)]}' { "log": [ { "commit": "27c3aeb25d1ef93c63e46443d73503353ba5e1a5", "tree": "725c850a398fe433b5078573a0b3b85eee1140a6", "parents": [ "7a58ab019b250a66287a1c91ba3693fef3e15b77" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Apr 30 21:34:51 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Apr 30 21:34:51 2018 -0500" }, "message": "xfsprogs: Release v4.16.1\n \nUpdate all the necessary files for a 4.16.1 release.\n \nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "7a58ab019b250a66287a1c91ba3693fef3e15b77", "tree": "46c48f273e067e24cb04a0d77011188859cd8b0a", "parents": [ "9048c8ba2bc22c1254892667be19d0d69fda2901" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Apr 30 21:26:24 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Apr 30 21:26:24 2018 -0500" }, "message": "fsck.xfs: do not use \u0027function\u0027 keyword\n\nIt was pointed out on irc that fsck.xfs uses the \u0027function\u0027 keyword\nalthough it invokes /bin/sh - \u0027function\u0027 is a bashism. It\u0027s not needed\nhere, so just remove it.\n\nFixes: 04a2d5d (\"fsck.xfs: allow forced repairs using xfs_repair\")\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "9048c8ba2bc22c1254892667be19d0d69fda2901", "tree": "9431f0f4f9f1f4d22372229b1ffc393165c0af2c", "parents": [ "6ae62d7bd5bc670a954b3740ac98a59849ae0087" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 26 15:13:34 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 26 15:13:34 2018 -0500" }, "message": "xfsprogs: Release v4.16.0\n \nUpdate all the necessary files for a 4.16.0 release.\n \nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "6ae62d7bd5bc670a954b3740ac98a59849ae0087", "tree": "0972293d7becc7ff2facde02f40a5f2bca5f7d61", "parents": [ "98884b660babfb3bb34587c97f811f856e75d462" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 16:30:13 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 16:30:13 2018 -0500" }, "message": "xfsprogs: Release v4.16.0-rc1\n \nUpdate all the necessary files for a 4.16.0-rc1 release.\n \nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "98884b660babfb3bb34587c97f811f856e75d462", "tree": "c08a40e947f0bacbc38fd62153a07e754c17c140", "parents": [ "74aed9c82b4b48d2a4be6ee3a72a38fdcc6d5a58" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_repair: fix getsubopt name definitions to use enums\n\nConvert the getsubopt usage in xfs_repair to use enums and explicitly\ninitialized array elements, similar to mkfs. This also fixes the hole\nin the o_opts table caused by 42fa89bc1b8dc8 (\"xfs_repair: remove\npre_65_beta option\") that causes segfaults in xfs/179 and xfs/202.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nFixes: 42fa89bc1b (\"xfs_repair: remove pre_65_beta option\")\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "74aed9c82b4b48d2a4be6ee3a72a38fdcc6d5a58", "tree": "cc091de4a0cf8c6624cfa2dd29bc99c78dd609f5", "parents": [ "07c6fd5956da012539ea876787ea1227cf1e8857" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub_all: use system encoding for lsblk output decoding\n\nDon\u0027t hardcode utf-8 as the decoding scheme for lsblk output, since the\nsystem could set it to anything else.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "07c6fd5956da012539ea876787ea1227cf1e8857", "tree": "46620bb03bf206c92ed7d5f229c732ce699c0f1a", "parents": [ "e4e022f9db759b5e45ff08974b12e050d5f4f769" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub_all: escape paths being passed to systemd service instances\n\nsystemd doesn\u0027t like unit instance names with slashes in them, so it\nreplaces them with dashes when it invokes the service. However, it\u0027s\nnot smart enough to convert the dashes to something else, so when it\nunescapes the instance name to feed to xfs_scrub, it turns all dashes\ninto slashes. \"/moo-cow\" becomes \"-moo-cow\" becomes \"/moo/cow\", which\nis wrong. systemd actually /can/ escape the dashes correctly if it is\ntold that this is a path (and not a unit name), but it didn\u0027t do this\nprior to January 2017, so fix this for them.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e4e022f9db759b5e45ff08974b12e050d5f4f769", "tree": "cefd5e933c73b86f15e86b134b9948af3e0d73c3", "parents": [ "3dd91472fd17dfbe7b431e9518f0659fb175602e" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: disable private /tmp for scrub service\n\nDon\u0027t make /tmp private when invoking xfs_scrub as a service, because\n/tmp might contain or itself be an xfs filesystem mountpoint.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "3dd91472fd17dfbe7b431e9518f0659fb175602e", "tree": "03ab78a7f8ab81ae1b50ac47221af732862ab061", "parents": [ "680eacaa3bcb6559d3381e6f1e6abf4603b55040" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub_all: report version\n\nMake xfs_scrub_all -V report its version like the other xfs tools.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "680eacaa3bcb6559d3381e6f1e6abf4603b55040", "tree": "ef43dc2867ac064ef3b2c10b648c66285877049e", "parents": [ "5fa5a86026ad3c0b38d254b467d739dbb31e7d73" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: refactor mountpoint finding code to use libfrog path code\n\nUse the libfrog path finding code to determine if the argument being\npassed in is a mountpoint, remove all mention of taking a block device\n(we have never supported that) from the documentation, and fix some\npotential memory leaks.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "5fa5a86026ad3c0b38d254b467d739dbb31e7d73", "tree": "0969158a1bfad3d10755acfb37cb9a6de9813925", "parents": [ "367db2f5537bc91d98a9f9d4ceb9e97d9ed50554" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: don\u0027t warn about confusing names if dir/file only writable by root\n\nIf we are scanning the directory entries or attribute names of a\ndir/file and the inode can only be written by root, don\u0027t warn about\nUnicode confusable names by default because the system administrator\npresumably made the system like that. Also don\u0027t warn about really\nshort confusable names because of the high chance of collisions. If\nthe caller really wants all the output, they can run in verbose mode.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nAcked-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "367db2f5537bc91d98a9f9d4ceb9e97d9ed50554", "tree": "399a4ebccf7cb125956d4aaefb11d3c176c04edd", "parents": [ "baa9ed8dca213fb286cb1af03d62746dcb4c36df" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: use Unicode skeleton function to find confusing names\n\nDrop the weak normalization-based Unicode name collision detection in\nfavor of the confusable name guidelines provided in Unicode TR36 \u0026 TR39.\nThis means that we transform the original name into its Unicode skeleton\nin order to do hashing-based collision detection.\n\nThe Unicode skeleton is defined as nfd(translation(nfd(string))), which\nis to say that it flattens sequences that render ambiguously into a\nunambiguous format. For example, \u0027l\u0027 and \u00271\u0027 can render identically in\nsome typefaces, so they\u0027re both squashed to \u0027l\u0027. From the skeletons we\ncan figure out if two names will look the same, and thereby complain\nabout them. The unicode spoofing is provided by libicu, hence the\nswitch away from libunistring.\n\nNote that potentially confusable names are only worth an informational\nwarning, since it\u0027s entirely possible that with the system typefaces in\nuse, two names will render distinctly enough that users can tell the\ndifference.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nAcked-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "baa9ed8dca213fb286cb1af03d62746dcb4c36df", "tree": "429d9b86e01d697fa52887d80a84452654e0463e", "parents": [ "bff5d1a4e8df8a23957e5739850754991ad2b9c8" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: check name for suspicious characters\n\nLook for suspicious characters in each name we process. This includes\ncontrol characters, text direction overrides, zero-width code points,\nand names that mix characters from different directionalities.