{"thread":{"id":"66821","subject":"[ANNOUNCE] Git 2.56.1 and friends","startedAt":"2026-11-10T18:03:38Z","lastAt":"2026-11-11T14:02:14Z","messageCount":4,"participants":["Junio C Hamano","Todd Zullinger","Kaartic Sivaraam","Toon Claes"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"558121","messageId":"xmqqh5hosgw5.fsf@gitster.g","threadId":"66821","inReplyTo":null,"subject":"[ANNOUNCE] Git 2.56.1 and friends","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2026-11-10T18:03:38Z","receivedAt":"2026-11-10T18:03:38Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"A maintenance release Git 2.56.1 and others for older maintenance\ntracks down to maint-2.43 are now available at the usual places.\n\nThis is a set of coordinated security fix releases.  Please update\nat your earliest convenience.\n\nThe tarballs are found at:\n\n    https://www.kernel.org/pub/software/scm/git/\n\nThe following public repositories all have a copy of the v2.50.2,\nv2.51.3, v2.52.1, v2.53.1, v2.54.1, v2.55.1, and v2.56.1 tags:\n\n  url = https://git.kernel.org/pub/scm/git/git\n  url = https://kernel.googlesource.com/pub/scm/git/git\n  url = git://repo.or.cz/alt-git.git\n  url = https://github.com/gitster/git\n\n\nCredit for organizing the set of these releases goes to Johannes\nSchindelin.  Credit for finding the vulnerabilities goes to Luke\nMartin and pwn2addr.  Credit for fixing them goes to Toon Claes,\nKarthik Nayak, and Kaartic Sivaraam.  The xdiff hardening is based\non a patch by Jeff King, with help from Toon Claes and Christian\nCouder.\n\n------------------------------------------------------------\nGit v2.56.1 Release Notes\n=========================\n\nThis release merges up the fix that appears in v2.50.2, v2.51.3,\nv2.52.1, v2.53.1, v2.54.1 and v2.55.1 to address CVE-2026-111760,\nCVE-2026-113408, CVE-2026-114466, CVE-2026-114467, CVE-2026-114469\nand CVE-2026-105437; see the release notes for these versions for\ndetails.\n\n\n------------------------------------------------------------\nGit v2.50.2 Release Notes\n=========================\n\nThis addresses the security issues CVE-2026-111760, CVE-2026-113408,\nCVE-2026-114466, CVE-2026-114467, CVE-2026-114469 and CVE-2026-105437.\n\nThis release also backports fixes necessary to let the CI builds pass\nsuccessfully.\n\nFixes since v2.50.1\n-------------------\n\n * CVE-2026-111760:\n\n   Recursive clones are susceptible to Remote Code Execution: an\n   attacker-controlled repository can be crafted in a way that exploits\n   a bug in Git whereby Git can be fooled into overwriting a gitfile\n   (used for submodules) with a symbolic link that then points to a\n   directory with malicious config or hooks. The config or hooks may\n   have attacker-specified code.\n\n   See <https://github.com/git/git/security/advisories/GHSA-vp2w-7xxj-jvxr>\n   for details.\n\n * CVE-2026-113408:\n\n   It is possible for a server to advertise an unborn HEAD that contains\n   path-traversal characters (e.g., `refs/heads/../../../TARGET`). With\n   the files ref backend that can lead to ref-updating actions to write\n   to unwanted files.\n\n   See <https://gitlab.com/gitlab-org/git/-/work_items/809> for details.\n\n * CVE-2026-114467:\n\n   When Git strips the GPG signatures from a commit message, it only\n   accounts for two signatures. A commit message containing more than\n   two signatures therefore causes a stack memory corruption.\n\n   This can be triggered when using `git merge` with a tag which\n   contains more than two signatures. As such, this makes way for a DoS\n   attack which can potentially also affect forges when such a tag is\n   pushed to a remote and a merge action is initiated.\n\n   See <https://gitlab.com/gitlab-org/git/-/work_items/810> for details.\n\n * CVE-2026-114469:\n\n   Git rejects patches larger than ~1 GiB, because its patch application\n   code does not work well with large files. That check, however, only\n   considered the size of the patch as it appears in the input, which for\n   binary patches is much smaller than the size of the file the patch\n   describes, as binary patches are compressed and encoded. A binary\n   patch that inflates to a very large size therefore slipped past the\n   check and crashed Git with an out-of-bounds read.\n\n   See <https://gitlab.com/gitlab-org/git/-/work_items/811> for details.