)]}' { "log": [ { "commit": "2150edc6c5cf00f7adb54538b9ea2a3e9cedca3f", "tree": "f72a0d85e66f500b4cead348a231e3d3b9f357bc", "parents": [ "cd764695b67386a81964f68e9c66efd9f13f4d29", "4b905671d2ea09fd48fed72c581df17e40823f39" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 08 17:14:59 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 08 17:14:59 2009 -0800" }, "message": "Merge branch \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4\n\n* \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (57 commits)\n jbd2: Fix oops in jbd2_journal_init_inode() on corrupted fs\n ext4: Remove \"extents\" mount option\n block: Add Kconfig help which notes that ext4 needs CONFIG_LBD\n ext4: Make printk\u0027s consistently prefixed with \"EXT4-fs: \"\n ext4: Add sanity checks for the superblock before mounting the filesystem\n ext4: Add mount option to set kjournald\u0027s I/O priority\n jbd2: Submit writes to the journal using WRITE_SYNC\n jbd2: Add pid and journal device name to the \"kjournald2 starting\" message\n ext4: Add markers for better debuggability\n ext4: Remove code to create the journal inode\n ext4: provide function to release metadata pages under memory pressure\n ext3: provide function to release metadata pages under memory pressure\n add releasepage hooks to block devices which can be used by file systems\n ext4: Fix s_dirty_blocks_counter if block allocation failed with nodelalloc\n ext4: Init the complete page while building buddy cache\n ext4: Don\u0027t allow new groups to be added during block allocation\n ext4: mark the blocks/inode bitmap beyond end of group as used\n ext4: Use new buffer_head flag to check uninit group bitmaps initialization\n ext4: Fix the race between read_inode_bitmap() and ext4_new_inode()\n ext4: code cleanup\n ...\n" }, { "commit": "cd764695b67386a81964f68e9c66efd9f13f4d29", "tree": "504e961ab6bad164c41f4b9c1ff00c0ce7f645ee", "parents": [ "97d61b8e3aef163a75f80f4762794c154572293d", "ffda8c7dc492e2170bb263f7c56f286992ceb54b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 08 16:27:31 2009 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 08 16:27:31 2009 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (45 commits)\n [SCSI] qla2xxx: Update version number to 8.03.00-k1.\n [SCSI] qla2xxx: Add ISP81XX support.\n [SCSI] qla2xxx: Use proper request/response queues with MQ instantiations.\n [SCSI] qla2xxx: Correct MQ-chain information retrieval during a firmware dump.\n [SCSI] qla2xxx: Collapse EFT/FCE copy procedures during a firmware dump.\n [SCSI] qla2xxx: Don\u0027t pollute kernel logs with ZIO/RIO status messages.\n [SCSI] qla2xxx: Don\u0027t fallback to interrupt-polling during re-initialization with MSI-X enabled.\n [SCSI] qla2xxx: Remove support for reading/writing HW-event-log.\n [SCSI] cxgb3i: add missing include\n [SCSI] scsi_lib: fix DID_RESET status problems\n [SCSI] fc transport: restore missing dev_loss_tmo callback to LLDD\n [SCSI] aha152x_cs: Fix regression that keeps driver from using shared interrupts\n [SCSI] sd: Correctly handle 6-byte commands with DIX\n [SCSI] sd: DIF: Fix tagging on platforms with signed char\n [SCSI] sd: DIF: Show app tag on error\n [SCSI] Fix error handling for DIF/DIX\n [SCSI] scsi_lib: don\u0027t decrement busy counters when inserting commands\n [SCSI] libsas: fix test for negative unsigned and typos\n [SCSI] a2091, gvp11: kill warn_unused_result warnings\n [SCSI] fusion: Move a dereference below a NULL test\n ...\n\nFixed up trivial conflict due to moving the async part of sd_probe\naround in the async probes vs using dev_set_name() in naming.\n" }, { "commit": "4d783b093cd4f3e2738527365979cbd1c5101065", "tree": "cd8b330894e63fe7e039884f56e3f710b95c5e38", "parents": [ "abda14189251563a50f56da5ea2e37e904ac4cba" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Jan 06 15:16:33 2009 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Tue Jan 06 15:16:33 2009 -0500" }, "message": "block: Add Kconfig help which notes that ext4 needs CONFIG_LBD\n\nSigned-off-by: \"Theodore Ts\u0027o\" \u003ctytso@mit.edu\u003e\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "3ada8b7e980dac7cc42937d42d90ee51b19204fe", "tree": "6c23e272997cd5b68459c5a92a4490504f1c4be7", "parents": [ "4383fc3d9a98759a1ec3c57a30781e95a4cfcfa9" ], "author": { "name": "Kay Sievers", "email": "kay.sievers@vrfy.org", "time": "Tue Jan 06 10:44:43 2009 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Tue Jan 06 10:44:43 2009 -0800" }, "message": "block: struct device - replace bus_id with dev_name(), dev_set_name()\n\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "97ae77a1cd332c7b011d71315c8faabce6840c72", "tree": "fc243ca28ea474eaf81729079eeb3b259cd1b81b", "parents": [ "56c451f4b583ccdf80c9e676179c9cb49de86745" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Thu Dec 18 14:49:38 2008 +0900" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Jan 02 11:10:35 2009 -0600" }, "message": "[SCSI] block: make blk_rq_map_user take a NULL user-space buffer for WRITE\n\nThe commit 818827669d85b84241696ffef2de485db46b0b5e (block: make\nblk_rq_map_user take a NULL user-space buffer) extended\nblk_rq_map_user to accept a NULL user-space buffer with a READ\ncommand. It was necessary to convert sg to use the block layer mapping\nAPI.\n\nThis patch extends blk_rq_map_user again for a WRITE command. It is\nnecessary to convert st and osst drivers to use the block layer\napping API.\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "56c451f4b583ccdf80c9e676179c9cb49de86745", "tree": "7ad9daeafb88e3c4390bf7a2db0c203fe2642728", "parents": [ "e623ddb4e940b266adc77ba1cc28a3554aa90e79" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Thu Dec 18 14:49:37 2008 +0900" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Jan 02 11:10:08 2009 -0600" }, "message": "[SCSI] block: fix the partial mappings with struct rq_map_data\n\nThis fixes bio_copy_user_iov to properly handle the partial mappings\nwith struct rq_map_data (which only sg uses for now but st and osst\nwill shortly). It adds the offset member to struct rq_map_data and\nchanges blk_rq_map_user to update it so that bio_copy_user_iov can add\nan appropriate page frame via bio_add_pc_page().\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "2ca1a615835d9f4990f42102ab1f2ef434e7e89c", "tree": "726cf3d5f29a6c66c44e4bd68e7ebed2fd83d059", "parents": [ "e12f0102ac81d660c9f801d0a0e10ccf4537a9de", "6a94cb73064c952255336cc57731904174b2c58f" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Wed Dec 31 23:05:57 2008 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Wed Dec 31 23:05:57 2008 +1030" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6\n\nConflicts:\n\n\tarch/x86/kernel/io_apic.c\n" }, { "commit": "33edcf133ba93ecba2e4b6472e97b689895d805c", "tree": "327d7a20acef64005e7c5ccbfa1265be28aeb6ac", "parents": [ "be4d638c1597580ed2294d899d9f1a2cd10e462c", "3c92ec8ae91ecf59d88c798301833d7cf83f2179" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Dec 30 08:02:35 2008 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Dec 30 08:02:35 2008 +1030" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6\n" }, { "commit": "62c1fe9d9f0a676fce89185b1513f0e5f473c72c", "tree": "28085a657fa93ed63b7dda20b5a84885e67d429e", "parents": [ "8ae30b895805a6e2bb725b1d78b12daabd7eadfe" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 15 21:19:25 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:52 2008 +0100" }, "message": "cfq-iosched: fix race between exiting queue and exiting task\n\nOriginal patch from Nikanth Karthikesan \u003cknikanth@suse.de\u003e\n\nWhen a queue exits the queue lock is taken and cfq_exit_queue() would free all\nthe cic\u0027s associated with the queue.\n\nBut when a task exits, cfq_exit_io_context() gets cic one by one and then\nlocks the associated queue to call __cfq_exit_single_io_context. It looks like\nbetween getting a cic from the ioc and locking the queue, the queue might have\nexited on another cpu.\n\nFix this by rechecking the cfq_io_context queue key inside the queue lock\nagain, and not calling into __cfq_exit_single_io_context() if somebody\nbeat us to it.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "b3a6ffe16b5cc48abe7db8d04882dc45280eb693", "tree": "7bd0860ca93beb9aa8c5cb54a3081617d8d12de9", "parents": [ "3c18ce71af754cefae75103dbae28817e04b2db4" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Dec 12 09:51:16 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:51 2008 +0100" }, "message": "Get rid of CONFIG_LSF\n\nWe have two seperate config entries for large devices/files. One\nis CONFIG_LBD that guards just the devices, the other is CONFIG_LSF\nthat handles large files. This doesn\u0027t make a lot of sense, you typically\nwant both or none. So get rid of CONFIG_LSF and change CONFIG_LBD wording\nto indicate that it covers both.\n\nAcked-by: Jean Delvare \u003ckhali@linux-fr.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "3c18ce71af754cefae75103dbae28817e04b2db4", "tree": "74e9ccf6478972c23f63c3304594dccaba7cd35b", "parents": [ "18af8b2ca34b831c32c6fa01e7ce33143c33bb63" ], "author": { "name": "Roel Kluin", "email": "roel.kluin@gmail.com", "time": "Wed Dec 10 15:47:33 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:51 2008 +0100" }, "message": "block: make blk_softirq_init() static\n\nSparse asked whether these could be static.\n\nSigned-off-by: Roel Kluin \u003croel.kluin@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "18af8b2ca34b831c32c6fa01e7ce33143c33bb63", "tree": "0feb71c891dea467c88181b1974d15cf58243f9a", "parents": [ "a6f23657d3072bde6844055bbc2290e497f33fbc" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Thu Dec 04 08:56:35 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:51 2008 +0100" }, "message": "block: use min_not_zero in blk_queue_stack_limits\n\nzero is invalid for max_phys_segments, max_hw_segments, and\nmax_segment_size. It\u0027s better to use use min_not_zero instead of\nmin. min() works though (because the commit 0e435ac makes sure that\nthese values are set to the default values, non zero, if a queue is\ninitialized properly).\n\nWith this patch, blk_queue_stack_limits does the almost same thing\nthat dm\u0027s combine_restrictions_low() does. I think that it\u0027s easy to\nremove dm\u0027s combine_restrictions_low.\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "a6f23657d3072bde6844055bbc2290e497f33fbc", "tree": "bd96916615d04228cc9492ae198ed5012d5ee86a", "parents": [ "30e0dc28bff9dc456cdfc2aae4aca78b8b1a1cec" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 24 12:52:42 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:51 2008 +0100" }, "message": "block: add one-hit cache for disk partition lookup\n\ndisk_map_sector_rcu() returns a partition from a sector offset,\nwhich we use for IO statistics on a per-partition basis. The\nlookup itself is an O(N) list lookup, where N is the number of\npartitions. This actually hurts performance quite a bit, even\non the lower end partitions. On higher numbered partitions,\nit can get pretty bad.