing a session +Daniel Stenberg (28 May 2010) +- statvfs: use libssh2_sftp_statvfs only, no "_ex" - Fix memoary leak: if there was an "output" still allocated when a - session was torn down it needs to be freed in session_free() + As the long-term goal is to get rid of the extensive set of + macros from the API we can just as well start small by not adding + new macros when we add new functions. Therefore we let the + function be libssh2_sftp_statvfs() plainly without using an _ex + suffix. - Patch by Yoichi Iwaki in bug #2929647 - -- the working version name is now 1.2.5_DEV + I also made it use size_t instead of unsigned int for the string + length as that too is a long-term goal for the API. -Version 1.2.4 (13 Feb 2010) +- [Grubsky Grigory brought this change] -Daniel Stenberg (13 Feb 2010) -- updated info for 1.2.4 + DSP: output lib name typo -Dan Fandrich (10 Feb 2010) -- Allow compiling with OpenSSL when AES isn't available. +- [Grubsky Grigory brought this change] -Peter Stuge (9 Feb 2010) -- [Dave McCaldon brought this change] + win32: provide a uint8_t typedef for better building on windows - Fix Tru64 socklen_t compile issue with example/direct_tcpip.c - - Building libssh2-1.2.3 on Tru64 fails at line 48 and 166 because socklen_t - isn't defined on Tru64 unless _POSIX_PII_SOCKET is defined. +- agent: win32: fix bad _libssh2_store_str call - This patch updates configure.ac to add -D_POSIX_PII_SOCKET when building - on Tru64 platform(s). + As pointed out by Grubsky Grigory , I + made a mistake when I added the _libssh2_store_str() call before + and I made a slightly different patch than what he suggested. + Based purely on taste. -- [Dave McCaldon brought this change] +Peter Stuge (24 May 2010) +- [Joey Degges brought this change] - Resolve compile issues on Solaris x64 and UltraSPARC - - Solaris builds of libssh2-1.2.3 failed on both x64 and UltraSPARC - platforms because of two problems: + Add libssh2_sftp_statvfs() and libssh2_sftp_fstatvfs() - 1) src/agent.c:145 sun is a reserved word when using the SUNWspro compiler - 2) example/direct_tcpip.c:84 INADDR_NONE is not defined + These can be used to get file system statistics from servers that + support the statvfs@openssh.com and fstatvfs@openssh.com extensions. -Daniel Stenberg (3 Feb 2010) -- towards 1.2.4 now +Alexander Lamaison (22 May 2010) +- [Jose Baars brought this change] -Version 1.2.3 (3 Feb 2010) + VMS specific: make sure final release can be installed over daily build -Daniel Stenberg (3 Feb 2010) -- Version 1.2.3 (February 3, 2010) +- [Jose Baars brought this change] -- fix building out of source tree by proving better include path - - when building out of source tree, we provide -I$(top_builddir)/example - since the libssh2_config.h gets generated in that dir + VMS: small improvement to the man2help utilities -Peter Stuge (1 Feb 2010) -- [Sofian Brabez brought this change] +Peter Stuge (22 May 2010) +- [Joey Degges brought this change] - Replace : in hexdump with " " (two spaces) + libssh2_exit and libssh2_sftp_readdir man page fixes -- Detect when the forwarded connection is closed in example/direct_tcpip.c +Daniel Stenberg (21 May 2010) +- spelling: s/sue/use -- Fix example/direct_tcpip.c to work also on WIN32 +Alexander Lamaison (21 May 2010) +- Change magic port number for generic knownhost check. - read() and write() are no good for WIN32 sockets, use recv() and send(). + libssh2_knownhost_checkp took 0 as a magic port number that indicated + a 'generic' check should be performed. However, 0 is a valid port + number in its own right so this commit changes the magic value to any + negative int. -- Ignore libssh2_config.h.in and stamp-h2 in example/ and remove .cvsignore +Mikhail Gusarov (5 May 2010) +- Add re-discovered copyright holders to COPYING -- Simplify WIN32 ifdefs in example/direct_tcpip.c to allow standalone compile +- Restoring copyright statements from pre-git era + + Eli Fant has contributed fragmenting SFTP requests -- Always #define INVALID_SOCKET -1 in libssh2_priv.h when not on win32 +- Restoring my copyright statements from pre-git era - Fix broken builds since commit abd9bd0bbe631efeada1f54552c70b54e1c490c1 - for all non-win32 platforms. + keyboard_interactive, 'exit-status' information packet, non-atomic read/write + under FreeBSD, multi-channel operation bugfixes. -- Include hmac-md5 and hmac-md5-96 only