\n\n * CVE-2026-105437:\n\n   When merging, Git computes the size of the buffer holding the merge\n   result in a way that can overflow for very large merges, which can be\n   triggered by a large number of conflicts combined with a large\n   `conflict-marker-size`. The resulting buffer can then be too small,\n   and Git writes past its end (and, before and after that, before its\n   start), corrupting the heap.\n\n   See <https://gitlab.com/gitlab-org/git/-/work_items/814> for details.\n\n * CVE-2026-114466:\n\n   In response to CVE-2019-1348, `fast-import` disallows the\n   `export-marks` feature by default. The features\n   `rewrite-submodules-to` and `rewrite-submodules-from` as well as the\n   option `export-pack-edges` that were introduced in the meantime also\n   allow reading from arbitrary paths, and are therefore now also\n   disallowed by default.\n\n   See <https://gitlab.com/gitlab-org/git/-/work_items/812> for details.\n\n * Defense-in-depth: read-cache: reject cache entries with over-long\n   names.\n\n * Various fix-ups on HTTP tests.\n\n * Test update.\n\n * HTTP Header redaction code has been adjusted for a newer version of\n   cURL library that shows its traces differently from earlier\n   versions.\n\n * The `imap-send` command was adjusted to work with OpenSSL 4.0.\n\n * Replace macos-13 used at GitHub CI with macos-14, and macos-14 with\n   macos-15.\n\n * Work around a bug in the iconv library shipped with macOS 14 and 15\n   that fails to correctly handle stateful ISO/IEC 2022 encoded strings.\n\n * ci: fix broken jobs on Ubuntu 25.10 caused by switch to sudo-rs(1).\n\n * ci(dockerized): raise the PID limit for private repositories.\n\n * ci: bump debian-11 job to debian-12.\n\n * ci: skip CVS and P4 tests in leaks job, too.\n\n * mingw: drop the -D_USE_32BIT_TIME_T option, stop using nedmalloc, and\n   only use -Wl,--large-address-aware for 32-bit builds.\n\n * Update GitHub Actions jobs to avoid warnings against using deprecated\n   version of Node.js.\n\n----------------------------------------------------------------\n\nChanges since v2.56.0 are as follows:\n\nBeat Bolli (3):\n      imap-send: use the OpenSSL API to access the subject alternative names\n      imap-send: use the OpenSSL API to access the subject common name\n      imap-send: move common code into function host_matches()\n\nColin Stagner (1):\n      ci: ubuntu: use GNU coreutils for dirname\n\nCollin Funk (4):\n      global: constify some pointers that are not written to\n      gpg-interface: remove an unnecessary NULL initialization\n      bloom: remove a misleading const qualifier\n      dir: avoid -Wdiscarded-qualifiers in remove_path()\n\nElijah Newren (1):\n      read-cache: reject cache entries with over-long names\n\nJeff King (30):\n      revision: make handle_dotdot() interface less confusing\n      convert: add const to fix strchr() warnings\n      http: add const to fix strchr() warnings\n      transport-helper: drop const to fix strchr() warnings\n      pager: explicitly cast away strchr() constness\n      run-command: explicitly cast away constness when assigning to void\n      pseudo-merge: fix disk reads from find_pseudo_merge()\n      rev-parse: simplify dotdot parsing\n      skip_prefix(): check const match between in and out params\n      revision: avoid writing to const string for parent marks\n      pkt-line: make packet_reader.line non-const\n      rev-parse: avoid writing to const string for parent marks\n      range-diff: drop const to fix strstr() warnings\n      config: store allocated string in non-const pointer\n      http: drop const to fix strstr() warning\n      ref-filter: factor out refname component counting\n      refs/files-backend: drop const to fix strchr() warning\n      ref-filter: simplify lstrip_ref_components() memory handling\n      git-compat-util: fix CONST_OUTPARAM typo and indentation\n      ref-filter: simplify rstrip_ref_components() memory handling\n      ref-filter: avoid strrchr() in rstrip_ref_components()\n      ref-filter: clarify lstrip/rstrip component counting\n      find_last_dir_sep(): convert inline function to macro\n      ci: bump debian-11 job to debian-12\n      test-lib: allow lazy prerequisite snippets as here-docs\n      t5551: handle trailing slashes in expected