)]}' { "log": [ { "commit": "429f731dea577bdd43693940cdca524135287e6a", "tree": "932337b7d23dd736612c6b6e6601c50bc24cfc5f", "parents": [ "85b375a613085b78531ec86369a51c2f3b922f95", "d2f5e80862d33a7746a40569840248a592dbc395" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 21 15:41:27 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 21 15:41:27 2008 -0700" }, "message": "Merge branch \u0027semaphore\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/willy/misc\n\n* \u0027semaphore\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/willy/misc:\n Deprecate the asm/semaphore.h files in feature-removal-schedule.\n Convert asm/semaphore.h users to linux/semaphore.h\n security: Remove unnecessary inclusions of asm/semaphore.h\n lib: Remove unnecessary inclusions of asm/semaphore.h\n kernel: Remove unnecessary inclusions of asm/semaphore.h\n include: Remove unnecessary inclusions of asm/semaphore.h\n fs: Remove unnecessary inclusions of asm/semaphore.h\n drivers: Remove unnecessary inclusions of asm/semaphore.h\n net: Remove unnecessary inclusions of asm/semaphore.h\n arch: Remove unnecessary inclusions of asm/semaphore.h\n" }, { "commit": "ec82687f29127a954dd0da95dc1e0a4ce92b560c", "tree": "952b4f3f7128e2f5798607d595fb43d89c458e17", "parents": [ "2f676cbc0d60ae806216c7a61c6971bd72dedde8" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Fri Feb 15 14:37:53 2008 -0800" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat Apr 19 00:29:26 2008 -0400" }, "message": "[PATCH] r/o bind mounts: elevate count for xfs timestamp updates\n\nElevate the write count during the xfs m/ctime updates.\n\nXFS has to do it\u0027s own timestamp updates due to an unfortunate VFS\ndesign limitation, so it will have to track writers by itself aswell.\n\n[hch: split out from the touch_atime patch as it\u0027s not related to it at all]\n\nAcked-by: Al Viro \u003cviro@ZenIV.linux.org.uk\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "42a74f206b914db13ee1f5ae932dcd91a77c8579", "tree": "24e3dbe55edaacc750067ab9e01778255a6bff08", "parents": [ "20ddee2c75339cc095f6191c3115f81da8955e96" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Fri Feb 15 14:37:46 2008 -0800" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat Apr 19 00:29:24 2008 -0400" }, "message": "[PATCH] r/o bind mounts: elevate write count for ioctls()\n\nSome ioctl()s can cause writes to the filesystem. Take these, and make them\nuse mnt_want/drop_write() instead.\n\n[AV: updated]\n\nAcked-by: Al Viro \u003cviro@ZenIV.linux.org.uk\u003e\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "6188e10d38b8d7244ee7776d5f1f88c837b4b93f", "tree": "6675e02c47fb684a613e10513816b5dcc5de28a9", "parents": [ "d7b1acd3b524b39f418e463e836b48ac041954d6" ], "author": { "name": "Matthew Wilcox", "email": "matthew@wil.cx", "time": "Fri Apr 18 22:21:05 2008 -0400" }, "committer": { "name": "Matthew Wilcox", "email": "willy@linux.intel.com", "time": "Fri Apr 18 22:22:54 2008 -0400" }, "message": "Convert asm/semaphore.h users to linux/semaphore.h\n\nSigned-off-by: Matthew Wilcox \u003cwilly@linux.intel.com\u003e\n" }, { "commit": "65e67f5165c8a156b34ee7adf65d5ed3b16a910d", "tree": "6b2e82205ae59b2304d38880c15e62a229d70691", "parents": [ "3b2816be271b8b364294a5b48721a3e68af46cfa" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:59:45 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:59:45 2008 +1000" }, "message": "[XFS] Fix merge failure\n" }, { "commit": "3b2816be271b8b364294a5b48721a3e68af46cfa", "tree": "1e40214091bb6736249d3ce442a13ef7e9dfdc82", "parents": [ "f6e9f28865552bd9d79a9df93cf120436b073223" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:43:35 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:43:35 2008 +1000" }, "message": "[XFS] The forward declarations for the xfs_ioctl() helpers and the\nassociated comment about gcc behavior really aren\u0027t needed; all of these\nfunctions are marked STATIC which includes noinline, and the stack usage\nwon\u0027t be a problem.\n\nThis effectively just removes the forward declarations and moves\nxfs_ioctl() back to the end of the file.\n\nSGI-PV: 971186\nSGI-Modid: xfs-linux-melb:xfs-kern:30534a\n\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "e687330b5ed1ea899fdaf0dea50aba196b6e019a", "tree": "ac74d31ff84023d03939c60e1e2acce59b4fe1e7", "parents": [ "f7d3c34788696f5ba9ac9fa414ad80e2a91d4b2e" ], "author": { "name": "Donald Douwsma", "email": "donaldd@sgi.com", "time": "Thu Apr 17 16:50:28 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:04:29 2008 +1000" }, "message": "[XFS] Remove unused HAVE_SPLICE macro.\n\nHAVE_SPLICE was part of the infrastructure for building 2.4 and 2.6\nkernels out of the same tree. Now we don\u0027t build 2.4 kernels this\n\nSGI-PV: 971046\nSGI-Modid: xfs-linux-melb:xfs-kern:30878a\n\nSigned-off-by: Donald Douwsma \u003cdonaldd@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "f7d3c34788696f5ba9ac9fa414ad80e2a91d4b2e", "tree": "0e8d743bdb6344a20b61a8daeb23bf3b508ff573", "parents": [ "6d1337b29bf09a97682d39db36ac2d0dfc6659c0" ], "author": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Thu Apr 17 16:50:22 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:04:19 2008 +1000" }, "message": "[XFS] Remove CONFIG_XFS_SECURITY.\n\nThere is no point to the CONFIG_XFS_SECURITY option; it disables the\nability to set security attributes at runtime, but it does not actually\nslim down or remove any code for runtime. Just remove it and always allow\nsecurity attributes to be set.\n\nSGI-PV: 980310\nSGI-Modid: xfs-linux-melb:xfs-kern:30877a\n\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Tim Shimmin \u003ctes@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "6d1337b29bf09a97682d39db36ac2d0dfc6659c0", "tree": "ead8e985e731f984b9ba1f31568518206ad2592a", "parents": [ "cb49dbb130e17a6f9af4cb4714cf6976cf09afdf" ], "author": { "name": "Tim Shimmin", "email": "tes@sgi.com", "time": "Thu Apr 17 16:50:16 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:04:08 2008 +1000" }, "message": "[XFS] xfs_bmap_compute_maxlevels should be based on di_forkoff\n\nFix up xfs_bmap_compute_maxlevels() to account for the case when we go\nfrom using attr2 to using attr1. In that case attr1 will no longer\nnecessarily be at m_attr_offset\u003e\u003e3, but could be at a different value for\ndi_forkoff. Therefore, we return the worst case scenario using MINDBTPTRS\nand MINABTPTRS, as this function is used for determining the maximum log\nspace.\n\nSGI-PV: 979606\nSGI-Modid: xfs-linux-melb:xfs-kern:30862a\n\nSigned-off-by: Tim Shimmin \u003ctes@sgi.com\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "cb49dbb130e17a6f9af4cb4714cf6976cf09afdf", "tree": "33e0c38d5b4f67455804c793e625a0f7e91aec5c", "parents": [ "f6485057c5cfbc84e5eff639ddea1ce0d668607b" ], "author": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Thu Apr 17 16:50:09 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:03:40 2008 +1000" }, "message": "[XFS] Always use di_forkoff when checking for attr space.\n\nIn the case where we mount a filesystem which was previously using the\nattr2 format as attr1, returning the default mp-\u003em_attroffset instead of\nthe per-inode di_forkoff for inline attribute fit calculations, may result\nin corruption, if for example, the data fork is already taking more space\nthan the default fork offset and we try to add an extended attribute. Fix\ntested by xfstests/186.\n\nSGI-PV: 979606\nSGI-Modid: xfs-linux-melb:xfs-kern:30861a\n\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Tim Shimmin \u003ctes@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "f6485057c5cfbc84e5eff639ddea1ce0d668607b", "tree": "5366b91b26379b32e900fdccbac26edf71bca1b9", "parents": [ "7e20694d91f817f8e9f62404aca793ae0df4d98a" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 17 16:50:04 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:03:26 2008 +1000" }, "message": "[XFS] Ensure the inode is joined in xfs_itruncate_finish\n\nOn success, we still need to join the inode to the current transaction in\nxfs_itruncate_finish(). Fixes regression from error handling changes.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30845a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "7e20694d91f817f8e9f62404aca793ae0df4d98a", "tree": "36831242dad3429ea4bb56f681ebd8c82d3d2200", "parents": [ "e6430037e9fd0b3d02ceaf5ab99bfe3ccb763be7" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 17 16:49:55 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:03:12 2008 +1000" }, "message": "[XFS] Remove periodic logging of in-core superblock counters.\n\nxfssyncd triggers the logging of superblock counters every 30s if the\nfilesystem is made with lazy-count\u003d1. This will prevent disks from idling\nand spinning down as there will be a log write every 30s. With the way\ncounter recovery works for lazy-count\u003d1, this code is unnecessary and\nprovides no real benefit, so just remove it.\n\nSGI-PV: 980145\nSGI-Modid: xfs-linux-melb:xfs-kern:30840a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Barry Naujok \u003cbnaujok@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "e6430037e9fd0b3d02ceaf5ab99bfe3ccb763be7", "tree": "82af4af56d4469ef8fef545921c560b765ff036f", "parents": [ "d4055947bd0913864f4d8ac96bf1197338071622" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 17 16:49:49 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:03:02 2008 +1000" }, "message": "[XFS] fix logic error in xfs_alloc_ag_vextent_near()\n\nFix a logic error in xfs_alloc_ag_vextent_near(). This is a regression\nintroduced by the error handling changes.