if crypto backend supports MD5 +Daniel Stenberg (3 May 2010) +- pedantic: make the code C90 clean -- Use LIBSSH2_HOSTKEY_HASH_SHA1 instead of _MD5 in examples and tests - - MD5 support is optional and may not always be available, while SHA1 is both - required and recommended. +Peter Stuge (3 May 2010) +- Do proper keyboard-interactive user dialog in the sftp.c example -- Update mailing list address in configure.ac to @cool.haxx.se +Daniel Stenberg (3 May 2010) +- added to tarball: libssh2_knownhost_checkp.3 -- Make example/direct_tcpip.c compile for win32 +- knownhost: support [host]:port in knownhost file + + OpenSSH has ways to add hosts to the knownhosts file that include + a specific port number which makes the key associated with only + that specific host+port pair. libssh2 previously did not support + this, and I was forced to add a new function to the API to + properly expose this ability to applications: + libssh2_knownhost_checkp() - One warning from FD_SET() remains, it is also in some other examples. + To *add* such hosts to the knownhosts file, you make sure to pass + on the host name in that manner to the libssh2_knownhost_addc() + function. -- Correctly check for an invalid socket in session_startup() +- init/exit: mention these were added in 1.2.5 -- Small documentation fix after Dave's _USERAUTH_FAILURE improvement +- libssh2_knownhost_check docs: correct the prototype -- [Dave McCaldon brought this change] +- examples: avoid use of uninitialized variable 'sock' - Handle SSH_MSG_USERAUTH_FAILURE for password and kbd-int authentication - - Neither libssh2_userauth_password_ex() nor - libssh2_userauth_keyboard_interactive_ex() would return a login failure - error if the server responded with a SSH_MSG_USERAUTH_FAILURE, instead - you would see whatever previous error had occurred, typically - LIBSSH2_ERROR_EAGAIN. +- KEX: stop pretending we negotiate language - This patch changes error code -18 to LIBSSH2_ERROR_AUTHENTICATION_FAILED - and makes LIBSSH2_ERROR_PUBLICKEY_UNRECOGNIZED an alias for - LIBSSH2_ERROR_AUTHENTICATION_FAILED. In addition, new logic in - userauth_password() properly handles SSH_MSG_USERAUTH_FAILURE and both - this function and userauth_keyboard_interactive() now properly return - LIBSSH2_ERROR_AUTHENTICATION_FAILED. + There was some stub-like parts of an implementation for + implementing kex language negotiation that caused clang-analyzer + to warn and as it did nothing I've now removed the dead code. -Simon Josefsson (28 Jan 2010) -- Fix. +- Uninitialized argument -- Also deal with GLOBAL_REQUEST keep-alives. +- sftpdir: removed dead assignment -- Make OpenSSH-style keepalive work against libssh2 clients. +- Makefile.am: include the VMS-specific config header as well -Daniel Stenberg (27 Jan 2010) -- clarified +- [Jose Baars brought this change] -Peter Stuge (26 Jan 2010) -- [Dave McCaldon brought this change] + Add VMS specific libssh2_config.h - Fix trace context lookup in libssh2_debug() - - The trace context is actually a bitmask so that tracing output can be - controlled by setting a bitmask using libssh2_trace(). However, the logic - in libssh2_debug() that converted the context to a string was using the - context value as an array index. Because the code used a bounds check on - the array, there was never a danger of a crash, but you would certainly - either get the wrong string, or "unknown". +- fix Value stored to 's' is never read warning - This patch adds a lookup that iterates over the context strings and uses - it's index to check for the corresponding bit in the context. + and moved variable declaration of s to be more local -- Fix typo in RELEASE-NOTES +- kexinit: simplify the code and avoid scan-build warning + + Previously it would say "Value stored to 's' is never read" due + fourth increment of 's'. -Daniel Stenberg (20 Jan 2010) -- updated for 1.2.3 with all the stuff I found in the log +Alexander Lamaison (28 Apr 2010) +- Removed unecessary brackets. -- ignore more generated files +- Changed sftp_attrsize macro to a static function. -- [Dave McCaldon brought this change] +Daniel Stenberg (28 Apr 2010) +- release: include the VMS-specific files - Pass user context through libssh2_trace_sethandler() to callback - - The libssh2_trace_sethandler() call allows the user to handle the output of libssh2 rather than having it written to stderr. This patch