cookies output\n      t5563: add missing end-of-line in HTTP header\n      revert: initialize const value\n      curl: add support for curl_global_trace() components\n      http: handle curl stripping creds from effective url\n\nJohannes Schindelin (19):\n      ci(dockerized): raise the PID limit for private repositories\n      ci: work around Debian 12's HTTP/2 authentication failures\n      ci(vs-build): adapt to Visual Studio 2026 default on windows-latest\n      mingw: drop the -D_USE_32BIT_TIME_T option\n      mingw: stop using nedmalloc\n      mingw: only use -Wl,--large-address-aware for 32-bit builds\n      t5564: use a short path for the SOCKS proxy socket\n      checkout: use size_t and st_add() for cache entry length in update_some()\n      read-cache: use size_t and st_add() for cache entry size in read_one_entry_opt()\n      cmake: stop trying to build the reftable and xdiff libraries\n      t1600-index: add expensive tests for over-long tree paths\n      ci(vs-build) make the VS-built git.exe adjust MSYSTEM/PATH\n      Git 2.50.2\n      Git 2.51.3\n      Git 2.52.1\n      Git 2.53.1\n      Git 2.54.1\n      Git 2.55.1\n      Git 2.56.1\n\nJunio C Hamano (2):\n      GitHub CI: macos-13 images are no more\n      ci: skip CVS and P4 tests in leaks job, too\n\nKaartic Sivaraam (3):\n      git-compat-util: add st_add5() helper to detect size_t overflow\n      xdiff: use size_t for merge buffer size computations\n      xdiff: check for overflow when computing merge buffer size\n\nKarthik Nayak (8):\n      base85: cleanup unused debug code\n      base85: introduce a new `chunk_and_encode85()` function\n      test-tool: add command to encode-chunk base85\n      apply: reject binary patches greater than MAX_APPLY_SIZE\n      apply: add tests mismatch in binary patch size\n      commit: rewrite `remove_signature()` to work with multiple signatures\n      fast-import: add safeguard around rewrite-submodules features\n      fast-import: add safeguard around 'export-pack-edges'\n\nPatrick Steinhardt (15):\n      build: tolerate use of _Generic from glibc 2.43 with Clang\n      ci: fix broken jobs on Ubuntu 25.10 caused by switch to sudo-rs(1)\n      t7528: work around ETOOMANY in OpenSSH 10.1 and newer\n      meson: simplify iconv-emits-BOM check\n      meson: detect broken iconv that requires ICONV_RESTART_RESET\n      gitlab-ci: update to macOS 15 images\n      README: add GitLab CI badge to make it more discoverable\n      t0021: skip EXPENSIVE test that is broken without SIZE_T_IS_64BIT\n      t4141: fix inefficient use of dd(1)\n      t5608: reduce maximum disk usage\n      t7508: skip EXPENSIVE test that is broken without SIZE_T_IS_64BIT\n      t7900: clean up large EXPENSIVE repository\n      t: use `test_bool_env` to parse GIT_TEST_LONG\n      gitlab-ci: disable RAM disk on macOS jobs\n      gitlab-ci: enable \"GIT_TEST_LONG\"\n\nPaul Tarjan (1):\n      t7800: fix racy \"difftool --dir-diff syncs worktree\" test\n\nRené Scharfe (2):\n      macOS: make Homebrew use configurable\n      macOS: use iconv from Homebrew if needed and present\n\nTamir Duberstein (2):\n      t4205: compare huge output without diff\n      ci: use twice the CPU count on both providers\n\nToon Claes (6):\n      meson: ignore subprojects/.wraplock\n      meson: only detect ICONV_OMITS_BOM if possible\n      meson: use is_cross_build() where possible\n      connect: validate unborn HEAD symref-target in protocol v2\n      connect: validate born ref symref-target in protocol v2\n      connect: validate ref names in protocol v2 ls-refs response\n\nTorsten Bögershausen (2):\n      utf8.c: prepare workaround for iconv under macOS 14/15\n      utf8.c: enable workaround for iconv under macOS 14/15\n\n\n\n"},{"id":"558148","messageId":"20261110221253.DBFncUIG@teonanacatl.net","threadId":"66821","inReplyTo":"xmqqh5hosgw5.fsf@gitster.g","subject":"Re: [ANNOUNCE] Git 2.56.1 and friends","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2026-11-10T22:12:53Z","receivedAt":"2026-11-10T22:12:53Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Junio C Hamano wrote:\n> Credit for organizing the set of these releases goes to Johannes\n> Schindelin.  Credit for finding the vulnerabilities goes to Luke\n> Martin and pwn2addr.  Credit for fixing them goes to Toon Claes,\n> Karthik Nayak, and Kaartic Sivaraam.  