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nAcked-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "bff5d1a4e8df8a23957e5739850754991ad2b9c8", "tree": "70023890ede6c9d59b797833730cb71b2cfb1c5b", "parents": [ "3029a02cad9c8daaeced03ac4abcbf53e1968e66" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: transition from libunistring to libicu for Unicode processing\n\nMove off of libunistring and onto libicu for Unicode name scanning.\nThis will make it easy to warn about unicode code points that do not\nbelong in identifiers (directional overrides, zero width elements) and\nwarn about names that could render similarly enough to cause confusion.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nAcked-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "3029a02cad9c8daaeced03ac4abcbf53e1968e66", "tree": "f85ceb509c57e924fd6afc21789e3eb072e8c9e7", "parents": [ "95c2f78bbc4cb13a96dc10ab112b315158254cab" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: make name_entry a first class structure\n\nInstead of open-coding the construction and hashtable insertion of name\nentries, make name_entry a first class object. This means that we now\nhave name_entry_ prefix functions that take care of computing Unicode\nnormalized names as part of name_entry construction, and we pass around\nthe name_entries when we\u0027re looking for suspicious characters and\nidentically rendering names.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nAcked-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "95c2f78bbc4cb13a96dc10ab112b315158254cab", "tree": "5cb58229004893d6bb601b4c8533cf8f1ed5f054", "parents": [ "d2cf5c2868c726b28a4c72dad0140c6735e5574f" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:11 2018 -0500" }, "message": "xfs_scrub: communicate name problems via flagset instead of booleans\n\nUse an unsigned int to pass around name error flags instead of booleans.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "d2cf5c2868c726b28a4c72dad0140c6735e5574f", "tree": "90b0a5372fcce86ebf0a5c8fe83019f6031c5304", "parents": [ "55290cfc7c08e92489e9564145596c0b063217b4" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:08 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:08 2018 -0500" }, "message": "xfs_scrub: don\u0027t complain about different normalization\n\nSince there are different ways to normalize utf8 names, don\u0027t complain\nwhen we find a name that is normalized in a different way than the NFKC\nthat we use to find duplicate names.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "55290cfc7c08e92489e9564145596c0b063217b4", "tree": "b1abfbdc95314ddc99afa3d87b7804d8e48b8817", "parents": [ "99ea301818779ff663599d00610d6cee6a57738c" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:08 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:08 2018 -0500" }, "message": "xfs_scrub: only run ascii name checks if unicode name checker\n\nSkip the ASCII name checks if the Unicode name checker is going to run,\nsince the latter covers everything that the former does.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "99ea301818779ff663599d00610d6cee6a57738c", "tree": "6074edabe0f69faa87379e76a9a8b87e6f1f093f", "parents": [ "a8502cc4b7e7a9ae829df1446887d98f87a441a8" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Apr 12 10:34:08 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Apr 12 10:34:08 2018 -0500" }, "message": "xfs_scrub: avoid buffer overflow when scanning attributes\n\nAvoid a buffer overflow when we\u0027re formatting extended attribute names\nfor name checking. The kernel headers provide us with XATTR_NAME_MAX,\nso we can rely on that.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "a8502cc4b7e7a9ae829df1446887d98f87a441a8", "tree": "c3dceb03c6aa9d96817b8b5c1e004e5d0712c0d1", "parents": [ "31506b322e438104a52cdafd824d3e0e77736ea7" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:58 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:58 2018 -0500" }, "message": "libxfs: warn about deprecation of irix, freebsd, darwin\n\nIt\u0027s not clear that anyone is using these platforms or if\nthey even build at this point. Get someone\u0027s attention if\nthey are trying to use it.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "31506b322e438104a52cdafd824d3e0e77736ea7", "tree": "0c237574f24a12cf235bad5d9da6d0d8a6c613d0", "parents": [ "28a4f9c05e57bcd6ce463ecb81e2f0efd284c46d" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "message": "xfs_repair: test XFS_SB_VERSION_SHAREDBIT only once\n\nRemove 2 of the 3 identical tests for XFS_SB_VERSION_SHAREDBIT\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "28a4f9c05e57bcd6ce463ecb81e2f0efd284c46d", "tree": "5188e9a07aab27aee626bead8cec817c38dfec64", "parents": [ "42fa89bc1b8dc8723b7556e0b843a3f030a3d6eb" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "message": "xfsprogs: remove unused delete_attr_ok\n\ndelete_attr_ok is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "42fa89bc1b8dc8723b7556e0b843a3f030a3d6eb", "tree": "e9726a119d1a854738ae1f8d9be35451d7740890", "parents": [ "ea94cb62ab18cbc0e807b3bfa7aeab7a80e1454f" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "message": "xfs_repair: remove pre_65_beta option\n\nIrix 6.5 was released 20 years ago. Remove this option from\nthe code. (nb: it\u0027s not present in the manpage.)\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "ea94cb62ab18cbc0e807b3bfa7aeab7a80e1454f", "tree": "9771e102fcedd5db63c48cc5333ccf12b0d26898", "parents": [ "a425d16af1dd6a0e486a1b7c56a197f3d8cb9284" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "message": "xfs_repair: remove unused fs_shared_allowed variable\n\nThe fs_shared_allowed global was set to 1 and then ignored, and\nin fact the feature is never actualy allowed. Remove it and\nthe last stragglers of the old features comment.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "a425d16af1dd6a0e486a1b7c56a197f3d8cb9284", "tree": "dc19d967c838fd92a0d5118dbf30209730d8c3fb", "parents": [ "644589b162bfb7181708903288f7ee237da0adff" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:57 2018 -0500" }, "message": "xfs_repair: remove unused fs_has_extflgbit_allowed\n\nfs_has_extflgbit_allowed is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "644589b162bfb7181708903288f7ee237da0adff", "tree": "49c4f9bc3721576bf4f9f43940dd592f226dfc7c", "parents": [ "dd17791c7f4668275e66d002d134871f018d8d11" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "message": "xfs_repair: remove unused fs_sb_feature_bits_allowed\n\nfs_sb_feature_bits_allowed is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "dd17791c7f4668275e66d002d134871f018d8d11", "tree": "65363a36622010bcd21af84566496be2c56fc07c", "parents": [ "99a1e0dac339c2c2885ba26371923570f7032654" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "message": "xfs_repair: remove unused fs_aligned_inodes_allowed\n\nfs_aligned_inodes_allowed is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "99a1e0dac339c2c2885ba26371923570f7032654", "tree": "697fb637cc02369faa120228cb1372fc84a4b191", "parents": [ "232470f233c24676414a1d3ad433d0429f7f3241" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "message": "xfs_repair: remove unused fs_has_extflgbit_allowed\n\nfs_has_extflgbit_allowed is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "232470f233c24676414a1d3ad433d0429f7f3241", "tree": "cd163adf765dbc7c3f471acef19f69ce88c54007", "parents": [ "497d0706d222700b3005ca0d4bd067293829b36d" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:56 2018 -0500" }, "message": "xfs_repair: remove unused fs_attributes2_allowed\n\nfs_attributes2_allowed is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "497d0706d222700b3005ca0d4bd067293829b36d", "tree": "a644b6264d058ac9e7fc4e80e79c3963dd8e3919", "parents": [ "8aee048310cb23a94bbee0ddd058c00e84982a55" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Apr 03 11:13:49 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Tue Apr 03 11:13:49 2018 -0500" }, "message": "xfs_repair: remove unused fs_attributes_allowed\n\nfs_attributes_allowed is never set to anything but 1;\nremove it and all associated code.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "8aee048310cb23a94bbee0ddd058c00e84982a55", "tree": "e38aeeca6cab4ccd91fa5329933021d8ee82edb3", "parents": [ "6d135e8444aa33de7818c730d5a71d4517dcbd4f" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "message": "libfrog: handle NULL dir \u0026\u0026 blkdev in __fs_table_lookup_mount\n\nIf neither dir nor blkdev is set, dpath never gets set,\nand then gets used (uninitalized) later on.