)]}' { "log": [ { "commit": "1193755ac6328ad240ba987e6ec41d5e8baf0680", "tree": "40bf847d7e3ebaa57b107151d14e6cd1d280cc6d", "parents": [ "4edebed86690eb8db9af3ab85baf4a34e73266cc", "0ef97dcfce4179a2eba046b855ee2f91d6f1b414" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jun 01 10:34:35 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jun 01 10:34:35 2012 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull vfs changes from Al Viro.\n \"A lot of misc stuff. The obvious groups:\n * Miklos\u0027 atomic_open series; kills the damn abuse of\n -\u003ed_revalidate() by NFS, which was the major stumbling block for\n all work in that area.\n * ripping security_file_mmap() and dealing with deadlocks in the\n area; sanitizing the neighborhood of vm_mmap()/vm_munmap() in\n general.\n * -\u003eencode_fh() switched to saner API; insane fake dentry in\n mm/cleancache.c gone.\n * assorted annotations in fs (endianness, __user)\n * parts of Artem\u0027s -\u003es_dirty work (jff2 and reiserfs parts)\n * -\u003eupdate_time() work from Josef.\n * other bits and pieces all over the place.\n\n Normally it would\u0027ve been in two or three pull requests, but\n signal.git stuff had eaten a lot of time during this cycle ;-/\"\n\nFix up trivial conflicts in Documentation/filesystems/vfs.txt (the\n\u0027truncate_range\u0027 inode method was removed by the VM changes, the VFS\nupdate adds an \u0027update_time()\u0027 method), and in fs/btrfs/ulist.[ch] (due\nto sparse fix added twice, with other changes nearby).\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (95 commits)\n nfs: don\u0027t open in -\u003ed_revalidate\n vfs: retry last component if opening stale dentry\n vfs: nameidata_to_filp(): don\u0027t throw away file on error\n vfs: nameidata_to_filp(): inline __dentry_open()\n vfs: do_dentry_open(): don\u0027t put filp\n vfs: split __dentry_open()\n vfs: do_last() common post lookup\n vfs: do_last(): add audit_inode before open\n vfs: do_last(): only return EISDIR for O_CREAT\n vfs: do_last(): check LOOKUP_DIRECTORY\n vfs: do_last(): make ENOENT exit RCU safe\n vfs: make follow_link check RCU safe\n vfs: do_last(): use inode variable\n vfs: do_last(): inline walk_component()\n vfs: do_last(): make exit RCU safe\n vfs: split do_lookup()\n Btrfs: move over to use -\u003eupdate_time\n fs: introduce inode operation -\u003eupdate_time\n reiserfs: get rid of resierfs_sync_super\n reiserfs: mark the superblock as dirty a bit later\n ...\n" }, { "commit": "c3b2da314834499f34cba94f7053e55f6d6f92d8", "tree": "7012b569ee9e0781761a8eb388190979441583c7", "parents": [ "033369d1af1264abc23bea2e174aa47cdd212f6f" ], "author": { "name": "Josef Bacik", "email": "josef@redhat.com", "time": "Mon Mar 26 09:59:21 2012 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Jun 01 12:07:25 2012 -0400" }, "message": "fs: introduce inode operation -\u003eupdate_time\n\nBtrfs has to make sure we have space to allocate new blocks in order to modify\nthe inode, so updating time can fail. We\u0027ve gotten around this by having our\nown file_update_time but this is kind of a pain, and Christoph has indicated he\nwould like to make xfs do something different with atime updates. So introduce\n-\u003eupdate_time, where we will deal with i_version an a/m/c time updates and\nindicate which changes need to be made. The normal version just does what it\nhas always done, updates the time and marks the inode dirty, and then\nfilesystems can choose to do something different.\n\nI\u0027ve gone through all of the users of file_update_time and made them check for\nerrors with the exception of the fault code since it\u0027s complicated and I wasn\u0027t\nquite sure what to do there, also Jan is going to be pushing the file time\nupdates into page_mkwrite for those who have it so that should satisfy btrfs and\nmake it not a big deal to check the file_update_time() return code in the\ngeneric fault path. Thanks,\n\nSigned-off-by: Josef Bacik \u003cjosef@redhat.com\u003e\n" }, { "commit": "3ed37648e1cbf1bbebc200c6ea8fd8daf8325843", "tree": "60f27fc77a1e5c9c48790243266367318a3267dc", "parents": [ "8bdc81c5069e43755d6e59e5e990e21ca200e8e2" ], "author": { "name": "Cong Wang", "email": "xiyou.wangcong@gmail.com", "time": "Tue May 15 14:57:33 2012 +0800" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed May 30 21:04:52 2012 -0400" }, "message": "fs: move file_remove_suid() to fs/inode.c\n\nfile_remove_suid() is a generic function operates on struct file,\nit almost has no relations with file mapping, so move it to fs/inode.c.\n\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "782182e53a6cdb3e3d04cc40516e173046942a32", "tree": "9e6f74d90b65ca12d4992c034ffa9b2415f37068", "parents": [ "4fb5ef089b288942c6fc3f85c4ecb4016c1aa4c3" ], "author": { "name": "Cong Wang", "email": "xiyou.wangcong@gmail.com", "time": "Tue May 29 15:06:43 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 29 16:22:23 2012 -0700" }, "message": "mm: move readahead syscall to mm/readahead.c\n\nIt is better to define readahead(2) in mm/readahead.c than in\nmm/filemap.c.\n\nSigned-off-by: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\nCc: Fengguang Wu \u003cfengguang.wu@intel.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0fc9d1040313047edf6a39fd4d7c7defdca97c62", "tree": "4503c084da1097e1cf66d8bac1ec548bc69ce69a", "parents": [ "cebbd29e1c2f7a969919f19f74583070840163d7" ], "author": { "name": "Konstantin Khlebnikov", "email": "khlebnikov@openvz.org", "time": "Wed Mar 28 14:42:54 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 28 17:14:37 2012 -0700" }, "message": "radix-tree: use iterators in find_get_pages* functions\n\nReplace radix_tree_gang_lookup_slot() and\nradix_tree_gang_lookup_tag_slot() in page-cache lookup functions with\nbrand-new radix-tree direct iterating. This avoids the double-scanning\nand pointer copying.\n\nIterator don\u0027t stop after nr_pages page-get fails in a row, it continue\nlookup till the radix-tree end. Thus we can safely remove these restart\nconditions.\n\nUnfortunately, old implementation didn\u0027t forbid nr_pages \u003d\u003d 0, this corner\ncase does not fit into new code, so the patch adds an extra check at the\nbeginning.\n\nSigned-off-by: Konstantin Khlebnikov \u003ckhlebnikov@openvz.org\u003e\nTested-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "aab008db8063364dc3c8ccf4981c21124866b395", "tree": "72914203f4decb023efdaabd0301a62d742dfa8c", "parents": [ "4f5b1affdda3e0c48cac674182f52004137b0ffc", "16c0cfa425b8e1488f7a1873bd112a7a099325f0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 22 19:52:47 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 22 19:52:47 2012 -0700" }, "message": "Merge tag \u0027stable/for-linus-3.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/mm\n\nPull cleancache changes from Konrad Rzeszutek Wilk:\n \"This has some patches for the cleancache API that should have been\n submitted a _long_ time ago. They are basically cleanups:\n\n - rename of flush to invalidate\n\n - moving reporting of statistics into debugfs\n\n - use __read_mostly as necessary.\n\n Oh, and also the MAINTAINERS file change. The files (except the\n MAINTAINERS file) have been in #linux-next for months now. The late\n addition of MAINTAINERS file is a brain-fart on my side - didn\u0027t\n realize I needed that just until I was typing this up - and I based\n that patch on v3.3 - so the tree is on top of v3.3.\"\n\n* tag \u0027stable/for-linus-3.4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/mm:\n MAINTAINERS: Adding cleancache API to the list.\n mm: cleancache: Use __read_mostly as appropiate.\n mm: cleancache: report statistics via debugfs instead of sysfs.\n mm: zcache/tmem/cleancache: s/flush/invalidate/\n mm: cleancache: s/flush/invalidate/\n" }, { "commit": "cc9a6c8776615f9c194ccf0b63a0aa5628235545", "tree": "0cbbf118e86541f8eb2fc7b717a0e08eaced986d", "parents": [ "e845e199362cc5712ba0e7eedc14eed70e144258" ], "author": { "name": "Mel Gorman", "email": "mgorman@suse.de", "time": "Wed Mar 21 16:34:11 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 21 17:54:59 2012 -0700" }, "message": "cpuset: mm: reduce large amounts of memory barrier related damage v3\n\nCommit c0ff7453bb5c (\"cpuset,mm: fix no node to alloc memory when\nchanging cpuset\u0027s mems\") wins a super prize for the largest number of\nmemory barriers entered into fast paths for one commit.\n\n[get|put]_mems_allowed is incredibly heavy with pairs of full memory\nbarriers inserted into a number of hot paths. This was detected while\ninvestigating at large page allocator slowdown introduced some time\nafter 2.6.32. The largest portion of this overhead was shown by\noprofile to be at an mfence introduced by this commit into the page\nallocator hot path.\n\nFor extra style points, the commit introduced the use of yield() in an\nimplementation of what looks like a spinning mutex.\n\nThis patch replaces the full memory barriers on both read and write\nsides with a sequence counter with just read barriers on the fast path\nside. This is much cheaper on some architectures, including x86. The\nmain bulk of the patch is the retry logic if the nodemask changes in a\nmanner that can cause a false failure.\n\nWhile updating the nodemask, a check is made to see if a false failure\nis a risk. If it is, the sequence number gets bumped and parallel\nallocators will briefly stall while the nodemask update takes place.\n\nIn a page fault test microbenchmark, oprofile samples from\n__alloc_pages_nodemask went from 4.53% of all samples to 1.15%. The\nactual results were\n\n 3.3.0-rc3 3.3.0-rc3\n rc3-vanilla nobarrier-v2r1\n Clients 1 UserTime 0.07 ( 0.00%) 0.08 (-14.19%)\n Clients 2 UserTime 0.07 ( 0.00%) 0.07 ( 2.72%)\n Clients 4 UserTime 0.08 ( 0.00%) 0.07 ( 3.29%)\n Clients 1 SysTime 0.70 ( 0.00%) 0.65 ( 6.65%)\n Clients 2 SysTime 0.85 ( 0.00%) 0.82 ( 3.65%)\n Clients 4 SysTime 1.41 ( 0.00%) 1.41 ( 0.32%)\n Clients 1 WallTime 0.77 ( 0.00%) 0.74 ( 4.19%)\n Clients 2 WallTime 0.47 ( 0.00%) 0.45 ( 3.73%)\n Clients 4 WallTime 0.38 ( 0.00%) 0.37 ( 1.58%)\n Clients 1 Flt/sec/cpu 497620.28 ( 0.00%) 520294.53 ( 4.56%)\n Clients 2 Flt/sec/cpu 414639.05 ( 0.00%) 429882.01 ( 3.68%)\n Clients 4 Flt/sec/cpu 257959.16 ( 0.00%) 258761.48 ( 0.31%)\n Clients 1 Flt/sec 495161.39 ( 0.00%) 517292.87 ( 4.47%)\n Clients 2 Flt/sec 820325.95 ( 0.00%) 850289.77 ( 3.65%)\n Clients 4 Flt/sec 1020068.93 ( 0.00%) 1022674.06 ( 0.26%)\n MMTests Statistics: duration\n Sys Time Running Test (seconds) 135.68 132.17\n User+Sys Time Running Test (seconds) 164.2 160.13\n Total Elapsed Time (seconds) 123.46 120.87\n\nThe overall improvement is small but the System CPU time is much\nimproved and roughly in correlation to what oprofile reported (these\nperformance figures are without profiling so skew is expected). The\nactual number of page faults is noticeably improved.\n\nFor benchmarks like kernel builds, the overall benefit is marginal but\nthe system CPU time is slightly reduced.\n\nTo test the actual bug the commit fixed I opened two terminals. The\nfirst ran within a cpuset and continually ran a small program that\nfaulted 100M of anonymous data. In a second window, the nodemask of the\ncpuset was continually randomised in a loop.\n\nWithout the commit, the program would fail every so often (usually\nwithin 10 seconds) and obviously with the commit everything worked fine.\nWith this patch applied, it also worked fine so the fix should be\nfunctionally equivalent.\n\nSigned-off-by: Mel Gorman \u003cmgorman@suse.de\u003e\nCc: Miao Xie \u003cmiaox@cn.fujitsu.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Christoph Lameter \u003ccl@linux.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9a3c531df9462df6cc2b060f749651723ffc180c", "tree": "448ad57ef6ea3b462849ea1bdd4732e0296e1f4f", "parents": [ "47a133339c332f9f8e155c70f5da401aded69948" ], "author": { "name": "Andi Kleen", "email": "ak@linux.intel.com", "time": "Wed Mar 21 16:34:09 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 21 17:54:58 2012 -0700" }, "message": "mm: update stale lock ordering comment for memory-failure.c\n\nWhen i_mmap_lock changed to a mutex the locking order in memory failure\nwas changed to take the sleeping lock first. But the big fat mm lock\nordering comment (BFMLO) wasn\u0027t updated. Do this here.\n\nPointed out by Andrew.\n\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1010bb1b80edb0713415dfe1f97114d320f58c4f", "tree": "70aeaba13b61aa442e0c48c99a1035e204910982", "parents": [ "a05b0855fd15504972dba2358e5faa172a1e50ba" ], "author": { "name": "Fengguang Wu", "email": "fengguang.wu@intel.com", "time": "Wed Mar 21 16:34:08 2012 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 21 17:54:58 2012 -0700" }, "message": "mm: don\u0027t set __GFP_WRITE on ramfs/sysfs writes\n\nThere is not much point in skipping zones during allocation based on the\ndirty usage which they\u0027ll never contribute to. And we\u0027d like to avoid\npage reclaim waits when writing to ramfs/sysfs etc.\n\nSigned-off-by: Fengguang Wu \u003cfengguang.wu@intel.com\u003e\nAcked-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nCc: Greg Thelen \u003cgthelen@google.com\u003e\nCc: Ying Han \u003cyinghan@google.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Mel Gorman \u003cmgorman@suse.de\u003e\nAcked-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9b04c5fec43c0da610a2c37f70c5b013101a6ad7", "tree": "f04767281b7067fba91cf0d37440bf454c492e38", "parents": [ "c3eede8e0a1292d95c051cf947738687b9c42322" ], "author": { "name": "Cong Wang", "email": "amwang@redhat.com", "time": "Fri Nov 25 23:14:39 2011 +0800" }, "committer": { "name": "Cong Wang", "email": "xiyou.wangcong@gmail.com", "time": "Tue Mar 20 21:48:27 2012 +0800" }, "message": "mm: remove the second argument of k[un]map_atomic()\n\nSigned-off-by: Cong Wang \u003camwang@redhat.com\u003e\n" }, { "commit": "16c0cfa425b8e1488f7a1873bd112a7a099325f0", "tree": "eeba5728a8072096279c55ffc10a5ca7ac0ddf05", "parents": [ "94574d9a4c236e8bd19721b4adb0ea30ef446901", "072611ed1f291053a74b28b813d683a09495eba7" ], "author": { "name": "Konrad Rzeszutek Wilk", "email": "konrad.wilk@oracle.com", "time": "Mon Mar 19 12:12:19 2012 -0400" }, "committer": { "name": "Konrad Rzeszutek Wilk", "email": "konrad.wilk@oracle.com", "time": "Mon Mar 19 12:12:19 2012 -0400" }, "message": "Merge branch \u0027stable/cleancache.v13\u0027 into linux-next\n\n* stable/cleancache.v13:\n mm: cleancache: Use __read_mostly as appropiate.\n mm: cleancache: report statistics via debugfs instead of sysfs.\n mm: zcache/tmem/cleancache: s/flush/invalidate/\n mm: cleancache: s/flush/invalidate/\n" }, { "commit": "3deaa7190a8da38453c4fabd9dec7f66d17fff67", "tree": "23c522d5661bf7e73a39718eaf4eb547cf02b086", "parents": [ "55ca6140e9bb307efc97a9301a4f501de02a6fd6" ], "author": { "name": "Shaohua Li", "email": "shaohua.li@intel.com", "time": "Fri Feb 03 15:37:17 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 03 16:16:41 2012 -0800" }, "message": "readahead: fix pipeline break caused by block plug\n\nHerbert Poetzl reported a performance regression since 2.6.39. The test\nis a simple dd read, but with big block size. The reason is:\n\nT1: ra (A, A+128k), (A+128k, A+256k)\nT2: lock_page for page A, submit the 256k\nT3: hit page A+128K, ra (A+256k, A+384). the range isn\u0027t submitted\nbecause of plug and there isn\u0027t any lock_page till we hit page A+256k\nbecause all pages from A to A+256k is in memory\nT4: hit page A+256k, ra (A+384, A+ 512). Because of plug, the range isn\u0027t\nsubmitted again.\nT5: lock_page A+256k, so (A+256k, A+512k) will be submitted. The task is\nwaitting for (A+256k, A+512k) finish.\n\nThere is no request to disk in T3 and T4, so readahead pipeline breaks.\n\nWe really don\u0027t need block plug for generic_file_aio_read() for buffered\nI/O. The readahead already has plug and has fine grained control when I/O\nshould be submitted. Deleting plug for buffered I/O fixes the regression.\n\nOne side effect is plug makes the request size 256k, the size is 128k\nwithout it. This is because default ra size is 128k and not a reason we\nneed plug here.