updates libssh2_trace_sethandler() to allow a user-defined void* context value to be passed back to the output handler. +- sftp_attrsize: protect the macro argument with proper parentheses -- [Dave McCaldon brought this change] +- ssh2_agent: avoid using 'session' uninitialized on failures - Add libssh2_trace_sethandler() to the API (even more) +- examples: remove assignments of variable rc that's never used -- [Dave McCaldon brought this change] +- publickey_init: remove useless variable increment - Add libssh2_trace_sethandler() to the API +- hostkey_method_ssh_rsa_init: remove useless variable increment -- cleanup includes +- packet_x11_open: removed useless variable increment - We now produce a local libssh2_config.h file in this dir for the - examples to use so I cleaned up the include path at the same time. + and made the declaration of a variable more local -- generate a libssh2_config.h in the example dir +- packet_queue_listener: removed useless variable increment - buildconf copies the template to example/ and configure makes sure - to generate a proper file from it and the direct_tcpip.c example - is the first one to use it - to make sure it builds fine on more - paltforms - -Simon Josefsson (13 Jan 2010) -- Remove redundant #includes and reorder sys/types.h include. - -Daniel Stenberg (10 Jan 2010) -- avoid a free(NULL) + and made the declaration of a variable more local -Simon Josefsson (7 Jan 2010) -- Make it simpler to get more debug info. +- sftp_read: move a read_responses array to where its used + + I find that this increases readability since the array is used + only in the function call just immediately below and nowhere + else. -Daiki Ueno (1 Jan 2010) -- Simplify the commit 63457dfa using type cast from size_t * to ulong *. +- sftp_readdir: turn a small array static const and move it -Alexander Lamaison (30 Dec 2009) -- Fixed memory leak in userauth_publickey(). +- sftp_attrsize: converted function to a macro - userauth_publickey_fromfile() reads the key from a - file using file_read_publickey() which returns two - allocated strings, the decoded key and the key - method (such as "ssh-dss"). The latter can be - derived from the former but returning both avoids a - later allocation while doing so. + This way, the macro can evaluate a static number at compile time + for two out of four uses, and it probably runs faster for the + other two cases too. + +- sftp_open: deal with short channel_write calls - Older versions of userauth_publickey_fromfile() used - this method string directly but when - userauth_publickey() was factored out of - userauth_publickey_fromfile() it derived the method - from the key itself. This resulted in the method - being allocated twice. + This was an old TODO that just wasn't done before. If + channel_write returns short, that is not an error. + +- sftp_open: clean up, better check of input data - This fix, which maintains the optimisation that - avoids an extra allocation, changes - userauth_publickey() so it doesn't allocate and - derive the method when userauth_pblc_method already - has a value. + The clang-analyzer report made it look into this function and + I've went through it to remove a potential use of an + uninitialized variable and I also added some validation of input + data received from the server. - Signed-off-by: Alexander Lamaison + In general, lots of more code in this file need to validate the + input before assuming it is correct: there are servers out there + that have bugs or just have another idea of how to do the SFTP + protocol. -Daiki Ueno (25 Dec 2009) -- Fix the return value description of libssh2_knownhost_free(). +- bugfix: avoid using the socket if it failed to create one -- Fix compiler warnings for size_t pointers on 32-bit Windows. +- bugfix: potential use of NULL pointer -- Define INVALID_SOCKET and use it instead of SOCKET_BAD. +- libssh2_userauth_password_ex: clarify errors somewhat - Revert the part of previous commit that defines SOCKET_BAD library wide. - -- Use libssh2_socket_t in the ssh-agent stuff. - Define a portability macro SOCKET_BAD which means "invalid socket". - -- Mark/unmark connection to Pageant is open/close. - -- Add test to check if the socket is connected. - -Peter Stuge (24 Dec 2009) -- Add libssh2.pc to top-level .gitignore - -- Fix publickey authentication regression + The errors mentioned in