\n\nSolve this by adding a one-hit cache for partition lookup.\nThis makes the lookup O(1) for the case where we do most IO to\none partition. Even for mixed partition workloads, amortized cost\nis pretty close to O(1) since the natural IO batching makes the\none-hit cache last for lots of IOs.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "30e0dc28bff9dc456cdfc2aae4aca78b8b1a1cec", "tree": "cb72eba9e27bf893928189c832d29e97bbe132ad", "parents": [ "31dcfab0ae6eb0340d456931df4d478e17efc65d" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Oct 20 15:44:28 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:51 2008 +0100" }, "message": "cfq-iosched: remove limit of dispatch depth of max 4 times quantum\n\nThis basically limits the hardware queue depth to 4*quantum at any\npoint in time, which is 16 with the default settings. As CFQ uses\nother means to shrink the hardware queue when necessary in the first\nplace, there\u0027s really no need for this extra heuristic. Additionally,\nit ends up hurting performance in some cases.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "b374d18a4bfce705e4a99ae9f501b53e86ecb283", "tree": "7e773464725a78e96a2017fa36cf9d4a4e29f73b", "parents": [ "abf137dd7712132ee56d5b3143c2ff61a72a5faa" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 31 10:05:07 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:29:50 2008 +0100" }, "message": "block: get rid of elevator_t typedef\n\nJust use struct elevator_queue everywhere instead.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "a31a97381cdf7dceb03b797a8faf9bc8a01c65d1", "tree": "268f86dc23b8338396faf5b891af60a7fb23929f", "parents": [ "a185eb4bc84155fff35b602ce99602c010de9634" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 13:58:29 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:45 2008 +0100" }, "message": "block: don\u0027t use plugging on SSD devices\n\nWe just want to hand the first bits of IO to the device as fast\nas possible. Gains a few percent on the IOPS rate.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "a185eb4bc84155fff35b602ce99602c010de9634", "tree": "e052052cbac813166b614325b32584d8bb9beff0", "parents": [ "58eea927d2de43dc6f03d1ba2c46e55854b31540" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Nov 28 13:32:07 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:45 2008 +0100" }, "message": "block: fix empty barrier on write-through w/ ordered tag\n\nEmpty barrier on write-through (or no cache) w/ ordered tag has no\ncommand to execute and without any command to execute ordered tag is\nnever issued to the device and the ordering is never achieved. Force\ndraining for such cases.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "58eea927d2de43dc6f03d1ba2c46e55854b31540", "tree": "bb43f43891a3154741bb6a116dc2956342160f15", "parents": [ "8f11b3e99a1136fcbb67316c3260f085299c0bff" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Nov 28 13:32:06 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:45 2008 +0100" }, "message": "block: simplify empty barrier implementation\n\nEmpty barrier required special handling in __elv_next_request() to\ncomplete it without letting the low level driver see it.\n\nWith previous changes, barrier code is now flexible enough to skip the\nBAR step using the same barrier sequence selection mechanism. Drop\nthe special handling and mask off q-\u003eordered from start_ordered().\n\nRemove blk_empty_barrier() test which now has no user.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "8f11b3e99a1136fcbb67316c3260f085299c0bff", "tree": "bb9b12fe23aceac19e24f674786612d0fcad2142", "parents": [ "f671620e7d895af221bdfeda751d54fa55ed9546" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Nov 28 13:32:05 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:45 2008 +0100" }, "message": "block: make barrier completion more robust\n\nBarrier completion had the following assumptions.\n\n* start_ordered() couldn\u0027t finish the whole sequence properly. If all\n actions are to be skipped, q-\u003eordseq is set correctly but the actual\n completion was never triggered thus hanging the barrier request.\n\n* Drain completion in elv_complete_request() assumed that there\u0027s\n always at least one request in the queue when drain completes.\n\nBoth assumptions are true but these assumptions need to be removed to\nimprove empty barrier implementation. This patch makes the following\nchanges.\n\n* Make start_ordered() use blk_ordered_complete_seq() to mark skipped\n steps complete and notify __elv_next_request() that it should fetch\n the next request if the whole barrier has completed inside\n start_ordered().\n\n* Make drain completion path in elv_complete_request() check whether\n the queue is empty. Empty queue also indicates drain completion.\n\n* While at it, convert 0/1 return from blk_do_ordered() to false/true.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "f671620e7d895af221bdfeda751d54fa55ed9546", "tree": "beeb843a4a356d94b6b4faec97e078b2a4ad1f09", "parents": [ "a7384677b2f4cd40948fd7ce024ba5e1821444ba" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Nov 28 13:32:04 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:45 2008 +0100" }, "message": "block: make every barrier action optional\n\nIn all barrier sequences, the barrier write itself was always assumed\nto be issued and thus didn\u0027t have corresponding control flag. This\npatch adds QUEUE_ORDERED_DO_BAR and unify action mask handling in\nstart_ordered() such that any barrier action can be skipped.\n\nThis patch doesn\u0027t introduce any visible behavior changes.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "a7384677b2f4cd40948fd7ce024ba5e1821444ba", "tree": "679af09dde4d2659bc6a7dcd1d3dc34752899fa6", "parents": [ "313e42999dbc0f234ca5909a236f78f082cb43b1" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Nov 28 13:32:03 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:44 2008 +0100" }, "message": "block: remove duplicate or unused barrier/discard error paths\n\n* Because barrier mode can be changed dynamically, whether barrier is\n supported or not can be determined only when actually issuing the\n barrier and there is no point in checking it earlier. Drop barrier\n support check in generic_make_request() and __make_request(), and\n update comment around the support check in blk_do_ordered().\n\n* There is no reason to check discard support in both\n generic_make_request() and __make_request(). Drop the check in\n __make_request(). While at it, move error action block to the end\n of the function and add unlikely() to q existence test.\n\n* Barrier request, be it empty or not, is never passed to low level\n driver and thus it\u0027s meaningless to try to copy back req-\u003esector to\n bio-\u003ebi_sector on error. In addition, the notion of failed sector\n doesn\u0027t make any sense for empty barrier to begin with. Drop the\n code block from __end_that_request_first().\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "313e42999dbc0f234ca5909a236f78f082cb43b1", "tree": "023ac251809e3926ebc6b6c2174d67f8c4ac535f", "parents": [ "ba744d5e290055d171c68067259fcc1e2721f542" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Nov 28 13:32:02 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:44 2008 +0100" }, "message": "block: reorganize QUEUE_ORDERED_* constants\n\nSeparate out ordering type (drain,) and action masks (preflush,\npostflush, fua) from visible ordering mode selectors\n(QUEUE_ORDERED_*). Ordering types are now named QUEUE_ORDERED_BY_*\nwhile action masks are named QUEUE_ORDERED_DO_*.\n\nThis change is necessary to add QUEUE_ORDERED_DO_BAR and make it\noptional to improve empty barrier implementation.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "64d01dc9e1927e6535627d73f2336c75d1dd3fe2", "tree": "1813333970f6ed1c1959edfc2842b00ed2b3a70e", "parents": [ "08bafc0341f2f7920e9045bc32c40299cac8c21b" ], "author": { "name": "Cheng Renquan", "email": "crquan@gmail.com", "time": "Wed Dec 03 12:41:39 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:44 2008 +0100" }, "message": "block: use cancel_work_sync() instead of kblockd_flush_work()\n\nAfter many improvements on kblockd_flush_work, it is now identical to\ncancel_work_sync, so a direct call to cancel_work_sync is suggested.\n\nThe only difference is that cancel_work_sync is a GPL symbol,\nso no non-GPL modules anymore.\n\nSigned-off-by: Cheng Renquan \u003ccrquan@gmail.com\u003e\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "08bafc0341f2f7920e9045bc32c40299cac8c21b", "tree": "ab4ad0f73f457409239a98c050bf7edd47c03cf1", "parents": [ "7c239517d9f18427fc2e7ed259fb3b866595f5af" ], "author": { "name": "Keith Mannthey", "email": "kmannth@us.ibm.com", "time": "Tue Nov 25 10:24:35 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:44 2008 +0100" }, "message": "block: Supress Buffer I/O errors when SCSI REQ_QUIET flag set\n\nAllow the scsi request REQ_QUIET flag to be propagated to the buffer\nfile system layer. The basic ideas is to pass the flag from the scsi\nrequest to the bio (block IO) and then to the buffer layer. The buffer\nlayer can then suppress needless printks.\n\nThis patch declutters the kernel log by removed the 40-50 (per lun)\nbuffer io error messages seen during a boot in my multipath setup . It\nis a good chance any real errors will be missed in the \"noise\" it the\nlogs without this patch.\n\nDuring boot I see blocks of messages like\n\"\n__ratelimit: 211 callbacks suppressed\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242847\nBuffer I/O error on device sdm, logical block 1\nBuffer I/O error on device sdm, logical block 5242878\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242879\nBuffer I/O error on device sdm, logical block 5242872\n\"\nin my logs.\n\nMy disk environment is multipath fiber channel using the SCSI_DH_RDAC\ncode and multipathd. This topology includes an \"active\" and \"ghost\"\npath for each lun. IO\u0027s to the \"ghost\" path will never complete and the\nSCSI layer, via the scsi device handler rdac code, quick returns the IOs\nto theses paths and sets the REQ_QUIET scsi flag to suppress the scsi\nlayer messages.\n\n I am wanting to extend the QUIET behavior to include the buffer file\nsystem layer to deal with these errors as well. I have been running this\npatch for a while now on several boxes without issue. A few runs of\nbonnie++ show no noticeable difference in performance in my setup.\n\nThanks for John Stultz for the quiet_error finalization.\n\nSubmitted-by: Keith Mannthey \u003ckmannth@us.