\n\nVivek said:\n\n: We submit some readahead IO to device request queue but because of nested\n: plug, queue never gets unplugged. When read logic reaches a page which is\n: not in page cache, it waits for page to be read from the disk\n: (lock_page_killable()) and that time we flush the plug list.\n:\n: So effectively read ahead logic is kind of broken in parts because of\n: nested plugging. Removing top level plug (generic_file_aio_read()) for\n: buffered reads, will allow unplugging queue earlier for readahead.\n\nSigned-off-by: Shaohua Li \u003cshaohua.li@intel.com\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nReported-by: Herbert Poetzl \u003cherbert@13thfloor.at\u003e\nTested-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nCc: Jens Axboe \u003caxboe@kernel.dk\u003e\nCc: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3167760f83899ccda312b9ad9306ec9e5dda06d4", "tree": "a040add1a9b267cb0fa92847c88526c1329a258b", "parents": [ "972b2c719990f91eb3b2310d44ef8a2d38955a14" ], "author": { "name": "Dan Magenheimer", "email": "dan.magenheimer@oracle.com", "time": "Wed Sep 21 11:56:28 2011 -0400" }, "committer": { "name": "Konrad Rzeszutek Wilk", "email": "konrad.wilk@oracle.com", "time": "Mon Jan 23 16:06:24 2012 -0500" }, "message": "mm: cleancache: s/flush/invalidate/\n\nPer akpm suggestions alter the use of the term flush to be\ninvalidate. The next patch will do this across all MM.\n\nThis change is completely cosmetic.\n\n[v9: akpm@linux-foundation.org: change \"flush\" to \"invalidate\", part 3]\n\nSigned-off-by: Dan Magenheimer \u003cdan.magenheimer@oracle.com\u003e\nCc: Kamezawa Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Jan Beulich \u003cJBeulich@novell.com\u003e\nReviewed-by: Seth Jennings \u003csjenning@linux.vnet.ibm.com\u003e\nCc: Jeremy Fitzhardinge \u003cjeremy@goop.org\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Nitin Gupta \u003cngupta@vflare.org\u003e\nCc: Matthew Wilcox \u003cmatthew@wil.cx\u003e\nCc: Chris Mason \u003cchris.mason@oracle.com\u003e\nCc: Rik Riel \u003criel@redhat.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n[v10: Fixed fs: move code out of buffer.c conflict change]\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n" }, { "commit": "ab936cbcd02072a34b60d268f94440fd5cf1970b", "tree": "d37e3e3c54cc4cc691a428b6ceb71b4b40e4f42b", "parents": [ "28d82dc1c4edbc352129f97f4ca22624d1fe61de" ], "author": { "name": "KAMEZAWA Hiroyuki", "email": "kamezawa.hiroyu@jp.fujitsu.com", "time": "Thu Jan 12 17:17:44 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 12 20:13:04 2012 -0800" }, "message": "memcg: add mem_cgroup_replace_page_cache() to fix LRU issue\n\nCommit ef6a3c6311 (\"mm: add replace_page_cache_page() function\") added a\nfunction replace_page_cache_page(). This function replaces a page in the\nradix-tree with a new page. WHen doing this, memory cgroup needs to fix\nup the accounting information. memcg need to check PCG_USED bit etc.\n\nIn some(many?) cases, \u0027newpage\u0027 is on LRU before calling\nreplace_page_cache(). So, memcg\u0027s LRU accounting information should be\nfixed, too.\n\nThis patch adds mem_cgroup_replace_page_cache() and removes the old hooks.\n In that function, old pages will be unaccounted without touching\nres_counter and new page will be accounted to the memcg (of old page).\nWHen overwriting pc-\u003emem_cgroup of newpage, take zone-\u003elru_lock and avoid\nraces with LRU handling.\n\nBackground:\n replace_page_cache_page() is called by FUSE code in its splice() handling.\n Here, \u0027newpage\u0027 is replacing oldpage but this newpage is not a newly allocated\n page and may be on LRU. LRU mis-accounting will be critical for memory cgroup\n because rmdir() checks the whole LRU is empty and there is no account leak.\n If a page is on the other LRU than it should be, rmdir() will fail.\n\nThis bug was added in March 2011, but no bug report yet. I guess there\nare not many people who use memcg and FUSE at the same time with upstream\nkernels.\n\nThe result of this bug is that admin cannot destroy a memcg because of\naccount leak. So, no panic, no deadlock. And, even if an active cgroup\nexist, umount can succseed. So no problem at shutdown.\n\nSigned-off-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nAcked-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.cz\u003e\nCc: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0faa70cb0180d45a06208e54b552a538aabb8a30", "tree": "7d734a556ad1568301cf802c5afbaa83cc0bdcd5", "parents": [ "a756cf5908530e8b40bdf569eb48b40139e8d7fd" ], "author": { "name": "Johannes Weiner", "email": "jweiner@redhat.com", "time": "Tue Jan 10 15:07:53 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 10 16:30:43 2012 -0800" }, "message": "mm: filemap: pass __GFP_WRITE from grab_cache_page_write_begin()\n\nTell the page allocator that pages allocated through\ngrab_cache_page_write_begin() are expected to become dirty soon.\n\nSigned-off-by: Johannes Weiner \u003cjweiner@redhat.com\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Mel Gorman \u003cmgorman@suse.de\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nReviewed-by: Michal Hocko \u003cmhocko@suse.cz\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Dave Chinner \u003cdavid@fromorbit.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nCc: Shaohua Li \u003cshaohua.li@intel.com\u003e\nCc: Chris Mason \u003cchris.mason@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "649fc7b1b046eb98bf9e3fe20c9d11f629293140", "tree": "3e80a456d1921766e004e5100783ecbaff3cee57", "parents": [ "52ef0c042bf06f6aef382fade175075627beebc1" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Jul 26 04:31:05 2011 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Jan 03 22:55:14 2012 -0500" }, "message": "should_remove_suid(): inode-\u003ei_mode is umode_t\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "e6f67b8c05f5e129e126f4409ddac6f25f58ffcb", "tree": "4df809de3016d2e84467bded094c192adaf5218a", "parents": [ "7b21a202b87b7f18587bd15ee6bdb477fed0fb43" ], "author": { "name": "Dave Kleikamp", "email": "dave.kleikamp@oracle.com", "time": "Wed Dec 21 11:05:48 2011 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Dec 21 17:02:46 2011 -0800" }, "message": "vfs: __read_cache_page should use gfp argument rather than GFP_KERNEL\n\nlockdep reports a deadlock in jfs because a special inode\u0027s rw semaphore\nis taken recursively. The mapping\u0027s gfp mask is GFP_NOFS, but is not\nused when __read_cache_page() calls add_to_page_cache_lru().\n\nSigned-off-by: Dave Kleikamp \u003cdave.kleikamp@oracle.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a50527b19c62c808a7fca022816fff88a50b948d", "tree": "6f9bb1a5176b1a428ce272d398a2c784ee12f2eb", "parents": [ "786228ab3095fe18dba3bc0d62055a123991d9d9" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Fri Dec 02 09:17:02 2011 +0800" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Fri Dec 02 09:17:05 2011 +0800" }, "message": "fs: Make write(2) interruptible by a fatal signal\n\nCurrently write(2) to a file is not interruptible by any signal.\nSometimes this is desirable, e.g. when you want to quickly kill a\nprocess hogging your disk. Also, with commit 499d05ecf990 (\"mm: Make\ntask in balance_dirty_pages() killable\"), it\u0027s necessary to abort the\ncurrent write accordingly to avoid it quickly dirtying lots more pages\nat unthrottled rate.\n\nThis patch makes write interruptible by SIGKILL. We do not allow write\nto be interruptible by any other signal because that has larger\npotential of screwing some badly written applications.\n\nReported-by: Kazuya Mio \u003ck-mio@sx.jp.nec.com\u003e\nTested-by: Kazuya Mio \u003ck-mio@sx.jp.nec.com\u003e\nAcked-by: Matthew Wilcox \u003cmatthew.r.wilcox@intel.com\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "b95f1b31b75588306e32b2afd32166cad48f670b", "tree": "b5496144e41b117cfe5ae70b145b5351709ec4d0", "parents": [ "b9e15bafdf1aa20791cdefdcbf1ccf7d7aa03aaa" ], "author": { "name": "Paul Gortmaker", "email": "paul.gortmaker@windriver.com", "time": "Sun Oct 16 02:01:52 2011 -0400" }, "committer": { "name": "Paul Gortmaker", "email": "paul.gortmaker@windriver.com", "time": "Mon Oct 31 09:20:12 2011 -0400" }, "message": "mm: Map most files to use export.h instead of module.h\n\nThe files changed within are only using the EXPORT_SYMBOL\nmacro variants. They are not using core modular infrastructure\nand hence don\u0027t need module.h but only the export.h header.\n\nSigned-off-by: Paul Gortmaker \u003cpaul.gortmaker@windriver.com\u003e\n" }, { "commit": "39be79c16f2b8eb07dd0d4e965cddfe39cc0534a", "tree": "821611221295d47c671ec72e1fb558efcedff03b", "parents": [ "c3b92c8787367a8bb53d57d9789b558f1295cc96" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Thu Oct 27 23:53:08 2011 +0200" }, "committer": { "name": "Christoph Hellwig", "email": "hch@serles.lst.de", "time": "Fri Oct 28 13:55:08 2011 +0200" }, "message": "vfs: iov_iter: have iov_iter_advance decrement nr_segs appropriately\n\nCurrently, when you call iov_iter_advance, then the pointer to the iovec\narray can be incremented, but it does not decrement the nr_segs value in\nthe iov_iter struct. The result is a iov_iter struct with a nr_segs\nvalue that goes beyond the end of the array.\n\nWhile I\u0027m not aware of anything that\u0027s specifically broken by this, it\nseems odd and a bit dangerous not to decrement that value. If someone\nwere to trust the nr_segs value to be correct, then they could end up\nwalking off the end of the array.\n\nChanging this might also provide some micro-optimization when dealing\nwith the last iovec in an array. Many of the other routines that deal\nwith iov_iter have optimized codepaths when nr_segs \u003d\u003d 1.\n\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n" }, { "commit": "cc39c6a9bbdebfcf1a7dee64d83bf302bc38d941", "tree": "5c8c8b127e8bd31300fba1f5aa155c6463c4c78f", "parents": [ "4f5b04800a224aadb6cffcbbc3d3fa26e2367c7f" ], "author": { "name": "Shaohua Li", "email": "shaohua.li@intel.com", "time": "Thu Sep 15 08:45:19 2011 +0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 14 18:17:56 2011 -0700" }, "message": "mm: account skipped entries to avoid looping in find_get_pages\n\nThe found entries by find_get_pages() could be all swap entries. In\nthis case we skip the entries, but make sure the skipped entries are\naccounted, so we don\u0027t keep looping.\n\nUsing nr_found \u003e nr_skip to simplify code as suggested by Eric.\n\nReported-and-tested-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: Shaohua Li \u003cshaohua.li@intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8079b1c859c44f27d63da4951f5038a16589a563", "tree": "677e77a4c9d3b9c33427196cb5f860ebcf530b15", "parents": [ "e504f3fdd63d486d45b18009e5a65f2e329acb0a" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Wed Aug 03 16:21:28 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 03 14:25:24 2011 -1000" }, "message": "mm: clarify the radix_tree exceptional cases\n\nMake the radix_tree exceptional cases, mostly in filemap.c, clearer.\n\nIt\u0027s hard to devise a suitable snappy name that illuminates the use by\nshmem/tmpfs for swap, while keeping filemap/pagecache/radix_tree\ngenerality. And akpm points out that /* radix_tree_deref_retry(page) */\ncomments look like calls that have been commented out for unknown\nreason.\n\nSkirt the naming difficulty by rearranging these blocks to handle the\ntransient radix_tree_deref_retry(page) case first; then just explain the\nremaining shmem/tmpfs swap case in a comment.\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "31475dd611209413bace21651a400afb91d0bd9d", "tree": "679c1cb33e211a4f2e5c0f47bb836393c9454d35", "parents": [ "69f07ec938712b58755add82dd3d0b35f01317cc" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Wed Aug 03 16:21:27 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 03 14:25:24 2011 -1000" }, "message": "mm: a few small updates for radix-swap\n\nRemove PageSwapBacked (!page_is_file_cache) cases from\nadd_to_page_cache_locked() and add_to_page_cache_lru(): those pages now\ngo through shmem_add_to_page_cache().\n\nRemove a comment on maximum tmpfs size from fsstack_copy_inode_size(),\nand add a comment on swap entries to invalidate_mapping_pages().\n\nAnd mincore_page() uses find_get_page() on what might be shmem or a\ntmpfs file: allow for a radix_tree_exceptional_entry(), and proceed to\nfind_get_page() on swapper_space if so (oh, swapper_space needs #ifdef).\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a2c16d6cb0e478812829ca84aeabd02e36af35eb", "tree": "7803a522da5deee7ce753dc5404dca01f42aa176", "parents": [ "6328650bb4d854a7dc1498d1c0048b838b0d340c" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Wed Aug 03 16:21:19 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 03 14:25:22 2011 -1000" }, "message": "mm: let swap use exceptional entries\n\nIf swap entries are to be stored along with struct page pointers in a\nradix tree, they need to be distinguished as exceptional entries.\n\nMost of the handling of swap entries in radix tree will be contained in\nshmem.c, but a few functions in filemap.c\u0027s common code need to check\nfor their appearance: find_get_page(), find_lock_page(),\nfind_get_pages() and find_get_pages_contig().\n\nSo as not to slow their fast paths, tuck those checks inside the\nexisting checks for unlikely radix_tree_deref_slot(); except for\nfind_lock_page(), where it is an added test. And make it a BUG in\nfind_get_pages_tag(), which is not applied to tmpfs files.\n\nA part of the reason for eliminating shmem_readpage() earlier, was to\nminimize the places where common code would need to allow for swap\nentries.\n\nThe swp_entry_t known to swapfile.c must be massaged into a slightly\ndifferent form when stored in the radix tree, just as it gets massaged\ninto a pte_t when stored in page tables.\n\nIn an i386 kernel this limits its information (type and page offset) to\n30 bits: given 32 \"types\" of swapfile and 4kB pagesize, that\u0027s a maximum\nswapfile size of 128GB. Which is less than the 512GB we previously\nallowed with X86_PAE (where the swap entry can occupy the entire upper\n32 bits of a pte_t), but not a new limitation on 32-bit without PAE; and\nthere\u0027s not a new limitation on 64-bit (where swap filesize is already\nlimited to 16TB by a 32-bit page offset). Thirty areas of 128GB is\nprobably still enough swap for a 64GB 32-bit machine.\n\nProvide swp_to_radix_entry() and radix_to_swp_entry() conversions, and\nenforce filesize limit in read_swap_header(), just as for ptes.\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6328650bb4d854a7dc1498d1c0048b838b0d340c", "tree": "0e265dc86f7c4451647c1e227843cdd1530f9651", "parents": [ "70d327198a434edb95b3d858bc8010b8add28e3e" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Wed Aug 03 16:21:18 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 03 14:25:22 2011 -1000" }, "message": "radix_tree: exceptional entries and indices\n\nA patchset to extend tmpfs to MAX_LFS_FILESIZE by abandoning its\npeculiar swap vector, instead keeping a file\u0027s swap entries in the same\nradix tree as its struct page pointers: thus saving memory, and\nsimplifying its code and locking.\n\nThis patch:\n\nThe radix_tree is used by several subsystems for different purposes. A\nmajor use is to store the struct page pointers of a file\u0027s pagecache for\nmemory management. But what if mm wanted to store something other than\npage pointers there too?\n\nThe low bit of a radix_tree entry is already used to denote an indirect\npointer, for internal use, and the unlikely radix_tree_deref_retry()\ncase.\n\nDefine the next bit as denoting an exceptional entry, and supply inline\nfunctions radix_tree_exception() to return non-0 in either unlikely\ncase, and radix_tree_exceptional_entry() to return non-0 in the second\ncase.\n\nIf a subsystem already uses radix_tree with that bit set, no problem: it\ndoes not affect internal workings at all, but is defined for the\nconvenience of those storing well-aligned pointers in the radix_tree.\n\nThe radix_tree_gang_lookups have an implicit assumption that the caller\ncan deduce the offset of each entry returned e.g. by the page-\u003eindex of\na struct page. But that may not be feasible for some kinds of item to\nbe stored there.