)]}' { "log": [ { "commit": "9acc1853519a0473620d424105f9d49ea5b4e62e", "tree": "fd4d03bf82ea604e71f8e805e557023226c2dd42", "parents": [ "f3cef7a99469afc159fec3a61b42dc7ca5b6824f" ], "author": { "name": "Jay Lan", "email": "jlan@engr.sgi.com", "time": "Sat Sep 30 23:28:58 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 01 00:39:29 2006 -0700" }, "message": "[PATCH] csa: Extended system accounting over taskstats\n\nAdd extended system accounting handling over taskstats interface. A\nCONFIG_TASK_XACCT flag is created to enable the extended accounting code.\n\nSigned-off-by: Jay Lan \u003cjlan@sgi.com\u003e\nCc: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nCc: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jes Sorensen \u003cjes@sgi.com\u003e\nCc: Chris Sturtivant \u003ccsturtiv@sgi.com\u003e\nCc: Tony Ernst \u003ctee@sgi.com\u003e\nCc: Guillaume Thouvenin \u003cguillaume.thouvenin@bull.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "f3cef7a99469afc159fec3a61b42dc7ca5b6824f", "tree": "b1805a0d0b4a273b8ce0a245c17570ff18abdc5b", "parents": [ "7d1bdca9b06acb3df07329eaff72d5eaf1543287" ], "author": { "name": "Jay Lan", "email": "jlan@engr.sgi.com", "time": "Sat Sep 30 23:28:55 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 01 00:39:29 2006 -0700" }, "message": "[PATCH] csa: basic accounting over taskstats\n\nAdd some basic accounting fields to the taskstats struct, add a new\nkernel/tsacct.c to handle basic accounting data handling upon exit. A handle\nis added to taskstats.c to invoke the basic accounting data handling.\n\nSigned-off-by: Jay Lan \u003cjlan@sgi.com\u003e\nCc: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nCc: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jes Sorensen \u003cjes@sgi.com\u003e\nCc: Chris Sturtivant \u003ccsturtiv@sgi.com\u003e\nCc: Tony Ernst \u003ctee@sgi.com\u003e\nCc: Guillaume Thouvenin \u003cguillaume.thouvenin@bull.net\u003e\nCc: \"Michal Piotrowski\" \u003cmichal.k.k.piotrowski@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "0ae646845b603e9df5711084436d389f8371ffb3", "tree": "1554ba465812cc10c10b9bd28020a34effe0d8d4", "parents": [ "17db952cd16cecc76937b138c685ae3d198ab17c" ], "author": { "name": "Balbir Singh", "email": "balbir@in.ibm.com", "time": "Sat Sep 30 23:28:53 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 01 00:39:29 2006 -0700" }, "message": "[PATCH] Fix taskstats size calculation (use the new genetlink utility functions)\n\nThe addition of the CSA patch pushed the size of struct taskstats to 256\nbytes. This exposed a problem with prepare_reply(), we were not allocating\nspace for the netlink and genetlink header. It worked earlier because\nalloc_skb() would align the skb to SMP_CACHE_BYTES, which added some additonal\nbytes.\n\nSigned-off-by: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jamal Hadi \u003chadi@cyberus.ca\u003e\nCc: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nCc: Thomas Graf \u003ctgraf@suug.ch\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Jay Lan \u003cjlan@engr.sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "fe4944e59c357f945f81bc67edb7ed1392e875ad", "tree": "9634365d416ac574442fc7e21e1eaa26a71a26d1", "parents": [ "e1ef4bf23b1ced0bf78a1c98289f746486e5c912" ], "author": { "name": "Thomas Graf", "email": "tgraf@suug.ch", "time": "Fri Aug 04 23:03:05 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 14:53:43 2006 -0700" }, "message": "[NETLINK]: Extend netlink messaging interface\n\nAdds:\n nlmsg_get_pos() return current position in message\n nlmsg_trim() trim part of message\n nla_reserve_nohdr(skb, len) reserve room for an attribute w/o hdr\n nla_put_nohdr(skb, len, data) add attribute w/o hdr\n nla_find_nested() find attribute in nested attributes\n\nFixes nlmsg_new() to take allocation flags and consider size.\n\nSigned-off-by: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d94a041519f3ab1ac023bf917619cd8c4a7d3c01", "tree": "6d7d476814db70fab691242c7663204c64dbb43e", "parents": [ "7d94dddd438bcba97db44f120da39bb001b5249f" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Sun Jul 30 03:03:11 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Jul 31 13:28:37 2006 -0700" }, "message": "[PATCH] taskstats: free skb, avoid returns in send_cpu_listeners\n\nAdd a missing freeing of skb in the case there are no listeners at all.