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "7c239517d9f18427fc2e7ed259fb3b866595f5af", "tree": "cd2149cc2f2eb0faa83d38fe64e1228f7c703a85", "parents": [ "42364690992e592c05f85c76fda4055820b48c1b" ], "author": { "name": "Wu Fengguang", "email": "fengguang.wu@intel.com", "time": "Tue Nov 25 09:08:39 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:43 2008 +0100" }, "message": "block: don\u0027t take lock on changing ra_pages\n\nThere\u0027s no need to take queue_lock or kernel_lock when modifying\nbdi-\u003era_pages. So remove them. Also remove out of date comment for\nqueue_max_sectors_store().\n\nSigned-off-by: Wu Fengguang \u003cwfg@linux.intel.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "c6a06f707cc29ea3a47588e4d2cd0bdcfa311a7d", "tree": "6a058cb9ab1703eb329021b586a885d7ac92257a", "parents": [ "8a3173de4ab4cdacc43675dc5c077f9a5bf17f5f" ], "author": { "name": "Qinghuang Feng", "email": "qhfeng.kernel@gmail.com", "time": "Mon Nov 24 10:43:36 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:43 2008 +0100" }, "message": "block/blk-tag.c: cleanup kernel-doc\n\nThere is no argument named @tags in blk_init_tags,\nremove its\u0027 comment.\n\nSigned-off-by: Qinghuang Feng \u003cqhfeng.kernel@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "2b91bafcc0fc545e489e9537a38f487706960ea5", "tree": "69ebe2bbfef7f95036fab184ff206208438caf43", "parents": [ "70ed28b92a786f44750ab64117b03d126dd14656" ], "author": { "name": "Milton Miller", "email": "miltonm@bga.com", "time": "Mon Nov 17 13:10:34 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:42 2008 +0100" }, "message": "scsi-ioctl: use clock_t \u003c\u003e jiffies\n\nConvert the timeout ioctl scalling to use the clock_t functions\nwhich are much more accurate with some USER_HZ vs HZ combinations.\n\nSigned-off-by: Milton Miller \u003cmiltonm@bga.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "70ed28b92a786f44750ab64117b03d126dd14656", "tree": "03e84c48c262b380fbcccb21f55ff979aaeb4e77", "parents": [ "65d3618ccfe686e8d7b3f01a838d0578182406df" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Nov 19 14:38:39 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:42 2008 +0100" }, "message": "block: leave the request timeout timer running even on an empty list\n\nFor sync IO, we\u0027ll often do them serialized. This means we\u0027ll be touching\nthe queue timer for every IO, as opposed to only occasionally like we\ndo for queued IO. Instead of deleting the timer when the last request\nis removed, just let continue running. If a new request comes up soon\nwe then don\u0027t have to readd the timer again. If no new requests arrive,\nthe timer will expire without side effect later.\n\nThis improves high iops sync IO by ~1%.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "65d3618ccfe686e8d7b3f01a838d0578182406df", "tree": "0a53a7d3649eb82a8e2af2c9f4e8de2d9b1b6616", "parents": [ "565e411d764eeda006738dfadbccca79d48381e1" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 30 08:53:02 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:42 2008 +0100" }, "message": "block: add comment in blk_rq_timed_out() about why next can not be 0\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "565e411d764eeda006738dfadbccca79d48381e1", "tree": "5bd835b9aed2df0dc91ed58740834d6e156e86ea", "parents": [ "66d352e1e410dcea22fecb9fa9ec09dd23a62e1c" ], "author": { "name": "malahal@us.ibm.com", "email": "malahal@us.ibm.com", "time": "Thu Oct 30 08:51:58 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Mon Dec 29 08:28:42 2008 +0100" }, "message": "block: optimizations in blk_rq_timed_out_timer()\n\nNow the rq-\u003edeadline can\u0027t be zero if the request is in the\ntimeout_list, so there is no need to have next_set. There is no need to\naccess a request\u0027s deadline field if blk_rq_timed_out is called on it.\n\nSigned-off-by: Malahal Naineni \u003cmalahal@us.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "be4d638c1597580ed2294d899d9f1a2cd10e462c", "tree": "2a4800b70c1e2a1249915f74f8e45f62de319457", "parents": [ "9be3eec2c83848a1ca57ebad13c63c95d0df01e2" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Fri Dec 26 22:23:43 2008 +1030" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Fri Dec 26 22:23:43 2008 +1030" }, "message": "cpumask: Replace cpu_coregroup_map with cpu_coregroup_mask\n\ncpu_coregroup_map returned a cpumask_t: it\u0027s going away.\n\n(Note, the sched part of this patch won\u0027t apply meaningfully to the\nsched tree, but I\u0027m posting it to show the goal).\n\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nSigned-off-by: Mike Travis \u003ctravis@sgi.com\u003e\nCc: Jens Axboe \u003caxboe@kernel.dk\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\n" }, { "commit": "30cd324e9787ccc9a5ede59742d5409857550692", "tree": "e53027e0283119de0ef4c3a431b1d20cd43003fb", "parents": [ "c71dd42db2c6f1637b92502a214587431c1a6ad2", "6d102bc68f3dd2ae0e305b09170b1751aa67baeb", "3d9101e92529e1ff6014f95a69afc82f37b9b13a" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Dec 19 09:42:40 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Dec 19 09:42:40 2008 +0100" }, "message": "Merge branches \u0027tracing/ftrace\u0027, \u0027tracing/ring-buffer\u0027 and \u0027tracing/urgent\u0027 into tracing/core\n\nConflicts:\n\tinclude/linux/ftrace.h\n" }, { "commit": "f2f1fa78a155524b849edf359e42a3001ea652c0", "tree": "8d38fc76e48fc219e393ee57ee5f76843b735258", "parents": [ "24920a79a27dd28e67088bbfcf36746d15587f4d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Dec 05 14:49:18 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Dec 05 14:49:18 2008 -0800" }, "message": "Enforce a minimum SG_IO timeout\n\nThere\u0027s no point in having too short SG_IO timeouts, since if the\ncommand does end up timing out, we\u0027ll end up through the reset sequence\nthat is several seconds long in order to abort the command that timed\nout.\n\nAs a result, shorter timeouts than a few seconds simply do not make\nsense, as the recovery would be longer than the timeout itself.\n\nAdd a BLK_MIN_SG_TIMEOUT to match the existign BLK_DEFAULT_SG_TIMEOUT.\n\nSuggested-by: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nCc: Jeff Garzik \u003cjeff@garzik.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "970987beb9c99ca806edc464518d411cc399fb4d", "tree": "51e2298b6eb7e6412009428b380a51e1f17b531b", "parents": [ "faec2ec505d397e9426754722b6e80d519c4938f", "1fd8f2a3f9a91b287a876cef830b21baafc8a799", "feaf3848a813a106f163013af6fcf6c4bfec92d9" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Dec 05 14:45:22 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Fri Dec 05 14:45:22 2008 +0100" }, "message": "Merge branches \u0027tracing/ftrace\u0027, \u0027tracing/function-graph-tracer\u0027 and \u0027tracing/urgent\u0027 into tracing/core\n" }, { "commit": "fd4ce1acd0f8558033b1a6968001552bd7671e6d", "tree": "7ce54f2b2c967fe2315917ffa2eb8eb7d47e7a2b", "parents": [ "ebbefc011e56bd85b4745d01e5b8d7d05d95ed5d" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Nov 05 14:58:42 2008 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Dec 04 04:22:57 2008 -0500" }, "message": "[PATCH 1/2] kill FMODE_NDELAY_NOW\n\nUpdate FMODE_NDELAY before each ioctl call so that we can kill the\nmagic FMODE_NDELAY_NOW. It would be even better to do this directly\nin setfl(), but for that we\u0027d need to have FMODE_NDELAY for all files,\nnot just block special files.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "1c925604e1038c7c65b91a92d14dc972b3a70a97", "tree": "5c4e2f24a7e972b053cd05a6c80dc319a35f6e9b", "parents": [ "50c396d38c1a7f0c693579ec88cb4be3c0b0645e" ], "author": { "name": "Andreas Schwab", "email": "schwab@suse.de", "time": "Fri Oct 31 22:39:46 2008 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Dec 04 04:22:55 2008 -0500" }, "message": "[PATCH] Fix block dev compat ioctl handling\n\nCommit 33c2dca4957bd0da3e1af7b96d0758d97e708ef6 (trim file propagation\nin block/compat_ioctl.c) removed the handling of some ioctls from\ncompat_blkdev_driver_ioctl. That caused them to be rejected as unknown\nby the compat layer.\n\nSigned-off-by: Andreas Schwab \u003cschwab@suse.de\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "0e435ac26e3f951d83338ed3d4ab7dc0fe0055bc", "tree": "8f208a3093de1a314a981ae47e5ef92a5909c13b", "parents": [ "53a08807c01989c6847bb135d8d43f61c5dfdda5" ], "author": { "name": "Milan Broz", "email": "mbroz@redhat.com", "time": "Wed Dec 03 12:55:08 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Dec 03 12:55:55 2008 +0100" }, "message": "block: fix setting of max_segment_size and seg_boundary mask\n\nFix setting of max_segment_size and seg_boundary mask for stacked md/dm\ndevices.\n\nWhen stacking devices (LVM over MD over SCSI) some of the request queue\nparameters are not set up correctly in some cases by default, namely\nmax_segment_size and and seg_boundary mask.\n\nIf you create MD device over SCSI, these attributes are zeroed.\n\nProblem become when there is over this mapping next device-mapper mapping\n- queue attributes are set in DM this way:\n\nrequest_queue max_segment_size seg_boundary_mask\nSCSI 65536 0xffffffff\nMD RAID1 0 0\nLVM 65536 -1 (64bit)\n\nUnfortunately bio_add_page (resp. bio_phys_segments) calculates number of\nphysical segments according to these parameters.\n\nDuring the generic_make_request() is segment cout recalculated and can\nincrease bio-\u003ebi_phys_segments count over the allowed limit. (After\nbio_clone() in stack operation.)\n\nThi is specially problem in CCISS driver, where it produce OOPS here\n\n BUG_ON(creq-\u003enr_phys_segments \u003e MAXSGENTRIES);\n\n(MAXSEGENTRIES is 31 by default.)\n\nSometimes even this command is enough to cause oops:\n\n dd iflag\u003ddirect if\u003d/dev/\u003cvg\u003e/\u003clv\u003e of\u003d/dev/null bs\u003d128000 count\u003d10\n\nThis command generates bios with 250 sectors, allocated in 32 4k-pages\n(last page uses only 1024 bytes).\n\nFor LVM layer, it allocates bio with 31 segments (still OK for CCISS),\nunfortunatelly on lower layer it is recalculated to 32 segments and this\nviolates CCISS restriction and triggers BUG_ON().\n\nThe patch tries to fix it by:\n\n * initializing attributes above in queue request constructor\n blk_queue_make_request()\n\n * make sure that blk_queue_stack_limits() inherits setting\n\n (DM uses its own function to set the limits because it\n blk_queue_stack_limits() was introduced later. It should probably switch\n to use generic stack limit function too.)\n\n * sets the default seg_boundary value in one place (blkdev.h)\n\n * use this mask as default in DM (instead of -1, which differs in 64bit)\n\nBugs related to this:\nhttps://bugzilla.redhat.com/show_bug.cgi?id\u003d471639\nhttp://bugzilla.kernel.org/show_bug.cgi?id\u003d8672\n\nSigned-off-by: Milan Broz \u003cmbroz@redhat.com\u003e\nReviewed-by: Alasdair G Kergon \u003cagk@redhat.com\u003e\nCc: Neil Brown \u003cneilb@suse.de\u003e\nCc: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nCc: Tejun Heo \u003chtejun@gmail.com\u003e\nCc: Mike Miller \u003cmike.miller@hp.