\n\nradix_tree_gang_lookup_slot() allow for an optional indices argument,\noutput array in which to return those offsets. The same could be added\nto other radix_tree_gang_lookups, but for now keep it to the only one\nfor which we need it.\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f01ef569cddb1a8627b1c6b3a134998ad1cf4b22", "tree": "29ea1a0942c8549c24411e976cd6891c7e995e89", "parents": [ "a93a1329271038f0e8337061d3b41b3b212a851e", "bcff25fc8aa47a13faff8b4b992589813f7b450a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 10:39:54 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 10:39:54 2011 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/writeback\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/writeback: (27 commits)\n mm: properly reflect task dirty limits in dirty_exceeded logic\n writeback: don\u0027t busy retry writeback on new/freeing inodes\n writeback: scale IO chunk size up to half device bandwidth\n writeback: trace global_dirty_state\n writeback: introduce max-pause and pass-good dirty limits\n writeback: introduce smoothed global dirty limit\n writeback: consolidate variable names in balance_dirty_pages()\n writeback: show bdi write bandwidth in debugfs\n writeback: bdi write bandwidth estimation\n writeback: account per-bdi accumulated written pages\n writeback: make writeback_control.nr_to_write straight\n writeback: skip tmpfs early in balance_dirty_pages_ratelimited_nr()\n writeback: trace event writeback_queue_io\n writeback: trace event writeback_single_inode\n writeback: remove .nonblocking and .encountered_congestion\n writeback: remove writeback_control.more_io\n writeback: skip balance_dirty_pages() for in-memory fs\n writeback: add bdi_dirty_limit() kernel-doc\n writeback: avoid extra sync work at enqueue time\n writeback: elevate queue_io() into wb_writeback()\n ...\n\nFix up trivial conflicts in fs/fs-writeback.c and mm/filemap.c\n" }, { "commit": "b85e0effd3dcbf9118b896232f59526ab1a39a74", "tree": "32200b2e4052d50d4eb1771e555eaf66d7c4cfc8", "parents": [ "8a549bea51138be2126a2cc6aabe8f17ef66b79b" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Mon Jul 25 17:12:25 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 25 20:57:10 2011 -0700" }, "message": "mm: consistent truncate and invalidate loops\n\nMake the pagevec_lookup loops in truncate_inode_pages_range(),\ninvalidate_mapping_pages() and invalidate_inode_pages2_range() more\nconsistent with each other.\n\nThey were relying upon page-\u003eindex of an unlocked page, but apologizing\nfor it: accept it, embrace it, add comments and WARN_ONs, and simplify the\nindex handling.\n\ninvalidate_inode_pages2_range() had special handling for a wrapped\npage-\u003eindex + 1 \u003d 0 case; but MAX_LFS_FILESIZE doesn\u0027t let us anywhere\nnear there, and a corrupt page-\u003eindex in the radix_tree could cause more\ntrouble than that would catch. Remove that wrapped handling.\n\ninvalidate_inode_pages2_range() uses min() to limit the pagevec_lookup\nwhen near the end of the range: copy that into the other two, although\nit\u0027s less useful than you might think (it limits the use of the buffer,\nrather than the indices looked up).\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5e5358e7cf48aa079b8761a7d806ad536023745c", "tree": "8a0668541c8b356b46c32eea7f28a2f72259eef6", "parents": [ "df077ac4687500e02a273a628057ff5ab17dc19f" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Mon Jul 25 17:12:23 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 25 20:57:10 2011 -0700" }, "message": "mm: cleanup descriptions of filler arg\n\nThe often-NULL data arg to read_cache_page() and read_mapping_page()\nfunctions is misdescribed as \"destination for read data\": no, it\u0027s the\nfirst arg to the filler function, often struct file * to -\u003ereadpage().\n\nSatisfy checkpatch.pl on those filler prototypes, and tidy up the\ndeclarations in linux/pagemap.h.\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "bd5fe6c5eb9c548d7f07fe8f89a150bb6705e8e3", "tree": "ef5341c7747f809aec7ae233f6e3ef90af39be5f", "parents": [ "f9b5570d7fdedff32a2e78102bfb54cd1b12b289" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Fri Jun 24 14:29:43 2011 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Jul 20 20:47:46 2011 -0400" }, "message": "fs: kill i_alloc_sem\n\ni_alloc_sem is a rather special rw_semaphore. It\u0027s the last one that may\nbe released by a non-owner, and it\u0027s write side is always mirrored by\nreal exclusion. It\u0027s intended use it to wait for all pending direct I/O\nrequests to finish before starting a truncate.\n\nReplace it with a hand-grown construct:\n\n - exclusion for truncates is already guaranteed by i_mutex, so it can\n simply fall way\n - the reader side is replaced by an i_dio_count member in struct inode\n that counts the number of pending direct I/O requests. Truncate can\u0027t\n proceed as long as it\u0027s non-zero\n - when i_dio_count reaches non-zero we wake up a pending truncate using\n wake_up_bit on a new bit in i_flags\n - new references to i_dio_count can\u0027t appear while we are waiting for\n it to read zero because the direct I/O count always needs i_mutex\n (or an equivalent like XFS\u0027s i_iolock) for starting a new operation.\n\nThis scheme is much simpler, and saves the space of a spinlock_t and a\nstruct list_head in struct inode (typically 160 bits on a non-debug 64-bit\nsystem).\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "f758eeabeb96f878c860e8f110f94ec8820822a9", "tree": "fea5a465aa0aa38c6c9263eb264acbeb7f722c02", "parents": [ "424b351fe1901fc909fd0ca4f21dab58f24c1aac" ], "author": { "name": "Christoph Hellwig", "email": "hch@infradead.org", "time": "Thu Apr 21 18:19:44 2011 -0600" }, "committer": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Wed Jun 08 08:25:21 2011 +0800" }, "message": "writeback: split inode_wb_list_lock into bdi_writeback.list_lock\n\nSplit the global inode_wb_list_lock into a per-bdi_writeback list_lock,\nas it\u0027s currently the most contended lock in the system for metadata\nheavy workloads. It won\u0027t help for single-filesystem workloads for\nwhich we\u0027ll need the I/O-less balance_dirty_pages, but at least we\ncan dedicate a cpu to spinning on each bdi now for larger systems.\n\nBased on earlier patches from Nick Piggin and Dave Chinner.\n\nIt reduces lock contentions to 1/4 in this test case:\n10 HDD JBOD, 100 dd on each disk, XFS, 6GB ram\n\nlock_stat version 0.3\n-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total\n-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\nvanilla 2.6.39-rc3:\n inode_wb_list_lock: 42590 44433 0.12 147.74 144127.35 252274 886792 0.08 121.34 917211.23\n ------------------\n inode_wb_list_lock 2 [\u003cffffffff81165da5\u003e] bdev_inode_switch_bdi+0x29/0x85\n inode_wb_list_lock 34 [\u003cffffffff8115bd0b\u003e] inode_wb_list_del+0x22/0x49\n inode_wb_list_lock 12893 [\u003cffffffff8115bb53\u003e] __mark_inode_dirty+0x170/0x1d0\n inode_wb_list_lock 10702 [\u003cffffffff8115afef\u003e] writeback_single_inode+0x16d/0x20a\n ------------------\n inode_wb_list_lock 2 [\u003cffffffff81165da5\u003e] bdev_inode_switch_bdi+0x29/0x85\n inode_wb_list_lock 19 [\u003cffffffff8115bd0b\u003e] inode_wb_list_del+0x22/0x49\n inode_wb_list_lock 5550 [\u003cffffffff8115bb53\u003e] __mark_inode_dirty+0x170/0x1d0\n inode_wb_list_lock 8511 [\u003cffffffff8115b4ad\u003e] writeback_sb_inodes+0x10f/0x157\n\n2.6.39-rc3 + patch:\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock: 11383 11657 0.14 151.69 40429.51 90825 527918 0.11 145.90 556843.37\n ------------------------\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 10 [\u003cffffffff8115b189\u003e] inode_wb_list_del+0x5f/0x86\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 1493 [\u003cffffffff8115b1ed\u003e] writeback_inodes_wb+0x3d/0x150\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 3652 [\u003cffffffff8115a8e9\u003e] writeback_sb_inodes+0x123/0x16f\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 1412 [\u003cffffffff8115a38e\u003e] writeback_single_inode+0x17f/0x223\n ------------------------\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 3 [\u003cffffffff8110b5af\u003e] bdi_lock_two+0x46/0x4b\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 6 [\u003cffffffff8115b189\u003e] inode_wb_list_del+0x5f/0x86\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 2061 [\u003cffffffff8115af97\u003e] __mark_inode_dirty+0x173/0x1cf\n \u0026(\u0026wb-\u003elist_lock)-\u003erlock 2629 [\u003cffffffff8115a8e9\u003e] writeback_sb_inodes+0x123/0x16f\n\nhughd@google.com: fix recursive lock when bdi_lock_two() is called with new the same as old\nakpm@linux-foundation.org: cleanup bdev_inode_switch_bdi() comment\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\n" }, { "commit": "9e1f1de02c2275d7172e18dc4e7c2065777611bf", "tree": "15e9d202e64275cdbff6ed1d54804da5966d7d8d", "parents": [ "1fa7b6a29c61358cc2ca6f64cef4aa0e1a7ca74c" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Jun 03 18:24:58 2011 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Jun 03 18:24:58 2011 -0400" }, "message": "more conservative S_NOSEC handling\n\nCaching \"we have already removed suid/caps\" was overenthusiastic as merged.\nOn network filesystems we might have had suid/caps set on another client,\nsilently picked by this client on revalidate, all of that *without* clearing\nthe S_NOSEC flag.\n\nAFAICS, the only reasonably sane way to deal with that is\n\t* new superblock flag; unless set, S_NOSEC is not going to be set.\n\t* local block filesystems set it in their -\u003emount() (more accurately,\nmount_bdev() does, so does btrfs -\u003emount(), users of mount_bdev() other than\nlocal block ones clear it)\n\t* if any network filesystem (or a cluster one) wants to use S_NOSEC,\nit\u0027ll need to set MS_NOSEC in sb-\u003es_flags *AND* take care to clear S_NOSEC when\ninode attribute changes are picked from other clients.\n\nIt\u0027s not an earth-shattering hole (anybody that can set suid on another client\nwill almost certainly be able to write to the file before doing that anyway),\nbut it\u0027s a bug that needs fixing.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "69b4573296469fd3f70cf7044693074980517067", "tree": "aea41eacb2a0f32748145a59bb8dc300b4485f36", "parents": [ "d76ee18a8551e33ad7dbd55cac38bc7b094f3abb" ], "author": { "name": "Andi Kleen", "email": "ak@linux.intel.com", "time": "Sat May 28 08:25:51 2011 -0700" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat May 28 12:02:09 2011 -0400" }, "message": "Cache xattr security drop check for write v2\n\nSome recent benchmarking on btrfs showed that a major scaling bottleneck\non large systems on btrfs is currently the xattr lookup on every write.\n\nWhy xattr lookup on every write I hear you ask?\n\nwrite wants to drop suid and security related xattrs that could set o\ncapabilities for executables. To do that it currently looks up\nsecurity.capability on EVERY write (even for non executables) to decide\nwhether to drop it or not.\n\nIn btrfs this causes an additional tree walk, hitting some per file system\nlocks and quite bad scalability. In a simple read workload on a 8S\nsystem I saw over 90% CPU time in spinlocks related to that.\n\nChris Mason tells me this is also a problem in ext4, where it hits\nthe global mbcache lock.\n\nThis patch adds a simple per inode to avoid this problem. We only\ndo the lookup once per file and then if there is no xattr cache\nthe decision. All xattr changes clear the flag.\n\nI also used the same flag to avoid the suid check, although\nthat one is pretty cheap.\n\nA file system can also set this flag when it creates the inode,\nif it has a cheap way to do so. This is done for some common file systems\nin followon patches.\n\nWith this patch a major part of the lock contention disappears\nfor btrfs. Some testing on smaller systems didn\u0027t show significant\nperformance changes, but at least it helps the larger systems\nand is generally more efficient.\n\nv2: Rename is_sgid. add file system helper.\nCc: chris.mason@oracle.com\nCc: josef@redhat.com\nCc: viro@zeniv.linux.org.uk\nCc: agruen@linbit.com\nCc: Serge E. Hallyn \u003cserue@us.ibm.com\u003e\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "3d08bcc887a1c8d12be8d81f747ffa2e8a44b67b", "tree": "258f4e3d384e451c921c5e0c503fd368f236a20a", "parents": [ "98702467f829177b3993f17da9fe5c202d160e5e" ], "author": { "name": "Darrick J. Wong", "email": "djwong@us.ibm.com", "time": "Fri May 27 12:23:34 2011 -0700" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat May 28 01:03:21 2011 -0400" }, "message": "mm: Wait for writeback when grabbing pages to begin a write\n\nWhen grabbing a page for a buffered IO write, the mm should wait for writeback\non the page to complete so that the page does not become writable during the IO\noperation. This change is needed to provide page stability during writes for\nall filesystems.\n\nSigned-off-by: Darrick J. Wong \u003cdjwong@us.ibm.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "456f998ec817ebfa254464be4f089542fa390645", "tree": "5976aa500638f0bbade1a672233cad71765b89b8", "parents": [ "406eb0c9ba765eb066406fd5ce9d5e2b169a4d5a" ], "author": { "name": "Ying Han", "email": "yinghan@google.com", "time": "Thu May 26 16:25:38 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 26 17:12:36 2011 -0700" }, "message": "memcg: add the pagefault count into memcg stats\n\nTwo new stats in per-memcg memory.stat which tracks the number of page\nfaults and number of major page faults.\n\n \"pgfault\"\n \"pgmajfault\"\n\nThey are different from \"pgpgin\"/\"pgpgout\" stat which count number of\npages charged/discharged to the cgroup and have no meaning of reading/\nwriting page to disk.\n\nIt is valuable to track the two stats for both measuring application\u0027s\nperformance as well as the efficiency of the kernel page reclaim path.\nCounting pagefaults per process is useful, but we also need the aggregated\nvalue since processes are monitored and controlled in cgroup basis in\nmemcg.\n\nFunctional test: check the total number of pgfault/pgmajfault of all\nmemcgs and compare with global vmstat value:\n\n $ cat /proc/vmstat | grep fault\n pgfault 1070751\n pgmajfault 553\n\n $ cat /dev/cgroup/memory.stat | grep fault\n pgfault 1071138\n pgmajfault 553\n total_pgfault 1071142\n total_pgmajfault 553\n\n $ cat /dev/cgroup/A/memory.stat | grep fault\n pgfault 199\n pgmajfault 0\n total_pgfault 199\n total_pgmajfault 0\n\nPerformance test: run page fault test(pft) wit 16 thread on faulting in\n15G anon pages in 16G container. There is no regression noticed on the\n\"flt/cpu/s\"\n\nSample output from pft:\n\n TAG pft:anon-sys-default:\n Gb Thr CLine User System Wall flt/cpu/s fault/wsec\n 15 16 1 0.67s 233.41s 14.76s 16798.546 266356.260\n\n +-------------------------------------------------------------------------+\n N Min Max Median Avg Stddev\n x 10 16682.962 17344.027 16913.524 16928.812 166.5362\n + 10 16695.568 16923.896 16820.604 16824.652 84.816568\n No difference proven at 95.0% confidence\n\n[akpm@linux-foundation.org: fix build]\n[hughd@google.com: shmem fix]\nSigned-off-by: Ying Han \u003cyinghan@google.com\u003e\nAcked-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Daisuke Nishimura \u003cnishimura@mxp.nes.nec.co.jp\u003e\nAcked-by: Balbir Singh \u003cbalbir@linux.vnet.ibm.com\u003e\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f8d613e2a665bf1be9628a3c3f9bafe7599b32c0", "tree": "98d4da8d0e1a5fb1d9064626b4b96d95ccf26375", "parents": [ "8a0599dd2471f2a2e409498c08a0ab339057ad06", "5bc20fc59706214d9591c11e1938a629d3538c12" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 26 10:50:56 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 26 10:50:56 2011 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/djm/tmem\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/djm/tmem:\n xen: cleancache shim to Xen Transcendent Memory\n ocfs2: add cleancache support\n ext4: add cleancache support\n btrfs: add cleancache support\n ext3: add cleancache support\n mm/fs: add hooks to support cleancache\n mm: cleancache core ops functions and config\n fs: add field to superblock to support cleancache\n mm/fs: cleancache documentation\n\nFix up trivial conflict in fs/btrfs/extent_io.c due to includes\n" }, { "commit": "c515e1fd361c2a08a9c2eb139396ec30a4f477dc", "tree": "cb1a27f27f065400fc79dc287594843e169d74ab", "parents": [ "077b1f83a69d94f2918630a882d74939baca0bce" ], "author": { "name": "Dan Magenheimer", "email": "dan.magenheimer@oracle.com", "time": "Thu May 26 10:01:43 2011 -0600" }, "committer": { "name": "Dan Magenheimer", "email": "dan.magenheimer@oracle.com", "time": "Thu May 26 10:01:43 2011 -0600" }, "message": "mm/fs: add hooks to support cleancache\n\nThis fourth patch of eight in this cleancache series provides the\ncore hooks in VFS for: initializing cleancache per filesystem;\ncapturing clean pages reclaimed by page cache; attempting to get\npages from cleancache before filesystem read; and ensuring coherency\nbetween pagecache, disk, and cleancache. Note that the placement\nof these hooks was stable from 2.6.18 to 2.6.38; a minor semantic\nchange was required due to a patchset in 2.6.39.