\n\nSGI-PV: 890084\nSGI-Modid: xfs-linux-melb:xfs-kern:30838a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Barry Naujok \u003cbnaujok@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "d4055947bd0913864f4d8ac96bf1197338071622", "tree": "260f91444acd5cc81999b08ba32dd40be4a078c0", "parents": [ "1bb7d6b5a82f1d9487fd44415484a368f7c87bed" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 17 16:49:35 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:02:41 2008 +1000" }, "message": "[XFS] Don\u0027t error out on good I/Os.\n\nxfsbdstrat() made all I/Os error out, good or bad. Fix it.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30836a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Donald Douwsma \u003cdonaldd@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "1bb7d6b5a82f1d9487fd44415484a368f7c87bed", "tree": "7893e95a0f4b4aaba61dc1fffd09f6fb9ed19822", "parents": [ "b911ca0472c3762d2bafc4d21e432a9056844064" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:24:38 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:02:30 2008 +1000" }, "message": "[XFS] Catch log unmount failures.\n\nUnmounting the log can fail. unlikely, but it can. Catch all the error\nconditions an make sure it\u0027s propagated upwards.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30833a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "b911ca0472c3762d2bafc4d21e432a9056844064", "tree": "4ed23766782938e7316d087fa542cdbfd5fcadd7", "parents": [ "234f56aca20a4f66b6ba3d3bf2787634dd9e0999" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:24:30 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:02:20 2008 +1000" }, "message": "[XFS] Sanitise xfs_log_force error checking.\n\nxfs_log_force() is declared to return an error, but we almost never check\nit. We don\u0027t need to check it in most cases; if there\u0027s a log I/O error\nthen we\u0027ll be shutting down the filesystem anyway and that means we\u0027ll\ncatch the error somewhere else.\n\nHowever, on certain calls we should be returning an error - sync\ntransactions, fsync, sync writes, etc. so this isn\u0027t a pure black and\nwhite distinction. Hence make xfs_log_force() a void function that issues\na warning to the syslog on error, and call _xfs_log_force() in all the\nplaces where we actually care about the error status returned.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30832a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "234f56aca20a4f66b6ba3d3bf2787634dd9e0999", "tree": "5830e04df2a24cf988ad22abe960fd45fb8a5850", "parents": [ "78e9da77f1bf265fe750b9223ec15707473fb6e8" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:24:24 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:02:10 2008 +1000" }, "message": "[XFS] Check for errors when changing buffer pointers.\n\nxfs_buf_associate_memory() can fail, but the return is never checked.\nPropagate the error through XFS_BUF_SET_PTR() so that failures are\ndetected.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30831a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "78e9da77f1bf265fe750b9223ec15707473fb6e8", "tree": "8191da162b698d5b7aee7ea382d3cef6a56a7f0d", "parents": [ "64bfe1bfae833e89ed77f72c61ded19f4b1976f8" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:24:17 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:01:58 2008 +1000" }, "message": "[XFS] Don\u0027t allow silent errors in xfs_inactive().\n\nxfs_inactive() fails to report errors when committing the inactive\ntransaction. Hence we can get silent failures either finishing off the\ntruncation or committing the transaction. Even if we get errors, we need\nto continue, so simply warn loudly to the system if we get errors here.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30830a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "64bfe1bfae833e89ed77f72c61ded19f4b1976f8", "tree": "7560fc6e97a493a2b9066a23a69cf44ac2df81ce", "parents": [ "7b07339048f7b020575706b492c004b5664b67ab" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:24:10 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:01:39 2008 +1000" }, "message": "[XFS] Catch errors from xfs_imap().\n\nCatch errors from xfs_imap() in log recovery when we might be trying to\nmap an invalid inode number due to a corrupted log.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30829a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "7b07339048f7b020575706b492c004b5664b67ab", "tree": "f96a0067ca65ba541b6dbbcdd147df392b47db7b", "parents": [ "e4ac967b117c5780760abbd9ae996210c31cb398" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:24:04 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:01:27 2008 +1000" }, "message": "[XFS] xfs_bulkstat_one_dinode() never returns an error.\n\nMark it void.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30828a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "e4ac967b117c5780760abbd9ae996210c31cb398", "tree": "56e6a0d49da0981be34b4c2908c138ae6214aa4c", "parents": [ "cc88466f3f67bb16fc91b0b974e51c2a43a9e597" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:23:58 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:01:11 2008 +1000" }, "message": "[XFS] xfs_iflush_fork() never returns an error.\n\nxfs_iflush_fork() never returns an error. Mark it void and clean up the\ncode calling it that checks for errors.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30827a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "cc88466f3f67bb16fc91b0b974e51c2a43a9e597", "tree": "a3cb1c8ff110046aaca771330810697702ebb30d", "parents": [ "958d4ec606d4af590f86a601a238613f21e878ee" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:23:52 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:00:58 2008 +1000" }, "message": "[XFS] Catch unwritten extent conversion errors.\n\nOn unwritten I/O completion, we fail to propagate an error when converting\nthe extent to a written extent. This means that the I/O silently fails.\npropagate the error onto the ioend so that the inode is marked with an\nerror appropriately.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30826a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "958d4ec606d4af590f86a601a238613f21e878ee", "tree": "06ac01c7a29cab542fa253fbd71915773f35fb1d", "parents": [ "db7a19f2c89d99b66874a7e0c0dc681ff1f37b4e" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:23:46 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:00:46 2008 +1000" }, "message": "[XFS] xfs_bdwrite() does not return errors.\n\nxfs_bdwrite() cannot return an error; it only queues buffers to the\ndelayed write list and as such never encounters anything that can fail.\nMark it void.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30825a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "db7a19f2c89d99b66874a7e0c0dc681ff1f37b4e", "tree": "dd242710a41839617eae7a8b7e71f6effb4d608c", "parents": [ "d64e31a2f53cdcb2f95b782196faacb0995ca0c0" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:22:24 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:00:35 2008 +1000" }, "message": "[XFS] Ensure xfs_bawrite() errors are checked.\n\nxfs_bawrite() can return immediate error status on async writes. Unlike\nxfsbdstrat() we don\u0027t ever check the error on the buffer after the call,\nso we currently do not catch errors at all here. Ensure we catch and\npropagate or warn to the syslog about up-front async write errors.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30824a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "d64e31a2f53cdcb2f95b782196faacb0995ca0c0", "tree": "d13230aeeb739f263305c686168823e161bdf619", "parents": [ "556b8b166c9514b5f940047a41dad8fe8cd9a778" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:22:17 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:00:24 2008 +1000" }, "message": "[XFS] Ensure errors from xfs_bdstrat() are correctly checked.\n\nxfsbdstrat() is declared to return an error. That is never checked because\nthe error is propagated by the xfs_buf_t that is passed through the\nfunction.\n\nMark xfsbdstrat() as returning void and comment the prototype on the\nmethods needed for error checking.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30823a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "556b8b166c9514b5f940047a41dad8fe8cd9a778", "tree": "22fdec0d211d9c93a6d8d60b591591f7c9a8de1b", "parents": [ "7c9ef85c5672ae316aafd7bbe0bbadebe90301e6" ], "author": { "name": "Barry Naujok", "email": "bnaujok@sgi.com", "time": "Thu Apr 10 12:22:07 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 12:00:12 2008 +1000" }, "message": "[XFS] remove bhv_vname_t and xfs_rename code\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30804a\n\nSigned-off-by: Barry Naujok \u003cbnaujok@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "7c9ef85c5672ae316aafd7bbe0bbadebe90301e6", "tree": "5f89b57f3872c4cceb534953d59a9a12ee60ef6a", "parents": [ "fc6149d8d9634814cdcd9283b8f2efd3359181df" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:59 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:59:45 2008 +1000" }, "message": "[XFS] Catch errors returned from xfs_bmap_last_offset().