\n\nIf we are asked where \"nothing\" is mounted, just return\n\"nowhere.\"\n\nFixes-coverity-id: 1433615\nFixes-coverity-id: 1433616\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "6d135e8444aa33de7818c730d5a71d4517dcbd4f", "tree": "248267d944c0864bc558fafe9848bd6b2609cb7b", "parents": [ "25992d9c1120ea9097dfb2c08bad2062ded2217d" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "message": "xfs_scrub: initialize movon in xfs_scrub_connections\n\nGiven the logic in xfs_scrub_connections, it\u0027s possible to\nfail all 3 tests and wind up checking an uninitialized moveon\nvariable at the end. Start out with \"true\" to avoid this and\nmove on even if all the conditions in the function are false.\n\nFixes-coverity-id: 1433617\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "25992d9c1120ea9097dfb2c08bad2062ded2217d", "tree": "a43dd55b3a693b75800831c471789cdacac92672", "parents": [ "19177e69bbb98bbaf3f64168e77848b7651cf64d" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "message": "xfs_scrub: synchronize error levels \u0026 logging\n\nHaving only a subset of the five error_levels present in\nthe log_level[] array is asking for trouble when someone\ntries to __str_log(S_PREEN ...) and overruns the array.\n\nTie it all together in a single structure that\u0027s\ninitialized together to make the mapping more obvious and\nidiot-proof.\n\nFixes-coverity-id: 1433618\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "19177e69bbb98bbaf3f64168e77848b7651cf64d", "tree": "3546c072004eceb036da9bee9c0588edcad19239", "parents": [ "04a2d5dcdc0486e2b0c9903428a17313b8c92b7d" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Tue Mar 27 17:43:37 2018 -0500" }, "message": "xfs_spaceman: remove incorrect linux/fs.h include\n\nRemove the direct linux/fs.h include from spaceman because all xfs\nutilities should include xfs.h so that we can wrap missing kernel header\ndeclarations properly.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "04a2d5dcdc0486e2b0c9903428a17313b8c92b7d", "tree": "9990f076b0b12b0dea979ca8c77ad58529329e16", "parents": [ "7c3e94a3001292946521fb4cfac841e9ac32691b" ], "author": { "name": "Jan Tulak", "email": "jtulak@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "message": "fsck.xfs: allow forced repairs using xfs_repair\n\nThe fsck.xfs script did nothing, because xfs doesn\u0027t need a fsck to be\nrun on every unclean shutdown. However, sometimes it may happen that the\nroot filesystem really requires the usage of xfs_repair and then it is a\nhassle. This patch makes the situation a bit easier by detecting forced\nchecks (/forcefsck or fsck.mode\u003dforce) and invoking xfs_repair.\n\nSigned-off-by: Jan Tulak \u003cjtulak@redhat.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "7c3e94a3001292946521fb4cfac841e9ac32691b", "tree": "2e6859dfaee013d61139a903fa65752a359e366c", "parents": [ "4aa3c02fde3547b24707d379f8d994c8e41208c4" ], "author": { "name": "Jan Tulak", "email": "jtulak@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "message": "xfs_repair: add flag -e to modify exit code for corrected errors\n\nxfs_repair without -n ends with a return code 0 if it finished ok,\nno matter if there were some errors in the fs, or not. The new flag\n-e means that we can avoid screenscraping and parsing text output to\ndetect if an error was found (and corrected).\n\nIf something could not be corrected or in any other case than the \"found\nsomething but fixed it all,\" the behaviour with this flag is unchanged.\n\nSigned-off-by: Jan Tulak \u003cjtulak@redhat.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\n[sandeen: make -e and -n exclusivity clear in manpage synopsis]\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "4aa3c02fde3547b24707d379f8d994c8e41208c4", "tree": "a51cb560a138b5b5082b68469bf53a8cba55231f", "parents": [ "9cf846b514ff335470f130d635eff525386bab21" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "message": "metadump/restore: don\u0027t use errno after fwrite/fread failures\n\nfread/fwrite don\u0027t set errno, so printing out strerror(errno)\nafter a failure leads to incorrect and confusing messages:\n\n# xfs_mdrestore pre_repair.meta pre_repair.img\nxfs_mdrestore: error reading from file: Success\n\nDon\u0027t return unset errno from write_index, either.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "9cf846b514ff335470f130d635eff525386bab21", "tree": "5b04a9d71a782b30a9e1c1041e81b9ab21b4390d", "parents": [ "938f7b708872d71c143be4d87f774293319c2776" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "message": "mkfs: enable sparse inodes by default\n\nEnable the sparse inode feature by default.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "938f7b708872d71c143be4d87f774293319c2776", "tree": "4fcfd724b2bf2f8eb473f98924446c46e96622e2", "parents": [ "be66eb8ef581ae50ec9fe6ff8a95a58fe62c2b45" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:31 2018 -0500" }, "message": "xfs_fsr: refactor mountpoint finding to use libfrog paths functions\n\nRefactor the mount-point finding code in fsr to use the libfrog helpers\ninstead of open-coding yet another routine.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "be66eb8ef581ae50ec9fe6ff8a95a58fe62c2b45", "tree": "6a83047d1ff039cb2b85cac5e7eee6209e49cb24", "parents": [ "f8cd7da8eec39dddbdeebac8617b64f2a1a2deb0" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "message": "libfrog: fs_table_lookup_mount should realpath the argument\n\nCall realpath on the dir argument so that we\u0027re comparing canonical\npaths when looking for the mountpoint. This fixes the problem where\n\u0027/home/\u0027 doesn\u0027t match \u0027/home\u0027 even though they refer to the same thing.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "f8cd7da8eec39dddbdeebac8617b64f2a1a2deb0", "tree": "87914a32b141a51805fb54f761d2ab3b318e3295", "parents": [ "3452f451b2ecd91ca2c6609c150779e32a8a1442" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "message": "xfs_repair: use custom ifork verifier in mv_orphanage\n\nNow that we have a custom verifier which can ignore parent\ninode numbers, use it in mv_orphanage() as well; orphan inodes\nmay have invalid parents, and we\u0027re about to reconnect\nthem anyway, so override that test when we get them off disk.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "3452f451b2ecd91ca2c6609c150779e32a8a1442", "tree": "634920f734010bd0fd536b8bb0a69f00b42dcdb8", "parents": [ "e2bc34de8f935443b4570244ade8d5da0af9c0b0" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "message": "xfs_repair: implement custom ifork verifiers\n\nThere are a few cases where an early stage of xfs_repair will write an\ninvalid inode fork buffer to signal to a later stage that it needs to\ncorrect the value. This happens in phase 4 when we detect an inline\nformat directory with an invalid .. pointer. To avoid triggering the\nifork verifiers on this, inject a custom verifier for phase 6 that lets\nthis pass for now.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e2bc34de8f935443b4570244ade8d5da0af9c0b0", "tree": "35d27186a9331a6ba78f91b95d0f0fbee35fd7e0", "parents": [ "40b2e29848f10ee2e99c8aa982e7574a58c43d94" ], "author": { "name": "Baruch Siach", "email": "baruch@tkos.co.il", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "message": "xfs_scrub: add missing paths header\n\nFix the following build failure with musl libc:\n\nxfs_scrub.c: In function ‘main’:\nxfs_scrub.c:670:11: error: ‘_PATH_MOUNTED’ undeclared (first use in this function)\n mtab \u003d _PATH_MOUNTED;\n ^~~~~~~~~~~~~\n\nSigned-off-by: Baruch Siach \u003cbaruch@tkos.co.il\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "40b2e29848f10ee2e99c8aa982e7574a58c43d94", "tree": "291645a2d72b38561435998ed0b8b6d0a694f938", "parents": [ "745952887e0ef0a0a5796bc2cffd3da887147ac0" ], "author": { "name": "Baruch Siach", "email": "baruch@tkos.co.il", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "message": "workqueue: add missing pthreads header\n\nFix the following build failure with musl libc:\n\nIn file included from read_verify.c:25:0:\n../include/workqueue.h:39:2: error: unknown type name \u0027pthread_t\u0027\n pthread_t *threads;\n ^~~~~~~~~\n../include/workqueue.h:42:2: error: unknown type name \u0027pthread_mutex_t\u0027\n pthread_mutex_t lock;\n ^~~~~~~~~~~~~~~\n../include/workqueue.h:43:2: error: unknown type name \u0027pthread_cond_t\u0027\n pthread_cond_t wakeup;\n ^~~~~~~~~~~~~~\n\nSigned-off-by: Baruch Siach \u003cbaruch@tkos.co.il\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "745952887e0ef0a0a5796bc2cffd3da887147ac0", "tree": "4a79833a63d80384edaa182ab55db351958e3292", "parents": [ "12ac6e048a0fbf987820ee613465eecf884a8528" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Mar 26 21:27:28 2018 -0500" }, "message": "xfs_io: fix operation time reporting\n\nCUrrently the 100th/sec units always report zero, such as:\n\n32 MiB, 8192 ops; 0:00:21.00 (1.476 MiB/sec and 377.9260 ops/sec)\n ^^\n\nThis is incorrect. Fix the maths that is wrong by removing all the\nunnecesary floating point maths and just using basic integer\ndivision...