\n\nAll hooks become no-ops if CONFIG_CLEANCACHE is unset, or become\na check of a boolean global if CONFIG_CLEANCACHE is set but no\ncleancache \"backend\" has claimed cleancache_ops.\n\nDetails and a FAQ can be found in Documentation/vm/cleancache.txt\n\n[v8: minchan.kim@gmail.com: adapt to new remove_from_page_cache function]\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\nSigned-off-by: Dan Magenheimer \u003cdan.magenheimer@oracle.com\u003e\nReviewed-by: Jeremy Fitzhardinge \u003cjeremy@goop.org\u003e\nReviewed-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Al Viro \u003cviro@ZenIV.linux.org.uk\u003e\nCc: Matthew Wilcox \u003cmatthew@wil.cx\u003e\nCc: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Rik Van Riel \u003criel@redhat.com\u003e\nCc: Jan Beulich \u003cJBeulich@novell.com\u003e\nCc: Andreas Dilger \u003cadilger@sun.com\u003e\nCc: Ted Ts\u0027o \u003ctytso@mit.edu\u003e\nCc: Mark Fasheh \u003cmfasheh@suse.com\u003e\nCc: Joel Becker \u003cjoel.becker@oracle.com\u003e\nCc: Nitin Gupta \u003cngupta@vflare.org\u003e\n" }, { "commit": "2cbea1d3ab11946885d37a2461072ee4d687cb4e", "tree": "aab301cb3da1e633bbd7df2acc4b4c2e4f777b35", "parents": [ "207d04baa3591a354711e863dd90087fc75873b3" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue May 24 17:12:30 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:27 2011 -0700" }, "message": "readahead: trigger mmap sequential readahead on PG_readahead\n\nPreviously the mmap sequential readahead is triggered by updating\nra-\u003eprev_pos on each page fault and compare it with current page offset.\n\nIt costs dirtying the cache line on each _minor_ page fault. So remove\nthe ra-\u003eprev_pos recording, and instead tag PG_readahead to trigger the\npossible sequential readahead. It\u0027s not only more simple, but also will\nwork more reliably and reduce cache line bouncing on concurrent page\nfaults on shared struct file.\n\nIn the mosbench exim benchmark which does multi-threaded page faults on\nshared struct file, the ra-\u003emmap_miss and ra-\u003eprev_pos updates are found\nto cause excessive cache line bouncing on tmpfs, which actually disabled\nreadahead totally (shmem_backing_dev_info.ra_pages \u003d\u003d 0).\n\nSo remove the ra-\u003eprev_pos recording, and instead tag PG_readahead to\ntrigger the possible sequential readahead. It\u0027s not only more simple, but\nalso will work more reliably on concurrent reads on shared struct file.\n\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nTested-by: Tim Chen \u003ctim.c.chen@intel.com\u003e\nReported-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "207d04baa3591a354711e863dd90087fc75873b3", "tree": "17498d55af5b2a588e7e7111e927a099236ca770", "parents": [ "275b12bf5486f6f531111fd3d7dbbf01df427cfe" ], "author": { "name": "Andi Kleen", "email": "ak@linux.intel.com", "time": "Tue May 24 17:12:29 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:26 2011 -0700" }, "message": "readahead: reduce unnecessary mmap_miss increases\n\nThe original INT_MAX is too large, reduce it to\n\n- avoid unnecessarily dirtying/bouncing the cache line\n\n- restore mmap read-around faster on changed access pattern\n\nBackground: in the mosbench exim benchmark which does multi-threaded page\nfaults on shared struct file, the ra-\u003emmap_miss updates are found to cause\nexcessive cache line bouncing on tmpfs. The ra state updates are needless\nfor tmpfs because it actually disabled readahead totally\n(shmem_backing_dev_info.ra_pages \u003d\u003d 0).\n\nTested-by: Tim Chen \u003ctim.c.chen@intel.com\u003e\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "275b12bf5486f6f531111fd3d7dbbf01df427cfe", "tree": "ebed0109eaa3e4a015bc1653f4d933f2e84c794f", "parents": [ "1495f230fa7750479c79e3656286b9183d662077" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue May 24 17:12:28 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:26 2011 -0700" }, "message": "readahead: return early when readahead is disabled\n\nReduce readahead overheads by returning early in do_sync_mmap_readahead().\n\ntmpfs has ra_pages\u003d0 and it can page fault really fast (not constraint by\nIO if not swapping).\n\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nTested-by: Tim Chen \u003ctim.c.chen@intel.com\u003e\nReported-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3d48ae45e72390ddf8cc5256ac32ed6f7a19cbea", "tree": "1f46db3a8424090dd8e0b58991fa5acc1a73e680", "parents": [ "97a894136f29802da19a15541de3c019e1ca147e" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Tue May 24 17:12:06 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:18 2011 -0700" }, "message": "mm: Convert i_mmap_lock to a mutex\n\nStraightforward conversion of i_mmap_lock to a mutex.\n\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: David Miller \u003cdavem@davemloft.net\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Russell King \u003crmk@arm.linux.org.uk\u003e\nCc: Paul Mundt \u003clethal@linux-sh.org\u003e\nCc: Jeff Dike \u003cjdike@addtoit.com\u003e\nCc: Richard Weinberger \u003crichard@nod.at\u003e\nCc: Tony Luck \u003ctony.luck@intel.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nCc: Namhyung Kim \u003cnamhyung@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "37b23e0525d393d48a7d59f870b3bc061a30ccdb", "tree": "467e6fcd785108a21d836e1aad8fc1a68aa72e17", "parents": [ "f62e00cc3a00bfbd394a79fc22b334c31f91bd5f" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Tue May 24 17:11:30 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:08 2011 -0700" }, "message": "x86,mm: make pagefault killable\n\nWhen an oom killing occurs, almost all processes are getting stuck at the\nfollowing two points.\n\n\t1) __alloc_pages_nodemask\n\t2) __lock_page_or_retry\n\n1) is not very problematic because TIF_MEMDIE leads to an allocation\nfailure and getting out from page allocator.\n\n2) is more problematic. In an OOM situation, zones typically don\u0027t have\npage cache at all and memory starvation might lead to greatly reduced IO\nperformance. When a fork bomb occurs, TIF_MEMDIE tasks don\u0027t die quickly,\nmeaning that a fork bomb may create new process quickly rather than the\noom-killer killing it. Then, the system may become livelocked.\n\nThis patch makes the pagefault interruptible by SIGKILL.\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nReviewed-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Matthew Wilcox \u003cwilly@linux.intel.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f62e00cc3a00bfbd394a79fc22b334c31f91bd5f", "tree": "29e0a3ac793e4a5a15bbbe4b4945470a41e820e7", "parents": [ "fa25c503dfa203b921199ea42c0046c89f2ed49f" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Tue May 24 17:11:29 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 25 08:39:08 2011 -0700" }, "message": "mm: introduce wait_on_page_locked_killable()\n\ncommit 2687a356 (\"Add lock_page_killable\") introduced killable\nlock_page(). Similarly this patch introdues killable\nwait_on_page_locked().\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nAcked-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Matthew Wilcox \u003cwilly@linux.intel.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d39dd11c3e6a7af5c20bfac40594db36cf270f42", "tree": "6384e07fa2f347b286cde9754c4507b5a738ab47", "parents": [ "30f5b28e7f937608e0407edaa459cc8161de81d9", "0b2d0724e26a335cd326eb7ad552c109116a8795" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 24 19:01:30 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 24 19:01:30 2011 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:\n fs: simplify iget \u0026 friends\n fs: pull inode-\u003ei_lock up out of writeback_single_inode\n fs: rename inode_lock to inode_hash_lock\n fs: move i_wb_list out from under inode_lock\n fs: move i_sb_list out from under inode_lock\n fs: remove inode_lock from iput_final and prune_icache\n fs: Lock the inode LRU list separately\n fs: factor inode disposal\n fs: protect inode-\u003ei_state with inode-\u003ei_lock\n autofs4: Do not potentially dereference NULL pointer returned by fget() in autofs_dev_ioctl_setpipefd()\n autofs4 - remove autofs4_lock\n autofs4 - fix d_manage() return on rcu-walk\n autofs4 - fix autofs4_expire_indirect() traversal\n autofs4 - fix dentry leak in autofs4_expire_direct()\n autofs4 - reinstate last used update on access\n vfs - check non-mountpoint dentry might block in __follow_mount_rcu()\n" }, { "commit": "a66979abad090b2765a6c6790c9fdeab996833f2", "tree": "e48b2d0fac8f96456286a503aeeb952620234961", "parents": [ "55fa6091d83160ca772fc37cebae45d42695a708" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Tue Mar 22 22:23:41 2011 +1100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Mar 24 21:17:51 2011 -0400" }, "message": "fs: move i_wb_list out from under inode_lock\n\nProtect the inode writeback list with a new global lock\ninode_wb_list_lock and use it to protect the list manipulations and\ntraversals. This lock replaces the inode_lock as the inodes on the\nlist can be validity checked while holding the inode-\u003ei_lock and\nhence the inode_lock is no longer needed to protect the list.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "250df6ed274d767da844a5d9f05720b804240197", "tree": "b74f49a86c4451d9e3e82f90e3f791163025be21", "parents": [ "3dc8fe4dca9cd3e4aa828ed36451e2bcfd2350da" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Tue Mar 22 22:23:36 2011 +1100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Mar 24 21:16:31 2011 -0400" }, "message": "fs: protect inode-\u003ei_state with inode-\u003ei_lock\n\nProtect inode state transitions and validity checks with the\ninode-\u003ei_lock. This enables us to make inode state transitions\nindependently of the inode_lock and is the first step to peeling\naway the inode_lock from the code.\n\nThis requires that __iget() is done atomically with i_state checks\nduring list traversals so that we don\u0027t race with another thread\nmarking the inode I_FREEING between the state check and grabbing the\nreference.\n\nAlso remove the unlock_new_inode() memory barrier optimisation\nrequired to avoid taking the inode_lock when clearing I_NEW.\nSimplify the code by simply taking the inode-\u003ei_lock around the\nstate change and wakeup. Because the wakeup is no longer tricky,\nremove the wake_up_inode() function and open code the wakeup where\nnecessary.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "6c5103890057b1bb781b26b7aae38d33e4c517d8", "tree": "e6e57961dcddcb5841acb34956e70b9dc696a880", "parents": [ "3dab04e6978e358ad2307bca563fabd6c5d2c58b", "9d2e157d970a73b3f270b631828e03eb452d525e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 24 10:16:26 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 24 10:16:26 2011 -0700" }, "message": "Merge branch \u0027for-2.6.39/core\u0027 of git://git.kernel.dk/linux-2.6-block\n\n* \u0027for-2.6.39/core\u0027 of git://git.kernel.dk/linux-2.6-block: (65 commits)\n Documentation/iostats.txt: bit-size reference etc.\n cfq-iosched: removing unnecessary think time checking\n cfq-iosched: Don\u0027t clear queue stats when preempt.\n blk-throttle: Reset group slice when limits are changed\n blk-cgroup: Only give unaccounted_time under debug\n cfq-iosched: Don\u0027t set active queue in preempt\n block: fix non-atomic access to genhd inflight structures\n block: attempt to merge with existing requests on plug flush\n block: NULL dereference on error path in __blkdev_get()\n cfq-iosched: Don\u0027t update group weights when on service tree\n fs: assign sb-\u003es_bdi to default_backing_dev_info if the bdi is going away\n block: Require subsystems to explicitly allocate bio_set integrity mempool\n jbd2: finish conversion from WRITE_SYNC_PLUG to WRITE_SYNC and explicit plugging\n jbd: finish conversion from WRITE_SYNC_PLUG to WRITE_SYNC and explicit plugging\n fs: make fsync_buffers_list() plug\n mm: make generic_writepages() use plugging\n blk-cgroup: Add unaccounted time to timeslice_used.\n block: fixup plugging stubs for !CONFIG_BLOCK\n block: remove obsolete comments for blkdev_issue_zeroout.\n blktrace: Use rq-\u003ecmd_flags directly in blk_add_trace_rq.\n ...\n\nFix up conflicts in fs/{aio.c,super.c}\n" }, { "commit": "5b280c0cc70062967bb9d630b216375b18db3a0b", "tree": "6242d234d08fdd433b99c425e25f6562cd51cd0f", "parents": [ "9d8aa4ea855e0d64bb6926acb5618e6d1e2ed344" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Tue Mar 22 16:33:07 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:04 2011 -0700" }, "message": "mm: don\u0027t return 0 too early from find_get_pages()\n\nCallers of find_get_pages(), or its wrapper pagevec_lookup() - notably\ntruncate_inode_pages_range() - stop looking further when it returns 0.\n\nBut if an interrupt comes just after its radix_tree_gang_lookup_slot(),\nespecially if we have preemptible RCU enabled, isn\u0027t it conceivable that\nall 14 pages returned could be removed from the page cache by\nshrink_page_list(), before find_get_pages() gets to process them? So\ncausing it to return 0 although there may be plenty more pages beyond.\n\nMake find_get_pages() and find_get_pages_tag() check for this unlikely\ncase, and restart should it occur; but callers of find_get_pages_contig()\nhave no such expectation, it\u0027s okay for that to return 0 early.\n\nI have not seen this in practice, just worried by the possibility.\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Salman Qazi \u003csqazi@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9d8aa4ea855e0d64bb6926acb5618e6d1e2ed344", "tree": "a21357f21084169e0a322f8fa8ce8abe343d4f06", "parents": [ "c033a93c0d961fc7ec5b0872649143e061d97dd4" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Tue Mar 22 16:33:06 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:04 2011 -0700" }, "message": "mm: remove worrying dead code from find_get_pages()\n\nThe radix_tree_deref_retry() case in find_get_pages() has a strange little\nexcrescence, not seen in the other gang lookups: it looks like the start\nof an abandoned attempt to guarantee forward progress in a case that\ncannot arise.\n\nret should always be 0 here: if it isn\u0027t, then going back to restart will\nleak references to pages already gotten. There used to be a comment\nsaying nr_found is necessarily 1 here: that\u0027s not quite true, but the\nradix_tree_deref_retry() case is peculiar to the entry at index 0, when we\nrace with it being moved out of the radix_tree root or back.\n\nRemove the worrisome two lines, add a brief comment here and in\nfind_get_pages_contig() and find_get_pages_tag(), and a WARN_ON in\nfind_get_pages() should it ever be seen elsewhere than at 0.\n\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Salman Qazi \u003csqazi@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e64a782fec684c29a8204c51b3cb554dce588592", "tree": "5ff0beb21b973f1ad0edc1e31b6a1c2ee4406bdc", "parents": [ "702cfbf93aaf3a091b0c64c8766c1ade0a820c38" ], "author": { "name": "Minchan Kim", "email": "minchan.kim@gmail.com", "time": "Tue Mar 22 16:32:44 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:02 2011 -0700" }, "message": "mm: change __remove_from_page_cache()\n\nNow we renamed remove_from_page_cache with delete_from_page_cache. As\nconsistency of __remove_from_swap_cache and remove_from_swap_cache, we\nchange internal page cache handling function name, too.