\n\nxfs_bmap_last_offset() can fail and return an error.\nxfs_iomap_write_allocate() fails to detect and propagate the error.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30802a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "fc6149d8d9634814cdcd9283b8f2efd3359181df", "tree": "601c3ad274aae19a264009a385b8eb95977ada9a", "parents": [ "d87dd6360dce86cad9099aed74f14b4dd0143301" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:53 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:59:23 2008 +1000" }, "message": "[XFS] Check for xfs_free_extent() failing.\n\nxfs_free_extent() can fail, but log recovery never bothers to check if it\nsuccessfully free the extent it was supposed to. This could lead to silent\ncorruption during log recovery. Abort log recovery if we fail to free an\nextent.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30801a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "d87dd6360dce86cad9099aed74f14b4dd0143301", "tree": "743ea201fa895e75be79f357b0246c35a0726dcc", "parents": [ "c2b1cba6833da77b1b478ac144f9cf5144d276ec" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:46 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:59:12 2008 +1000" }, "message": "[XFS] Warn if errors come from block_truncate_page().\n\nblock_truncate_page() can return errors that we currently ignore and\nsilently discard. We should not ever get errors reported here - an error\nindicates a bug somewhere else. Hence catch the error and issue a stack\ndump to the syslog because we cannot propagate the error any further up\nthe call chain.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30800a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "c2b1cba6833da77b1b478ac144f9cf5144d276ec", "tree": "6e5bd6ccdc83cb6dbc13ea2807b7d54e9279868c", "parents": [ "12375c82375ec39ec948a3ad62e5e77533515e83" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:40 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:58:46 2008 +1000" }, "message": "[XFS] xfs_bmap_adjacent() never returns an error.\n\nMark it void.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30798a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "12375c82375ec39ec948a3ad62e5e77533515e83", "tree": "0122f66735d75583dcb6267be96175271a1e81d3", "parents": [ "f4586e40613a9f8bb9f7f9c8a796062a9ab1614c" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:32 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:58:36 2008 +1000" }, "message": "[XFS] Make xfs_alloc_compute_aligned() void.\n\nxfs_alloc_compute_aligned() returns a value based on a comparison of the\ncomputed extent length and the minimum length allowed. This is only used\nby some callers - the other four return parameters are used more often.\nHence move the comparison to the code that actually needs to do it and\nmake xfs_alloc_compute_aligned() a void function.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30797a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "f4586e40613a9f8bb9f7f9c8a796062a9ab1614c", "tree": "4c8594c056057afd26bb5af1ee41dcd11c4d4dfc", "parents": [ "e5720eec0548c08943d759e39db0388d8fe59287" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:25 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:58:27 2008 +1000" }, "message": "[XFS] Clean up xfs_alloc_search_busy() return values.\n\nxfs_alloc_search_busy() returns an index into the busy array if the extent\nwas found in the array. This is never checked, and the\nxfs_alloc_search_busy() does a log force to prevent reuse of the extent\nbefore the free transaction hits the disk. Hence the return value is\nuseless. Declare the function void and remove the slot number from the\ntracing as well.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30796a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "e5720eec0548c08943d759e39db0388d8fe59287", "tree": "e38b474f0dbac30aee7141878953223a2a588c69", "parents": [ "3c1e2bbe5bcdcd435510a05eb121fa74b848e24f" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:18 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:58:17 2008 +1000" }, "message": "[XFS] Propagate errors from xfs_trans_commit().\n\nxfs_trans_commit() can return errors when there are problems in the\ntransaction subsystem. They are indicative that the entire transaction may\nbe incomplete, and hence the error should be propagated as there is a good\npossibility that there is something fatally wrong in the filesystem. Catch\nand propagate or warn about commit errors in the places where they are\ncurrently ignored.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30795a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "3c1e2bbe5bcdcd435510a05eb121fa74b848e24f", "tree": "1b53868f47e95afde9317964c642db0faa4b36ba", "parents": [ "5ca1f261a08d5cff5f29eaa0887b59baae2ae7f7" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:11 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:58:08 2008 +1000" }, "message": "[XFS] Propagate xfs_trans_reserve() errors.\n\nxfs_trans_reserve() reports errors that should not be ignored. For\nexample, a shutdown filesystem will report errors through\nxfs_trans_reserve() to prevent further changes from being attempted on a\ndamaged filesystem. Catch and propagate all error conditions from\nxfs_trans_reserve().\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30794a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "5ca1f261a08d5cff5f29eaa0887b59baae2ae7f7", "tree": "4973597a434d7b4422f102097010e35020b317e0", "parents": [ "0c928299676c8df2b00e75d5691cd4846e6c0868" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:21:04 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:57:57 2008 +1000" }, "message": "[XFS] Catch errors from xfs_acl_vremove().\n\nRemoving an ACL can return an error. Propagate it.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30793a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "0c928299676c8df2b00e75d5691cd4846e6c0868", "tree": "48c4bd6b8157afa7016cb633dc872b15028c2294", "parents": [ "88ab02085363b7c45935d66ab3e969b4fec9a20c" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:58 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:57:46 2008 +1000" }, "message": "[XFS] Catch errors from xfs_acl_setmode().\n\nPropagate the error status from xfs_acl_setmode() so that callers know if\nthe ACl was set correctly or not.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30792a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "88ab02085363b7c45935d66ab3e969b4fec9a20c", "tree": "933391b62fb0712c87feb182ded9219079f46167", "parents": [ "cb6edc26c386d2268dcf61bcdec02b6fb50b6ba2" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:51 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:57:36 2008 +1000" }, "message": "[XFS] Propagate quota file truncation errors.\n\nTruncating the quota files can silently fail. Ensure that truncation\nerrors are propagated to the callers.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30791a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "cb6edc26c386d2268dcf61bcdec02b6fb50b6ba2", "tree": "420f39455fd5665dd1c61ee3d2fde135c1cf5bd0", "parents": [ "31d5577b35d8397dea19f2ba7550e9225605a785" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:45 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:57:26 2008 +1000" }, "message": "[XFS] Catch errors when turning off quotas.\n\nWhen turning off quota, we need to write various transactions to the log\nto ensure that they are cleanly removed in the case of a crash. We need to\ncheck that the transactions hit the disk correctly. If we fail to write\nthe final quota off transaction, we are corrupt in memory and so the only\noption is to shut the filesystem down at this point.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30790a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "31d5577b35d8397dea19f2ba7550e9225605a785", "tree": "9702baeb058a786b3022475dccc9bac81defdb8f", "parents": [ "53aa7915d67b9d0f5986c9f08e76846fedc520d4" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:38 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:57:16 2008 +1000" }, "message": "[XFS] Catch errors resetting quota flags.\n\nWarn to the syslog if we fail to reset the quota flags in the superblock\nwhen a quota check fails.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30789a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "53aa7915d67b9d0f5986c9f08e76846fedc520d4", "tree": "d80c449c1369de8bb92a8abe34506f2810732227", "parents": [ "3c56836f92683cb871ebbf44c512069b0d48a08f" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:31 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:57:05 2008 +1000" }, "message": "[XFS] Clean up quotamount error handling.\n\nxfs_qm_mount_quotas() returns an error status that is ignored. If we fail\nto mount quotas, we continue with quota\u0027s turned off, which is all handled\ninside xfs_qm_mount_quotas(). Mark it as void to indicate that errors need\nnot be returned to the callers.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30788a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "3c56836f92683cb871ebbf44c512069b0d48a08f", "tree": "ac0ef5d76e086aa6e2542fb5548db26af5967297", "parents": [ "4b8879df8c21bed3efd1eb2da5d72501199aba29" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:24 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:56:55 2008 +1000" }, "message": "[XFS] Check for dquot flush errors\n\nxfs_qm_dqflush() can fail, but the return is not checked anywhere. Hence\nwe never know if we\u0027ve failed to flush a dquot to disk. Propagate the\nerror and warn to the syslog if a flush ever fails.