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "53a08807c01989c6847bb135d8d43f61c5dfdda5", "tree": "e2297de901dc82a0336e6af28b594864792c585d", "parents": [ "bf91db18ac2852a3ff39fe25ff56c5557c0fff78" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Wed Dec 03 12:41:26 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Dec 03 12:41:26 2008 +0100" }, "message": "block: internal dequeue shouldn\u0027t start timer\n\nblkdev_dequeue_request() and elv_dequeue_request() are equivalent and\nboth start the timeout timer. Barrier code dequeues the original\nbarrier request but doesn\u0027t passes the request itself to lower level\ndriver, only broken down proxy requests; however, as the original\nbarrier code goes through the same dequeue path and timeout timer is\nstarted on it. If barrier sequence takes long enough, this timer\nexpires but the low level driver has no idea about this request and\noops follows.\n\nTimeout timer shouldn\u0027t have been started on the original barrier\nrequest as it never goes through actual IO. This patch unexports\nelv_dequeue_request(), which has no external user anyway, and makes it\noperate on elevator proper w/o adding the timer and make\nblkdev_dequeue_request() call elv_dequeue_request() and add timer.\nInternal users which don\u0027t pass the request to driver - barrier code\nand end_that_request_last() - are converted to use\nelv_dequeue_request().\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nCc: Mike Anderson \u003candmike@linux.vnet.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "bf91db18ac2852a3ff39fe25ff56c5557c0fff78", "tree": "42ce67fceb2b3072717cde31c8b4474f4c4a050d", "parents": [ "53cc0b2948bcb8a084982e6c1f9bd7b337e0df38" ], "author": { "name": "Cheng Renquan", "email": "crquan@gmail.com", "time": "Thu Nov 20 08:37:37 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Dec 03 12:41:20 2008 +0100" }, "message": "block: set disk-\u003enode_id before it\u0027s being used\n\ndisk-\u003enode_id will be refered in allocating in disk_expand_part_tbl, so we\nshould set it before disk-\u003enode_id is refered.\n\nSigned-off-by: Cheng Renquan \u003ccrquan@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "53cc0b2948bcb8a084982e6c1f9bd7b337e0df38", "tree": "e2d4ca231035de3a14f47b97f6c9196fc375340c", "parents": [ "f6f7b52e2f6149d2ee365717afff315b05720162" ], "author": { "name": "Petr Vandrovec", "email": "petr@vandrovec.name", "time": "Wed Nov 19 11:12:14 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Dec 03 12:41:20 2008 +0100" }, "message": "When block layer fails to map iov, it calls bio_unmap_user to undo\nmapping. Which is good if pages were mapped - but if they were provided\nby someone else and just copied then bad things happen - pages are\nreleased once here, and once by caller, leading to user triggerable BUG\nat include/linux/mm.h:246.\n\nSigned-off-by: Petr Vandrovec \u003cpetr@vandrovec.name\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "0bfc24559d7945506184d86739fe365a181f06b7", "tree": "5c152128faac7080f4802ce03d1c6b6bc7173227", "parents": [ "5f3ea37c7716db4e894a480e0c18b24399595b6b" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Nov 26 11:59:56 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Nov 26 13:04:35 2008 +0100" }, "message": "blktrace: port to tracepoints, update\n\nPort to the new tracepoints API: split DEFINE_TRACE() and DECLARE_TRACE()\nsites. Spread them out to the usage sites, as suggested by\nMathieu Desnoyers.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nAcked-by: Mathieu Desnoyers \u003cmathieu.desnoyers@polymtl.ca\u003e\n" }, { "commit": "5f3ea37c7716db4e894a480e0c18b24399595b6b", "tree": "db6784635d024894f641b340dcd7c5060c446077", "parents": [ "509dceef6470442d8c7b8a43ec34125205840b3c" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Thu Oct 30 08:34:33 2008 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Nov 26 12:13:34 2008 +0100" }, "message": "blktrace: port to tracepoints\n\nThis was a forward port of work done by Mathieu Desnoyers, I changed it to\nencode the \u0027what\u0027 parameter on the tracepoint name, so that one can register\ninterest in specific events and not on classes of events to then check the\n\u0027what\u0027 parameter.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "c26156b2534c75bb3cdedf76f6ad1340971cf5bd", "tree": "b95ed72f569c105fcc9e8f38df3d47628e689059", "parents": [ "98ba4031ab2adc8b394295e68aa4c8fe9d5060db" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Nov 18 15:07:05 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Nov 18 15:08:56 2008 +0100" }, "message": "block: hold extra reference to bio in blk_rq_map_user_iov()\n\nIf the size passed in is OK but we end up mapping too many segments,\nwe call the unmap path directly like from IO completion. But from IO\ncompletion we have an extra reference to the bio, so this error case\ngoes OOPS when it attempts to free and already free bio.\n\nFix it by getting an extra reference to the bio before calling the\nunmap failure case.\n\nReported-by: Petr Vandrovec \u003cvandrove@vc.cvut.cz\u003e\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "561ec68e4de7947167937c49c451728e6b19e63b", "tree": "4391af5bc00e5a2039e703df43ec3e84a21b63d5", "parents": [ "55e8e30c382d25c34f8aafcc78efec948571a941" ], "author": { "name": "Zhang, Yanmin", "email": "yanmin_zhang@linux.intel.com", "time": "Fri Nov 14 08:26:30 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Nov 18 15:08:56 2008 +0100" }, "message": "block: fix boot failure with CONFIG_DEBUG_BLOCK_EXT_DEVT\u003dy and nash\n\nWe run into system boot failure with kernel 2.6.28-rc. We found it on a\ncouple of machines, including T61 notebook, nehalem machine, and another\nHPC NX6325 notebook. All the machines use FedoraCore 8 or FedoraCore 9.\nWith kernel prior to 2.6.28-rc, system boot doesn\u0027t fail.\n\nI debug it and locate the root cause. Pls. see\nhttp://bugzilla.kernel.org/show_bug.cgi?id\u003d11899\nhttps://bugzilla.redhat.com/show_bug.cgi?id\u003d471517\n\nAs a matter of fact, there are 2 bugs.\n\n1)root\u003d/dev/sda1, system boot randomly fails. Mostly, boot for 5 times\nand fails once. nash has a bug. Some of its functions misuse return\nvalue 0. Sometimes, 0 means timeout and no uevent available. Sometimes,\n0 means nash gets an uevent, but the uevent isn\u0027t block-related (for\nexmaple, usb). If by coincidence, kernel tells nash that uevents are\navailable, but kernel also set timeout, nash might stops collecting\nother uevents in queue if current uevent isn\u0027t block-related. I work\nout a patch for nash to fix it.\nhttp://bugzilla.kernel.org/attachment.cgi?id\u003d18858\n\n2) root\u003dLABEL\u003d/, system always can\u0027t boot. initrd init reports\nswitchroot fails. Here is an executation branch of nash when booting:\n (1) nash read /sys/block/sda/dev; Assume major is 8 (on my desktop)\n (2) nash query /proc/devices with the major number; It found line\n\t\"8 sd\";\n (3) nash use \u0027sd\u0027 to search its own probe table to find device (DISK)\n\ttype for the device and add it to its own list;\n (4) Later on, it probes all devices in its list to get filesystem\n\tlabels; scsi register \"8 sd\" always.\n\nWhen major is 259, nash fails to find the device(DISK) type. I enables\nCONFIG_DEBUG_BLOCK_EXT_DEVT\u003dy when compiling kernel, so 259 is picked up\nfor device /dev/sda1, which causes nash to fail to find device (DISK)\ntype.\n\nTo fixing issue 2), I create a patch for nash and another patch for\nkernel.\n\nhttp://bugzilla.kernel.org/attachment.cgi?id\u003d18859\nhttp://bugzilla.kernel.org/attachment.cgi?id\u003d18837\n\nBelow is the patch for kernel 2.6.28-rc4. It registers blkext, a new\nblock device in proc/devices.\n\nWith 2 patches on nash and 1 patch on kernel, I boot my machines for\ndozens of times without failure.\n\nSigned-off-by Zhang Yanmin \u003cyanmin.zhang@linux.intel.com\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "ba32929a91fe2c0628f5be62d1597b379c8d3062", "tree": "09ea54ed2d2299d25b92fee6a6dde4d7527a3d37", "parents": [ "eb60fa1066622ddb2278732cf61e0c4544e82c6f" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Mon Nov 10 15:29:58 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Nov 18 15:08:56 2008 +0100" }, "message": "block: make add_partition() return pointer to hd_struct\n\nMake add_partition() return pointer to the new hd_struct on success\nand ERR_PTR() value on failure. This change will be used to fix md\nautodetection bug.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nCc: Neil Brown \u003cneilb@suse.de\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "7838c15b8dd18e78a523513749e5b54bda07b0cb", "tree": "7fa295ce51f263c55756d3693f5f5ed7beb68a87", "parents": [ "9c133c469d38043d5aadaa03f2fb840d88d1cf4f" ], "author": { "name": "Alan Stern", "email": "stern@rowland.harvard.edu", "time": "Thu Nov 06 08:42:49 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Nov 06 08:42:49 2008 +0100" }, "message": "Block: use round_jiffies_up()\n\nThis patch (as1159b) changes the timeout routines in the block core to\nuse round_jiffies_up(). There\u0027s no point in rounding the timer\ndeadline down, since if it expires too early we will have to restart\nit.\n\nThe patch also removes some unnecessary tests when a request is\nremoved from the queue\u0027s timer list.\n\nSigned-off-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "e78042e5b83936b1d12a4b5bbb492bdd88ad76c6", "tree": "bd65da0f0f87c844fb4a22c5c56822a5ff15165b", "parents": [ "2920ebbd65f3e80c318adf5191ac0987142bda80" ], "author": { "name": "Mike Anderson", "email": "andmike@linux.vnet.ibm.com", "time": "Thu Oct 30 02:16:20 2008 -0700" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Nov 06 08:41:56 2008 +0100" }, "message": "blk: move blk_delete_timer call in end_that_request_last\n\nMove the calling blk_delete_timer to later in end_that_request_last to\naddress an issue where blkdev_dequeue_request may have add a timer for the\nrequest.\n\nSigned-off-by: Mike Anderson \u003candmike@linux.vnet.ibm.com\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "2920ebbd65f3e80c318adf5191ac0987142bda80", "tree": "1f45e880b1cfe86040b4acf834a1d04faf9804f3", "parents": [ "f92131c3dd567fc6df18ce3f46fcf57ecbdefbe0" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Thu Oct 30 08:32:29 2008 +0100" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Nov 06 08:41:55 2008 +0100" }, "message": "block: add timer on blkdev_dequeue_request() not elv_next_request()\n\nBlock queue supports two usage models - one where block driver peeks\nat the front of queue using elv_next_request(), processes it and\nfinishes it and the other where block driver peeks at the front of\nqueue, dequeue the request using blkdev_dequeue_request() and finishes\nit. The latter is more flexible as it allows the driver to process\nmultiple commands concurrently.