\nAlso remove the returning of error values by the function as it is unused\nby the sole caller.\n\nSigned-off-by: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nSigned-off-by: Chandra Seetharaman \u003csekharan@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7d94dddd438bcba97db44f120da39bb001b5249f", "tree": "edd737ba4e06ec086f0296a7d99a13e240c7ab93", "parents": [ "15a647eba94c3da27ccc666bea72e7cca06b2d19" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Sun Jul 30 03:03:10 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Jul 31 13:28:37 2006 -0700" }, "message": "[PATCH] make taskstats sending completely independent of delay accounting on/off status\n\nComplete the separation of delay accounting and taskstats by ignoring the\nreturn value of delay accounting functions that fill in parts of taskstats\nbefore it is sent out (either in response to a command or as part of a task\nexit).\n\nAlso make delayacct_add_tsk return silently when delay accounting is turned\noff rather than treat it as an error.\n\nSigned-off-by: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "bb129994c3bff9c5e8df91f05d7e9b6402fbd83f", "tree": "8d8c4e27af727c9de42ac6aeae04f70537e6bd01", "parents": [ "f9fd8914c1acca0d98b69d831b128d5b52f03c51" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Fri Jul 14 00:24:47 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jul 14 21:53:57 2006 -0700" }, "message": "[PATCH] Remove down_write() from taskstats code invoked on the exit() path\n\nIn send_cpu_listeners(), which is called on the exit path, a down_write()\nwas protecting operations like skb_clone() and genlmsg_unicast() that do\nGFP_KERNEL allocations. If the oom-killer decides to kill tasks to satisfy\nthe allocations,the exit of those tasks could block on the same semphore.\n\nThe down_write() was only needed to allow removal of invalid listeners from\nthe listener list. The patch converts the down_write to a down_read and\ndefers the removal to a separate critical region. This ensures that even\nif the oom-killer is called, no other task\u0027s exit is blocked as it can\nstill acquire another down_read.\n\nThanks to Andrew Morton \u0026 Herbert Xu for pointing out the oom related\npitfalls, and to Chandra Seetharaman for suggesting this fix instead of\nusing something more complex like RCU.\n\nSigned-off-by: Chandra Seetharaman \u003csekharan@us.ibm.com\u003e\nSigned-off-by: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "f9fd8914c1acca0d98b69d831b128d5b52f03c51", "tree": "18c2506bb31e49a9e2c4a483d37b0394c815fe9d", "parents": [ "c8924363da07aec213e5d359f23eeae1fff91951" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Fri Jul 14 00:24:47 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jul 14 21:53:57 2006 -0700" }, "message": "[PATCH] per-task delay accounting taskstats interface: control exit data through cpumasks\n\nOn systems with a large number of cpus, with even a modest rate of tasks\nexiting per cpu, the volume of taskstats data sent on thread exit can\noverflow a userspace listener\u0027s buffers.\n\nOne approach to avoiding overflow is to allow listeners to get data for a\nlimited and specific set of cpus. By scaling the number of listeners\nand/or the cpus they monitor, userspace can handle the statistical data\noverload more gracefully.\n\nIn this patch, each listener registers to listen to a specific set of cpus\nby specifying a cpumask. The interest is recorded per-cpu. When a task\nexits on a cpu, its taskstats data is unicast to each listener interested\nin that cpu.\n\nThanks to Andrew Morton for pointing out the various scalability and\ngeneral concerns of previous attempts and for suggesting this design.