\n\nSigned-off-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nReviewed-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nReviewed-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "702cfbf93aaf3a091b0c64c8766c1ade0a820c38", "tree": "05416ad8133cbe64cb3b0ae475f486dd6ade8457", "parents": [ "5adc7b518b54f7af2b8395d2035898340d96b1d5" ], "author": { "name": "Minchan Kim", "email": "minchan.kim@gmail.com", "time": "Tue Mar 22 16:32:43 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:02 2011 -0700" }, "message": "mm: goodbye remove_from_page_cache()\n\nNow delete_from_page_cache() replaces remove_from_page_cache(). So we\nremove remove_from_page_cache so fs or something out of mainline will\nnotice it when compile time and can fix it.\n\nSigned-off-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nReviewed-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nReviewed-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "97cecb5a254fec22d28ef32235d888bfbfd7c783", "tree": "186c6bb9c1f999f807e2ec68dc6a0cf16c9d0d73", "parents": [ "ef6a3c63112e865d632ff7c478ba7c7160cad0d1" ], "author": { "name": "Minchan Kim", "email": "minchan.kim@gmail.com", "time": "Tue Mar 22 16:30:53 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:02 2011 -0700" }, "message": "mm: introduce delete_from_page_cache()\n\nPresently we increase the page refcount in add_to_page_cache() but don\u0027t\ndecrease it in remove_from_page_cache(). Such asymmetry adds confusion,\nrequiring that callers notice it and a comment explaining why they release\na page reference. It\u0027s not a good API.\n\nA long time ago, Hugh tried it (http://lkml.org/lkml/2004/10/24/140) but\ngave up because reiser4\u0027s drop_page() had to unlock the page between\nremoving it from page cache and doing the page_cache_release(). But now\nthe situation is changed. I think at least things in current mainline\ndon\u0027t have any obstacles. The problem is for out-of-mainline filesystems\n- if they have done such things as reiser4, this patch could be a problem\nbut they will discover this at compile time since we remove\nremove_from_page_cache().\n\nThis patch:\n\nThis function works as just wrapper remove_from_page_cache(). The\ndifference is that it decreases page references in itself. So caller have\nto make sure it has a page reference before calling.\n\nThis patch is ready for removing remove_from_page_cache().\n\nSigned-off-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nReviewed-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nReviewed-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Edward Shishkin \u003cedward.shishkin@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ef6a3c63112e865d632ff7c478ba7c7160cad0d1", "tree": "d0bd3ee2b79674e22b8dd3f318814cd4789697b8", "parents": [ "318b275fbca1ab9ec0862de71420e0e92c3d1aa7" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Tue Mar 22 16:30:52 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:02 2011 -0700" }, "message": "mm: add replace_page_cache_page() function\n\nThis function basically does:\n\n remove_from_page_cache(old);\n page_cache_release(old);\n add_to_page_cache_locked(new);\n\nExcept it does this atomically, so there\u0027s no possibility for the \"add\" to\nfail because of a race.\n\nIf memory cgroups are enabled, then the memory cgroup charge is also moved\nfrom the old page to the new.\n\nThis function is currently used by fuse to move pages into the page cache\non read, instead of copying the page contents.\n\n[minchan.kim@gmail.com: add freepage() hook to replace_page_cache_page()]\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nSigned-off-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "318b275fbca1ab9ec0862de71420e0e92c3d1aa7", "tree": "aa4984469443ed53b4e7fa23d3f91966e536a803", "parents": [ "5fda1bd5b8869574dad8e1f9f71e23bf0c186274" ], "author": { "name": "Gleb Natapov", "email": "gleb@redhat.com", "time": "Tue Mar 22 16:30:51 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 22 17:44:02 2011 -0700" }, "message": "mm: allow GUP to fail instead of waiting on a page\n\nGUP user may want to try to acquire a reference to a page if it is already\nin memory, but not if IO, to bring it in, is needed. For example KVM may\ntell vcpu to schedule another guest process if current one is trying to\naccess swapped out page. Meanwhile, the page will be swapped in and the\nguest process, that depends on it, will be able to run again.\n\nThis patch adds FAULT_FLAG_RETRY_NOWAIT (suggested by Linus) and\nFOLL_NOWAIT follow_page flags. FAULT_FLAG_RETRY_NOWAIT, when used in\nconjunction with VM_FAULT_ALLOW_RETRY, indicates to handle_mm_fault that\nit shouldn\u0027t drop mmap_sem and wait on a page, but return VM_FAULT_RETRY\ninstead.\n\n[akpm@linux-foundation.org: improve FOLL_NOWAIT comment]\nSigned-off-by: Gleb Natapov \u003cgleb@redhat.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: Michel Lespinasse \u003cwalken@google.com\u003e\nCc: Avi Kivity \u003cavi@redhat.com\u003e\nCc: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "55602dd66f5353981b53f734e26c307f310ced37", "tree": "c1478f1f0fd8f8e08ce5dceb352fa6b18a534745", "parents": [ "7eaceaccab5f40bbfda044629a6298616aeaed50" ], "author": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Jun 24 15:05:37 2010 +0200" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:26 2011 +0100" }, "message": "fs: make generic file read/write functions plug\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "7eaceaccab5f40bbfda044629a6298616aeaed50", "tree": "33954d12f63e25a47eb6d86ef3d3d0a5e62bf752", "parents": [ "73c101011926c5832e6e141682180c4debe2cf45" ], "author": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:07 2011 +0100" }, "committer": { "name": "Jens Axboe", "email": "jaxboe@fusionio.com", "time": "Thu Mar 10 08:52:07 2011 +0100" }, "message": "block: remove per-queue plugging\n\nCode has been converted over to the new explicit on-stack plugging,\nand delay users have been converted to use the new API for that.\nSo lets kill off the old plugging along with aops-\u003esync_page().\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n" }, { "commit": "c585a2678d83ba8fb02fa6b197de0ac7d67377f1", "tree": "b722f3c6fa28d1228e34adbcb85fee8683b34069", "parents": [ "e20e87795834f2f14cb53baf657b91d9c39f92c8" ], "author": { "name": "Steven Rostedt", "email": "srostedt@redhat.com", "time": "Thu Jan 13 15:46:18 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 17:32:36 2011 -0800" }, "message": "mm: remove likely() from grab_cache_page_write_begin()\n\nRunning the annotated branch profiler on a box doing average work\n(firefox, evolution, xchat, distcc farm), the likely() used in\ngrab_cache_page_write_begin() was incorrect most of the time:\n\n correct incorrect % Function File Line\n ------- --------- - -------- ---- ----\n 1924262 71332401 97 grab_cache_page_write_begin filemap.c 2206\n\nAdding a trace_printk() and running the function tracer limited to\njust this function I can see:\n\n gconfd-2-2696 [000] 4467.268935: grab_cache_page_write_begin: page\u003d (null) mapping\u003dffff8800676a9460 index\u003d7\n gconfd-2-2696 [000] 4467.268946: grab_cache_page_write_begin \u003c-ext3_write_begin\n gconfd-2-2696 [000] 4467.268947: grab_cache_page_write_begin: page\u003d (null) mapping\u003dffff8800676a9460 index\u003d8\n gconfd-2-2696 [000] 4467.268959: grab_cache_page_write_begin \u003c-ext3_write_begin\n gconfd-2-2696 [000] 4467.268960: grab_cache_page_write_begin: page\u003d (null) mapping\u003dffff8800676a9460 index\u003d9\n gconfd-2-2696 [000] 4467.268972: grab_cache_page_write_begin \u003c-ext3_write_begin\n gconfd-2-2696 [000] 4467.268973: grab_cache_page_write_begin: page\u003d (null) mapping\u003dffff8800676a9460 index\u003d10\n gconfd-2-2696 [000] 4467.268991: grab_cache_page_write_begin \u003c-ext3_write_begin\n gconfd-2-2696 [000] 4467.268992: grab_cache_page_write_begin: page\u003d (null) mapping\u003dffff8800676a9460 index\u003d11\n gconfd-2-2696 [000] 4467.269005: grab_cache_page_write_begin \u003c-ext3_write_begin\n\nWhich shows that a lot of calls from ext3_write_begin will result in the\npage returned by \"find_lock_page\" will be NULL.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\nAcked-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "212260aa07135b327752dc02625c68cf4ce04caf", "tree": "7ea25dddf5f69e91ec4b3d22f940e6c9604c7f93", "parents": [ "dabb16f639820267b3850d804571c70bd93d4e07" ], "author": { "name": "Rik van Riel", "email": "riel@redhat.com", "time": "Thu Jan 13 15:46:06 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 17:32:35 2011 -0800" }, "message": "mm: clear PageError bit in msync \u0026 fsync\n\nTemporary IO failures, eg. due to loss of both multipath paths, can\npermanently leave the PageError bit set on a page, resulting in msync or\nfsync returning -EIO over and over again, even if IO is now getting to the\ndisk correctly.\n\nWe already clear the AS_ENOSPC and AS_IO bits in mapping-\u003eflags in the\nfilemap_fdatawait_range function. Also clearing the PageError bit on the\npage allows subsequent msync or fsync calls on this file to return without\nan error, if the subsequent IO succeeds.\n\nUnfortunately data written out in the msync or fsync call that returned\n-EIO can still get lost, because the page dirty bit appears to not get\nrestored on IO error. However, the alternative could be potentially all\nof memory filling up with uncleanable dirty pages, hanging the system, so\nthere is no nice choice here...\n\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Valerie Aurora \u003cvaurora@redhat.com\u003e\nAcked-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nCc: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nAcked-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9cbb4cb21b19fff46cf1174d0ed699ef710e641c", "tree": "56480b1f4082d6158f077378db36c99d6d3ff209", "parents": [ "f0bc0a60b13f209df16062f94e9fb4b90dc08708" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Thu Jan 13 15:45:51 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 17:32:32 2011 -0800" }, "message": "mm: find_get_pages_contig fixlet\n\nTesting -\u003emapping and -\u003eindex without a ref is not stable as the page\nmay have been reused at this point.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nReviewed-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b5c84bf6f6fa3a7dfdcb556023a62953574b60ee", "tree": "7a2c299a180713e21d5cb653cb933121adf53c31", "parents": [ "949854d02455080d20cd3e1db28a3a18daf7599d" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:49:38 2011 +1100" }, "committer": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:50:23 2011 +1100" }, "message": "fs: dcache remove dcache_lock\n\ndcache_lock no longer protects anything. remove it.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "6072d13c429373c5d63b69dadbbef40a9b035552", "tree": "a2bf745efaa4092f2a8d7d9a9b160c2a7a3b303f", "parents": [ "0aded708d125a3ff7e5abaea9c2d9c6d7ebbfdcd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Dec 01 13:35:19 2010 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Dec 02 09:55:21 2010 -0500" }, "message": "Call the filesystem back whenever a page is removed from the page cache\n\nNFS needs to be able to release objects that are stored in the page\ncache once the page itself is no longer visible from the page cache.\n\nThis patch adds a callback to the address space operations that allows\nfilesystems to perform page cleanups once the page has been removed\nfrom the page cache.\n\nOriginal patch by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n[trondmy: cover the cases of invalidate_inode_pages2() and\n truncate_inode_pages()]\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "27d20fddc8af539464fc3ba499d6a830054c3bd6", "tree": "23514cfe88f90150a8635c47586a8a378fb905e3", "parents": [ "eaf06b241b091357e72b76863ba16e89610d31bd" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Thu Nov 11 14:05:19 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 12 07:55:32 2010 -0800" }, "message": "radix-tree: fix RCU bug\n\nSalman Qazi describes the following radix-tree bug:\n\nIn the following case, we get can get a deadlock:\n\n0. The radix tree contains two items, one has the index 0.\n1. The reader (in this case find_get_pages) takes the rcu_read_lock.\n2. The reader acquires slot(s) for item(s) including the index 0 item.\n3. The non-zero index item is deleted, and as a consequence the other item is\n moved to the root of the tree. The place where it used to be is queued for\n deletion after the readers finish.\n3b. The zero item is deleted, removing it from the direct slot, it remains in\n the rcu-delayed indirect node.\n4. The reader looks at the index 0 slot, and finds that the page has 0 ref\n count\n5. The reader looks at it again, hoping that the item will either be freed or\n the ref count will increase. This never happens, as the slot it is looking\n at will never be updated. Also, this slot can never be reclaimed because\n the reader is holding rcu_read_lock and is in an infinite loop.\n\nThe fix is to re-use the same \"indirect\" pointer case that requires a slot\nlookup retry into a general \"retry the lookup\" bit.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\nReported-by: Salman Qazi \u003csqazi@google.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8d056cb965b8fb7c53c564abf28b1962d1061cd3", "tree": "b422c16d0fe4f647e5c1fec990d610ec28c06a44", "parents": [ "834b40380e93e36f1c9b48ec1d280cebe3d7bd8c" ], "author": { "name": "Dave Hansen", "email": "dave@linux.vnet.ibm.com", "time": "Thu Nov 11 14:05:15 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 12 07:55:31 2010 -0800" }, "message": "mm/vfs: revalidate page-\u003emapping in do_generic_file_read()\n\n70 hours into some stress tests of a 2.6.32-based enterprise kernel, we\nran into a NULL dereference in here:\n\n\tint block_is_partially_uptodate(struct page *page, read_descriptor_t *desc,\n\t unsigned long from)\n\t{\n----\u003e\t\tstruct inode *inode \u003d page-\u003emapping-\u003ehost;\n\nIt looks like page-\u003emapping was the culprit. (xmon trace is below).\nAfter closer examination, I realized that do_generic_file_read() does a\nfind_get_page(), and eventually locks the page before calling\nblock_is_partially_uptodate(). However, it doesn\u0027t revalidate the\npage-\u003emapping after the page is locked. So, there\u0027s a small window\nbetween the find_get_page() and -\u003eis_partially_uptodate() where the page\ncould get truncated and page-\u003emapping cleared.\n\nWe _have_ a reference, so it can\u0027t get reclaimed, but it certainly\ncan be truncated.\n\nI think the correct thing is to check page-\u003emapping after the\ntrylock_page(), and jump out if it got truncated. This patch has been\nrunning in the test environment for a month or so now, and we have not\nseen this bug pop up again.