\n\nThese two inconsistent usage models affect the block layer\nimplementation confusing. For some, elv_next_request() is considered\nthe issue point while others consider blkdev_dequeue_request() the\nissue point.\n\nTill now the inconsistency mostly affect only accounting, so it didn\u0027t\nreally break anything seriously; however, with block layer timeout,\nthis inconsistency hits hard. Block layer considers\nelv_next_request() the issue point and adds timer but SCSI layer\nthinks it was just peeking and when the request can\u0027t process the\ncommand right away, it\u0027s just left there without further processing.\nThis makes the request dangling on the timer list and, when the timer\ngoes off, the request which the SCSI layer and below think is still on\nthe block queue ends up in the EH queue, causing various problems - EH\nhang (failed count goes over busy count and EH never wakes up),\nWARN_ON() and oopses as low level driver trying to handle the unknown\ncommand, etc. depending on the timing.\n\nAs SCSI midlayer is the only user of block layer timer at the moment,\nmoving blk_add_timer() to elv_dequeue_request() fixes the problem;\nhowever, this two usage models definitely need to be cleaned up in the\nfuture.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "43381785a5ba1cb424b36812373a6a04054b5c3c", "tree": "46917d4723f5cf167913597346c0377ab8da59a6", "parents": [ "75fa67706cce5272bcfc51ed646f2da21f3bdb6e" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Mon Oct 20 15:43:43 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Nov 06 08:41:55 2008 +0100" }, "message": "block: remove unused ll_new_mergeable()\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "88ed86fee6651033de9b7038dac7869a9f19775a", "tree": "38b638d2e7cba110ec271275738f221feb7e0a37", "parents": [ "3856d30ded1fe43c6657927ebad402d25cd128f4", "59c7572e82d69483a66eaa67b46548baeb69ecf4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 23 12:04:37 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 23 12:04:37 2008 -0700" }, "message": "Merge branch \u0027proc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/proc\n\n* \u0027proc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/proc: (35 commits)\n proc: remove fs/proc/proc_misc.c\n proc: move /proc/vmcore creation to fs/proc/vmcore.c\n proc: move pagecount stuff to fs/proc/page.c\n proc: move all /proc/kcore stuff to fs/proc/kcore.c\n proc: move /proc/schedstat boilerplate to kernel/sched_stats.h\n proc: move /proc/modules boilerplate to kernel/module.c\n proc: move /proc/diskstats boilerplate to block/genhd.c\n proc: move /proc/zoneinfo boilerplate to mm/vmstat.c\n proc: move /proc/vmstat boilerplate to mm/vmstat.c\n proc: move /proc/pagetypeinfo boilerplate to mm/vmstat.c\n proc: move /proc/buddyinfo boilerplate to mm/vmstat.c\n proc: move /proc/vmallocinfo to mm/vmalloc.c\n proc: move /proc/slabinfo boilerplate to mm/slub.c, mm/slab.c\n proc: move /proc/slab_allocators boilerplate to mm/slab.c\n proc: move /proc/interrupts boilerplate code to fs/proc/interrupts.c\n proc: move /proc/stat to fs/proc/stat.c\n proc: move rest of /proc/partitions code to block/genhd.c\n proc: move /proc/cpuinfo code to fs/proc/cpuinfo.c\n proc: move /proc/devices code to fs/proc/devices.c\n proc: move rest of /proc/locks to fs/locks.c\n ...\n" }, { "commit": "5f4f0c4d3fc95cb7d509e4469a22aed40532f0fc", "tree": "f780c6faa1ea3995c1b2a626fae595b078ffde2e", "parents": [ "54d822a6169b76b807b8cdbbf76ff2812a88947f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 23 10:28:25 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Oct 23 10:28:25 2008 -0700" }, "message": "compat_blkdev_driver_ioctl: Remove unused variable warning\n\nVariable \u0027ret\u0027 is no longer used. Don\u0027t declare it.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "31d85ab28e71b0c938e0ef48af45747e80d99b53", "tree": "1cd97bee97abb008dd3591da2a001e3e85356fb6", "parents": [ "5c9fe6281b75832e8d2555ec8700ea763d9a865e" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Mon Oct 06 12:55:38 2008 +0400" }, "committer": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Thu Oct 23 17:57:37 2008 +0400" }, "message": "proc: move /proc/diskstats boilerplate to block/genhd.c\n\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "f500975a3f3ecf3611d79f1d933906753460b9f2", "tree": "df3728998ca1638ecbccf4ab6f401f9a2cbdba88", "parents": [ "8591cf43224980a0bc9216a4e50b0a740f8cba35" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Sat Oct 04 23:53:21 2008 +0400" }, "committer": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Thu Oct 23 15:07:31 2008 +0400" }, "message": "proc: move rest of /proc/partitions code to block/genhd.c\n\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "56b26add02b4bdea81d5e0ebda60db1fe3311ad4", "tree": "3e9b5877f77ce24a6d17574ff3d32ae8351fa0ef", "parents": [ "6af3a56e1dd4d95836a47214e5c60d5b749a5501" ], "author": { "name": "Al Viro", "email": "al@aretha.pdmi.ras.ru", "time": "Fri Sep 19 03:17:36 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:49:14 2008 -0400" }, "message": "[PATCH] kill the rest of struct file propagation in block ioctls\n\nNow we can switch blkdev_ioctl() block_device/mode\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "6af3a56e1dd4d95836a47214e5c60d5b749a5501", "tree": "757b01fe8668783fcf52a4e8fe82c5d8ebbc6b72", "parents": [ "45048d0961cdbf7ddfba906797ad8cb42489e070" ], "author": { "name": "Al Viro", "email": "al@aretha.pdmi.ras.ru", "time": "Fri Sep 19 03:08:13 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:49:12 2008 -0400" }, "message": "[PATCH] get rid of struct file use in blkdev_ioctl() BLKBSZSET\n\nWe need to do bd_claim() only if file hadn\u0027t been opened with O_EXCL\nand then we have no need to use file itself as owner.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "45048d0961cdbf7ddfba906797ad8cb42489e070", "tree": "d471933ccf9e8c52f608d0cef88de0569982a933", "parents": [ "e436fdae70a31102d2be32969b80fe8545edebd9" ], "author": { "name": "Al Viro", "email": "al@aretha.pdmi.ras.ru", "time": "Thu Sep 18 15:53:24 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:49:10 2008 -0400" }, "message": "[PATCH] get rid of blkdev_locked_ioctl()\n\nMost of that stuff doesn\u0027t need BKL at all; expand in the (only) caller,\nmerge the switch into one there and leave BKL only around the stuff that\nmight actually need it.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "e436fdae70a31102d2be32969b80fe8545edebd9", "tree": "ab07320851fc5ae82bc422ab54dbb922a8057452", "parents": [ "572c48921574dbe6dceb958cf965aa962baefde4" ], "author": { "name": "Al Viro", "email": "al@aretha.pdmi.ras.ru", "time": "Thu Sep 18 03:38:12 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:49:08 2008 -0400" }, "message": "[PATCH] get rid of blkdev_driver_ioctl()\n\nconvert remaining callers to __blkdev_driver_ioctl()\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "33c2dca4957bd0da3e1af7b96d0758d97e708ef6", "tree": "435f24f2c0b4f15b267502e4a35c7fbd6a610058", "parents": [ "90b8f2824ce68dd87d304641a1d5a048dfff39f5" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Feb 22 22:02:50 2008 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:48:54 2008 -0400" }, "message": "[PATCH] trim file propagation in block/compat_ioctl.c\n\n... and remove the handling of cases when it falls back to native\nwithout changing arguments.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "90b8f2824ce68dd87d304641a1d5a048dfff39f5", "tree": "b55ddcb619aac1e16d845dd6796c51721dd6ee56", "parents": [ "40cc51be699e37b78102ce7b2196ec9c32fe1908" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sun Mar 02 10:43:36 2008 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:48:52 2008 -0400" }, "message": "[PATCH] end of methods switch: remove the old ones\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "d4430d62fa77208824a37fe6f85ab2831d274769", "tree": "5d4d0bca31e63eb208fbebe4f39c912b964c1e4d", "parents": [ "badf8082c33d18b118d3a6f1b32d5ea6b97d3839" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sun Mar 02 09:09:22 2008 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:47:32 2008 -0400" }, "message": "[PATCH] beginning of methods conversion\n\nTo keep the size of changesets sane we split the switch by drivers;\nto keep the damn thing bisectable we do the following:\n\t1) rename the affected methods, add ones with correct\nprototypes, make (few) callers handle both. That\u0027s this changeset.\n\t2) for each driver convert to new methods. *ALL* drivers\nare converted in this series.\n\t3) kill the old (renamed) methods.\n\nNote that it _is_ a flagday; all in-tree drivers are converted and by the\nend of this series no trace of old methods remain. The only reason why\nwe do that this way is to keep the damn thing bisectable and allow per-driver\ndebugging if anything goes wrong.\n\nNew methods:\n\topen(bdev, mode)\n\trelease(disk, mode)\n\tioctl(bdev, mode, cmd, arg)\t\t/* Called without BKL */\n\tcompat_ioctl(bdev, mode, cmd, arg)\n\tlocked_ioctl(bdev, mode, cmd, arg)\t/* Called with BKL, legacy */\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "633a08b81206122469365b4c72eaeb71f04f2cb4", "tree": "9cd30507efd30645d30d351c7ad0d5966f806532", "parents": [ "a0eb62a0a4470fef5a5f41e7f1442fdd667220ef" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Aug 29 20:34:12 2007 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:47:26 2008 -0400" }, "message": "[PATCH] introduce __blkdev_driver_ioctl()\n\nAnalog of blkdev_driver_ioctl() with sane arguments. For\nnow uses fake struct file, by the end of the series it won\u0027t\nand blkdev_driver_ioctl() will become a wrapper around it.