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30787a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "4b8879df8c21bed3efd1eb2da5d72501199aba29", "tree": "4a6991a366e93cd1aebf834f8172dbb83d636c6c", "parents": [ "5b1397385bf536cbdb60f3362f44079d15d5f5ee" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:17 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:54:56 2008 +1000" }, "message": "[XFS] Propagate xfs_qm_dqflush_all() errors.\n\nxfs_qm_dqflush_all() can return flush errors. Ensure they are propagated\ninto the quotacheck code to determine if the quotacheck succeeded or not.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30786a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "5b1397385bf536cbdb60f3362f44079d15d5f5ee", "tree": "44ced2daad34684d65744d787425eef364c3660c", "parents": [ "714082bc12b6c305f825411df02177efcb0085f1" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:10 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:54:06 2008 +1000" }, "message": "[XFS] xfs_qm_reset_dqcounts() does not return errors.\n\nDeclare it void.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30785a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "714082bc12b6c305f825411df02177efcb0085f1", "tree": "2fd7f10bc2f3aeca24457534cf716db6a47fdf26", "parents": [ "36fbe6e6bd5408b09341043dfece978b4a7a0f34" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:20:03 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:53:51 2008 +1000" }, "message": "[XFS] Report errors from xfs_reserve_blocks().\n\nxfs_reserve_blocks() can fail in interesting ways. In neither case is it a\nfatal error, but the result can lead to sub-optimal behaviour. Warn to the\nsyslog if the call fails but otherwise continue.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30784a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "36fbe6e6bd5408b09341043dfece978b4a7a0f34", "tree": "0d04ed7a9906e2a464f660c6c6420f42549b1056", "parents": [ "a414047fc97aea7db6237176ce00013117839cd5" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:19:56 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:52:43 2008 +1000" }, "message": "[XFS] xfs_icsb_counter_disabled() never returns an error.\n\nMark it void.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30782a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "a414047fc97aea7db6237176ce00013117839cd5", "tree": "126e4bbd1162b40c2b29ef91e1d4a0a6912e5755", "parents": [ "3c85c36cc2e87018d38fcd033f41bbdf1360c07a" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:19:47 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:51:58 2008 +1000" }, "message": "[XFS] Remove useless whitespace in function prototypes\n\nMakes it simpler to annotate function prototypes with __must_check via sed\nscripts.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30781a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "3c85c36cc2e87018d38fcd033f41bbdf1360c07a", "tree": "c0beaf03831f981a6ea22aca84ccb4df2270ec54", "parents": [ "b6ddc4e6fed9c6f4adb273c8b36e1731f90ec17e" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:19:40 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:51:46 2008 +1000" }, "message": "[XFS] xfs_quiesce_fs() never returns an error. Mark it void.\n\nSGI-PV: 980084\nSGI-Modid: xfs-linux-melb:xfs-kern:30780a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "b6ddc4e6fed9c6f4adb273c8b36e1731f90ec17e", "tree": "d12b3d657b9fac6f5c33b271a093be25a9fa2284", "parents": [ "34a622b2e1c8e11c8990184634f101c1aad42fec" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Apr 10 12:19:27 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:51:36 2008 +1000" }, "message": "[XFS] Don\u0027t validate symlink target component length\n\nThis target component validation is not POSIX conformant and it is not\ndone by any other Linux filesystem so remove it from XFS.\n\nSGI-PV: 980080\nSGI-Modid: xfs-linux-melb:xfs-kern:30776a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "34a622b2e1c8e11c8990184634f101c1aad42fec", "tree": "4a82ae6c61bf5d4cf90b931a84362b154e04184d", "parents": [ "0225da1f35df46c67785eb08526995d7cdb4e3b0" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Thu Apr 10 12:19:21 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:51:26 2008 +1000" }, "message": "[XFS] replace remaining __FUNCTION__ occurrences\n\n__FUNCTION__ is gcc-specific, use __func__\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30775a\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "0225da1f35df46c67785eb08526995d7cdb4e3b0", "tree": "6af0fb2b99f249b90748d8f5edd36e918a35acb4", "parents": [ "6b1d1a732f886936fe515d911b1a01d9cc50e179" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Thu Apr 10 12:19:10 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:51:15 2008 +1000" }, "message": "[XFS] Replace __inline with inline\n\nRemove the remaining uses of __inline in the XFS code base.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30774a\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "6b1d1a732f886936fe515d911b1a01d9cc50e179", "tree": "ecfbef2372320ebf527dc7a70100d8d208d4e078", "parents": [ "4679b2d36d53ed508c956337972fbbea8db99a77" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:19:02 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:51:04 2008 +1000" }, "message": "[XFS] Fix lock inversion in forced shutdown.\n\nRecent changes to xlog_state_release_iclog() placed the grant_lock inside\nthe icloglock. forced unmount of the log does this the opposite way\naround, but does not depend on the order for correct working. Fix the\ninversion by changing the order locks are gained in\nxfs_log_force_umount().\n\nSGI-PV: 979661\nSGI-Modid: xfs-linux-melb:xfs-kern:30773a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "4679b2d36d53ed508c956337972fbbea8db99a77", "tree": "a9b0449c2b0566bb8e8398ff6644def57f7455e3", "parents": [ "eb01c9cd87c7a9998c2edf209721ea069e3e3652" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:18:54 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:50:53 2008 +1000" }, "message": "[XFS] Reorganise xlog_t for better cacheline isolation of contention\n\nTo reduce contention on the log in large CPU count, separate out different\nparts of the xlog_t structure onto different cachelines. Move each lock\nonto a different cacheline along with all the members that are\naccessed/modified while that lock is held.\n\nAlso, move the debugging code into debug code.\n\nSGI-PV: 978729\nSGI-Modid: xfs-linux-melb:xfs-kern:30772a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "eb01c9cd87c7a9998c2edf209721ea069e3e3652", "tree": "201fc07dac684d6a5a789995c8c35918ea16de7f", "parents": [ "114d23aae51233b2bc62d8e2a632bcb55de1953d" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:18:46 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:50:39 2008 +1000" }, "message": "[XFS] Remove the xlog_ticket allocator\n\nThe ticket allocator is just a simple slab implementation internal to the\nlog. It requires the icloglock to be held when manipulating it and this\ncontributes to contention on that lock.\n\nJust kill the entire allocator and use a memory zone instead. While there,\nallow us to gracefully fail allocation with ENOMEM.\n\nSGI-PV: 978729\nSGI-Modid: xfs-linux-melb:xfs-kern:30771a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "114d23aae51233b2bc62d8e2a632bcb55de1953d", "tree": "39aa3e7e6dd32c39a416e34c4a663cb329315685", "parents": [ "2abdb8c88110bab78bfe17e51346e735560daa02" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Apr 10 12:18:39 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:50:22 2008 +1000" }, "message": "[XFS] Per iclog callback chain lock\n\nRather than use the icloglock for protecting the iclog completion callback\nchain, use a new per-iclog lock so that walking the callback chain doesn\u0027t\nrequire holding a global lock.\n\nThis reduces contention on the icloglock during transaction commit and log\nI/O completion by reducing the number of times we need to hold the global\nicloglock during these operations.\n\nSGI-PV: 978729\nSGI-Modid: xfs-linux-melb:xfs-kern:30770a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "2abdb8c88110bab78bfe17e51346e735560daa02", "tree": "0fad752a8f931991f0a788d52aeaaee508c0c3ea", "parents": [ "433550990e6c2e94995239bac6a52b4df454cae0" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@sgi.com", "time": "Thu Mar 27 18:01:14 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:49:51 2008 +1000" }, "message": "[XFS] Prevent xfs_bmap_check_leaf_extents() referencing unmapped memory.\n\nWhile investigating the extent corruption bug I ran into this bug in debug\nonly code. xfs_bmap_check_leaf_extents() loops through the leaf blocks of\nthe extent btree checking that every extent is entirely before the next\nextent. It also compares the last extent in the previous block to the\nfirst extent in the current block when the previous block has been\nreleased and potentially unmapped. So take a copy of the last extent\ninstead of a pointer. Also move the last extent check out of the loop\nbecause we only need to do it once.