\n\nSigned-Off-By: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "12ac6e048a0fbf987820ee613465eecf884a8528", "tree": "1c336f8f869bc35ff8e8f21f3a0f28124b967eec", "parents": [ "ad9ac929a32cc8f09bfab7d4f1c5f13f1fa9adc7" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_repair: don\u0027t fail directory repairs when grabbing inodes\n\nThere are a few places where xfs_repair needs to be able to load a\ndamaged directory inode to perform repairs. Since inline data fork\nverifiers can now be customized, refactor libxfs_iget to enable\nrepair to get at this so that we don\u0027t crash in phase 6.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "ad9ac929a32cc8f09bfab7d4f1c5f13f1fa9adc7", "tree": "390c35d9cb9e4d47270a6ae5e69b5dbd9bac8cb3", "parents": [ "e24ee6b654599b5c129994921777cc3d3ad2b420" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_db: print transaction reservation type information\n\nCreate a new xfs_db command to print the transaction reservation info for\na given filesystem. This will make it easier to compare the calculations\nmade by the kernel and xfsprogs in case there is a discrepancy.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e24ee6b654599b5c129994921777cc3d3ad2b420", "tree": "242c7c6da37ae809fadb09d22b5b026b37c8113c", "parents": [ "a440f87737944bc2b4f1ce0b4dae7698703f0fdc" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_scrub: don\u0027t try to scan xattrs if bstat says there aren\u0027t any\n\nOnly try to scan the extended attributes of a file if bstat says that\nthe file actually has any. Surprisingly, this reduces the phase 5\nruntime by 40% if most of the files don\u0027t have attrs.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "a440f87737944bc2b4f1ce0b4dae7698703f0fdc", "tree": "f58b80e3cdbd341d0312bb6a5acd213e457d8eb5", "parents": [ "c767c5aee0427c117b9338a62a5f140befd15aaa" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_scrub: fix #include ordering to avoid build failure\n\nFix the ordering of the header includes in all the scrub source. We put\nxfs.h first so that it will pull in include/linux.h which pulls in\nlinux/fs.h + whatever overrides are necessary (currently limited to\nstruct fsxattr) to make things work on this platform, and then we remove\nthe #includes for anything that will get pulled (directly or indirectly)\nby xfs.h for cleanliness. Without this, a user compiling new xfsprogs\non a system with a 4.7 kernel gets this:\n\nBuilding scrub\n [CC] disk.o\nIn file included from ../include/xfs.h:37:0,\n from disk.c:40:\n../include/xfs/linux.h:185:8: error: redefinition of \u0027struct fsxattr\u0027\n struct fsxattr {\n ^~~~~~~\nIn file included from disk.c:31:0:\n/usr/include/linux/fs.h:155:8: note: originally defined here\n struct fsxattr {\n ^~~~~~~\ngmake[2]: *** [../include/buildrules:60: disk.o] Error 1\ngmake[1]: *** [include/buildrules:36: scrub] Error 2\nmake: *** [Makefile:77: default] Error 2\n\nReported-by: Mikael Magnusson \u003cmikachu@gmail.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "c767c5aee0427c117b9338a62a5f140befd15aaa", "tree": "96f6d5c9d642fef7c27b9de4bac13701fd216342", "parents": [ "7c309151c9ca733fd6319250abafb447cb4d5d61" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_scrub: don\u0027t ask user to run xfs_repair for only warnings\n\nDon\u0027t advise the user to run xfs_repair on a filesystem that triggers\nwarnings but no errors; there\u0027s no corruption for it to fix.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "7c309151c9ca733fd6319250abafb447cb4d5d61", "tree": "1c5a08d558e57664c36e96187deb894cca6e92bf", "parents": [ "de24d640d5e30bbaf25ac314f95cbcbec9a97966" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_scrub: log operational messages when interactive\n\nRecord the summary of an interactive session in the system log so that\nfuture support requests can get a better picture of what happened.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "de24d640d5e30bbaf25ac314f95cbcbec9a97966", "tree": "ec6498ed2caa75cd1e99800ca624969308e34163", "parents": [ "a93fcc042a7c79786dfe72f1c14737290e629c27" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:23 2018 -0600" }, "message": "xfs_db: don\u0027t crash in ablock if there\u0027s no inode\n\nMake sure we actually have an inode selected before trying to unwrap its\nattribute fork. Found via a crash in xfs/288 with project quotas\nenabled.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "a93fcc042a7c79786dfe72f1c14737290e629c27", "tree": "66482ae14a902f0ac99593690c3e66abd320dcf5", "parents": [ "f50eb2c32b50c9bc4913f2374869380c5969573c" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "misc: fix gcc 7.3 warnings\n\nFix various compiler warnings that pop up in 7.3.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "f50eb2c32b50c9bc4913f2374869380c5969573c", "tree": "01f754d0595c430a6d1b297c888ad96843cd2fcc", "parents": [ "2ce8bff527b559cb0257042c99f4e51efb877132" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "xfsprogs: new libxfs-apply option for Signed-off-by: tag\n\nTechnically when a maintainer moves a patch from another project,\nthey should add their Signed-off-by: tag. Get that info automatically\nfrom git-config, and add an option to to override it if desired,\nto make that easy when cross-porting libxfs patches\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "2ce8bff527b559cb0257042c99f4e51efb877132", "tree": "392eb40df725c0c07c3f1163626dd35773efbffb", "parents": [ "444884913bd52a843eb49a757e6e65681ab2e90b" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "xfsprogs: call libxfs_destroy from other utilities\n\nCall libxfs_destroy() from xfs_copy, xfs_db, mkfs.xfs, and\nxfs_repair to allow us to detect leaked items in these\nutilities as well.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "444884913bd52a843eb49a757e6e65681ab2e90b", "tree": "f943a2fcb303b1c01d2d45017918032c04b03a0f", "parents": [ "bf0e024fb10d623e8da98a0c41625a2007437a6a" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "libxfs: Catch non-empty zones on destroy\n\nCreate and use a kmem_zone_destroy which warns if we are\nreleasing a non-empty zone when the LIBXFS_LEAK_CHECK\nenvironment variable is set, wire this into libxfs_destroy(),\nand call that when various tools exit.\n\nThe LIBXFS_LEAK_CHECK environment variable also causes\nthe program to exit with failure when a leak is detected,\nuseful for failing automated tests if leaks are encountered.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "bf0e024fb10d623e8da98a0c41625a2007437a6a", "tree": "f9e96342c3803f3a0c70f4b91b8f85c4949bb14f", "parents": [ "864028ed3ae482212b41f83d85b2be42aedb3aac" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "libxfs: move xfs_inode_zone to rdwr.c\n\nThe zone itself is created in rdwr.c, so define it there as\nwell, and add it to the list of externs in manage_zones along\nwith all the rest, for consistency.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "864028ed3ae482212b41f83d85b2be42aedb3aac", "tree": "6bd3ccf583b2155ac9ea3cfca2c4ab86235f99fd", "parents": [ "fc9f709f5beeb2b89136662013fc0bb65828ca8f" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "libxfs: add function to free all buffers in bcache\n\nlibxfs_bcache_purge simply moves all \"free\" buffers\nonto the xfs_buf_freelist mru list; add a new function to\nactually free them when we tear everything down, so leak\ncheckers don\u0027t go nuts about lots of unfreed xfs_bufs\nat exit.