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "74f3c8aff36ad6552ea609c8b20bfd588fa16f38", "tree": "c046823751971ff37b66499282ca2b2e263cc71a", "parents": [ "e915e872ed921d707bc32b3f2184d43abfa8c9e2" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Mon Aug 27 15:38:10 2007 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:47:14 2008 -0400" }, "message": "[PATCH] switch scsi_cmd_ioctl() to passing fmode_t\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "e915e872ed921d707bc32b3f2184d43abfa8c9e2", "tree": "698142e0a56622c3a3c0740522537dbff703732a", "parents": [ "5842e51ff2e8d660756248db80ad24e4f41977d9" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Sep 02 17:16:41 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:47:12 2008 -0400" }, "message": "[PATCH] switch sg_scsi_ioctl() to passing fmode_t\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "5842e51ff2e8d660756248db80ad24e4f41977d9", "tree": "245de9d6178271f9caa4f52f3abde6455ac4221f", "parents": [ "86d434dede14108dd917b25af0f29c0cb28b8d18" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Sep 02 16:35:55 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:47:10 2008 -0400" }, "message": "[PATCH] pass mode instead of file to sg_io()\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "aeb5d727062a0238a2f96c9c380fbd2be4640c6f", "tree": "51dae8a071fcf42e4431a66d37c5b843c8e99cf6", "parents": [ "2515ddc6db8eb49a79f0fe5e67ff09ac7c81eab4" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Sep 02 15:28:45 2008 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Oct 21 07:47:06 2008 -0400" }, "message": "[PATCH] introduce fmode_t, do annotations\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "c53dbf54863e7f3b0b8810dda2bdd0290006bdac", "tree": "f783074f1bec1112bf1148a077e0114a38403ad4", "parents": [ "b73b636e8987f8728c6c700377615757691b9a55", "f73e2d13a16cc88c4faa4729967f92bfeec8a142" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Oct 17 09:29:55 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Oct 17 09:29:55 2008 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.dk/linux-2.6-block\n\n* \u0027for-linus\u0027 of git://git.kernel.dk/linux-2.6-block:\n block: remove __generic_unplug_device() from exports\n block: move q-\u003eunplug_work initialization\n blktrace: pass zfcp driver data\n blktrace: add support for driver data\n block: fix current kernel-doc warnings\n block: only call -\u003erequest_fn when the queue is not stopped\n block: simplify string handling in elv_iosched_store()\n block: fix kernel-doc for blk_alloc_devt()\n block: fix nr_phys_segments miscalculation bug\n block: add partition attribute for partition number\n block: add BIG FAT WARNING to CONFIG_DEBUG_BLOCK_EXT_DEVT\n softirq: Add support for triggering softirq work on softirqs.\n" }, { "commit": "ed09441dacc2a2d6c170aa3b1f79a041291a813f", "tree": "95c35bdf4f0b679806984093dce627a66d0d7cf1", "parents": [ "b225ee5bed70254a100896c473e6dd8c2be45c18", "4c393e6e457fb41169dd110c1b96a138394c2d7b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Oct 17 09:00:23 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Oct 17 09:00:23 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (39 commits)\n [SCSI] sd: fix compile failure with CONFIG_BLK_DEV_INTEGRITY\u003dn\n libiscsi: fix locking in iscsi_eh_device_reset\n libiscsi: check reason why we are stopping iscsi session to determine error value\n [SCSI] iscsi_tcp: return a descriptive error value during connection errors\n [SCSI] libiscsi: rename host reset to target reset\n [SCSI] iscsi class: fix endpoint id handling\n [SCSI] libiscsi: Support drivers initiating session removal\n [SCSI] libiscsi: fix data corruption when target has to resend data-in packets\n [SCSI] sd: Switch kernel printing level for DIF messages\n [SCSI] sd: Correctly handle all combinations of DIF and DIX\n [SCSI] sd: Always print actual protection_type\n [SCSI] sd: Issue correct protection operation\n [SCSI] scsi_error: fix target reset handling\n [SCSI] lpfc 8.2.8 v2 : Add statistical reporting control and additional fc vendor events\n [SCSI] lpfc 8.2.8 v2 : Add sysfs control of target queue depth handling\n [SCSI] lpfc 8.2.8 v2 : Revert target busy in favor of transport disrupted\n [SCSI] scsi_dh_alua: remove REQ_NOMERGE\n [SCSI] lpfc 8.2.8 : update driver version to 8.2.8\n [SCSI] lpfc 8.2.8 : Add MSI-X support\n [SCSI] lpfc 8.2.8 : Update driver to use new Host byte error code DID_TRANSPORT_DISRUPTED\n ...\n" }, { "commit": "f73e2d13a16cc88c4faa4729967f92bfeec8a142", "tree": "6844c0e85b9b4ecdc8f2fc30ce8e7d2152d0c417", "parents": [ "713ada9ba94f2ad874cffd074b83e3dc681ca82f" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 14:03:08 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 14:03:08 2008 +0200" }, "message": "block: remove __generic_unplug_device() from exports\n\nThe only out-of-core user is IDE, and that should be using\nblk_start_queueing() instead.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "713ada9ba94f2ad874cffd074b83e3dc681ca82f", "tree": "e94cae4cb7d6d8bee404feb869bceee74b402f93", "parents": [ "0997f1c5fec0b540784611036d458a84a1f7029a" ], "author": { "name": "Peter Zijlstra", "email": "a.p.zijlstra@chello.nl", "time": "Thu Oct 16 13:44:57 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 08:46:57 2008 +0200" }, "message": "block: move q-\u003eunplug_work initialization\n\nmodprobe loop; rmmod loop effectively creates a blk_queue and destroys it\nwhich results in q-\u003eunplug_work being canceled without it ever being\ninitialized.\n\nTherefore, move the initialization of q-\u003eunplug_work from\nblk_queue_make_request() to blk_alloc_queue*().\n\nReported-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nSigned-off-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "496aa8a98f5ab22ced46be5dc2087cdf3d029bd7", "tree": "5c4335ca12bbc8f93f7630c2ef0733973099cd17", "parents": [ "80a4b58e36b63d7b0b592beb1bd6410aadeeb63c" ], "author": { "name": "Randy Dunlap", "email": "randy.dunlap@oracle.com", "time": "Thu Oct 16 07:46:23 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 08:46:57 2008 +0200" }, "message": "block: fix current kernel-doc warnings\n\nFix block kernel-doc warnings:\n\nWarning(linux-2.6.27-git4//fs/block_dev.c:1272): No description found for parameter \u0027path\u0027\nWarning(linux-2.6.27-git4//block/blk-core.c:1021): No description found for parameter \u0027cpu\u0027\nWarning(linux-2.6.27-git4//block/blk-core.c:1021): No description found for parameter \u0027part\u0027\nWarning(/var/linsrc/linux-2.6.27-git4//block/genhd.c:544): No description found for parameter \u0027partno\u0027\n\nSigned-off-by: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "80a4b58e36b63d7b0b592beb1bd6410aadeeb63c", "tree": "a37df7feb2db67254cc8927730828e87706b2205", "parents": [ "ee2e992cc28553f6c4dd1ab5483c8733c393626b" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Oct 14 09:51:06 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 08:46:57 2008 +0200" }, "message": "block: only call -\u003erequest_fn when the queue is not stopped\n\nCallers should use either blk_run_queue/__blk_run_queue, or\nblk_start_queueing() to invoke request handling instead of calling\n-\u003erequest_fn() directly as that does not take the queue stopped\nflag into account.\n\nAlso add appropriate comments on the above functions to detail\ntheir usage.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "ee2e992cc28553f6c4dd1ab5483c8733c393626b", "tree": "1947e8cb26f1018c20f89db93e06698cc17353f5", "parents": [ "e6d63840ba55ffd3a79aea6792aac6f29f338083" ], "author": { "name": "Li Zefan", "email": "lizf@cn.fujitsu.com", "time": "Tue Oct 14 08:49:56 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 08:46:57 2008 +0200" }, "message": "block: simplify string handling in elv_iosched_store()\n\nstrlcpy() guarantees the dest buffer is NULL teminated.\n\nSigned-off-by: Li Zefan \u003clizf@cn.fujitsu.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "e6d63840ba55ffd3a79aea6792aac6f29f338083", "tree": "6fa14f4a89167e142385369a51e4d09f5c5233f8", "parents": [ "8677142710516d986d932d6f1fba7be8382c1fec" ], "author": { "name": "Li Zefan", "email": "lizf@cn.fujitsu.com", "time": "Tue Oct 14 08:49:34 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 08:46:56 2008 +0200" }, "message": "block: fix kernel-doc for blk_alloc_devt()\n\nNo argument \u0027gfp_mask\u0027 for blk_alloc_devt().\n\nSigned-off-by: Li Zefan \u003clizf@cn.fujitsu.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "8677142710516d986d932d6f1fba7be8382c1fec", "tree": "5221f48b4e7e62590f8bbb832b4065886681d6cf", "parents": [ "0fc71e3d6520ba7abad5cfbc9a33db0190e4d5b8" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Mon Oct 13 14:19:05 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Fri Oct 17 08:46:56 2008 +0200" }, "message": "block: fix nr_phys_segments miscalculation bug\n\nThis fixes the bug reported by Nikanth Karthikesan \u003cknikanth@suse.de\u003e:\n\nhttp://lkml.org/lkml/2008/10/2/203\n\nThe root cause of the bug is that blk_phys_contig_segment\nmiscalculates q-\u003emax_segment_size.\n\nblk_phys_contig_segment checks:\n\nreq-\u003ebiotail-\u003ebi_size + next_req-\u003ebio-\u003ebi_size \u003e q-\u003emax_segment_size\n\nBut blk_recalc_rq_segments might expect that req-\u003ebiotail and the\nprevious bio in the req are supposed be merged into one\nsegment. blk_recalc_rq_segments might also expect that next_req-\u003ebio\nand the next bio in the next_req are supposed be merged into one\nsegment. In such case, we merge two requests that can\u0027t be merged\nhere. Later, blk_rq_map_sg gives more segments than it should.\n\nWe need to keep track of segment size in blk_recalc_rq_segments and\nuse it to see if two requests can be merged. This patch implements it\nin the similar way that we used to do for hw merging (virtual\nmerging).\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "1ff9f542e5f87c299226557ce5e67a402ed4b502", "tree": "964bd7293fbf607f6a7a329d9fc40e0a2b798a96", "parents": [ "ae87221d3ce49d9de1e43756da834fd0bf05a2ad" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Jul 21 20:03:34 2008 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Oct 16 09:24:41 2008 -0700" }, "message": "device create: block: convert device_create_drvdata to device_create\n\nNow that device_create() has been audited, rename things back to the\noriginal call to be sane.\n\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "6000a368cd8e6da1caf101411bdb494cd6fb8b09", "tree": "4cd3333af00182e915aa96ffa49069f5f76976dc", "parents": [ "056a44834950ffa51fafa6c76a720fa32e86851a" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Tue Aug 19 18:45:30 2008 -0500" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Mon Oct 13 09:28:52 2008 -0400" }, "message": "[SCSI] block: separate failfast into multiple bits.\n\nMultipath is best at handling transport errors. If it gets a device\nerror then there is not much the multipath layer can do. It will just\naccess the same device but from a different path.\n\nThis patch breaks up failfast into device, transport and driver errors.\nThe multipath layers (md and dm mutlipath) only ask the lower levels to\nfast fail transport errors. The user of failfast, read ahead, will ask\nto fast fail on all errors.\n\nNote that blk_noretry_request will return true if any failfast bit\nis set. This allows drivers that do not support the multipath failfast\nbits to continue to fail on any failfast error like before. Drivers\nlike scsi that are able to fail fast specific errors can check\nfor the specific fail fast type. In the next patch I will convert\nscsi.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "ad7fce93147d32ae53d25d9ea1a8ba31a239deee", "tree": "baa8c613c1d34ec247f13b860c3bf7671b07f586", "parents": [ "0c032ab889e7b20b8a5a7d09313e4aca214a15f7" ], "author": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Wed Oct 01 03:38:39 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:21 2008 +0200" }, "message": "block: Switch blk_integrity_compare from bdev to gendisk\n\nThe DM and MD integrity support now depends on being able to use\ngendisks instead of block_devices when comparing integrity profiles.