\n\n[akpm@osdl.org: build fix]\nSigned-off-by: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nSigned-off-by: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nSigned-off-by: Chandra Seetharaman \u003csekharan@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "ad4ecbcba72855a2b5319b96e2a3a65ed1ca3bfd", "tree": "a2f5b98598948525de77ab594e4432f09a230388", "parents": [ "25890454667b3295f67b3372352be90705f8667c" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Fri Jul 14 00:24:44 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jul 14 21:53:57 2006 -0700" }, "message": "[PATCH] delay accounting taskstats interface send tgid once\n\nSend per-tgid data only once during exit of a thread group instead of once\nwith each member thread exit.\n\nCurrently, when a thread exits, besides its per-tid data, the per-tgid data\nof its thread group is also sent out, if its thread group is non-empty.\nThe per-tgid data sent consists of the sum of per-tid stats for all\n*remaining* threads of the thread group.\n\nThis patch modifies this sending in two ways:\n\n- the per-tgid data is sent only when the last thread of a thread group\n exits. This cuts down heavily on the overhead of sending/receiving\n per-tgid data, especially when other exploiters of the taskstats\n interface aren\u0027t interested in per-tgid stats\n\n- the semantics of the per-tgid data sent are changed. Instead of being\n the sum of per-tid data for remaining threads, the value now sent is the\n true total accumalated statistics for all threads that are/were part of\n the thread group.\n\nThe patch also addresses a minor issue where failure of one accounting\nsubsystem to fill in the taskstats structure was causing the send of\ntaskstats to not be sent at all.\n\nThe patch has been tested for stability and run cerberus for over 4 hours\non an SMP.\n\n[akpm@osdl.org: bugfixes]\nSigned-off-by: Shailabh Nagar \u003cnagar@watson.ibm.com\u003e\nSigned-off-by: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jay Lan \u003cjlan@engr.sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "6f44993fe1d7b2b097f6ac60cd5835c6f5ca0874", "tree": "0f349f4e6c28cc5d11b7083273543a294c437216", "parents": [ "c757249af152c59fd74b85e52e8c090acb33d9c0" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Fri Jul 14 00:24:41 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jul 14 21:53:56 2006 -0700" }, "message": "[PATCH] per-task-delay-accounting: delay accounting usage of taskstats interface\n\nUsage of taskstats interface by delay accounting.\n\nSigned-off-by: Shailabh Nagar \u003cnagar@us.ibm.com\u003e\nSigned-off-by: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jes Sorensen \u003cjes@sgi.com\u003e\nCc: Peter Chubb \u003cpeterc@gelato.unsw.edu.au\u003e\nCc: Erich Focht \u003cefocht@ess.nec.de\u003e\nCc: Levent Serinol \u003clserinol@gmail.com\u003e\nCc: Jay Lan \u003cjlan@engr.sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "c757249af152c59fd74b85e52e8c090acb33d9c0", "tree": "78495f661fe537bf5087b24e6577659de8725b5a", "parents": [ "fb0ba6bd021248b6bdc58a7b1213a55a6776a38a" ], "author": { "name": "Shailabh Nagar", "email": "nagar@watson.ibm.com", "time": "Fri Jul 14 00:24:40 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jul 14 21:53:56 2006 -0700" }, "message": "[PATCH] per-task-delay-accounting: taskstats interface\n\nCreate a \"taskstats\" interface based on generic netlink (NETLINK_GENERIC\nfamily), for getting statistics of tasks and thread groups during their\nlifetime and when they exit. The interface is intended for use by multiple\naccounting packages though it is being created in the context of delay\naccounting.\n\nThis patch creates the interface without populating the fields of the data\nthat is sent to the user in response to a command or upon the exit of a task.\nEach accounting package interested in using taskstats has to provide an\nadditional patch to add its stats to the common structure.\n\n[akpm@osdl.org: cleanups, Kconfig fix]\nSigned-off-by: Shailabh Nagar \u003cnagar@us.ibm.com\u003e\nSigned-off-by: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jes Sorensen \u003cjes@sgi.com\u003e\nCc: Peter Chubb \u003cpeterc@gelato.unsw.edu.au\u003e\nCc: Erich Focht \u003cefocht@ess.nec.de\u003e\nCc: Levent Serinol \u003clserinol@gmail.com\u003e\nCc: Jay Lan \u003cjlan@engr.sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" } ] }