\n\nxmon info:\n\n 1f:mon\u003e e\n cpu 0x1f: Vector: 300 (Data Access) at [c0000002ae36f770]\n pc: c0000000001e7a6c: .block_is_partially_uptodate+0xc/0x100\n lr: c000000000142944: .generic_file_aio_read+0x1e4/0x770\n sp: c0000002ae36f9f0\n msr: 8000000000009032\n dar: 0\n dsisr: 40000000\n current \u003d 0xc000000378f99e30\n paca \u003d 0xc000000000f66300\n pid \u003d 21946, comm \u003d bash\n 1f:mon\u003e r\n R00 \u003d 0025c0500000006d R16 \u003d 0000000000000000\n R01 \u003d c0000002ae36f9f0 R17 \u003d c000000362cd3af0\n R02 \u003d c000000000e8cd80 R18 \u003d ffffffffffffffff\n R03 \u003d c0000000031d0f88 R19 \u003d 0000000000000001\n R04 \u003d c0000002ae36fa68 R20 \u003d c0000003bb97b8a0\n R05 \u003d 0000000000000000 R21 \u003d c0000002ae36fa68\n R06 \u003d 0000000000000000 R22 \u003d 0000000000000000\n R07 \u003d 0000000000000001 R23 \u003d c0000002ae36fbb0\n R08 \u003d 0000000000000002 R24 \u003d 0000000000000000\n R09 \u003d 0000000000000000 R25 \u003d c000000362cd3a80\n R10 \u003d 0000000000000000 R26 \u003d 0000000000000002\n R11 \u003d c0000000001e7b60 R27 \u003d 0000000000000000\n R12 \u003d 0000000042000484 R28 \u003d 0000000000000001\n R13 \u003d c000000000f66300 R29 \u003d c0000003bb97b9b8\n R14 \u003d 0000000000000001 R30 \u003d c000000000e28a08\n R15 \u003d 000000000000ffff R31 \u003d c0000000031d0f88\n pc \u003d c0000000001e7a6c .block_is_partially_uptodate+0xc/0x100\n lr \u003d c000000000142944 .generic_file_aio_read+0x1e4/0x770\n msr \u003d 8000000000009032 cr \u003d 22000488\n ctr \u003d c0000000001e7a60 xer \u003d 0000000020000000 trap \u003d 300\n dar \u003d 0000000000000000 dsisr \u003d 40000000\n 1f:mon\u003e t\n [link register ] c000000000142944 .generic_file_aio_read+0x1e4/0x770\n [c0000002ae36f9f0] c000000000142a14 .generic_file_aio_read+0x2b4/0x770 (unreliable)\n [c0000002ae36fb40] c0000000001b03e4 .do_sync_read+0xd4/0x160\n [c0000002ae36fce0] c0000000001b153c .vfs_read+0xec/0x1f0\n [c0000002ae36fd80] c0000000001b1768 .SyS_read+0x58/0xb0\n [c0000002ae36fe30] c00000000000852c syscall_exit+0x0/0x40\n --- Exception: c00 (System Call) at 00000080a840bc54\n SP (fffca15df30) is in userspace\n 1f:mon\u003e di c0000000001e7a6c\n c0000000001e7a6c e9290000 ld r9,0(r9)\n c0000000001e7a70 418200c0 beq c0000000001e7b30 # .block_is_partially_uptodate+0xd0/0x100\n c0000000001e7a74 e9440008 ld r10,8(r4)\n c0000000001e7a78 78a80020 clrldi r8,r5,32\n c0000000001e7a7c 3c000001 lis r0,1\n c0000000001e7a80 812900a8 lwz r9,168(r9)\n c0000000001e7a84 39600001 li r11,1\n c0000000001e7a88 7c080050 subf r0,r8,r0\n c0000000001e7a8c 7f805040 cmplw cr7,r0,r10\n c0000000001e7a90 7d6b4830 slw r11,r11,r9\n c0000000001e7a94 796b0020 clrldi r11,r11,32\n c0000000001e7a98 419d00a8 bgt cr7,c0000000001e7b40 # .block_is_partially_uptodate+0xe0/0x100\n c0000000001e7a9c 7fa55840 cmpld cr7,r5,r11\n c0000000001e7aa0 7d004214 add r8,r0,r8\n c0000000001e7aa4 79080020 clrldi r8,r8,32\n c0000000001e7aa8 419c0078 blt cr7,c0000000001e7b20 # .block_is_partially_uptodate+0xc0/0x100\n\nSigned-off-by: Dave Hansen \u003cdave@linux.vnet.ibm.com\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nReviewed-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nCc: \u003carunabal@in.ibm.com\u003e\nCc: \u003csbest@us.ibm.com\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d88c0922fa0e2c021a028b310a641126c6d4b7dc", "tree": "f6e74b391a526c845cbe44bbc5e6585256a4d097", "parents": [ "eb8abb927ae2fd1730e24ea94cd9527f3c086292" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Tue Nov 02 13:05:18 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 02 17:02:31 2010 -0400" }, "message": "Release page reference during page fault retry\n\nThis slipped by when unifying the filemap and swap versions of\nlock_page_or_retry()...\n\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0116651c85e671a693dd2f56e95dd651f746c973", "tree": "d09f274105bc756b5fbd09dcb55302c0191e7e74", "parents": [ "68da336a14e16c2de95e987f3200995b707d7038" ], "author": { "name": "Namhyung Kim", "email": "namhyung@gmail.com", "time": "Tue Oct 26 14:21:58 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 26 16:52:09 2010 -0700" }, "message": "mm: remove temporary variable on generic_file_direct_write()\n\n\u0027end\u0027 shadows earlier one and is not necessary at all. Remove it and use\n\u0027pos\u0027 instead. This removes following sparse warnings:\n\n mm/filemap.c:2180:24: warning: symbol \u0027end\u0027 shadows an earlier one\n mm/filemap.c:2132:25: originally declared here\n\nSigned-off-by: Namhyung Kim \u003cnamhyung@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d065bd810b6deb67d4897a14bfe21f8eb526ba99", "tree": "f58c59075732ec4ccba336278c9bdc7ff61bef94", "parents": [ "b522c94da5d9cbc73f708be5e530ebc3bbd4a031" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Tue Oct 26 14:21:57 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 26 16:52:09 2010 -0700" }, "message": "mm: retry page fault when blocking on disk transfer\n\nThis change reduces mmap_sem hold times that are caused by waiting for\ndisk transfers when accessing file mapped VMAs.\n\nIt introduces the VM_FAULT_ALLOW_RETRY flag, which indicates that the call\nsite wants mmap_sem to be released if blocking on a pending disk transfer.\nIn that case, filemap_fault() returns the VM_FAULT_RETRY status bit and\ndo_page_fault() will then re-acquire mmap_sem and retry the page fault.\n\nIt is expected that the retry will hit the same page which will now be\ncached, and thus it will complete with a low mmap_sem hold time.\n\nTests:\n\n- microbenchmark: thread A mmaps a large file and does random read accesses\n to the mmaped area - achieves about 55 iterations/s. Thread B does\n mmap/munmap in a loop at a separate location - achieves 55 iterations/s\n before, 15000 iterations/s after.\n\n- We are seeing related effects in some applications in house, which show\n significant performance regressions when running without this change.\n\n[akpm@linux-foundation.org: fix warning \u0026 crash]\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nReviewed-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Ying Han \u003cyinghan@google.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nAcked-by: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: \u003clinux-arch@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b522c94da5d9cbc73f708be5e530ebc3bbd4a031", "tree": "d29cc4887bca05fde5b86a367f02a3ea70043e11", "parents": [ "182fea8f48332de085c0ae936605cb72671db9f2" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Tue Oct 26 14:21:56 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Oct 26 16:52:09 2010 -0700" }, "message": "mm: filemap_fault: unique path for locking page\n\nIntroduce a single location where filemap_fault() locks the desired page.\nThere used to be two such places, depending if the initial find_get_page()\nwas successful or not.\n\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nReviewed-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Ying Han \u003cyinghan@google.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: \u003clinux-arch@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "4e60c86bd9e5a7110ed28874d0b6592186550ae8", "tree": "9fb60e9f49b44b293a0c0c7d9f40e1a354a22b5a", "parents": [ "627295e492638936e76f3d8fcb1e0a3367b88341" ], "author": { "name": "Andi Kleen", "email": "andi@firstfloor.org", "time": "Mon Aug 09 17:19:03 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Aug 09 20:44:58 2010 -0700" }, "message": "gcc-4.6: mm: fix unused but set warnings\n\nNo real bugs, just some dead code and some fixups.\n\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "003386fff3e02e51cea882e60f7d28290113964c", "tree": "253001a9a0fc609b757362708edc2dcaab9e2a14", "parents": [ "092405cdb66f060918160ae730640b449ed7b827", "51921cb746f56983db5a373ca68deb2b0d3ddf01" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 30 09:16:14 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 30 09:16:14 2010 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:\n mm: export generic_pipe_buf_*() to modules\n fuse: support splice() reading from fuse device\n fuse: allow splice to move pages\n mm: export remove_from_page_cache() to modules\n mm: export lru_cache_add_*() to modules\n fuse: support splice() writing to fuse device\n fuse: get page reference for readpages\n fuse: use get_user_pages_fast()\n fuse: remove unneeded variable\n" }, { "commit": "105a048a4f35f7a74c7cc20b36dd83658b6ec232", "tree": "043b1110cda0042ba35d8aae59382bb094d0af3f", "parents": [ "00b9b0af5887fed54e899e3b7f5c2ccf5e739def", "9aeead73782c4b8e2a91def36dbf95db28605c95" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 10:43:44 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 27 10:43:44 2010 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable: (27 commits)\n Btrfs: add more error checking to btrfs_dirty_inode\n Btrfs: allow unaligned DIO\n Btrfs: drop verbose enospc printk\n Btrfs: Fix block generation verification race\n Btrfs: fix preallocation and nodatacow checks in O_DIRECT\n Btrfs: avoid ENOSPC errors in btrfs_dirty_inode\n Btrfs: move O_DIRECT space reservation to btrfs_direct_IO\n Btrfs: rework O_DIRECT enospc handling\n Btrfs: use async helpers for DIO write checksumming\n Btrfs: don\u0027t walk around with task-\u003estate !\u003d TASK_RUNNING\n Btrfs: do aio_write instead of write\n Btrfs: add basic DIO read/write support\n direct-io: do not merge logically non-contiguous requests\n direct-io: add a hook for the fs to provide its own submit_bio function\n fs: allow short direct-io reads to be completed via buffered IO\n Btrfs: Metadata ENOSPC handling for balance\n Btrfs: Pre-allocate space for data relocation\n Btrfs: Metadata ENOSPC handling for tree log\n Btrfs: Metadata reservation for orphan inodes\n Btrfs: Introduce global metadata reservation\n ...\n" }, { "commit": "91803b499cca2fe558abad709ce83dc896b80950", "tree": "5665b06dcacf14c2e5e979ed1c7066633b259f80", "parents": [ "63a6440326e4cd01d6a663069208a0e68e9b833f" ], "author": { "name": "Jeff Moyer", "email": "jmoyer@redhat.com", "time": "Wed May 26 11:49:40 2010 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 26 10:20:27 2010 -0700" }, "message": "do_generic_file_read: clear page errors when issuing a fresh read of the page\n\nI/O errors can happen due to temporary failures, like multipath\nerrors or losing network contact with the iSCSI server. Because\nof that, the VM will retry readpage on the page.\n\nHowever, do_generic_file_read does not clear PG_error. This\ncauses the system to be unable to actually use the data in the\npage cache page, even if the subsequent readpage completes\nsuccessfully!\n\nThe function filemap_fault has had a ClearPageError before\nreadpage forever. This patch simply adds the same to\ndo_generic_file_read.\n\nSigned-off-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Larry Woodman \u003clwoodman@redhat.com\u003e\nCc: stable@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c0ff7453bb5c7c98e0885fb94279f2571946f280", "tree": "8bb2b169a5145f0496575dbd2f48bb4b1c83f819", "parents": [ "708c1bbc9d0c3e57f40501794d9b0eed29d10fce" ], "author": { "name": "Miao Xie", "email": "miaox@cn.fujitsu.com", "time": "Mon May 24 14:32:08 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 25 08:06:57 2010 -0700" }, "message": "cpuset,mm: fix no node to alloc memory when changing cpuset\u0027s mems\n\nBefore applying this patch, cpuset updates task-\u003emems_allowed and\nmempolicy by setting all new bits in the nodemask first, and clearing all\nold unallowed bits later. But in the way, the allocator may find that\nthere is no node to alloc memory.\n\nThe reason is that cpuset rebinds the task\u0027s mempolicy, it cleans the\nnodes which the allocater can alloc pages on, for example:\n\n(mpol: mempolicy)\n\ttask1\t\t\ttask1\u0027s mpol\ttask2\n\talloc page\t\t1\n\t alloc on node0? NO\t1\n\t\t\t\t1\t\tchange mems from 1 to 0\n\t\t\t\t1\t\trebind task1\u0027s mpol\n\t\t\t\t0-1\t\t set new bits\n\t\t\t\t0\t \t clear disallowed bits\n\t alloc on node1? NO\t0\n\t ...\n\tcan\u0027t alloc page\n\t goto oom\n\nThis patch fixes this problem by expanding the nodes range first(set newly\nallowed bits) and shrink it lazily(clear newly disallowed bits). So we\nuse a variable to tell the write-side task that read-side task is reading\nnodemask, and the write-side task clears newly disallowed nodes after\nread-side task ends the current memory allocation.\n\n[akpm@linux-foundation.org: fix spello]\nSigned-off-by: Miao Xie \u003cmiaox@cn.fujitsu.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: Hugh Dickins \u003chugh.dickins@tiscali.co.uk\u003e\nCc: Ravikiran Thirumalai \u003ckiran@scalex86.org\u003e\nCc: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e9d6c157385e4efa61cb8293e425c9d8beba70d3", "tree": "fca2452b46328c9005b8a4043a22b7b1b4d47d0c", "parents": [ "1f0a738868cbfe20ae53a00b7c302c04ef7ab8fc" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Mon May 24 14:31:48 2010 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 25 08:06:56 2010 -0700" }, "message": "tmpfs: insert tmpfs cache pages to inactive list at first\n\nShaohua Li reported parallel file copy on tmpfs can lead to OOM killer.\nThis is regression of caused by commit 9ff473b9a7 (\"vmscan: evict\nstreaming IO first\"). Wow, It is 2 years old patch!\n\nCurrently, tmpfs file cache is inserted active list at first. This means\nthat the insertion doesn\u0027t only increase numbers of pages in anon LRU, but\nit also reduces anon scanning ratio. Therefore, vmscan will get totally\nconfused. It scans almost only file LRU even though the system has plenty\nunused tmpfs pages.\n\nHistorically, lru_cache_add_active_anon() was used for two reasons.\n1) Intend to priotize shmem page rather than regular file cache.\n2) Intend to avoid reclaim priority inversion of used once pages.\n\nBut we\u0027ve lost both motivation because (1) Now we have separate anon and\nfile LRU list. then, to insert active list doesn\u0027t help such priotize.\n(2) In past, one pte access bit will cause page activation. then to\ninsert inactive list with pte access bit mean higher priority than to\ninsert active list. Its priority inversion may lead to uninteded lru\nchun. but it was already solved by commit 645747462 (vmscan: detect\nmapped file pages used only once). (Thanks Hannes, you are great!)\n\nThus, now we can use lru_cache_add_anon() instead.\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nReported-by: Shaohua Li \u003cshaohua.li@intel.com\u003e\nReviewed-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nReviewed-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nReviewed-by: Rik van Riel \u003criel@redhat.com\u003e\nReviewed-by: Minchan Kim \u003cminchan.kim@gmail.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Henrique de Moraes Holschuh \u003chmh@hmh.eng.br\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "66f998f611897319b555364cefd5d6e88a205866", "tree": "3d2a46624bf6cf1ea1645cc8dad975af858dc114", "parents": [ "3fd0a5585eb98e074fb9934549c8d85c49756c0d" ], "author": { "name": "Josef Bacik", "email": "josef@redhat.com", "time": "Sun May 23 11:00:54 2010 -0400" }, "committer": { "name": "Chris Mason", "email": "chris.mason@oracle.com", "time": "Tue May 25 10:34:55 2010 -0400" }, "message": "fs: allow short direct-io reads to be completed via buffered IO\n\nThis is similar to what already happens in the write case. If we have a short\nread while doing O_DIRECT, instead of just returning, fallthrough and try to\nread the rest via buffered IO. BTRFS needs this because if we encounter a\ncompressed or inline extent during DIO, we need to fallback on buffered. If the\nextent is compressed we need to read the entire thing into memory and\nde-compress it into the users pages. I have tested this with fsx and everything\nworks great. Thanks,\n\nSigned-off-by: Josef Bacik \u003cjosef@redhat.com\u003e\nSigned-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n" }, { "commit": "a52116aba5b3eed0ee41f70b794cc1937acd5cb8", "tree": "cedff5545f4cac8c2fd56fc0bd7a5e536cc677fd", "parents": [ "47846b0650f2f62fc4217cfb36efc94b8d919727" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Tue May 25 15:06:06 2010 +0200" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Tue May 25 15:06:06 2010 +0200" }, "message": "mm: export remove_from_page_cache() to modules\n\nThis is needed to enable moving pages into the page cache in fuse with\nsplice(..., SPLICE_F_MOVE).\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "5a0e3ad6af8660be21ca98a971cd00f331318c05", "tree": "5bfb7be11a03176a87296a43ac6647975c00a1d1", "parents": [ "ed391f4ebf8f701d3566423ce8f17e614cde9806" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Wed Mar 24 17:04:11 2010 +0900" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Tue Mar 30 22:02:32 2010 +0900" }, "message": "include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h\n\npercpu.h is included by sched.h and module.h and thus ends up being\nincluded when building most .c files. percpu.h includes slab.h which\nin turn includes gfp.h making everything defined by the two files\nuniversally available and complicating inclusion dependencies.