this man page are possible return codes + but not necessarily the only return codes that this can return. - Commit 70b199f47659a74b8778c528beccf893843e5ecb introduced a parsing - bug in file_read_publickey() which made the algorithm name contain an - extra trailing space character, breaking all publickey authentication. - -- Add a direct-tcpip example which shows local port forwarding - -- Add session parameter and LIBSSH2_TRACE_SOCKET to libssh2_trace(3) man page - -- Add TODO: Expose error messages sent by the server - -Daiki Ueno (23 Dec 2009) -- Fix doc comments. + Also reformatted the typ prototypes somewhat. -- Add man pages for ssh-agent API. +- examples: fixed and made them more similar + + The channel read/write functions can return 0 in legitimate cases + without it being an error, and we need to loop properly if they + return short. -- Don't request userauthlist after authentication. +- [Jose Baars brought this change] -Simon Josefsson (21 Dec 2009) -- Add. + VMS port of libssh2; changes in the libssh2 common code -- [Daiki Ueno brought this change] +- Makefile: added the two news headers userauth.h and session.h - Add an example to use ssh-agent API. +- cleanup: prefer the internal functions - Signed-off-by: Simon Josefsson - -- [Daiki Ueno brought this change] - - Add ssh-agent API. + To get the blocking vs non-blocking to work as smooth as possible + and behave better internally, we avoid using the external + interfaces when calling functions internally. - Signed-off-by: Simon Josefsson + Renamed a few internal functions to use _libssh2 prefix when not + being private within a file, and removed the libssh2_ for one + that was private within the file. -- [Daiki Ueno brought this change] +- session_free: remove dead code - Add callback-based API for publickey auth. +- libssh2_publickey_init: fixed to work better non-blocking - Signed-off-by: Simon Josefsson - -- Move examples from example/simple to example/. - -- Move examples from example/simple to example/. - -Daniel Stenberg (17 Dec 2009) -- _libssh2_list_insert() fixed to work + This was triggered by a clang-analyzer complaint that turned out + to be valid, and it made me dig deeper and fix some generic non- + blocking problems I disovered in the code. - While this is code not currently in use, it is part of the generic linked - list code and since I found the error I thought I'd better fix it since we - might bring in this function into the code one day. + While cleaning this up, I moved session-specific stuff over to a + new session.h header from the libssh2_priv.h header. -Simon Josefsson (16 Dec 2009) -- Silence compiler warnings. +- channel: reduce duplicated free and returns - Based on patch by Kamil Dudka in - . - -- [Kamil Dudka brought this change] + Simplified the code by trying to free data and return on a single + spot. - libgcrypt: simplify code of _libssh2_dsa_sha1_sign +- channel: make variables more local - Signed-off-by: Simon Josefsson + By making 'data' and 'data_len' more local in several places in + this file it will be easier to spot how they are used and we'll + get less risks to accidentally do bad things with them. -- [Kamil Dudka brought this change] +Mikhail Gusarov (24 Apr 2010) +- Fix typos in manpages, catched by Lintian - libgcrypt: follow-up for ssh-dss padding fix +Daniel Stenberg (24 Apr 2010) +- channel_request_pty: simplify the code - Signed-off-by: Simon Josefsson - -Dan Fandrich (15 Dec 2009) -- Check for the right environment variable in the test app + clang-analyzer pointed out how 'data' could be accessed as a NULL + pointer if the wrong state was set, and while I don't see that + happen in real-life the code flow is easier to read and follow by + moving the LIBSSH2_FREE() call into the block that is supposed to + deal with the data pointer anyway. -Simon Josefsson (14 Dec 2009) -- Silence warning about unused function parameter. +- libssh2_channel_process_startup: simplify the code - Reported by Steven Van Ingelgem . + clang-analyzer pointed out how 'data' could be accessed as a NULL + pointer if the wrong state was set, and while I don't see that + happen in real-life the code flow is easier to read and follow by + moving the LIBSSH2_FREE() call into the block that is supposed to + deal with the data pointer anyway. -Daniel Stenberg (10 Dec 2009) -- avoid returning data to memory already freed +- sftp_close_handle: add precation to not access NULL pointer - In case of failure we must make sure that the data we return - doesn't point to a memory area already freed. Reported anonymously - in the bug report #2910103. + clang-analyzer pointed this out as a "Pass-by-value argument in + function call is undefined" but while I can't see exactly how + this can ever happen in reality I think a little check for safety + isn't such a bad thing here. -Peter Stuge (8 Dec 2009) -- Use LIBSSH2_TRACE_* internally and remove redundant LIBSSH2_DBG_* +- scp_write_nonblock: Value stored to 'nread' is never read -- Add LIBSSH2_TRACE_SOCKET context for tracing send() and recv() - - Helpful in debugging the -39 errors. +- scp_write: Value stored to 'ptr' is never read -- Another transport layer fix for bogus -39 (LIBSSH2_ERROR_BAD_USE) errors - - Commit 683aa0f6b52fb1014873c961709102b5006372fc made send_existing() send - more than just the second part of a packet when the kernel did not accept - the full packet, but the function still overlooked the SSH protocol - overhead in each packet, often 48 bytes. - - If only the last few bytes of a packet remained, then the packet would - erroneously be considered completely sent, and the next call to write - more data in the session would return a -39 error. +- scp_write_nonblock: Value stored to 'ptr' is never read -Daniel Stenberg (6 Dec 2009) -- move local variable to be more localized +- sftp_mkdir: less silly output but show failures -- fixed some indent mistakes +- [Jose Baars brought this change] -Peter Stuge (6 Dec 2009) -- Fix padding in ssh-dss signature blob encoding - - DSA signatures consist of two 160-bit integers called r and s. In ssh-dss - signature blobs r and s are stored directly after each other in binary - representation, making up a 320-bit (40 byte) string. (See RFC4253 p14.) - - The crypto wrappers in libssh2 would either pack r and s incorrectly, or - fail, when at least one integer was small enough to be stored in 19 bytes - or less. - - The patch ensures that r and s are always stored as two 160 bit numbers. + VMS port of libssh2 including VMS specific build procedures -- Don't always clear write direction blocking flag +- two variable types changes, made lines less than 80 columns - When libssh2_transport_write() is called to continue sending a - partially sent packet the write direction flag must not be cleared - until the previous packet has been completely sent, or the app would - hang if the packet still isn't sent completely, since select() gets - called by the internal blocking emulation layer in libssh2 but would - then not be watching the socket for writability. + The two variable type changes are only to match type variable + fields actually read from the binary protocol. + +- remove check for negative padding_length - Clear the flag only once processing of previous packet data is - complete and a new packet is about to be prepared. + It was silly, since it is read as an unsigned char... -Alexander Lamaison (24 Nov 2009) -- Detabify. +- hostkey_method_ssh_dss_init: Value stored to 's' is never read -- [Daniel Stenberg brought this change] +- libssh2_banner_set: avoid unnecessary increment and explain code + +- agent_transact_unix: remove unused variable - Fixed memory leak in sftp_fstat(). +- remove two unnecessary increments -Simon Josefsson (17 Nov 2009) -- Mark date of 1.2.2 release. +- more code converted to use _libssh2_store_*() -- Merge branch 'master' of ssh://git.stuge.se/var/lib/git/libssh2 +- libssh2_publickey_list_fetch: removed unused variables -Version 1.2.2 (16 Nov 2009) +- libssh2_publickey_init: remove unused variables -Daniel Stenberg (16 Nov 2009) -- prepared for 1.2.2 +- libssh2_scp_send64: added to API to provide large file transfers + + The previously existing libssh2_scp_send_ex() function has no way + to send files that are larger than 'size_t' which on 32bit + systems mean 4GB. This new API uses a libssh2_int64_t type and + should thus on most modern systems be able to send enormous + files. -Simon Josefsson (16 Nov 2009) -- Improve NEWS items. +- sftp_init: remove unused variables and assignments -- Support AES-Counter ciphers. +- libssh2_knownhost_check: Value stored to 'keylen' is never read -- Silence compiler warning. - - Reported by Steven Van Ingelgem - in . +- hostkey: fix compiler warning -- cgit v1.2.3