)]}' { "log": [ { "commit": "009789f040b71699278e70a6664701c10065e430", "tree": "83692d9eec2773e1d78628120f8e36807efe233b", "parents": [ "b72b71c6cb6ecc564d4d5f9c512a7df269837846" ], "author": { "name": "Chris Peterson", "email": "cpeterso@cpeterso.com", "time": "Tue Jun 16 15:33:43 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jun 16 19:47:49 2009 -0700" }, "message": "slow-work: use round_jiffies() for thread pool\u0027s cull and OOM timers\n\nRound the slow work queue\u0027s cull and OOM timeouts to whole second boundary\nwith round_jiffies(). The slow work queue uses a pair of timers to cull\nidle threads and, after OOM, to delay new thread creation.\n\nThis patch also extracts the mod_timer() logic for the cull timer into a\nseparate helper function.\n\nBy rounding non-time-critical timers such as these to whole seconds, they\nwill be batched up to fire at the same time rather than being spread out.\nThis allows the CPU wake up less, which saves power.\n\nSigned-off-by: Chris Peterson \u003ccpeterso@cpeterso.com\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b415c49a864dab8ee90713833d642dd461eccae9", "tree": "f5e763afdab9a43ff632993d7be49e73d8a6df6c", "parents": [ "6adc74b7d03c06a8e15d51fe33c3d924ada9271a" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Thu Jun 11 13:12:55 2009 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jun 11 11:26:38 2009 -0700" }, "message": "slow_work_thread() should do the exclusive wait\n\nslow_work_thread() sleeps on slow_work_thread_wq without WQ_FLAG_EXCLUSIVE,\nthis means that slow_work_enqueue()-\u003e__wake_up(nr_exclusive \u003d\u003e 1) wakes up all\nkslowd threads. This is not what we want, so we change slow_work_thread() to\nuse prepare_to_wait_exclusive() instead.\n\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "418df63c2d94f238ac7e1d1d53be35dd6b7a7252", "tree": "d9ab346ae46a45a4bff12fe7dbc615b7469d17a5", "parents": [ "3e241ff0c57cb610301009fe8c3b9da4c6877800" ], "author": { "name": "Jonathan Corbet", "email": "corbet@lwn.net", "time": "Wed Apr 22 12:01:49 2009 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 24 07:47:59 2009 -0700" }, "message": "Delete slow-work timers properly\n\nSlow-work appears to delete its timer as soon as the first user\nunregisters, even though other users could be active. At the same time, it\nnever seems to delete slow_work_oom_timer. Arrange for both to happen in\nthe shutdown path.\n\nSigned-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8f0aa2f25b31ba27db84259141e52ee6ec0d2820", "tree": "f2935c02ac8508dacd125bb1c8bb6f4879fe778e", "parents": [ "12e22c5e4bc08ab4b05ac079fe40d9891c5e81a0" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "message": "Document the slow work thread pool\n\nDocument the slow work thread pool.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Steve Dickson \u003csteved@redhat.com\u003e\nAcked-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nAcked-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nTested-by: Daire Byrne \u003cDaire.Byrne@framestore.com\u003e\n" }, { "commit": "12e22c5e4bc08ab4b05ac079fe40d9891c5e81a0", "tree": "e3d34a8f21d4c00dff311dfef564c59a76e1ae70", "parents": [ "109d9272c423f46604d45fedfe87e21ee0b25180" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "message": "Make the slow work pool configurable\n\nMake the slow work pool configurable through /proc/sys/kernel/slow-work.\n\n (*) /proc/sys/kernel/slow-work/min-threads\n\n The minimum number of threads that should be in the pool as long as it is\n in use. This may be anywhere between 2 and max-threads.\n\n (*) /proc/sys/kernel/slow-work/max-threads\n\n The maximum number of threads that should in the pool. This may be\n anywhere between min-threads and 255 or NR_CPUS * 2, whichever is greater.\n\n (*) /proc/sys/kernel/slow-work/vslow-percentage\n\n The percentage of active threads in the pool that may be used to execute\n very slow work items. This may be between 1 and 99. The resultant number\n is bounded to between 1 and one fewer than the number of active threads.\n This ensures there is always at least one thread that can process very\n slow work items, and always at least one thread that won\u0027t.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nAcked-by: Steve Dickson \u003csteved@redhat.com\u003e\nAcked-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nAcked-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nTested-by: Daire Byrne \u003cDaire.Byrne@framestore.com\u003e\n" }, { "commit": "109d9272c423f46604d45fedfe87e21ee0b25180", "tree": "b6bd1d94fd3e018ee97dedb577aa6757787db1a9", "parents": [ "07fe7cb7c7c179f473fd9c823348fd3eb5dad369" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "message": "Make slow-work thread pool actually dynamic\n\nMake the slow-work thread pool actually dynamic in the number of threads it\ncontains. With this patch, it will both create additional threads when it has\nextra work to do, and cull excess threads that aren\u0027t doing anything.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nAcked-by: Steve Dickson \u003csteved@redhat.com\u003e\nAcked-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nAcked-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nTested-by: Daire Byrne \u003cDaire.Byrne@framestore.com\u003e\n" }, { "commit": "07fe7cb7c7c179f473fd9c823348fd3eb5dad369", "tree": "516c959de929a4c39870629e550b3307601fa73f", "parents": [ "8fe74cf053de7ad2124a894996f84fa890a81093" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri Apr 03 16:42:35 2009 +0100" }, "message": "Create a dynamically sized pool of threads for doing very slow work items\n\nCreate a dynamically sized pool of threads for doing very slow work items, such\nas invoking mkdir() or rmdir() - things that may take a long time and may\nsleep, holding mutexes/semaphores and hogging a thread, and are thus unsuitable\nfor workqueues.\n\nThe number of threads is always at least a settable minimum, but more are\nstarted when there\u0027s more work to do, up to a limit. Because of the nature of\nthe load, it\u0027s not suitable for a 1-thread-per-CPU type pool. A system with\none CPU may well want several threads.\n\nThis is used by FS-Cache to do slow caching operations in the background, such\nas looking up, creating or deleting cache objects.\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nAcked-by: Serge Hallyn \u003cserue@us.ibm.com\u003e\nAcked-by: Steve Dickson \u003csteved@redhat.com\u003e\nAcked-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nAcked-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nTested-by: Daire Byrne \u003cDaire.Byrne@framestore.com\u003e\n" } ] }