\n\npercpu.h -\u003e slab.h dependency is about to be removed. Prepare for\nthis change by updating users of gfp and slab facilities include those\nheaders directly instead of assuming availability. As this conversion\nneeds to touch large number of source files, the following script is\nused as the basis of conversion.\n\n http://userweb.kernel.org/~tj/misc/slabh-sweep.py\n\nThe script does the followings.\n\n* Scan files for gfp and slab usages and update includes such that\n only the necessary includes are there. ie. if only gfp is used,\n gfp.h, if slab is used, slab.h.\n\n* When the script inserts a new include, it looks at the include\n blocks and try to put the new include such that its order conforms\n to its surrounding. It\u0027s put in the include block which contains\n core kernel includes, in the same order that the rest are ordered -\n alphabetical, Christmas tree, rev-Xmas-tree or at the end if there\n doesn\u0027t seem to be any matching order.\n\n* If the script can\u0027t find a place to put a new include (mostly\n because the file doesn\u0027t have fitting include block), it prints out\n an error message indicating which .h file needs to be added to the\n file.\n\nThe conversion was done in the following steps.\n\n1. The initial automatic conversion of all .c files updated slightly\n over 4000 files, deleting around 700 includes and adding ~480 gfp.h\n and ~3000 slab.h inclusions. The script emitted errors for ~400\n files.\n\n2. Each error was manually checked. Some didn\u0027t need the inclusion,\n some needed manual addition while adding it to implementation .h or\n embedding .c file was more appropriate for others. This step added\n inclusions to around 150 files.\n\n3. The script was run again and the output was compared to the edits\n from #2 to make sure no file was left behind.\n\n4. Several build tests were done and a couple of problems were fixed.\n e.g. lib/decompress_*.c used malloc/free() wrappers around slab\n APIs requiring slab.h to be added manually.\n\n5. The script was run on all .h files but without automatically\n editing them as sprinkling gfp.h and slab.h inclusions around .h\n files could easily lead to inclusion dependency hell. Most gfp.h\n inclusion directives were ignored as stuff from gfp.h was usually\n wildly available and often used in preprocessor macros. Each\n slab.h inclusion directive was examined and added manually as\n necessary.\n\n6. percpu.h was updated not to include slab.h.\n\n7. Build test were done on the following configurations and failures\n were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my\n distributed build env didn\u0027t work with gcov compiles) and a few\n more options had to be turned off depending on archs to make things\n build (like ipr on powerpc/64 which failed due to missing writeq).\n\n * x86 and x86_64 UP and SMP allmodconfig and a custom test config.\n * powerpc and powerpc64 SMP allmodconfig\n * sparc and sparc64 SMP allmodconfig\n * ia64 SMP allmodconfig\n * s390 SMP allmodconfig\n * alpha SMP allmodconfig\n * um on x86_64 SMP allmodconfig\n\n8. percpu.h modifications were reverted so that it could be applied as\n a separate patch and serve as bisection point.\n\nGiven the fact that I had only a couple of failures from tests on step\n6, I\u0027m fairly confident about the coverage of this conversion patch.\nIf there is a breakage, it\u0027s likely to be something in one of the arch\nheaders which should be easily discoverable easily on most builds of\nthe specific arch.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nGuess-its-ok-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\n" }, { "commit": "59e99e5b9706867f18d4a36c1e4645fbaacbec2e", "tree": "e977fb5eecccf1446296fd196072bd1287b0a92f", "parents": [ "06f9d8c2b50060543fb6e0af87ddb86e654dee6b" ], "author": { "name": "Jiri Slaby", "email": "jslaby@suse.cz", "time": "Fri Mar 05 13:41:44 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 06 11:26:24 2010 -0800" }, "message": "mm: use rlimit helpers\n\nMake sure compiler won\u0027t do weird things with limits. E.g. fetching them\ntwice may return 2 different values after writable limits are implemented.\n\nI.e. either use rlimit helpers added in\n3e10e716abf3c71bdb5d86b8f507f9e72236c9cd (\"resource: add helpers for\nfetching rlimits\") or ACCESS_ONCE if not applicable.\n\nSigned-off-by: Jiri Slaby \u003cjslaby@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "2ecdc82ef0b03e67ce5ecee79d0d108177a704df", "tree": "66887e3e0e7328c4030888905a713a995bb1654b", "parents": [ "270ba5f7c5dac0bfb564aa35a536fb31ad4075bd" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Tue Jan 26 17:27:20 2010 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Mar 03 14:07:55 2010 -0500" }, "message": "kill unused invalidate_inode_pages helper\n\nNo one is calling this anymore as everyone has switched to\ninvalidate_mapping_pages long time ago. Also update a few\nreferences to it in comments. nfs has two more, but I can\u0027t\neasily figure what they are actually referring to, so I left\nthem as-is.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "931e80e4b3263db75c8e34f078d22f11bbabd3a3", "tree": "47c735f039d2c8623a1fc4d20333a9899ac8e99c", "parents": [ "bc173f7092c76a7967f135c2b3a54052ad99733b" ], "author": { "name": "anfei zhou", "email": "anfei.zhou@gmail.com", "time": "Tue Feb 02 13:44:02 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 02 18:11:21 2010 -0800" }, "message": "mm: flush dcache before writing into page to avoid alias\n\nThe cache alias problem will happen if the changes of user shared mapping\nis not flushed before copying, then user and kernel mapping may be mapped\ninto two different cache line, it is impossible to guarantee the coherence\nafter iov_iter_copy_from_user_atomic. So the right steps should be:\n\n\tflush_dcache_page(page);\n\tkmap_atomic(page);\n\twrite to page;\n\tkunmap_atomic(page);\n\tflush_dcache_page(page);\n\nMore precisely, we might create two new APIs flush_dcache_user_page and\nflush_dcache_kern_page to replace the two flush_dcache_page accordingly.\n\nHere is a snippet tested on omap2430 with VIPT cache, and I think it is\nnot ARM-specific:\n\n\tint val \u003d 0x11111111;\n\tfd \u003d open(\"abc\", O_RDWR);\n\taddr \u003d mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);\n\t*(addr+0) \u003d 0x44444444;\n\ttmp \u003d *(addr+0);\n\t*(addr+1) \u003d 0x77777777;\n\twrite(fd, \u0026val, sizeof(int));\n\tclose(fd);\n\nThe results are not always 0x11111111 0x77777777 at the beginning as expected. Sometimes we see 0x44444444 0x77777777.\n\nSigned-off-by: Anfei \u003canfei.zhou@gmail.com\u003e\nCc: Russell King \u003crmk@arm.linux.org.uk\u003e\nCc: Miklos Szeredi \u003cmiklos@szeredi.hu\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nCc: \u003clinux-arch@vger.kernel.org\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0531b2aac59c2296570ac52bfc032ef2ace7d5e1", "tree": "4ca454bdc03c24654529bab9c882c1cd5f99a88c", "parents": [ "caf0801e0cc482497c14a9ce349469c33c60beec" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jan 27 09:20:03 2010 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jan 27 09:20:03 2010 -0800" }, "message": "mm: add new \u0027read_cache_page_gfp()\u0027 helper function\n\nIt\u0027s a simplified \u0027read_cache_page()\u0027 which takes a page allocation\nflag, so that different paths can control how aggressive the memory\nallocations are that populate a address space.\n\nIn particular, the intel GPU object mapping code wants to be able to do\na certain amount of own internal memory management by automatically\nshrinking the address space when memory starts getting tight. This\nallows it to dynamically use different memory allocation policies on a\nper-allocation basis, rather than depend on the (static) address space\ngfp policy.\n\nThe actual new function is a one-liner, but re-organizing the helper\nfunctions to the point where you can do this with a single line of code\nis what most of the patch is all about.\n\nTested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c05c4edd876b7ae92787d1295868afcb89b6a348", "tree": "8dc4ce17c0a9be223a70e99d246035fee0a7168e", "parents": [ "2cfd30adf6130dab3fbb130eb5f7b1fd42a70e31" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Sep 23 15:07:30 2009 +0200" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Dec 16 12:16:50 2009 -0500" }, "message": "direct I/O fallback sync simplification\n\nIn the case of direct I/O falling back to buffered I/O we sync data\ntwice currently: once at the end of generic_file_buffered_write using\nfilemap_write_and_wait_range and once a little later in\n__generic_file_aio_write using do_sync_mapping_range with all flags set.\n\nThe wait before write of the do_sync_mapping_range call does not make\nany sense, so just keep the filemap_write_and_wait_range call and move\nit to the right spot.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "94004ed726f38a841cc51f97c4a3f9eda9fbd0d9", "tree": "786689e9fc9e686a5b54c1a1b1efcbb7d043b5ab", "parents": [ "6b2f3d1f769be5779b479c37800229d9a4809fc3" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Sep 30 22:16:33 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Thu Dec 10 15:02:50 2009 +0100" }, "message": "kill wait_on_page_writeback_range\n\nAll callers really want the more logical filemap_fdatawait_range interface,\nso convert them to use it and merge wait_on_page_writeback_range into\nfilemap_fdatawait_range.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "af901ca181d92aac3a7dc265144a9081a86d8f39", "tree": "380054af22521144fbe1364c3bcd55ad24c9bde4", "parents": [ "972b94ffb90ea6d20c589d9a47215df103388ddd" ], "author": { "name": "André Goddard Rosa", "email": "andre.goddard@gmail.com", "time": "Sat Nov 14 13:09:05 2009 -0200" }, "committer": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Fri Dec 04 15:39:55 2009 +0100" }, "message": "tree-wide: fix assorted typos all over the place\n\nThat is \"success\", \"unknown\", \"through\", \"performance\", \"[re|un]mapping\"\n, \"access\", \"default\", \"reasonable\", \"[con]currently\", \"temperature\"\n, \"channel\", \"[un]used\", \"application\", \"example\",\"hierarchy\", \"therefore\"\n, \"[over|under]flow\", \"contiguous\", \"threshold\", \"enough\" and others.\n\nSigned-off-by: André Goddard Rosa \u003candre.goddard@gmail.com\u003e\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "f0f37e2f77731b3473fa6bd5ee53255d9a9cdb40", "tree": "3c26d3ed1a453156e9c208ccb5567a8954dba064", "parents": [ "6f5071020d5ec89b5d095aa488db604adb921aec" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Sun Sep 27 22:29:37 2009 +0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Sep 27 11:39:25 2009 -0700" }, "message": "const: mark struct vm_struct_operations\n\n* mark struct vm_area_struct::vm_ops as const\n* mark vm_ops in AGP code\n\nBut leave TTM code alone, something is fishy there with global vm_ops\nbeing used.\n\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6c5daf012c9155aafd2c7973e4278766c30dfad0", "tree": "33959d7b36d03e1610615641a2940cb2de5e8603", "parents": [ "6d39b27f0ac7e805ae3bd9efa51d7da04bec0360", "c08d3b0e33edce28e9cfa7b64f7fe5bdeeb29248" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 08:32:11 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 08:32:11 2009 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:\n truncate: use new helpers\n truncate: new helpers\n fs: fix overflow in sys_mount() for in-kernel calls\n fs: Make unload_nls() NULL pointer safe\n freeze_bdev: grab active reference to frozen superblocks\n freeze_bdev: kill bd_mount_sem\n exofs: remove BKL from super operations\n fs/romfs: correct error-handling code\n vfs: seq_file: add helpers for data filling\n vfs: remove redundant position check in do_sendfile\n vfs: change sb-\u003es_maxbytes to a loff_t\n vfs: explicitly cast s_maxbytes in fiemap_check_ranges\n libfs: return error code on failed attr set\n seq_file: return a negative error code when seq_path_root() fails.\n vfs: optimize touch_time() too\n vfs: optimization for touch_atime()\n vfs: split generic_forget_inode() so that hugetlbfs does not have to copy it\n fs/inode.c: add dev-id and inode number for debugging in init_special_inode()\n libfs: make simple_read_from_buffer conventional\n" }, { "commit": "db16826367fefcb0ddb93d76b66adc52eb4e6339", "tree": "626224c1eb1eb79c522714591f208b4fdbdcd9d4", "parents": [ "cd6045138ed1bb5d8773e940d51c34318eef3ef2", "465fdd97cbe16ef8727221857e96ef62dd352017" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:53:22 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Sep 24 07:53:22 2009 -0700" }, "message": "Merge branch \u0027hwpoison\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ak/linux-mce-2.6\n\n* \u0027hwpoison\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ak/linux-mce-2.6: (21 commits)\n HWPOISON: Enable error_remove_page on btrfs\n HWPOISON: Add simple debugfs interface to inject hwpoison on arbitary PFNs\n HWPOISON: Add madvise() based injector for hardware poisoned pages v4\n HWPOISON: Enable error_remove_page for NFS\n HWPOISON: Enable .remove_error_page for migration aware file systems\n HWPOISON: The high level memory error handler in the VM v7\n HWPOISON: Add PR_MCE_KILL prctl to control early kill behaviour per process\n HWPOISON: shmem: call set_page_dirty() with locked page\n HWPOISON: Define a new error_remove_page address space op for async truncation\n HWPOISON: Add invalidate_inode_page\n HWPOISON: Refactor truncate to allow direct truncating of page v2\n HWPOISON: check and isolate corrupted free pages v2\n HWPOISON: Handle hardware poisoned pages in try_to_unmap\n HWPOISON: Use bitmask/action code for try_to_unmap behaviour\n HWPOISON: x86: Add VM_FAULT_HWPOISON handling to x86 page fault handler v2\n HWPOISON: Add poison check to page fault handling\n HWPOISON: Add basic support for poisoned pages in fault handler v3\n HWPOISON: Add new SIGBUS error codes for hardware poison signals\n HWPOISON: Add support for poison swap entries v2\n HWPOISON: Export some rmap vma locking to outside world\n ...\n" }, { "commit": "25d9e2d15286281ec834b829a4aaf8969011f1cd", "tree": "e4329a481ca197afae30f04335e023c7d04f7d67", "parents": [ "eca6f534e61919b28fb21aafbd1c2983deae75be" ], "author": { "name": "npiggin@suse.de", "email": "npiggin@suse.de", "time": "Fri Aug 21 02:35:05 2009 +1000" }, "committer": { "name": "al", "email": "al@dizzy.pdmi.ras.ru", "time": "Thu Sep 24 08:41:47 2009 -0400" }, "message": "truncate: new helpers\n\nIntroduce new truncate helpers truncate_pagecache and inode_newsize_ok.\nvmtruncate is also consolidated from mm/memory.c and mm/nommu.c and\ninto mm/truncate.c.\n\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "4b02108ac1b3354a22b0d83c684797692efdc395", "tree": "9f65d6e8e35ddce940e7b9da6305cf5a19e5904e", "parents": [ "c6a7f5728a1db45d30df55a01adc130b4ab0327c" ], "author": { "name": "KOSAKI Motohiro", "email": "kosaki.motohiro@jp.fujitsu.com", "time": "Mon Sep 21 17:01:33 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Sep 22 07:17:27 2009 -0700" }, "message": "mm: oom analysis: add shmem vmstat\n\nRecently we encountered OOM problems due to memory use of the GEM cache.\nGenerally a large amuont of Shmem/Tmpfs pages tend to create a memory\nshortage problem.\n\nWe often use the following calculation to determine the amount of shmem\npages:\n\nshmem \u003d NR_ACTIVE_ANON + NR_INACTIVE_ANON - NR_ANON_PAGES\n\nhowever the expression does not consider isolated and mlocked pages.\n\nThis patch adds explicit accounting for pages used by shmem and tmpfs.\n\nSigned-off-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nReviewed-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nAcked-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Hugh Dickins \u003chugh.dickins@tiscali.co.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6a46079cf57a7f7758e8b926980a4f852f89b34d", "tree": "efd72e830201370d6273bd436dda5a3c4cd6ed9b", "parents": [ "4db96cf077aa938b11fe7ac79ecc9b29ec00fbab" ], "author": { "name": "Andi Kleen", "email": "andi@firstfloor.org", "time": "Wed Sep 16 11:50:15 2009 +0200" }, "committer": { "name": "Andi Kleen", "email": "ak@linux.intel.com", "time": "Wed Sep 16 11:50:15 2009 +0200" }, "message": "HWPOISON: The high level memory error handler in the VM v7\n\nAdd the high level memory handler that poisons pages\nthat got corrupted by hardware (typically by a two bit flip in a DIMM\nor a cache) on the Linux level. The goal is to prevent everyone\nfrom accessing these pages in the future.