--- - branch: pkgsrc-2016Q4 date: Thu Jan 19 19:55:17 UTC 2017 files: - new: 1.167.4.1 old: '1.167' path: pkgsrc/security/gnutls/Makefile pathrev: pkgsrc/security/gnutls/Makefile@1.167.4.1 type: modified - new: 1.53.4.1 old: '1.53' path: pkgsrc/security/gnutls/PLIST pathrev: pkgsrc/security/gnutls/PLIST@1.53.4.1 type: modified - new: 1.121.4.1 old: '1.121' path: pkgsrc/security/gnutls/distinfo pathrev: pkgsrc/security/gnutls/distinfo@1.121.4.1 type: modified - new: '0' old: '1.1' path: pkgsrc/security/gnutls/patches/patch-tests_mini-server-name.c pathrev: pkgsrc/security/gnutls/patches/patch-tests_mini-server-name.c@0 type: deleted id: 20170119T195517Z.bc6d30e4401dea9bd8e06bb077c4123c1aa03fa8 log: "Pullup ticket #5185 - requested by wiz\nsecurity/gnutls: security fix\n\nRevisions pulled up:\n- security/gnutls/Makefile 1.168-1.169\n- security/gnutls/PLIST 1.54\n- security/gnutls/distinfo \ 1.122\n- security/gnutls/patches/patch-tests_mini-server-name.c \ deleted\n\n---\n Module Name:\tpkgsrc\n Committed By:\tmaya\n Date:\t\tSat Jan 7 18:49:16 UTC 2017\n\n Modified Files:\n \tpkgsrc/security/gnutls: Makefile\n\n \ Log Message:\n gnutls: don't redefine max_align_t on FreeBSD. It incorrectly fails the\n configure test because the type in stddef.h is guarded by a c11 macro\n (most likely).\n\n Force the configure test to pass.\n\n From David Shao in PR pkg/51793 (originally from FreeBSD ports).\n\n---\n Module Name:\tpkgsrc\n \ Committed By:\twiz\n Date:\t\tTue Jan 10 16:23:50 UTC 2017\n\n Modified Files:\n \tpkgsrc/security/gnutls: Makefile PLIST distinfo\n Removed Files:\n \ \tpkgsrc/security/gnutls/patches: patch-tests_mini-server-name.c\n\n Log Message:\n Updated gnutls to 3.5.8.\n\n * Version 3.5.8 (released 2016-01-09)\n\n \ ** libgnutls: Ensure that multiple calls to the gnutls_set_priority_*\n functions will not leave the verification profiles field to an\n undefined state. The last call will take precedence.\n\n ** libgnutls: Ensure that GNUTLS_E_DECRYPTION_FAIL will be returned\n by PKCS#8 decryption functions when an invalid key is provided. This\n addresses regression on decrypting certain PKCS#8 keys.\n\n ** libgnutls: Introduced option to override the default priority string\n used by the library. The intention is to allow support of system-wide\n priority strings (as set with --with-system-priority-file). The\n configure option is --with-default-priority-string.\n\n \ ** libgnutls: Require a valid IV size on all ciphers for PKCS#8 decryption.\n \ This prevents crashes when decrypting malformed PKCS#8 keys.\n\n ** libgnutls: Fix crash on the loading of malformed private keys with certain\n parameters set to zero.\n\n ** libgnutls: Fix double free in certificate information printing. If the PKIX\n extension proxy was set with a policy language set but no policy specified,\n that could lead to a double free.\n\n ** libgnutls: Addressed memory leaks in client and server side error paths\n (issues found using oss-fuzz project)\n\n ** libgnutls: Addressed memory leaks in X.509 certificate printing error paths\n (issues found using oss-fuzz project)\n\n ** libgnutls: Addressed memory leaks and an infinite loop in OpenPGP certificate\n parsing. Fixes by Alex Gaynor. (issues found using oss-fuzz project)\n\n ** libgnutls: Addressed invalid memory accesses in OpenPGP certificate parsing.\n (issues found using oss-fuzz project)\n\n ** API and ABI modifications:\n No changes since last version.\n\n * Version 3.5.7 (released 2016-12-8)\n\n ** libgnutls: Include CHACHA20-POLY1305 ciphersuites in the SECURE128\n and SECURE256 priority strings.\n\n \ ** libgnutls: Require libtasn1 4.9; this ensures gnutls will correctly\n operate with OIDs which have elements that exceed 2^32.