\nChange function parameters accordingly.\n\nAlso update comparison logic so that two NULL profiles are a valid\nconfiguration.\n\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "0c032ab889e7b20b8a5a7d09313e4aca214a15f7", "tree": "921d41ff5a31f9b538b43bea6fb56a6e2467530d", "parents": [ "74aa8c2cc010035a7eef2b4ca4d6430e0dae206a" ], "author": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Wed Oct 01 03:38:38 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:21 2008 +0200" }, "message": "block: Fix double put in blk_integrity_unregister\n\n- kobject_del already puts the parent.\n\n - Set integrity profile to NULL to prevent stale data.\n\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "d00e29fd99dd63d1c51917604e35dee824ed567f", "tree": "461c725d7ffaf2d5a69bc854ca4aa9977f530ea2", "parents": [ "99cd3386f290eaf61f2b7596d5a4cc2007771174" ], "author": { "name": "Kiyoshi Ueda", "email": "k-ueda@ct.jp.nec.com", "time": "Wed Oct 01 10:14:46 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:21 2008 +0200" }, "message": "block: remove end_{queued|dequeued}_request()\n\nThis patch removes end_queued_request() and end_dequeued_request(),\nwhich are no longer used.\n\nAs a results, users of __end_request() became only end_request().\nSo the actual code in __end_request() is moved to end_request()\nand __end_request() is removed.\n\nSigned-off-by: Kiyoshi Ueda \u003ck-ueda@ct.jp.nec.com\u003e\nSigned-off-by: Jun\u0027ichi Nomura \u003cj-nomura@ce.jp.nec.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "99cd3386f290eaf61f2b7596d5a4cc2007771174", "tree": "90f93246c060a33b90d0ef3d98f327768f167909", "parents": [ "7afb3a6e752503d5ebeb038336aa0fa886a51b44" ], "author": { "name": "Kiyoshi Ueda", "email": "k-ueda@ct.jp.nec.com", "time": "Wed Oct 01 10:13:44 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:21 2008 +0200" }, "message": "block: change elevator to use __blk_end_request()\n\nThis patch converts elevator to use __blk_end_request() directly\nso that end_{queued|dequeued}_request() can be removed.\nRelated \u0027uptodate\u0027 arguments is converted to \u0027error\u0027.\n\nSigned-off-by: Kiyoshi Ueda \u003ck-ueda@ct.jp.nec.com\u003e\nSigned-off-by: Jun\u0027ichi Nomura \u003cj-nomura@ce.jp.nec.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "0497b345e7d067109e0dd9bf9f4978a6847ee13b", "tree": "8c181a200c9e883a9c85169dff887bfb9109ee5e", "parents": [ "ef9e3facdf1fe1228721a7c295a76d1b7a0e57ec" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Oct 01 16:16:25 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:20 2008 +0200" }, "message": "blktrace: use BLKTRACE_BDEV_SIZE as the name size for setup structure\n\nDefine as 32, which is is what BDEVNAME_SIZE is/was as well. This keeps\nthe user interface the same and gets rid of the difference between\nkernel and user api here.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "ef9e3facdf1fe1228721a7c295a76d1b7a0e57ec", "tree": "33847b1e267895ea58c2399c06cdd23365c7dc73", "parents": [ "336c3d8ce771608815b65bcfa27a17a83b297328" ], "author": { "name": "Kiyoshi Ueda", "email": "k-ueda@ct.jp.nec.com", "time": "Wed Oct 01 16:12:15 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:20 2008 +0200" }, "message": "block: add lld busy state exporting interface\n\nThis patch adds an new interface, blk_lld_busy(), to check lld\u0027s\nbusy state from the block layer.\nblk_lld_busy() calls down into low-level drivers for the checking\nif the drivers set q-\u003elld_busy_fn() using blk_queue_lld_busy().\n\nThis resolves a performance problem on request stacking devices below.\n\nSome drivers like scsi mid layer stop dispatching request when\nthey detect busy state on its low-level device like host/target/device.\nIt allows other requests to stay in the I/O scheduler\u0027s queue\nfor a chance of merging.\n\nRequest stacking drivers like request-based dm should follow\nthe same logic.\nHowever, there is no generic interface for the stacked device\nto check if the underlying device(s) are busy.\nIf the request stacking driver dispatches and submits requests to\nthe busy underlying device, the requests will stay in\nthe underlying device\u0027s queue without a chance of merging.\nThis causes performance problem on burst I/O load.\n\nWith this patch, busy state of the underlying device is exported\nvia q-\u003elld_busy_fn(). So the request stacking driver can check it\nand stop dispatching requests if busy.\n\nThe underlying device driver must return the busy state appropriately:\n 1: when the device driver can\u0027t process requests immediately.\n 0: when the device driver can process requests immediately,\n including abnormal situations where the device driver needs\n to kill all requests.\n\nSigned-off-by: Kiyoshi Ueda \u003ck-ueda@ct.jp.nec.com\u003e\nSigned-off-by: Jun\u0027ichi Nomura \u003cj-nomura@ce.jp.nec.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "336c3d8ce771608815b65bcfa27a17a83b297328", "tree": "91841449cf24ee051ff1daefcb60a8fbefe2522a", "parents": [ "c0ddffa84a7d12da9943a94d04dadbfb1883b904" ], "author": { "name": "Elias Oltmanns", "email": "eo@nebensachen.de", "time": "Wed Oct 01 16:02:33 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:20 2008 +0200" }, "message": "block: Fix blk_start_queueing() to not kick a stopped queue\n\nblk_start_queueing() should act like the generic queue unplugging\nand kicking and ignore a stopped queue. Such a queue may not be\nrun until after a call to blk_start_queue().\n\nSigned-off-by: Elias Oltmanns \u003ceo@nebensachen.de\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "e3ba9ae58a5599226e3976b29c8093041ae7c332", "tree": "5e8a7c0c1b675e19fa6209489479fe9b799a1b3e", "parents": [ "f7d7b7a7a3db6526a84ea755c1c54a051e9a52de" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Sep 25 11:42:41 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:19 2008 +0200" }, "message": "block: reserve some tags just for sync IO\n\nBy only allowing async IO to consume 3/4 ths of the tag depth, we\nalways have slots free to serve sync IO. This is important to avoid\nhaving writes fill the entire tag queue, thus starving reads.\n\nOriginal patch and idea from Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "f7d7b7a7a3db6526a84ea755c1c54a051e9a52de", "tree": "4d02c3abfca1c6175527473dff12769b09b630bb", "parents": [ "8bff7c6b0f63c7ee9c5e3a076338d74125b8debb" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Sep 25 11:37:50 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:19 2008 +0200" }, "message": "block: as/cfq ssd idle check update\n\nWe really need to know about the hardware tagging support as well,\nsince if the SSD does not do tagging then we still want to idle.\nOtherwise have the same dependent sync IO vs flooding async IO\nproblem as on rotational media.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "a68bbddba486020c9c74825ce90c4c1ec463e0e8", "tree": "ce0acea1e609185968310316d3d1ac6f730db983", "parents": [ "9e49184c82e9ec3ab4d45f9ea5a17ccaf43869f0" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Wed Sep 24 13:03:33 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:19 2008 +0200" }, "message": "block: add queue flag for SSD/non-rotational devices\n\nWe don\u0027t want to idle in AS/CFQ if the device doesn\u0027t have a seek\npenalty. So add a QUEUE_FLAG_NONROT to indicate a non-rotational\ndevice, low level drivers should set this flag upon discovery of\nan SSD or similar device type.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "4ee5eaf4516a60f8ef64d3c246c64c6be0cf8c3a", "tree": "3097d305534cd3a0fbfcca31ab5f4079b90e0fc3", "parents": [ "82124d60354846623a4b94af335717a5e142a074" ], "author": { "name": "Kiyoshi Ueda", "email": "k-ueda@ct.jp.nec.com", "time": "Thu Sep 18 10:46:13 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:18 2008 +0200" }, "message": "block: add a queue flag for request stacking support\n\nThis patch adds a queue flag to indicate the block device can be\nused for request stacking.\n\nRequest stacking drivers need to stack their devices on top of\nonly devices of which q-\u003erequest_fn is functional.\nSince bio stacking drivers (e.g. md, loop) basically initialize\ntheir queue using blk_alloc_queue() and don\u0027t set q-\u003erequest_fn,\nthe check of (q-\u003erequest_fn \u003d\u003d NULL) looks enough for that purpose.\n\nHowever, dm will become both types of stacking driver (bio-based and\nrequest-based). And dm will always set q-\u003erequest_fn even if the dm\ndevice is bio-based of which q-\u003erequest_fn is not functional actually.\nSo we need something else to distinguish the type of the device.\nAdding a queue flag is a solution for that.\n\nThe reason why dm always sets q-\u003erequest_fn is to keep\nthe compatibility of dm user-space tools.\nCurrently, all dm user-space tools are using bio-based dm without\nspecifying the type of the dm device they use.\nTo use request-based dm without changing such tools, the kernel\nmust decide the type of the dm device automatically.\nThe automatic type decision can\u0027t be done at the device creation time\nand needs to be deferred until such tools load a mapping table,\nsince the actual type is decided by dm target type included in\nthe mapping table.\n\nSo a dm device has to be initialized using blk_init_queue()\nso that we can load either type of table.\nThen, all queue stuffs are set (e.g. q-\u003erequest_fn) and we have\nno element to distinguish that it is bio-based or request-based,\neven after a table is loaded and the type of the device is decided.\n\nBy the way, some stuffs of the queue (e.g. request_list, elevator)\nare needless when the dm device is used as bio-based.\nBut the memory size is not so large (about 20[KB] per queue on ia64),\nso I hope the memory loss can be acceptable for bio-based dm users.\n\nSigned-off-by: Kiyoshi Ueda \u003ck-ueda@ct.jp.nec.com\u003e\nSigned-off-by: Jun\u0027ichi Nomura \u003cj-nomura@ce.jp.nec.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "82124d60354846623a4b94af335717a5e142a074", "tree": "af7b98d5ef57483cfac53cc4866194391c3ff7b7", "parents": [ "32fab448e5e86694beade415e750363538ea5f49" ], "author": { "name": "Kiyoshi Ueda", "email": "k-ueda@ct.jp.nec.com", "time": "Thu Sep 18 10:45:38 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:18 2008 +0200" }, "message": "block: add request submission interface\n\nThis patch adds blk_insert_cloned_request(), a generic request\nsubmission interface for request stacking drivers.\nRequest-based dm will use it to submit their clones to underlying\ndevices.\n\nblk_rq_check_limits() is also added because it is possible that\nthe lower queue has stronger limitations than the upper queue\nif multiple drivers are stacking at request-level.