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30718a\n\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\n" }, { "commit": "433550990e6c2e94995239bac6a52b4df454cae0", "tree": "4536c63306b8e6656b969dc71b099e1b2c149454", "parents": [ "df26cfe849d8fd767b26fcd4bfebfff67bda9f3a" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 27 18:01:08 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:49:08 2008 +1000" }, "message": "[XFS] remove most calls to VN_RELE\n\nMost VN_RELE calls either directly contain a XFS_ITOV or have the\ncorresponding xfs_inode already in scope. Use the IRELE helper instead of\nVN_RELE to clarify the code. With a little more work we can kill VN_RELE\naltogether and define IRELE in terms of iput directly.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30710a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "df26cfe849d8fd767b26fcd4bfebfff67bda9f3a", "tree": "1a4aa2887786f11d787c119cd16cd301974640bc", "parents": [ "f3dcc13f6fa20af1171eac7a537a4b89b1a84849" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:44:03 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:44:03 2008 +1000" }, "message": "[XFS] split xfs_ioc_xattr\n\nThe three subcases of xfs_ioc_xattr don\u0027t share any semantics and almost\nno code, so split it into three separate helpers.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30709a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "f3dcc13f6fa20af1171eac7a537a4b89b1a84849", "tree": "4f1fb766af5a1a5110a5054894af1902ef02cb63", "parents": [ "59a33f9f776b051018ec98af95bd9fe8ba9d0f3e" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 27 18:00:54 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:42:36 2008 +1000" }, "message": "[XFS] cleanup root inode handling in xfs_fs_fill_super\n\n- rename rootvp to root for clarify\n- remove useless vn_to_inode call\n- check is_bad_inode before calling d_alloc_root\n- use iput instead of VN_RELE in the error case\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30708a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "59a33f9f776b051018ec98af95bd9fe8ba9d0f3e", "tree": "d8d93fcd6ef6a77a7efe722ed61febd3b4051bf7", "parents": [ "75de2a91c98a6f486f261c1367fe59f5583e15a3" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 27 18:00:45 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:42:21 2008 +1000" }, "message": "[XFS] Ensure a btree insert returns a valid cursor.\n\nWhen writing into preallocated regions there is a case where XFS can oops\nor hang doing the unwritten extent conversion on I/O completion. It turns\nout that the problem is related to the btree cursor being invalid.\n\nWhen we do an insert into the tree, we may need to split blocks in the\ntree. When we only split at the leaf level (i.e. level 0), everything\nworks just fine. However, if we have a multi-level split in the btreee,\nthe cursor passed to the insert function is no longer valid once the\ninsert is complete.\n\nThe leaf level split is handled correctly because all the operations at\nlevel 0 are done using the original cursor, hence it is updated correctly.\nHowever, when we need to update the next level up the tree, we don\u0027t use\nthat cursor - we use a cloned cursor that points to the index in the next\nlevel up where we need to do the insert.\n\nHence if we need to split a second level, the changes to the tree are\nreflected in the cloned cursor and not the original cursor. This\nclone-and-move-up-a-level-on-split behaviour recurses all the way to the\ntop of the tree.\n\nThe complexity here is that these cloned cursors do not point to the\noriginal index that was inserted - they point to the newly allocated block\n(the right block) and the original cursor pointer to that level may still\npoint to the left block. Hence, without deep examination of the cloned\ncursor and buffers, we cannot update the original cursor with the new path\nfrom the cloned cursor.\n\nIn these cases the original cursor could be pointing to the wrong block(s)\nand hence a subsequent modification to the tree using that cursor will\nlead to corruption of the tree.\n\nThe crash case occurs when the tree changes height - we insert a new level\nin the tree, and the cursor does not have a buffer in it\u0027s path for that\nlevel. Hence any attempt to walk back up the cursor to the root block will\nresult in a null pointer dereference.\n\nTo make matters even more complex, the BMAP BT is rooted in an inode, so\nwe can have a change of height in the btree *without a root split*. That\nis, if the root block in the inode is full when we split a leaf node, we\ncannot fit the pointer to the new block in the root, so we allocate a new\nblock, migrate all the ptrs out of the inode into the new block and point\nthe inode root block at the newly allocated block. This changes the height\nof the tree without a root split having occurred and hence invalidates the\npath in the original cursor.\n\nThe patch below prevents xfs_bmbt_insert() from returning with an invalid\ncursor by detecting the cases that invalidate the original cursor and\nrefresh it by do a lookup into the btree for the original index we were\ninserting at.\n\nNote that the INOBT, AGFBNO and AGFCNT btree implementations also have\nthis bug, but the cursor is currently always destroyed or revalidated\nafter an insert for those trees. Hence this patch only address the problem\nin the BMBT code.\n\nSGI-PV: 979339\nSGI-Modid: xfs-linux-melb:xfs-kern:30701a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "75de2a91c98a6f486f261c1367fe59f5583e15a3", "tree": "599d1024b34a6649f8c589100f72adf0f30b40b5", "parents": [ "535f6b3735db6ef6026537bfe55ae00c3d9cc1ee" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 27 18:00:38 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:42:09 2008 +1000" }, "message": "[XFS] Account for inode cluster alignment in all allocations\n\nAt ENOSPC, we can get a filesystem shutdown due to a cancelling a dirty\ntransaction in xfs_mkdir or xfs_create. This is due to the initial\nallocation attempt not taking into account inode alignment and hence we\ncan prepare the AGF freelist for allocation when it\u0027s not actually\npossible to do an allocation. This results in inode allocation returning\nENOSPC with a dirty transaction, and hence we shut down the filesystem.\n\nBecause the first allocation is an exact allocation attempt, we must tell\nthe allocator that the alignment does not affect the allocation attempt.\ni.e. we will accept any extent alignment as long as the extent starts at\nthe block we want. Unfortunately, this means that if the longest free\nextent is less than the length + alignment necessary for fallback\nallocation attempts but is long enough to attempt a non-aligned\nallocation, we will modify the free list.\n\nIf we then have the exact allocation fail, all other allocation attempts\nwill also fail due to the alignment constraint being taken into account.\nHence the initial attempt needs to set the \"alignment slop\" field so that\nalignment, while not required, must be taken into account when determining\nif there is enough space left in the AG to do the allocation.\n\nThat means if the exact allocation fails, we will not dirty the freelist\nif there is not enough space available fo a subsequent allocation to\nsucceed. Hence we get an ENOSPC error back to userspace without shutting\ndown the filesystem.\n\nSGI-PV: 978886\nSGI-Modid: xfs-linux-melb:xfs-kern:30699a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "535f6b3735db6ef6026537bfe55ae00c3d9cc1ee", "tree": "49be6c12a19332a0ce0288143f67b44e4b8538b8", "parents": [ "a45c796867df8dabc8eed6e72898d7ba1609bd7e" ], "author": { "name": "Josef \u0027Jeff\u0027 Sipek", "email": "jeffpc@josefsipek.net", "time": "Thu Mar 27 17:58:27 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:41:57 2008 +1000" }, "message": "[XFS] Replace custom AIL linked-list code with struct list_head\n\nReplace the xfs_ail_entry_t with a struct list_head and clean the\nsurrounding code up. Also fixes a livelock in xfs_trans_first_push_ail()\nby terminating the loop at the head of the list correctly.\n\nSGI-PV: 978682\nSGI-Modid: xfs-linux-melb:xfs-kern:30636a\n\nSigned-off-by: Josef \u0027Jeff\u0027 Sipek \u003cjeffpc@josefsipek.net\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "a45c796867df8dabc8eed6e72898d7ba1609bd7e", "tree": "81e67643e1bc0577dccbb7f28a35437b951b2521", "parents": [ "dfa18b117974d7667a2d5b941853fac3f2e256db" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:49:36 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:41:46 2008 +1000" }, "message": "[XFS] Remove superflous xfs_readsb call in xfs_mountfs.\n\nWhen xfs_mountfs is called by xfs_mount xfs_readsb was called 35 lines\nabove unconditionally, so there is no need to try to read the superblock\nif it\u0027s not present. If any other port doesn\u0027t have the superblock read at\nthis point it should just call it directly from it\u0027s xfs_mount equivalent.