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "fc9f709f5beeb2b89136662013fc0bb65828ca8f", "tree": "f6c78175ac9e32ea2a0d68d8723af0142d63b023", "parents": [ "d97638251b67861786b3dbfeab6555c5ea3660b0" ], "author": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "libxfs: Replace XFS_BUF_SET_PTR with xfs_buf_associate_memory\n\nWe test the return value of the macro, but it returns\nreturns a side-effect which looks like failure. Write\na userspace-libxfs-specific version of xfs_buf_associate_memory\nto make this code a tad more like the kernel, with a proper\nreturn value to boot.\n\nSigned-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "d97638251b67861786b3dbfeab6555c5ea3660b0", "tree": "e2f8b2db1975ea1f400288191def6b321b016269", "parents": [ "9d50331ab193c26d133bc8fd164f1054fb87312f" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "xfs_io: add RWF_DSYNC support to pwrite\n\nEnable testing write behaviour with the per-io RWF_DSYNC flag.\n\nSigned-Off-By: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "9d50331ab193c26d133bc8fd164f1054fb87312f", "tree": "5ad0b4f7d2e0aafce855d2c5d69cf033b7d3490c", "parents": [ "a2e81058c167b6b9761b46883cdd6371616a2de6" ], "author": { "name": "Luis R. Rodriguez", "email": "mcgrof@kernel.org", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:22 2018 -0600" }, "message": "xfs_scrub: make interpreter explicit to python3\n\nUsing #!/usr/bin/env makes some package dependency tools\nsuch as rpm complain given that it cannot verify package\ndependencies. Making it explicit resolves this lint rant.\n\nSigned-off-by: Luis R. Rodriguez \u003cmcgrof@kernel.org\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "a2e81058c167b6b9761b46883cdd6371616a2de6", "tree": "6d0d53b08341375721832d1af1da945b93ceab0b", "parents": [ "57e7138aa513b89813d042a5410d335baf39a009" ], "author": { "name": "Xiao Yang", "email": "yangx.jy@cn.fujitsu.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "message": "xfs_repair: Add missing braces to allow zeroing of corrupt log\n\nWhen xlog_find_tail() fails to find the head or the tail, the missing\nbraces leads that an unparseable log always exits with status 2, even\nif we\u0027ve asked for -n or -L which should proceed. We can expose this\nissue by xfstests case xfs/098.\n\nFixes: commit b04647edea32 (\"xfs_repair: exit with status 2 if log dirtiness is unknown\")\n\nSigned-off-by: Xiao Yang \u003cyangx.jy@cn.fujitsu.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "57e7138aa513b89813d042a5410d335baf39a009", "tree": "dcc2565e95112e2fbf2772de280685daa6cd30db", "parents": [ "99d6e84ff181d9944cd3eb4fe13c95d8277e026b" ], "author": { "name": "Brian Foster", "email": "bfoster@redhat.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "message": "xfs_io: support a basic extent swap command\n\nExtent swap is a low level mechanism exported by XFS to facilitate\nfilesystem defragmentation. It is typically invoked by xfs_fsr under\nconditions that will atomically adjust inode extent state without\nloss of file data.\n\nWhile xfs_fsr provides some debug capability to tailor its behavior,\nit is not flexible enough to facilitate low level tests of the\nextent swap mechanism. For example, xfs_fsr may skip swaps between\ninodes that consist solely of preallocated extents because it\nconsiders such files already 100% defragmented. Further, xfs_fsr\ncopies data between files where doing so may be unnecessary and thus\ninefficient for lower level tests.\n\nAdd a basic swapext command to xfs_io that allows userspace\ninvocation of the command under more controlled conditions. This\nfacilites targeted tests without interference from xfs_fsr policy,\nsuch as using files with only preallocated extents, known/expected\nfailure cases, etc. This command makes no effort to retain data\nacross the operation. As such, it is for testing purposes only.\n\nSigned-off-by: Brian Foster \u003cbfoster@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "99d6e84ff181d9944cd3eb4fe13c95d8277e026b", "tree": "016c461617226ee72f5989d2f209f9ae74668b82", "parents": [ "e61f1552aae4819194ad603222eedef4b08710a2" ], "author": { "name": "Goldwyn Rodrigues", "email": "rgoldwyn@suse.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "message": "xfs_io: Add missing perror for write_once (-O)\n\nThis got missed in the last set of patches.\n\nSigned-off-by: Goldwyn Rodrigues \u003crgoldwyn@suse.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e61f1552aae4819194ad603222eedef4b08710a2", "tree": "e485b289b77d405f9b8ebe1bc9350bf681838f57", "parents": [ "1b0adc71229ba8611b06abc2bffa03babba5157d" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "message": "misc: enable link time optimization, if requested\n\nEnable link time optimization (LTO) if the builder requests it. The\nextra link optimization results in smaller binaries.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "1b0adc71229ba8611b06abc2bffa03babba5157d", "tree": "eeb51d705966ef5c7a304c97050256b594e59150", "parents": [ "ffe5df5574f3655ab30a1cedacb3c7cb37b3f7a6" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Thu Mar 08 20:35:20 2018 -0600" }, "message": "misc: enable retpolines across all xfsprogs utilities\n\nDetect and enable retpolines for all code, to mitigate Spectre v2\n(branch target injection) on x86.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "ffe5df5574f3655ab30a1cedacb3c7cb37b3f7a6", "tree": "f99d862da1158d48f94e0743573e053bfcbfb580", "parents": [ "c38464ff97af428ce9036c8613b0a8d3232129f7" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Feb 28 13:46:11 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Wed Feb 28 13:46:11 2018 -0600" }, "message": "xfs: fix u32 type usage in sb validation function\n\nSource kernel commit: 131fa58d391fc0939f6c66b23776ad5df5db20f9\n\nDon\u0027t use u32, use uint32_t, because this won\u0027t work in xfsprogs.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Eric Sandeen \u003csandeen@redhat.com\u003e\n[sandeen: no-op commit, fixed previously to keep build working]\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "c38464ff97af428ce9036c8613b0a8d3232129f7", "tree": "f99d862da1158d48f94e0743573e053bfcbfb580", "parents": [ "d07cc7245eaaebd6a9073936965a4d9737773ccb" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "message": "xfs: don\u0027t screw up direct writes when freesp is fragmented\n\nSource kernel commit: 6d8a45ce29c7d67cc4fc3016dc2a07660c62482a\n\nxfs_bmap_btalloc is given a range of file offset blocks that must be\nallocated to some data/attr/cow fork. If the fork has an extent size\nhint associated with it, the request will be enlarged on both ends to\ntry to satisfy the alignment hint. If free space is fragmentated,\nsometimes we can allocate some blocks but not enough to fulfill any of\nthe requested range. Since bmapi_allocate always trims the new extent\nmapping to match the originally requested range, this results in\nbmapi_write returning zero and no mapping.\n\nThe consequences of this vary -- buffered writes will simply re-call\nbmapi_write until it can satisfy at least one block from the original\nrequest. Direct IO overwrites notice nmaps \u003d\u003d 0 and return -ENOSPC\nthrough the dio mechanism out to userspace with the weird result that\nwrites fail even when we have enough space because the ENOSPC return\noverrides any partial write status. For direct CoW writes the situation\nwas disastrous because nobody notices us returning an invalid zero-length\nwrong-offset mapping to iomap and the write goes off into space.\n\nTherefore, if free space is so fragmented that we managed to allocate\nsome space but not enough to map into even a single block of the\noriginal allocation request range, we should break the alignment hint in\norder to guarantee at least some forward progress for the direct write.\nIf we return a short allocation to iomap_apply it\u0027ll call back about the\nremaining blocks.