\n\n ** libgnutls: The DN decoding functions output the traditional DN format\n rather than the strict RFC4514 compliant textual DN. This reverts the\n 3.5.6 introduced change, and allows applications which depended on the\n previous format to continue to function. Introduced new functions which\n output the strict format by default, and can revert to the old one using\n a flag.\n\n ** libgnutls: Improved TPM key handling. Check authorization requirements\n prior to using a key and fix issue on loop for PIN input. Patches by\n James Bottomley.\n\n ** libgnutls: In all functions accepting UTF-8 passwords, ensure that\n passwords are normalized according to RFC7613. When invalid UTF-8\n passwords are detected, they are only tolerated for decryption.\n This introduces a libunistring dependency on GnuTLS. A version of\n libunistring is included in the library for the platforms that do\n not ship it; it can be used with the '--with-included-unistring'\n \ option to configure script.\n\n ** libgnutls: When setting a subject alternative name in a certificate\n which is in UTF-8 format, it will transparently be converted to IDNA form\n prior to storing.\n\n ** libgnutls: GNUTLS_CRT_PRINT_ONELINE flag on gnutls_x509_crt_print()\n will print the SHA256 key-ID instead of a certificate fingerprint.\n\n ** libgnutls: enhance the PKCS#7 verification capabilities. In the case\n signers that are not discoverable using the trust list or input, use\n the stored list as pool to generate a trusted chain to the signer.\n\n ** libgnutls: Improved MTU calculation precision for the CBC ciphersuites\n under DTLS.\n\n ** libgnutls: [added missing news entry since 3.5.0]\n No longer tolerate certificate key usage violations for\n TLS signature verification, and decryption. That is GnuTLS will fail\n to connect to servers which incorrectly use a restricted to signing certificate\n for decryption, or vice-versa. This reverts the lax behavior introduced\n in 3.1.0, due to several such broken servers being available. The %COMPAT\n priority keyword can be used to work-around connecting on these servers.\n\n ** certtool: When exporting a CRQ in DER format ensure no text data are\n intermixed. Patch by Dmitry Eremin-Solenikov.\n\n ** certtool: Include the SHA-256 variant of key ID in --certificate-info\n options.\n\n ** p11tool: Introduced the --initialize-pin and --initialize-so-pin\n options.\n\n ** API and ABI modifications:\n gnutls_utf8_password_normalize: Added\n gnutls_ocsp_resp_get_responder2: Added\n gnutls_x509_crt_get_issuer_dn3: Added\n gnutls_x509_crt_get_dn3: Added\n gnutls_x509_rdn_get2: Added\n gnutls_x509_dn_get_str2: Added\n gnutls_x509_crl_get_issuer_dn3: Added\n gnutls_x509_crq_get_dn3: Added\n\n \ * Version 3.5.6 (released 2016-11-04)\n\n ** libgnutls: Enhanced the PKCS#7 parser to allow decoding old\n (pre-rfc5652) structures with arbitrary encapsulated content.\n\n ** libgnutls: Introduced a function group to set known DH parameters\n \ using groups from RFC7919.\n\n ** libgnutls: Added more strict RFC4514 textual DN encoding and decoding.\n Now the generated textual DN is in reverse order according to RFC4514,\n and functions which generate a DN from strings such gnutls_x509_crt_set_*dn()\n set the expected DN (reverse of the provided string).\n\n ** libgnutls: Introduced time and constraints checks in the end certificate\n in the gnutls_x509_crt_verify_data2() and gnutls_pkcs7_verify_direct()\n \ functions.