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30603a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Donald Douwsma \u003cdonaldd@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "dfa18b117974d7667a2d5b941853fac3f2e256db", "tree": "ec6035762d8c7b75435f76a7bae6688a6d218941", "parents": [ "5f90150abad61b49dbb4a6ca1087fe0a75001ef9" ], "author": { "name": "Niv Sardi", "email": "xaiki@sgi.com", "time": "Thu Mar 06 13:49:26 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:41:35 2008 +1000" }, "message": "[XFS] kill t_sema member of struct xfs_trans\n\nIt\u0027s completely unused so we might aswell kill it. Note that there is\nanother t_sema in struct xlog_ticket, which is used and actually an sv_t\ndespite the name. That one is left untouched by this patch.\n\nSGI-PV: 971186\nSGI-Modid: xfs-linux-melb:xfs-kern:30591a\n\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "5f90150abad61b49dbb4a6ca1087fe0a75001ef9", "tree": "21a1a0fb9edc473ab0be08ad0158ed658c086c01", "parents": [ "af048193fcfe2650e7ed3b1ab3d48b1ed0efb467" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:49 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:41:25 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_bmap.c\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30553a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "af048193fcfe2650e7ed3b1ab3d48b1ed0efb467", "tree": "0e02606bb4db074fdf18c3e4af2d9b28544a900f", "parents": [ "dcf49cc5cfbbc0070ad4307428f8282dc7e04e58" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:43 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:41:14 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_iops.c\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30552a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "dcf49cc5cfbbc0070ad4307428f8282dc7e04e58", "tree": "ca789975aade7e70fed8642eecd6d11e81eff748", "parents": [ "ef1f5e7ad38e5414d016983a8cc5a8db7654a61d" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:37 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:41:04 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_lrw.c\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30551a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "ef1f5e7ad38e5414d016983a8cc5a8db7654a61d", "tree": "6257477d68089b51a5dc352ef573344c42af0115", "parents": [ "3937be5ba836a204d3d1df96b518eecd6cdacbb9" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:25 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:40:55 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_lookup\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30550a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "3937be5ba836a204d3d1df96b518eecd6cdacbb9", "tree": "89c21be058d0b1b43718f49025600346524cd68e", "parents": [ "a3da789640871c897901c5f766e33be78d56f35a" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:19 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:40:45 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_symlink and xfs_rename\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30548a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "a3da789640871c897901c5f766e33be78d56f35a", "tree": "cbc8cda24404077100e2d029b21531b219affbfb", "parents": [ "979ebab11623894528d4d37b947533ea4e8649d1" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:12 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:40:35 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_link\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30547a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "979ebab11623894528d4d37b947533ea4e8649d1", "tree": "b7d2554818c1c0e8b2d5a7c581567be3c740247c", "parents": [ "bc4ac74a4e5bd7db02976eb1b681e1d11f81c9ce" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:46:05 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:40:25 2008 +1000" }, "message": "[XFS] cleanup vnode use in xfs_create/mknod/mkdir\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30546a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "bc4ac74a4e5bd7db02976eb1b681e1d11f81c9ce", "tree": "580c59e724938a57874c05298c8da4e685971401", "parents": [ "d234154125197053d5215711b5df867979e55ebd" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:45:58 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:40:15 2008 +1000" }, "message": "[XFS] cleanup vnode use in dmapi calls\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30545a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "d234154125197053d5215711b5df867979e55ebd", "tree": "366a6ff91b514a8f9846ac82926e0bbfd0a1fbed", "parents": [ "6ee4752ffe782be6e86bea1403a2fe0f682aa71a" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:45:43 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:40:04 2008 +1000" }, "message": "[XFS] Use power-of-2 sized buffers to reduce overhead\n\nNow that the ktrace_enter() code is using atomics, the non-power-of-2\nbuffer sizes - which require modulus operations to get the index - are\nshowing up as using substantial CPU in the profiles.\n\nForce the buffer sizes to be rounded up to the nearest power of two and\nuse masking rather than modulus operations to convert the index counter to\nthe buffer index. This reduces ktrace_enter overhead to 8% of a CPU time,\nand again almost halves the trace intensive test runtime.\n\nSGI-PV: 977546\nSGI-Modid: xfs-linux-melb:xfs-kern:30538a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "6ee4752ffe782be6e86bea1403a2fe0f682aa71a", "tree": "aea65e2a9f1f6ca5c3f8dd44b78c5a9dd5adaee3", "parents": [ "44d814ced4cffbfe6a775c5bb8b941a6e734e7d9" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:45:35 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:39:55 2008 +1000" }, "message": "[XFS] Use atomic counters for ktrace buffer indexes\n\nktrace_enter() is consuming vast amounts of CPU time due to the use of a\nsingle global lock for protecting buffer index increments. Change it to\nuse per-buffer atomic counters - this reduces ktrace_enter() overhead\nduring a trace intensive test on a 4p machine from 58% of all CPU time to\n12% and halves test runtime.\n\nSGI-PV: 977546\nSGI-Modid: xfs-linux-melb:xfs-kern:30537a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "44d814ced4cffbfe6a775c5bb8b941a6e734e7d9", "tree": "94e0cecd2b18cf2b126b702b30a8a982de3931e0", "parents": [ "24bd861d1c3fff5248de7ba3bdddb3369087ad46" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:45:29 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:39:45 2008 +1000" }, "message": "[XFS] Update c/mtime correctly on truncates\n\nXFS changes the c/mtime of an inode when truncating it to the same size.\nThe c/mtime is only supposed to change if the size is changed. Not to be\nconfused with ftruncate, where the c/mtime is supposed to be changed even\nif the size is not changed.\n\nThe Linux VFS encodes this semantic difference in the flags it sends down\nto -\u003esetattr, which XFS currently ignores. We need to make XFS pay\nattention to the VFS flags and hence Do The Right Thing.\n\nSGI-PV: 977547\nSGI-Modid: xfs-linux-melb:xfs-kern:30536a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "24bd861d1c3fff5248de7ba3bdddb3369087ad46", "tree": "11092d964b9323c9eaef3df36bd77ef7ff03281b", "parents": [ "126468b1156211e26d97f74b2f1767acd141005a" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:45:16 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:39:35 2008 +1000" }, "message": "[XFS] don\u0027t encode parent in nfs filehandles unless nessecary\n\nAs Dave pointed out after the export ops changes we now always encode the\nparent into the filehandle for regular files, but it\u0027s not actually needed\nwhen the filesystem is export with no_subtree_check. This one-liner fixes\nxfs_fs_encode_fh to skip encoding the parent unless nessecary.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30535a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "126468b1156211e26d97f74b2f1767acd141005a", "tree": "0bf60d932dad0b4e7bc7c092782b531e550a11b2", "parents": [ "43973964a386348af0a392266f008ba24170aa30" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:44:57 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:39:25 2008 +1000" }, "message": "[XFS] kill xfs_rwlock/xfs_rwunlock\n\nWe can just use xfs_ilock/xfs_iunlock instead and get rid of the ugly\nbhv_vrwlock_t.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30533a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "43973964a386348af0a392266f008ba24170aa30", "tree": "17d21f1c23b4bdcd04b1057eaa8509ccad095d1d", "parents": [ "a8b3acd57e3aaaf73a863a28e0e9f6cca37cd8e3" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:44:50 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:39:14 2008 +1000" }, "message": "[XFS] kill xfs_get_dir_entry\n\nInstead of of xfs_get_dir_entry use a macro to get the xfs_inode from the\ndentry in the callers and grab the reference manually.\n\nOnly grab the reference once as it\u0027s fine to keep it over the dmapi calls.\n(And even that reference is actually superflous in Linux but I\u0027ll leave\nthat for another patch)\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30531a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "a8b3acd57e3aaaf73a863a28e0e9f6cca37cd8e3", "tree": "d759912cd15f3e7a439f9b9b63bea2335d8faa62", "parents": [ "db0bb7baa1533db156d8af3ebeda1f0473a0197a" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:44:41 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:39:03 2008 +1000" }, "message": "[XFS] vnode cleanup in xfs_fs_subr.c\n\nCleanup the unneeded intermediate vnode step in the flushing helpers and\ngo directly from the xfs_inode to the struct address_space.