\nNot only for blk_insert_cloned_request()\u0027s internal use, the function\nwill be used by request-based dm when the queue limitation is\nmodified (e.g. by replacing dm\u0027s table).\n\nSigned-off-by: Kiyoshi Ueda \u003ck-ueda@ct.jp.nec.com\u003e\nSigned-off-by: Jun\u0027ichi Nomura \u003cj-nomura@ce.jp.nec.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "32fab448e5e86694beade415e750363538ea5f49", "tree": "6eab6189afba893c302fa6acca1a89406fd620c6", "parents": [ "e3335de94067dbebe22e3962632ead34e832cb60" ], "author": { "name": "Kiyoshi Ueda", "email": "k-ueda@ct.jp.nec.com", "time": "Thu Sep 18 10:45:09 2008 -0400" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:18 2008 +0200" }, "message": "block: add request update interface\n\nThis patch adds blk_update_request(), which updates struct request\nwith completing its data part, but doesn\u0027t complete the struct\nrequest itself.\nThough it looks like end_that_request_first() of older kernels,\nblk_update_request() should be used only by request stacking drivers.\n\nRequest-based dm will use it in bio-\u003ebi_end_io callback to update\nthe original request when a data part of a cloned request completes.\nFollowings are additional background information of why request-based\ndm needs this interface.\n\n - Request stacking drivers can\u0027t use blk_end_request() directly from\n the lower driver\u0027s completion context (bio-\u003ebi_end_io or rq-\u003eend_io),\n because some device drivers (e.g. ide) may try to complete\n their request with queue lock held, and it may cause deadlock.\n See below for detailed description of possible deadlock:\n \u003chttp://marc.info/?l\u003dlinux-kernel\u0026m\u003d120311479108569\u0026w\u003d2\u003e\n\n - To solve that, request-based dm offloads the completion of\n cloned struct request to softirq context (i.e. using\n blk_complete_request() from rq-\u003eend_io).\n\n - Though it is possible to use the same solution from bio-\u003ebi_end_io,\n it will delay the notification of bio completion to the original\n submitter. Also, it will cause inefficient partial completion,\n because the lower driver can\u0027t perform the cloned request anymore\n and request-based dm needs to requeue and redispatch it to\n the lower driver again later. That\u0027s not good.\n\n - So request-based dm needs blk_update_request() to perform the bio\n completion in the lower driver\u0027s completion context, which is more\n efficient.\n\nSigned-off-by: Kiyoshi Ueda \u003ck-ueda@ct.jp.nec.com\u003e\nSigned-off-by: Jun\u0027ichi Nomura \u003cj-nomura@ce.jp.nec.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "e3335de94067dbebe22e3962632ead34e832cb60", "tree": "93d88fd70ce8af33d9444e9af91a4e0635a64c96", "parents": [ "9246b5f06deeea541e7c62437c2ad19a0b1172c0" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Sep 18 09:22:54 2008 -0700" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:18 2008 +0200" }, "message": "block: blk_cleanup_queue() should call blk_sync_queue()\n\nWhen a driver calls blk_cleanup_queue(), the device should be fully idle.\nHowever, the block layer may have pending plugging timers and the IO\nschedulers may have pending work in the work queues. So quisce the device\nby waiting for the timer and flushing the work queues.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "7ba1fbaa4a478f72fbaf5a56af9c82a77966b4c7", "tree": "275c1d86386d7db9fa75840363230098ca6bddd3", "parents": [ "581d4e28d9195aa8b2231383dbabc288988d615e" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Sep 16 09:54:11 2008 -0700" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:17 2008 +0200" }, "message": "block: use rq complete marking in blk_abort_request()\n\nWe cannot abort a request if we raced with the timeout handler already,\nor with the IO completion. So make blk_abort_request() mark the request\nas complete, and only continue if we succeeded.\n\nFound and suggested by Mike Anderson \u003candmike@linux.vnet.ibm.com\u003e\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "581d4e28d9195aa8b2231383dbabc288988d615e", "tree": "093bcac9a538ef67edcf91826f7532bdb88e6248", "parents": [ "0a0d96b03a1f3bfd6bc3ea08008699e8e59fccd9" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Sun Sep 14 05:56:33 2008 -0700" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:17 2008 +0200" }, "message": "block: add fault injection mechanism for faking request timeouts\n\nOnly works for the generic request timer handling. Allows one to\nsporadically ignore request completions, thus exercising the timeout\nhandling.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "3e6053d76dcbd92b2f9f4ad5ece9bce83149523e", "tree": "0e7ab62872071c39e2475decb769b9715de75376", "parents": [ "4677735f03f5b6b6f2182f457a921855cadfb85b" ], "author": { "name": "Hugh Dickins", "email": "hugh@veritas.com", "time": "Thu Sep 11 10:57:55 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:17 2008 +0200" }, "message": "block: adjust blkdev_issue_discard for swap\n\nTwo mods to blkdev_issue_discard(), thinking ahead to its use on swap:\n\n1. Add gfp_mask argument, so swap allocation can use it where GFP_KERNEL\n might deadlock but GFP_NOIO is safe.\n\n2. Enlarge nr_sects argument from unsigned to sector_t: unsigned long is\n enough to cover a whole swap area, but sector_t suits any partition.\n\nChange sb_issue_discard()\u0027s nr_blocks to sector_t too; but no need seen\nfor a gfp_mask there, just pass GFP_KERNEL down to blkdev_issue_discard().\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "11914a53d2ec2974a565311af327b8983d8c820d", "tree": "72713d2661ed56fa5d9e29203541bad32c88a3ba", "parents": [ "242f9dcb8ba6f68fcd217a119a7648a4f69290e9" ], "author": { "name": "Mike Anderson", "email": "andmike@linux.vnet.ibm.com", "time": "Sat Sep 13 20:31:27 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:13 2008 +0200" }, "message": "block: Add interface to abort queued requests\n\nSigned-off-by: Mike Anderson \u003candmike@linux.vnet.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "242f9dcb8ba6f68fcd217a119a7648a4f69290e9", "tree": "1bfe245ffbc50d204d76665cd8f90d85100f86a1", "parents": [ "608aeef17a91747d6303de4df5e2c2e6899a95e8" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Sun Sep 14 05:55:09 2008 -0700" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:13 2008 +0200" }, "message": "block: unify request timeout handling\n\nRight now SCSI and others do their own command timeout handling.\nMove those bits to the block layer.\n\nInstead of having a timer per command, we try to be a bit more clever\nand simply have one per-queue. This avoids the overhead of having to\ntear down and setup a timer for each command, so it will result in a lot\nless timer fiddling.\n\nSigned-off-by: Mike Anderson \u003candmike@linux.vnet.ibm.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "243294dae09c909c0442c8f04d470b69c3c19d6e", "tree": "b6178beadd19247b81d2b09e063e856943c83177", "parents": [ "fad7f01e61bf737fe8a3740d803f000db57ecac6" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Thu Sep 04 09:17:31 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:12 2008 +0200" }, "message": "block: fix duplicate headers for /proc/partitions\n\nseqf can be started multiple times for a read and the header should be\nprinted only for the initial one. Fix it.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "818827669d85b84241696ffef2de485db46b0b5e", "tree": "694d09728733e65d604bf3e1f13679db73fc1d9a", "parents": [ "839e96afba87117befd39cf4e43f156edc8047a7" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Tue Sep 02 16:20:19 2008 +0900" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:11 2008 +0200" }, "message": "block: make blk_rq_map_user take a NULL user-space buffer\n\nThis patch changes blk_rq_map_user to accept a NULL user-space buffer\nwith a READ command if rq_map_data is not NULL. Thus a caller can pass\npage frames to lk_rq_map_user to just set up a request and bios with\npage frames propely. bio_uncopy_user (called via blk_rq_unmap_user)\ndoesn\u0027t copy data to user space with such request.\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "839e96afba87117befd39cf4e43f156edc8047a7", "tree": "ae091a3a8ed386e080cd18569abc4335c1adde84", "parents": [ "55dc7db70a73a3809a2334063c9b5b0d8ccebdaa" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Sep 02 09:25:21 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:11 2008 +0200" }, "message": "block: update comment on end_request()\n\nIt refers to functions that no longer exist after the IO completion\nchanges.\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "2bbedcb4c1abac498f18e5770d62ae66ff235ada", "tree": "2609360651b9a0d137c5c564a971152c90d52001", "parents": [ "759f8ca3048f7438aa3129268d7252552505d662" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Aug 29 11:41:51 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:11 2008 +0200" }, "message": "block: don\u0027t test for partition size in bdget_disk() and blk_lookup_devt()\n\nbdget_disk() and blk_lookup_devt() never cared whether the specified\npartition (or disk) is zero sized or not. I got confused while\nconverting those not to depend on consecutive minor numbers in commit\n5a6411b1178baf534aa9138052864dfa89d3eada and later when dev0 was added\nit broke callers which expected to get valid return for zero sized\ndisk devices.\n\nSo, they never needed nr_sects checks in the first place. Kill them.\n\nThis problem was spotted and debugged by Bartlmoiej Zolnierkiewicz.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nCc: Bartlomiej Zolnierkiewicz \u003cbzolnier@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "aeb3d3a81e81c6323a17fe914e91eb228b3f1aa1", "tree": "9cacfa989eaad1611f61e5707fd432e7f1fde685", "parents": [ "01cfcddd98f09e05a2e36031654ed46643b76f23" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Thu Aug 28 09:27:42 2008 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Thu Oct 09 08:56:11 2008 +0200" }, "message": "block: kmalloc args reversed, small function definition fixes\n\nNoticed by sparse:\nblock/blk-softirq.c:156:12: warning: symbol \u0027blk_softirq_init\u0027 was not declared. Should it be static?\nblock/genhd.c:583:28: warning: function \u0027bdget_disk\u0027 with external linkage has definition\nblock/genhd.c:659:17: warning: incorrect type in argument 1 (different base types)\nblock/genhd.c:659:17: expected unsigned int [unsigned] [usertype] size\nblock/genhd.c:659:17: got restricted gfp_t\nblock/genhd.c:659:29: warning: incorrect type in argument 2 (different base types)\nblock/genhd.c:659:29: expected restricted gfp_t [usertype] flags\nblock/genhd.c:659:29: got unsigned int\nblock: kmalloc args reversed\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" } ], "next": "879040742cf09f2360a9ac41846288707e4e567c" }