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "d07cc7245eaaebd6a9073936965a4d9737773ccb", "tree": "8ad3237f3dfb0d3fb9d7a3ddea920081e0a68607", "parents": [ "3cb68815a885ca9c39e7d288ed6f4bd026b22b9c" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "message": "xfs: treat CoW fork operations as delalloc for quota accounting\n\nSource kernel commit: 4b4c1326fd7c7210d23d9dd3bfc51f2b6477bb9e\n\nSince the CoW fork only exists in memory, it is incorrect to update the\non-disk quota block counts when we modify the CoW fork. Unlike the data\nfork, even real extents in the CoW fork are only delalloc-style\nreservations (on-disk they\u0027re owned by the refcountbt) so they must not\nbe tracked in the on disk quota info. Ensure the i_delayed_blks\naccounting reflects this too.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "3cb68815a885ca9c39e7d288ed6f4bd026b22b9c", "tree": "f6f8975dafb016a19e1387552729d20d39b3e26f", "parents": [ "1d3bac1f3b6b4aa8b4de1ed6a95b0440b1bcd953" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "message": "xfs: refactor accounting updates out of xfs_bmap_btalloc\n\nSource kernel commit: 751f3767c245f9adf4f0a4f8f04aae9ae1d675a0\n\nMove all the inode and quota accounting updates out of xfs_bmap_btalloc\nin preparation for fixing some quota accounting problems with copy on\nwrite.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "1d3bac1f3b6b4aa8b4de1ed6a95b0440b1bcd953", "tree": "160670996cba555bbf1c47f66dc1b66f1060f318", "parents": [ "885ba5ce1dd446a82f2ecf1f03ad834d253b05d7" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "message": "xfs: refactor inode verifier corruption error printing\n\nSource kernel commit: 22431bf3dfbf44d7356933776eb486a6a01dea6f\n\nRefactor inode verifier error reporting into a non-libxfs function so\nthat we aren\u0027t encoding the message format in libxfs. This also\nchanges the kernel dmesg output to resemble buffer verifier errors\nmore closely.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "885ba5ce1dd446a82f2ecf1f03ad834d253b05d7", "tree": "b01799b5221db41cb5993813751a7c8b078dc6ff", "parents": [ "37d086ca9ac33917d90aa883decef8bb24524ba1" ], "author": { "name": "Shan Hai", "email": "shan.hai@oracle.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "message": "xfs: bmap code cleanup\n\nSource kernel commit: 6ca30729c206d62d88730a904af7d543a56273d8\n\nRemove the extent size hint and realtime inode relevant code from\nthe xfs_bmapi_reserve_delalloc since it is not called on the inode\nwith extent size hint set or on a realtime inode.\n\nSigned-off-by: Shan Hai \u003cshan.hai@oracle.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "37d086ca9ac33917d90aa883decef8bb24524ba1", "tree": "711290d4fb9e9debfa6e2c033daf64a7e6ca3593", "parents": [ "d5105a897ca973d43e82d54bda4ea40692a94769" ], "author": { "name": "Carlos Maiolino", "email": "cmaiolino@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:19 2018 -0600" }, "message": "Split buffer\u0027s b_fspriv field\n\nSource kernel commit: fb1755a645972ed096047583600838f6cf414e2b\n\nBy splitting the b_fspriv field into two different fields (b_log_item\nand b_li_list). It\u0027s possible to get rid of an old ABI workaround, by\nusing the new b_log_item field to store xfs_buf_log_item separated from\nthe log items attached to the buffer, which will be linked in the new\nb_li_list field.\n\nThis way, there is no more need to reorder the log items list to place\nthe buf_log_item at the beginning of the list, simplifying a bit the\nlogic to handle buffer IO.\n\nThis also opens the possibility to change buffer\u0027s log items list into a\nproper list_head.\n\nb_log_item field is still defined as a void *, because it is still used\nby the log buffers to store xlog_in_core structures, and there is no\nneed to add an extra field on xfs_buf just for xlog_in_core.\n\nSigned-off-by: Carlos Maiolino \u003ccmaiolino@redhat.com\u003e\nReviewed-by: Bill O\u0027Donnell \u003cbillodo@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n[darrick: minor style changes]\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n[sandeen: b_li_list unused in userspace]\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "d5105a897ca973d43e82d54bda4ea40692a94769", "tree": "3357cfa05471b67fadec02d35d726294c3bc253e", "parents": [ "e91df8449abbbb2087c64428c8e4af25c445ca0e" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: convert to new i_version API\n\nSource kernel commit: f0e28280629e0ec7921f3179409a179b1ea41f24\n\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nAcked-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nAcked-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e91df8449abbbb2087c64428c8e4af25c445ca0e", "tree": "3968b158d387e53c1361b30b17aed2afd16f5557", "parents": [ "faaad1df3f51c03131bd5bc527c9ffdcecb0f0f1" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: check sb_agblocks and sb_agblklog when validating superblock\n\nSource kernel commit: 4bb73d014785cc55225686f9f46e7192fb59d26b\n\nCurrently, we don\u0027t check sb_agblocks or sb_agblklog when we validate\nthe superblock, which means that we can fuzz garbage values into those\nvalues and the mount succeeds. This leads to all sorts of UBSAN\nwarnings in xfs/350 since we can then coerce other parts of xfs into\nshifting by ridiculously large values.\n\nOnce we\u0027ve validated agblocks, make sure the agcount makes sense.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\n[sandeen: fix up u32 usage now so we keep building]\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n" }, { "commit": "faaad1df3f51c03131bd5bc527c9ffdcecb0f0f1", "tree": "4735f27e2cdf6898236a15e87c7204ae02e273f1", "parents": [ "e4b963e9ae98a1c7522e97eec7015d33f7eb8ab4" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: recheck reflink / dirty page status before freeing CoW reservations\n\nSource kernel commit: be78ff0e72778eb4df4aac66edb9e97462bfe00d\n\nEryu Guan reported seeing occasional hangs when running generic/269 with\na new fsstress that supports clonerange/deduperange. The cause of this\nhang is an infinite loop when we convert the CoW fork extents from\nunwritten to real just prior to writing the pages out; the infinite\nloop happens because there\u0027s nothing in the CoW fork to convert, and so\nit spins forever.\n\nThe fundamental issue here is that when we go to perform these CoW fork\nconversions, we\u0027re supposed to have an extent waiting for us, but the\nlow space CoW reaper has snuck in and blown them away! There are four\nconditions that can dissuade the reaper from touching our file -- no\nreflink iflag; dirty page cache; writeback in progress; or directio in\nprogress. We check the four conditions prior to taking the locks, but\nwe neglect to recheck them once we have the locks, which is how we end\nup whacking the writeback that\u0027s in progress.\n\nTherefore, refactor the four checks into a helper function and call it\nonce again once we have the locks to make sure we really want to reap\nthe inode. While we\u0027re at it, add an ASSERT for this weird condition so\nthat we\u0027ll fail noisily if we ever screw this up again.\n\nReported-by: Eryu Guan \u003ceguan@redhat.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nTested-by: Eryu Guan \u003ceguan@redhat.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e4b963e9ae98a1c7522e97eec7015d33f7eb8ab4", "tree": "4043bf35bed0c3c850e402e1f968943fb012f5c0", "parents": [ "f596a90d3d7a03f42cedcbc9b505fefe5da5df08" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: btree format ifork loader should check for zero numrecs\n\nSource kernel commit: 55e45429ce3e4ac9dd2bf4937b1a499a69ccc4ca\n\nA btree format inode fork with zero records makes no sense, so reject it\nif we see it, or else we can miscalculate memory allocations. Found by\nzeroes fuzzing {a,u3}.bmbt.numrecs in xfs/{374,378,412} with KASAN.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "f596a90d3d7a03f42cedcbc9b505fefe5da5df08", "tree": "f2f3f37a544638a4f50f13be99085546a90d2b69", "parents": [ "b67317ef3c22d69e0c29a5c624a84dc37e9c6ffd" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: attr leaf verifier needs to check for obviously bad count\n\nSource kernel commit: 79a69bf8dc240ebeb105226a8a8540df136bf987\n\nIn the attribute leaf verifier, we can check for obviously bad values of\nfirstused and count so that later attempts at lasthash don\u0027t run off the\nend of the memory buffer. Found by ones fuzzing hdr.count in xfs/400 with\nKASAN.