\n\n ** libgnutls: Set limits on the maximum number of alerts handled. That is,\n applications using gnutls could be tricked into an busy loop if the\n peer sends continuously alert messages. Applications which set a maximum\n handshake time (via gnutls_handshake_set_timeout) will eventually recover\n but others may remain in a busy loops indefinitely. This is related but\n not identical to CVE-2016-8610, due to the difference in alert handling\n \ of the libraries (gnutls delegates that handling to applications).\n\n ** libgnutls: Reverted the change which made the gnutls_certificate_set_*key*\n functions return an index (introduced in 3.5.5), to avoid affecting programs\n which explicitly check success of the function as equality to zero. In order\n for these functions to return an index an explicit call to gnutls_certificate_set_flags\n \ with the GNUTLS_CERTIFICATE_API_V2 flag is now required.\n\n ** libgnutls: Reverted the behavior of sending a status request extension even\n without a response (introduced in 3.5.5). That is, we no longer reply to a\n client's hello with a status request, with a status request extension. Although\n that behavior is legal, it creates incompatibility issues with releases in\n the gnutls 3.3.x branch.\n\n ** libgnutls: Delayed the initialization of the random generator at\n the first call of gnutls_rnd(). This allows applications to load\n on systems which getrandom() would block, without blocking until\n \ real random data are needed.\n\n ** certtool: --get-dh-params will output parameters from the RFC7919\n groups.\n\n ** p11tool: improvements in --initialize option.\n\n ** API and ABI modifications:\n GNUTLS_CERTIFICATE_API_V2: Added\n \ GNUTLS_NO_TICKETS: Added\n gnutls_pkcs7_get_embedded_data_oid: Added\n gnutls_anon_set_server_known_dh_params: Added\n gnutls_certificate_set_known_dh_params: Added\n gnutls_psk_set_server_known_dh_params: Added\n gnutls_x509_crt_check_key_purpose: Added\n\n * Version 3.5.5 (released 2016-10-09)\n\n ** libgnutls: enhanced gnutls_certificate_set_ocsp_status_request_file()\n \ to allow importing multiple OCSP request files, one for each chain\n provided.\n\n \ ** libgnutls: The gnutls_certificate_set_key* functions return an\n index of the added chain. That index can be used either with\n gnutls_certificate_set_ocsp_status_request_file(), or with\n gnutls_certificate_get_crt_raw() and friends.\n\n ** libgnutls: Added SHA*, AES-GCM, AES-CCM and AES-CBC optimized implementations\n for the aarch64 architecture. Uses Andy Polyakov's assembly code.\n\n ** libgnutls: Ensure proper cleanups on gnutls_certificate_set_*key()\n failures due to key mismatch. This prevents leaks or double freeing\n on such failures.\n\n \ ** libgnutls: Increased the maximum size of the handshake message hash.\n This will allow the library to cope better with larger packets, as\n the ones offered by current TLS 1.3 drafts.\n\n ** libgnutls: Allow to use client certificates despite them containing\n disallowed algorithms for a session. That allows for example a client\n to use DSA-SHA1 due to his old DSA certificate, without requiring him\n to enable DSA-SHA1 (and thus make it acceptable for the server's certificate).\n\n ** libgnutls: Reverted AESNI code on x86 to earlier version as the\n latest version was creating position depending code. Added checks\n \ in the CI to detect position depending code early.\n\n ** guile: Update code to the I/O port API of Guile >= 2.1.4\n This makes sure the GnuTLS bindings will work with the forthcoming 2.2\n stable series of Guile, of which 2.1 is a preview.\n\n ** API and ABI modifications:\n gnutls_certificate_set_ocsp_status_request_function2: Added\n gnutls_session_ext_register: Added\n gnutls_session_supplemental_register: Added\n GNUTLS_E_PK_INVALID_PUBKEY: Added\n GNUTLS_E_PK_INVALID_PRIVKEY: Added\n" module: pkgsrc subject: 'CVS commit: [pkgsrc-2016Q4] pkgsrc/security/gnutls' unixtime: '1484855717' user: bsiegert