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30530a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "db0bb7baa1533db156d8af3ebeda1f0473a0197a", "tree": "39247febd2c597ad58f754b0fd3ce35cd2c4e796", "parents": [ "155cc6b784a959ed456fe46dca522e1d28b3b718" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Mar 06 13:44:35 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:38:53 2008 +1000" }, "message": "[XFS] cleanup xfs_vn_mknod\n\n- use proper goto based unwinding instead of the current mess of\n multiple conditionals\n- rename ip to inode because that\u0027s the normal convention for Linux\n inodes while ip is the convention for xfs_inodes\n- remove unlikely checks for the default_acl - branches marked unlikely\n might lead to extreme branch bredictor slowdons if taken and for some\n workloads a default acl is quite common\n- properly indent the switch statements\n- remove xfs_has_fs_struct as nfsd has a fs_struct in any semi-recent\n kernel\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30529a\n\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "155cc6b784a959ed456fe46dca522e1d28b3b718", "tree": "e88b9041570f299497a6f9aac7e01797affed205", "parents": [ "b589334c7a1fff85d2f009d5db4c34fad48925e9" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:44:14 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:38:10 2008 +1000" }, "message": "[XFS] Use atomics for iclog reference counting\n\nNow that we update the log tail LSN less frequently on transaction\ncompletion, we pass the contention straight to the global log state lock\n(l_iclog_lock) during transaction completion.\n\nWe currently have to take this lock to decrement the iclog reference\ncount. there is a reference count on each iclog, so we need to take þhe\nglobal lock for all refcount changes.\n\nWhen large numbers of processes are all doing small trnasctions, the iclog\nreference counts will be quite high, and the state change that absolutely\nrequires the l_iclog_lock is the except rather than the norm.\n\nChange the reference counting on the iclogs to use atomic_inc/dec so that\nwe can use atomic_dec_and_lock during transaction completion and avoid the\nneed for grabbing the l_iclog_lock for every reference count decrement\nexcept the one that matters - the last.\n\nSGI-PV: 975671\nSGI-Modid: xfs-linux-melb:xfs-kern:30505a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Tim Shimmin \u003ctes@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "b589334c7a1fff85d2f009d5db4c34fad48925e9", "tree": "bc46a7f50dbc5f8c09e86374cea63192ee2f146b", "parents": [ "3354040897f828644be6ca5783588e9f64a53b8e" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:44:06 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:38:01 2008 +1000" }, "message": "[XFS] Prevent AIL lock contention during transaction completion\n\nWhen hundreds of processors attempt to commit transactions at the same\ntime, they can contend on the AIL lock when updating the tail LSN held in\nthe in-core log structure.\n\nAt the moment, the tail LSN is only needed when actually writing out an\niclog, so it really does not need to be updated on every single\ntransaction completion - only those that result in switching iclogs and\nflushing them to disk.\n\nThe result is that we reduce the number of times we need to grab the AIL\nlock and the log grant lock by up to two orders of magnitude on large\nprocessor count machines. The problem has previously been hidden by AIL\nlock contention walking the AIL list which was recently solved and\nuncovered this issue.\n\nSGI-PV: 975671\nSGI-Modid: xfs-linux-melb:xfs-kern:30504a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Tim Shimmin \u003ctes@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "3354040897f828644be6ca5783588e9f64a53b8e", "tree": "c41cfb4584e13225b1f4578e64c8739d88a8537c", "parents": [ "bad5584332e888ac40ca13584e8c114149ddb01e" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:43:59 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:37:51 2008 +1000" }, "message": "[XFS] Use xfs_inode_clean() in more places\n\nRemove open coded checks for the whether the inode is clean and replace\nthem with an inlined function.\n\nSGI-PV: 977461\nSGI-Modid: xfs-linux-melb:xfs-kern:30503a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "bad5584332e888ac40ca13584e8c114149ddb01e", "tree": "f66ff83686cd28967d8b9fc0a8b8d6411a081071", "parents": [ "a3f74ffb6d1448d9a8f482e593b80ec15f1695d4" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:43:49 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:37:41 2008 +1000" }, "message": "[XFS] Remove the xfs_icluster structure\n\nRemove the xfs_icluster structure and replace with a radix tree lookup.\n\nWe don\u0027t need to keep a list of inodes in each cluster around anymore as\nwe can look them up quickly when we need to. The only time we need to do\nthis now is during inode writeback.\n\nFactor the inode cluster writeback code out of xfs_iflush and convert it\nto use radix_tree_gang_lookup() instead of walking a list of inodes built\nwhen we first read in the inodes.\n\nThis remove 3 pointers from each xfs_inode structure and the xfs_icluster\nstructure per inode cluster. Hence we reduce the cache footprint of the\nxfs_inodes by between 5-10% depending on cluster sparseness.\n\nTo be truly efficient we need a radix_tree_gang_lookup_range() call to\nstop searching once we are past the end of the cluster instead of trying\nto find a full cluster\u0027s worth of inodes.\n\nBefore (ia64):\n\n$ cat /sys/slab/xfs_inode/object_size 536\n\nAfter:\n\n$ cat /sys/slab/xfs_inode/object_size 512\n\nSGI-PV: 977460\nSGI-Modid: xfs-linux-melb:xfs-kern:30502a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "a3f74ffb6d1448d9a8f482e593b80ec15f1695d4", "tree": "e7a9ea7ba4032340e771605000002da4349719cb", "parents": [ "4ae29b4321b99b711bcfde5527c4fbf249eac60f" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:43:42 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:37:32 2008 +1000" }, "message": "[XFS] Don\u0027t block pdflush when writing back inodes\n\nWhen pdflush is writing back inodes, it can get stuck on inode cluster\nbuffers that are currently under I/O. This occurs when we write data to\nmultiple inodes in the same inode cluster at the same time.\n\nEffectively, delayed allocation marks the inode dirty during the data\nwriteback. Hence if the inode cluster was flushed during the writeback of\nthe first inode, the writeback of the second inode will block waiting for\nthe inode cluster write to complete before writing it again for the newly\ndirtied inode.\n\nBasically, we want to avoid this from happening so we don\u0027t block pdflush\nand slow down all of writeback. Hence we introduce a non-blocking async\ninode flush flag that pdflush uses. If this flag is set, we use\nnon-blocking operations (e.g. try locks) whereever we can to avoid\nblocking or extra I/O being issued.\n\nSGI-PV: 970925\nSGI-Modid: xfs-linux-melb:xfs-kern:30501a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "4ae29b4321b99b711bcfde5527c4fbf249eac60f", "tree": "a508c4574d61a9d1713e6674a5b6c6817616120b", "parents": [ "e9a56b7cdaf6129892fd7c8d950b71a1a4304bb0" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:43:34 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:37:19 2008 +1000" }, "message": "[XFS] Factor xfs_itobp() and xfs_inotobp().\n\nThe only difference between the functions is one passes an inode for the\nlookup, the other passes an inode number. However, they don\u0027t do the same\nvalidity checking or set all the same state on the buffer that is returned\nyet they should.\n\nFactor the functions into a common implementation.\n\nSGI-PV: 970925\nSGI-Modid: xfs-linux-melb:xfs-kern:30500a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "e9a56b7cdaf6129892fd7c8d950b71a1a4304bb0", "tree": "f7686735965a6c2cb81c7be75f14f58794c61983", "parents": [ "163d3686bb09d88e2120bffe780a3f2d7cc4c948" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@sgi.com", "time": "Thu Mar 06 13:43:27 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:37:06 2008 +1000" }, "message": "[XFS] Fix regression due to refcache removal\n\nSGI-PV: 971186\nSGI-Modid: xfs-linux-melb:xfs-kern:30490a\n\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\nSigned-off-by: Donald Douwsma \u003cdonaldd@sgi.com\u003e\n" }, { "commit": "163d3686bb09d88e2120bffe780a3f2d7cc4c948", "tree": "8503962ad2c4641f0796b5bb864996994c1ca137", "parents": [ "461aa8a22595e3bd3e6f4dc2894d7c4315ea2bb9" ], "author": { "name": "Donald Douwsma", "email": "donaldd@sgi.com", "time": "Thu Mar 06 13:43:20 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:36:55 2008 +1000" }, "message": "[XFS] Remove the xfs_refcache\n\nRemove the xfs_refcache, it was only needed while we were still\nbuilding for 2.4 kernels.\n\nSGI-PV: 971186\nSGI-Modid: xfs-linux-melb:xfs-kern:30472a\n\nSigned-off-by: Donald Douwsma \u003cdonaldd@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "461aa8a22595e3bd3e6f4dc2894d7c4315ea2bb9", "tree": "52cef56ec70610b9e0dd1fac355d646e827c0102", "parents": [ "e12070a5dca8bfeee352e9655ae79772a96b32f8" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@sgi.com", "time": "Thu Mar 06 13:43:11 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:34:54 2008 +1000" }, "message": "[XFS] make inode reclaim synchronise with xfs_iflush_done()\n\nOn a forced shutdown, xfs_finish_reclaim() will skip flushing the inode.