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "b67317ef3c22d69e0c29a5c624a84dc37e9c6ffd", "tree": "d309ef7e48fa9d79cb3b4413e6515fe1f484c37d", "parents": [ "221b1ddc29415a673750665696f95990c53d8d2e" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: directory scrubber must walk through data block to offset\n\nSource kernel commit: ce92d29ddf9908d397895c46b7c78e9db8df414d\n\nIn xfs_scrub_dir_rec, we must walk through the directory block entries\nto arrive at the offset given by the hash structure. If we blindly\ntrust the hash address, we can end up midway into a directory entry and\nstray outside the block. Found by lastbit fuzzing lents[3].address in\nxfs/390 with KASAN enabled.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "221b1ddc29415a673750665696f95990c53d8d2e", "tree": "9c601e98171825d58b59237e55aa5d909b718297", "parents": [ "1048d0e209ca4ad387485e6aeed07dd5870b0f36" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: cross-reference the realtime bitmap\n\nSource kernel commit: 46d9bfb5e706493777b9dfed666cd8967f69e6fd\n\nWhile we\u0027re scrubbing various btrees, cross-reference the records\nwith the other metadata.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "1048d0e209ca4ad387485e6aeed07dd5870b0f36", "tree": "141b56d2ca3ca766265db7629185fb943bf26e71", "parents": [ "556c739daebc238e585aae4b4c20a34760f6b7ae" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: add scrub cross-referencing helpers for the refcount btrees\n\nSource kernel commit: 49db55eca5665e32c9d3e67a7d5694bcc6c274de\n\nAdd a couple of functions to the refcount btrees that will be used\nto cross-reference metadata against the refcountbt.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "556c739daebc238e585aae4b4c20a34760f6b7ae", "tree": "be54b1c1bd1d7c2b4fd14cd2a5d05612238cbfb9", "parents": [ "9deb7f653cf80aa158aee121cbb655d8d88c2413" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: add scrub cross-referencing helpers for the rmap btrees\n\nSource kernel commit: ed7c52d4bf92ac1f05b8c251a44a8bf4688f8786\n\nAdd a couple of functions to the rmap btrees that will be used\nto cross-reference metadata against the rmapbt.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "9deb7f653cf80aa158aee121cbb655d8d88c2413", "tree": "fe60e7d724cd8f1b8f75fb86b265b2c806c16eb6", "parents": [ "1fe41a732560e8e7f029390fa9da5a9e83a3fe40" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: add scrub cross-referencing helpers for the inode btrees\n\nSource kernel commit: 2e001266b67c865ad904e1889658282d0773b207\n\nAdd a couple of functions to the inode btrees that will be used\nto cross-reference metadata against the inobt.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "1fe41a732560e8e7f029390fa9da5a9e83a3fe40", "tree": "ed4f01e98545d025462bd058e5e29bd67643fc7d", "parents": [ "b09294b35f2e4690e718bd1b4bc99a1d50591776" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: add scrub cross-referencing helpers for the free space btrees\n\nSource kernel commit: ce1d802e6a889b8ee53b3444c6d7e8cfecadac50\n\nAdd a couple of functions to the free space btrees that will be used\nto cross-reference metadata against the bnobt/cntbt, and a generic\nbtree function that provides the real implementation.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "b09294b35f2e4690e718bd1b4bc99a1d50591776", "tree": "13a3d1089009cda08a8576b9d913461ab2749cae", "parents": [ "8504509d9e0d93c7ed1dd0bb44925c9c17efe8d5" ], "author": { "name": "Brian Foster", "email": "bfoster@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: cancel tx on xfs_defer_finish() error during xattr set/remove\n\nSource kernel commit: c468562879a766de2c2fbedd41b653a7bf4c157d\n\nChris Dunlop reports a problem where an xattr operation fails,\nreports the following error to syslog and hangs during unmount:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ BUG: lock held when returning to user space! ]\n...\n------------------------------------------------\n\u003cPID\u003e is leaving the kernel with locks still held!\n1 lock held by \u003cPID\u003e:\n#0: (sb_internal){......}, at: [\u003cffffffffa07692a3\u003e] xfs_trans_alloc+0xe3/0x130 [xfs]\n\nThe failure/shutdown occurs during deferred ops processing which\nleads to an error return from xfs_defer_finish() via\nxfs_attr_leaf_addname(). While the root cause of the failure is\nunknown corruption, the cause of the subsequent BUG above and\nunmount hang is failure to cancel the transaction before returning\nto userspace.\n\nThe transaction is not cancelled because the out_defer_cancel error\nhandling paths in the xfs_attr_[leaf|node]_[add|remove]name()\nfunctions clear args.trans without releasing the transaction. The\ncallers therefore lose the reference to the transaction and fail to\ncancel it.\n\nSince xfs_attr_[set|remove]() always cancel args.trans when !\u003d NULL\nand xfs_defer_finish()-\u003e...-\u003exfs_trans_roll() should always return\nwith a valid transaction, update the leaf/node xattr functions to\nnot reset args.trans in the error path responsible for cancelling\ndeferred ops.\n\nReported-by: Chris Dunlop \u003cchris@onthe.net.au\u003e\nSigned-off-by: Brian Foster \u003cbfoster@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "8504509d9e0d93c7ed1dd0bb44925c9c17efe8d5", "tree": "2a88ecb84c9cbd271a81f674e57a00d595c4cf58", "parents": [ "dfcfc5f1aac7eb29d94eca4d45d9254eb96499a8" ], "author": { "name": "Brian Foster", "email": "bfoster@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: account finobt blocks properly in perag reservation\n\nSource kernel commit: ad90bb585c45917b6c1bb01c812fba337e689362\n\nXFS started using the perag metadata reservation pool for free inode\nbtree blocks in commit 76d771b4cbe33 (\"xfs: use per-AG reservations\nfor the finobt\"). To handle backwards compatibility, finobt blocks\nare accounted against the pool so long as the full reservation is\navailable at mount time. Otherwise the -\u003em_inotbt_nores flag is set\nand the filesystem falls back to the traditional per-transaction\nfinobt reservation.\n\nThis commit has two problems:\n\n- finobt blocks are always accounted against the metadata\nreservation on allocation, regardless of -\u003em_inotbt_nores state\n- finobt blocks are never returned to the reservation pool on free\n\nThe first problem affects reflink+finobt filesystems where the full\nfinobt reservation is not available at mount time. finobt blocks are\nessentially stolen from the reflink reservation, putting refcountbt\nmanagement at risk of allocation failure. The second problem is an\nunconditional leak of metadata reservation whenever finobt is\nenabled.\n\nUpdate the finobt block allocation callouts to consider\n-\u003em_inotbt_nores and account blocks appropriately. Blocks should be\nconsistently accounted against the metadata pool when\n-\u003em_inotbt_nores is false and otherwise tagged as RESV_NONE.\n\nSigned-off-by: Brian Foster \u003cbfoster@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "dfcfc5f1aac7eb29d94eca4d45d9254eb96499a8", "tree": "f30d99336d4fbda8689b0c6749f877c699d29bbe", "parents": [ "d7e71605e73f5c05ac029046d56781b675311592" ], "author": { "name": "Colin Ian King", "email": "colin.king@canonical.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: fix check on struct_version for versions 4 or greater\n\nSource kernel commit: a8789a5ae28f69d7f3791a0e74f8c44222f3108b\n\nIt appears that the check for versions 4 or more is incorrect and is\noff-by-one. Fix this.\n\nDetected by CoverityScan, CID#1463775 (\"Logically dead code\")\n\nFixes: ac503a4cc9e8 (\"xfs: refactor the geometry structure filling function\")\nSigned-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "d7e71605e73f5c05ac029046d56781b675311592", "tree": "297be88297f165feb76e23eeb28cd3da1665e72c", "parents": [ "ef4109d1b66633808ed945c13e6478a5eb09952f" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: use %px for data pointers when debugging\n\nSource kernel commit: c96900435fa9fdfd9702a60cd765bd85e380303e\n\nStarting with commit 57e734423ad (\"vsprintf: refactor %pK code out of\npointer\"), the behavior of the raw \u0027%p\u0027 printk format specifier was\nchanged to print a 32-bit hash of the pointer value to avoid leaking\nkernel pointers into dmesg. For most situations that\u0027s good.