\n\nThis done at the VM level by marking a page hwpoisoned\nand doing the appropriate action based on the type of page\nit is.\n\nThe code that does this is portable and lives in mm/memory-failure.c\n\nTo quote the overview comment:\n\nHigh level machine check handler. Handles pages reported by the\nhardware as being corrupted usually due to a 2bit ECC memory or cache\nfailure.\n\nThis focuses on pages detected as corrupted in the background.\nWhen the current CPU tries to consume corruption the currently\nrunning process can just be killed directly instead. This implies\nthat if the error cannot be handled for some reason it\u0027s safe to\njust ignore it because no corruption has been consumed yet. Instead\nwhen that happens another machine check will happen.\n\nHandles page cache pages in various states. The tricky part\nhere is that we can access any page asynchronous to other VM\nusers, because memory failures could happen anytime and anywhere,\npossibly violating some of their assumptions. This is why this code\nhas to be extremely careful. Generally it tries to use normal locking\nrules, as in get the standard locks, even if that means the\nerror handling takes potentially a long time.\n\nSome of the operations here are somewhat inefficient and have non\nlinear algorithmic complexity, because the data structures have not\nbeen optimized for this case. This is in particular the case\nfor the mapping from a vma to a process. Since this case is expected\nto be rare we hope we can get away with this.\n\nThere are in principle two strategies to kill processes on poison:\n- just unmap the data and wait for an actual reference before\nkilling\n- kill as soon as corruption is detected.\nBoth have advantages and disadvantages and should be used\nin different situations. Right now both are implemented and can\nbe switched with a new sysctl vm.memory_failure_early_kill\nThe default is early kill.\n\nThe patch does some rmap data structure walking on its own to collect\nprocesses to kill. This is unusual because normally all rmap data structure\nknowledge is in rmap.c only. I put it here for now to keep\neverything together and rmap knowledge has been seeping out anyways\n\nIncludes contributions from Johannes Weiner, Chris Mason, Fengguang Wu,\nNick Piggin (who did a lot of great work) and others.\n\nCc: npiggin@suse.de\nCc: riel@redhat.com\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nReviewed-by: Hidehiro Kawai \u003chidehiro.kawai.ez@hitachi.com\u003e\n" }, { "commit": "18f2ee705d98034b0f229a3202d827468d4bffd9", "tree": "1da282e5eb90dc8e8a1e46214e0e7639138d3568", "parents": [ "2f3d675bcd4a84251d6e8eea8096ec8fc795e5d6" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Tue Aug 18 18:43:15 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Sep 14 17:08:17 2009 +0200" }, "message": "vfs: Remove generic_osync_inode() and sync_page_range{_nolock}()\n\nRemove these three functions since nobody uses them anymore.\n\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "148f948ba877f4d3cdef036b1ff6d9f68986706a", "tree": "c07963f08bf8c2119ec00df64e4293e2e60acaa1", "parents": [ "eef99380679e20e7edc096aa4d8a98b875404d79" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Aug 17 19:52:36 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Sep 14 17:08:15 2009 +0200" }, "message": "vfs: Introduce new helpers for syncing after writing to O_SYNC file or IS_SYNC inode\n\nIntroduce new function for generic inode syncing (vfs_fsync_range) and use\nit from fsync() path. Introduce also new helper for syncing after a sync\nwrite (generic_write_sync) using the generic function.\n\nUse these new helpers for syncing from generic VFS functions. This makes\nO_SYNC writes to block devices acquire i_mutex for syncing. If we really\ncare about this, we can make block_fsync() drop the i_mutex and reacquire\nit before it returns.\n\nCC: Evgeniy Polyakov \u003czbr@ioremap.net\u003e\nCC: ocfs2-devel@oss.oracle.com\nCC: Joel Becker \u003cjoel.becker@oracle.com\u003e\nCC: Felix Blyakher \u003cfelixb@sgi.com\u003e\nCC: xfs@oss.sgi.com\nCC: Anton Altaparmakov \u003caia21@cantab.net\u003e\nCC: linux-ntfs-dev@lists.sourceforge.net\nCC: OGAWA Hirofumi \u003chirofumi@mail.parknet.co.jp\u003e\nCC: linux-ext4@vger.kernel.org\nCC: tytso@mit.edu\nAcked-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "eef99380679e20e7edc096aa4d8a98b875404d79", "tree": "358a39148e8513eed9ba6aaff13f6bf660a2ce1e", "parents": [ "918941a3f3d46c2a69971b4718aaf13b1be2f1a7" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Thu Aug 20 17:43:41 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Sep 14 17:08:15 2009 +0200" }, "message": "vfs: Rename generic_file_aio_write_nolock\n\ngeneric_file_aio_write_nolock() is now used only by block devices and raw\ncharacter device. Filesystems should use __generic_file_aio_write() in case\ngeneric_file_aio_write() doesn\u0027t suit them. So rename the function to\nblkdev_aio_write() and move it to fs/blockdev.c.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "c7b50db21fe8c295092518e224d60b95e69da3b0", "tree": "ed73757b475462c4cfccdcf61d95929f4cb32637", "parents": [ "e4dd9de3c66bc7e26c5c7f149a060c5a67cf06a0" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Tue Aug 18 16:18:20 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Sep 14 17:08:15 2009 +0200" }, "message": "vfs: Remove syncing from generic_file_direct_write() and generic_file_buffered_write()\n\ngeneric_file_direct_write() and generic_file_buffered_write() called\ngeneric_osync_inode() if it was called on O_SYNC file or IS_SYNC inode. But\nthis is superfluous since generic_file_aio_write() does the syncing as well.\nAlso XFS and OCFS2 which call these functions directly handle syncing\nthemselves. So let\u0027s have a single place where syncing happens:\ngeneric_file_aio_write().\n\nWe slightly change the behavior by syncing only the range of file to which the\nwrite happened for buffered writes but that should be all that is required.\n\nCC: ocfs2-devel@oss.oracle.com\nCC: Joel Becker \u003cjoel.becker@oracle.com\u003e\nCC: Felix Blyakher \u003cfelixb@sgi.com\u003e\nCC: xfs@oss.sgi.com\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "e4dd9de3c66bc7e26c5c7f149a060c5a67cf06a0", "tree": "3270180b085c86afe4834fba5567464cd3932eb8", "parents": [ "d3bccb6f4b886060aa0f58976b92b77d951f5434" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Aug 17 18:10:06 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Sep 14 17:08:14 2009 +0200" }, "message": "vfs: Export __generic_file_aio_write() and add some comments\n\nRename __generic_file_aio_write_nolock() to __generic_file_aio_write(), add\ncomments to write helpers explaining how they should be used and export\n__generic_file_aio_write() since it will be used by some filesystems.\n\nCC: ocfs2-devel@oss.oracle.com\nCC: Joel Becker \u003cjoel.becker@oracle.com\u003e\nAcked-by: Evgeniy Polyakov \u003czbr@ioremap.net\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "d3bccb6f4b886060aa0f58976b92b77d951f5434", "tree": "b5f9eb43890da173a078e883d51c23b8c27ceaa6", "parents": [ "86d710146fb9975f04c505ec78caa43d227c1018" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Aug 17 19:30:27 2009 +0200" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Mon Sep 14 17:08:14 2009 +0200" }, "message": "vfs: Introduce filemap_fdatawait_range\n\nThis simple helper saves some filesystems conversion from byte offset\nto page numbers and also makes the fdata* interface more complete.\n\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "c8236db9cd7aa492dcfcdcca702638e704abed49", "tree": "75eb83051298fd1f9df426cafaaa6d3c422d1b68", "parents": [ "a65e7bfcd74e4c0939f235d2bf9f48ddb3a57991" ], "author": { "name": "Josef Bacik", "email": "josef@redhat.com", "time": "Sun Jul 05 12:08:18 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 06 13:57:03 2009 -0700" }, "message": "mm: mark page accessed before we write_end()\n\nIn testing a backport of the write_begin/write_end AOPs, a 10% re-read\nregression was noticed when running iozone. This regression was\nintroduced because the old AOPs would always do a mark_page_accessed(page)\nafter the commit_write, but when the new AOPs where introduced, the only\nplace this was kept was in pagecache_write_end().\n\nThis patch does the same thing in the generic case as what is done in\npagecache_write_end(), which is just to mark the page accessed before we\ndo write_end().\n\nSigned-off-by: Josef Bacik \u003cjbacik@redhat.com\u003e\nAcked-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6484eb3e2a81807722c5f28efef94d8338b7b996", "tree": "10ce36f412c2ff0c7eb399af1a189f8e354f56db", "parents": [ "b3c466ce512923298ae8c0121d3e9f397a3f1210" ], "author": { "name": "Mel Gorman", "email": "mel@csn.ul.ie", "time": "Tue Jun 16 15:31:54 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jun 16 19:47:32 2009 -0700" }, "message": "page allocator: do not check NUMA node ID when the caller knows the node is valid\n\nCallers of alloc_pages_node() can optionally specify -1 as a node to mean\n\"allocate from the current node\". However, a number of the callers in\nfast paths know for a fact their node is valid. To avoid a comparison and\nbranch, this patch adds alloc_pages_exact_node() that only checks the nid\nwith VM_BUG_ON(). Callers that know their node is valid are then\nconverted.\n\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nReviewed-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nReviewed-by: KOSAKI Motohiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nReviewed-by: Pekka Enberg \u003cpenberg@cs.helsinki.fi\u003e\nAcked-by: Paul Mundt \u003clethal@linux-sh.org\u003e\t[for the SLOB NUMA bits]\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nCc: Dave Hansen \u003cdave@linux.vnet.ibm.com\u003e\nCc: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7ffc59b4d0bdfa00e882339f85b8a969bb7021e2", "tree": "6b6d96208f08bc394c8e64efed6767b9a95e7a6d", "parents": [ "61b7cbdba2f3c588a0cf3db574c562805454b09b" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue Jun 16 15:31:38 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jun 16 19:47:30 2009 -0700" }, "message": "readahead: enforce full sync mmap readahead size\n\nNow that we do readahead for sequential mmap reads, here is a simple\nevaluation of the impacts, and one further optimization.\n\nIt\u0027s an NFS-root debian desktop system, readahead size \u003d 60 pages.\nThe numbers are grabbed after a fresh boot into console.\n\napproach pgmajfault RA miss ratio mmap IO count avg IO size(pages)\n A 383 31.6% 383 11\n B 225 32.4% 390 11\n C 224 32.6% 307 13\n\ncase A: mmap sync/async readahead disabled\ncase B: mmap sync/async readahead enabled, with enforced full async readahead size\ncase C: mmap sync/async readahead enabled, with enforced full sync/async readahead size\nor:\nA \u003d vanilla 2.6.30-rc1\nB \u003d A plus mmap readahead\nC \u003d B plus this patch\n\nThe numbers show that\n- there are good possibilities for random mmap reads to trigger readahead\n- \u0027pgmajfault\u0027 is reduced by 1/3, due to the _async_ nature of readahead\n- case C can further reduce IO count by 1/4\n- readahead miss ratios are not quite affected\n\nThe theory is\n- readahead is _good_ for clustered random reads, and can perform\n _better_ than readaround because they could be _async_.\n- async readahead size is guaranteed to be larger than readaround\n size, and they are _async_, hence will mostly behave better\nHowever for B\n- sync readahead size could be smaller than readaround size, hence may\n make things worse by produce more smaller IOs\nwhich will be fixed by this patch.\n\nFinal conclusion:\n- mmap readahead reduced major faults by 1/3 and no obvious overheads;\n- mmap io can be further reduced by 1/4 with this patch.\n\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "61b7cbdba2f3c588a0cf3db574c562805454b09b", "tree": "657756b1ab5dc5fd63b341d999d492b868f4309c", "parents": [ "10be0b372cac50e2e7a477852f98bf069a97a3fa" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue Jun 16 15:31:36 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jun 16 19:47:30 2009 -0700" }, "message": "readahead: remove redundant test in shrink_readahead_size_eio()\n\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d30a11004e3411909f2448546f036a011978062e", "tree": "c1980adb410d9fabd2c2eb8af9f0ed8ee4b656da", "parents": [ "2fad6f5deee5556f511eab58da78737a23ddb35d" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue Jun 16 15:31:30 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jun 16 19:47:29 2009 -0700" }, "message": "readahead: record mmap read-around states in file_ra_state\n\nMmap read-around now shares the same code style and data structure with\nreadahead code.\n\nThis also removes do_page_cache_readahead(). Its last user, mmap\nread-around, has been changed to call ra_submit().\n\nThe no-readahead-if-congested logic is dumped by the way. Users will be\npretty sensitive about the slow loading of executables. So it\u0027s\nunfavorable to disabled mmap read-around on a congested queue.\n\n[akpm@linux-foundation.org: coding-style fixes]\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Fengguang Wu \u003cwfg@mail.ustc.edu.cn\u003e\nCc: Ying Han \u003cyinghan@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "2fad6f5deee5556f511eab58da78737a23ddb35d", "tree": "eca8262062c4fda63cb3bd34f9478f3fbfd7f518", "parents": [ "70ac23cfa31f68289d4b720c6162b3929ab4de36" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue Jun 16 15:31:29 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jun 16 19:47:29 2009 -0700" }, "message": "readahead: enforce full readahead size on async mmap readahead\n\nWe need this in one particular case and two more general ones.\n\nNow we do async readahead for sequential mmap reads, and do it with the\nhelp of PG_readahead. For normal reads, PG_readahead is the sufficient\ncondition to do a sequential readahead. But unfortunately, for mmap\nreads, there is a tiny nuisance:\n\n[11736.998347] readahead-init0(process: sh/23926, file: sda1/w3m, offset\u003d0:4503599627370495, ra\u003d0+4-3) \u003d 4\n[11737.014985] readahead-around(process: w3m/23926, file: sda1/w3m, offset\u003d0:0, ra\u003d290+32-0) \u003d 17\n[11737.019488] readahead-around(process: w3m/23926, file: sda1/w3m, offset\u003d0:0, ra\u003d118+32-0) \u003d 32\n[11737.024921] readahead-interleaved(process: w3m/23926, file: sda1/w3m, offset\u003d0:2, ra\u003d4+6-6) \u003d 6\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~\n\nAn unfavorably small readahead. The original dumb read-around size could\nbe more efficient.\n\nThat happened because ld-linux.so does a read(832) in L1 before mmap(),\nwhich triggers a 4-page readahead, with the second page tagged\nPG_readahead.\n\nL0: open(\"/lib/libc.so.6\", O_RDONLY) \u003d 3\nL1: read(3, \"\\177ELF\\2\\1\\1\\0\\0\\0\\0\\0\\0\\0\\0\\0\\3\\0\u003e\\0\\1\\0\\0\\0\\340\\342\"..., 832) \u003d 832\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\nL2: fstat(3, {st_mode\u003dS_IFREG|0755, st_size\u003d1420624, ...}) \u003d 0\nL3: mmap(NULL, 3527256, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) \u003d 0x7fac6e51d000\nL4: mprotect(0x7fac6e671000, 2097152, PROT_NONE) \u003d 0\nL5: mmap(0x7fac6e871000, 20480, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x154000) \u003d 0x7fac6e871000\nL6: mmap(0x7fac6e876000, 16984, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) \u003d 0x7fac6e876000\nL7: close(3) \u003d 0\n\nIn general, the PG_readahead flag will also be hit in cases\n\n- sequential reads\n\n- clustered random reads\n\nA full readahead size is desirable in both cases.\n\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Wu Fengguang \u003cfengguang.wu@intel.com\u003e\nCc: Ying Han \u003cyinghan@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ], "next": "70ac23cfa31f68289d4b720c6162b3929ab4de36" }