\nIf the inode flush lock is not already held and there is an outstanding\nxfs_iflush_done() then we might free the inode prematurely. By acquiring\nand releasing the flush lock we will synchronise with xfs_iflush_done().\n\nSGI-PV: 909874\nSGI-Modid: xfs-linux-melb:xfs-kern:30468a\n\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\n" }, { "commit": "e12070a5dca8bfeee352e9655ae79772a96b32f8", "tree": "76ebdb3044efec2699090ac70fe7aa9edbf1a311", "parents": [ "4b119e21d0c66c22e8ca03df05d9de623d0eb50f" ], "author": { "name": "Niv Sardi", "email": "xaiki@sgi.com", "time": "Thu Mar 06 13:43:03 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Apr 18 11:34:47 2008 +1000" }, "message": "[XFS] actually check error returned by xfs_flush_pages, clean up and\nbailout if fails.\n\nSGI-PV: 973041\nSGI-Modid: xfs-linux-melb:xfs-kern:30462a\n\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "e6957ea48429aeda4f5d51af4238231d44b0dc4a", "tree": "860c95c0e9446191262bb44ad436d38afb8cb7d6", "parents": [ "ee1c090825bf2f1cb9712489c234eabf69a9a78f" ], "author": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Thu Apr 10 12:19:34 2008 +1000" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Thu Apr 10 16:25:26 2008 +1000" }, "message": "[XFS] Ensure \"both\" features2 slots are consistent\n\nSince older kernels may look in the sb_bad_features2 slot for flags,\nrather than zeroing it out on fixup, we should make it equal to the\nsb_features2 value.\n\nAlso, if the ATTR2 flag was not found prior to features2 fixup, it was not\nset in the mount flags, so re-check after the fixup so that the current\nsession will use the feature.\n\nAlso fix up the comments to reflect these changes.\n\nSGI-PV: 980085\nSGI-Modid: xfs-linux-melb:xfs-kern:30778a\n\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "ee1c090825bf2f1cb9712489c234eabf69a9a78f", "tree": "c261cbeeda46937e6c65994b4348227505665f66", "parents": [ "6211870992502efdccf9cd6bae155c83277835fa" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:45:50 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Thu Apr 10 16:25:15 2008 +1000" }, "message": "[XFS] Fix superblock features2 field alignment problem\n\nDue to the xfs_dsb_t structure not being 64 bit aligned, the last field of\nthe on-disk superblock can vary in location This causes problems when the\nfilesystem gets moved to a different platform, or there is a 32 bit\nuserspace and 64 bit kernel.\n\nThis patch detects the defect at mount time, logs a warning such as:\n\nXFS: correcting sb_features alignment problem\n\nin dmesg and corrects the problem so that everything is OK. it also\nblacklists the bad field in the superblock so it does not get used for\nsomething else later on.\n\nSGI-PV: 977636\nSGI-Modid: xfs-linux-melb:xfs-kern:30539a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "6211870992502efdccf9cd6bae155c83277835fa", "tree": "1e8a5606550a4cccc7d8e82dd23e681077f223a6", "parents": [ "7180c4c9e09888db0a188f729c96c6d7bd61fa83" ], "author": { "name": "Eric Sandeen", "email": "sandeen@sandeen.net", "time": "Thu Mar 06 13:44:28 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Thu Apr 10 16:24:45 2008 +1000" }, "message": "[XFS] remove shouting-indirection macros from xfs_sb.h\n\nRemove macro-to-small-function indirection from xfs_sb.h, and remove some\nwhich are completely unused.\n\nSGI-PV: 976035\nSGI-Modid: xfs-linux-melb:xfs-kern:30528a\n\nSigned-off-by: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Donald Douwsma \u003cdonaldd@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "72772a3b5b158cddcfbbff3ef13b26b03a905158", "tree": "e87303c54834c58dca859c881d693babbf0fce5e", "parents": [ "92d9cd1059f80b9c89dee191ffb88b0872e6a7ae" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:49:43 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Thu Mar 06 16:38:50 2008 +1100" }, "message": "[XFS] fix inode leak in xfs_iget_core()\n\nIf the radix_tree_preload() fails, we need to destroy the inode we just\nread in before trying again. This could leak xfs_vnode structures when\nthere is memory pressure. Noticed by Christoph Hellwig.\n\nSGI-PV: 977823\nSGI-Modid: xfs-linux-melb:xfs-kern:30606a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\n" }, { "commit": "92d9cd1059f80b9c89dee191ffb88b0872e6a7ae", "tree": "972a8e148775ec20358f6e59a01807417be9b7a1", "parents": [ "29e8c3c304b62f31b799565c9ee85d42bd163f80" ], "author": { "name": "David Chinner", "email": "dgc@sgi.com", "time": "Thu Mar 06 13:45:10 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Thu Mar 06 16:38:17 2008 +1100" }, "message": "[XFS] 977545 977545 977545 977545 977545 977545 xfsaild causing too many\nwakeups\n\nIdle state is not being detected properly by the xfsaild push code. The\ncurrent idle state is detected by an empty list which may never happen\nwith mostly idle filesystem or one using lazy superblock counters. A\nsingle dirty item in the list that exists beyond the push target can\nresult repeated looping attempting to push up to the target because it\nfails to check if the push target has been acheived or not.\n\nFix by considering a dirty list with everything past the target as an idle\nstate and set the timeout appropriately.\n\nSGI-PV: 977545\nSGI-Modid: xfs-linux-melb:xfs-kern:30532a\n\nSigned-off-by: David Chinner \u003cdgc@sgi.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "1bd960ee2b1231759bd485aad0fa483c2f793a3b", "tree": "557463c5faac349791e14890f35668e4c1bcb639", "parents": [ "7704a8b6fc4a8f51599eb2af4dcf1e2ac9c7e576" ], "author": { "name": "Josef Jeff Sipek", "email": "jeffpc@josefsipek.net", "time": "Fri Feb 29 13:58:40 2008 +1100" }, "committer": { "name": "Niv Sardi", "email": "xaiki@oss.sgi.com", "time": "Thu Feb 28 20:37:56 2008 -0800" }, "message": "[XFS] If you mount an XFS filesystem with no mount options at all, then\nthe \"ikeep\" option is set rather than \"noikeep\".\n\nThis regression was introduced in 970451.\n\nWith no mount options specified, xfs_parseargs() does the following:\n\nint ikeep \u003d 0;\n\nargs-\u003eflags |\u003d XFSMNT_BARRIER;\n\nargs-\u003eflags2 |\u003d XFSMNT2_COMPAT_IOSIZE;\n\nif (!options)\n\ngoto done;\n\nIt only sets the above two options by default and before, it also used to\nset XFSMNT_IDELETE by default.\n\nIf options are specified, then\n\nif (!(args-\u003eflags \u0026 XFSMNT_DMAPI) \u0026\u0026 !ikeep)\n\nargs-\u003eflags |\u003d XFSMNT_IDELETE;\n\nis executed later on which is skipped by the \"goto done;\" above.\n\nThe solution is to invert the logic.\n\nSGI-PV: 977771\nSGI-Modid: xfs-linux-melb:xfs-kern:30590a\n\nSigned-off-by: Niv Sardi \u003cxaiki@sgi.com\u003e\nSigned-off-by: Barry Naujok \u003cbnaujok@sgi.com\u003e\nSigned-off-by: Josef \u0027Jeff\u0027 Sipek \u003cjeffpc@josefsipek.net\u003e\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "ef8ece55d9b6825c28a5c1a4bd89b94040cb7b32", "tree": "c086be0bf7ef1139db75d4fd084f2390545629e8", "parents": [ "db69c915e67705daac25cad06d816c09be634de0" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@sgi.com", "time": "Tue Feb 26 17:00:22 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Tue Feb 26 17:05:44 2008 +1100" }, "message": "[XFS] Undo bit ops cleanup mod due to regression on 32-bit powermac\nplatform.\n\nSGI-PV: 971186\nSGI-Modid: xfs-linux-melb:xfs-kern:30559a\n\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "db69c915e67705daac25cad06d816c09be634de0", "tree": "8befd8c9eb1763345a031dfcd64a7559e0c388e0", "parents": [ "91e229bbad6524aabaac8717b2f559283670c37a" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@sgi.com", "time": "Tue Feb 26 17:00:14 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Tue Feb 26 17:05:37 2008 +1100" }, "message": "[XFS] Undo bit ops cleanup mod due to regression on 32-bit powermac\nplatform.\n\nSGI-PV: 974005\nSGI-Modid: xfs-linux-melb:xfs-kern:30558a\n\nSigned-off-by: Lachlan McIlroy \u003clachlan@sgi.com\u003e\n" }, { "commit": "6e5e93424dc66542c548dfaa3bfebe30d46d50dd", "tree": "232aeb0a00995ed2cd0213694d3a635f4aef41a0", "parents": [ "c58310bf4933986513020fa90b4190c7492995ae" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Feb 22 15:36:19 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Fri Feb 22 15:39:10 2008 +1100" }, "message": "Remove empty file fs/xfs/Makefile-linux-2.6.\n" }, { "commit": "c58310bf4933986513020fa90b4190c7492995ae", "tree": "143f2c7578d02ebef5db8fc57ae69e951ae0e2ee", "parents": [ "269cdfaf769f5cd831284cc831790c7c5038040f", "1309d4e68497184d2fd87e892ddf14076c2bda98" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Mon Feb 18 13:51:42 2008 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Mon Feb 18 13:51:42 2008 +1100" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6 into for-linus\n" }, { "commit": "269cdfaf769f5cd831284cc831790c7c5038040f", "tree": "d7e4ff963d0bf73f5ba61500f1bae0fcfca9d9d6", "parents": [ "794f744b225aaf35742aac9e7b9dda96a9943413" ], "author": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Wed Nov 28 18:28:09 2007 +1100" }, "committer": { "name": "Lachlan McIlroy", "email": "lachlan@redback.melbourne.sgi.com", "time": "Mon Feb 18 13:06:17 2008 +1100" }, "message": "[XFS] Added quota targets and removed dmapi directory\n\nFixes build failures introduced by bad merge to mainline.\n" } ], "next": "794f744b225aaf35742aac9e7b9dda96a9943413" }