\n\nThis is /undesirable/ behavior when we\u0027re trying to debug XFS, however,\nso define a PTR_FMT that prints the actual pointer when we\u0027re in debug\nmode.\n\nNote that %p for tracepoints still prints the raw pointer, so in the\nlong run we could consider rewriting some of these messages as\ntracepoints.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "ef4109d1b66633808ed945c13e6478a5eb09952f", "tree": "d9c8d1aa3a1d474fbb6f664e00078581225c2131", "parents": [ "493b46c4d9a1dbf636a1f3452c1ac36232ede6f4" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:18 2018 -0600" }, "message": "xfs: change 0x%p -\u003e %p in print messages\n\nSource kernel commit: 3d170aa24283568b1ed92a09daa0e05a8788c6a4\n\nSince %p prepends \"0x\" to the outputted string, we can drop the prefix.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "493b46c4d9a1dbf636a1f3452c1ac36232ede6f4", "tree": "9dde8a64d39617d2d408ae76cced5cf683c2b35d", "parents": [ "fdef0e8b6a5e20f405526d5a2979fc2695f8d62d" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: harden directory integrity checks some more\n\nSource kernel commit: 46c59736d8090e602f960aeaf1c6b8292151bf38\n\nIf a malicious filesystem image contains a block+ format directory\nwherein the directory inode\u0027s core.mode is set such that\nS_ISDIR(core.mode) \u003d\u003d 0, and if there are subdirectories of the\ncorrupted directory, an attempt to traverse up the directory tree will\ncrash the kernel in __xfs_dir3_data_check. Running the online scrub\u0027s\nparent checks will tend to do this.\n\nThe crash occurs because the directory inode\u0027s d_ops get set to\nxfs_dir[23]_nondir_ops (it\u0027s not a directory) but the parent pointer\nscrubber\u0027s indiscriminate call to xfs_readdir proceeds past the ASSERT\nif we have non fatal asserts configured.\n\nFix the null pointer dereference crash in __xfs_dir3_data_check by\nlooking for S_ISDIR or wrong d_ops; and teach the parent scrubber\nto bail out if it is fed a non-directory \"parent\".\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "fdef0e8b6a5e20f405526d5a2979fc2695f8d62d", "tree": "0bf051e658323dc2eacfbfc31a098f838f0bc238", "parents": [ "a2a523844910af899af620bfe25aef42455c6d66" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: refactor the geometry structure filling function\n\nSource kernel commit: ac503a4cc9e8ab574032e3e217ffb555f5bf2341\n\nRefactor the geometry structure filling function to use the superblock\nto fill the fields. While we\u0027re at it, make the function less indenty\nand use some whitespace to make the function easier to read.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "a2a523844910af899af620bfe25aef42455c6d66", "tree": "7a8888718c56089ac4569e8e7d811e7fcacdef2f", "parents": [ "079e8efb3ef7e5491cd28b10857dd14b58b380d6" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: hoist xfs_fs_geometry to libxfs\n\nSource kernel commit: c368ebcd4cc3bbc08602adce083ad3cc76a15258\n\nMove xfs_fs_geometry to libxfs so that we can clean up the fs geometry\nreporting in xfsprogs.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "079e8efb3ef7e5491cd28b10857dd14b58b380d6", "tree": "a84a3d75459f4d8d146a1ca3dcb258508f2453a2", "parents": [ "71ffd552b499cc420aa2599ba0a2a8699551d966" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: trace log reservations at mount time\n\nSource kernel commit: b872af2c8700e9d64af8e13811b7679ede26ca00\n\nAt each mount, emit the transaction reservation type information via\ntracepoints. This makes it easier to compare the log reservation info\ncalculated by the kernel and xfsprogs so that we can more easily diagnose\nminimum log size failures on freshly formatted filesystems.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "71ffd552b499cc420aa2599ba0a2a8699551d966", "tree": "b3907fcaca22fd61c2a26c1009e5c5d76e8c368b", "parents": [ "055b84e39cabe4e2242b37bd2554908e10dd9d06" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: standardize quota verification function outputs\n\nSource kernel commit: eebf3cab9c5eac7fdb54fb9e9fb38c06f46f17f3\n\nRename xfs_dqcheck to xfs_dquot_verify and make it return an\nxfs_failaddr_t like every other structure verifier function.\nThis enables us to check on-disk quotas in the same way that we check\neverything else. Callers are now responsible for logging errors, as\nXFS_QMOPT_DOWARN goes away.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "055b84e39cabe4e2242b37bd2554908e10dd9d06", "tree": "69eb9e94dbd358df6414e5f4db067420560cc91a", "parents": [ "95d9582b620a1e91ef4cf532597b9edbfb8ea793" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: separate dquot repair into a separate function\n\nSource kernel commit: eeea79802871fef82a8ca6ab1220515855e5cdcc\n\nMove the dquot repair code into a separate function and remove\nXFS_QMOPT_DQREPAIR in favor of calling the helper directly. Remove\nother dead code because quotacheck is the only caller of DQREPAIR.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "95d9582b620a1e91ef4cf532597b9edbfb8ea793", "tree": "d1fcb9d273b952f51e15f7cb86a8f5d09d998aff", "parents": [ "e9998c63dc86706bee6fd61d2dd5d01530f8d1cc" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: create a new buf_ops pointer to verify structure metadata\n\nSource kernel commit: b55725974c9d3a5afcdf83daff6fba7d3f91ffca\n\nExpose all metadata structure buffer verifier functions via buf_ops.\nThese will be used by the online scrub mechanism to look for problems\nwith buffers that are already sitting around in memory.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "e9998c63dc86706bee6fd61d2dd5d01530f8d1cc", "tree": "45f203e984656d9d5594ba54faa29b9c84c53155", "parents": [ "20e882d4fe6a7fdd0efbb61b3f10fb5b8ee8316a" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: fail out of xfs_attr3_leaf_lookup_int if it looks corrupt\n\nSource kernel commit: 8ba92d43d499f4920af983a7c16e02304dd36932\n\nIf the xattr leaf block looks corrupt, return -EFSCORRUPTED to userspace\ninstead of ASSERTing on debug kernels or running off the end of the\nbuffer on regular kernels.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "20e882d4fe6a7fdd0efbb61b3f10fb5b8ee8316a", "tree": "e985a33de3b5703f7c1ebfd14f9f2d50e883242f", "parents": [ "1f98251ffd56f2d78fff582568e984dec952d89c" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: provide a centralized method for verifying inline fork data\n\nSource kernel commit: 9cfb9b47479e237d217dbcfafe034cbf98f45909\n\nReplace the current haphazard dir2 shortform verifier callsites with a\ncentralized verifier function that can be called either with the default\nverifier functions or with a custom set. This helps us strengthen\nintegrity checking while providing us with flexibility for repair tools.\n\nxfs_repair wants this to be able to supply its own verifier functions\nwhen trying to fix possibly corrupt metadata.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "1f98251ffd56f2d78fff582568e984dec952d89c", "tree": "80a2f611826575a01725767655487ef29c88d620", "parents": [ "6db3a8007a2493d6887cf45a5edfaa6a770b8492" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: refactor short form directory structure verifier function\n\nSource kernel commit: dc042c2d8ff629dd411e9a60bce9c379e2f8aaf8\n\nChange the short form directory structure verifier function to return\nthe instruction pointer of a failing check or NULL if everything\u0027s ok.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" }, { "commit": "6db3a8007a2493d6887cf45a5edfaa6a770b8492", "tree": "7cc638c851f98cd563236a261e780f6619f3e597", "parents": [ "c7a710b7beeb21c5d84ab5e3c6a6aa9c96631fcc" ], "author": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "committer": { "name": "Eric Sandeen", "email": "sandeen@redhat.com", "time": "Mon Feb 26 22:43:17 2018 -0600" }, "message": "xfs: create structure verifier function for short form symlinks\n\nSource kernel commit: 0795e004fd4f2723f3dbf09a195cd7ccf3c74c58\n\nCreate a function to check the structure of short form symlink targets.\n\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\n\n" } ], "next": "c7a710b7beeb21c5d84ab5e3c6a6aa9c96631fcc" }