The xdiff hardening is based\n> on a patch by Jeff King, with help from Toon Claes and Christian\n> Couder.\n\nMany thanks to all for the effort, including Junio.\n\n> This addresses the security issues CVE-2026-111760, CVE-2026-113408,\n> CVE-2026-114466, CVE-2026-114467, CVE-2026-114469 and CVE-2026-105437.\n> \n>  * CVE-2026-113408:\n> \n>    It is possible for a server to advertise an unborn HEAD that contains\n>    path-traversal characters (e.g., `refs/heads/../../../TARGET`). With\n>    the files ref backend that can lead to ref-updating actions to write\n>    to unwanted files.\n> \n>    See <https://gitlab.com/gitlab-org/git/-/work_items/809> for details.\n\nFWIW, this and the other GitLab work item links appear to be\ninaccessible to normal users.  I'm guessing that's not the\nintent.\n\nI added Patrick to the Cc as one of the folks who may be\nable to poke the right folks at GitLab to open them up?\n\nCheers and thanks again,\n\n-- \nTodd\n\n"},{"id":"558156","messageId":"e12a9cd6-d61a-4e80-a459-954c4b26724f@gmail.com","threadId":"66821","inReplyTo":"20261110221253.DBFncUIG@teonanacatl.net","subject":"Re: [ANNOUNCE] Git 2.56.1 and friends","fromName":"Kaartic Sivaraam","fromEmail":"kaartic.sivaraam@gmail.com","sentAt":"2026-11-11T06:24:18Z","receivedAt":"2026-11-11T06:24:18Z","isPatch":false,"sender":{"key":"kaartic.sivaraam@gmail.com","avatar":"https://avatars.githubusercontent.com/u/12448084?v=4"},"body":"Hi Todd,\n\nOn 11/11/26 03:42, Todd Zullinger wrote:\n> Junio C Hamano wrote:\n>> Credit for organizing the set of these releases goes to Johannes\n>> Schindelin.  Credit for finding the vulnerabilities goes to Luke\n>> Martin and pwn2addr.  Credit for fixing them goes to Toon Claes,\n>> Karthik Nayak, and Kaartic Sivaraam.  The xdiff hardening is based\n>> on a patch by Jeff King, with help from Toon Claes and Christian\n>> Couder.\n> > Many thanks to all for the effort, including Junio.\n>\n\nIndeed. Thank you all for the efforts!\n>> This addresses the security issues CVE-2026-111760, CVE-2026-113408,\n>> CVE-2026-114466, CVE-2026-114467, CVE-2026-114469 and CVE-2026-105437.\n>>\n>>   * CVE-2026-113408:\n>>\n>>     It is possible for a server to advertise an unborn HEAD that contains\n>>     path-traversal characters (e.g., `refs/heads/../../../TARGET`). With\n>>     the files ref backend that can lead to ref-updating actions to write\n>>     to unwanted files.\n>>\n>>     See <https://gitlab.com/gitlab-org/git/-/work_items/809> for details.\n> > FWIW, this and the other GitLab work item links appear to be\n> inaccessible to normal users.  I'm guessing that's not the\n> intent.\n> > I added Patrick to the Cc as one of the folks who may be\n> able to poke the right folks at GitLab to open them up?\n> Thank you for pointing it out. It was indeed not the intention. Those work items are accessible now.\n\n-- \nSivaraam\n\n\n"},{"id":"558213","messageId":"874idn5uvt.fsf@dev.null.iotcl.com.invalid","threadId":"66821","inReplyTo":"e12a9cd6-d61a-4e80-a459-954c4b26724f@gmail.com","subject":"Re: [ANNOUNCE] Git 2.56.1 and friends","fromName":"Toon Claes","fromEmail":"toon@iotcl.com","sentAt":"2026-11-11T14:02:14Z","receivedAt":"2026-11-11T14:02:14Z","isPatch":false,"sender":{"key":"toon@iotcl.com","avatar":"https://avatars.githubusercontent.com/u/121621?v=4"},"body":">>>   * CVE-2026-113408:\n>>>\n>>>     It is possible for a server to advertise an unborn HEAD that contains\n>>>     path-traversal characters (e.g., `refs/heads/../../../TARGET`). With\n>>>     the files ref backend that can lead to ref-updating actions to write\n>>>     to unwanted files.\n>>>\n>>>     See <https://gitlab.com/gitlab-org/git/-/work_items/809> for details.\n>> \n>> FWIW, this and the other GitLab work item links appear to be\n>> inaccessible to normal users.  I'm guessing that's not the\n>> intent.\n>> \n>> I added Patrick to the Cc as one of the folks who may be\n>> able to poke the right folks at GitLab to open them up?\n>> \n>\n> Thank you for pointing it out. It was indeed not the intention. Those \n> work items are accessible now.\n\nYeah, I was planning to open them up ASAP, but I wasn't around my\ncomputer around the time this release came out. I did open them around\nmidnight CET. Thanks for pointing that out.\n\n-- \nLaters,\nToon\n\n"}]}